Speaking: From Intention to Articulation

2/17/2011

Home

Books: Speaking

Library

What's New

Departments

Subscriber : University of Pennsylvania Library » LOG IN space For Librarians

GO

REGISTER

Advanced Search

Selected Title Details OPTIONS

Speaking Willem J. M. Levelt

RELATED TOPICS Linguistics

"Required reading for anyone w orking in the field." -- Joseph Paul Sternberger, Contemporary Psychology

Aug 1993 ISBN 0262620898

In Speaking, Willem "Pim" Levelt, Director of the Max-Planck-Institut für Psycholinguistik, accomplishes the formidable task of covering the entire process of speech production, from constraints on conversational appropriateness to articulation and self-monitoring of speech. Speaking is unique in its balanced coverage of all major aspects of the production of speech, in the completeness of its treatment of the entire speech process, and in its strategy of exemplifying rather than formalizing theoretical issues.

584 pp. "[Levelt] synthesizes the state of the art w ith remarkable breadth, depth, and clarity. In 500 pages of highly readable text and forty pages of bibliography, he integrates, as no one has done previously, many lines of research that have often been pursued in ignorance of one another.... An outstandingly instructive book." -- Dan Sperber, Times Literary Supplement "A w ork of massive scholarship and authority." -- Brian Butterw orth, Times Higher Education Supplement

Table of Contents Preface Acknow ledgments Author's Notes 1 The Speaker as Information Processor 2 The Speaker as Interlocutor 3 The Structure of Messages 4 The Generation of Messages 5 Surface Structure 6 Lexical Entries and Accessing Lemmas 7 The Generation of Surface Structure 8 Phonetic Plans for Words and Connected Speech 9 Generating Phonetic Plans for Words 10 Generating Phonetic Plans for Connected Speech 11 Articulating 12 Self-Monitoring and Self-Repair Appendix: Symbols From the International Phonetic Alphabet, With Examples Bibliography Author Index Subject Index

Terms of Use | Privacy Policy | Contact | FAQ © 2010 The MIT Press

proxy.library.upenn.edu:2580/…/view?is…

1/1

) 1- 4. North-Hollanri Lingua92 ( 1994

Preface For many years, the topic of lexical acquisition was a stepchild in linguistic inquiry . While" the acquisition of syntax was acknowledged to be organized according to a set of deep and highly structured innate principles, word meanings were assumed to be acquired by a simple and particularistic associativeprocedure that mapped perceptual experienceonto phonetic entities . We believe this stance was anomalous from the start, as had been pointed out eloquently by Plato and, in modem times, by Quine. Not all words are perceptually based; even those that are map quite abstractly from ' ' perceptual categories; and neither perception nor learning is simple . During the past decade, it has becomeclear from linguistic inquiry that the lexicon is more highly structured than heretofore thopght ; moreover, that much of grammar turns on critical - and universal - links between syntactic and lexical-semanticphenomena. Hence the present volume. The papers grew out of a workshop held at the University of Pennsylvania. Its aim was to bring together psychologists, computer scientists, and linguists whosejoint concern was the lexicon and its acquisition; these researchershave typically worked quite separately and on problems thought to be disparate. The volume is organized in six sections: 1. Nature of the mental lexicon.. The two essaysthat open the collection are from linguists gleaning general perspectiveson lexical learning from the linguistic facts themselves. The first , by Edwin Williams , arguesfrom the real complexity and idiosyncratic nature of the mental lexicon (including lexical phrases) that innate linguistic and categorization principles are not enough, that real children require a learning theory more sophisticated than usually has been supposed. A detailed analysis of English causativealternation verbs by Levin and Hovav eloquently supports Williams on the complexity of lexical structure. 2. Discovering the word units.. To acquire a vocabulary, the learner requires some proceduresfor segmentingthe continuously varying sound wave into word -sizedpieces. Cutler discusses this problem, presentingevidencethat a bias toward rhythmic alternation servesas a powerful bootstrap for infants (and adults , whatever their specific language background) solving this problem. Kelly and Martin then show that , armed with such blases, learners B.V. All riahtsreserved 0024 - 3841 /94/ 507.00 ~ 1994- ElsevierScience SSDI 0024- 3841( 93) EO036- 7

2

Preface

exploit multiple probabilistic cuesto discover the word units: domain-general learning procedurescan be used to discover many domain-specificfacts about language. ' 3. Categorizing the world: The word meaning culled from children s interaction with the world ultimately depends on how they represent their ' observations. A first question is whether children s mental representations (categories, and the theories that bind them) are the same, no matter the age ' of the learner. In her article, Carey makes a strong casefor a ' discontinuity theory, in which the very basis of inference from observation onto meaningidentification changesover developmental time. Keil , in contrast, argues that the mentalities of children and adults hardly differ. For both , his view is that concept and word acquisition involves both the discovery of multiple probabilistic surface factors (reminiscent of Kelly and Martin ) and some theorydriven generalizations. 4. Categories, words, and language: A further set of articles again documents ' how the child usesthe evidenceof observation to decide on a word s meaning. But theseemphasizethe role ~f linguistic principles as an additional determining factor. Thus Markman points out that the learner is biased ' toward a ' whole object assumption - a new word probably namesthe whole animal rather than one of its parts or properties. But then how could a child learn ' nose' or ' white' ? She provides evidence for a linguistic principle ' ' ( mutual exclusivity ) that leads children to avoid two labels for the same object. Waxman shows experimentally that such blases, even in the toddler , are specifically related to the characteristic semantic properties typical of lexical classes- nouns name objects but adjectives name properties. Landau ' ' ' ' shows how the separation and respectivestructures of the what and where systems(as documentedby neuropsychologists) are related, in learning, to the child ' s acquisition of object namesvs. place expressions. All thesearticles are neutral concerning the direction of the causal links here: the extent to which the perceptual- conceptual facts determine lexical class assignment, or the other way round. Bloom takes a stronger view in discussingthe acquisition of nominal subtypes (count vs. mass noun, noun phrase) : He holds that very early in life children consider syntactic form classas relevant to determining a novel word ' s meaning. 5. The caseof verbs: The papers mentioned thus far make a strong case that the form -meaning linkages that typify language design are exploited by children acquiring the vocabulary of their native tongue. Grimshaw ( 1981) and Pinker ( 1984) emphasized, on the basis of these linkages, that children could derive lexical class assignment from experience-based learning of the

Preface

word ' s meaning; Pinker showed further that such a procedure would be ' marvelously useful for the construction of phrase structure ( semantic boots' . Landau and Gleitman 1985 ( ) emphasized that learning might trapping ) sometimes, even usually, be the other way round ; specifically, that syntactic structure plays a necessary role in narrowing the search space for verb ' ' meanings ( syntactic bootstrapping ) . The three articles in this section hotly debate how the semantics-syntax links (specifically, the relation betweenverb argument structures and subcategorization structures) are implicated in the learning processfor verbs. Fisher, Hall , Rakowitz , and Gleitman emphasize the role of syntax in narrowing the range of interpretations made available by experience; Pinker argues that logically the child must work from experience to determine the syntax in the first place; and Grimshaw stalwartly proposes a reconciliation of these two views. 6. Proceduresfor verb learning: The positions just described, even if correct, can suggestonly some boundary conditions on a learning theory for verbs. Fisher et al., asjust described, assertthat some rudiments of syntax are necessaryfor learning the verb meanings. But as Pinker points out , the question then would be: How did the child acquire the syntax if not by exploiting the word meanings themselves? Brent, in the best tradition of computer science, offers a discovery procedure for verb subcategorization that usesstring-local surface cues only , and does surprisingly well. The final article in the volume, from Steedman, was designedas a specific commentary on Brent' s work . But this short article is much more general than that. It strikes a number of sane cautionary notes about current theories of verb learning, of which the most important is that real acquisition is bound to be a messybusiness, with syntactic, semantic, and prosodic cues recruited by the child more or less catch-as-catch-can. Steedmanalso points out , in what we find an appropriately laudatory comment, that it is only in the presenceof ' explicit computational models such as Brent s that the bootstrapping theories can ever be developedand seriously evaluated. It is the hope of the editors that this compendium of topics and views on lexical learning will be of particular use to the linguists who constitute the Lingua audience, in two ways. First , we suppose it will be useful to see the kinds of methodological and substantive contributions that scientific psychology can make to the question of languagelearning - and henceto the theory of language. Second- as we have emphasizedearlier - becausesyntactic and lexical structure are so closely entwined, we expect that the various articles will be informative as to how theselinkages enter into the learning procedure for vocabulary.

Preface

We thank Lingua for the opportunity to air theseworks in this Special Issue, and we particularlythank the LinguaChief Editor, Teun Hoekstra, for this continuingaid and supportof this project. We thank also the National ScienceFoundation which, through an STC grant to the University of Institutefor Researchin Cognitivescience Pennsylvania , madethe Workshop . In that regard, we areespeciallygratefulto the staffof IRCS (Trisha possible Yanuzziand Chris Sandy) and to Carol Miller, Kimberly Cassidy , and Sally Davis (graduatestudentsin the Departmentof Psychology who ran the ) with verve and considerable , efficiency Workshop , good humor. Finally, we thank Stevenand Marcia Roth for a grant to Lila Gleitmanwhichaidedus in the preparationof the Volume.

, June1993 Philadelphia

Lila Gleitman Barbara Landau

Acknowledgments

This book emergedin the stimulating circle of the Max -Planck-lnstitut fiir ' Psycholinguistik and Nijmegen University . Numerous colleagues can t give up educating me in matters linguistic and psycholinguistic, and I would be at a losswithout them. Also , many distinguishedvisitors from all over the world have given me help and advice. I am sincerelygrateful to all of them. Some, from within and without this circle, have taken the effort to read drafts of chapters and to formulate comments, either orally or in writing . Before mentioning them, I must say that each of theseefforts has deeply touched me. Time and again I experiencedit asa great privilege to be taken so seriously and with so much friendship. In alphabetical order, these angelswere Manfred Bierwisch, Melissa Bowerman, Herbert Clark , Anne Cutler , Jane Edwards, Lyn Frazier, Merrill Garrett , Gerard Kempen, Wolfgang Klein , Aditi Lahiri , and John Marshall . I would also like to thank Gerard van Galen, who gave me a specialtutorial on psychomotor theory. Though I typed the manuscript myself with two index fingers on a terminal , finishing toucheshave beenmade by Uschi de Pagter and Edith Sjoerdsma, especiallyin completing the bibliography . Our librarian , Karin Kastens, also helped me on countlessoccasions. Many thanks to all three of them. The graphic work was done by Wil Maas during three months of ideal cooperation.

NiJmegen . February 29, 1988

Author ' s Notes

On Pronominalization in to the present chaos of person pronominalization male whether : conventions , Speakers, English by adhering to the following or . Hearers female , or generic , will receive masculine pronominalization addressees will be treated as female . When there are two or more interlocutors (i .e., speakers/ hearers) , the first one will be male , the second one female , and so on in alternation . General use of these conventions in psycholinguistics will , given the bias for language - comprehension research,

I will contribute

make most person reference female . On Tr ~

ription

I will follow the transcription conventions of the International Phonetic Alphabet . The phonetic symbols used are listed in the appendix .

) 7- 34. North-Holland Lingua92 ( 1994

Remarks on lexical knowledge EdwinWilliams CognitiveScienceLaboratory, Princeton University, 221 NassauStreet, Princeton, NJ08540, USA ' ' The abstractnessof lexical knowledge, and its independencefrom the words of a language, are is the the lexicon that idioms demonstrate . Abstract repository not only generally underestimated of fixed expressions, but of abstract language-particular phrase- and word-patterns as well. Paradigmsare language-particular patternings dimensionedand structured in languageparticular two abstract of these burden . The of with learning languageparticular patterns syncretism ways, but learned features of lexical knowledge falls ouside of what current thinking about language learning would allow.

1. Introduction I will survey two aspects of lexical knowledge which I feel pose special problems for learning. In both casesthe remarks will tend to emphasizethe extensivenessabstractness , and at the sametime the language-particularity of lexical knowledge, and will consequentlymagnify the learning problem. One conclusion that could be drawn from these observations is that intricate structures can be learned, and this learning is not adequately modelled by . either parameter setting or list learning. I t is useful to distinguish two notions of lexicon, one Bloomfieldian and the other grammatical. The Bloomfieldian lexicon is the repository of all of a ' is the linguist' s theory of languages idiosyncracies. The grammatical 'lexicon ' the category of linguistic object we call a word . These are quite different conceptions, but have come to be identified in modern times. The result is a picture of grammar in which we have a clean streamlined syntax, where syntax is the theory of phrase, and a notion of word that entails that words are idiosyncratic and irregular at heart, with partial regularities expressedas ' ' redundancy rules whose name implies low level generalizationswhere exception is the rule. I think that this picture is wrong. A more correct picture is I think given in Disciullo and Williams ( 1986) . In that view, both the word formation and the ScienceB.V. All rightsreserved -3841 0024 /94/$07.00~ 1994- Elsevier -G SSDI0024-3841(93)EOOS9

E. Williams / Remarkson lexical knowledge

' ' syntactic system are clean streamlined systems, independent of the lexicon. The lexicon is the repository of forms about which something specialmust be learned. This certainly includes, for example, all the monomorphemic words, but it also includes composed units of both syntax and word formation , composed units whose properties do not all follow from the rules which compose them. This would include a great deal of words, but also a great deal of phrases. In fact, I now think there are more lexical phrasesthan there are lexical words, but this re~ ains a speculation. In addition , I think that lexical knowledge includes knowledge of complex abstract.structures that cannot be arrived at through parameter setting, and which must be learned from the data in a strong sense. I will discusstwo of these: ' abstract idioms' in section 2, and paradigms in section 3. If my view of theseis correct, the rich structure each exhibits results not from a rich innate predefined linguistic structure, but rather from a richly structured learning strategy. If nothing else, then, the remarks that follow draw attention to the learning problems that might arise if first , more ~f linguistic knowledge is lexical than has been thought, and second, if acquired lexical knowledge is more abstract and structured in language-particular ways than has been thought . I think that the sort of structure that is found in each case reveals the hand of the learning strategy.

2. Idioms

The phrasesin the lexicon are called ' idioms' . We generally think of idioms as ' frozen' expressions. I will use the term idiom to refer to any defined unit whose definition does not predict all of its properties. This will include the phrase kick the bucket, a phrase, whose idiomatic meaning is die, but also transmission, which unpredictably means ' such and such car part ' . We don' t think of transmissionas an idiom , but it will be useful and I believecorrect to include it . It has been commonly assumedthat idioms are well-formed structures, but I think lately this view has been questioned. To say that an idiom is well formed is to say simply that it conforms to the rules of formation for the type that it is. Of course, the idiom does not conform to all of the rules, else it would not be an idiom . What rules do idioms obey? It is useful to consider the rules of ' form ' apart from the rules of interpretation . We find idioms violating both types of rules.

E. Williams I Remarks on lexical knowledge

The idioms that do not obey the rules of interpretation are deviant in meaning. The rules of argument structure are not obeyed in some cases, and in others, the rules of referenceare not obeyed. We might refer to both these ' ' types of idioms as semantic idioms. It is perhaps misleading for me to say that there are idioms which do not obey the rules of form , for it is an overwhelming fact that idioms obey the basic rules of fonn it) a language. For example, all idioms in English obey the ' head-initial ' setting of the head position parameter. On the other hand, there are unpredictable, language-particular exploitations of the formal possibilities ' in a language; we might call theseidioms ' formal idioms. A widely shared set of assumptionsabout idioms is the following : ( 1) usual assumptions: idioms are listed idioms are well-formed phrases idioms have empty parts : The cat has Xs tongue relation of idioms to syntax: insert the idiom , then fill in the parts ' So, for example, The cat has got X s tongue is an idiomatic phrase, fully consistent with the laws of English syntax, which is inserted in a phrase marker for an S position ; further substitution of a referential NP for the position of X will yield an English sentence. These assumptions, though common, lead to some surprising conclusions, or at least some surprising conclusions, or at least some surprising questions. . In particular , they lead to the view that a great deal of what we have called ' ' ' rule' in syntax might really be idiom . To give an example, consider the fact that embeddedquestions in English must begin with a WH word. It is well known that this is not to be describedby making WH movement obligatory in English. What kind of information is it then that [ Wh-phraseS ] is a (potential) embeddedquestion in English? It COULD be that this form is an idiom : ' ' (2) [[ Wh-phraseS ]. , : embeddedquestion Under this conception, (2) could be listed in the lexicon, awaiting insertion into some embedded S position, where it is an embedded question (as the slogan after the colon above indicates) . Further substitution of some wh

E. Williams I Remarkson lexical knowledge

phrase for the wh-phrase position in (2) and of an S for the S position in (2) will complete the sentence, exactly parallel to The cat has got X 's tongue. One does not ordinarily think of this as idiomatic information , even though it is languageparticular ; rather, one thinks of this feature of English as one of the ' parametric' possibilities. (2) is not the ordinary way to encode this information . Example (2) may seem sufficiently different from The cat has got your tongue that it would never be misidentified as an idiom in the same sense. However, I intend to supply enough examplesof casesintermediate between the two ' that I cannot see how it cannot be questioned whether (2) is the correct description or not. 2.1. Idioms are instances of we//- formed structures

The view that I am presentingherewould have large numbersof objects of a sometimesquite abstract character stored in the idiom lexicon of language. This view might then be seenas convergi.ng with severalrecent proposals that . there are 'constructions' in language, a backlash against the principles and parameters model. There is, though, a fundamental difference between the proposal here and those proposals - specifically, I hold that idioms are well formed, and that the rules of well-formednessare simple ' parameterized' rules. 2.1.1. Filmore's examples For example: C. Filmore (p.c.) has sugg~sted to me that while most prepositions in English are prepositions , some are not , and consequentlythe notion that there is a uniform head position in English is not correct ; the most ,:>ne can say is that there are a number of constructions in which the head is leftmost, and some others in which it is not. His candidates for postposition in English are notwithstandingand ago: (3a) John notwithstanding, we will go there tomorrow . (3b) John left 3 days ago. I believe that neither of these is a preposition, and that the notion that English is head initial can be maintained in a strong form . Notwithstandingcan be assimilated to the following constructions:

(4a) John aside, ... That noted, ...

on.lexical E. Williams / Remarks knowledge which are clearly small clauseconstructionsof somekind. The difference and asideis that asidecan appearin other small betweennotwithstanding whereas clauseconstructions , notwithstandingis restrictedto the context indicated: (Sa) put that aside, * notwithstanding . (Sb) We may ~xpress this restriction on notwithstanding by not listing it in the ' ' lexicon on its own , but only as a part of the following idiom :

' ' ]sc S]s: evenwith NP, S (6) [[NP notwithstanding Importantly, (6) is an instanceof a structurewhich is well fonnedin English , namelythe structureof (4) : independently

(7) rNP - AP]- S > >

' ' [ NPnotwithstandingS: even

' I will usethe doublecarat sign' > > ' to mean' hasa well-formedinstance. I think ago as well has a better analysisthan the postpositionalone. Consider: (8a) long (8b) 5 minutes (8c) a few days

ago in the past before the party

It appears that time prepositions in general can take some sort of extent specification; however, this specification in general precedesthe time preposition , as we might expect specifiers to. (8c, d) shows clearly that the extent specification is not a part of the complement structure, as the prepositions in and before have complements to their right . Now , ago differs from these in two ways. First , it cannot have a complement to the right . This however shows nothing except that ago is intransitive. Secondly, in the caseof ago, the extent specification is obligatory . Here, if we think that extent specification is ordinarily outside of the subcategorizational reach of a head, we might appeal to an idiom to capture this exceptional feature of ago:

E. Williams/ Remarkson lexicalknowledge

' I complements right , specs left ' > > assign [(Extent-spec) P (NP)]pp> > [Extent-specago]pp Here, the idiom is a well-formed subinstanceof prepositional phrase, which itself is a ' subinstance' of all the structures that fall under the parametric specializationsof English. In sum then, although the notion of idiom we are employing is quite general, it is at the same time quite restrictive, in that all idioms conform to more abstract (though perhaps parameterizedlanguageparticular) principles. 2.1.2. What rules.do idioms obey? Idioms basically obey the (rules determining the) phrase structure of the language, but not the rules of reference; so for example, kick the bucket is a well-formed VP, but no referenceto any bucket is made. These two observations set someouter bounds. I think a more interestingprobe is theta structure, where it turns out in fact that idioms diff ~r in the extent of their conformity . . It is well known that idioms differ in their syntactic ' transparency' - ' kick the bucket' does not passivize, whereas' cross that bridge when we come to it ' does. Fiengo ( 1974) connected this difference to a further difference, namely, a difference in whether the object can be quantified : ( l Oa) * They kicked the buckets (as an idiom ) ( lOb) We will cross those bridges when we get to them. In (b), ' bridges' refers to ' problems' , and (b) says that we are talking about more than one problem. In (a) on the other hand, it is impossible to refer to more than one death. Is this an arbitrary difference? I do not think so; I think it follows from the fact that the meaning of ' kick the bucket' is intransitive. This intransitivity of meaning meansthat there can not be a correspondencebetweenthe syntactic and semanticargument structures, sincesyntax is transitive here. In (b) on the other hand, the verb cross can be assigneda transitive theta structure, and one of its theta roles can be assignedto bridges. So what is wrong with the meaning of (b) is simply the reference, not the theta structure: cross refers to solve and bridges refers. to problems. But in (a), it is not just reference, but . theta structure itself, which is idiosyncratic. If this is so, then (b) has a transitive theta structure, and its passivizability is no surprise, if passiveis an operation on theta structures, as I believe it is.

E. Wi/Iiams / Remarkson lexical knowledge

And the fact (a) is not passivizableis no surprise either, sincethere is no theta relation betweenthe direct object and the rest of the sentence. If this account is correct, then theta structure is not necessarilyrespectedin idioms, though it may be. A child learning an idiom will aggressivelyassignit as much structure as possible according to its rules. If it can assign it a semanticargument structure basedon its syntactic argument structure, it will , but if not , then the idiom will lack argument structure.

2.2. Abstractidioms 2.2.1. Idioms with instances If we begin with the idea that an idiom is a phrase, and that it can contain empty parts, then we immediately face the question, how empty can an idiom be? Can it , for example, be mostly empty? The question arisessharply for an example like the English noun pants. This noun, as is well known , is an ' ' arbitrary plural ; a shirt , for example, of the same genus topologically (at least when the fly of the pants is unzipped) is singular. This is a trivial piece of .idiomatic , that is unpredictable, information about this noun, that it must be plural. Strikingly , though, this is not true of this noun alone, but in fact of every word that has the meaning that pants has: something worn on the legs in such and such a way : ( 11) pants jeans shorts cutoffs culottes bermudas ... New, made-up terms for lower trunk wear must conform as well. One exception, bathing -suit , is an exception only in that it does not specifically refer to lower - trunk wear, but rather means whatever one wears to bathe in ; it is an accidental fact of current fashion that this refers to lower -trunk wear. Now , what sort of information is this? It is information about a general restriction on form that follows from meaning; specifically, if a noun is going to have such and such a meaning, then it must be plural ;

' ' ( 12) Ns +- : lower trunk wear I .have drawn the arrow from right to left to mean, if an item is going to have the meaning on the right , then it MUST have the form on the left. This is a different sort of idiomatic information from knowing that kick the bucket CAN mean die, and so the different notation.

E. Williams / Remarkson lexical knowledgE

Another example like pants is fish . Fish names , with some exceptions , are all unmarked plurals or have that as an option : trout , bass, perch , bream , yellow tail , mahimahi . The exceptions are not really fish , by and large : whale , guppy , minnow . Other animal families are untouched by this idio syncracy : bee, wasp , ant . As far as I can tell , this is an unpredictable , but ' ' information very general fact about English , and counts as idiomatic about the language . A further case of the same sort is the language - particular patterns of lexicalization identified by Talmy ( 1985) . He found that languages systematically differed in the kinds of verb meanings they allowed . For example , English allows verbs of motion to indicate a means of motion . Float can be used as a directional verb , but at the same time , it indicates the manner of motion : John floated under the bridge can mean that John moved under the bridge by Boating . Spanish and French lack entirely verbs of this kind . F/otter in French and flotar in Spanish ( float ) can mean only to float stationarily , and the restriction is apparently a hard and fast one. Similarly , verbs of posture ( sit , kneel , lie , etc.) differ systematically from language to language , in whether the stative , inchoative , or causative is the basic or underived form ; English , Japanese, and Spanish systematically differ in this choice - see Talmy ( 1985) for details . Again , we have language -particular variation of a quite general sort . ' ' Again , the information is idiomatic , but the question remains , how to ' ' represent it . We might again represent it as an idiom with a hole in it :

( 13a) [inchoative)v +- : posture verb (Japane~ where V is atomic. 13b motion verb]v +- : manner (English ) ( ) [ Now , the representation here I think is not so important as the question of what this sort of information is, and especiallyhow it is acquired. I think that language-particular patterns of the kind that havejust been discussed- pants in English, the motion verbs in Romance, the posture verbs, etc. - fall outside of the ' parametric' core, and yet, they are quite general, and basically exceptionless. This means some sort of general induction , the kind that is meant to extract ' lexical reduncancyrules' , must be capable of acquiring from ' of the rule. Some substitute for the data the ' exceptionlessness negative evidence, such as counting, and statisticking, is required. And somelimitation on the spaceof searches must hold , in order for the induction to remain in the realm of possibility .

E. Williams I Remarkson lexical knowledge

The most surprising casesare the casesthat achieveexceptionlessness . The ' idiom is pants exceptionless, in that any noun that means the right thing must participate in the idiom . The presentparticiple is another case: there are no present participles that do not end in -ing whereaspast tense forms, as is well known , are quite varied. The learner learns more than that there are no exceptions; he learns that there can be no exceptions. We will discuss such cases further in sec~ion 4. For the moment, we note that the formalism , though it at the same time proposed suffices to express exceptionlessness hides the learning problem implied by them. '

2.2.2. Formal idioms: Exploited and unexploitedavenues The pants phenomenon just examined is I think more widespread. Formally ' , the grammar permits singular nouns to have the meaning such and ' such a type of legwear , but this avenue is unexploited in English, thanks to ' ' ( 12) . By unexploited avenues I mean possibilities that the formal system would seemto allow , but which it does not use. In identifying such casesone always risks missing the formal explana~ion for the missing possibility, though a case like pants I think clearly shows that there will not always be one. As one example, consider compound terms in English and French. Both languageshave meansof putting together words from further words, or kind denoting terms. In French, the syntactic system is exploited; so, one has compound terms of the following kind : ( 14

)

VP

>

>

V

>

NP

>

V

N

>

>

essuie

'

'

glace

wipe

window

' ,

windshield '

wipers VP

>

>

V

PP

>

>

V

P

N

>

>

laisse

pour

'

'

- conte

left

for

count

,

'

' abandoned

one

Here, the double carat means ' has as an instance' . So, compound terms in French are instancesof syntactic constructions. As word level items they have their own limitations (e.g., no referential material may occur in them, and so, for example, no determiners are allowed) , but they are neverthelesswellformed syntactic objects. English , on the other hand , exploits a different system to form its compound terms :

( 15) [ X Y] y > > [ NN] N

E. WUliamsI Remarkson lexical knowledge

The system exploited here is the affixation system in the lexicon, which is . Ordinarily , Y is a suffix, forming the head of a word. English right -headed ' lets Y be a full noun, giving us compounds. Importantly , both languageshave both resources- English has the same (left -headed) syntax as French, and French has the same right -headedaffixation systemas English; however, they each exploit a different one of thesefor their compound terms. I assumethat this is ' idiomatic ' - that is, language' ' particular , but perhaps not parametric .

2.3. Syntacticidioms English embeddedquestions must begin with a wh-phrase. This is not to say that Wh-movement is obligatory in English, as it clearly is not ; not only do matrix questions not necessarilyundergo Wh -movement, but even a whword in an embeddedquestion need not move, when, for example, another has moved, or the complementizeris already wh, like whether:

( 16) Who wonderswhetherGeorge saw who ? The correct generalization is as stated: a Wh -word must appear at the beginning of an embeddedQuestion. What sort of information is that ? We might describeit as an idiom , in the sensedevelopedherein:

( 17) [wh-phraseS +- : embeddedquestion That is, a sentencewith a Wh-phrase at the beginning is a question, and nothing else is. The arrow goes backwards, as any embeddedquestion must have this form . ' Idiom ( 17) is a good candidate for a ' parameter , in that there is a small number of ways that question words can be dealt with : ( I ) moved to front ; (2) moved to verb (as in Hungarian); or (3) left in situ. However, I think there is good evidencethat idioms just like ( 17) must be countenanced, ones that are not reducible to parametersof variation . One caseis the ' amount' relatives. These have the following form :

) [wh-phraseS] (18a '

> > N what S : Little (amount relatives) [ ] I him food I had. 18b what ( ) gave I him what I had. 19 ( ) give

E. Williams I Remarks on lexical knowledge

( 18b) has an implication that ( 19) does not have, namely, that there was little food in question. Now , where does this implication come from ? It does not come from what, which does not have this implication in general, not even in free relatives, exceptin the context in ( 18b) . Furthermore, it does not inhere in free relatives in general. In fact, it occurs only in the structure in ( 18), it is idiosyncratic to that structure. Assuming that there is no parameter to set here, then this is a learned fact about this structure. What is interesting is how formally similar ( 18) is to ( 17) . The only difference is that the semantics of ( 18) is very particular, and therefore plausibly idiomatic , whereas the semantics of ( 17) is very general. But the formal means may be the same in the two cases: a feature of meaning and form are connected in an idiomatic entry in the lexicon. If a learner can induce ( 18), it would seem that ( 19) would be accessible to the same mechanism. A related sort of case arises from Subject Aux Inversion in English; the following is an idiom of English:

(20) [ V S]. +- : matrix yes/no question This is comparable to ( 17) - an obligatory idiom (that is, of the pants variety) has the effect of forcing syntactic rules to apply. More interesting are the casesof inversion which receivea conditional interpretation : (2Ia ) Had I been there, this would not have happened (21b) S > > [ V S]s: conditional > > [ had...]s > > [were...]vp The rule of inversion gives a large number of forms which are ungrammatical in this context :

* (22) Could I write poetry, I would not be a linguist . In fact, inversion in the conditional context works only for the auxiliaries had a~d were. What sort of infonnation is this? Importantly , the casesallowed in the construction are a subset of the casesallowed in general; hence, what is learned is that not all the formally allowed possibilities are realized. We will adopt the following convention for representingthis situation :

E. Williams I Remarks on lexical knowledge

(23) Instanceprinciple : If a fonn to which a meaning is assignedhas listed subinstances, then those subinstancesare exhaustive. 2.4. Idiom families

At one end of the language-particular information that a learner must acquire are the completely fixed expressions; at the other end are the broad ' . I have suggestedthat there are intermediate ' abstract typologiCalparameters 0 idioms _ pants and amount relatives, for example, which link theseendpoints with a graded continuum. In compensation for this more complicated situation , and the more complex learning problem that it poses, I have suggested that each level of ' abstractness' must conform to the level above it ; thus we ' ' do not have a wholesale theory of constructions , but still a broadly parametric model. This says that a construction - say, passive must conform to the typo logical pattern that is determined by some parameter but leaves open the possibility 0that not all features of the passive .settings, construction will be determined by this conformation . As further evidence of this view, I will discuss here some idiom families, that is, language-particular idiom patterns. Each idiom pattern has a number of idioms as instances. The principle reason for recognizing the existenceof the idiom family is that somelanguageswill have idioms of a certain sort, and others will lack them altogether, apparently in a way not related to the parametric typology of the languages, though of course one could always be wrong about that for any particular case. The most interesting sort of case I know of was pointed out to me by Martin Everaert (pic.) ; the idioms are of the form :

(24) NyPNy by side The two Ns are meant to be identical tokens of the samenoun, as in side by side. French, English, and Dutch have these idioms, Japaneselacks them ' ' entirely. For this reason, we would want to call (24) itself an abstract idiom of the sort discussedin previous sections. Idiom (24) probably has as its most imniediate instantiations not actual idioms, but further , more concrete, idiom families, one for each P that participates:

E. Williams I Remarkson lexical knowledge

(25) Nx P Nx > > Nx by Nx Nx to Nx Nx on Nx Nx for Nx Nx fromNx Someof the manyexamplesof instancesin thesesubfamiliesare:

(26) N to.N > >

hour minute second year

window door station head hand cheek

NbyN > >

minute, hour , etc. ; layer house cave

N for N > > dollar for dollar N from N > > limb from limb Non N > > layer on layer If these are idioms, then we expect them to be instances of well-formed structures in the language. One is at first hard-pressed to identify what structures these are well-formed instances of , but I think the most likely "candidate is the compound prepositional phrases, asfrom . .. to:

(27) John went [from a to b] The compound prepositions generally require both prepositions to be overtly present, but this is not always so, as in the following : (28) I do not know what he is doing [one day to the next] . T1')us the compound prepositions are somewhat like conjunctions, where we find both simple (and) and compound (both ... and) varieties. It so happens that the idiom family so productive in English is a subspeciesof compound prepositional phraseswhere the first preposition is not present. In French, on

E. .Williams I Remarkson lexical know/ed~E

the other hand, both prepositions must be present: depart en part (' limb from limb ' ) ( V. Deprez, p .c. ., T . Hoekstra has pointed out to me the existenceof heurepar heure and cote a cote) . The various subcasesof (24) do not have a common element of meaning. So, for example, cheekto cheekrefers to the pressingtogether of two cheeks, as in dancing; but minute to minute and day by day refers to a seriesof days in sequence . Even the instancessharing a common proposition do not have a completely shared element of meaning, as the two instanceswith to just cited show. Therefore, these forms are not compositional, despite clear patterns in the meaning. On the other hand, from the fact that they are so prevalent in one language, and absent altogether in another, we know that they are present as a group in some sense. It seemsunlikely that there is a parameter for this property alone; perhaps it follows from some other parameters, though it is hard to seehow. An alternative is that the structure in (25)-( 27) is induced from the data of the language. How could this happen? Supposethat at a certain point in the . course of acquisition, some number of forms with the shape NxPNx have been learned. As a class, they conform to the shapeof left headedcompound prepositional phrases, and so do not fall outside of the language altogether. On the other hand, their properties are not entirely explicable in terms of the general principles of the grammar; in particular, the use of bare singular count Ns as the objects of prepositions is not a general feature of prepositional phrases in English. So these remain idiomatic ; however, they are idiomatic as a class, not as individuals. Another family of idioms is illustrated in the following : (29) John hunts bear snares rabbit traps monkey

(a) *John huntsbook * 1 am goingto gradepaper , (b) *Johncountsmonkey (c) * hunt sleepyelephant What is special about this case is the use of the bare singular as object. Normally , this is not allowed for English count nouns, but is allowed here. So, we have an abstract idiom , of the following form :

E. Williams I Remarks on lexical knowledge

(30)

N] vp [V + HUNT + ANIMAL

The form in (30) is a special case of the general form of VP (and so, for ' ' ' ' example, is V N and not N V ), and thus conforms to our overall claim that abstract idioms are always instancesof more generalpatterns of the language. The limits of this idiom are somewhat roughly indicated by (b) and (c) - the ) shows verb must be a verb of hunting, and the object must be an animald a further restriction - not only are determiners excluded, but adjectives as ' well (unless"sleepy is a kind of elephant) . One might conclude from this the observation that this construction was ' ' in fact a lexically compound verb - to hunt monkey then would be syntactically intransitive . I doubt this, since English in general disallows compound verbs, and particularly disallows left -headed compound verbs; but even if the conjecture were correct , then the problem posed by these examples is not solved, but simply delivered to the lexicon, with fee still unpaid. I . assume that it is not at all predictable that English would have this pattern ; and in fact, inspection of (b) and (c) might lead one to not expect this pattern . I conclude therefore, that learning ( 30) entails generalizing over exampleslike those in (29), and the limits of the generalization must follow in some way from the actual mechanism of generalization. The existence of these idioms of intermediate abstractness argues that learning languagedoes not reduce to (a) learning parameter settings, and (b) learning the properties of particular lexical items. Rather, there are structures between these two extremes, what I have called abstract idioms , which .can only be learned as language-particular generalizations of particular forms.

3. Paradigms A paradigm is a multidimensional array of linguistic forms - for example, a verb conjugation, or a Latin noun declension. A paradigm is not just a convenient way to display linguistic information ; rather, it is a basic form of linguistic knowledge. It interests here becauseit is highly languageparticular , and at the same time, quite abstract in structure. Paradigmatic structure is also pervasive. Example (a) below is a slice of the Latin verbal conjugation :

E. Williams / Remarkson lexical knowledge

(31) Latin : (a) + - finite + - indicative + - passive + - perfective presjimperfjfut + - plural Ij2j3person

(b) amo amamus amas amatis amat amant

The Latin verbal conjugation is 8-dimensional, with the dimensions listed on the left. It is not possibleto say in a general way how many dimensionsa paradigm will have, nor how many points on a given dimension, nor what the dimensionswill ' mean' , or what syntactic or semanticcategoriesthe points on a dimension will be taken as signifying. There are severallevels of abstraction involved in paradigms. At the lowest level, we have word -paradigms, such as in (31b). At a slightly higher level of abstraction, we have paradigms in the traditional sense, roughly speaking, setsof endings: ( 32) - 0 - as -t

-mus - atis - ant

At a slightly greater level of abstraction, one might regard a paradigm as a set of rules which, when applied to a stem, derive a word paradigm, by, for example, adding endings. So there is a rule for forming the past tense, a rule for forming the third person present tense, etc. In fact, though, I believe that a paradigm is even more abstract than that - a paradigm is a patterning which is more abstract than any set of forms, any set of endings, or any set of rules for filling the slots in a paradigm. That the paradigm is a real object, and not the epiphenomenalproduct of various rules, is shown by the phenomenaof blocking, syncretism, suppletion, and paradigm defectiveness , as argued in Halle ( 1973) . To consider only the first of these: if there are two rules for filling a slot in a paradigm, ' only one may be used; thus, we have bit , not bited, and in general, only a single past tense form for a given verb , despite multiple ways to form past tenses. This reveals that there is a target slot to fill , which .is independent of the rules for filling it , and that slot is given by the paradigm .

E. Williams / Remarkson lexical knowledge

3.1. Extensiveness

We tend to think of paradigms as a means of displaying inflectional information about the parts of speech. But in fact, the notion of paradigm is much broader than this. To begin with , paradigms must include syntactic items, or phrases, as well as words. To seethis, we need only examine a slice of the Latin verb paradigm: (33)

active passive present amo amor perfect amavi amatus sum

Here the forms are all 1st singular, present-perfective crossed with activepassive. One comer of this square contains a phrase, amatus sum, while the other three comers contain words. This shows that phrases form an inextricable part of paradigmatic information . If we removed the perfective passive form , we would have destroyed the symmetry of the paradigm, which is otherwise perfectly symmetrical. We can see the same thing in an English paradigm, the comparative paradigm:

COMP SUPER (34) A long longer longest compact morecompact most compact better best good The rule is, if an adjective is mono- (or nearly) syllabic, then form the comparative with -er ; if not , then the comparative and superlative are formed ' phrasally. The existenceof this paradigm is what pennits us to speak of the ' of an comparative adjective, even though there are two ways of forming . comparatives Many languageslack any way at all to fonn the comparative; English has two ways, one morphological, the other syntactic. Paradigms include not just inflectional dimensions, but what have been called ' derivational' processes as well. I am sure that there is no distinction be~ween derivational and inflectional morphology , but if there is, then paradigms are found in both morphologies.

E. Williams I Remarkson lexical knowledge

3.2. The abstractness of paradigm structure

In general, there is not a one-to- one relation between the slots in a paradigm and the rules for creating the fonDSwhich fill the paradigm. In one circumstance, there are more slots than rules, and therefore, severalslots will be filled with the same form - this is the phenomenon of syncretism. The pattern of syncretism that is found reveals the method of assigningforms to slots. The method consistsof the following : . ( I ) Hierarchalize the dimensionsof the paradigm. Assign forms to nodes in the hierarchy. For example, the English verbal paradigm is 4-dimensional (personX number X present-past X finite), and there are 4 distinct forms. We may hierarchize the tree and assignthe forms to it as follows :

(35)

.

I pres m sg I

2 3*

m pi

I

2 3

V * = run

V' I

I

I

finite I

infinitive I

I

I

I

past.

perf*

infinitive

rfl

rfl

I 2 3

1 2 3

rfl

rfl

I 2 3

1 2 3

* past = ran

3 * = runs

1 2 3 rfl rfl

I 2 3

perf * = run

The tenninal nodes are the actual cells of the paradigm. The starred nodes are the nodes to which actual forms are assigned. By convention, a cell is filled by the nearestspecifiedform above it . The identity of the forms and the points in the tree at which they are mapped are given in (35) . The assignment ' shown is the most economical, as each form is assignedonce. ' We might call the starred nodes entry points' - these are the points at which concrete forms are specified. The tree along with the starred nodes we

E. Williams / Remarkson lexical knowledge

' ' might call the pattern of syncretism . This is a pattern which is independent of the rules for creating the forms in the pattern. It is a part of the formal structure of the paradigm. Strikingly , the pattern of syncretism holds generally in a category, not just for particular verbs, as we will see. If we look at the pattern of syncretism for a variety of English verbs, a startling pattern emerges:

V$ I

(36)

I I pres r +- , * sg IA 2 3 r +- ,

. pI I 2 3

finite ~

I

perf *

infinitive

r +- ,

r +- ,

pastS

I 2 3

I 2 3

r +- ,

~

m

IA 2 JA

I 2 3

1 2 3

I

r +- , I 2 J

Entry points : $ modals $* regular; go-went $*" be have, says, does - irregular in 3prsg: * haves Here, marked on the same tree, are the entry points for severalcategoriesof verbs, including irregular verbs. As can be seen, the sets of entry points form a nested set; the verb be shows the most distinctions, and consequently has the most entry points, but all other verbs, including all irregulars, have some subset of the entry points of be. It is far from obvious that such a relation should exist - if a verb is going to be irregular, why should it not be irregular in having a different pattern of syncretism, a different set of entry points ? But this does not happen, even irregular verbs respectthe pattern of the language as a whole. In fact, even suppletive verbs, the limiting case of irregularity , respectsthe pattern of syncretism; the verb go has went as its past tenseform . Things could have been different: went could have been the third past plural form , with goed (or something else) for all the other forms ; but then, go-went would have violated the language-wide pattern of syncretism.

E. Williams / Remarks on lexical knowledge

Hence, the pattern of syncretism is a quite abstract structure, standing above particular words, particular rules, particular suppletive relationships. We can see this further in the noun declensions of Latin . Latin has 5 declensions, each with its own set of endings (we ignore here the genitive and the vocative) : Latin

declension structure :

< 1> 1 .

" s ~

direct r- l -, nom

indirect r - L- ,

acc

< 2>

abl s ~

,

dat

abl

acc

direct r--l --w nom acc

dat

direct r=

nom

nom

indirect . r- L- ,

acc

< 3>

abl s ~

.

nom

I

s ~

: L - ,

acc

direct . r - L- , nom

dat

indirect . r - L- , abl

dat

acc

abl

dat

direct r-- L- , nom

dat

indirect . r - L- , abl

dat

-;Il I I p

I

direct indirect r-- L, r - Li nom acc abl dat

acc

direct r ~ nom

acc

indirect . r - L- , abl

N Ac D Ab

-a - am - ae -a

-IS . - IS

N Ac D Ab

-us -urn -0 -0

-i - os - is - is

etc.

indirect . r - L- ,

I 1 1 -pi

I

abl

acc

< 4>

dat

indirect r-- L- ,

direct r - L- ,

abl

-;I1 I I p

< 2n > ~ ____~ - -- , direct r -- L- ,

indirect . r~

I I I -pi

I

indirect . r-- - L- ,

direct , . J--, nom

-;Il 1 I p

I

dat

-aeas

E. WilliamsI Remarkson lexicalknowledge < 4>

s ,

direct r- l - ,

nom acc

~

-pII

1

,

1

indirect r-- L- , abl

dat

nom

acc

abl

dat

principle syncretisms: 1st decl: pl . indirect (-is) 2nd decl:pl . indirect (is), sg. indirect (-0) neuter: above + direct sg. (-urn). pl . direct (-a) 3rd decl. :pl . indirect(-ibus) 4th decl. :pl . direct (-es) pl . indirect(-ibus) neuter: sg. (direct = indirect) (-0), pl . indirect, pl . direct Here, the nominative and accusativehave beengrouped together as ' direct' , and the ablative and dative as ' indirect ' . The reason for this is that this reflects the patterns of syncretism: nominative and accusative fall together sometimes, and dative and ablative do as well. A striking thread that runs through the entire set of declensionsis the indirect plural syncretism. In the singular, there is an indirect entry point for 2nd and 4th neuter; the plural syncretism holds across the board. Importantly , this generalization is independent of the rules for forming the indirect forms, for in fact there are two different rules for that : in the first and second, -is is affixed; whereasin the 3rd and 5th, -;bus is affixed. Hence, the pattern is more abstract than the rules or affixes. I would suggest that a pattern is abstracted that applies to all the declensions, essentially the tree structure in -< I > . Other declensionswill make further syncretisms, but this one will hold for all. A general conclusion we may draw is that when there are multiple related paradigms, there will be one instantiated paradigm, and all others will have its syncretic structure, and perhaps some more. But no other related paradigm will have a contrary syncretic structure, making distinctions where that one does not. We will call that one paradigm the basic paradigm. For the Latin nominal declension, the first declension is the basic paradigm . For English verbs, the verb to be is the basic paradigm. L,et us now consider Latin verbs. Every finite Latin verb form has a different form , so there is no syncretism at this level. But at the level of stem, there is a paradigm structure with a limited number of entry points. Below is a chart of the stem forms for various classesof Latin verbs:

E. Williams I Remarkson lexical knowledge

.. (37) Latin verbstemsuppletion Vx I I . I

perf *

I

I

A imperfect

future .

Z'

.

I

sg ------ ---+- , , 1

partic

I

I pres I

I

I

r

' 3

I

plur - - ------" , 1

2'

3

Entry points for stem : * A' esse edo , * A' volo *A regularized volo * A do * regulars , including fero feri Tull latus x amo

The asterisked positions represent the entry points for the regular verbs of all four conjugations. This alone is striking , for again, there are different rules in the different conjugations for yielding the forms at these entry ' ' points . The first and second conjugation form the future by suffixing -b- , whereas the third suffixes nothing , but switches the conjugational class of the basic stem. Neverthess, that the future is an entry point is common to all the conjugations. The verb essehas the most entry points, and every other verb uses some subset of those entry points. Thus the conjugation of esse is the basic conjugation, to which all the others are related. ' As a final example, we consider Anderson s ( 1984) description of Georgian ' verb conjugation. The system is quite complex; Anderson s account uses blocks of rules, both conjunctively and disjunctively ordered. It is my contention that such a systemwill fail to capture the most abstract patterning of a paradigm, which, as we have seen, is generally independent of affixes or

E. Williams / Remarkson lexical knowledge

rules. Someindicationthat this is so can be derivedfrom the following table of affixes(or, to useAnderson's accountmore directly, of rules for adding affixes ) for presentand past: (38) present -en 3pl subj -t pi -s 3rd subjects

past -es 3pl subj -t pi 3rd subject

' ' The rules are.quite similar, and Anderson s remark This -es rule has the same status as the -en rule in this block , and also takes precedenceof the -s and the -1 rules ...' ( 1984: 8) shows I believe that an abstract structure is being replicated in different parts of the paradigm, thus underscoring the independence of that structure from actual rules or affixes. 3.3. Learning paradigm structure We ask at this point , why are these patte .rns of syncretism in language ? Why is a pattern of syncretism replicated accross different modes of realizing paradigm cells? I speculate that it is the acquisition of paradigm structurethat is respon

sible for this arrangement. Pinker has demonstrated how the blocking principle will give rise to the developmentof paradigm structure in the languagelearner. The basic idea is that wheneverthe languagelearner has been forced to posit two items to fill a single cell, he is then motivated to split the paradigm (really, to double it ) so as to avoid violating the blocking principle : (39)

D

The paradigm now has a new dimension, and a whole new set of cells; the ' ' language learner now must learn the significance of the dimension, and fill in the rest of the cells. I think that this algorithm for building paradigms, combined with the notion of paradigm structure I have just outlined, will predict some of the patterns we have observed.

E. WilliamsI Remarkson lexicalknow/edgE

There are two sorts of splitting than can take place - intraand interparadigm ' ' splitting . Given a present tense paradigm for some verb, see, for ' ' ' ' example, the recognition that saw , like see, can be used for the 1st person ' ' singular, leads to a postulation of a past tenseplane, in which to locate saw . This is intraparadigm splitting . The other kind of splitting might be called ' declensional' splitting : given a set of endings, say the endings for the 1st declensionin Latin (-a, -am, -De, -a, etc.), the recognition that the ending -us can signify nominative singular, just as -a does, triggers the splitting of the nouns into (at least) two declensions. The differencebetweenthis and the first caseis that in the first case, a given word will have forms in every cell of the new and old plane, but in the second, a word will have forms in only a single declension. We may neverthelessconsider the splitting to be formally the same in the two cases, and this is supported by the observation that syncretism patterns the samein the two cases. An example of a ' declensional' split is the comparative paradigm in English; the paradigm is a linear three-point paradigm: adjective, comparative , superlative. But there are two modes of forming members, as we have . seen- one for monosyllabics and Simple disyllabics, and another for everything .else. A telling point which showsthat we are dealing with a declensional here is that the same criterion that is used to determine whether the split ' ' comparative is A -er or more A is used to determine whether the superlative ' ' is A -est or most A . This criterion is not therefore a part of the rules themselves , but is rather a general criterion of membership in the two declensions; much as ' feminine' is a criterion for membershipin the Latin 1st declension. Now , supposethat a learner has learned a piece of paradigm structure, and has learned not only the labels for the dimensions, but has also learned the positioning of the entry points, which I have called the pattern of syncretism. Supposefurther that when the paradigm is split, that this abstract pattern of syncretismis replicated along with the cells themselves:

(40)

0 If this is done, then we will expect to seepatterns of syncretismrecurring. The following prediction is made: whatever paradigm is learned first will embody

E. Williams / Remarkson lexical knowledge

the most distinctions. This is becausecells which are designated identical by the pattern of syncretism of the first pattern will remain identical in latter versions. There may be fewer distinctions in later folds , but not more. Thus for example when the learner learns the first declension, with its indirect plural syncretism, and then learns that there is a second declension, that same indirect plural syncretism will show up in the second declension as well , copied as a part of the abstract structure of the first declension.

4. Learning .words 4.1. Learning morphemes

The notion of an abstract paradigm, the blocking principle , and paradigm splitting may account for how a paradigm is elaborated, but what accounts for the identification of a (potential) paradigm in the first place? One ordinarily thinks of verbal paradigms, re.alizing person, number, tense reference, etc., as the typical par~digm, but in fact languages have novel paradigms that it is unlikely coul.d be specifically anticipated in UG . Supposethat one component of the learner is a device that usesextra cycles in the child ' s computational life to track down statistical correlations among various properties of its thus far stored linguistic units. What the set of properties is will not detain us here. The child learning English, for example, will discover a correlation betweenwords ending in -yand adjectivehood:

, etc. (4Ia) fishy, lumpy, lucky, speedy

(41b) dainty

, pretty , happy , etc .

Many adjectives do not end in -y , and many words ending in -yare not adjectives, but the probability that a word is an adjective increasesonce one knows that it ends in -yo Note that this is true for a large class of adjectives where the -y does not serve as a suffix (the second group) . Even excluding the caseswhere -y is an affix, there is a correlation. A battery of such correlations could serve the next step: to identify m~rphemes, and assign them properties. In the case of -y - + A , there are sufficient casesto warrant postulating a suffix with the category adjective, and assign it wherever possible - that is, wherever an independent stem exists.

E. Williams I Remarkson lexical knowledge

Even where the analysis into morphemesdoes not hold , the information of the correlation has been noted; it seemsunlikely that this information is forgotten once the morpheme has been established. Another example in English of a no- go morpheme is the suffixtude ; most of the 50 or so membersdo not admit an analysis into morphemes:

(42) altitudeattitudeplatitude, And yet, this suffix so strongly marks nounhood that there is not a single verb or -adjective with this ending. The number of examples it takes to establish a correlation between two properties is quite small. For example, the English noun-deriving suffix -af has fewer than 20 instances; and yet, the correlation between suffix and category seemsfirmly established for all speakers, as well as the restriction that the stem must be stressfinal , a separatelearned correlation : (43) betrothal, avowal, approval, removal ... The confidenceof theseidentifications is surprising given the small number of casesinvolved. S6, there are two levels of analysis so far ; first , the identification of correlating properties, and second, the postulation of analysisthat arisesfrom this. Units will be postulated that account for the correlation. There is a further level of analysis, I believe. In somecases, it is detennined by the learner not just that two properties correlate, but that one of the properties is uniquely correlated with the other property . We might suppose that children are built to look for this especially. Severalexampleshave been examined in this paper. One is the idiom Ns : ' ' ' wear on the ' legs ; the property of meaning wear them on your legs is ' ' uniquely correlated with nouns that end in s . Another is the progressive, which is uniquely corellated with V + ing. I believe that the uniqueness has special salience. It is what we might consider the ' ideal' case, and so the first sought and most readily accepted. We know , for example, that the past tense forms of verbs in English are irregular, and some are in fact suppletive. Given this, we are not surprised so much to learn new verbs with irregular past tenses. However, I think we would resist entirely learning a new verb with a suppletive progressive, a progressivethat did not end in -ing.

E. Williams I Remarks on lexical knowledge

4.2. Learningsubsystems In orderto geta paradigmgoing, onemustnoticeparallelcorrelations . For , in English, thereis a Latinateparadigm, of the following shape: example (44) verb

adjective

de- verbal noun

de-adjectival noun

conclude conclusive conclusion conclusivity

The strength.of this paradigm is great enough that when a coined word meets the criterion for membership , it immediatelyparticipatesin the entire paradigm: (45) subduct: subductive: subduction: subductivity How do we know that we are dealing with a paradigm here, and not simply with several distinguishable morphological rules? Becausethe correlations amongst the rules striking : for example, 80%' of the words that take -ive to form adjectivestake -ion to form nouns; this is as compared to the baserate of 6% of nouns that take. -ion. Surely speakersnotice this sort of correlation , and use it to construct a paradigm like (44), which can then be exercisedon new forms. The criteria for membership in this paradigm are somewhat rough, but include at least the following :

(46) word = prefix-stem prefix = {trans, sub, super, de, dis, con, etc.} stem= CVC(C). The force of thesecriteria can be seenin the fact that 26% of the nouns that begin with trans- fonn nominalizations in -ion, against the base rate of 6% for nouns in general. Incidentally, this sort of thing is not limited to the English Latinate vocabulary. 39% of the verbs beginning with the causative prefix enencroach , endear, encase , ensnare) form their nominalization in -ment, as opposed to the background rate of lessthan 5% . ' ' We ordinarily think of -ion as an unproductive suffix, compared, say, to the suffixness . However, as we have seen, in the domain of certain classesof stems, -ion approachescomplete productivity , and even becomesapplicable to new forms. The other side of the coin is that -ness is actually not so

E. Williams I Remarkson lexical knowledge

productive over the entire set of stemsin English - it is of course only in the adjectives that it shines. So -nessand -ion are equally productive, we might say, over their own classes- the class for -nessis adjectives, and the class for -ion is Latinate verbs. Given this, we might wonder, why isn' t every affix completely productive within some arbitrarily drawn subclassof the lexicon, say the class of things to which it does attach? I think the answer is that subclasses cannot be arbitrary . The Latinate class is productive for -ion precisely becauseit can be identified independently of the occurencesof -ion: it can be identified as the claSsto which -ive attaches, or perhaps it can be identified in some way along the lines of (46) . In either case, we must attribute to the language learner the ability and the inclination to compare subclasses, and look for high matches. When a high match is found , then a dimension of a paradigm has been identified.

References . NLLT 2, 157-218. in morphoiogy . On representation . Anderson , S., 1984 . . On the definitionof word. Cambridge Disciullo, A.M. and E. Williams, 1986 , MA : MIT Press . . Semanticconditionson surfa(:e structure. MIT dissertation Fiengo, R., 1974 . Prolegomena to a theoryof word formation. LinguisticInquiry 4, 3- 16. Halle, M., 1973 . Patternsoflexicalization.In: T. Shopen(ed.), Languagetypologyand language Talmy, L., 1985 . . Cambridge , MA : CambridgeUniversityPress description . On the notions'lexicallyrelated'and ' headofa word' . LinguisticInquiry 12, Williams, E., 1981 245-274.

) 35- 77. North-Holland Lingua92 ( 1994

A preliminary analysis of causative verbs English*

in

BethLevina.and Maika RappaportHovavb . Department , Northwestern , 2016SheridanRoad, Evanston , IL 60208 of Linguistics University 4090, USA . Department , Bar lIan University , RaniatGan, Israel52900 of English This paper investigates the phenomena that come under the label ' causative alternation ' in English, as illustrated in the transitive and intransitive sentencepair Antonia broke the vase/ The vasebroke. Central to our analysis is a distinction between verbs which are inherently monadic and verbs which are inherently dyadic. Given this distinction , much of the relevant data is explained by distinguishing two processes that give rise.to causative alternation verbs. The first , and .by far more pervasiveprocess, forms lexical detransitive verbs from certain transitive verbs with a causativemeaning. The secondprocess, which is more restricted in its scope, results in the existence of causative transitive verbs related to some intransitive verbs. Finally , this study provides further insight into the semantic underpinnings of the Unaccusativity Hypothesis (Perlmutter 1978).

I . Introduction is particularly rich in verbs with both transitive and intransitive . English uses where the meaning of the transitive use of a verb V can be roughly ' ' paraphrased as cause to V- in transit ive . Such verbs are illustrated in ( 1) and ( 2), where the transitive (a) sentences might be paraphrased in terms of the ' ' intransitive ( b ) sentences; that is , as Antonia caused the vase to break and ' ' Pat caused the door to open . . Thiswork waspresented at theWorkshopon theAcquisitionof the Lexiconat the University in January, 1992 . We would like to thank the other workshopparticipantsand of Pennsylvania ' . This work was also , for their comments particularly Tony Kroch, the papers discussant in March at Hebrew , 1992; we are grateful to the audiencefor their University presented . This paperhasalso benefitedfrom the commentsof Mary Laughren comments , StevePinker, . We wouldlike to thankJohnWickbergfor helpingus BetsyRitter, andan anonymousreviewer wassupportedin part by NSF Grant BNSin on-line texts. This research find relevantexamples . 8919884 - 3841/94/$07.00 (Q 1994- ElsevierScienceB.V. All rightsreserved 0024 -9 SSDIOO24-3841( 93) EOO38

B. Levin, M. RappaportHovavI Causative verbsin English

( la ) ( I b) (2a) (2b)

Antonia broke the vase. The vase broke. Pat opened the door. The door opened.

We refer to this alternation as the causative alternation and to verbs with both uses as causative alternation verbs. 1

Such transitive/ intransitive pairs have receivedconsiderableattention from linguists . working in a variety of linguistic frameworks: analyses of this phenomenon have ranged from the primarily syntactic (e.g., Burzio 1986) to the primarily semantic (e.g., Fillmore 1968) and from the wholly lexical (e.g., Wasow 1977, Keyser and Roeper 1984) to the partly lexical (e.g., Borer 1991). 2 There is a sensein which, at least descriptively, the phenomenon is taken to be well-understood, and the same handful of acceptedfacts regarding this phenomenon are frequently cited. 3 Although the analysis of this phenomenon has been the focus of many studies (see, for example, the referencescited in Levin 1993), relatively little has been said about the phenomenonexcept in relation to the multifarious theoretical concernsit has been used to shed light on. It turns out , however, that many important questions about the phenomenon itself remain unanswered. And as long as the phenomenon itself is still not well-understood, a complete analysis of the alternation cannot be devel' causative which is 1 Besidesthe causativealternation. English also has a ' periphrastic . with the verbsmakeor have. asillustratedin Antoniamadethevasebreak. It hasoften expressed ' that entersinto the relationbetweenthe transitiveand beenremarkedthat the notion of 'cause than intransitiveusesof the alternatingverbsallowsfor a morerestrictedrangeof interpretations that found in English periphrasticcausatives . The type of causationassociatedwith the . . manipulative alternatingverbswhicharethesubjectof our studyis termeddirect(or. sometimes allow indirectas well as contact. or immediate . while Englishperiphrasticcausatives ) causation directcausation(Comrie1981 . among . Cruse1972 . NedjalkovandSilnitsky1973 . Shibatani1976 with alternatingverbsin Englishis not others). As we shallsee. the type of causativeexpressed availableto all verbs. contrastingwith the type of causativeexpressedby the periphrastic . In somelanguages both directand constructionin English.whichis generallyavailable causative indirectcausationaremorphologicallyencoded . the two typicallyinvolve . but in suchlanguages distinctmorphologicaldevices . We referto the kind of causativewearefocusingon in this paper of a languageand as the lexicalcausative . sinceit is usuallyformedusingthe lexicalresources showsthe hallmarksof a lexicalprocess( Wasow1977 ). 2 The causativealternationhas also attractedconsiderableattention outsidethe theoretical and child linguisticsliterature. It is the subjectof a numberof studiesin psycholinguistics ) for a reviewof this literature. languageacquisition; seePinker( 1989 3 Two notableexceptions are Pinker( 1989 ). ) and Haspelmath( 1993

verbsin English B. Levin, M. RappaportHovavI Causative

oped. For this reason, before presenting our analysis of the causativealternation , we provide a survey of certain properties of the causative alternation aimed at providing a contribution towards filling this gap in our understanding . This facet of our investigation focuseson two related questions: (i ) Is it possibleto delimit semantically the classof verbs which participate in the alternation? and (ii ) Do all examplesof the causative alternation as defined above represent instances of a single phenomenon? Answers to these questions will not only help us understand the causativealternation itself, but they should also deepenour understanding of the nature of lexical representation and its relation to syntactic structure. In this paper, we hope to show that the phenomenathat fall under the label 4causativealternation' are on the one hand lessidiosyncratic and on the other hand less uniform than is typically believed. We suggestthat much of the data we investigate is explained once we distinguish two processes that give rise to transitive and intransitive verb pairs.4 The first , and by far more ' pervasive process, is the one which forms lexical 4detransitive verbs from some transitive causative verbs. The second, which is more restricted in its sCQpeforms causative verbs from some intransitive verbs. With respect to intransitivity , we hope to provide further insight into the semantic underpinnings of the Unaccusativity Hypothesis, the hypothesis proposed by Perlmutter ( 1978) that the class of intransitive verbs consists of two subclasses, each associatedwith a distinct syntactic configuration . Finally , as in our previous work , we hope to show that if the relevant aspectsof meaning of a verb (or class of verbs) are properly identified, many of the apparent idiosyncratic properties of that verb (or verb class) fall into place.

: The propertiesof the alternation 2. Background that need to This section sets out the properties of the causative alternation that made observation the often be accounted for . We begin by repeating 4 That is, disregarding those verbs which participate in the unspecifiedobject alternation (e.g., the verb eat as in Terry ate her lunch/ Terry ate). In this paper, we also do not discussthe middle construction (e.g., Bread cuts easily) ; this construction involves intransitive uses of transitive verbs which at least on the surface show some similarity to the intransitive variant of the causativealternation. We consider the middle construction to be a distinct phenomenon since it differs from the intransitive variant of the causative alternation in interpretation and other properties (Keyser and Roeper 1984, Ruwet 1972, among others) ; however, seeHale and Keyser ( 1987) for an analysis that treats the intransitive variant of the causative alternation as a special caseof the middle construction.

B. Levin. M . Rappaport Hovav I Causativeverbsin English

there are many verbs in English which occur in the transitive/ intransitive pairs characteristic of this alternation. A preliminary list of such verbs is given below. ( 3) bake , bounce , blacken , break , close, cook , cool , dry , freeze, melt , move , open, roll, rotate, shatter , spin , thaw , thicken , whiten , widen , ...

Furthennore, the counterparts of these verbs in other languages occur in transitive/ intransitive pairs characterized by the same semantic relationship. In some.languages, as in English, the relation is not morphologically mediated ; see the Basque example in (4). 5 In other languages, the relation is morphologically mediated in some way, as in the French example in (5), where the reflexive clitic se is associatedwith the intransitive member of the pair . 6

(4a) Mirenekateaireki duo Miren-NORK door-NOR open3sNOR-have-3sNORK ' Miren

(4b)

(Sa) (Sb)

' opened the door. Atea ireki da. door-NOR open 3sNOR-be ' The door ' opened. Marie a ouvert la porte. ' Marie ' " opened the door. ' La porte s est ouverte. ' The door ' opened.

The existenceof this phenomenonin a wide range of languagessuggeststhat the causativealternation is not idiosyncratic to English. Studies of the causative alternation going at least as far back as Jespersen ( 1927) have suggested that this alternation is found with a semantically 5 In Basque the change jn transitivity is accompanied by a change in the auxiliary accompanying the verb. Simplifying somewhat, the transitive use selectsthe transitive auxiliary ukan ' have' while the intransitive use selectsthe intransitive auxiliary izan ' be' . Thus the difference in , auxiliary reflects general properties of Basqueand not properties of the alternation. The labels ' NO R' and ' NO R K ' are the traditional namesfor the casesassociatedwith the noun phrasesin . the examples. SeeLevin ( 1989) for more discussion. 6 For more on the morphological relationships between the verb forms in the transitive and intransitive variants of the causative alternation, see the discussion of Nedjalkov ( 1969) and Haspelmath ( 1993) at the end of section 4.

B. Levin. M . Rappaport Hovav / Causativeverbsin English

coherent classof verbs. In order to determine whether this suggestionreceives support, we can ask the following rather simplistic questions: (i ) Do all intransitive verbs have transitive counterparts with the appropriate paraphrase ? and (ii ) Do all transitive verbs with a causative meaning have intransitive counterparts with the appropriate meaning? We begin with a discussionof the first question. The following examplesshow that there are undoubtedly intransitive verbs which do not have transitive causativecounterparts. 7 (6a) The children played. * (6b) The parents played the children. (cf. The parents made the children play.) (7a) The actor spoke. (7b) * The director spoke the actor. (cf. The director made the actor speak.) Sa The audiencelaughed. ( ) * (Sb) The comedian laughed the audience. (cf. The comedian made the audiencelaugh.) These examples might suggestthat agentivity is the crucial factor and that agentive verbs do not participate in the alternation , while non-agentive verbs do. As it happens, both suggestionsare wrong. There are agentive verbs which do show the causativealternation, as in (9) and ( 10), and non-agentive verbs which do not , as in ( 11)-( 14) . (9a) (9b) ( l Oa) ( lOb) ( 11a) ( 11b) ( 12a) ( 12b)

The soldiers marched to the tents. The general marched the soldiers to the tents. The horsejumped over the fence. The rider jumped the horse over the fence. The cactus bloomed/ blossomed/flowered early. * The warm weather bloomedjblossomed/flowered the cactus early . The neglectedwound festered. * The heat and dirt festeredthe neglectedwound.

'7 Some English intransitive verbs without transitive causativecounterparts are used transitively in the resultative construction, but in this construction such verbs do not have the transitive causative meaning which the alternating verbs have. Consider the verb laugh in the resultative construction The crowd laughedthe actor off the stage. This construction does not mean that the crowd made the actor laugh, which would be the interpretation that would parallel the intended interpretation of (8b), but rather that the crowd laughed.

B. Levin Hovav/ Causative in English verbs , M. Rappaport

(13a) The jewels glittered/sparkled. (13b) * The queen glittered/sparkled the jewels. (14a) The stream burbled/ roared. (14b) * The rocks burbled/ roared the stream. The examples in ( 15) and ( 16) illustrate a further complication involving the transitive use of agentive verbs of manner of motion : the directional phraseswhich are optional in the intransitive use of theseverbs are obligatory in their transitive use. 8

(:I Sa) (:I Sb) (:I Sc) (:16a) (:16b) (:16c)

The soldiers marched (to the tents) . The general marched the soldiers to the tents. " The general marched the soldiers. The horsejumped (over the fence) . The rider jumped the horse over the fence. ' The rider jumped the horse.

.The behavior of the agentive verbs of manner of motion contrasts with that of non-agentive verbs of manner of motion , which, as shown in ( 17), do not require a directional phrase in either their transitive or intransitive use. ( 17a ) The ball bounced / rolled ( into the room ) . ( 17b ) The boys bounced / rolled the ball ( into the room ) . Although manifested

various

researchers

by agentive

have

verbs of manner

commented of motion

that

the alternation

is qualitatively

as

different

8 There may be some disagreement about whether the directional phrases are absolutely necessaryin the transitive causative usesof these verbs, particularly with a verb like jump . But even if thesephrasesneed not be expressedin certain circumstances, they are always understood in the transitive causative use. A speaker who accepts ( 16c) stin cannot give this sentencethe interpretation that the rider made the horse jump in place; rather this sentencereceives the interpretation involving the directional phrase: the rider made the horse jump over something. We look at this issuein more detail in section 8, where we also discusssome verbs of manner of motion that do not have causativeforms even in the presenceof directional phrases. Verbs of manner of motion are not unique in imposing the directional phraserequirement. The behavior of agentiveverbs of position parallels that of agentiveverbs of manner of motion in that they can have a causativevariant only in the presenceof a directional phrase, which givesthem an ' ' assume . position reading: MDUdestood the baby versusMDUdestood the baby on the table. We do not discussthis data here becausethis classof verbspresentsa number of complications. SeeLevin and Rappaport Hovav (to appear) for more discussionof verbs of position, as well as a discussion of a directional phrase requirementthat surfacesin certain circumstanceswith verbs of emission.

B. Levin. M. Rappaport Hovav/ Causative verbsin English

from that shown by verbs such as break (Cruse 1972, Hale and Keyser 1987, among others), we include this alternation among the data that needsto be accounted for since the general form of the alternation is the same: the transitive and intransitive usesof theseverbs differ with respectto the notion of ' cause'. Aside from Pinker ( 1989), previous researchershave taken the central property of these verbs to be that when intransitive they require agentive subjects, noting that this property appears to be carried over to the object of their transitive causative use. This work disregards the change in status of the. directional phrase. In contrast, we believe that the directional phrase is the key to explaining why theseverbs show the alternation. On the other hand, the contrast between ( 15)-( 16) and ( 17) suggeststhat , although there are agentive verbs which participate in the alternation as we have initially defined it , this alternation may be an instance of a different phenomenon , as we propose in section 8. Jespersen ( 1927) calls the classof causativealternation verbs the ' move and ' change verbs, becauseit includes a variety of verbs of motion and verbs of change of state. The list of alternating verbs presented in (3) can easily be divided into two subclasses along theselines: ( 18a) bake, blacken, break, close, cook, cool, dry, freeze, melt, open,

shatter, thaw, thicken, whiten, widen.

(ISb) bounce, move, roll, rotate, spin,

...

To the extent that verbs of motion involve a change of position (though not ' ' necessarilya translation through space), the set of move and change verbs ' ' might be given the unified characterization verbs of change. . This semanticcharacterization, although on the right track , is nevertheless inadequate. As we will see, change of state verbs do constitute the core of the class of intransitive verbs which alternate. However, to the extent that verbs of manner of motion like run are verbs of motion , it remains to be explained why they cannot appear in this alternation without directional phrases (in contrast to non-agentive manner of motion verbs like ro//). There are also verbs manifestingthe causativealternations which cannot be readily characterized as verbs of change. These include verbs of sound and light emission and verbs of position . ( 19a) The bell buzzed/rang. ( 19b) The postman buzzed/rang the bell. (20a) The flashlight beamed/shone.

B. Levin Hovav/ Causative verbsin English , M. Rappaport (20b) We beamed/shone the flashlight. (21a) Tony hung the laundry on the clothesline. (21b) The laundry hung on the clothesline. Furthennore, different classesof verbs participate in the alternation to varying degrees, a fact which itself is in need of an explanation. Verbs of change figure most prominently and most regularly in the alternation. Some, though by no means all , verbs of emission - whether they describe the emission. of sound, light , smell, or substance - can alternate. We have presented examples that show that among the verbs of light emission, the verbs beam and shine alternate, but the verbs glitter and sparkle do not. Similarly, among verbs of sound emission, the verbs buzz and ring can alternate, but the verbs burble and roar do not. Verbs of position allow the alternation rather freely. Not only hang, but also the verbs lean, sit , and stand allow the alternation, although a few verbs of position, including slouch and loom, do not. The behavior of slouchis particularly interesting since this verb is rather close in meaning to lean. (22a) (22b) (23a) (23b) (24a) (24b)

The ladder leaned against the wall. 1 was leaning the ladder against the wall. The surly youth slouchedagainst the wall. * 1 slouched the surly youth against the wall. The bear loomed over the sleepingchild. * The giant loomed the bear over the sleepingchild.

To summarize, our discussion so far has focused on the first question: whether all intransitive verbs have transitive counterparts with the paraphrase appropriate to the causative alternation. We have seenthat the intransitivity of a verb is not sufficient to ensureits participation in the alternation. Nor is the semantic notion 'change' sufficient, since although verbs of change are generally found in this alternation, intransitive verbs of other types differ in their behavior with respectto the alternation, even when they are membersof the same semantic class. Some other properties besides intransitivity and ' ' change must be found, and presumably the properties isolated will help to explain the behavior of the verbs in the different classes. We turn now to the second question: whether all transitive verbs whose ' ' meaning involves a notion of cause have related intransitive usesthat lack this notion . Again , the answer is ' no' . There are verbs which meet the semanticcriterion , but which do not have related intransitive uses. Examples

B. Levin, M . Rappaport Hovav I Causativeverbsin English

include the verb cut, which Hale and Keyser ( 1987) define as in (25), or kill , which has been defined - albeit controversially - as ' cause to die' (Lakoff 1970, McCawley 1968, among others).

in materialintegrity (25) cut: [x cause[y developlinearseparation ], by sharpedgecominginto contactwith latter] : (10 (HaleandKeyser1987 (26a) Thebakercut thebread. ' ' (26b) *Thebreadcut. (on theinterpretationThebreadcameto becut) . (27a) Theterroristkilledthepolitician * 27b The killed . ( ) politician Verbs close in meaning to cut such as slice or carve do not show the alternation ; neither do verbs related to kill , such as murder and assassinate.

(28a) Thechiefsliced / carved the turkey . . (28b) *Theturkeysliced /carved (29a) The terroristassassinated /murderedthe politician. (29b) *The politician assassinated /murdered. Moving to other domains, verbs of creation also do not participate in the alternation, although creation is sometimesdescribed as ' cause to exist' or ' causeto come to be' e. . ( g , Dowty 1979: 91) . (30a) (30b) (3Ia ) (3Ib )

Anita Brookner just wrote a new novel. * A new novel wrote. The contractor built another house. * Another house built .

Even more interesting is the fact that many morphologically complex English verbs formed with the suffixes -ize and -iff lack intransitive counterparts ' ' , 9 although these suffixes can be consideredto be causative affixes. (In fact, -iff comes from the Latin word for ' make/ do' .) Consider the examples below: .(32a) The farmer homogenized/pasteurizedthe milk .

/pasteurized (32b) *Themilk hom0genized

9 This property of -;ze is also noted by Keyser and Roeper ( 1984).

B. Levin, M . Rappaport Bovav I Causativeverbsin English

(33a) Carlahumidifiedher

. apartment * ( 33b) Her apartment humidified . However , some of these : type: counterparts of the appropriate

(34a) (34b) (35a) (35b)

complexverbs have intransitive

I solidifiedthe mixture. The mixturesolidified. The cook caramelizedthe sugar. The sugarcaramelized .

The behavior of -ify and -ize verbs contrasts strikingly with that of English verbs formed with the suffix -en. The suffix -en is also arguably a causative suffix, but verbs with this suffix appear to show the causative alternation rather more freely.

./ Thebana~as ripened. (36a) I ripenedthe bananas . (36b) I loosenedthe rope./ Theropeloosened ./ Thesaucethickened. (36c) John thickenedthe sauce (Lakoff 1968: (37a), (4a As part of a study that attempted to identify causative alternation verbs automatically in a machine-readable version of the Longman Dictionary of Contemporary English (Procter et al. 1978), Fontenelle and Vanandroye ( 1989) found that only 14 out of the 82 -ify verbs in that dictionary participated in the alternation , contrasting with 46 out of the 84 -en verbs. Unfortunately , they did not provide figures for -ize verbs, but an examination of the machine-readable version of a comparable dictionary, the Oxford AdvancedLearner's Dictionary (Hornby 1974), suggeststhat 14 out of the 78 -ize verbs listed as headwords in this dictionary participate in such pairs. 10 The contrasting behavior of these morphologically complex verbs formed with ' causative' suffixes again calls into question the existenceof a correlation betweenthe presenceof a notion of 'cause' in a verb' s meaning and a verb' s ability to show the alternation. It appears that neither intransitivity nor a ' ' meaninginvolving cause is sufficientto ensureparticipation in the alternation. 10 Thesmallnumber of -ify and-izeverbslistedin thesedictionaries canbeattributed to their intended function : thesedictionaries arerelativelysmalldictionaries for learners of designed . However examination of a moreextensive list of suchverbssuggests that . a preliminary English thenumber of alternating verbsreallyis notthathigh.

B. Levin, M. RappaportBovav/ Causative verbsin Englisll

Before presenting our own account of the alternation, we turn to an examination of an additional factor that intervenesin determining participation : selectionalrestrictions. The shared semantic relation betweenthe transitive and intransitive variants of causative alternation verbs has sometimes beendemonstrated via the existenceof selectionalrestrictions that are shared by the subject of the intransitive use and the object of the transitive use (Fillmore 1967, among others) . For example, only physical objects with certain characteristics can break, a property reflected in the set of possible objects of transitive break and possible subjectsof intransitive break. (37a) (37b) (38a) (38b)

Antonia broke the vase/ the glass/ the dish/ the radio. The vase/the glass/ the dish/ the radio broke. * Antonia broke the cloth the / paper/ the innocence. * The cloth the the innocence broke. / paper/

Assuming that selectional restrictions reflect the meaning of a verb, then this pattern of selectional restrictions reflects the fact that both variants share a common core of meaning. However, the extent to which selectionalrestrictions are shared acrosssuch pairs is not as great as is often thought . Smith ( 1970), whose study of the factors that determine participation in this alternation we come back to in section 3, points out that some intransitive verbs that typically do not enter into such alternations may enter into them for certain specific choices of subjectsof the intransitive use, as shown in the following examples.

(39a) (39b) (40a) (40b) (4Ia) (4Ib) (42a) (42b)

The baby burped. The nurseburpedthe baby. (Smith1970 : (36a The doctor burped. *The nurseburpedthe doctor. (Smith1970 : (36c The bell buzzed . The postmanbuzzedthe bell. The beesbuzzed . *The postmanbuzzedthe bees.

The exampleswith the verbs burp and buzz show that selectional restrictions need not be identical for the corresponding arguments in the transitive and intransitive uses. In theseexamples, the set of possibleobjects of the transitive use are a subsetof the set of possible subjectsof the intransitive use.

B. Levin. M . Rappaport Hovav I Causativeverbsin English

The lack of common selectional restrictions is even more pervasive. There are also instances of the reverse phenomenon: a verb which when used transitively is found with a set of objects that is larger than the set of subjects the sameverb allows when used intransitively . To take one example, consider the verb clear, a deadjectival verb that presumably means ' cause to become clear' . This verb is found in causativepairs as in (43), yet, although one can clear a table or a sidewalk, the table and sidewalkcan' t 'clear' , as shown in (44). (43a) (43b) (44a) (44b)

The wind cleared (up) the sky. The sky cleared (up). The men cleared the table/ the sidewalk. * The table the sidewalk cleared. /

A similar example involves the verb peel. This verb does not alternate at all in its most literal sense' remove peel from a fruit or a vegetable', although it can be used intransitively to describethe removal of skin - a ' peel' -like covering from a body part . The intransitive use ~f peel seemseven to be preferred in .the use in (46) . 11 (45a ') (45b) (46a) (46b)

1 peeledthe orange. * The orange peeled. 11peeledmy nose. My nose was peeling.

The examplesin (43)-(46) show that for some causativealternation verbs the selectional restrictions on the object of the transitive and the subject of the intransitive do not always coincide exactly. 12 The transitive object or the intransitive subject may show narrower restrictions. Presumably, for those choicesof arguments where these do not have transitive or intransitive uses, they lack them for the same reason that some verbs never have them. To summarize, an account of the causative alternation as defined in the broadest sensemust explain why someverbs show this alternation freely, why 11 This examplewas inspired by a similar examplein Rothemberg( 1974 ), a study of a comparablephenomenonin French, which includesmany examplesof divergingselectional restrictions . 12 It is possiblethat a closerexaminationof a widerangeof verbsmayshowthat theselectional restrictionsdo not coincidefor any verb. For instance and Ritter , as pointedout by Brousseau of the verb breakwherethe overlapis not complete : H~ brokehis ( 1991 ), thereare evensenses promisebut * His promisebroke.

B. Levin, M.

Hovav / Causativeverbsin English

someverbsdo not showit at all, and why someverbsshowit underrestricted circumstances . Finally, such an account must grapple with the issue of whetherthe data discussed in this sectionrepresenta unifiedphenomenon or not.

3. Towards an account of the alternation

The phenomenon we are dealing with falls under the general rubric of diathesis alternations, alternations which involve changes in the syntactic expression of the arguments of a predicator as well as in its adicity (the number of arguments it requires) . Therefore, our account of the alternation should ideally be embeddedin a generaltheory which accounts for the adicity of predicatesand the expressionof their arguments. We follow much current researchon the lexicon (Jackendoff 1990, Rappaport et al. 1988, Rappaport and Levin 1988, Hale and Keyser 1986, 1987; Pinker 1989, among others) in assuming that the properties of diathesis alternations in general can be predicted from the formulation of appropriate lexical semantic representations for the alternating verbs together with a set of linking rules (Carter 1988), rules which determine the syntactic expression of arguments of a predicate. As assumedin much of this literature, we postulate two levels of lexical representation. The first is a lexical semantic representation, a representation of the syntactically-relevant aspectsof verb meaning, which probably takes the form of a predicate decomposition. The second is a lexical syntactic representation or argument structure, which encodesthe syntactic expressionof the arguments of a verb. We assumethat the lexical syntactic . representationis derived from the lexical semantic representationby a set of linking rules. Since the causative alternation verbs can be found with either one or two arguments, a question which arises in the context of determining the lexical semantic representation of these verbs is whether they are basically one argument or two argument verbs. That is, are the causativeusesformed from the non-causative ones or vice versa? We assumethat the basic use of the verb will impose less stringent restrictions on its arguments than other uses since the conditions associatedwith a derived use might impose additional co~straints on the arguments of the verb. In those instanceswhere there are different selectionalrestrictions on the transitive and intransitive uses, the use with the looser selectional restrictions, if there is one, is the basic one. This means that given the data discussed above, for the verb buzz it is the

B. Levin, M. RappaportHovav / Causativeverbsin English

intransitive use that is basic, but for the verb peel it is the transitive use that is basic. The question to be asked in such instances is what aspect of verb meaning determines that peel is basically transitive, while buzz is basically intransitive. The selectionalrestriction criterion still leavesopen the issueof those verbs that appear to have similar selectionalrestrictions for both the transitive and intransitive uses, such as break or open. (Although given the comment in footnote 12, it is possible that for all verbs the selectional restrictions in one variant are looser than those in the other.) In order to isolate the meaning components which determine the (in )transitivity of a verb, we compare verbs like break that permit transitive and intransitive uses, to verbs such.as laugh, cry, or glitter that permit only intransitive uses (except perhaps under very special circumstances). (In section 6 we will address the issue of what distinguishes the break verbs from transitive verbs like cut and write, which have only transitive, but not intransitive , uses.) The question is what makes verbs like break on their intransitive use different from these other verbs? Here we draw on Smith' s ( 1970) insightful discussionof the semantic factors that playa part in determining which verbs that can be used intransitively have transitive causativeuses. Smith characterizesthe difference between those intransitive verbs which ' do and do not have transitive causativeusesby meansof a notion of external ' control . Verbs like break, Smith proposes, denote eventualitiesthat are under the control of some external causewhich typically brings such an eventuality about. Such intransitive verbs have transitive usesin which the external cause is expressedas subject. Verbs like laugh and cry do not have this property : ' ' the eventualities each one denotes ' cannot be externally controlled but can ' ' be control led only by the person engaging in it ; that is, control cannot be ' ' relinquished ( 1970: 107) . Smith takes the lack of a causativetransitive use for these verbs and other verbs such as shudder, blush, tremble, malinger, and hesitate, to be a reflection of the presenceof internal control ; we return in section 4 to the question of why verbs of internal control should have this property . (47a) (47b) (47c)

Mary shuddered. * The green monster shudderedMary . The green monster made Mary shudder.

: (35a-c (Smith1970

Similar distinctions have been recognized in other work on English (e.g ., Hale and Keyser 1987) and other languages (e.g ., Guerssel1986 on Berber ) .

B. Levin, M . Rappaport Hovav / Causativeverbsin English

For reasons which we explain below, we will not use Smith' s notion of 'control' for distinguishingamong intransitive verbs which do and do not have causativeuses. Rather, we use a related notion, distinguishing between' inter' ' ' nally and externally caused eventualities. With an intransitive verb denoting an internally causedeventuality, someproperty inherent to the argument of the verb is ' responsible' for bringing about the eventuality. On this approach, the concept of internal cause subsumesagency. For agentive verbs such as play , speak, or work, the inherent property responsiblefor the eventuality is the will or volition of the agent who performs the activity . However, an internally caused eventuality need not be agentive. For example, the verbs blush and tremble are not agentive, but they, nevertheless , can be consideredto denote internally causedeventualities, becausethese eventualities arise from internal properties of the arguments, typically an emotional reaction. 13 Verbs with an inanimate, clearly non-agentive subject, may also denote internally causedeventualitiesin the sensethat theseeventualitiesare possible becauseof inherent properties of their subjects. In particular , the notion of internal cause can be straightforwardly e~tended to encompass verbs of emission. It is an internal physical property of the argument of such a verb which brings about the eventuality denoted by the verb. This property is reflected in the strong restrictions that these verbs impose on possible subjects. For example, only very few things have the properties that are necessary to sparkle, and the same holds for other verbs of emission. Consistent with the classification of theseverbs as internally causedis the fact that , as mentioned in section 2, verbs of emission generally do not have causativecounterparts, as illustrated in (48) . (We return in section 7 to cases in which they do.) * (48a) The jeweller sparkled the diamond . ' * (48b ) Max glowed Jenny s face with excitement . * (48c) We buzzed the bee when we frightened it . * (48d ) The cook bubbled the stew. 13 The verbsshudder andshake , whichat first glanceappearto havethe samemeaning , present an interestingminimalpair. Only shake , and not shudder , showsa transitivecausativeuse. Our accountwouldsuggestthat shakingis externallycausedandshudderingis internallycaused . This . The proposalreceivessupportfrom an examinationof the thingsthat can shakeand shudder tw~ setsarenot co-extensive ; thesetof thingsthat shudderis to a largeextenta subsetof the set of thingsthat shake . Thingsthat shudderusuallycan be thoughtof as havinga ' self-controlled' include ; , animals , and, perhapsby forcedextension , the earth or a car. In body they people contrast, leaves , teacups , or furniture can only shake. This difference , like the internal versus externalcausedistinction, reflectsthe way weconceptualize the world.

B. Levin, M . Rappaport Bovav I Causativeverbsin English

Since verbs of emission pattern with other verbs without causative counterparts , we use the notion internal versusexternal causerather than the notion of control . It seems inappropriate to attribute control to the inanimate emitter argument of a verb of emission. In contrast to internally caused verbs, verbs which are externally caused inherently imply the existenceof an external cause with immediate control over bringing about the eventuality denoted by the verb: an agent, an instrument, a natural force, or a circumstance. Thus something breaks becauseof the existenceof some external cause; something does not break solely because of its own properties. Some of these verbs can be used intransitively without the expressionof an external cause, but , even when no cause is specified, our knowledge of the world tells us that the eventuality theseverbs denote could not have happenedwithout an external cause.

(49a) The vasebroke.j Antonia broke the vase. .jPat openedthe door. (49b) The door opened .We thus assume that the intransitive verbs which have transitive uses are externally caused, while those intransitive verbs which do not are internally caused. A closer look at the class of alternating verbs will bear out this suggestion. The change of state verbs that figure prominently among the alternating verbs describe changes in the physical shape or appearanceof some entity that can be brought about by an external cause, be it an agent, a natural force, or an instrument. Many of theseverbs are deadjectival; they are based on stage-level adjectives which describe properties of entities that can be caused to change, such as their physical characteristics, color , and temperature (Dixon 1982) . Some examples of such deadjectival verbs taken from Levin ( 1993) are given below in (SO); these verbs fall into two major groups, one in which the verbs are zero-related to adjectives, as in (a), and the second in which the verbs are formed from adjectivesthrough the use of the affix -en, as in (b) . ( 50a) brown , clear, clean, cool, crisp, dim , dirty , dry, dull , empty, even, finn , level, loose, mellow, muddy, narrow , open, pale, quiet, round , shut, slack, slim, slow, smooth, sober, sour, steady, tame, tan, tense, thin , warm, yellow, ... (SOb ) awaken, blacken, brighten, broaden, cheapen, coarsen, dampen, darken, deepen, fatten, flatten, freshen, gladden, harden, hasten,

B. Levin. M. RappaportHovav I Causativeverbsin English

heighten, lengthen, lessen, lighten, loosen, moisten, neaten, quicken, quieten, redden, ripen, roughen, sharpen, shorten, sicken, slacken, smarten, soften, steepen, stiffen, straighten, strengthen, sweeten, tauten, thicken, tighten, toughen, waken, weaken, whiten, widen, worsen, ... The verb smarten provides a particularly interesting illustration of the constraints on the adjectives that can serve as the base for verbs. Although the ' ' ' , intelligent and well and fashionably adjective smart has two senses ' dressed, the verb smarten is related to the second adjectival sense, reflecting the fact that it is typically only in this sensethat the adjective denotesa stagelevel property, and, hence, a property that might be causedto change. l4 That is, individual -level properties are typically not acquired as a result of an external cause, whereasstage-level properties are. The distinction between internally versus externally causedeventualities is not relevant only to verbs of change. l 5 It also explains the behavior of verbs of position with respect to the causative alternation. As noted above, verbs like hang, lean, sit , and stand have causative uses, but verbs like loom and slouchdo not. It seemsto us that the differencebetweeninternal and external cause is the key to their differing behavior. Looming and slouching are postures that are necessarilyinternal caused, unlike hanging, leaning, sitting, or standing, which are postures that can be brought about by an external cause. Many studiesassumethat the intransitive variant of a causativealternation verb is basic and the transitive variant derived. This assumption probably seemsjustified becausethe meaning of the transitive verb includes that of the 14

Betsy Ritter has pointed out to us the expressionSmarten up! Here the verb is related to the ' ' adjectival sense intelligent , but interestingly the verb is related to a stage-level use of the . It that this adjective appears adjective, like rnany other basically individual -level adjectives, can sornetirnesbe used as a stage-level predicate. Dowty ( 1979: 129, fn. 4) discusses other instancesin which deadjectival verbs lose sorneof the sensesof their baseadjective. For exarnple, he notes that although the adjective tough can rnean ' ' either ' difficult or ' resistant to tearing , the verb toughencannot rnean ' rnake difficult ' . We think that the stage-level versusindividual -level distinction could be responsiblefor at least sorneof the differencesin available sensesthat Dowty cites including the toughenexarnple. 15 There seernsto be a gap in the English verb inventory : there appear to be no agentive verbs of changeof state. We do not have an explanation for their absence. In fact, we are aware of very few internally causedverbs of changeof state at all , and those we have found , such asflower and blo.r.rom, and, in sorne languages, blush are non-agentive. We discuss this type of verb in Levin and Rappaport Hovav (to appear) .

B. Levin, M. RappaportBovavI Causative verbsin English

intransitive verb. For example, while transitive break means' causeto become broken' , intransitive break means ' become broken' . We suggest that this is not the case. A scrutiny of the range of verb classesin Levin ( 1993) reveals that there are no externally causedverbs without a transitive variant. That is, all externally causedverbs have a transitive causative use, but not all of them need have an intransitive use in which the external causeis unspecified(e.g., write or murder) . Given this generalization, we offer the following analysis: internally caused verbs are inherently monadic predicates, and externally causedverbs are inherently dyadic predicates, taking as arguments both the external cause and the passive participant , which is often referred to as the patient or theme. The adicity of the predicate is then a direct reflection of a semantic property of the verb. Externally caused verbs only detransitivize under specific circumstances; we discuss the circumstances that license the non-expressionof the causeargument of externally causedverbs in section 6. But it is important to stress that on our analysis externally causedverbs do not undergo a processof causativization - they are inherently causative- but rather a processof detransitivization. Since the majority of causativealternation verbs are externally caused, it is the processof detransitivization that is most pervasive in English. The following lexical semantic representationsfor the two types of verbs reflect the type of distinction we suggest. BROKEN ]] ( 51) break-transitive: [x CAUSE [y BECOME ( 52) laugh: [x LAUGH ] The representation for a verb like break is a complex lexical semantic ; it represents the meaning of representation involving the predicate CAUSE such verbs as involving two subevents, with each of the argumentsof the verb associated with a distinct subevent. The representation for an internally caused verb such as laugh does not involve the predicate CAUSE ; such verbs have only one subeventand are taken to be basically monadic. We discussthe rules that determine the syntactic expressionof the arguments in theselexical in the next section. However, it is clear that the semantic representations ' intransitive fo rm of break involves an operation which prevents the external cause from being projected to the lexical syntactic representation (the argument structure) . We do not discussthis operation in this paper, but seeLevin and Rappaport Hovav (to appear) for discussion. In light of the discussion above, certain facts about the formation of causativesacross languagescited by Nedjalkov ( 1969) are not surprising. In

verbsin English B. Levin, M. RappaportHovavI Causative

this study, which is basedon a survey of 60 languages, Nedjalkov looks at the morphological relation between the causative and non-causative uses of the verbs break and laugh (as well as two other verbs) in each of theselanguages. Nedjalkov points out that in the majority of his sample, the transitive causative form of the verb break is morphologically unmarked, with the intransitive form being identical to the transitive form ( 19 out of 60 languages ) or derived from this form (22 out of 60 languages). If verbs such as break are appropriatelycharacterizedas denoting externallycausedeventualities , then the monadic use is in some sensederived and indeed morphological marking has a function : it is needed to indicate the non-expression of the external cause. 16 Nedjalkov also considers the verb laugh. As a monadic verb which is internally caused, the verb laugh does not denote an eventuality that involves an external cause and can, therefore, be assumed to be basically a single argument verb. lit fact, Nedjalkov does not cite any languagesin which this verb has a transitive counterpart which is identical in form to or morpho logically less complex than the intransitive and which receives a causative 17 interpretation . Nedjalkov reports that in .54 of the 60 languagessurveyed, the causative form is morphologically more complex than the non-causative form ; seealso Hale and Keyser ( 1987) for discussionof some similar data. ' Haspelmath ( 1993) follows up on Nedjalkov s study and discusses verbs which tend not to show consistent patterns cross-linguistically . For example, verbs corresponding to English melt tend to be basically transitive in most languages, with the intransitive form being the derived form , but the opposite pattern is found in a few languages. It is likely that this variability arises becausethe meaning of a verb such as melt is consistent with classification as either internally or externally caused. 18 Pinker ( 1989) also points out that 16 Of course a is usedto create wherethereverse , therearesomelanguages typeof morphology 's in Nedjalkov . 9 of the60languages fromthemonadic sample predicate dyadiccausative predicate 's themorpheme . However showthisproperty , it is difficultto tell fromNedjalkov paperwhether in thelanguages in general of causatives breakisthatusedfor thederivation usedto formtransitive thatin the to hispapersuggests citesin theappendix concerned , althoughthedataNedjalkov . the causative of used to form the it is at least not the of laugh morpheme languages majority 17 Nedjalkov and wheretheverblaughhasbothtransitive ( 1969 ) notesthatin thoselanguages ' ' ' intransitive uses , this verbis likelyto mean'laughat ratherthan makelaugh whenused . transitively 18 Nedjalkov , burnandboil, findingthat theirbehavior ) alsolooksat two otherverbs (1969 thanthatof break wasmuchmorevariable formation across to causative withrespect languages withtheverbmelt, could observes . Thisvariation andlaugh . likethevariationthatHaspelmath . classification of theseverbs alsobeattributedto thevariable

B. Levin. M. RappaportBovavI Causative verbsin English

there are certain classesof verbs which denote eventualities which can be construed on cognitive grounds to be either internally or externally caused. It is precisely with respect to these kinds of verbs that cross-linguistic variation is expected. In fact, appropriately formulated linking rules should predict which kinds of verbs are most likely to exhibit cross-linguistic variation . The distinction betweeninternal and external causation seemsto do just this, and we take it to corroborate our approach.

4. Formulatingthe linking rules Although the number of arguments that a verb requires in its lexical semantic representationis determined by whether it describesan internally or an externally causedeventuality, we must also posit linking rules that ensure that these arguments have the appropriate syntactic expression. As we describein Rappaport et al. ( 1988), we seelinking rules as creating the lexical syntactic representation or argument structure of a verb from its lexical ' semantic representation. As we also outline in that paper, a verb s argument structure in turn relatively straightforwardly determines the d-structure syntactic configuration that the verb is found in. We propose that the following linking rules are among those that determine the lexical syntactic representation of a verb: (53) Immediate CauseLinking Rule: The argument of a verb that denotes the immediate cause of the eventuality denoted by that verb is its external argument. ( 54) Directed ChangeLinking Rule: The argument of a verb that denotes an entity undergoing a directed changedenoted by the verb is its direct internal argument. We have stated theselinking rules in tenDs of the argument structure notions ' ' ' ' external argument and direct internal argument ; these argument structure ' ' positions are then projected into syntax as the d-structure grammatical relations of subject and object, respectively. In the next section we explain why we have stated these rules in tenDS of argument structure notions that correspond most closely to d-structure grammatical relations rather than to sstructure grammatical relations. In this section we discuss the linking rules and their application to the data we have discussed.

B. Levin, M . Rappaport Hovav I Causativeverbsin

Englisll

The Immediate Cause Linking Rule is intended to apply to the argument that causesthe eventuality denoted by both internally and externally caused verbs. First , we consider internally caused verbs such as laugh or play . The verb laugh' s single argument is the cause of the eventuality that the verb denotes and will be expressedas an external argument as a consequenceof the Immediate Cause Linking Rule. This rule will also explain why laugh and other internally caused verbs do not have a simple transitive causative use. Such a use would involve the introduction of an additional cause, external to the eventuality denoted by the verb. Such an external causewould have to be expressedas the external argument due to the Immediate Cause Linking Rule. The external causewould thus compete with the verb' s own argument for external argument. As a verb has only a single external argument, such causative useswould be ruled out. On this account, the lack of a causative variant for an internally caused verb receivesan explanation in terms of the properties of argument structure; this explanation only indirectly appeals to the semanticsof the verbs involved. 19 The only way to introduce an external causeis to expressthe causative use of internally caused verbs periphrastically. And across languages, verbs like laugh, cry, speak or play are causativized through the use of a causative affix or verb. ( 55a) (55b) (56a) (56b) ( 57a) ( 57b) (58a) ( 58b)

* The clown

laughed me. The clown made me laugh. * The bad news cried me. The bad news made me cry . * The director spoke the actor. The director made the actor speak. * The parents played the children. The parents made the children play .

Following Baker ( 1988), Marantz ( 1984), S. Rosen ( 1989), and others, we assumethat the causative morpheme or verb comes with its own argument structure, so that the Immediate Cause Linking Rule does not have to 19 Pinker ( 1989 ) points out that internally causedverbs are not expectedto have causativeuses becausethe eventuality they denote cannot have an external causewhich is at the same time an immediate cause; that is, such eventualities cannot be construed as being directly caused. Although this property is probably implicated in the non- causativizability of such verbs, the existenceof ii1;ternally causedverbs which do causativizeunder certain syntactic conditions, such as those discussedin section 8, suggeststhat syntactic factors enter into the explanation as well.

verbsin English B. Levin. M. RappaportBovav/ Causative

associate two arguments from a single argument structure with the same argument structure position . General principles will determine that in languages with causativeaffixes or verbs the introduced causewill be first in line for being chosen as the external argument in its clause. The Directed Change Linking Rule is similar in spirit to familiar linking rules which associatea patient or a theme (or an equivalent notion ) with the direct object grammatical function (Anderson 1977, Fillmore 1968, Marantz 1984, among others) . Our formulation is meant to give specific semantic content to the notions ' patient' and ' theme' . The Directed Change Linking Rule is meant to apply to verbs of change of state and verbs of change of location. This second class includes verbs of directed motion such as come, go, rise, andfa // but NOTverbs of manner of motion such as roll , run, jog , and bounce. This difference follows because, although the action denoted by a verb of manner of motion inherently involves a kind of change, it is not a directed change. Tenny suggeststhat there are certain kinds of changeswhich can be characterized ' ... as a change in a single parameter or a change on a scale' ( 1987: 189) . We call such changes' directed changes'. Tenny. arguesthat an argument denoting an entity which is specifiedto undergo such a changeis realized in the syntax as a direct object. This property distinguishes a change of state verb like dry from both agentiveand non-agentive verbs of manner of motion like walk and roll . The verb dry specifiesa change characterizable in tenDS of a single parameter, dryness, whereas walk and roll do not specify such a change. In contrast, for verbs of directed motion there IS a directed 2o change: a movement in a particular direction. The argument of a nonagentive manner of motion verb such as roll will be a direct internal argument, as we will see, but this linking will be effected by another linking rule. The justification for this will be given in section 7. The linking rules we have formulated also ensure that when a verb like break is used transitively, the external cause will be the external argument, and the patient, since it undergoes a specified change, will be the direct internal argument. When a verb like break is used intransitively with only the patient argument, the Directed Change Linking Rule will apply, and this 10 As fonnulatedherethe DirectedChangeLinking Rule, unlikesomeother proposedlinking rulesthat aresimilarin scope , suchas widenor cool. , will applyto certainatelicverbsof change We arguethat this propertyis desirablein Levin and RappaportHovav (to appear ), wherewe Rule with other a more detailed of the Directed linking comparison ChangeLinking provide rules, especiallythosewhichmakereferenceto conceptssuchas telicity. We alsocompareour ' ) which makesuseof the rather similar notion of approachwith one such as Dowty s ( 1991 ' incrementaltheme'.

B. Levin, M . Rappaport Hovav I Causativeverbsin English

. argument will be the direct internal argument. Since these verbs have sstructure subjects when intransitive , this argument must assumethe subject grammatical relation at s-structure, presumably as a consequenceof independent syntactic principles. The typical GB -framework account of the expression of the arguments of such verbs makes reference to the Case Filter , Durzio' s Generalization, and the Extended Projection Principle (e.g., Burzio 1986); we do not go into details here. 21 Together the Immediate Cause and Directed Change Linking Rules can be used to predict whether the membersof the verb classesthat we discussedin section2 will have causativeusesor not. Verbs of changeof state are inherently dyadic verbs, so they will always have causativeuses, although not as a result of causativization; in section 6 we elaborate on the circumstancesin which these verbs can have monadic ' detransitive' uses. Internally caused verbs are not expectedto have causativeuses, explaining the behavior we observedfor verbs of emission; we discuss in section 7 why some verbs of emission neverthelessdo have causatives. Agentive verbs of manner of motion , as internally causedverbs, are also not expected~o have causativeuses. As seenin section 2 these verbs do not typically have causative uses in isolation; we discussin section 8 why theseverbs may have causativeusesin the presenceof a directional phrase. We attribute the mixed behavior of verbs of position to a split in the class: some of theseverbs are internally causedand others are not , and the internally causedverbs are not expectedto have a causativeuse. These linking rules leave open the question of what happens with an argument that falls under neither of the linking rules introduced in this section. Here we make the assumption, which we justify in Levin and Rappaport Hovav (to appear), that an argument that is not linked by one of these two linking rules will be a direct internal argument rather than an external argument.22 ( 59) Default Linking Rule: An argument of a verb that does not fall under the scopeof the other linking rules is its direct internal argument. 21 SeeBresnanand Zaenen( 1990 ) for an accountwithin LFG. 22 In Levinand RappaportHovav(to andexistence ), wearguethat verbsof appearance appear requiretheir own linking rule. We formulatean additionalrule that appliesto theseverbs, linking the argumentwhoseexistence is asserted to directinternalargument . The DefaultLinking Ruleis supposedto apply to thoseargumentsthat do not fall under the scopeof any linking rule, whetherit is this additionallinking rule or the previouslyfonnulatedrules, the ImmediateCause and DirectedChangeLinking Rules.

verbsin English B. Levin, M. RappaportHovavI Causative

The Default Linking Rule will apply to the theme (located) argument of transitive sit , stand and other externally caused verbs of position, since this argument neither causes the eventuality denoted by the verb nor does it 23 undergo a specified change. We return to the Default Linking Rule in sections7 and 8, where we illustrate its applicability more fully .

5. The Un8ccaativeHypothesis We have introduced a distinction between internally and externally caused eventualities. As we discussin greaterdetail in Levin and Rappaport Hovav (to appear), the two classesof intransitive verbs describedhere are preciselythose that are implicated in phenomenawhich fall under the rubric of the Unaccusative Hypothesis ( perlmutter 1978). Since the linking rules do not make referenceto the adicity of a predicate, they predict that a verb like break, even when the external causeis not expressed , still takes a direct internal argument, as can be seenfrom the application of the linking rules to the representations we introduced above for the two classesof intransitive verbs. With intransitive break, only the y variable in (51) is expressed ; as the argument undergoing a directed change, it will be a direct internal argument, and hencead -structure object in the syntax. In contrast, the x variable in (52), as an immediate cause, will be an external argument, and hencea d-structure subjectin the syntax. The syntactic expressionof the argumentsof thesetwo verbs is given below.

-intransitive : - [vpV NP] (60a) break NP : 60b ( ) laugh [vpV] Given the definitions of unaccusativeverbs as verbs taking a single direct internal argument and unergative verbs as verbs taking a single external argument, the linking rules proposed in section 4 will receivesupport if there is evidencethat internally caused verbs are unergative and externally caused . tive diagnostics . tive. We review two uoa(x:usa verbs, when monadic, are una(x:usa ; see Levin and discussion this claim for further be used to that can ; support Rappaport Hovav (to appear) . 23 Wedo notdiscuss of of theapplication theseverbsfurtherin thispapersincea full account in their the linkingrulesto theseverbswouldrequireus to introducecertaincomplications Hovav(to appear in LevinandRappaport . Wediscuss thesecomplexities behavior , ). However verbsof position caused thattheexternally wewouldliketo pointout thatouraccount suggests . transitive shouldbebasically

B. Levin, M . Rappaport Hovav / Causativeverbsin English

Work on the Unaccusative Hypothesis has establishedthat the resultative construction can be used as an unaccusativediagnostic (Hoekstra 1984, Levin and Rappaport Hovav (to appear), Simpson 1983, among others) . Although both unaccusativeand unergative verbs are found in this construction, they pattern differently due to an interaction of verb type with a syntactic constraint requiring the resultative phrase to be predicated of ad -structure object. What matters for our purposes is that when an unaccusativeverb is found in the resultative construction, the resultative phrase is predicated directly of its surface subject, as in (61), but a resultative phrase cannot be predicateddirectly of the surface subject of an unergative verbs, as in (62a). A resultative phrase may only be predicated of the subject of an unergative verb ' ' through the mediation of what Simpson ( 1983) calls a fake reflexive object, as in (62b). Alternatively , a resultative phrase may be predicated of a nonsubcategorizedobject found with an unergative verb, as in (63), an option not available to unaccusativeverbs, as shown in (64) . 24

(61) (62a) (62b) (63) (64)

The bag broke open. *We yelledhoarse. We yelledourselveshoarse. The dog barkedthemawake. *The bag broke the groceriesall over the floor.

Thus the different patterns of the resultative construction correlate with the status of a verb as unaccusativeor unergative: a monadic verb which allows a resultative phrase to be predicated directly of its subject is unaccusative, while a monadic verb which allows such a phrase to be predicated of an object - either a ' fake' reflexive or a non-subcategorizedobject - is unergative. The closely related X ' s way construction is also an unaccusativediagnostic. This construction, in which a resultative phrase is predicated of the subject of a verb through the use of the phrase ' X ' s way' in object position, is found with unergative. verbs, but not with unaccusative verbs (Jackendoff 1990, Marantz 1992) .

(65) They workedtheir way to the top. (66) *The Arctic explorersfroze their way to fame. 24 SeeLevin and Rappaport Hovav (to appear) for an explanation of the differential behavior of the two classesof verbs in the resultative construction, and Hoekstra ( 1992) f~r an alternative account.

B. Levin, M . Rappaport Hovav / Causativeverbsin English

The resultative and X 's way constructions distinguish internally caused verbs from externally causedverbs as predicted. An examination of the set of tokens of theseconstructions we have collected over the last few years shows that internally causedverbs like laugh, play , and work are regularly found in the X 's way construction and the unergative resultative pattern, while instances of monadic externally caused verbs are attested only in the unaccusative resultative pattern. The behavior of verbs of emission in the resultative construction is of particular interest since the classification of these verbs has been the subject of controversy. Perlmutter ( 1978) originally classifiedtheseverbs as unaccusative , but this classification has been challenged (see for example Zaenen 1993) . We have classified these verbs as internally caused verbs, and hence, we predict that they will pattern with unergative verbs in general, and in the resultative and X 's way constructions in particular . The examples below verify this prediction . 25

(67a) The beaconsflared the newsthrough the land. (HendersonI 92;

cited in K.-G. Lindkvist, A Comprehensive Study of Conceptions of , 1976, p. 89, sec. Locality, Almqvist & Wiksell, Stockholm, Sweden 233, 4) ' (67b) ... you cant just let the thing ring itself to death, can you? (Observer ; TraceThat Call No More!, New York Times, March 8, 1989 ) (67c) The very word was like a bell that tolled me back to childhood summers... (Hers; Child' s Play, Women's Sway, New York Times, July 17, 1988) (67d) Then he watchedas it gurgledits way into a whiskeytumbler. (M. Grimes, TheFive Bellsand Bladestone , Little, Brown, Boston, 1987, . 200 ) p (67e) To counterthe uneasethat was oozing its way betweenthem. (P. Chute, Castine , Doubleday, GardenCity, NY , 1987, p. 214) In Levin and Rappaport Hovav (to appear) we look at a wide range of tests and find that they corroborate the results of the two tests that we have discussedin this section, further supporting the linking rules formulated in 25 Given their unergative classification. we would not expect these verbs to pattern as unaccusativeverbs with respectto the resultative construction. In actual fact. some of theseverbs are found in the unaccusativeresultative construction. but as we discussin Levin and Rappaport Hovav (to appear) their unaccusative behavior correlates with a shift in meaning. with the additional meaning being one that is typically associatedwith an unaccusativeclassification.

B. Levin, M . Rappaport Hovav I Causativeverbsin English

section 4. In that work , we also show that there are some verbs which are compatible with both internal and external causation. Theseverbs include the non-agentive verbs of manner of motion such as roll and bounce and the verbs of position . As we show in that work , with such verbs external causation is correlated with unaccusative status, while internal causation is correlated with unergative status.

6. When can an externally causedverb detransitivize? The next question we addressis the following : if externally causedeventualities are basically dyadic, when can verbs denoting such eventualities turn up as intransitive , and why is this possibility open to some verbs only for ' certain choices of arguments? Again we draw on the insights in Smith s ( 1970) paper to reach an understanding of this phenomenon. In trying to identify the factors that permit detransitivization (that is, the non-expressionof the external cause), it is useful to look at the characteristics . of the subjects of externally caused verbs. Among the verbs that never detransitivize are verbs that require an animate intentional and volitional agent as subject, such as the verbs murder and assassinateor the verbs of creation write and build. / murdered the candidate. (68) The terrorist assassinated wrote a letter to the editor of the local newspaper. 69a ( ) Tony (69b) That architect also built the new high school. Smith proposesthat the verbs of change that may be used intransitively are ' precisely those in which the change can come about independently in the ' sensethat it can occur without an external agent ( 1970: 102) . She identifies independenceand external control - the notion which we have subsumed under our notion external cause - as the two features which characterize verbs of change. Independence allows for the possibility of intransitive counterparts, and external control or causation allows for the possibility of a ' transitive causativeuse. Smith s observation can also be recast as follows : the transitive verbs that detransitivize are those in which the eventuality can happen spontaneously without the volitional intervention of an agent. We believe that this property is reflected in the ability of such verbs to allow natural forces or causes, as well as agents or instruments, as external. causes, and, hence, as subjects, as illustrated with the alternating verb break.

B. Levin, M. RappaportHovav/ Causative verbsin English ( 70) The vandals / the rocks / the stonn broke the windows .

Verbs such as break contrast with verbs such as murder, assassinate , write. and build. These four verbs, as well as any other verbs which, like them, denote eventualities that require the participation of a volitional agent and do not admit natural force subjects, will not detransitivize, despite the fact that their meaningsinvolve a notion of ' cause'.

(7]a) (7] b) (7]c)

*The candidateassassinated / murdered. *The letter wrote. *The housebuilt.

In fact , these four verbs are among those that require an agent in the strongest sense: they do not even allow an instrument as subject .

thecandidate . (72a) *Theknifeassassinated /murdered (72b) *Thepenwrotetheletter. cranebuilt thehouse . (72c) 11The A verb like cut shows that the set of verbs that do not detransitivize is not limited to verbs which restrict their subjects to volitional agents. Although this .verb does not typically allow natural force subjects, it does allow instruments in addition to agents as subjects. 26

(73) The baker/that knife cut the bread. Sentence(74), however, cannot be used to describe the bringing about of a separation in the material integrity of some object. ' ' (74) * The bread cut. (on the interpretation The bread came to be cut ) The behavior of a verb like cut can receive an explanation. Its meaning includes a specification of the means involved in bringing the action it denotes about, which in turn implies the existence of a volitional agent. Specifically, the very meaning of the verb cut implies the existenceof a sharp instrument that must be used by a volitional agent to bring about the change 26 SeeBrousseauand Ritter ( 1991) for further discussionof the circumstancesthat allow verbs to take both instruments and agents as subjects.

B. Levin, M . Rappaport Bovav / Causativeverbs"in English

of state denoted by the verb. If the samechange of state were to come about without the use of a sharp instrument, then it could not be said to have come about through cutting , showing that the choice of instrument makes cutting cutting . Perhaps the same considerations can explain the behavior of the verb remove, which does not have an intransitive form . Its non-existence might seemsomewhat surprising since at a first approximation this verb' s meaning ' ' might be paraphrased as cause to become not at some location . A closer look at the verb remove's meaning reveals that the eventuality it denotes is brought about by a volitional agent, as shown by the oddnessof the examples in (75), which have inanimate non-volitional subjects. (75a) ??The wind removed the clouds from the sky. (cf. The wind cleared the clouds from the sky.) (75b) ??The water removed the sand from the rocks. (cf. The water washed the sand from the rocks.) We assume that the same factors explain why most morphologically complex verbs formed with the suffixes -ize and -iff cannot typically detransitivize, as the data repeatedhere illustrates.

(76a) The fanner homogenized/pasteurizedthe milk . (76b) * The milk homogenized/pasteurized. (77a) Carla humidified her apartment. (77b) * Her apartment humidified. Most of theseverbs cannot detransitivize, we propose, becausethey describe eventualities such as being pasteurized or homogenized that cannot come about spontaneouslywithout the external intervention of an agent. It appears to be preciselythose -ify and -ize verbs which allow for this possibility that do detransitivize. (78a) I solidified the mixture./ The mixture solidified. (78b) The cook caramelizedthe sugar./ The sugar caramelized. Again , the -iff and -ize verbs that do and do not permit intransitive uses contrast ,with respectto the range of subjectsthey permit when transitive. The verbs that resist detransitivization show a narrower range of subjects when transitive ; specifically, they appear to exclude natural force subjects.

B. Levin, M . Rappapor ' Hollall I CausativelIerbsin English

(79a) * The rainy weather humidified the apartment. (79b) The intense heat caramelizedthe sugar. If we look more closely at some of the alternating verbs in -iff and -ize listed in (80), we see that many of these verbs, such as intensify or equalize, are deadjectival and are very similar in meaning to the previously mentioned alternating deadjectival verbs in (50) . (80a) acetify, acidify , alkalify , calcify, carbonify , emulsify, gasify, intensify, lignify , liquefy, nitrify , ossify, petrify , putrefy, silicify, solidify, stratify, vitrify , ... (80b) caramelize, carbonize, crystallize, decentralize, demagnetize, depressurize , destabilize, equalize, fossilize, gelatinize, glutenize, harmonize, ionize, magnetize, neutralize, oxidize, polarize, pulverize, regularize, staQilize, vaporize, ... Other alternating -iff and -ize verbs are denominal; their meaning may be ' paraphrasedroughly as causeto turn into the substancenamed by the noun that the verb is based on' : caramel for caramelize, powder for pulverize, gas for gasify, and so on. The non-alternating -iff and -ize verbs also include some denominal verbs whose stems are nouns that name substances: zincify , carbonize, and lodize. But -what is interesting is that the meaning of these non-alternating verbs is different from that of the alternating verbs: it could be paraphrased as ' ' ' process or treat using the substance rather than cause to turn into the ' substance. We suggest that due to this difference in meaning, these verbs require an agent and hence do not detransitivize. In fact, if zincify meant ' turn to zinc' rather than ' ' processwith zinc , we would predict that the verb could alternate, and our own intuitions , as well as those of others we have consulted, is that it would. A preliminary examination of a wider range of non-alternating -iff and -ize verbs suggeststhat many describe changesthat involve a particular type of processing or treatment, as with the previously cited verbs homogenize and pasteurize or as with the verbs sterilize or vulcanize. Other non-alternating verbs involve changesof state that only come about through the active intervention of an agent, such as legalizeor sanctify. The constraint on detransitivization also explains why some verbs have intransitive usesonly for certain choicesof patient : it is only for thesechoices of patient that the change can come about without the intervention of an agent. For instance, in section 2 we noted the following contrasts involving the verb clear:

inEnglish verbs Hovav B. Levin , M. Rappaport I Causative (8Ia) (81b) (82a) (82b)

The menclearedthe table/ the sidewalk. *The table/ the sidewalkcleared. The wind cleared(up) the sky. The sky cleared(up).

Our knowledge of the world tell us that tables and sidewalks are things that are cleared (of dishes and snow, respectively) through the intervention of an animate agent. The sky, however, can clear through the intervention of natural forces, such as the wind . Thus the difference in the possibility of intransitive counterparts. Similarly, peeling - causing an entity to lose an outer layer is typically brought about through the actions of a volitional agent, particularly if a fruit or vegetable is involved. However, there are certain entities that lose their outer layers due to natural causesrather than through the action of an agent, and in these instancesthe verb peel can be used intransitively , as in the case of the loss of skin from a person, as illustrated in (84) .

(83a) (83b) (84a) (84b)

I peeledthe orange. . The orangepeeled . 'II peeledmy nose. My nosewaspeeling.

The verblengthencanbe usedto presentanothercontrastof the sametype: (85a) (85b) (86a) (86b)

The dressmaker lengthenedthe skirt. *The skirt lengthened . The mad scientistlengthenedthe days. . The dayslengthened

Typically skirts are only lengthenedthrough the intervention of an agent, and 27 hence the verb lengthen as applied to skirts is not found intransitively . Days, on the other hand, become longer as the earth progresses through a certain part of its orbit around the sun, something that happens without the intervention of an outside agent. And lengthenas applied to days is typically used intransitively, although in a science fiction context where artificial Z7 Of course, it is possible to construct contexts in which a skirt might be lengthenedby being washed. As Mary Laughren has pointed out to us, the intransitive use should be possible in such circumstances.

B. Levin, M . Rappaport Hovav / Causativeverbsin English

manipulation of the length of days is possible, transitive usesare also found , as in (86a) . These examplesshow yet again that detransitivization is possible precisely where an externally causedeventuality can come about without the intervention of an agent. In this sense, detransitivization is a productive process, since it appearsto be possible wherever this condition is met. In trying to pin down a verb' s transitivity , we have suggestedthat verbs can be categorized according to whether or not they denote an eventuality with an external cause and according to whether or not they denote an eventuality which can occur spontaneously. Since these two distinctions are rather similar, we might ask whether there is any need to distinguish between them. In fact, Haspelmath ( 1993) has independently developed an analysis similar to the one we present here, except that he does not make a clear distinction between the two notions. Although Haspelmath is not explicit about this, it appears that he takes the likelihood of spontaneousoccurrence ' for an event to be the opposite of external causation for that event. It seems to us that there is evidence favoring our approach, which takes the two notions to be distinct. Haspelmath links the likelihood of spontaneous occurrenceto intransitivity , without distinguishing betweenunaccusativeand unergative intransitive verbs as we do. For Haspelmath, those verbs which denote events which are likely to occur spontaneously will have an intransitive form , while those which are not likely to occur spontaneously will have only .a transitive form . However, Haspelmath does note that acrosslanguages certain intransitive verbs like break tend to be the morphologically marked member of a causative alternation pair of verbs, while others like laugh tend to be the morphologically unmarked member. It turns out , as he notes, that those verbs which, like break, are both spontaneously occurring and externally caused, are the ones which tend to have the intransitive form as the morphologically marked one. Those which, like laugh, are spontaneously occurring and internally caused tend to have the transitive member of a causative alternation pair morphologically marked. This difference justifies the retention of both notions. In some sense, Haspelmath' s study provides cross-linguistic corroboration of the results we obtained from the in -depth study of a single language.

7. Why can some intern aBy caused verbs have a causative use?

We now return briefly to the question of why some internally causedverbs sometimeshave causative uses. For instance, in section 2 we discussedhow

B. Levin, M . Rappaport Hovav I Causativeverbsin English

the verbs burp and buzz, which we have seen are internally caused, can be used transitively for certain types of arguments, as in the examplesbelow.

(87a) (87b) (88a) (88b)

The nurseburpedthe baby. *The nurseburpedthe doctor. (Smith1970 : (36a,c The postmanbuzzedthe bell. *The postmanbuzzedthe bees.

This phenomenonis sparselyand unevenly distributed acrossthe English verb inventory . For instance, the verb burp may be the only bodily process verb with a causative transitive use. The existenceof causative transitive uses is somewhat more widely attested with verbs of emission, particularly verbs of sound emission. This property might be attributable to the fact that , unlike verbs of bodily process, verbs of emission are typically predicated of inanimates ; therefore, some verbs of emission can describe either internally or externally causedeventualities. Among the verbs of emission that can be used transitively are a few verbs of light emission, including beam and shine, and a somewhat larger number of verbs of sound emission, including buzz, jingle , ring, and rustle. The verb buzz describesa type of sound that is emitted by certain animals - bees - or by certain types of devices- bells and buzzers. This verb can only be used transitively when the emitter of the sound is a device, and only if the device can be causedto emit the sound through direct manipulation by an external cause. Similarly , the verb of light emission beam may be used transitively when the object of the verb is a flashlight, again a ' 28 manipulatable device, but not a person s face. (89a) He beamedthe flashlight in the dark . (89b) * He beamedher face with satisfaction. The following generalization appears to hold of all the verbs of emission with causative transitive uses: they can be used transitively only with an emitter that is directly manipulated by an external cause, and when used in this way, the interpretation must be one in which the emission is directly brought about by an external cause. There are fewer verbs of light emission 28 StevePinkerhaspointedout to us that the transitiveuseof verbsof light emissiongenerally ' hasa meaningwhich includes' aimingin a particulardirection, renderinga directionalphrase that . He suggests perhapsthe analysisof these eitherobligatorily presentor at leastunderstood verbsshouldbe similar to the one we proposefor the causativeformsof the agentiveverbsof mannerof motionin the nextsection.

B. Levin, M. RappaportHovavI Causative verbsin English

with transitive causative usesthan there are verbs of sound emission since in most instances the entities of which verbs of light emission are predicated emit light without the intervention of an external cause, unless theseentities are devices. More verbs of sound emission than verbs of light emission are predicated of entities which emit a sound only under manipulation by an external cause. Some verbs of emission, such as sparkle and burble cited in section 2, never have causative transitive uses. It is unclear to us at this point whether some verbs of emission lack causative uses because they denote eventualities in which causation simply cannot be assumed by an external cause - that is, they are necessarily internally caused - or because, even though external causation may be possible, the set of verbs denoting eventualities compatible with both internal and external causation is explicitly learned from examples. We can now propose an explanation for why burp is apparently the only verb denoting a bodily processwith a transitive causative use. One of the few feasible instances of external causation of a bodily process is burping as it applies to babies. Babies are incapable of burping by themselves, so that the person caring for the baby must assumecontrol of the burping . Thus the verb burp can be used transitively only when babies are involved. We propose then that the eventualities denoted by a small number of English verbs are compatible with either internal or external causation, giving rise to both an intransitive use and a transitive causative use of these verbs. Sincethe causativeuse, when available, is associatedwith direct manipulation of the emitter by an external cause, we assume that in such instances the emitter is no longer viewed as the causeof the eventuality, and that the only cause is the external cause which manipulates the emitter. The Immediate Cause Linking Rule will apply to the external cause, so that it will be the external argument. The Default Linking Rule will apply to the emitter, since it does not meet the conditions on the other linking rules, and it will be the direct internal argument. As mentioned earlier, certain verbs of manner of motion have meanings compatible with either internal or external causation. These verbs include the set of verbs of manner of motion which are not necessarilyagentive, such as swing, bounce, or roll . In Levin and Rappaport Hovav (to appear) we provide evidence that a verb like roll is in fact unaccusativewhen predicated of an inanimate entity, as in The ball rolled ( on the floor ) , but unergative when used agentively, as in The dog rolled ( on thefloor ) . This behavior is just what our analysis predicts. When internal causation is involved, the Immediate Cause Linking Rule will ensure that the single argument, as the internal

verbsin English B. Levin. M. RappaportHovavI Causative

cause, will be the external argument, and the verb will be unergative. When external causation by an agent or a force, such as a push or gravity, is involved but no overt cause is expressed, the single argument will be the direct internal argument due to the Default Linking Rule, and the verb will be unaccusative. (The Directed Change Linking Rule does not apply since there is no directed change; the verb roll is atelic in the absence of a directional phrase.)

8. The interaction of directional phrasesand tr88itivity In this section we return to the last type of causatives rnentioned in the survey in section 2: the causativeusesof agentive verbs of rnanner of rnotion such as march and jump , illustrated in exarnples( I Sb) and ( 16b), which are repeatedbelow. (90a) The general rnarched the soldiers to the tents. (90b) The rider jumped the horse over the fence. These verbs are internally caused monadic predicates. By the linking rules, their single argument should be an external argument; therefore, contrary to fact, these verbs are not expected on our analysis to have the transitive causativeuseswhich some of them do manifest. In this section we provide an account of why some members of this set of internally caused verbs have causativeuses. In the process, we will also provide an answer to another question that is posed by the linking rules that figure in our account of causatives. We have formulated two linking rules which associatearguments with the notion of direct internal argument, and one which associatesarguments with the notion of external argument. Since one of the rules linking arguments to direct internal argument is a default rule, a natural question to ask is why we need the other rule that links arguments to direct internal argument, the Directed ' Change Linking Rule, at all. Couldn t the Default Linking Rule alone yield the same results? For example, if we dispensed with the Directed Change Linking Rule, the Default Linking Rule could be applied to the verb break with the desired result. This section explains why both linking rules are needed. We illustrate the necessityof the Directed Change Linking Rule using the behavior of agentive verbs of manner of motion with respectto causativization.

B. Levin, M . Rappaport Hovav / Causativeverbsin English

We propose that the key to understanding the unexpectedbehavior of the agentiveverbs of manner of motion is the fact that in English, such verbs can be used as verbs of directed motion in the presenceof a directional phrase ( Talmy ] 975, ] 985, among others) . (9Ia ) The soldiers marched to the tents. (91b) The horsejumped over the fence. When an agentive verb of manner of motion is used in a directed motion sense, then both the Immediate CauseLinking Rule and the Directed Change Linking Rule are applicable to the agentive argument. If we assumethat the Directed Change Linking Rule takes precedenceover the Immediate Cause Linking Rule - something that a default linking rule could by definition not do - then the single argument of a verb like run would be a direct internal argument when the verb is usedin a directed motion sense. And indeed many studies of unaccusativity have established that English agentive verbs of manner of motion are unaccusativein the presenceof a directional phrase. 29 Given the unaccusativityof theseverbs with directional phrases, it is possible to give an explanation for why agentiveverbs of manner of motion may have a transitive causative use when they are unaccusative: there is no external argument, so that the external causecan be linked to external argument. Since this alternative linking, which allows us to explain the existenceof the causative use of these verbs, cannot be accomplished by a default rule, we do not dispensewith the Directed Change Linking Rule. This account also explains why a directional phrase is neededor - at the very least - understood when agentiveverbs of manner of motion are usedcausatively. The presenceof such a phrase sanctionsthe alternative linking of the theme argument that permits the introduction of an external cause, explaining the contrasts below. (92a) (92b) (93a) (93b)

The general marched the soldiers to the tents. " The general marched the soldiers. The rider jumped the horse over the fence. * The rider jumped the horse.

Theexamplein (94)I shows that a phrasewith a directionalinterpretation.and not any type of locative phrase, is needed for the causative use. 29 We do not repeat this evidence here; see Hoekstra ( 1984), L. Levin ( 1986), Levin and " Rappaport Hovav ( 1992), C. Rosen ( 1984), among others.

verbsin English B. Levin, M. RappaportHovav/ Causative ( 94) 111ran the mouse around the box .

This example is unacceptable on the locative interpretation, which would involve the mouse running aimlessly around inside the box, but it improves on the directional interpretation where the mouse runs around the perimeter of the box. The constraint against locative phrasesreflects the fact that only directional phrasesallow for a directed change. The process which makes manner of motion verbs into verbs of directed motion is fully productive in English. Therefore, we would expect that the process which transitivizes these directed verbs of manner of motion to be fully productive as well, so that all class members should have a transitive causativeuse in the presenceof a directional phrase. 3o In fact, a wide variety of agentive verbs of manner of motion are attested in causative uses with directional phrases. 31 (95a) ... severalstrong Teamsters... shuffled Kit out of the room ... (L . Matera, A Radical Departure, 1988; Ballatine, New York , 1991, p . 79) " (95b) ... I promised Ms. Cain I would ride her around the ranch . . ." (N . Pickard, Bum Steer, Pocket Books, New York , 1990, p . 92) However, some of theseverbs do not seemto have such causatives. (96a) (96b)

* The trainer logged the runners to the finish line. * The leader climbed the team to the summit.

30 This account leavesunexplained the fact, noted also in Pinker ( 1989), that verbs of directed motion which are not verbs of manner of motion do not have causative counterparts: . She arrived the package ( at the store) . We believethat the lack of causativeswith theseverbs may not be a problem for our accou~t of causativesof verbs of motion . We suspectthat theseverbs are best not characterized as verbs of motion for several reasons, but rather they should be considered verbs of appearance. Interestingly, verbs of appearance, for reasons that we do not . fully understand, do not permit causative uses: The magician appeareda dove (from his sleeve) . Pinker suggeststhat the semantic conditions we formulate here are only necessaryconditions for participation in the alternations. He proposes that membership in lexically specified semantic subclasses of verbs determinesthe sufficient conditions for participation indiathesis alternations in general. These subclasses are implicated in what Pinker caDs narrow-range lexical rules. It remains to be seenwhether the lack of causative usesfor certain classesneeds to be stipulated lexically as Pinker suggestsor can be shown to follow from more general principles. 31 It is clear from the context that in (9Sb) the riders are actually on separatehorses; that is, the exampledoes not have the accompanimentinterpretation found in sentencessuch as I walkedmy dog, which might be argued to instantiate a distinct phenomenon from the phenomenon being discussedhere.

B. Levin. M. RappaportHovavI Causative verbsin English

The unavailability of certain causativescan be attributed to the Immediate Cause Linking Rule itself, which is fonnulated in tenDS of immediate causation . All of the sentenceswith transitive causativeusesof agentive manner of motion verbs imply somesort of coercion (a fact also noted in Pinker 1989). In fact, in the absenceof a particular context, these verbs sound best when the subject is human and the object is an animal, or else when the subject is someonein a position of authority and the object is under that authority . We attribute thesepreferencesto a need to construe such examplesin a way that the subjectcan be interpreted as the immediate causeof the eventuality. Some verbs of manner of motion describe types of motion that do not lend themselvesto an interpretation involving coercion, and such verbs are unacceptab in the causativeuse. This additional condition on causativizationis illustrated by verbs that describeaimlessmotion , such as stroll , mosey, meander and wander. Typically aimless motion cannot be brought about by coercion and, indeed, theseverbs appear not to have a causativeuse.

However, a search of text corpora did yield the following example of a causative use of stroll , suggestingthat in the right circumstanceseven these verbs can causativize, although a reviewer found this example unacceptable, as our analysis would suggest. (98) Julie Smith will stroll you through the Garden District , in New

OrleansM Durning...

(NewYorkTimes )

On this account, agentive verbs of manner of motion enter into a real ' ' processof causativization , in the sensethat the causativeform is the derived form. The account of the causativeforms of theseverbs contrasts with that of the causative forms of verbs like break, which we have argued are basically dyadic and enter into a process of detransitivization. This analysis, as we mentioned in section 3, is corroborated by the fact, noted in Hale and Keyser ( 1987), that cross-linguistically it is the causative form of such verbs which tends to be morphologically marked. 32 32 Another fact which suggests that the pr~ involvedwith verbsof mannerof motion is differentfrom theoneinvolvedwith verbsof changeofstatc is pointedout by Reinhart( 1991 ). She notesthat the introducedsubjectin a transitiveuseof a verb of mannerof motion must be an agent, not an instrumentor naturalfon:e. CompareTheriderjumpedthehorseoverthefencewith . Thewhip/thelightningjumpedthehorseoverthe fence. Thispropertyis alsonotedin Cruse( 1972 )

verbsin English B. Levin, M. RappaportBovav/ Causative

Our account of the transitive use of verbs of manner of motion is different from our account of the transitive use of verbs of emission. We have claimed that certain verbs of emission can be construed as being externally caused; becauseof that, a directional phraseis not required to effect a changefrom an unergativeto an unaccusativeverb. In contrast, verbs of manner of motion are never really consideredexternally caused, so that a directional phrase must be introduced to effect t.hechangein the classificationof the verb from unergative to unaccusative. The introduction of the new external causeis constrained in that the external causemust somehowbe construed as an immediate cause. In fact, as we discuss at greater length in Levin and Rappaport Hovav (to appear), even with certain inanimate emitters, the emission of the sound does not come about by direct manipulation of the emitter, so that causativesare only possible in the presenceof a directional phrase (e.g., The driver roared/ screechedthe car into the driveway). That is, the situation reducesto precisely the situation observedwith agentiveverbs of manner of motion. The crucial part that the directional phrase plays in sanctioning the causative use of agentive verbs of manner of motion is brought out by comparing the behavior of the verbs run and roll . The verb roll , although a manner of motion verb, is not necessarily agentive and falls rather into ' Jespersen' s ' move class. As discussedat the end of section 7, the type of motion that roll denotes can be either internally caused OT, when brought about by an agent or a force such as a push or gravity , externally caused. Depending on whether the verb roll is understood as internally or externally caused, monadic roll would be predicted to behave either as an unaccusative or as an unergative verb. When the verb takes an animate agentive argument, it would be expectedto show unergative behavior since the rolling would be internally caused. In fact, when the verb takes an animate subject, it can be found in the prepositional passive construction, a construction that Perlmutter and Postal ( 1984) argue is only possible with unergative verbs. (99) This carpet has been rolled on by three generationsof children. When its argument is inanimate, the eventuality denoted by the verb would be externally caused. The argument would be an internal argument by the Default Linking Rule, since neither of the other two linking rules would be applicable, and the verb would be expected to show unaccusativebahavior. In fact, the verb cannot be found in an unergative type resultative construction with an inanimate subject, as shown in ( I OOa), though it can be found in an unaccusativetype resultative construction, as shown in ( l00b ) .

B. Levin HovavI Causative verbsin English , M. Rappaport ( I OOa) * The bowling balls rolled the markings off the floor . (cr. The basketball players dribbled the markings off the floor .) I ( OOb) The door rolled open/ shut. This account of why roll can be unaccusativedoes not make referenceto a directional phrase, contrasting with the account of why run can be unaccusative . However, like unaccusative run, unaccusative roll should allow for a causativecounterpart, though again without the necessaryaccompanimentof a directional phrase. As predicted, the verb roll can be used causatively even in the absenceof a directional phrase. ( I Ola) The bowling ball rolled (into the room ) . ( I Olb) The bowler rolled the bowling ball (into the room) . The contrasting behavior of the verbs roll and run supports our account of thesetwo verbs.

9. Conclusion

In this paper we have unravelled someof the puzzlesconcerningthe causative alternation. Central to our analysis is the distinction betweenverbs which are inherently monadic and verbs which are inherently dyadic. This distinction is related to - but not reducible t033 - the distinction betweenunaccusativeand unergativeverbs. With respectto the phenomenathat havecome under the label 'causativealternation' we have , suggestedthat the more productive processin 'detransitive' verbs from lexical transitive causative is one which forms English verbs, as in the case of the verb break. Some verbs, such as buzz have both transitive and intransitive usessincethe meaningof the verb is compatible with both internal and externalcausation; this phenomenonis restrictedonly to those verbs which are indeedcompatible with both interpretations. Transitivization of agentiveverbs of manner of motion involves the introduction of an agent to an inherently monadic verb when, due to the presenceof a directional phrase, the verb no longer takesan externalargument. We hope that this study of causative verbsin English will help to illuminate our understandingof the much discussed , still elusive notion of . , though transitivity 33 We argue in Levin and Rappaport Hovav (to appear) that verbs of inherently directed motion such as arrive are unaccusativeand monadic.

B. Levin, M . Rappaport Hovav I Causativeverbsin English

References Anderson . Commentson the paper by Wasow. In: P. Culicover, A. Akmajian, , S.R., 1977 T. Wasow(eds.), 1977 , 361- 377. Baker, M., 1988 . Incorporation: A theory of grammaticalfunction changing . Chicago , IL : . Universityof ChicagoPress -inchoativealternation: A casestudyin parallelmorphology . TheCausative Borer, H., 1991 . The LinguisticReview8, 119- 158. Bresnan . Deep unaccusativityin LFG. In: K. Dziwirek, P. Farrell, . J. and A. Zaenen , 1990 E. Mejias-Bikandi (eds.), Grammaticalrelations: A cross-theoreticalperspective , 45- 57. Stanford, CA: CSLI, StanfordUniversity. Brousseau . A non-unifiedanalysisofagentiveverbs. WCCFL 10. , A.-M. and E. Ritter, 1991 -Bindingapproach Burzio. L., 1986 . Italian syntax: A Government . Dordrecht: Reidel. . On linking: Papersby RichardCarter. (LexiconProjectWorking Papers25, Carter, R., 1988 editedby B. Levinand C. Tenny.) Cambridge , MA : Centerfor CognitiveScience , MIT . . Languageuniversalsand linguistic typology. Chicago Comrie, B., 1981 , IL : University of .' ChicagoPress Cruse. D.A., 1972 . A noteon Englishcausatives . LinguisticInquiry 3, 522- 528. Culicover . Formal syntax. New York: Academic . P., A. Akmajianand T. Wasow(eds.), 1977 . Press Dixon, R.M.W., 1982 . Wherehaveall the adjectivesgone? In : R.M.W. Dixon, Wherehaveall the adjectives ? and syntax, 1- 62. Berlin: Mouton. gone and otheressaysin semantics . Word meaningand Montaguegrammar. Dordrecht: Reidel. Dowty, D.R., 1979 . Thematicproto-rolesand argumentselection . Language67, 547- 619. Dowty, D.R., 1991 Fillmore, C.J., 1967 . The grammarof hilling and breaking . In: R. Jacobs , P. Rosenbaum (eds.), Readingsin Englishtransformationalgrammar, 120- 133. Waltham, MA : Ginn. Fillmore, C.J.. 1968 . The casefor case. In: E. Bach, R.T. Harms(eds.), Universalsin linguistic 1 88 . New York: Holt, Rinehartand Winston. , theory Fontenelle . Retrievingergativeverbs from a lexical data base. , T. and J. Vanandroye , 1989 DictionariesII , 11- 39. . On Berberverbsof change: A study of transitivity alternations Guerssel . (Lexicon , M., 1986 , MA : Centerfor CognitiveScience , MIT . ProjectWorkingPapers9.) Cambridge Hale. K.L. and S.J. Keyser . Sometransitivity alternationsin English. (LexiconProject , 1986 7. , , MIT . ) WorkingPapers CambridgeMA : Centerfor CognitiveScience Hale, k .L. and S.J. Keyser, 1987 . A viewfrom the middle. (LexiconProjectWorkingPapers10.) , MA : Centerfor CognitiveScience , MIT . Cambridge . More on the typology of inchoative . In: , M., 1993 /causativeverb alternations Haspelmath B. Comrie(ed.), Causatives : Benjamins . and transitivity. Amsterdam Hoekstra . Transitivity. Dordrecht: Foris. , T., 1984 . Smallclauseresults. Lingua74, 101- 139. Hoekstra , T., 1988 Hoekstra . Aspectand thetatheory. In: I.M. Roca(ed.), 1992 . T., 1992 , 145- 174. . Oxford advancedlearner's dictionary of current English. Oxford: Hornby, A.S. (ed.), 1974 . Oxford UniversityPress . Semanticstructures Jackendoff . Cambridge . , R.S., 1990 , MA : MIT Press Jesperse n, 0 ., 1927 . A modemEnglishgrammaron historicalprinciples : , Volume3. Heidelberg Carl Winters.

B. Levin, M. Rappaport Hovav/ Causative verbsin English . On the middleandergativeconstructions in English. Linguistic , 1984 Keyser,S.J. andT. Roeper Inquiry 15, 381- 416. Lakoff, G., 1968 . Someverbsof changeand causation . (Report NSF-20.) Cambridge , MA : Aiken ComputationLaboratory, HarvardUniversity. Lakoff, G., 1970 . Irregularityin syntax. New York: Holt, Rinehartand Winston. Levin, B., 1989 . The Basqueverbalinventoryandconfigurationality . In: L. Maracz , P. Muysken : The typologyof asymmetries , 39- 62. Dordrecht: Foris. (eds.), Configurationality Levin, B., 1993 . Englishverbclasses . Chicago and alternations : A preliminaryinvestigation , IL : . Universityof ChicagoPress Levin, B. and M. RappaportHovav, 1992 . The lexical semanticsof verbs of motion: The from unaccusativity . In: I.M. Roca(ed.), 1992 , 247- 269. perspective Levin, B. and M. RappaportHovav, to appear. Unaccusativity : At the syntax-lexicalsemantics interface . Cambridge MA : MIT . Press , Levin, L., 1986 . Operationson lexicalfonDS: Unaccusative rulesin Germaniclanguages . PhiD. thesis , MIT . Marantz, A.P., 1984 . On the natureof grammaticalrelations.Cambridge . , MA : MIT Press Marantz, A.P., 1992 . The wayconstructions in English. In: and thesemantics of directarguments T. Stowell, E. Wehrli (eds.), Syntaxand semantics , Vol. 26: Syntaxand the lexicon, 179- 188. New York: AcademicPress . . Lexicalinsertionin a transfonnationalgrammarwithout deepstructure. , J.D., 1968 McCawley CLS 4, 71- 80. Reprintedwith notesin: J.D. McCawley , 1973 , Grammarand meaning , 154166. Tokyo: Taishukan . . Nekotoryeverojatnostnye universaliiv glagol'nom slovoobrazovanii . In: Nedjalkov, V.P., 1969 ' I.F. Vardul (ed.), Jazykovyeuniversaliii lingvisticheskajatipologija, 106- 114. Moscow: Nauka. . The typologyof morphologicaland lexicalcausatives . Nedjalkov, V.P. and GiG. Silnitsky, 1973 In: F. Kiefer (ed.), Trendsin Soviettheoreticallinguistics , 1- 32. Dordrecht: Reidel. Perlmutter . Impersonalpassives and the Unaccusative . BLS4, 157- 189. , D.M., 1978 Hypothesis Perlmutter . The I -AdvancementExclusiveness Law. In: D.M. , D.M. and P. Postal, 1984 Perlmutter , C. Rosen(eds.), 1984 , 81- 125. Perlmutter . Studiesin RelationalGrammar2. Chicago, IL : , D.M. and C. Rosen(eds.), 1984 . Universityof ChicagoPress Pinker, S., 1989 . Learnabilityand cognition: The acquisitionof argumentstructure.Cambridge , MA : MIT Press . . Longmandictionaryof contemporaryEnglish. London: Longman Procter, P., et al. (eds.), 1978 Group. . Semantics and the lexicon. Dordrecht: Kluwer. , J. (ed.), 1993 Pustejovsky . What to do with theta-roles. In: W. Wilkins(ed.), Syntaxand , M. and B. Levin, 1988 Rappaport semantics . , Vol. 21: Thematicrelations, 7- 36. New York: AcademicPress . Lexique7, . Niveauxde representation lexicale , , 1988 RappaportM., B. Levinand M. Laughren 13- 32. Appearsin Englishas: Levelsof lexicalrepresentation . In: J. Pustejovsky , (ed.), 1993 37- 54. . Lexicalpropertiesof ergativity. Paperpresented Reinhart, T., 1991 at the Workshopon lexical and lexicalinsertion, December9- 11, 1991 . Utrecht, The Netherlands : Research specification Institutefor Languageand Speech , Universityof Utrecht. Roca, I.M., 1992 . Thematicstructure: Its role in grammar. Berlin: Walterde Gruyter.

B. Levin. M . Rappaport Hovav I Causativeverbsin English

. The interfacebetweensemanticroles and initial grammaticalrelations . In : Rosen , C., 1984 D.M. Perlmutter , C. Rosen(eds.), 1984 , 38- 77. . Argumentstructureand complexpredicates . BrandeisUniversity. (PhiD. Rosen , S.T., 1989 thesis .) . Lesverbesa la fois transitifset intransitifs en Fran~ s contemporain . , M., 1974 Rothemberg The Hague: Mouton. . Theoriesyntaxiqueet syntaxedu Fran~ is. Paris: Editionsdu Seuil. Ruwet, N., 1972 . The grammarof causativeconstructions : A conspectus . In: M. Shibatani Shibatani , M., 1976 , Vol. 6: Thegrammarof causativeconstructions , 1- 40. NewYork: (ed.), Syntaxandsemantics . AcademicPress . Resultatives . In: L. Levin, M. Rappaport , J., 1983 , A. Zaenen(eds.), PapersinLexicalSimpson FunctionalGrammar, 143 --157. Bloomington , IN : IndianaUniversityLinguisticsClub. ' classand causativeverbs in . Jesperse n's ' move and change Smith, C.S., 1970 English. In : M.A. Jazayery , E.C. Palome , W. Winter (eds.), Linguisticand literary studiesin honor of ArchibaldA. Hill , Vol. 2: Descriptivelinguistics , 101- 109. The Hague: Mouton. of motion . In : J.P. Kimball (ed.), Syntaxand semantics L. 1975 . Semantics and , syntax Talmy, , . Vol. 4: 181-238. New York: AcademicPress . Lexicalizationpatterns: Semanticstructurein lexicalforms. In: T. Shopen(ed.), Talmy, L., 1985 and the lexicon, , Vol. 3: Grammaticalcategories Languagetypologyand syntacticdescription . : CambridgeUniversityPress 57- 149. Cambridge . PhiD. thesis . Grammaticalizing , MIT . aspectand affectedness Tenny, C., 1987 . Transformations and the lexicon. In: P. Culicover, A. Akmajian, T. Wasow Wasow , T., 1977 , 327- 360. (eds.), 1977 . In: J. Pustejovsky(ed.), . Unaccusativityin Dutch: An integratedapproach Zaenen , A., 1993 1993 , 129- 161.

Section 2

thewordunits Discovering

) 81- 104. North-Holland Lingua92 ( 1994

Segmentation

problems , rhythmic

solutions *

Anne Cutler . . MedicalResearch Council.AppliedpsychologyUnit. 15 ChaucerRoad. Cambridge CB22EF. UK Thelexiconcontainsdiscreteentries,whichmustbe locatedin speechinput in orderfor speech to be understood ; but the continuity of speechsignalsmeansthat lexicalaccessfrom spoken . The speechenvironmentof prelinguistic input involvesa segmentation problemfor listeners infantsmaynot providespecialinformationto assistthe infant listenersin solvingthis problem. Maturelanguageusersin possession of a lexiconmight be thoughtto be able to avoid explicit segmentationof speechby relying on information from successfullexical access ; however , evidencefrom adult perceptualstudiesindicatesthat listenersdo use explicit segmentation -specific . Theseproceduresdiffer acrosslanguagesand seemto exploit language procedures . structure Efficient as these are not have been rhythmic procedures , they may developedin responseto statisticalpropertiesof the input, becausebilinguals, equally competentin two onerhythmicsegmentation . Theorigin of rhythmic , apparentlyonly possess procedure ;languages ' segmentation may thereforelie in the infant s exploitationof rhythm to solvethe segmentation . Recentevidencefrom speechproduction problemand gain a first toeholdon lexicalacquisition and perceptionstudieswith prelinguisticinfantssupportsthe claim that infantsare sensitiveto . rhythmicstructureand its relationshipto lexicalsegmentation

I . Introduction: Using the lexicon in listening to speech The lexicon, which is the focus of the present collection of papers, is assumed to be an essential component of every language user' s linguistic apparatus. The contents of a lexicon are so patently language-specific that it goeswithout saying that a lexicon cannot be inborn - it must be acquired, on the basis of linguistic experience. Such experienceusually consists of hearing utterances, which typically are many words in length. But we can safely . Financialsupportis acknowledged from the Human Frontier ScienceProgram. Thanksas with whom the many and variedexperimentssummarised herewere alwaysto the colleagues carriedout: Sally Butterfield, David Carter, Giyoo Hatano, PeterJusczyk , JamesMcQueen , JacquesMehler, Dennis Norris, TakasmOtake and Juan Segui. For discussionsand other with this paperthanksto MelissaBowerman assistance , Sally Butterfield, Mike Kelly, James . , JacquesMehler, DennisNorris and Ann Peters McQueen . . AnneCutler is now at Max PlanckInstitutefor Psycholinguistics , P.O. Box 310, 6500AH . , The Netherlands Nijmegen - 3841/94/507.00 @ 1994- ElsevierScienceB.V. All rightsreserved 0024 -A SSDIOO24-3841( 93) EOO39

A. Cutler/ Segmentation problems

assumethat complete utterancesare not what lexicons consist of. The number of utterances is potentially infinite , so that to store all the utterances we might ever hear our lexicon would also have to have infinite capacity. Even if we set an arbitrary length limit , the number of possible utterances is enormously large; for instance, Miller ( 1967) calculated that there must be at least 1020possible English sentencesof twenty words or less - a total which, he drily added, would take considerably longer than the estimated age of the earth to speak. Instead, we assume that the contents of a lexicon consist of sound-tomeaning mappings in discrete chunks. ( Wecan refer to lexical entries by the shorthand term ' words' , although of course not aU lexical entries necessarily correspond to what would be written as a single separate word. Some subword forms such as affixes or stem morphemes may well have lexical representation, as may particles which are conjoined with other words in writing ; likewise, multi -word idiomatic expressionsand frequently occurring phrasesmay be representedby a single entry .) Thus using a lexicon requires the separation of utterances into the lexically relevant chunks of which they are made up - producing speechrequires the languageuser to string together lexical entries to make a whole utterance, and recognising speech requires division of an utterance into units which can be looked up in the lexicon. Likewise, acquiring a lexicon eventually involves acquiring the ability to use it in theseways. The present contribution focusseson the very start of lexicon-building : how the infant might find out what words in the input languageare like , and might assemblean initial stock of known words. The initial task is perceptual. What exactly does it involve? For instance, does it involve (as mature useof a lexicon in speechrecognition involves) division of multiword utterancesinto lexically relevant chunks? And if so, how difficult is this task? To answer thesequestions we need to consider the nature of the speechinput with which the infant is most likely to be confronted. Comparative studies of various types of speechare consideredin the next section.

2. Stylesof adult - directetl

speech

Most of the speechany listener hears is spoken spontaneously- the speech signalswhich occur in the majority of everydaysituations have beenconceived and composed by their speakerseven as they are uttered. For most listeners, spontaneous speech is encountered far more often than other styles such

A. Cutler / Segmentationproblems

as rehearsed speech (heard in the theatre or on radio or television, for example), read speech (in news broadcasts, or , too often, in lectures) or computer-synthesisedspeech. Whatever the style of speech, words in isolation occur only rarely - nearly all utterances are multiword . A lot is known about the phonetics of multiword utterances, and a fair summary of our knowledge is that words are strongly affected by the contexts in which they occur; moreover, these contextual assimilation processes operate to obscure word boundaries, with the result that there are few reliable cues in a continuous speechsignal to where one word ends and the next begins. Klatt ( 1989) provides a telling overview of the problems which this causesfor the lexical accessprocess so essentialfor speechrecognition. Nevertheless, the majority of such phonetic studieshave beenconducted on speechproduced in laboratory situations, which is normally read speech. Is this a fair representation of the speech which most listeners usually hear? Motivated by this question, speechscientists have undertaken a number of studies aimed at describing spontaneous speech, and becauseof the underlying motivation , most of the studies have been comparative: spontaneous speech has been contrasted with read speech. These studies have revealed systematic differences between the two types of speech. Some of these differences might render the listener' s problems even worse in spontaneous than in read speech. For example, casual spontaneousspeechis particularly prone to phonological elisions and assimilations (G. Brown 1977, Labov 1972, Milroy 1980) and to syntactic simplifications and, occasionally, incompleteness (Cheshire 1982, Labov 1972) . Other differences, however, might make life easier for listeners to spontaneous speech. These are principally differences in the prosodic domain. Thus spontaneous speech tends to be produced at a slower rate than read speech(Barik 1977, Johns-Lewis 1986, Levin et al. 1982), and to have longer and more frequent pauses and hesitations (Barik 1977, Crystal and Davy 1969, Kowal et al. 1975, Levin et al. 1982) and shorter prosodic units (Crystal and Davy 1969) . Listeners can distinguish spontaneous utterances either from read speech (Levin et al. 1982, Remezet al. 1985, Blaauw 1991) or from rehearsedspeech (Johns- Lewis 1987) ; their judgements are most likely based on prosodic aspectsof the speech, becauseaccuracy is still high when the speechextracts have been low-pass filtered (Levin et al. 1982), while the distinction can not be as accurately made on written versions of the text (Johns- Lewis 1987) . Fluent spontaneous speech can be identified as accurately as disftuent (Blaauw 1991).

A. CullerI Segmentation problems

The prosodic differencesbetweenspontaneousand read speechhave consequences for the way speechin each mode is processedby listeners. McAllister ( 1991) examined word recognition in spontaneousand read speechusing the gating task, in which listeners hear successivelylarger fragments of a word. She found that word identification (in context) occurred earlier for a word stressedon the first rather than on the secondsyllable in spontaneousspeech, but not in read speech. In a word -by-word gating study of spontaneous speech Bard et ale ( 1988) and Shillcock et ale ( 1988) similarly found that words containing strong syllables were easier to identify than words which were realised as weak syllables. Mehta and Cutler ( 1988) investigated phoneme detection reaction time in spontaneous and read speech, and compared in particular the relative strength in the two speech modes of a number of previously established effects. They found no overall difference in responsetime between the two speech modes, and also no difference between the two modes on the one semanticvariable in the study, the effectsof the transitional probability of the target-bearing word. However, four other effects differed across modes. In read speechbut not in spontaneous speech, late targets were detected more rapidly than early targets, and targets precededby long words were detected more rapidly than targets preceded by short words. In contrast, in spontaneous speechbut not in read speech, targets were detected more rapidly in accentedthan in unaccentedwords and in strong than in weak syllables. Mehta and Cutler explained these differencesin terms of prosodic differences betweenthe two speechmodes. The greater frequency of hesitations in spontaneousspeech, for example, results in shorter prosodic units, which in turn reduces the average span over which rhythmic predictability will hold. So becauseprosodic units are long - generally clause-length - in read speech, but usually short in spontaneous speech, the opportunity for rhythmic prediction in the latter case is much smaller. Mehta and Cutler thus argued that position in the sentence is not , strictly speaking, what affects target detection time ; rather, the effective variable is position in the prosodic unit . Similarly, becausehesitations tend to be more frequent and longer in spontaneous speech,,it is much more likely that a particular target-bearing word will be precededby a hesitation in the spontaneousthan in the read mode. Where a target is immediately preceded by a hesitation, any effects of incomplete processingof the previous word will be nullified by the extra processingtime provided by the hesitation, so that effectsof precedingword length, which are held to reflect just such processinghangovers from the preceding word , will be lesslikely . Finally , becauseaccent patterns in spontaneousutteranceswere

A. Cutler / Segmentationproblems

more varied and less likely to expressdefault accenting than those in read utterances, and the acoustic differences between strong and weak syllables were greater in spontaneousthan in read speech, there was greater opportunity for processing effects of both sentence accent and syllable stress to appear in the spontaneousthan in the read speech; this would account for the finding of significant facilitation due to sentenceaccent and syllable stressin the fonner but not in the latter. The results of the gating studies described above provide similar evidenceof the perceptual importance of syllable stress in spontaneousspeech. These findings speak to the majority case for speech processing. Most speech that adult listeners hear is spontaneously produced. Such speech is characterised by a fairly slow overall rate of speech, short prosodic units, frequent pausesand, in English, a clear opposition betweenstrong and weak syllables. These factors affect the way the speechis processed.

3. The infant's speechenvironment Is the infant ' s speechenvironment the sameas that of an adult listener? In one respect it is not , becausespeechdirected to the infant as chosen listener exhibits a pattern which is systematicenough to warrant calling it a separate speechstyle. This conclusion emergesfrom a number of studies which have investigated the characteristics of speech addressed to young children at various stagesof development; again, the studies have mainly been comparative , with infant -directed spontaneousspeechbeing typically contrasted with speechfrom the same speakersto adults. In European languages, infant -directed speech tends to be spoken at a slower rate, to have more frequent stresses, shorter continuous sequencesand longer pauses, and to be higher in pitch than adult-directed speech(Fernald and Kuhl 1987, Fernald and Simon 1984, Fernald et al. 1989, Garnica 1977, Stern et al. 1983) . Slower rate, more frequent prosodic demarcation, and longer pauses, it will immediately be noted, are the primary factors which the review in the previous section revealed as distinguishing spontaneous from ' non-spontaneousspeech. To the present author s knowledge thesetwo literatures have not been directly compared; there would seemto be a defensible case, however, for considering adult-directed and it:tfant-directed spontaneous speechin terms of a single continuum , with infant ~directed speechoccupying a more extreme position on most measures.

A. CutlerI Segmentation problems

The one exception is that infant -directed speechis reported to have higher pitch and a wider fundamental frequency range ( Fernald and Simon 1984, Garnica 1977) . In contrast, the fundamental frequency range of spontaneous speechhas been reported in at least some studies to be relatively narrow , at least in intimate conversation (Johns-Lewis 1986, Blaauw 1991) . Pitch is a particularly important dimension of infant -directed speech, since the fact that infants prefer to listen to this style of speech(Fernald 1985) has been found to be principally due to its pitch characteristics (Fernald and Kuhl 1987, Sullivan and Horowitz 1983) . Ohala ( 1983, 1984) has argued that raised pitch is an ethologically universal signal of smallness, ingratiation and non-threatening attitude. From such a perspective it would be possible to argue that raised pitch might not be a phonologically relevant manipulation in speechto infants, but might simply arise from universal expressionof affection or nurturance on the part of an adult to an infant . Against this conclusion, on the other hand, might be cited the more recent findings that the pitch manipulations found in infant -directed speech in American English and related languagesare apparently not universal . Although rising contours predominate in infant -directed speechin the stresslanguagesEnglish (Sullivan and Horowitz 1983) and German (Fernald and Simon 1984), falling contours are more prevalent in the tone languages Mandarin (Grieser and Kuhl 1988) and Thai (Tuaycharoen 1978) . In a comprehensive review of the literature on pitch in infant -directed speech, Shute ( 1987) concluded that pitch modifications are not only clearly not universal across languages, but may also differ within one language as a function of sex of the speaker, age of the child addressee , frequency of the ' s interaction with children and other factors. speaker In fact a recognisablestyle of infant -directed speechis itself not universal, contrary to the confident expectations of researchersin the 70s that it would prove not only to be universal (Ferguson 1977) but absolutely necessaryfor successfulacquisition (R. Brown 1977) . It is now clear that there are cultures where infants are exposedto much normal adult speechbut no speechin any special infant -directed mode (Heath 1983, Schieffelin 1985, Schieffelin and Ochs 1983) . Even where infant -directed speech appears to conform to the pattern obserVedin English and like languages, this may not constitute a specialisedmode; thus infant -directed speechin Quiche Mayan has relatively high pitch , but so, in this language, does adult -directed speechfrom the same informants (Bernstein-Ratner and Pye 1984) . Thus it is reasonable to conclude that infants in the earliest stages of - and for some language acquisition receive at least some of their input

A. Cutler/ Segmentation problems

infants, perhaps all of their input - in a fonn that at least closely resembles nonnal spontaneous speech between adults. One of the characteristics of spontaneousspeech, it will be recalled, is the high frequency of phonological elisions and assimilations (G. Brown 1977, Labov 1972, Milroy 1980) . Some studies have reported that child -directed speech, too , is replete with such distorting processes (Bard and Anderson 1983, 1991; Shockey and Bond 1980), which is consistent with the view that this style of speechlies on a generalcontinuum with adult-directed casual speech. Other studies, however, have reported lower frequency of distorting phonological transformations in speechto infants than in speechto adults (Bernstein- Ratner 1984a,b) . In an attempt to resolve this apparent contradiction , Stoel-Gammon ( 1984) transcribed five hours of speechto one-year-olds; her results strongly support the view of a continuum , since she effectively discovereda continuum within her own data, from very clear articulation (e.g. releaseof word -final stop bursts, clear articulation of unstressedvowels) to very casual forms (frequent vowel reduction, omissions of whole syllables such as [sko] as a pronunciation of ' let s go) . Stoel-Gammon concluded that the phonological characteristics of speech to children depend on such factors as contextual redundancy, the function of the individual utterance, and the situational context - the same factors that determine the phonologicalforms of adult spontaneous speech (Liebennan 1963, Cheshire 1982) . There is to my knowledge no evidence, from any culture, of a greater incidence of isolated words in speech to children than in other forms of speech. Even though phrases may be short, they are still phrases. Thus a speechsegmentationproblem, as describedin the introductory section to this paper, seemsto exist for the infant as for the adult languageuser. The speech ' that the infant hears is continuous; much of the speech of the infant s environment will be speech among mature language users; in perhaps a majority of cultures speechaddressedspecifically to the infant would fonn only a small proportion of the input ; even then, such speech may not necessarilybe clearly articulated. The problem is compounded for the infant by the necessity of compiling a lexicon, and this added difficulty does not trade off against reduced segmentation difficulty in the input . In fact, the scaleof the segmentationproblem in the structure of the input is remarkably similar for the infant and for the adult.

A. CutlerI Segmentation problems 4. The segmentation problem for adult listeners

The adult listener typically hears continuous multiword utterances, and must therefore segmentthe speechstream in order to understand them. The importance of this segmentation problem has long been acknowledged, and many, widely varying, experimental approaches have tackled the question (e.g. Hayes and Clark 1970, Wakefield et al. 1974, Pilon 1981) . But the adult ' s situation differs from that of the child becauseas a mature language user the adult is already in possession of a lexicon. This has led some researchersto claim that the adult listener has no need of explicit segmentation procedures, since the successfulrecognition of a word will ensure that whatever immediately follows that word will be known to be word -initial . For example, Cole and Jakimik ( 1978) proposed that recognition of spoken utterances proceeds in strictly temporal order, and ' one word ' s recognition ' automatically 'directs segmentationof the immediately following word ' ( 1978: 93) . We could call such a model ' segmentationby default' . On closer inspection, though, it becomes obvious that segmentation by default could not work , at least for English. Firstly , the model relies on listeners being indeed able to determine where a word ends. It is true that sometimesphonetic sequenceconstraints will be of use in this. Somesequences of phonemes- [au k ], [m g] , [ai S] , for example - never occur word -internally in English (Lamel and Zue 1984, Harrington et al. 1987) ; thus a sentencelike How comeGuy shaved? should prove very easyto segment. But unfortunately such helpful sequencesare rare. As McQueen and Cutler ( 1992) have recently demonstrated, the English vocabulary contains few totally distinct words ; most long words have other words embedded in them (as reconciliation contains wreck, reckon, sill , silly , ill , etc.), while most short words occur within longer words (as late can be found in latency, collate, belated, ' translatability, etc.) . Moreover, McQueen and Cutler s computations showed that the majority of such overlaps occur at the beginnings of the longer words, a particular problem for segmentationby default. Of course, in typical English speechthe majority of words are monosyllabic Cutler and Carter 1987), which will certainly reduce the problems causedby ( such embeddings. But most monosyllabic English words do not become unique until at or after their final phoneme (Luce 1986) ; and in fact many words - especially monosyllabic words - can not be recognised until after their acoustic offset. Post-offset recognition has been demonstrated both with laboratory -produced (i.e. carefully read) speech (Grosjean 1985), and with spontaneously produced speech(Bard et al. 1988, Shillcock et al. 1988) . If

A. Cutler / Segmentationproblems

words cannot be recognisedtill after their ends then segmentationby default would lose its very basis. Secondly, models such as segmentationby default are far from robust ; they assumethat prelexical processingof the speechsignal will be accurate. But in practice speechsignals are not always fully clear. Background noise, distance between speaker and listener, distortion of the speaker's vocal tract , foreign accents, slips of the tongue - all these, and similar factors, conspire to make the listener' s phonetic interpretation task harder. A much more robust model is neededto account for what is obviously true, namely that human speech recognition is extremely successful even under noisy conditions or with previously unfamiliar voices or accents.

S. A solutionfor English: Rhythmicsegmentation In fact there is a good deal of evidencefrom human speechrecognition in English that explicit segmentationprocedures.are employed by adult listeners. Cutler and Norris ( 1988) suggestedthat the characteristic rhythmic structure of English could form the basis for an effective segmentation procedure, becauseEnglish speech shows a systematic relationship between rhythmic patterns and word boundary locations. The rhythm of English is based on stress, with syllables of the language being either strong or weak; strong syllables contain full vowels, while weak syllables contain reduced vowels (usually schwa) . Cutler and Carter ( 1987) demonstrated that English lexical words are far more likely than not to begin with strong syllables - in a -word phonetically transcribed dictionary (the MRC Psycholinguistic 33,OOO Database: Coltheart 1981, Wilson 1988), 73% of all entries had strong initial syllables. But the frequency of occurrenceof individual words differs widely; lexical, or content words, are sometimesvery common (e.g. people), but more often are very rare (e.g. peon, steeple), while some words which inrunning speech are usually realised as weak syllables - grammatical, or function words, such as of or the - occur very frequently. Cutler and Carter examined -word natural speech sample, the Corpus of English Conversation a 190,OOO Svartvik and Quirk 1980), using the frequency count of this corpus prepared ( by Brown ( 1984) ; they found that in this corpus 90% of the lexical words have strong initial syllables. However, the grammatical words in the corpus were actually in the majority , and they were virtually all weak monosyllables. Cutler and Carter computed that about three-quarters of all strong syllables in the sample were the sole or initial syllables of lexical words, while more

A. Cutler / Segmentationproblems

than two-thirds of all weak syllables w~re the sole or initial syllables of grammatical words. This means that a listener encountering a strong syllable in spontaneous English conversation would seem to have about a three to one chance of finding that strong syllable to be the onset of a new lexical word. A weak syllable, on the other hand, would be most likely to be a grammatical word. English speech should therefore lend itself to a segmentation procedure whereby strong syllablesare assumedto be the onsetsof lexical words. Cutler and Norris interpreted results of an experiment they ran as evidencefor such a procedure. They used a task which they called word -spotting, in which listeners were asked to detect real words embeddedin nonsensebisyllables; detection times were slower to the embeddedword in , say, mintayf (in which the second vowel is strong) than in mintef (in which the second vowel is schwa) . Cutler and Norris interpreted this as evidence that listeners were segmenting mintayf prior to the second syllable, so that detection of mint therefore required combining speechmaterial from parts of the signal which had been segmentedfrom one another. No such difficulty would arise for the detection of mint in mintef, since the weak second syllable would not be divided from the preceding material. Further evidence for such a procedure was produced by Cutler and Butterfield ( 1992), who investigated the way in which word boundaries tend to be misperceived. In both spontaneousand experimentally elicited misperceptions they found that erroneous insertions of a word boundary before a strong syllable (e.g. achievebeing heard as a cheap) and deletions of a word boundary before a weak syllable (e.g. bird in being heard as burgling) were far more common than erroneous insertions of a boundary before a weak syllable (e.g. effective being heard as effect oJ> or deletions of a boundary before a strong syllable (e.g. were waiting being heard as awaken) . This is exactly what would be expected if listeners deal with the segmentation problem by assuming that strong syllables are likely to be word initial , but weak syllables are not. As Cutler and Norris point out , the strong syllable is defined by the quality of its vowel (full , in comparison to the reduced vowels of weak syllables) ; thus spotting strong syllables cannot provide a complete solution to the segmentation problem since word boundaries actually occur prior to the onset of syllables. A strong syllable spotter must be supplementedby some means of estimating actual syllable onset; Cutler and Norris suggest that more than one alternative realisation of such a device would be feasible. Assuming that a rhythmically based segmentation procedure is indeed prac

A. Cutler I Segmentationproblems

tical , its advantages are considerable. For instance, such a procedure is obviously not going to be affected by the frequency of words embedded within other words in speech, or by the relative frequency of monosyllables versus polysyllables. Only where polysyllabic words contain strong syllables in non-initial position will the procedure produce a non-optimal result (i.e. it will signal a word boundary but this will be a false alarm) . However, polysyllabic words with non-initial strong syllables occur relatively rarely (Cutler and Carter 1987), and in only a small minority of them will a false alarm actually produce a real word unrelated to the embedding word (e.g. late in collate; Cutler and McQueen, in press). Thus rhythmic segmentationis a relatively efficient procedure for English. It is also quite robust - in fact, it is precisely with uncertain input that rhythmic segmentation proves particularly useful. R.esearchersin automatic speechrecognition (e.g. Shipman and Zue 1982) have developed systematic representationsof phonetic uncertainty, namely transcriptions in which only general classesof phoneme are provided (e.g. glide, nasal, stop consonant, etc.). Two studies using uncertain input of this kind have produced further evidence in favour of rhythmic segmentation. In the first study, Briscoe ( 1989) implemented four segmentation algorithms and tested their performance -word on a (phonetically transcribed) continuous input , using a 33,OOO lexicon. The algorithms postulated potential lexical boundaries: (a) at the ' ' end of each success fully identified word ( segmentation by default ); (b) at each phoneme boundary; (c) at each syllable onset; and (d) at each strong syllable onset (the rhythmic segmentation proposal) . The measureof performance was the number of potential lexical hypothesesgenerated (the fewer the better) . With completely specifiedphonetic input all algorithms naturally performed quite well. However, significant differencesbetweenthe algorithms emerged when some or all of the input was phonetically uncertain; most affected were ' segmentationby default' and the phonemic algorithm , both of which generated huge numbers of potential parses of incomplete input . Far better results were produced by the algorithms which constrained possible word onset positions in some way, and the more specific the constraints, the better the performance: the rhythmic segmentationalgorithm performed best of all with the uncertain input . In the second study, Harrington et al. ( 1989) compared the rhythmic segmentation algorithm with a segmentation algorithm based on permissible phoneme sequences(Lamel and Zue 1984, Har rington et al. 1987), using as a metric the proportion of word boundaries correctly identified in a 145-utterance corpus. With phonetically uncertain , but the rhythmic segmeninput, sequenceconstraints proved virtually useless

A . Cutler I Segmentationproblems

tation algorithm still performed effectively (in fact it correctly detected more word boundaries in uncertain input tharl the phoneme sequenceconstraints had detectedin completely specifiedinput) . The efficiency and robustness of rhythmic segmentation therefore suggest that listeners profit from employing an explicit segmentation procedure of this kind . A striking fact about this procedure, however, is its languagespecificity: as described for English, the procedure is based on stressrhythm , i.e. the opposition of strong and weak syllables. Clearly, it therefore cannot be a universal strategy, becausemany (indeed most) languagesof the world do not have stress rhythm . However, all languages have rhythm - speech rhythm need not be stress-based. In the next section alternative forms of rhythmic segmentation are described, supported by the results from experiments in languageswhich do not have stressrhythm .

6. Rhythmicsegmentation in French and Japanese Mehler (e.g. 1981) and his colleagues(e.g. Segui 1984) have used a variety of psycholinguistic tasks to demonstrate processing advantagesfor syllables in speech comprehension. In one experiment, which launched a series of cross-linguistic comparisons, Mehler et al. ( 1981) had French subjects listen to lists of unrelated words and press a responsekey as fast as possible when they heard a specifiedword -initial sequenceof sounds. This target was either a consonant-vowel (CV) sequencesuch as ba- or a consonant-vowel-consonant (CYC) sequencesuch as hal- . The words which began with the specified sound sequencehad one of two syllabic structures: the initial syllable was either open (CY ), as in balance, or closed (CYC ), as in balcon. Mehler et al. found that responsetime was significantly faster when the target sequence corresponded exactly to the initial syllable of the target-bearing word than when the target sequenceconstituted more or less than the initial syllable. Thus responsesto ba- were faster in balancethan in balcon, whereasresponses to bal- were faster in balcon than in balance. Mehler et al. claimed that this result supported a syllabically basedsegmentationstrategy in speechrecognition in French. Similarly, Segui et al. ( 1981) found that listeners are faster to detect syllable targets than to detect targets corresponding to the individual phonemeswhich make up those samesyllables. Further evidencefrom French that polysyllabic words, whether they are heard in isolation or in connected speech, are analysed syllable by syllable came from studies by Segui ( 1984) and by Dupoux and Mehler ( 1990) .

A. Cutler / Segmentationproblems

If speechsegmentation in French proceedssyllable by syllable, there is an interesting parallel to the results from English reported in the previous section. Just as use of the opposition between strong and weak syllables in ' segmentingEnglish exploits the English languages characteristic stress-based rhythmic pattern, so does use of the syllable in segmenting French exploit rhythmic patterns, since the characteristic rhythm of French is syllable-based. Recent results from studies of speech segmentation in a third language, , confinn the connection between segmentation and speechrhythm . Japanese In Japanese , speech rhythm is based on a subsyllabic unit called the mora which can a vowel, an onset-vowel sequence be ( , or a syllabic coda). Otake et al. ( 1993) conducted an experiment in Japanesewhich was directly analogous to the French experiment by Mehler et al. ( 1981) ; they compared detection of CV (e.g. la-) and CVC (e.g. tan-) targets in Japanesewords beginning with open ( tanishl) versusclosed ( tanshl) syllables. In both words the first mora is ' the initial CV sequence ta ; and detection of CV targets was equally fast in both words (had the Japanesesubjects been using a syllabic segmentation procedure, the CV targets should have been harder to detect in closed than in open syllables) . CVC targets constitute two morae, and correspond to the first two morae of the words with initial closed syllables; however, they do not correspond properly to a mora-based segmentation of words like tanishi (CV-CV -CV) . Indeed, the Japaneselisteners respondedto the CVC targets in words like. tanshi, but usually failed to respond in words like tanishi. Thus rhythmic segmentation seemsto be quite a widespread phenomenon across languages, with the nature of the rhythmic processing being determined ' by the nature of each languages characteristic rhythmic structure: stress-based, syllabic, or moraic rhythm can all be used in speechsegmentation by adult listeners. However, there turn out to be strict limitations on the way any listener can exploit speech rhythm in segmentation; and these limitations may illuminate the questions with which we started this chapter, namely those pertaining to how the prelinguistic infant first solves the segmentationproblem.

7. Limits to rhythmicsegmentation First of all , the cross- linguistic differences in speech segmentation are characteristics of the listeners and do not simply follow from the nature of the speech input . English monolinguals do not use syllabic segmentation in performing the target detection task with either English or French ( Cutler et

A. CutlerI Segmentation problems

al. 1986) ; neither English nor French listeners use moraic segmentationwhen performing the same task with Japanese(Otake et al. 1993). In other words, syllabic segmentationseemsto be specific to French listeners, moraic segmentation to Japanese listeners. (In fact the French listeners segmented both English and Japanesespeechby syllables, just as they segmentFrench!) Moreover, under appropriate conditions listeners can be seen to abandon the rhythmic segmentation procedures characteristic of their language community . When responding very fast, French listeners can base their responseson subsyllabic units (Dupoux 1993) . CVV sequencesare apparently less conducive to application of moraic segmentation by Japaneselisteners than the (more common) CVCV and CVN sequences(Otake 1992). The failure to find processing disadvantages for English words beginning with weak syllables when the words are carefully read, reported in the second section of this paper, may reflect a similar case: if the input is very clear, stress-basedsegmentationmay not need to be called into play. Thus it is quite clear that none of the rhythmic segmentation procedures constitutes an ' absolutely necessarycomponent of adult listeners speechprocessing. The strongestevidencethat this is so comes, however, from studiesof bilingual . Cutler et al. ( 1992) testedFrench-English bilinguals with the techniques processing which had demonstratedsyllabic respondingin French listeners(Mehler et al. 1981) and stress-based responding in English listeners(Cutler and Norris 1988). Their subjectswereas bilingual as they could find - eachhad learnedboth languagesfrom the earlieststagesof acquisition, spokeboth languagesdaily, and was acceptedas a native speakerby monolingualsin eachlanguage. Yet these bilinguals did not necesarilyproduce the pattern of results which monolinguals had shown on each previous experiment. Instead, their response patterns could be predicted from a measureof what Cutler et al. called ' ' language dominance , which amounted in essenceto a decision as to which of their two languagesthe bilinguals would be most sorry to lose. On Mehler et alis target detection task with French materials, only those bilinguals who ' chose French as their ' dominant language showed a syllabic pattern of responding; the English-dominant bilinguals showed no trace of syllabic ' effects. On Cutler and Norris word -spotting task, in contrast, a stress-based responsepattern appeared only with those bilinguals who chose English; the responsesof the French-dominant bilinguals were unaffected by the rhythmic pattern of the embedding nonsense word. Apparently , these maximally competent bilinguals had available to them in these tasks only one rhythmic segmentationprocedure - either that which was characteristic of one of their native languages, or that which was characteristic of the other, but not both.

A. Cutler/ Segmentation problems

Of course, it should be rememberedthat this conclusion is based only on the results of laboratory experiments, and may not reflect the full extent of the resourceswhich bilinguals can apply to the processing of , for example, spontaneous speech; as earlier sections of this chapter described, different ' speechstyles may call differentially upon a listener s processing repertoire. However, the experiments undeniably show that in the laboratory some bilinguals can exploit a given rhythmic segmentation procedure, and do exploit it , while others certainly do not exploit the same procedure, and possibly cannot do so. A claim that , for example, French-dominant FrenchEnglish bilinguals are capable of stress-based segmentation, but abandon it when processing laboratory speech, ought therefore to be accompanied by an account of why Eoglish-dominant bilinguals, and monolingual English speakers, do not abandon this procedure in the laboratory . On the basis of the laboratory results alone, it would surely appear that bilinguals simply do not have available to them the segmentationprocedure characteristic of their non-dominant language. This is a remarkable finding in the light of the undoubted competenceof these bilingual speakersin both their languages. The English-dominant bilinguals spoke and understood French just as well as the French-dominant bilinguals did , and the latter group spoke and understood English just as well as the former. For those bilinguals who used stress-based segmentationwith English, the apparent unavailability of syllabic segmentation for use with French seemed to have no adverse effect on their linguistic competence; likewise, for those bilinguals who used syllabic segmentationwith French, the unavailability of stress-based segmentation seemednot to reduce in any way their demonstrated competencein English. These results may therefore indicate that the rhythmic segmentation procedures are not a necessarycomponent of a languageuser' s processingmechanism; one can demonstrate native competencewithout them. This in turn would imply that rhythmic segmentation procedures are not simply developed in responseto experiencewith the statistical properties of the native language, as the arguments made by, for instance, Cutler and Carter ( 1987) with respectto stress-basedsegmentationin English contended. There is no doubt that stress-based segmentation does work efficiently with English; but despite having been exposedto English since their earliest years, and despite using English with native competenceall their lives, the Frenchdominant bilinguals neverthelessdo not , in the word -spotting experiment, show evidenceof segmenting by stress. The question must be posed, therefore , of how the rhythmic segmentation procedures could arise, if it is

96

A. Cutler/ Segmentation problems

arguably the casethat they may not result automatically from experiencewith the statistical properties of the native language. A possible answer to this question, proposed by Cutler et al. ( 1992) and by Cutler and Mehler ( 1993), is discussedin the next section.

Supposethat the rhythmic segmentati.on proceduresused by adult listeners exist not for purposes of adult processingat all , but are simply traces which remain from a peri.od when the segmentationproblem dominated the infant ' s language processing. Perhaps it is precisely the characteristic rhythm of the input languagewhich offers the infant a first toehold into lexical acquisition, by suggestinga possible segmentation of the continuous speechstream into discrete units. In the caseof the syllable in French, in fact, just such a model has been put forward by Mehler et al. ( 1990) . More generally, Cutler and Mehler ( 1993) have suggestedthat the infant enters the world already armed with what they have called aperiodicity bias' . The task of lexical acquisition is primed in that the infant expects that meaning will map to form ; the task is made possible by the fact that this expectation is targeted towards a particular kind of form : input which is periodically structured. Speechsignals have periodic structure, and for the majority of children speechwill be among the most salient forms of input available. In the first few months and even days of life infants prefer to listen to speech rather than to other auditory input (Colombo and Bundy 1981, Glenn et al. 1981). The contrast between the periodic structure of speech as opposed to random noise is only one level of structure, however; regular periodic structure in speechexists at several levels. At the level of the speechsound, some sounds are relatively ' more periodic' than others - for instance, vowels are relatively steady-state sounds, while consonants are often more transient. As Cutler and Mehler ( 1993) point out , this could account for the findings that infants acquire language-specific vowel prototypes at about six months of age (Kuhl et al. 1992), well before they acquire the consonantal phonology of their language ( Werker and Polka 1993) . At a higher level again is the rhythmic structure of language discussedin the preceding sections. It is not unreasonableto propose that at some point during the infant ' s prelinguistic period this level of rhythmic structure is also attended to , and that the processing that goes on at that point is intimately connected with lexical

A. Cutler / Segmentationproblems

segmentation- dividing the continuous speechinto lexical units. Nor need the rhythmic structure be exclusively expressedin the auditory domain ; as Pettito and Marentette ( 1991) demonstrate, gestural language acquisition by congenitally deaf infants follows a developmentalpath with noticeable similarities to spoken languageacquisition. Because , as we have described, rhythmic structure differs even across spoken languages, the infant exposed to stress rhythm will focus upon a different regularity than the infant exposed to , say, syllabic or moraic rhythm . As Cutler et al. ( 1992) argue, this can be conceived of as the infant attending to the smallest level of regularity occurring in the spoken input . What is remarkable about this processis that it seemsto happen only once, if the evidence from the bilingual studies is reliable. That is, exposure to two differing rhythmic regularities (syllabic and stressrhythm , for instance) does not result in the ability to use both types of rhythm in speechsegmentation; a language user appears to be able to command only one rhythmic segmentation procedure. This type of all-or -none instantiation of a languageprocessing procedure is distinctly reminiscent of the notion of parameter-setting in syntactic processing(e.g. Wexier and Manzini 1987) .

9. Prosodyandthe preHnguisticinfant The notion of rhythmic segmentation by infants as a ' bootstrap' into the beginningsof lexical acquisition has not been directly tested. In this concluding section, however, researchon prosodic processingby prelinguistic infants will be reviewed in an attempt to discover whether there is evidence which could support the notions proposed above. First , though , it should be acknowledged that involvement of speechprosody in this level of language acquisition has been proposed by others in several forms. The suggestionby Gleitman and Wanner ( 1982; seealso Gleitman et al. 1987) that words are initially identified as units via their stressedsyllables is closely related to the application of the present proposal to languages with stress rhythm , for instance. Likewise, Jusczyk (e.g. 1993) has suggestedthat prosodic structure is the dimension which infant listeners exploit to accomplish speechsegmentation . In a series of experiments Jusczyk and his colleagues have shown that infants are sensitive to prosodic marking of syntactic structure, with sensitivity to clausal prosody emerging by four to five months of age (Hirsch -Pasek et al. 1987) , and sensitivity to phrasal prosody by nine months (Jusczyk et al. 1992); Jusczyk interprets this pattern as

A. Cutler/ Segmentation problems

evidenceof infants' exploitation of utterance prosody to structure speechinto interpretable units. Indirect evidencefor the present proposal can be found in both perceptual and production evidencefrom prelinguistic infants. For instance, it has been shown that the characteristic rhythmic pattern of speech is salient to the newborn child. Condon and Sander ( 1974) found that neonatesare able to synchronise their movements with speech structure, whether the speech is spoken directly to the child or played from a tape recorder, and whether it is in the parental languageor a foreign language. ( Tapping sounds, on the other hand, did not invoke synchrony in the infant ' s movement.) The ability to discriminate the contrasts involved in rhythmic patterning appearsearly ; thus two-month-olds can discriminate rhythmic groupings of tones (Demany et ale 1977) . These early discriminatory abilities also apply to the particular contrasts involved in speech rhythm : very young infants can discriminate stresscontrasts (Spring and Dale 1977, Jusczyk and Thompson 1978, Karzon 1985), and neonatescan make discriminations based on number of syllables (Bijeljac-Babic et ale 1993) . Speech to infants tends to have more regular rhythm than speech to adults, as evidenced in English by more frequent occurrenceof stresses (Garnica 1977) and more regular alternation of vocalisation and pause (Stem et ale 1983); however, the relevanceof this is unclear given that durational features of infant -directed speechdo not appear to be involved in infant preferencesfor this speechstyle (Fernald and Kuhl 1987) . More important would seemto be recent evidenceof rhythmic patterning in the speechproduction of prelinguistic infants. Cross-linguistic studies of babbling have pointed to increasing language-specificity in babbling during the secondhalf of the first year of life (e.g. de Boysson- Bardies et al 1984, de Boysson-Bardies and Vihman 1991, Halle et ale 1991, Blake and de BoyssonBardies 1992), including language-specificity in prosodic structure (Whalen et ale 1991) . Rhythmic structure is one of the language-specific patterns which appear in speechat this age. Levitt and Wang ( 1991) and Levitt and Utman ( 1992) found that reduplicative babbling of infants from French-speaking homes showed a gradually increasing regularity of timing of non-final syllables across the first year of life , while the speechof infants of the same age from English-speaking homes showed a gradually increasing variability of syllable structure and timing . This suggeststhat the characteristic rhythm of ' speechis incorporated into infants linguistic competencebefore they acquire their first words. It appears that infants also become aware of the characteristic word prosody of their language before acquiring their first words. Jusczyk et ale

A. Cutler / Segmentationproblems

( 1993) found that nine-month -old infants acquiring English preferred to listen to lists of bisyllabic words with initial stress (crossing, former , cab/e) than bisyllables with final stress (across, before, decay), although no such preferences appeared with six-month-olds. Even when the lists were low-pass filtered to remove most of the segmental information , nine-month -olds still preferred the initial -stress lists, suggestingthat their preferenceswere based on prosodic structure. Jusczyk et ale argued that during the second half of their first year, infants exercisetheir ability to segmentspeechwith the result that they acquire knowledge of the typical prosodic structure of words in the input language. At later ages, language-specific exploitation of rhythmic structure by children is established: children learning English use stressrhythm in segmentation (Gerken 1991, Gerken et ale 1990, Peters 1985) ; children learning French and other languageswith syllable rhythm use syllables (Alegria et ale 1982, Content et ale 1986) ; children learning Japanese use morae (Mann 1986). The hypothesis proposed here is that language rhythm is also what allows infants to accomplish their very first segmentation of speech. An ability to processrhythm is inborn . By using this ability , infants are enabled to overcomethe segmentationproblem and hencetake their first step towards compilation of their very own lexicon.

References . Phoneticanalysisof speechand memorycodesin Alegria, J., E. Pignotand J. Morals, 1982 . Memoryand Cognition10, 451- 456. beginningreaders . The unintelligibilityof speechto children. Journalof Child Bard, E.G. and A. Anderson , 1983 . 10 265 292 , , Language Bard, E.G. andA. Anderson . The unintelligibilityof speechto children: Effectsof referent , 1991 of the Twelfth InternationalCongressof PhoneticSciences , Aix-enProvence availability. Proceedings , Vol. 4, 458- 461. Bard, E.G., R.C. Shillcockand G. Altmann, 1988.The recognitionof wordsafter their acoustic 44, offsetsin spontaneous : Effectsof subsequent context. Perceptionand Psychophysics speech 395- 408. . Cross-linguisticstudyof temporalcharacteristics Barik, H.C., 1977 of differenttypesof speech materials . Languageand Speech20, 116- 126. -Ratner, N., 1984a Bernstein . Patternsof vowelmodificationin mother-child speech . Journalof Child Language11, 557- 578. -Ratner, N., 1984b Bernstein . Phonologicalrule usagein mother-child speech . Journalof Phone tics 12, 245-254. -Ratner, N. and C. Pye, 1984 . Higher pitch in BT is nOI universal: Acousticevidence Bernstein from QuicheMayan. Journalof Child Language11, 515- 522.

A. Cutler / Segmentationproblems

. How do 4- day-old infants categorize Bijeljac-Babic, R., J. Bertonciniand J. Mehler, 1993 ? Developmental multisyllabicutterances Psychology29, 711- 721. -Bardies Blake, J. and B. de Boysson . Patternsin babbling: A cross-linguisticstudy. Journal , 1992 of Child Language19, 51- 74. Blaauw, E., 1991. Phoneticcharacteristics of spontaneous and read-aloud speech . Proceedings of the ESCAWorkshopon Phoneticsand Phonologyof SpeakingStyles, Barcelona , 12.112.5. -Bardies de Boysson . Adaptation to language : Evidencefrom , B. and M.M. Vihman, 1991 and first wordsin four languages . Language67, 297- 319. babbling -Bardies de Boysson . Discernibledifferences in the babblingof , B., L. SagartandC. Durand, 1984 infantsaccordingto targetlanguage . Journalof Child LanguageII , 1- 15. Briscoe . Lexical accessin connectedspeechrecognition . Proceedings of the 27th , E.J., 1989 , Associationfor ComputationalLinguistics , Vancouver , 84- 90. Congress Brown, G., 1977 . Listeningto spokenEnglish. London: Longman. Brown, G.D.A., 1984. A frequencycount of 190,000 words in the London-Lund corpusof . BehaviorResearch Methods, Instrumentationand Computers16, 502Englishconversation 532. Brown, R., 1977 . Introduction. In: C.E. Snow, C.A. Ferguson(eds.), Talking to children: : CambridgeUniversityPress . , 1- 27. Cambridge Languageinput and acquisition Cheshire . Variationin an Englishdialect. Cambridge : CambridgeUniversityPress . , J., 1982 Cole, R.A. and J. Jakimik, 1978 . Understanding : How wordsare heard. In: G. Underwood speech of informationprocessing . , 67- 116. London: AcademicPress (ed.), Strategies . A methodfor the measurement Colombo, J. and R. Bundy, 1981 of infant auditoryselectivity . Infant Behaviorand Development , 4, 219- 233. . The MRC psycholinguistic Coltheart, M., 1981 database . QuarterlyJournalof Experimental Psychology33A, 497- 505. . Synchronydemonstratedbetweenmovementof the Condon, W.S. and L.W. Sander , 1974 . Child Development neonateand adult speech 45, 456- 462. Content, A., R. Kolinsky, J. MoralsandP. Bertelson . Phoneticsegmentation in prereaders : , 1986 Effectof correctiveinformation. Journalof ExperimentalChild Psychology 42, 49- 72. . InvestigatingEnglishstyle. London: Longman. CrystalD . and D. Davy, 1969 . Rhythmiccues to speechsegmentation : Evidencefrom Cutler, A. and S. Butterfield, 1992 . Journalof Memoryand Language31, 218- 236. juncturemisperception . The predominance Cutler, A. and DiM . Carter, 1987 of stronginitial syllablesin the English . ComputerSpeechand Language2, 133- 142. vocabulary . The recognitionof lexicalunits in speech . In: B. de Cutler, A. and JiM. McQueen , in press . Cambridge . Gelder, J. Morals(eds.), From spokento written language , MA : MIT Press . The periodicitybias. Journalof Phonetics21, 103- 108. Cutler, A. and J. Mehler, 1993 . The role of strongsyllablesin segmentation . for lexicalaccess Cutler, A. and DiG. Norris, 1988 : HumanPerceptionand Performance 14, 113- 121. Journalof ExperimentalPsychology 's . The syllable Cutler, A., J. Mehler, D.G. Norris and J. Segui, 1986 differing role in the . of of French and Journal and Memory segmentation English Language25, 385- 400. . The monolingualnature of speech Cutler, A., J. Mehler, D. Norris and J. Segui, 1992 24, 381- 410. segmentation by bilinguals. CognitivePsychology . Rhythmperceptionin early infancy. Nature , L., B. McKenzieand E. Vurpillot, 1977 Demany 266, 718- 719.

A. Cutler / Segmentationproblems

. The time courseof lexicalprocessing : The syllabichypothesisrevisited . In: Dupoux, E., 1993 G.T.M. Altmann, R.C. Shillcock(eds.), Cognitivemodelsof speechprocessing : The Sperlonga . MeetingII , 81- 114. CambridgeMA : MIT Press . Monitoring the lexiconwith normal and compressed : Dupoux, E. and J. Mehler, 1990 speech Frequencyeffectsand the prelexicalcode. Journalof Memoryand Language29, 316- 335. . Baby talk as a simplifiedregister . In : C.E. Snow, C.A. Ferguson(eds.), , C., 1977 Ferguson : CambridgeUniversity , 209-235. Cambridge Talkingto children: Language input andacquisition Press . Fernald, A., 1985 . Four-month-old infants prefer to listen to motherese . Infant Behaviorand 8, 181- 195. Development Fernald, A. and P. Kuhl, 1987 . Acousticdetenninantsof infant preference for motheresespeech . Infant Behaviorand Development10, 279- 293. ' Fernald, A. andT. Simon, 1984 . Expandedintonationcontoursin mothers . speechto newborns 20, 104- 113. Developmental Psychology -Bardiesand I. Furui, 1989 Fernald, A., T. Taeschner .A , J. Dunn, M. Papousek , B. de Boysson ' and fathers' cross-languagestudy of prosodicmodificationsin mothers speechto preverbal infants. Journalof Child Language16, 477- 501. Garnica, 0 ., 1977 . Someprosodicand paralinguisticfeaturesof speechto youngchildren. In: C.E. Snow, C.A. Ferguson(eds.), Talkingto children: Languageinput and acquisition , 63- 88. : . Press CambridgeCambridgeUniversity Gerken, L., 1991 . The metricalbasisfor children's subjectless sentences . Journalof Memoryand Language30, 431- 451. Gerken, L., B. Landauand R.E. Remez . Functionmorphemes in youngchildren's speech , 1990 . and 26 204 216 . , perception production Developmental Psychology Gleitman, L.R. and E. Wanner, 1982.Languageacquisition:The stateof the stateof the art. In: E. Wanner, L.R. Gleitman(eds.), Language : acquisition:Thestateof theart, 3- 48. Cambridge . CambridgeUniversityPress Gleitman, L.R., H. Gleitman, B. Landauand E. Wanner, 1987 . Wherelearningbegins: Initial for languagelearning.In: F. Newmeyer(ed.), TheCambridgelinguisticsurvey, representations : CambridgeUniversityPress 150- 193. Cambridge . Glenn, S.M., C.C. Cunningham and P.F. Joyce, 1981 . A study of auditory preferences in ' nonhandicap . Child Development52, 1303 ped infants and infants with Downs Syndrome . 1307 Grieser . Maternalspeechto infantsin a tonal language : Supportfor , D.L. and P.K. Kuhl, 1988 universalprosodicfeaturesin motherese . Developmental Psychology24, 14- 20. . The recognitionof wordsafter their acousticoffset: Evidenceand implications , F., 1985 Grosjean . Perceptionand Psychophysics 38, 299- 310. -Bardiesand M.M. Vihman, 1991 Halle, P., B. de Boysson . Beginnings of prosodicorganisation : Intonation and duration patternsof disyllablesproducedby Japanese and Frenchinfants. Languageand Speech34, 299- 318. . The applicationof phonemesequence Harrington, J.M., I. Johnsonand M. Cooper, 1987 constraintsto word boundaryidentificationin automatic . , continuousspeechrecognition of the First EuropeanConference on Speech , Edinburgh,Vol. I , 163Proceedings Technology 167. Harrington, J.M., G. Watsonand M. Cooper, 1989.Word boundarydetectionin broadclassand phonemestrings. ComputerSpeechand Language3, 367- 382.

A. CutlerI Segmentation problems . Experimentson the segmentation of an artificial speech Hayes, J.R. and H.H. Clark, 1970 . In : J.R. Hayes(Ed.), Cognitionand the developmentof language , 221- 234. New analogue York: Wiley. Heath, S.B., 1983 . Wayswith words: Language . , life and work in communitiesand classrooms : CambridgeUnivenity Press . Cambridge Hirsch-Pasek , K., D.G. KemlerNelson, P.W. Jusczyk , , K.W. Cassidy , B. Drussand L. Kennedy . Clausesare perceptualunits for younginfants. Cognition26, 269- 286. 1987 . Prosodicdifferentiationof discoursemodes Johns-Lewis, C., 1986 . In: C. Johns-Lewis (ed.), Intonationand discourse , 199-219. London: CroomHelm. Johns-Lewis, C., 1987 . The perceptionof discorusemodes . In: M. Coulthard(ed.), Discussing discourse . Universityof Binningham: DiscourseAnalysisResearch MonographNo. 14, 249271. . How word recognitionmay evolvefrom infant speechrecognitioncapaci, P.W., 1993 Jusczyk : The tien. In: G.TiM. Altmann, R.C. Shillcock(eds.), Cognitivemodelsof speechprocessing . SperlongaMeetingII , 27- 55. CambridgeMA : MIT Press . Perceptionof a phoneticcontrast in multisyllabic , P.W. and E. Thompson , 1978 Jusczyk utterances 23, 105- 109. by,2-month-old infants. Perceptionand Psychophysics . Infants' preference for the predominantstress , P.W., A. Cutler and N.J. Redanz , 1993 Jusczyk . 64 675 687 . of words Child , Development patterns English , L. Kennedy , A. Woodwardand J. Piwoz, , P.W., DiG. KemlerNelson, K. Hirsch-Pasek Jusczyk . Perceptionof acousticcorrelatesof major phrasalunits by young infants. Cognitive 1992 24, 252- 293. Psychology . Discriminationof polysyllabicsequences Karzon, R.G., 1985 by one- to four-month-old infants. . Journalof ExperimentalChild Psychology39, 326- 342 . In: W.D. Manlen-Wilson . Reviewof selectedmodelsof speechperception Klatt, D.H., 1989 . and process , MA : MIT Press , 169-226. Cambridge (ed.), Lexicalrepresentation of readingaloud . Temporalaspects Kowal, S., D. O' Connell, E. A. O' Brienand ET . Bryant, 1975 88, 549- 569. : Threeexperiments . AmericanJournalof Psychology and speaking . Linguistic Kuhl, P.K., K.A. Williams, F. Lacerda , K.N. Stevensand B. Lindblom, 1992 255, 606- 608. alten phoneticperceptionin infantsby six monthsof age. Science experience Press . . Sociolinguistic Labov, W., 1972 , PA : UDiversityof Pennsylvania patterns.Philadelphia within wordsand acrossword . Propertiesof consonantsequences Lamel, L. andV.W. Zue, 1984 and Signal on Acoustics . Proceedings of the 1984InternationalConference boundaries , Speech 42.3.4. 42.3.1 Processing . Theprosodicand paralinguisticfeaturesof reading Levin, H., Schaffer , C.A. andC. Snow, 1982 and tellingstories. Languageand Speech25, 43- 54. of Englishand . From babblingtowardsthe soundsystems Levitt, A.G. and JiG.A. Utman, 1992 French: A longitudinaltwo-casestudy. Journalof Child Language19, 19- 49. . Evidencefor language Levitt, A.G. and Q. Wang, 1991 -specificrhythmic influencesin the . Languageand Speech34, 235infants and of French learning English babbling reduplicative 249. . Someeffectsof semanticand grannnaticalcontexton the productionand Lieberman , P., 1963 . Languageand Speech6, 172- 187. of perception speech . . A computationalanalysisof uniqueness Luce, P.A., 1986 pointsin auditoryword recognition 39, 155- 158. Perceptionand Psychophysics . Cognition24, 65-92. : Therole of readingexperience . Phonological awareness Mann, V.A., 1986

A. Cutler I Segmentationproblems

. of lexicallystreSsed . Theprocessing speech McAllister, J., 1991 syllablesin readandspontaneous 1 26 . 34 Languageand Speech , . . Wordswithin words: Lexicalstatisticsand lexicalaccess , J.M. and A. Cutler, 1992 McQueen on SpokenLanguageProcessing , Banff, of the SecondInternationalConference Proceedings Canada , Vol. I , 221- 224. . PhilosophicalTransactionsof the . The role of syllablesin speechprocessing Mehler, J., 1981 , B295, 333- 352. RoyalSociety : The onsetof . Constrainingmodelsof lexicalaccess Mehler, J., E. Dupouxand J. Segui, 1990 : Psycholinguistic . In: G.TiM. Altmann(ed.), Cognitivemodelsof speechprocessing word recognition . , MA : MIT Press and computationalperspectives , 236- 262. Cambridge ' and J. Segui , 1981.The syllables role in speech Mehler, J., J.-Y. Dommergues , U. Frauenfelder . Journalof VerbalLearningand VerbalBehavior20, 298- 305. segmentation . and readspeech in spontaneous . Detectionof targetphonemes Mehta, G. and A. Cutler, 1988 . 156 31 135 and Language Speech , . . London: Penguin . The psychologyof communication Miller, G.A., 1967 . . Languageand socialnetworks.Oxford: Blackwell Milroy, L., 1980 . Cr,?ss-languageuseof pitch: An ethologicalview. Phonetica40, 1- 18. Ohala, J.J., 1983 . An ethologicalperspectiveon commoncross-languageutilization of FO of Ohala, J.J., 1984 voice. Phonetica41, 1- 16. . . Paperpresentedto the of Japanese . Morae and syllablesin the segmentation Otake, T., 1992 of Psychology , July. XXV InternationalCongress , Brussels in . Mora or syllable? Speechsegmentation Otake, T., G. Hatano, A. Cutlerand J. Mehler, 1993 378 . 358 32 and of . Journal , Language Memory Japanese : Operatingprinciplesfor the perceptionand analysis . Languagesegmentation Peters , A.M., 1985 , Vol. 2: . In: D.I Slobin (ed.), The crosslinguisticstudy of languageacquisition of language . Erlbaum NJ: Hillsdale . 1029 1067 Theoreti~ 1issues , , . Babbling in the manual mode: Evidencefor the Pettito, L.A. and P.F. Marentette , 1991 - 1496. . Science251, 1493 ontogenyof language . Journal of Psycholinguistic of speechin a foreign language . Segmentation Pilon, R., 1981 Research10, 113- 122. utterancesand intonationin spontaneous . Sentence Remez , R.E., PiE. Rubin and S. Ball, 1985 l09th of America Acoustical to the , Meeting, text. Society Paperpresented fluently spoken Austin, Texas, April. . The acquisitionof Kaluli. In: D.I. Siobin(ed.), The crosslinguisticstudy Schieffelin , B.B., 1985 of languageacquisition , Vol. 1: The data, 525- 593. Hilisdale, NJ: Erlbaum. on the transitionfrom prelinguisticto . A culturalperspective Schieffelin , B.B. and E. Ochs, 1983 The transition from prelinguisticto ed. R.M. Golinkoff . In: ( ), linguistic communication . : Erlbaum . London 131 115 communication , linguistic . In : H. Bouma, Segui, J., 1984. The syllable: A basic perceptualunit in speechprocessing . Hillsdale, NJ: 165 181 10 Vol. and . Attention , , D.G. Bouwhuis (eds), performance Erlbaum. . Phonememonitoring, syllablemonitoringand , J., U. Frauenfelderand J. Mehler, 1981 Segui 72 of . BritishJournal Psychology , 471- 477. lexicalaccess . Someprosodiceffectson human word . G. Bard and F. Spensley , 1988 Shillcock, R.CE of SPEECH' 88 (SeventhSymposiumof the . Proceedings recognitionin continuousspeech of Europe), Edinburgh, 819- 826. Federationof AcousticSocieties

A. Cutler/ Segmentation problems . Propertiesof large lexicons: Implicationsfor advanced , D.W. and V.W. Zue, 1982 Shipman isolatedword recognitionsystems . Proceedings of the 1982InternationalConferenceon Acoustics , Speechand SignalProcessing , Paris, 546- 549. . Phonologicalprocess , L. and Z.S. Bond, 1980 es in speechaddressedto children. Shockey Phonetica37, 267- 274. Shute, H.B., 1987 . Vocalpitch in motherese . EducationalPsychology 7, 187-205. . Discriminationof linguisticstressin earlyinfancy. Journalof SpringD .R. and P.S. Dale, 1977 20, 224- 232. Speechand HearingResearch Stem, D.N. S. Spieker . The prosodyof maternalspeech , R.K. Barnettand K. MacKain, 1983 : Infant ageand context-relatedchanges . Journalof Child Language10, 1- 15. Stoel-Gammon,C., 1984 . Phonological . Phonetica41, 208- 214. variabilityin mother-childspeech Sullivan,J.W. and F.D. Horowitz, 1983 . Theeffectsof intonationon infant attention: Theroleof the risingintonationcontour. Journalof Child Language10, 521- 534. Svartvik, J. and R. Quirk, 1980 . A corpusof Englishconversation . Lund: Gleerup. . The babblingof a Thai baby: Echoesand responses , P., 1978 Tuaycharoen to the soundsmade . by adults. In : N. Waterson , C. Snow(eds.), The developmentof communication , 111- 125. Chichester : ~ iley. Wakefield , J.A., E.B. Doughtie and B.-H.L. Yom, 1974 . The identificationof structural in an unknownlanguage . Journalof Psycholinguistic components Research 3, 261- 269. Werker, J.F. and L. Polka, 1993 . Developmental : New challenges changesin speechperception and newdirections . Journalof Phonetics21, 83- 101. Wexier, K. and R. ManziniParameters and learnabilityin bindingtheory. In: T. Roeper , E. Williams(eds.), Parametersetting, 41- 76. Dordrecht: Reidel. Whalen, D.H., A.G. Levitt and Q. Wang, 1991 . Intonationaldifferences betweenthe reduplicative babblingof French- and English-learninginfants. Journalof Child Language18, 501- 516. Wilso~, M.D., 1988 . MRC psycholinguistic database : Machine-usabledictionary, Version2.0. BehaviorResearch Methods, Instrumentationand Computers20, 6- 10.

) 105- 140. North-Holland Lingua92 ( 1994

Domain -general abilities applied to domain -specific tasks: Sensiti vi ty to probabilities in perception , cognition , and * language Michael H . Kelly and Susanne Martin , Philadelphia , PA , 3815WalnutStreet , University of Pennsylvania Department of Psychology -6196 19104 . USA Perceptual and cognitive abilities that are species- and domain-specific may nonethelesshave componentsthat are widespreadacrossspeciesand apply to numerousdomains. For example, all theories of sentenceparsing are constrained by the operations of a limited -capacity memory that is a generalcharacteristic of cognition . This paper discussesanother ability that is general across speciesand, within a species, across numerous cognitive and perceptual domains. We review evidencefrom the animalleaming and human cognitive literature that animals (a) possessfinegrained sensitivity to probabilistic patterns in their environment and (b) usemultiple probabilistic cues to solye particular problems. Such sensitivity is advantageousbecausethe structure of the environment itself can often be characterized as probabilistic . The chancesof successat solving various problems, from foraging to depth perception, would therefore increaseif animals were sensitiveto probabilistic cues and could determine whether multiple cuesconvergeon a solution. We discussthe implications of theseclaims for languageprocessing, and argue that the domaingeneral ability to detect and exploit probabilistic information is brought to bear on numerous language-specific problems.

1. Introduction

A common theme in evolutionary biology concernsthe distinction between inherited and specific adaptations. Both structures may aid an organism, but ' the former derive from the organism s ancestry whereas the latter may have evolved to suit its specific needs. Consider the duck-billed platypus, for example. These strangest of all mammals possess numerous anatomical . Preparationof this paper was supportedby National Institutesof Health Grant I R29 Mehler, and Bob Rescoriafor helpful HD23385to M.H. Kelly. Thanksto AnneCutler, Jacques commentson earlierversionsof this paper. - 3841/94/507.00 @ 1994- ElsevierScienceB.V. All rightsreserved 0024

-E SSDI0024-3841 (93)EO040

-generalabilities M.H. Kelly, S. Martin / Domain

structures that assist their aquatic lifestyle. Thus, waterproof fur helps to insulate the platypus from the cold waters in which it normally swims and its famous bill is actually a highly-developedsenseorgan that is used in foraging. However, the history of these two adaptations is quite different. Whereas the bill seems to be a specific adaptation of the platypus and monotreme relatives, the fur , of course, is a general mammalian trait that the platypus inherits (seeGould , 1991, for further discussion). This trait is general in two ways. First , many speciesshare it and, second, it is relatively general purpose. Thus hair is not only involved in insulation, but also in sensationsof touch and the transduction of sound in the inner ear. As Darwin emphasized in many of his works, the same evolutionary concepts that apply to anatomical structures can also extend to thought and behavior. Thus, some of our own cognitive and perceptual abilities are inherited from our ancestors(and henceit is no coincidencethat we can learn something about, for instance, our visual system by studying the visual systemsof other animals - they are branches from the same tree) . Still other structures might be unique to the evolutionary history of human beings. Language is, of course, the most obvious human duck-bill , and much ink has been spilled over the years in arguments over whether this obvious intuition should really be believed. That controversy will be avoided here. Instead, we will focus on the following general theme: Whatever the evolutionary history of the language faculty , language processing must involve cognitive and perceptual abilities that are domain-general in nature. For an example, consider the parsing literature. Both linguists (e.g., Frazier and Fodor 1978) and psychologists (e.g., Just and Carpenter 1992) have devoted considerable attention to how listeners identify sentenceconstituents and their grammatical relations. Radically different viewpoints are often proposed about how these tasks are accomplished. However, throughout all the controversy, one unifying point stands clear: The parser is fundamentally a slave to a limited capacity short-term (or working ) memory system. Thus, parsing preferences from Kimball ' s ( 1973) ' fast phrase closure' to Frazier and Fpdor ' s ( 1978) ' minimal attachment' are all justified by reference to constraints on shortterm memory capacity. Indeed, potential individual differences in parsing blaseshave recently been related to individual differencesin memory capacity (Just and Carpenter 1992) . In sum, then, a system that is shared by other animals, namely a limited capacity memory, is critically involved in a systemprobably unique to human beings, namely language. Furthermore, an understanding of this ancestral memory system is crucial to a full understanding of how linguistic compe-

M .H . Kelly , S. Martin / Domain-general abilities

tence gets put to use in perfonnance. Finally , since this ancestral system is shared with our non-human relatives, we could gain insights into the human systemthrough memory experimentswith animals as well as experimentswith human beings. We will extend these claims here to another domain potentially shared with non-human animals. This domain is reflected in the old adage that death and taxes are the only certainties in life. In less depressing tenDs, much of the information that animals learn during their lives, and which must be used to guide behavior, is probabilistic in nature. Should the human being bet on the Flyers to win the NHL championship now that they have signed the Michael Jordan of hockey in Eric Lindros ? Is the breeding seasonpromising enough for a bird to raise two clutches of young rather than just one? Will the squirrel find more nuts foraging by this tree or that one? Should a listener who hears ' The fruit flies ...' assumethat ' fruit ' is a noun and ' flies' is a verb? Though these questions vary in their subject matter and importance, they all nonethelesscontain some level of uncertainty, and the types of information that could be used to decide a course of action are probabilistic in nature. Historically , attention to probabilities and their use has been either ignored or looked upon with aversion bordering on plain disgust. Thus, mathematical treatments of probability are relatively recent developments, and were fonnulated in the dubious area of gambling. Even physics, the ' paragon of scientific certainty, has not escapedfrom the clutches of probability , and the probabilistic component inherent to quantum mechanicsdrove Einstein to exclaim that God could simply NOT be a closet gambler. One possible reason for negative views of probability could be an implication that the world is full of mass confusion that cannot lead to anything ' beyond ignorance. However, an animal s environment is not a homogeneous ' ' soup, although it rarely provides sure things . The basic fact of the matter is that the world is awash with stuff best described as ' tendencies', ' maybes', ' estimates' and ' ' , generally-speakings. Furthermore, thesetenDs characterizea wide array of domains, from foraging to parsing, and must be confronted by creatures from aardvarks to zebus. Finally , animals must often make rapid decisions while minimizing the likelihood and severity of errors. These considerations lead to the following conclusion: Animals that have the capacity to detect probabilistic patterns and exploit them will have an advantageover those that do not. Furthermore, confidencein the solution to a problem should increaseif multiple cues converge on that solution. Hence, one might expect to see widespread sensitivity to multiple probabilistic infonnation sources in the animal kingdom . Since many aspects of the

-generalabilities M.H. Kelly, S. Marlin I Domain

environment are characterizedby probabilistic relations among variables, one would also expect this sensitivity to be relatively domain-independent. Indeed, such sensitivity might permeate the activities of specializedabilities, including language. In the remainder of this paper, we will attempt to justify these claims. We will begin by highlighting some phenomena in the area of animal learning and cognition that support the view, perhaps surprising, that rather precise sensitivity to probabilistic information is prevalent in many species. In fact, it seemsto be a rather everyday, ho-hum ability . We will then turn to human perception and cognition , and review some evidencein nonlinguistic domains suggestingthat , like rats, human beings exploit probabilistic information in a variety of tasks. Furthermore, even when sufficient information exists to apparently deduce the solution to a problem, human beings still use additional , probabilistic information that might also be present. We will discuss some of the possible advantages to this strategy. Finally , we will summarize evidencefrom the domain of languageprocessing that also supports widespread sensitivity to probabilistic relations among linguistic variables, as well as the types of tasks that could be assisted by exploiting this information . The researchhere will suggestthat , just as human memory is critically involved in areas ranging from visual pattern recognition to parsing, so the ability to learn and exploit probabilistic information is widespreadin perception and cognition, including traits that might be highly specializedand species-specific. 2. The ' statistical ' rat

Since the cognitive revolution , the once dominant area of animal learning has become of peripheral interest to many experimental psychologists. This major shift of both theoretical and experimental emphasiscould be attributed to a variety of factors, ranging from the development of symbolic computing devices to Chomsky' s critique of stimulus- response accounts of language. Hence, in the cognitive sciencesin general, and languagestudies in particular , traditional investigations of animal learning and their possible implications for human cognition have been virtually , if not completely, ignored. (For some recent exceptions, see Alloy and Tabachnik 1984, Estes 1985, Gluck and Bower 1988.) We believe that this abandonment of animal learning is a sad state of affairs. Certain basic processes of learning are clearly homologous across speciesranging from insects to human beings, and suciJ universality should hardly be of peripheral interest to cognitive scientists. Our

M .H . Kelly , S. Martin I Domain-general abilities

purpose here is not , however, to instigate a general rapprochement between divorced parties. Rather, we will attempt to direct attention to certain fundamental abilities that animals apparently must possessin order to behave the way they do in learning experiments. These abilities can be characterized as a keen sensitivity to the rate at which certain probabilistic events occur in the animal' s environment. 2.1. Behavioral decisionsbasedon detectionof rate information Consider the following situation : On each of a series of trials , a rat is placed in the long axis of a T -shaped maze. If the rat traverses down the correct arm of the T , it will receivea morsel of food as a reward. The correct arm varies from trial to trial , but the right arm is biased to be correct 70% of the time, with the left arm correct 30% of the time. Across a wide range of studies, the results of this situation are clear: The rat will distribute its choices in accord with the particular probabilistic bias establishedby the experimenter . In this case, the rat will travel down the right arm 70% of the time and the left arm 30% of the time. If the probabilities changeto , say, 85- 15 or 6040, then the rat ' s choices will adjust accordingly. The rat ' s behavior reveals precisesensitivity to the rate of reward in the two arms (seeGallistel, 1990, for more extensivediscussion) . ' Despite this sensitivity to rate information , the rat s behavior in this paradigm seemsparadoxical. After all , in order to optimize its food intake, the rat should always choose the side with the higher rate of return. 1 Although this conclusion is true in the laboratory , where the animal facesno competition for the food resource, it is not true in the wild . Supposethat two food patchesexist in a local habitat , with one of them being twice as rich as the other. If all of the animals went to the richer patch, then selection pressureswould favor animals who would exploit the food -poor , but competitorless patch. Evolutionary biologists have argued that , in the long run , the most stable evolutionary strategy would involve dividing one' s time between the two food patchesin accord with their rates of return. Investigations of the spatial distributions of animals while foraging indicates that this strategy is indeed used in the natural environment (seeGallistel , 1990, for summary) . Of course, as Gallistel emphasizes , this strategy presumesthe ability to detect the I For instance, supposethat the probability of reward on the right arm is 75%. In 100trials , the rat on averagewould receivea reward on O.75( 100) + 0.25(0), or 75, of the trials if it always chose the right arm. If it choosesthe arms in accord with their reward probabilities, then the rat can expect reward on 0.75(75) + 0.25(25), or only 62.5 of the trials.

M .H . Kelly , S. Martin I Domain-general abilities

rate of food return in different parts of the environment. Both laboratory studiesand natural observations indicate that a range of species, such as rats, pigeons, ducks, and fish possessthis ability . This sensitivity to probabilistic patterns reveals itself in a variety of behaviors besides the choice of where a food resource might be located. .Furthermore, the pervasivenessof this sensitivity across speciesis shown by the fact that different speciesproduce analogous results in the same task. Consider, for example, the speedwith which an action is performed after a stimulus is perceived. Gradual increasesin the likelihood of a certain event leads to corresponding decreasesin the time neededto react to that event. A notable example in the field of language researchis the inverse relationship between word frequency and the reaction time needed to make judgments about a word. However, numerous events vary in their frequency of occurrence, and animals generally become faster at reacting to more frequent events. Indeed, procedures have been developed to test human beings and other animals in very similar circumstances, and the relationship between frequency of an event and speedof responseis virtually identical across the speciesgroups (Pang et al. 1992) . Such results clearly imply that (a) a learning mechanism exists that permits adaptive adjustments to the likelihoods of different events, (b) the mechanism is domain-general in that it appears for -general, and numerous eventsand behaviors, and (c) the mechanismis species so representsa basic aspect of animal learning, including learning by human beings. 2.2. Catching the contingencies

In the famous learning experiments conducted by Ivan Pavlov, a dog was presented with a neutral stimulus, such as a tone, paired with food. As a result of these pairings, the dog began to salivate when the tone was presentedalone. In the technical terms of the field, an unconditioned stimulus (US), the food , normally elicits an innate unconditioned response (UR ), salivation. By pairing the tone with the food , the former becomesa conditioned stimulus (CS) that produces a conditioned response(CR ), in this case salivation. Thousands of subsequentexperiments, involving a huge assortment of stimuli , responses , and specieshave verified that this type of learning is quite basic throughout the animal kingdom . Extensiveresearchhas tried to understand the mechanismsthat are responsible for learning in the Pavlovian paradigm. We will focus here on one theme that has emergedfrom this research. This theme is, in many ways, reflectedin

M .H . Kel/y , S. Martin I Domain-general abilities

the description of Pavlovian conditioning with which we opened this section. This description, which is commonly used in introductory psychology texts, assumesthat learning dependson explicit pairings betweenan unconditioned stimulus and a conditioned stimulus. Various predictions follow from this assumption. Thus, learning should increase with the number of CS-US pairings, and learning should not occur in the absence of such pairings. However, as Rescorl.a ( 1988) has emphasized, this assumption and its empirical consequencesare untrue. Under certain conditions, even multiple pairings of a CS and US will not causethe animal to assumethat the CS predicts the US. On the other hand, learning can be achieved without a single CS-US pairing . Thus, CS-US pairings are neither sufficient nor necessaryto produce learning. These facts, and the conditions that create them, have led to major revisions in learning theory over the past two decades, and indicate that animals have far more sophisticated learning abilities than were previously imagined. In a series of classic experiments, Rescoria ( 1968) demonstrated that animals are sensitivenot to the raw frequency of CS-US pairings, but rather the general contingency between the variables. In his studies, rats were first trained to press a bar for food. In the next phase, the animals received a number of pairings betweena tone and a mild electric shock. To measurethe extent of learning, the rats were then permit ted to press the food bar once again. During this period, the tone was occasionally presented, and the animal' s behavior monitored. If the animals learned that the tone signals shock, they would decreasetheir levels of bar pressingand hold a motionless posture. Slight levels of learning would produce slight decreasesin the rates of bar pressingwhereashigher levels of learning would produce correspondingly greater suppressionof the bar presses. The critical manipulation in the experiments concerned the likelihood that the shocks in the second phase occurred in the absence of the tone. That is, in addition to the explicit pairings between the tone and shock, a number of other shocks could be administered without a corresponding tone. Traditional learning theory would predict that these additional unpaired shocks would be irrelevant to the learning process since the number of US-CS pairs would be unaltered. However, instead of responding just to explicit pairings of the CS and US, supposethat the rat is sensitiveto the appearanceof the US in the absenceof the CS. If the animal can detect these general contingencies between the shock and tone, then the signaling power of the tone should decreaseas higher rates of independent shocks are encountered. Indeed, a certain point should be reached at which the shock and the tone are statistically indepen-

-generalabilities M.H. Kelly, S. Martin / Domain

dent. That is, given the overall base rate of shocks, one would expect a certain number of tone-shock pairs JUST BY CHANCE . Given that such encountersare mere coincidences, one should make no conclusion about the relevanceof the tone to the shock, and hence the tone should be ignored in the final test phase. Rescorla's results strongly supported the contingency interpretation of learning. As the predictive power of the CS dropped, so did the effectsof the tone on the rats' bar pressing. When the tone and the shock reached statistical independence , the presenceof the tone had no effect on behavior. In sum, Rescorla' s experiments refute the traditional belief, and one still generally held outside of the animal learning field, that the number of CS-US pairings is the crucial variable in learning. Rather, animals are sensitiveto the statistical relationship between the variables. If no statistical relationship exists, then even the presenceof numerous CS-US pairings will not causethe animal to assume that the CS predicts the US. Furthermore, statistical relationships can exist and be learned even in the absence of any CS-US ' pairings. For instance, supposeRescorla s experiment is repeated, but with all of the scheduled tone-shock pairs removed. Thus, the rat would receive a number of shocks, but none accompaniedby the tone. Since no CS-US pairs are presented, the stereotyped view of Pavlovian conditioning would predict that the animal would learn no relation betweenthesevariables. However, the rats do in fact learn something useful in this situation , namely that the presenceof the tone predicts the absenceof the shock. Hence, the rats treat the tone as a safety signal and actually show lessfear reactions when the tone is present than when it is absent (Rescoria 1969) . So far , we have discussedsome evidence that animals are sensitive to the probabilistic relationship betweena single CS and a US. However, numerous experimentsdemonstrate that animals can attend to multiple predictors of an event. For example, suppose that an animal has learned that a tone and a light both signal shock. If the stimuli are now presentedtogether, then, up to a certain point , the animal will exhibit a stronger fear reaction to the joint ' than to the separatepresentation of the stimuli. 2 Thus, an animal s behavior Z The' up to a certainpoint' phraseturnsout to becrucial. If two stimuliarestronglypredictiveof another , canactuallyreduce , evenif coupledwith US presentation , thenrepeated joint presentations of theCSs. As Uebennan( 1990 thereactionto individualpresentations ) pointsout, this apparently . One might expectthat joint makessensefor the followingreason counterintuitive phenomenon the US or strengthof the US. If of an increased the CSs would of probability signal presentation . These are not confirmed theseexpectations , then they must be adjusteddownwardaccordingly of theCSs. in reactionsto theseparate canthenbe observed downwardadjustments presentations

M .H . Kelly , S. Martin / Domain-general abilities

in a particular situation is best predicted by what it has learned about the set of informative cues present in its environment rather than any single cue examined in isolation. Indeed, what an animal learns about a novel stimulus will depend on what it has learned about other stimuli present at the same time. Consider, for example, the following situation. An animal learns that a tone is associatedwith a mild shock. After this relationship is established, the animal now is presentedwith two stimuli - the original tone and a novel light . No shocks are experiencedduring this phrase of compound stimulus presentation . Later, the light can be presented alone, and the animal' s behavior monitored to determine what , if anything, has been learned about the light . A number of possibilities are conceivable. Since the light was never experienced with the shock, one might expect that the former would later be perceivedas a neutral stimulus. Alternatively , some of the fear induced by the tone might spread to the light . However, the animal actually treats the light as a safety signal, and the strength of the safety reaction varies with the strength of the fear associatedwith the tone (seeRescoria and Wagner, 1972, for summary) . In learning about the impact of the light , the animal apparently takes into consideration its knowledge about the tone. Given that the latter normally signals shock, the absenceof the latter becomes associated with the novel ' light . As Rescoria and Wagner ( 1972: 73) state, The effect of a reinforcement or nonreinforcement in changing the associative strength of a stimulus depends-upon the existing associativestrength, not only of that stimulus, but also of other stimuli concurrently present' . In sum, then, numerous experiments in animal learning indicate (a) widespread sensitivity to probabilistic relations betweenevents and (b) the use of multiple cues in identifying the probable course of imminent events, and the types of behaviors neededto deal with them. Nonetheless, we must emphasize that even though animals behave in accordance with statistical patterns in their environment, this behavior does not lead to strong conclusions about the underlying mechanismsthat support it . Animals might reveal sensitivity to statistical patterns without actually representingor manipulating probabilistic data. For example, each learning trial could lead to the adjustment of the association strength between a CS and a US. This association strength would be increasedwhen the US appearswith the CS and decreasedwhen the US appears without the CS. Over time, the probability of a CS-US pairing would become more and more correlated with the association strength, though the probability itself would not be representedby the animal. Thus variations in the probabilities of different eventscould produce behavior that is attuned to those probabilities without entailing that these values are

M .H . Kelly , S. Martin / Domain-general abilities

explicitly represented or that the animal even remembers each individual learning trial . In fact, the most influential model of Pavlovian conditioning over the past two decadesdoes reproduce apparent sensitivi~;Y to statistical patterns by using trial by trial updates of association strengths betweena US and potential CSs (Rescoria and Wagner 1972) . But regardlessof the exact nature of the mechanismsthat underlie it , animals clearly have evolved some form of sensitivity to probabilistic patterns in their environment, and this sensitivity is finely tuned to the actual probabilities of various events.

2.3. Someimplications for languagelearning Let us translate some of theseanimal learning studies into a more familiar setting, though with controversial implications. During languageacquisition, children must learn the subcategorization privileges of the verbs in their ' ' language. Thus, English children must learn that give can appear in both ' ' prepositional (e.g., John gave the money to the charity ) and double object ' ' ' dative structures (' John gave the charity the money ), but that donate can only appear in the prepositional form . How do children learn these relationships ? The ' give' caseseemsstraightforward in that the child will observe the ' verb in both structures and conclude that ' give is legal in those frames. Thus, the child makes conclusions based on positive evidence. However, caseslike ' donate' seemto create more problems for languageacquisition. One possibility is that general principles of learning could apply to the linguistic domain. ' ' Thus, if children get explicit reinforcement that donate is ungrammatical in the double object form , then such negative evidencecould help them to learn ' ' syntactic restrictions on donate . However, numerous studies have documented that children do not get such explicit negative evidence(see Pinker, 1989, for review) . Given the lack of reinforcement coupled with successful learning, the dominant conclusion has been that general principles of learning will not apply to languageacquisition. This argument rests on incorrect assumptions about current theoretical views in animal learning, or the experimentson which those views are based. As the preceding discussion has emphasized, animals can readily learn not only that tones predict the presenceof shock, but also that tones can predict the absenceof shock. They need no additional feedback for justification or refutation of their conclusions. If learning is considered, at least in part , as the detection of statistical regularities in the environment, then children might ' learn that ' donate systematically predicts the absenceof double-object structures in the same manner as rats can learn that tones systematically predict

-generalabilities M.H. Kelly, S. Martin I Domain

the absence of shocks. Just as a species-general memory system can be -general learning abilities, in critically involved in languageparsing, so species this casethe ability to detect the statistical texture of the environment, could be involved in languagelearning. Referencesto statiscial sensitivity have recently grown in critiques of the negative evidence problem (e.g., Liberman 1991). However, we believe that prima facie incredulity over the potential of this approach, or even the mere existenceof the statistical sensitivity, would decline if animal learning studies such as those reviewedabove were more seriously consideredin discussionsof language learning. The evidence from the animal literature is now quite conclusive in that speciesafter specieshas been found not only sensitive to probabilistic information , but even to the precise quantitative nature of this information (bearing in mind once again that the exact mechanismsresponsible for this sensitivity must be identified) . Furthermore, they show such sensitivity in a variety of learning situations that use quite arbitrary stimuli , suggestingthat the ability is domain-general, and not tied to information that the animal is biologically prepared to learn. 2.4. Objections

Numerous objections could, of course, be raised against the conjectureswe have just made. We do not pretend to have definitive answers to these problems, but the problems themselveshave not beenworked out in sufficient detail to determine their scope or their resistanceto repair. ' 2.4.1. How to define what s missing? For a language learner to determine that a verb has not appeared in a structure, and that this absenceis statistically meaningful, the nature of the expected structure must be defined. As Pinker ( 1989) has pointed out , the structures cannot be individual sentencetokens, becausethere are an infinite number of such structures, and the odds that one of them would be encounteredin the corpus of speechheard by the learner is vanishingly small. ' Thus, the items that enter into the learner s statistical tabulations must be . Hence reference to the statistical more abstract than individual sentences structure of the environment will not be sufficient to characterize learning since the nature of the environment is partly defined by how the learner categorizes items within it . Furthermore, these categories may in fact be intrinsically linguistic in nature. Hence we are back to a basic problem in language learning : What notions about the nature of language do children

-generalabilities M.H. Kelly, S. Martin / Domain

bring to the task at hand? Referenceto statistical regularitieshas not eliminated this problem at all. We believe this characterization of the learning task is correct, but note that it is not specificto languagelearning. The useof statistical regularities by animals also depends on abstracting away from individual tokens. Furthermore , the categoriesinvolved in this abstraction processmay depend on the learning domain in question. For example, suppose a rat learns that a tone signals the absenceof shock. Each instance of a tone is a unique event, and yet the animal must abstract away from this uniquenessto similarities across the various tokens. What counts as similar may depend on the nature of the auditory system just as what counts as similar in language learning may depend on the nature of a species-specificlinguistic system. As we emphasized at the beginning of this paper, the use of some general purpose abilities (e.g., memory) in a particular task (e.g., parsing) does not preclude the important role of domain-specific factors. A complete account of the task will require a specification of the general and the specific, along with how they interact. 2.4.2. Outcomesimilarity despiteinput variability Given that languagelearning will depend in part on the statistical structure of the input , why is the outcome of the learning so robust in the sensethat all normal children attain the same basic level of competencedespite (probably) wide ' variations in what they have heard? One answer could focus on the domain-specific languagefaculty , which makescertain assumptionsabout the nature of the language (in statistical terms, the population) from which the sample was drawn (e.g., all of the sentencesin the language are either right branching or left-branching) . We would also suggesta statistical answer as an accompaniment. In particular , the learning is robust becauseof the robust nature of statistical sampling. One of the fundamental theoremsof statistics is that as the size of a sample increases, it reflects the actual nature of the population from which it was drawn with greater accuracy. Hence, one can make reliable conclusions about the population (which is too large to be examined in its entirety) from a relatively small sample. Although no two samples from , such a population will be alike, they demonstrably will 3 convergeon the sameconclusionsgiven a sufficiently large sample size. Note 3 Here is an easy way to dernonstrate this statistical principle : Instruct your cornputer to generate sarnples of randorn nurnbers between 0 and 100. Start off with two sarnples. each containing 5 cases. The rneans of these sarnples will probably differ greatly. Now continue generating sarnple pairs. but increasethe nurnber of casesin each pair . As the nurnber of cases

-general M.H. Kelly, S. Martin/ Domain abilities

117

that such an account would predict that the greatest variability in linguistic judgments should occur for relatively rare constructions, where the idiosyncrasies of the sample will have a much greater impact on learning. 2.4.3. Going beyondidiosyncrasies Instead of relying on some general learning mechanism, such as one that involves feedback regarding ungrammatical sentences , children might be able to exploit regularities about the sentenceframes that verbs could take (Pinker 1989). For example, the absenceof ' donate' from double-object datives is not an isolated fact. Rather, polysyllabic English verbs and verbs with certain stresspatterns are, in general, barred from double-object dative structures. If children could learn such regularities, they could then generalizethem to new instances. In fact, Gropen et al. ( 1989) have found that children learning English were re.luctant to generalize polysyllabic nonsense verbs from a propositional dative to a double-object dative, though they were more willing to do so for monosyllabic nonsenseverbs. This effect of syllable number did not extend to other syntactic frames, suggestingthat the children had learned about the blasesconcerning the English dative. Although Pinker ( 1989) casts the general learning and regularities approaches as conflicting alternatives, we see them as quite similar. Englishspeakingchildren may in fact detect regularities in the structure of English by being sensitive to the systematic absenceof certain verbs from double-object datives. They would , of course, have to recognize dative structures despite variations in their lexical content, and they would have to recognize that certain phonological properties of verbs are correlated with the dative structures . As noted above, some of these classification abilities may derive from characteristicsof the child ' s native linguistic capacity. However, granted that theseclassification abilities exist, the results found by Gropen et al. ( 1989) are quite consistent with the claim that children are sensitive to statistical patterns in their linguistic environment. The children in the Gropen et al. , rather than discrete. experiments were also probabilistic in their responses That is, although they were biased against generalizing polysyllabic words to double-object datives, they nonethelessdid so. This pattern follows from the fact that the English restriction is also probabilistic in nature (e.g., ' offer' can appear in the double-object dative) . Given a probabilistic input , a probabilistic output is expected. This leads to the next point . , convergingon a valueof 50. Yet getslarger, the meansof the sampleswill moveclosertogether the sampleswill obviouslydiffer in their exactsequence of numbers .

-generalahiliti P.~ M.H. Kelly, S. Marlin I Domain

2.4.4. Probabilistic responsesinherent to the useof statistical patterns Two items might both be members of a population , but the use of finite samplesto infer characteristicsof the population could lead to some differences in how they are treated. Confidencein the membershipof item I might be lessthan confidencein the membership of item 2. When applied to language, this aspectof statistics means that two sentencescould, in fact, be grammatical in the language, but speakersmight differ slightly in their opinions. One of the sentencesmight be judged more strange than the other, or rated as grammatical more slowly than the other. In fact, when subjectsare asked to rate sentenceson scalesof grammaticality, they find the task fairly easy, they do rate items as somewherebetween grammatical and ungrammatical, and they agree among themselveson these ratings (e.g., Fisher et al. 1991) . 2.4.5. Contradictory evidencefrom the domain of human decisionmaking Over the past two decades, voluminous evidence has been gathered that human beings are actually very poor in statistical reasoning (seeNisbett and Ross, 1980, for a summary) . For example, they often ignore base rate information , have incorrect ideas about randomness, and do not realize that large samples provide better estimates of a population than small samples. Given thesewell-documented problems, how could human beings use statistical information as an effective learning tool ? We will offer two parries to this objection. First , lack of explicit understanding of statistical principles does not entail lack of some type of richer implicit knowledge. This knowledge might be relatively inaccessibleto conscious access , and yet still be useful in perception and action. For instance, human being have a very poor understanding of physics when problems are presentedin a verbal format (e.g., McCloskey 1983) . Thus, they often believe that certain impossible descriptions of moving objects are, in fact, accurate statements. However, when presentedwith computer-generateddynamic displays , they now correctly recognize the natural and unnatural motions (Shannon 1976) . Of course, we normally base our actions on visual perception of actual motion rather than verbal descriptions of it . Second, the manner in w~ich the statistical information is presented in the decisionmaking experiments differs markedly from the way that information is normally experienced. In the former , information about base rate (i.e., frequency of occurrence) is given as an explicit sum. Thus, subjectsmight be told that a city contains 100cabs, with 80 being blue and 20 green. However, we rarely receive explicit information about frequency, and animals literally never receivesummary data tables. Instead, most of the frequency knowledge

M.H. Kelly, S. Martin / Domain-generalahiliti2.!

that we and other animals have is probably obtained through numerous encounters with individual objects or events. Hence, we cannot use experiments in decision making as a strong basis for conclusions about human sensitivity to statistical patterns, especially since studies have found differences in the way people treat the samedata when presentedin summary format or as a seriesof events(e.g., Wassermanand Shaklee 1984) . Rather, we must examine direct investigations of human sensitivity to frequency information , and the potential roles of such knowledge in solving problems in language and other domains. We turn to theseissuesin the following sections.

3. The useof multipleandprobabilisticinformationsourcesby humanbeings As the precedingdiscussionillustrates, non-human animals have well-documented ' abilities to detect the statistical texture of their environment. Such statistical sensitivity has been documented in numerous species and in a variety of settings, ranging from highly constrained laboratory environments to open field situations. The evolution of these abilities along with their speciesand domain generality should not be surprising given their obvious use in a variety of tasks such as foraging. In this section, we will turn to human beings and summarize some of the evidencethat we also are sensitive to and rely extensivelyon probabilistic information from multiple sourcesin order to solve a variety of problems. This section will focus primarily on nonlinguistic domains, with the next section turning to implications for language processing. The topics that we will discuss here should not be considered exHaustive, but instead viewed as a sampling from the available evidence. 3.1. General sensitivity to frequency information

In order to effectively use probabilistic infonnation , human beings must, of course, be sensitive to the relative frequencies of various events. Numerous experimentsacrossthe past four decadeshave establishedthat human beings possess this prerequisite ability (though controversy continues over the mechanismsthat might underlie it , cf. Hintzman 1988, Jonides and Jones 1992) . Furthennore, this frequency sensitivity seemsto be general in nature, as it extends across numerous domains (see Hasher and Zacks, 1984, for a more detailed review) . For instance, studies have shown that people are quite sensitive to the relative frequencies of right and left -handers (Coren and Porac 1977), encounters with different people (Saege~ et al. 1973), the

M .H . Kelly , S. Martin / Domain-general abilities

lethality of different events (Lichenstein et al. 1978), and even fast-food restaurant chains (Shedier et al. 1985) . This frequency sensitivity appears not only across domains, but also across tasks. Whereas some studies explicitly ask subjects to rate items in terms of frequency, others use more indirect probes. For instance, in numerous experiments, Zajonc ( 1968) has found that the more frequently something is experienced, the more it is liked (even though the subjectshave no control over theseexperiences ) . Hence frequency sensitivity can be revealedby subjective evaluations. As with fast-food restaurants, numerous experiments have shown that human beings are sensitive to the relative frequencies of many linguistic events, such as words (Shapiro 1969), syllables (Rubin 1974), and letters (Attneave 1953). 4 Indeed, this sensitivity to frequency is quite strong. For example, Attneave asked subjectsto estimate the frequency of English letters per thousand tokens. He found that fully three-quarters of the variance in the ' subjects responsescould be accounted for by the actual frequencies of the letters in English text. Using different techniquesand word frequency as the object of study, Shapiro ( 1969) found that over 85% of the variance in judgments of word frequency could be accounted for by measuresof actual 5 frequency. In fact, this frequency sensitivity might be even stronger than these experimentsindicate, given that our measuresof actual frequency are subject to typical errors of sampling. These errors will be particularly apparent for low frequency items. For example, Gernsbacher ( 1984) points out that the words ' boxer' , ' icing' , and 'joker ' have the samefrequencies(namely, one per million ) as ' Loire ' , ' gnome' , and ' assay' in the two most widely used frequency norms for English. However, there is clearly no doubt that the former words are ' really' more commonly encountered, and that a sufficiently large sample of English would confirm these intuitions . In fact, the validity of the 4 Thesevariablesare, of course , correlatedin that relativelycommonsyllablestend to contain . However . , the variablescan be statisticallyseparated relativelyfrequentlettersand phonemes Thus, Rubin ( 1974 aresensitiveto syllablefrequencyindependently ) foundthat Englishspeakers of phonemefrequency . 5 Thestrengthof thesecorrelationscan, of course , beinfluencedby the rangeof materialsused. ' and ' zinc' were If only the words' the' , ' house , judged, the correlationbetweenratedand actual frequencywould obviously(and trivially) be perfect. SinceAttneaveusedall the lettersof the , the resultsfrom his experimenton letter frequencycannot be criticizedas Englishalphabet artifactsarisingfrom the typesof lettersselected . Shapiro( 1969 ) used60 words for his adult . Using455 , and theserangedfrom very high throughintermediateto low frequencies subjects wordsin a lexicaldecisiontask, Gemsbacher found that ratedfamiliarity accountedfor ( 1984 ) . morethan 71% of the variancein reactiontime.

-generalabilities M.H. Kelly, S. Martin / Domain

intuitions in this casesuggeststhat human ratings of frequency might in fact be more useful in experiments than listings in word frequency norms, and Gemsbacher ( 1984) found that subjects agreed quite well on their ratings of word familiarity , and that these ratings were better predictors of reaction times in a lexical decision task than supposedmeasuresof ' actual' frequency. Given that human beings are sensitive to frequency, sampling considerations might lead one to expect that their frequency judgments would be more accurate than current frequency norms. After all , the Francis and Kucera frequency norms are based on an analysis of one million word tokens, which is a relatively small sample. Given conservative assumptions (speech/reading rates of ISO words per minute along with an eight-hour sample per day), a typical person will be exposedto a million -word sample in about two weeks. In sum, human beings are sensitiveto frequency information in a variety of domains. This domain-independencesuggeststhat this sensitivity is a general ability and hence might be exploited in a variety of domain-specific tasks. Nonetheless, we must emphasizethat human beings are far from infallible in their frequency judgments. For example, .they tend to overestimate the frequency of rare events and underestimate the frequency of common events (seeBaron, 1988, for a summary) . In addition , frequency estimatesare based at least in part on factors other than frequency. For instance, one category is often consideredto be larger than another if instancesof the former class are more easily retrieved from memory. Now , these retrieval effects might be correlated with actual frequency, and so be reasonably accurate. However, they also might reflect aspects of memory organization itself. Thus, people ' ' generally and falsely believe that more English words begin with r than have ' r ' as the third letter. Theseintuitions probably reflect the fact that the mental lexicon is organized by first rather than third letter and/ or phoneme, and henceexemplars of the former class are more easily retrieved from memory. however, despite these inconsistencies, the evidence as a whole supports the view that human beings, like other animals, can learn probabilistic patterns in their environment. We will now turn to a sample of some areas where multiple , probabilistic information sourcesare available and might be used to solve particular problems in perception and cognition . 3.2. Depth perception One of the fundamental questions in the history of visual perception concerns how we perceive a three-dimensional world based on a two- dimensional image. Centuries of research on this question have focused on two

-generalabilities M.H. Kelly, S. Martin I Domain

issues, one optical and the other psychological. First , what types of optical patterns exist in the two-dimensional image that are correlated with the threedimensional world that reflected the image to the eye? Second, which of these information sourcesare actually used in visual perception, and how are they weighed? The second question becomes particularly important given that numerous potential cues to depth exist, such as those listed in table I .

TableI Somevariablesthat arecorrelatedwith distance Variable

Description

Binocular disparity Motion parallax Accommodation Relative size Height in picture plane Texture gradients Occlusion

Differencesin the image of an objected projected to the two eyes The relative velocity of imagesacross the retina Changesin the shape of the lens The size of an image on the retina The location of an image along the vertical axis The e , shape, and densityof textureelementslike pebbles , tiles on a floor An object blocking the view of a more distant object

Of course, these cues vary in their range and predictive power. Thus accommodation is not effective beyond 10 feet, and height in the projection ' place can easily be violated (e.g., a person s head will be higher in the projection place than the feet, yet is not further away from an observer) . Nonetheless, many cuesare simultaneouslyavailable, and hencean important issueconcernshow the various cuesare weighedand whether an available cue might even be used. The situation could be complicated further becausethe weights given to a set of cuesmight vary with context. Consider, for example, the caseof textural cuesto depth. Many surfacesconsist of textured elements, such as pebbles along a path or tiles in a room , that have roughly the same size and shapeand are distributed randomly . However, becausethe elements can systematically vary in their distance from an observer, their projected imageswill vary also. As distance from the observer increases, the images of the texture elements will decreasein size, compress in shape, and become more densely packed. Investigations of these variables have found that human observers(a) are sensitive to all three dimensions, but (b) respond to some dimensions more strongly than others, and (c) change their weightings with context (Cutting and Millard 1984). In particular , for flat horizontal surfaces, the size gradient had the greatest effect in creating impressions of depth, followed by the density and shape gradients. However, for curved surfaces, the shapegradient dominated over the other variables.

M .H . Kelly , S. Martin / Domain-general abilities

Given that cues vary in their effectiveness , one might consider the extreme whether some cues case, and inquire , though present, might receivea weight of zero, and hence be totally ignored. Such total neglect might appear most clearly when another more powerful cue is present. In such a situation , visual processing might engage in what Bruno and Cutting ( 1988: 162) call cue ' selection' in which ' observers ... use the , single most effective available ' . For instance a static information source and the others , , pictorial [ ] disregard cue to depth, like height in the image plane, might become ignored when motion parallax is present, given that the former is lessconsistently predictive of distance. However, using a variety of experimental methods, Bruno and Cutting examined the impact of motion parallax, height in the image plane, occlusion, and relative size on distancejudgments, and found that all of these variables contributed significantly to those judgments. Such results indicate that human beings use a variety of cuesin three-dimensional perception, even cuesthat are not completely reliable.

3.3. Categorization When subjectsare asked to judge the truth fulness of category membership statementslike ' An x is a V ' , where Y is a category term and x is a possible member, their time to respond consistently varies depending on x (cf. Smith and Medin , 1981, for summary) . For example, subjects take longer to ' ' ' ' ' ' respond true to An ostrich is a bird than to A robin is a bird . These reaction time patterns can be predicted by examining the extent to which an es features typically associatedwith the category in question instance possess e. . Rosch and Mervis 1975). In the case of birds, these might be flight ( g, capacity, relatively small size, song, etc. This interpretation of the reaction time phenomenaindicates that numerous information sourcesare used in the processof categorization. Indeed, some have claimed that the reaction time (and other) data indicate that membership in categories is not discrete, but rather continuous, or that human beings believe that category membership is usually fuzzy. However, subsequentresearchhas cast serious doubt on these conclusions. For example, the status of an integer as even or odd is clearly discrete. A simple, straightforward definition sufficesto unambiguously classify any integer as even or odd. If the classification decision data reflect the fact that human beings typically lack knowledge of category definitions and/ or believe that category membershipis generally fuzzy, then the decision data should look rather different if subjects have to classify integers as even or odd. If subjects know the mathematical definition and explicitly admit that

M .H . Kelly , S. Marlin I Domain-general abililies

category membership is completely and unambiguously determined by that definition , then all integers should be classified as even or odd with equal speed, so long as confounding variables like word frequency are control led. However, Arm strong et al. ( 1983) found that some numbers are in fact classified as even or odd faster than other numbers. Indeed, for a range of categoriesthat had clear, agreed-upon criteria for membership, Arm strong et al. found that some members were consistently classified faster than others. Given these results, Arm strong et al. argue that , whatever the reaction time data indicate, they do not imply that items are members of categories to varying degrees, or that people believe in graded category membership. We agree completely with the rationale for the Arm strong et al. experiments and their conclusions. However, we nonetheless believe that the reaction time data reflect certain aspectsof the processof categorization. In particular , even when a criterion exists that is necessaryand sufficient for category membership, people cannot help but weigh other factors that are correlated with membership, though in the final analysis such features are irrelevant. For instance, prime numbers are almost always odd (2 being the only counterexample). This feature is, however, neither necessarynor sufficient for being classified as odd. Still , the feature might be weighed in classification, with prime numbers being classified as odd faster than nonprime . Consideration of such other factors might be beneficial by speedingup decision times while keeping error rates low. This advantage might also be illustrated in the following phenomenon. 3.4. The word superiority effect

' ' Suppose subjects view a clearly printed letter k either by itself or in a ' word , and they must identify whether a ' k or some other letter was presented. Numerous experiments using variations on this task have repeatedly found that subjectsidentify the target letter more accurately if it appears in a word than if it appears by itself (e.g., Reicher 1969, Johnston 1978) . ' ' Thus, ' k ' would be identified more accurately in the context of work than alone (or in a string of letters that do not form a word ) . Why is the word context advantageouseven though the letter in isolation is clearly printed ? One possibility concerns the high speedswith which skilled readers process orthographic material. Given that the context in which a letter appears is perceptual partly predictive of its identity , observers could bypass complete ' encoding of a letter, and use partial information about a letter s appearance plus partial information about the identity of surrounding letters to speed

-generalabilities M.H. Kelly, S. Martin I Domain

overall processing while minimizing error rates. Thus, observers could circumvent the usual tradeoff between speed and accuracy by using multiple sources of information about letter identity . Of course, this additional information is probabilistic in nature, given that words other than ' work ' ' ' ' ' ' ' begin with wor , such as word and worm . Nonetheless, human beings are apparently sensitive to this probabilistic information , and use it to increase processingspeed.

4. Probabilisticinformationandlanguage process ina In this section, we will review some evidence that multiple , probabilistic information sourcesare used to solve various tasks in language processing. We will focus on phoneme perception, word boundary identification , and the assignmentof words to grammatical categories. Other potential examplesalso exist, such as word identification in sentencecontexts and assignmentsof words to agent and patient roles. SeeMassaro ( 1991) for discussion of these and other cases, along with mathematical models of some of these phenomena . 4.1. Speech perception

As in the case of visual depth perception, numerous acoustic cues for phonemeidentification exist. Indeed, Lisker ( 1978) lists fully sixteen variables that could be used by listeners to determine whether a / b/ or /p/ is present in ' ' . Furthermore numerous rud , experimentshave shown that listenersweigh in variables multiple speechperception. Many of these experiments exploit the possibility of trading relationships between different acoustic cues for phonemecategories. In particular , supposethat the values along two acoustic dimensions could be used to distinguish between two phonemesA and B. If these two dimensions are in fact weighed by the listener, then shifts in the value of one dimension toward the B category should be offset by compensatory changestoward the A category in the value of the other dim ~nsion. This paradigm has been used to document the existence of a variety of trading relationships. For example, both voice onset time ( VOT) and r~lative aspiration are potential cues for distinguishing voiced from voicelessstop consonants . With amplitude of aspiration held constant, a particular VOT value can be identified as the approximate boundary betweena voicelessand voiced stop consonant. However, if the aspiration amplitude is then raised, thus

-generalabilities M.H. Kelly, S. Martin I Domain

increasing the evidence for an unvoiced consonant, the VaT boundary betweenthe two phoneme classeswill shift toward shorter VaTs . Hence, an ' ' aspiration value that indicates unvoiced can be offset by a suitable VaT ' ' indicating voiced . Such results indicate that both VaT and aspiration are weighed in distinguishing voiced from voiceless consonants (Repp 1979) . Similar trading relations have been found for formant transition duration and the duration of the following vowel, which are used to distinguish jbj from jwj (Miller and Liberman 1979), silence duration and Fl onset frequency, ' which are involved in the distinction between ' say' and ' stay (Best et al. 1981), and many other contrasts (seeRepp, 1982, for summary) . Furthermore experimentsindicate that infants as well as adults weigh multiple variables in speechcategorization (Miller and Elmas 1983) . The huge mass of this speech ' perception evidence leads to the conclusion that listeners will make use of ' any cue for a given phonetic distinction (Repp and Liberman 1987: 98) . Even further , listeners will combine multiple cues to identify phonetic segments. The search for multiple information sourcesin speechperception has been driven in large part by the failure to find invariant relationships between acoustic structure and phoneme categorization. If such invariants did exist, then they would presumably be sufficient in themselvesto distinguish between phonemes. Other, probabilistic cues, though present, might then be considered superfluous. One possible invariant that has been identified and explored over the past decadeinvolves the spectral shapeof a stop consonant release burst. Blumstein and Stevens ( 1979) have argued that place of articulation distinctions among stop consonants are invariantly signaled by this variable. For example, alveolar stops exhibit a gradual rise in amplitude as frequency increases, whereas labial stops show a falling or flat amplitude pattern. Experiments have, in fact, shown that listeners are sensitive to this information (Blumstein and Stevens1980) . However, even if this information is invariant , its presenceapparently does not eliminate the effects of other, probabilistic cues for stop consonant identity . Indeed, when the invariant and probabilistic information conflict , the latter dominates (Walley and Carrell 1983). These results are analogous to the literature in visual depth perception, in which probabilistic cues to depth are exploited even if more powerful cues are also present. They also emphasize the danger of assuming that the measuredstrength of a cue will map directly onto cue weights in perception. 4.2. Identification of word boundaries Identifying

word boundaries in continuous

speech is one of the major

M.H. Kelly, S. Martin / Domain-generalabilities

problems in speechperception, languageacquisition, and the development of speechrecognition devices. In contrast to our intuitions regarding our native language, invariant cuesto word segmentationdo not appear to exist. Hence, our impressionsof an unfamiliar languagemight better capture the true state of affairs in the sound stream: Words seemto be tightly knit together, with no obvious seamsthat can be used to teasethem apart . In fact, errors in word ' segmentationoccur in learning and listening to one s native language, testifying to the difficulties involved in this task. Segmentationproblems are wellattested in the acquisition literature (e.g., Gleitman et al. 1988), and uncertainties ' ' ' over whether ' a napron or an apron was spoken may have caused the former to have lost its initial / n/ over the course of English history . Despite its clear difficulties, listeners nonethelessbecome quite skilled at word segmentation, and numerous researchershave tried to identify the types of cues listeners can and do use to solve this problem. These investigations have uncovered cues that exist at various levels of language structure, from constraints on phoneme sequencesto patterns in the prosody. None of these cuescan guarantee successin word segmentaition, either alone or in concert, but their joint effectsmay conspire to make word segmentationby and large successful. 4.2.1. Phonemesequences Languagesoften have restrictions on phonemesand/ or phoneme sequences at syllable, morpheme, and word boundaries. Some of these restrictions are strong constraints, such as the impossibility of obstruent + nasal sequences within English syllables. However, large-scale corpora analyses might also reveal probabilistic relationships between the distributions of phonemesand various boundary types. Speakersdo appear to have some knowledge of such probabilistic relations. Thus, Cutler et al. ( 1987) have shown that English speakers have learned that English consonant vowel sequencesare more likely to follow a CVCV than a CVCC pattern . Experiments on human sensitivity to frequency information have found that people are not only sensitiveto the frequenciesof individual letters (Attneave 1953) and syllables (Rubin 1974), but also to letter combinations (Underwood 1983) . Given that letter combinations will be correlated with phoneme combinations, such data indicate that speakers also have knowledge of the relative frequencies of . However, experiments need to be performed to see if phoneme sequences not just to the overall frequency of phoneme sequences sensitive , are speakers but also to significant interactions between those frequencies and syllable, morpheme, and word boundaries.

-generalabilities M.H. Kelly, S. Martin / Domain

4.2.2. Syllable cues Individual syllables may vary in the likelihood with which they signal word boundaries. In a recent analysis, we examined four common English syllables - / 9~/, / b /, / m/, and / ~/ - and estimated their probability of occurring before a word boundary . For example, / ~/ occurs before a word boundary in ' a . maze' and ' Louisiana cooking' , but not in ' amaze' or ' astounding' . The analysesrevealed that / 9~/ and / b / almost always appear just prior to word boundary, followed by / ~/ and then / m/ . As a result, it is far easier to find ' ' ' ' ' ' ' ' English words like increase, invade , applause, and approach than words ' ' ' ' like today and topeka . We recently conducted an experiment to test whether English speakershave knowledge of theserelationships. This experiment used a one-word / two -word judgment task. In this procedure, subjects hear a sequenceof two syllables and must judge as rapidly as possible whether the sequencecorrespondswith one word or two. The critical items consisted of quadruplets such as ' the mind ' , ' a mind ' , ' to mind ' , and ' in mind ' . In thesecases, a monosyllabic open class word is precededby one of four monosyllabic closed class words, with the closedclasswords varying in the likelihood with which they signal a word boundary. If English speakers have learned these probabilities and exploit them in word segmentation, then we would expect faster ' two -word ' judgments for those syllablesthat most reliably correspond with word boundaries. In addition , listeners should make fewer errors with these items. The filler items consisted of disyllabic words such as ' typhoon ' . All of the disyllabic words had stresson the secondsyllable, and so matched the stresspatterns of the two-word cases. This control prevented the subjects from using stress pattern as a guide to the correct answer. The results from this experiment are presented in table 2. Statistical analysesrevealedthat subjectswere significantly faster and made fewer errors for the / 9~/ and / b / conditions than for the / ~/ and then / m/ conditions. In Table2 Mean reactiontimes(in msec ) and error rates(pereight items) to identify a disyllabicpattern ' ' (e.g., the mind) as two wordsdependingon the initial word in the pattern

-generalabilities M.H. Kelly, S. Martin / Domain

addition , I ~I was significantly easierthan Iml on both the speedand accuracy measures. The 19~1 and Ibl conditions were not significantly different. We are currently planning a secondexperiment involving Ib / , I ~/, and Im/ . The same one-word/ two -word task will be used, but in this case, the target syllableswill ' ' ' ' ' ' actually be part of a word as in tomorrow , abandon and infection . The ' ' correct answer here is one word , but the predictions are exactly the opposite ' of those made for the samesyllablesin the ' two word context. Now , listeners should be fastest and make the fewest errors with Iml items and be slowest and make the most errors with the Ib / items. 4.2.3. ~rosodic structure Cutler and her colleagueshave recently argued that the distinction between strong and weak syllables, which is a major characteristic of English prosody, can provide powerful cues to word segmentation. Whereas strong syllables have a full vowel, weak syllables have a reduced vowel, which is typically though not exclusively realized as a schwa. In extensive analysesof English speechcorpora, Cutler and Carter ( 1987) found that over 70% of strong syllables coincided with a word boundary. Our own unpublished analysis of parental speechto children between 12 and 25 months of age is even more strongly marked. Across the 14 mothers in our sample, strong syllables corresponded with word boundaries 95% of the time, with 93% being the LOWEST value in the corpus. 6 Hence, a productive word segmentation strategy for English would divide an utterance at the beginning of each strong syllable, and submit the resulting units to a lexical search. In general, these units will match words in the lexicon, and the failures can be subjected to some type of re-analysis. Current evidence indicates that listeners do, in fact, use some version of this strategy. First , Cutler and Butterfield ( 1992) predicted that when listeners make word segmentationerrors, they should remove weak syllables from the beginnings of words and attach them to the ends of words. Analyses of 6 Of course, the mothers are not deliberately making their speech more predictive of word boundaries so as to assist their children in acquiring English. Rather, parental speechcontains a strikingly high percentageof words that refer to concrete objects and easily perceivableevents, for the good reason that children are more interested in these entities rather than more abstract themes like religion or politics . Given that the vocabulary of everyday objects contains a high proportion of monosyllabic words and is drawn from a core Germanic lexicon that is character ized by polysyllabic words beginning with strong syllables, a greater correspondencebetween strong syllables and word boundaries will appear in parental speech than in speech between adults.

-generalabilities M.H. Kelly, S. Martin / Domain

naturally -occurring and experimentally-induced segmentationerrors supported thesehypotheses. For example, a listener misheard ' bought a Mercedes' as ' Mortimer Sadies' , which involved detaching the weak initial syllable of ' Mercedes' and ' ' appending it along with the weak a to the strong first syllable. (Of course, some phoneme misperceptionsoccurred as well.) Second, Cutler and Norris ( 1988) found that listeners took longer to identify the real word 'mint ' in ' mintayf ' than in ' mintef' . Cutler and Norris argued that since ' mintayf ' contains two strong syllables, listeners would initially segmentthe syllablesinto separateword-units after the In/ . As a result, the listenerswould have to overrule the initial word boundary and reunite the It I with the ' preceding syllable. In mintef' , on the other hand, no word boundary would be inserted between the Inl and It I since the final syllable is weak. Thus, the subjectswould not have to rectify incorrect word boundaries in this case, and their reaction time would be faster. Finally , although Cutler and Carter' s corpus analysis indicated that a strong syllable generally marked a word boundary, numerous segmentation errors would still be made with a simple heuristic that always and only placed word boundaries before strong syllables . However, Cutler and Carter point out that most of the errors in this case would involve closed-class items as these are the principal words that violate the strong syllable heuristic. Indeed, they systematically violate the heuristic becauseword boundaries generally appear before weak syllables of closed class words but strong syllables of open class words. If English speakershave learned this pattern, and supplementedtheir word segmentation strategies with it , they could further reduce the chances of a missegment . In fact, segmentationerrors themselvesprovide some evidence that English speakershave learned these prosodic patterns. Although Cutler and Butter worth ( 1982) found that segmentationerrors tended to place word boundaries before strong syllables, some did in fact occur before weak syllables. However, most of these casesposited a closed-classword immediately after the boundary.

4.3. Parsing In determining the grammatical structure of . a sentence, listeners and readers are often faced with local ambiguities. For example, in a sentence ' ' beginning with The judge knew the law ... , the final noun phrase could be ' ' the direct object of knew or the subject of a complement clause. In attempting to understand how such ambiguities are dealt with , two major questions have been intensely investigated: What information is brought to

M .H . Kelly , S. Martin I Domain-general abilities

bear on the ambiguity , and when is that information available? Over the past two decades, the dominant answershave claimed that only a restricted range of information contributes to initial parsing decisions, with other classesof information only coming into play later to rectify errors in the first parse (e.g., Frazier and Fodor 1978, Frazier and Rayner 1982, Ferreira and Clifton 1986, Ferreira and Henderson 1990) . In particular , preferencesfor certain phrase structure geometries are used to guide initial parsing, such as a preference to create the simplest phrase structure tree consistent with the preceding input . These preferences are reputedly blind to other types of information , such as the identity of the verb in the sentence. The impenetrability of certain information classesto parsing mechanisms createsvery interesting predictions. Consider, for example, the sentences' The student forgot the solution was in the back of the book ' and ' The student ' ' , the hoped the solution was in the back of the book . In both sentences ' ' solution is the subject of a complement clause. However, a complement clauseinterpretation would create a more complex phrase structure tree than an interpretation that initially categorized' the solution ' as the direct object of the respectiveverbs ' forgot ' and ' hoped' . Hence, upon reaching ' the solution ' , a reader might opt initially for the simpler direct object parse, which would have to be rejectedwhen later disconfirming information is encountered. Still , one might entertain the possibility that the identity of the verbs themselves could affect theseinitial parsing blases. Although ' forgot ' can be followed by an NP object or an object clause, ' hoped' cannot appear with the former structure. If such facts about verb argument types can influence the earliest stages of parsing, then one would predict that readers would not posit a direct object interpretation for ' hoped' , and so would not need to reject such a hypothesis when subsequentportions of the sentenceare read. One would therefore predict that reading times for the disambiguating areas of the sentencewith ' hoped' would be faster than those for the same areas of the sentence with ' forgot ' . This prediction and analogous ones using other dependentmeasuresand experimental procedureswere supported in a set of studies by Trueswell et al. (in press b) . The results could not be attributed to ' ' ' ' generally faster processingof forgot than hoped becausethe reading time differencesbetween the verbs were significantly reduced in control sentences ' ' containing the complementizer that , which made the sentencesunambiguous . Although some studies have been reported that did not find effects of verb subcategorization preferenceson parsing (e.g., Ferreira and Henderson 1990), Trueswell et al. argue that thesestudies have a number of methodological flaws. For instance, the verbs used by Ferreira and Henderson ( 1990)

-generalabilities M.H. Kelly, S. Martin I Domain

did not differ very much in their preferencefor noun phrase versus sentential objects. Consequently, their methods were not sensitive enough to detect differences. In sum, then, the experiments conducted by Trueswell et al. (in press b) indicate that subcategorization information about a verb has immediate effects on parsing (see also Shapiro et al. 1993), contrary to models that severely restrict the class of information available to parsing mechanisms. Indeed, Trueswell et al. argue that a wide range of information sources contribute to parsing decisions. As one final illustration , Trueswell et al. (in press a) have recently provided evidence that semantic properties of the sentencesubject can influence parsing. They contrasted sentenceslike ' The defendant examined by the lawyer turned out to be unreliable' with ' The evidenceexamined by the lawyer turned out to be unreliable' . Parsing models that rely solely on syntactic preferenceswould claim that readerswould have ' difficulty with these sentencesbecausethey would initially treat defendant/ ' evidenceexamined to be the subject and verb of the main clause. Although such syntactic preferencesmight exist, Trueswell et al. explore the possibility that they could be modified by the relative prototypicality of the preceding nouns as the subject for the verb. Although ' defendant' is a reasonable ' ' ' ' subject of the verb examined , evidence is not , and Trueswell et al. provide evidence that readers weigh such information in their initial parsing processes. Thus, like visual depth perception, parsing appears to proceed by considering multiple sourcesof information , which may range from general syntactic preferencesto properties of the specific lexical items composing a sentence. 4.4. Grammatical category assignments

One problem that listeners must solve during language comprehension is the assignmentof words to the correct grammatical classes , such as noun and verb. These assignments must be made rapidly given that conversational speechproceedsat about 150 words per minute (Maclay and Osgood 1959). What types of information might listeners exploit in order to make these categorizations? Both semantic and syntactic factors are certainly available. Semantically, nouns tend to denote concrete objects whereas verbs tend to denote actions. These patterns are not invariant , but children do show sensitivity to them in experiments (e.g., Brown 1957) . Syntactically, words from different grammatical classesvary in their distributional requirementsin sentences . Thus, English nouns, but not verbs, can appear in the sentence

M .H . Kelly , S. Martin / Domain-general abilities

stole the base' whereas verbs, but not nouns, can appear in the ' the base' . sentence The runner While not disputing the important role of semantic and syntactic information for grammatical class, we would like to explore a relatively neglected domain of information . In particular , a large number of phonological cuesto grammatical classexist, and experimentshave repeatedly shown that speakers are sensitiveto them. For example, nouns and verbs in English differ in stress pattern, syllable number, duration , vowel characteristics, and other phonolo' gical dimensions (see Kelly , 1992, for a review) . English speakers have revealed their sensitivity to these and other correlations in a wide range of tasks. For instance, if listeners hear disyllabic pseudowords that differ in stressand are asked to use each of these words in a sentence, they will use words with first syllable stress more often as nouns and words with second syllable stressmore often as verbs (Kelly 1988b) . Another set of experiments took advantage of the fact that English words often develop uses in " other ' ' grammatical categories. Thus, police originated in English as a noun, but ' ' subsequentlydevelopeda verb use, whereas fumble originated as a verb, but later developed a noun use. Despite the frequency with which these lexical extensionsoccur, blasesagainst certain types of extensions exist (Clark and Clark 1979). Although the cited blases tend to be based on semantic and pragmatic factors, Kelly ( 1988b) speculatedthat the phonological characteristics of nouns and verbs could influence the easewith which they develop uses in the other category. In particular , if English speakers have learned the correlation between stressand grammatical class in English, then they might ' ' use this knowledge as a measureof fit betweena current noun and possible verb use and vice versa. Thus, they might consider nouns to be better verb candidates to the extent that they have the prototypical verb stress pattern. This possibility was tested by presenting one group of subjectswith pairs of disyllabic nouns that differed in stress but were control led for other factors , ' The

7 Although thesecorrelationsare well- documented . However , their originsmay seemmysterious evaluated . , explanationsfor someof thesepatternshavebeenproposedand experimentally For example , Kelly ( 1992 ) hasarguedthat threefactorsmight be involvedin the evolutionof differ phonologicalpredictorsof grammaticalclass: ( I ) Wordsfrom differentgrammaticalclasses . (2) Thesedistributionaldifferencesmay havephonological in their distributionsin sentences in the waywordsfrom various with them, whichcreatecontextualdifferences reflexesassociated . (3) Overtime, listenersviewthesecontextualeffectsas classes are pronouncedand/or perceived . This accounthasbeenappliedto the Englishnoun-verb context-freepronunciations permissible stressdifference(Kelly I 988a, 1989 ), and might be ) and duration difference( Daviset al. 1992 of other, currentlyunknownphonologicalcorrelates usedas a heuristicto predictthe existence with grammaticalclass.

-generalabilities M.H. Kelly, S. Martin / Domain

such as word frequency (e.g., llama and gazelle) . Another group of subjects were presented with pairs of disyllabic verbs that differed in stress (e.g., grovel and beseech ) . None of the words had uses in the other category as determined by the most recently available Webster's Collegiate Dictionary . Subjects in the two groups were asked to select one noun (verb) from each pair and use it as a verb (noun) in a sentence. The choices were significantly affected by the stress patterns of the words, as nouns with second syllable stress were used as verbs more often than nouns with first syllable stress whereas verbs with first syllable stress were used as nouns more often than verbs with second syllable stress. These patterns are not simply a laboratory artifact . Analyses of the history of English also found the same patterns in actual grammatical category extensions down through the centuries (see Kelly , 1988b, for details) . Although such experiments demonstrate that English speakers possess implicit knowledge of the noun- verb stressdifference, and that such knowledge may even have affectedthe history of English, they do not entail that the initial identification of a word as a noun or verb is at all affected by stressor other phonological variables. That is, the subject might have recognizedthat ' ' ' ' gazelle is a noun just as easily as they recognized llama . Stressthen came into play only later as a metalinguistic variable, but did not affect initial categorization of the words. In order to determine whether stress actually affects grammatical category assignmentsearly in speech processing, Kelly and Martin ( 1993) recently conducted a noun- verb categorization task that we will summarize here. The task was straightforward : On each trial , the subjectsheard a word that they had to classify as a noun or a verb as quickly as possible. The words were disyllabic and differed in stress, with half of the nouns and half of the verbs having first syllable stress and the other half having second syllable stress. In addition to varying a phonological correlate to grammatical class, we also manipulated a semantic cue. Given that nouns generally denote concrete objects and verbs readily perceived actions, we selectedour words such that half of the nouns and half the verbs had the prototypical meanings of their class, whereas the other half denoted abstractions. It is conceivable that the phonological cue to grammatical classwould only influence grammatical category assignmentswhen the semantic cue is absent. Given that the semanticcue is universal whereasthe phonological cuesare languagespecific, one might expect the former to be more basic and perhaps completely eliminate effects of the phonological cue. The results indicated that both semantic and phonological factors affected the subjects' judgments. Nouns

M .H . Kelly , S. Martin / Domain-general abilities

and verbs were classified faster if they had meanings typical of their classes. In addition , the phonological variable of stressinteracted with grammatical classsuch that nouns were classifiedfaster if they had first syllable stressbut verbs were classified faster if they had second syllable stress. Most importantly , the phonological effects were not eliminated for words that had concrete meanings. Indeed, these effects were significantly magnified by the presenceof semantic features that convergedon the samecategorization. The effect of concretenesswas likewise magnified by the presenceof appropriate phonological features. For example, concrete nouns with first or second syllable stress were classified faster than abstract nouns with either stress pattern. However, the differencebetweenthe concrete and abstract nouns was larger for items with first syllable stress. For verbs, on the other hand, the concrete/ abstract difference was larger for items with second syllable stress. Thus, listenersappear to use a conspiracy of cues to identify the grammatical category of a word . These cues can either be language-universal, like the semanticcue to grammatical class, or language-specific, like stressin English. Finally , the language-specific cues are not overwhelmed or even damped by the language-universal factors. Instead, they mutually reinforce one another.

5. Conclusions

In this paper, we have submitted arguments and evidencefor the following claims: ( I ) Multiple sources of information are available to solve problems in perception and cognition . 2 ( ) These information sources, though plentiful in number, are often probabilistic in nature. Nonetheless, if multiple probabilistic sourcesconverge on the solution to a problem, that solution is likely to be correct. Reliance on multiple cues should therefore produce greater successthan reliance on an individual cue. 3 ( ) Given ( 1) and (2), one would expect to see widespread sensitivity to probabilistic information throughout the animal kingdom . Furthermore, since many problem domains are characterizedby probabilistic solutions, sensitivity to variables such as frequency and rate of return should be a domain-general ability . This general ability could, like memory, be involved -specificproblems that have domain- dependentcharacteristics in species . The current evidence from the human and non-human animal literature strongly supports the hypothesis concerning sensitivity to pro -

M .H . Kelly , S. Martin I Domain-general abilities

babilistic information . Its domain-independence and interactions with -specific abilities need to be heshed out in much species greater detail, however. (4) Sensitivity to multiple sources of information for solving particular problems should be widespread and domain-independent. We have reviewed some areas, such as visual depth perception, where this claim appears to be correct. Language appears to be no different from other domains in this regard as numerous studies have shown that speakersare sensitive to statistical regularities in their language, even though some of these patterns are quite subtle. Extensive research remains to be performed to determine if such knowledge is actually exploited in language acquisition, comprehension, and production . In conclusion, over the past few decades, cognitive scientists have moved away from t ~e position that general principles operate across cognitive and ' perceptual domains. Instead, research and theory ranging from cognitive development (e.g., Chi 1978, Keil 1989) to animal learning (e.g., Seligman 1970) have emphasizedthat domain- dependentprinciples might constrain the ways in which human beings gather, organize, and draw inferences from different kinds of information . This approach recognizes that organisms confront many problems whose solutions require specific kinds of information that must be manipulated in highly constrained ways. As a result, the manner in which, say, migrating birds solve their navigational problems will probably not provide us much guidance in determining how children acquire the meaningsof words in their language. However, the value of the domainspecific approach should not causeus to neglect the fact that similar types of problems are in fact encounteredacross domains, and hencedomain-general solutions might have evolved to deal with them. The inherent probabilistic nature of an animal' s environment is one such invariant , and sensitivity to probablistic patterns appears to permeate perception and cognition, even in areas like language that seemto be highly constrained by domain-dependent principles.

References . Assessment of covariationby humansand animals: The . 1984 Alloy. L.B. and N. Tabachnik andcurrentsituationalinformation. Psychological Review joint influenceof prior expectations 91. 112- 149. ArmstrongS.. L. GleitmanandH. Gleitman. 1983 . Whatsomeconceptsmightnot be. Cognition 13. 263- 308.

M .H . Kelly , S. Martin I Domain-general abilities

. Journalof . Psychological Attneave , F., 1953 probabilityasa functionof experienced frequency 46 81 86. , Experimental Psychology . . Thinkingand deciding . New York: CambridgeUniversityPress Baron, J., 1988 . Perceptualequivalence of acousticcuesin Best, C.T., B. Morrongielloand R. Robson, 1981 . Perceptionand Psychophysics 29, 191-211. perception speechand nonspeech . Acousticinvariancefor placeof articulationin speech Blumstein , S.E. and KiN. Stevens , 1979 . of stop consonants of the spectralcharacteristics production: Evidencefrom measurements - 1017 . Journalof the AcousticalSocietyof America66, 1001 . Perceptualinvarianceand onset spectrafor stop Blumstein , S.E. and KiN . Stevens , 1980 . Journalof the AcousticalSocietyof America67, consonants in differentvowelenvironments 648- 662. . Journalof Abnormaland Social . Linguisticdetenninismand part of speech Brown, R., 1957 49, 454- 462. Psychology . Minimodularityand the perceptionof layout. Journal of Bruno, N. and J.E. Cutting, 1988 : General117, 161- 170. Experimental Psychology . In: R.S. Siegier(ed.), . Knowledgestructuresand memorydevelopment Chi, M.T.H., 1978 ?, 73-96. Hinsdale Children's thinking: What develops , NJ: Erlbaum. . Whennounssurfaceasverbs. Language55, 767- 811. Clark, E.V. and H.H. Clark, 1979 : The historicalrecord. Science . Fifty centuriesof right-handedness Coren, S. and C. Porac, 1977 198, 631- 632. : Evidencefrom . Rhythmiccuesto speechsegmentation Cutler, A. and S. Butterfield, 1992 . Journalof Memoryand Language31, 218- 236. juncturemisperception . The predominance of stronginitial syllablesin the English Cutler, A. and DiM. Carter, 1987 . ComputerSpeechand Language2, 133- 142. vocabulary . for lexicalacxess . The role of strongsyllablesin segmentation Cutler, A. and DiG. Norris, 1988 : HumanPerceptionand Performance14, 113- 121. Journalof ExperimentalPsychology . A noteon the role of phonologicalexpectationin Cutler, A., D. Norris andJ.N. Williams, 1987 and - 487. . Journal of Memory Language26, 48G speechsegmentation . Threegreadientsand the perceptionof flat and curved Cutting, J.E. and R.T. Millard, 1984 : General113, 198-216. . Journalof ExperimentalPsychology surfaces betweenEnglish . The causesof durationdifferences Davis, S., J. Morris and M.H. Kelly, 1992 . nounsand verbs. Unpublishedmanuscript . Somecommonaspectsof modelsfor learningand memoryin lower animals Estes , W.K., 1985 on learningand memory, 151- 166. and man. In: L. Nilsson, T. Archer (eds.), Perspectives Hillsdale, NJ: Erlbaum. . Journalof Memory of syntacticprocessing . The independence Ferreira, F. and C. Clifton, 1986 and Language25, 348- 368. . The useof verb informationin syntacticparsing: A Ferreira, F. and JiM. Henderson , 1990 andword-by word self-pacedreading.Journalof from eyemovements comparisonof evidence : , , and Cognition16, 555- 568. Memory Learning Psychology Experimental . On the semanticcontentof subcategorization Fisher, C., H. Gleitmanand L.R. Gleitman, 1991 23, 331- 392. . CognitivePsychology frames . . Frequency Francis,W.N. andH. Kucera, 1982 analysisof Englishusage:Lexiconand grammar Boston, MA : Houghton-Mifflin. : A new parsingmodel. Cognition6, . The sausagemachine Frazier, L. and J.D. Fodor, 1978 291- 326.

M .H . Kelly , S. Martin I Domain-general abilities

Fra2ier . Makingandcorrectingerrorsduringsentence , L. andK. Rayner : Eye , 1982 comprehension movements in theanalysis of structurallyambiguous senten ~ . CognitivePsychology 14, 17S - 210. Frazier, L. andK. Rayner, 1987 . Resolutionof syntacticcategoryambiguities : Eyemovements in . Journalof Memoryand Language26, 505- 526. parsinglexicallyantbiguoussentences Gallistel, C.R., 1990 . The organizationof learning. Cambridge . , MA : MIT Press Gernsbacher . Resolving20 years of inconsistentinteractionsbetweenlexical , M.A., 1984 . Journalof Experimental : , concreteness , andpolysemy familiarityandorthography Psychology General113, 256- 281. Gleitman, L.R., H. Gleitman, B. Landauand E. Wanner, 1988 . Wherelearningbegins: Initial for languagelearning. In: F.J. Newmeyer(ed.), Linguistics : The Cambridge representations : CambridgeUniversityPress . survey, Vol. 3, 150- 193. Cambridge Gluck, M.A. and G.H. Bower, 1988 . From conditioningto categoryleaming: An adaptive networkmodel. Journalof ExperimentalPsychology : General117, 227- 247. Gould, S.J., 1991 . Bully for brontosaurus : Reflectionsin naturalhistory. New York: Norton. . The learnabilityand Gropen, J., S. Pinker, M. Hollander, R. Goldbergand R. Wilson, 1989 acquisitionof the dativealternationin English. Language65, 203- 257. Hasher,L. and R.T. Zacks, 1984 . Automaticprocessing of fundamentalinformation: Thecaseof - 1388 . AmericanPsychologist 39, 1372 . frequencyof occurrence Hintzman, D.L., 1988 . Judgmentsof frequencyand recognitionmemoryin a multiple-trace Review95, 52S-551. memorymodel. Psychological Johnston . A testof the sophisticated . Cognition , J.C., 1978 guessingtheoryof word perception 10, 123- 153. Jonides . Direct coding of frequencyof occurrence . Journal of , J. and C.M. Jones , 1992 : Learning, Memory, and Cognition18, 36S-378. ExperimentalPsychology Just, M.A. and P.A. Carpenter . A capacitytheoryof comprehension : Individualdifferences , 1992 in workingmemory. Psychological Review99, 122- 149. Keil, F.C., 1989 . Concepts . Cambridge . , kinds, and cognitivedevelopment , MA : MIT Press . Rhythmicalternationand lexicalstressdifferences in English. Cognition30, Kelly, M.H., 1988a 107- 137. . Phonologicalblasesin grammaticalcategoryshifts. Journalof Memoryand Kelly, M.H., 1988b Language27, 343- 358. . Rhythmand languagechangein English. Journalof Memoryand Language Kelly, M.H., 1989 28, 690- 710. . Using sound to solve syntactic problems : The role of phonologyin Kelly, M.H., 1992 . Psychological Review99, 349- 364. grammaticalcategoryassignments . Phonological cuesto grammatical . Unpublished class . KeUy,M.H. andS. Martin, 1993 manuscript Kimball, J., 1973 . Sevenprinciplesof surfacestructureparsingin naturallanguage . Cognition2, 15- 47. . It ' s frogsall the waydown. Cognition15, 75-93. Levy, Y., 1983 Liberman, M., 1991 . Colloquiumpresentedto the Departmentof Psychology , Universityof . Pennsylvania Lichenstein, S., P. Slovic, B. Fischoff, M. Laymanand B. Combs, 1978 . Judgedfrequencyof lethalevents . Journalofex~ rimentalPsychology : HumanLearningand Memory4, 551- 578. Lieberman . Learning. Belmont, CA: Wadsworth . , D.A., 1990 Lisker, L., 1978 . Rapidvs. rabid: A catalogueof acousticfeaturesthat maycuethe distinction. HaskinsLaboratoriesStatusReporton SpeechResearch , SR-54, 127- 132.

-generalabilities M.H. Kelly, S. Martin / Domain

139

. Hesitationphenomena in spontaneous . Word , 1959 Maclay, H. andC.E. Osgood Englishspeech IS, 19- 44. . The acquisitionof morphonology . Monographsof the Societyfor , B., 1978 MacWhinney Research in Child Development 43 ( 1/2), 174. . Languageprocessingand informationintegration. In: N.H. Anderson Massaro , D.W., 1991 (ed.), Contributionsto informationintegrationtheory, Vol. I : Cognition, 259- 292. Hillsdale, NJ: Erlbaum. . Intuitive physics . ScientificAmerican24, 122- 130. , M., 1983 McCloskey . Studieson thecategorization of speechby infants. Cognition Miller, J.L. and P.D. Elmas, 1983 13, 135- 165. . Someeffectsof later- occurringinformation on the Miller, J.L. and AiM . Libennan, 1979 . Perceptionand Psychophysics 25, 457- 465. perceptionof stopconsonantand semivowel : Strategiesand shortcomingsof human . Human inference Nisbett, R.E. and L. Ross, 1980 -Hall. judgment. EnglewoodCliffs, NJ: Prentice . Expectancyand stimulusfrequency :A Pang, K., F. Merkel, H. Egethand D.S. Olton, 1992 . Perceptionand Psychophysics 51, 607- 615. comparativeanalysisin rats and humans . Lemabilityand cognition. Cambridge . Pinker, S., 1989 , MA : MIT Press . Grammaticalelementsof languagein the speechof preschoolchildren. In : , M.I., 1973 Popova C.A. Ferguson , 269- 280. NewYork: , D.I. Siobin(eds.), Studiesof child language development Holt, Rinehart, &; Winston. of stimulusmaterial. Reicher . Perceptual , G.M., 1969 recognitionasa functionof meaningfulness 81, 275-280. Journalof ExperimentalPsychology -initial stop . Relativeamplitudeof aspirationnoiseasa voicingcuefor syllable Repp, B.H., 1979 . Languageand Speech22, 173- 189. consonants evidencefor . Phonetictradingrelationsand contexteffects:New experimental Repp, B.H., 1982 ' modeof perception . Psychological Bulletin92, 81- 110. a speech . Phoneticcategoryboundariesare flexible. In: S. Hamad Repp, B.H. and AiM . Libennan, 1987 : Cambridge : The groundworkof cognition, 89- 112. Cambridge (ed.), Categoricalperception . UniversityPress . Probabilityof shockin the presenceand absenceof CS in fear conditioning Rescoria , R.A., 1968 66, 1- 5. . Journalof Comparativeand Physiological Psychology W.K. . In: . inhibition of fear R.A. 1969 Conditioned Rescoria , , Honig, N.J. Mackintosh(eds.), . issuesin associative Fundamental learning. Halifax: DalhousieUniversityPress . Pavlovianconditioning: It 's not what you think it is. AmericanPsychologist Rescoria , R.A., 1988 43, 151- 160. . A theoryof Pavlovianconditioning: Variationsin the Rescoria , R.A. and AiR. Wagner, 1972 . In : A.H. Black, W.F. Prokasy(eds.), of reinforcementand nonreinforcement effectiveness 2: research and Vol. Current aassical conditioning , theory, 64-99. New York: AppletonCenturyCrofts. : Studiesin the internal structureof Rosch, E. and C.B. Mervis, 1975. Family resemblances 7 . 573 605 . categoriesCognitivePsychology, . Perception andPsychophysics . Thesubjective estimationof syllablefrequency Rubin, D.C., 1974 16, 193- 196. : A contributionbased of inflectionin child language . On the emergence Ruke-Dravina, V., 1973 on Latvian speechdata. In : C.A. Ferguson , D.I. Siobin (eds.), Studiesof child language , 252- 267. New York: Holt, Rinehart, &; Winston. development

-generalabilities M.H. Kelly, S. Martin I Domain . Exposure , S., W. Swapand RiB. Zajonc, 1973 , context, and interpersonalattraction. Saegert Journalof personalityand SocialPsychology 25, 234- 242. . On the generalityof the laws of learning. PsychologicalReview77, , M.E.P., 1970 Seligman 406- 418. Shannon . Aristotelianism , B., 1976 , Newtonianism , and the physicsof the layman. Perception5, 241- 243. . The subjectiveestimationof relativeword frequency . Journal of Verbal Shapiro, B.J., 1969 Learningand VerbalBehavior13, 638- 643. . Preferences for a verb's complements andtheir Shapiro, L.P., H.N. Nageland B.A. Levine, 1993 usein sentence . Journalof Memoryand Language32, 96- 114. processing Shedier . Availability: Plausiblebut questionable . Paper , J.K., J. Jonidesand M. Manis 1985 at the 26thannualmeetingof the Psychonomic , Boston, MA. presented Society Smith, E.E. and D. Medin, 1981 . Categories andconcepts . Cambridge , MA : HarvardUniversity Press . Trueswell and S.M. Garnsey on parsing: , J.C., M.K. Tanenhaus , in pressa. Semanticinfluences Useof thematicrole informationin syntacticambiguityresolution . Journalof Experimental : Learning, Memory, and Cognition. Psychology Trueswell and C. Kello, in pressb. Verb-specificconstraintsin sentence , J.C., M.K. Tanenhaus -paths. Journalof Experimental : Separating effectsof lexicalpreference from garden processing : and . , , PsychologyLearning Memory Cognition . Attributesof memory. Glenview Underwood . , B.J., 1983 , IL : Scott, Foresman ' Walley, A.C. and TiD. Carrell, 19.83. Onsetspectraand formant transitionsin the adult s and child' s perceptionof placeof articulationin stopconsonants . Journalof the AcousticalSociety - 1022 of America73, 1011 . -outcomerelations: The role of E.A. Wasserman and H. Shaklee . Judging response , , 1984 outcome outcome and method of informationpresentation . , , respense contingency probability Memoryand Cognition12, 276- 286. . Attitudinal effectsof mere exposure . Journal of Personalityand Social Zajonc, R.B., 1968 2 Part 2 1 28. , PsychologyMonographSupplement 9( ),

Section 3

the Categorizing

world

-Holland 92(1994 - 167 . North Lingua ) 143

Does learning

a language

require the child to reconceptualize the world ? Susan Carey Instituteof Technology , Massachusetts , 77 Massachusetts Departmentof Brain and CognitiveScience Ave., 25-406, Cambridge . MA 02139 , USA Sortslconcepts with a count/massdistinction, provide , lexicalizedascountnounsin languages ' ' criteriafor individuationandnumericalidentity. This paperexamines Quines andPiagets claims that babiesand youngchildrenlack the logicalresources to representsortalconcepts . Evidenceis marshalledagainstthe Quine/Piagetposition, in favor of a view that evenyoung infants representat least one sortal concept , physicalobject, which providesspatiotemporal criteriafor individuationand identity. Evidenceis alsoprovidedthat babiesbelow II monthsof agemay not representmore specificsortalssuchas cup, animal, bottle, or book. Rather, they theseentitiesin a waycloselyrelatedto Quine's hypothesis . mayconceptualize

1. Intr04uctio D

Many students of language acquisition and cognitive development argue that the continuity hypothesisshould be the default, to be defeatedonly in the face of extraordinary evidence (e.g., Pinker 1984, Macnamara 1982). The continuity hypothesis is that representational format is constant throughout development; that the child has innately the logical and conceptual resources to representhis or her world as do adults. The continuity hypothesis denies ' stage changesof the sort envisioned by Piaget, denies changesin the child s ' linguistic representations such as the putative semantic category/syntactic ' category shift posited some years ago. According to the continuity hypothesis , language learning is a very complex mapping process; the child must learn which syntactic devices his/ her language employs, and which of a universal set of semantic distinctions are expressedin the syntax of his/ her language. What the child need not do, on the continuity hypothesis, is construct genuinely new representationalresources. Of course, whether the continuity hypothesis is true or not is an empirical question, and to examine it , one must entertain possibilities as to what types -3841 Science B.V. All rightsreserved 0024 /94/$07.00@ 1994- Elsevier -F SSDI0024-3841(93)EOOS8

S. Carey / Reconceptualizingthe world

of discontinuities could possibly obtain in the course of development. If evidence for discontinuities is found , several further questions are then licensed, including : ( 1) by what mechanism is the change effected (e.g., maturational , learning by some other process than currently understood parametersetting or hypothesis testing methods) . (2) What is the relation betweenthe discontinuity and languagelearning? Is somechange in representational resources required as a prerequisite to some aspect of language learning? Alternatively , does language learning playa role in causing the change? ' Here I examine an important discontinuity proposal of Quine s, versions of which are endorsed by thinkers as diverse as the British empiricists and Piaget. Quine, Piaget, and others maintain that early representationsof the world are formulated over a perceptual quality space(Quine, the empiricists) ' or sensori -motor representational system (Piaget) . On both Quine s and ' Piaget s views, the baby is not capable of formulating any representations with the properties of adult concepts such as object, dog, table. ' Quine s proposal is that the ontology that underlies language is a cultural construction . ' Our conceptual firsts are middle -sized, middle distanced objects, and our introduction to them and to everything comesmidway in the ' cultural evolution of the race (Quine 1960: 5). Before the child has mastered ' this cultural construction, the child s conceptual universeconsistsof representations of histories of sporadic encounters, a scattered portion of what goes ' on. Quine speculatesas to the representationsunderlying the toddler s usesof ' ' ' ' ' ' ' the words water , red , and Mama . His first learning of the three words is uniformly a matter of learning how much of what goes on about him counts as the mother, or as red, or as water. It is not for the child to say in the first ' ' ' case, ' Hello , Mama again , in the secondcase Hello , another red thing , and ' ' in the third case, Hello , more water . They are all on a par : Hello , more ' Mama , more red, more water (Quine 1960: 92) . The child mastersthe notion of an object, and of particular kinds of objects, in the course of getting the ' ' hang of what Quine calls divided reference, and this through the processof ' ' ' mastering quantifiers and words like same. The contextual learning of these various particles goes on simultaneously, we may suppose, so that they are gradually adjusted to one another and a coherent pattern of usageis evolved ' matching that of one s elders. This is a major step in acquiring the conceptual schemethat we all know so well. For it is on achieving this step, and only ' then, that there can be any general talk of objects as such (Quine 1969: 910) . And in another place he finishes the same idea with a bootstrapping metaphor, underlining the degreeof conceptual changehe thinks is occurring :

S. Carey/ Reconceplua Ilzing lhe world

' The child scrambles up an intellectual chimney, supporting himself against each side by pressureagainst the others' (Quine 1960: 93) . Quine also states that once the child has masteredthe notion of an object, and got the trick of divided reference, he goes back and reanalyzes' Mama ' , so that it is now the name of a unique enduring person. ' Quine s view can be schematizedas follows. Imagine a portion of bottle experience that we adults would conceptualize as a single bottle. Babies respond to bottleness or bottlehood also, and can learn many things about bottlehood ; for instance, they can come to associatebottlehood with milk , or with the word ' bottle' . Now imagine a portion of bottle experiencethat we would conceptualizeas three bottles. The infant would also expect to obtain milk (indeed, more milk ) from this bottleness and could also refer to it with the word ' bottle ' . Note that shape is important to the identification of bottlehood, just as the shape of the individual grains is important for distinguishing riCefrom spaghetti from macaroni. Similary, even if Mama is a scattered portion of what goes on, shape is important for distinguishing Mama from Rover or from Papa. That shape is important for distinguishing what scattered portion of experienceconstitutes bottlehood does not mean that the baby is capable of representing' a bottle ' , ' two bottles' , or ' the same bottle I had yesterday'. Thus, demonstrations that toddlers are sensitive to shape in inductions of word meanings when new words are ostensively defined over objects (e.g., Landau, this volume) do not bear on Quine' s proposal. In this discussion I will not make contact with Quine' s radical philosophical views such as the indeterminacy of translation. I assumethat we can characterize the adult ' s onto logical commitments, that these include middle' sized physical objects, and that words such as ' table' , ' dog' and ' person , ' function as sortals in the adult lexicon, in Wiggins ( 1980) sense. Sortals refer to kinds of individuals (i.e., divide reference), providing conditions for individuation (establishing the boundaries of entities) and for numerical identity (establishing when an individuated entity is the same one as one experiencedat someother time, or in somecounterfactual world ) . One way of ' stating Quine s hypothesis, as I construe it , is that babies and toddlers represent no sortal concepts, no concepts that provide conditions of individuation and numerical identity , no concepts that divide reference. Two reviewers of this paper raised the objection that representations of shapes presuppose representations of individuals that have those shapes, ' claiming therefore that Quine s proposal (at least as construed above) is ' incoherent. This is not so. Pleasedwell on the spaghetti, macaroni case. It s

S. Carey I Reconceptualizingthe world

true that if the contrast between the two types of stuff is basedon the shape differencesof individual pieces, then some representationof those individual pieces must enter into the representation of shape. But our concepts of spaghetti and macaroni (and the words spaghetti, macaronI) do not quantify over those individuals. Similarly , we can represent the shape of a scattered portion of sand, arranged, for example, into an S, and when we refer to it as ' ' ' 'a portion or an S we are quantifying over that individual . But when we ' ' think of it as sand, we are not. Quine s proposal is that the child s conceptual/ linguistic system has only the capacity to represent the world in terms of ' concepts like furniture , sand, bottlehood. Of course the child s perceptual system must pick out individuals in order to represent shape, to determine what to grasp, and so on. This is part of what Quine meant when he claimed ' ' that the child is inherently body minded (Quine 1974) . Piaget, like Quine, believed that that baby must construct the concept of enduring objects, although he differed from Quine as to the' mechanismshe envisioned underlying this construction. Quine saw the child s mastery of the linguistic devices of noun quantification , the machinery by which natural languagessuch as English manage divided reference, as the' processthrough ' which the child s ontology comes to match his or her elders . Piaget held that the baby constructs the concept object during the course of sensori -motor development by the age of IS-months or so, and that this construction is the ' basis for the child s mastery of natural language. Since Piaget did not frame his discussion in terms of an analysis of the logic of sortals, it is not clear when he would attribute full sortals to the child. 1 ' The Quine/Piaget conjecture about the baby s representational resourcesis a serious empirical claim, and as I will show, it is difficult to bring data to ' bear on it . In what follows , I first consider Quine s views, contrasting his hypothesis that children come to represent sortals only upon 'learning the' linguistic devicesof noun quantification with what I will call the Sortal First hypothesis. The Sortal First hypothesis is that babies represent sortal concepts, that toddler lexicons include words that expresssortals, and that these representations underly the capacity for learning quantifiers rather than resulting from learning them. I then turn to early infancy, and explore the contrast betweenthe Quine/Piaget hypothesis and the Sortal First hypothesis as regards the earliest phases of word learning. A preview of my conclusions: whereas the Sortal First hypothesis is ultimately favored, evi1 For example, Piaget thought that the logical prerequisitesfor representingthe adult concepts all and someare not acquired until after age S.

S. CareyI Reconceptualizing theworld

147

denceis presentedfor a decidedly Quinian discontinuity in infant conceptual development.

's mastery Thetoddler

of

syntax

' Quine s hypothesis is that the child masters the logic of sortals through a processof adjusting the meanings of nouns and of natural language quantifiers to each other (scrambling up an intellectual chimney, the walls of which are the child ' s currently worked out representationsof the quantifiers he/ she knows). To address Quine' s conjecture experimentally, we must first know when in the child ' s life the putative scrambling is going on. Even by age 3 the child is not producing all the quantifiers that constitute the sides of Quine' s chimney. The very beginnings of the English count/ mass distinction are masteredin the months leading up to age 21/. 2. Many children age 2:0 produce nouns with no determiners or plurals, but some have begun to produce plurals and a few determiners and quantifiers (usually possessivessuch as ' ' ' ' ' ' my , plus a and the ) . Many 2-year-olds beginning to use determiners do not distribute them differently according to the noun' s count/ mass status in the adult lexicon. They still omit many determiners, and use others like ' the' and ' my' that do not differentiate count nouns and mass nouns. By 21.J2 ' virtually all children distinguish in some ways the syntactic contexts in which words like ' table' and ' dog' appear from those in which words like ' water' and ' playdoh' appear (Gordon 1985, Sola et al. 1991) . Gordon ( 1982) showed that between 21/. 2 and 3 years of age the distinction becomes marked in ' syntax, as the child s speechabruptly comes to reflect the arbitrary rule that detenniners are obligatory for singular count nouns, but not for mass nouns ' ' ' .' (that is, one can say I like juice , but not 1 like dog ) . The developmentalfacts summarizedabove determine the relevant agesfor an empirical test of Quine' s speculations. Data bearing against Quine' s claims could be of several types: e.g., data showing children age 2 or under take proper nouns to refer to individuals of a kind or that they take count nouns to refer to kinds of individuals. But, as already mentioned, the trick is ' ' ' figuring out how we can know whether toddlers Mama refers to entities ' ' they conceptualize as individuals or whether their bottle divides reference, referring to each individual of a certain kind , as opposed to bott/ehood. Another type of evidencecould be relevant. If it can be shown that upon first learning ' a' or the plural ' -s' , toddlers interpret them correctly, as signalling an individuated entity of a kind or a plurality of individuals of a

148

theworld S. CareyI Reconceptualizing

kind , respectively, this would tell against Quine. This is becausetheseare the first relevant quantifiers the child learns. If he or she interprets them correctly from the beginning, the interpretation could not have been acquired through an adjustment process involving the entire set of quantificational devices of noun syntax. This last point is important . In the beginnings of language ' learning, on Quine s view, children will not interpret those few quantifiers in their lexicons as adults do. The scramble will have just begun. Data showing ' that children use ' a' and plurals will not be itself relevant to Quine s hypothesis; it must be shown that such quantificational devicesare doing the samework as they do in the adult language.

3. The composition of the toddler lexicon

' A large proportion of the baby s first words are words for middle-sized ' ' ' ' ' ' ' ' ' ' physical objects, such as bottle , book , dog , cup , and banana . But that babies have words in their lexicons that refer to object kinds in the adult lexicon tells us nothing of what these words mean to the babies. Many have argued that the earliest words are often complexive (e.g., Bowerman 1978, Dromi 1987, Vygotsky 1962). That is, children appear to extend words to new referentson the basisof any of the salient perceptualproperties of the original experiencesin which the word was heard. These complexive uses often cut ' ' across what are for adults distinct ontological categories, as when paper apparently refers to the act of cutting, the act of drawing, to pens and pencils and to paper (Dromi 1987). If such complexive uses reflect unconstrained (from the point of view of adult lexical categories) projection of word ' meanings, Quine s views receivesupport . But it is important to seethat such ' complexive usesare not necessaryfor Quine s conjecture to be correct. Indeed, others deny that toddlers construct complexive meanings; Huttenlocher and Smiley, 1987, for example, presentevidencethat from the beginning ' ' babiesuseeach word for middle-sizedobjects appropriately: bottle to refer to bottles, ' book' to books, and so on. But even if Huttenlocher and Smiley are ' right , this fact .does not disconfirm Quine s conjecture. In fact, Quine presupposes that the baby usesthe words in contexts adults would. His point is that, ' evenso, the baby might not be individuating the words referentsas we do. The baby could refer only to what we conceptualize as bottles when she uses ' bottle' but could be , referring to bottlehoods. She could be using the word to refer to a scattered portion of what goes on, determined by perceptual ' ' similarity to the portions of her experiencewhen adults use bottle .

S. Carey I Reconceptua/izing the world

4. Toddler se. itivity to noun syntax

Children as young as 17 months (at least girls that young) are sensitive to the syntactic context in which a new noun is heard in their projection of noun meaning (Katz et al. 1974, Macnamara 1982) . Specifically, if ostensively taught a new word in a count noun context, referring to an unfamiliar doll . ( See this. This is a dax. Can you touch the dax? Can you put the dax on ' your head ... ), they assumethat other dolls of the sametype are also daxes. But if taught in a proper noun context (.Seethis. This is Dax. Can you touch Dax. Can you put Dax on your head ...' ), they assumethat other dolls of the sametype are not Dax , reserving ' Dax ' for the original doll only . Do thesedata establish that young children distinguish kinds from individuals , and use count nouns to achieve divided reference? Certainly not. They do establish that toddlers are sensitive to the syntactic distinction between nouns following determiners and those not following determiners, but this distinction could be signalling a different semantic distinction than that between individuals and kinds. For a sample Quinian interpretation : babies ' ' ' ' could take nouns without determiners such as ' Dax , ' Rover , and Joan , to refer to portions of experiencedefined by a stricter similarity metric than that referred to by nouns with determiners. Supposea Quinian baby, Alice , has a ' ' ' ' brother whom she hears called both Rupert and a boy . Supposealso that she relies on shape to determine Rupertness and boyness. She could have ' ' learned from others' usage of the words that to be called Rupert , a given portion of experiencemust be very similar in shapeto the original portions of ' ' experienceto which the term was heard to refer, whereasto be called a boy , ' .the boy , something need look only somewhat like the original referent. A ' ' ' ' generalization of this pattern of distinction , across Alice and a baby , ' and so on could .Rover' and ' a , , underly the patterns of projection dog , found by Katz et al., 1974, and subsequentreplications. This interpretation of the Katz et al. data attributes to the baby a different ' ' ' ' ' ' meaning for a from the adult as well as different meaningsfor bottle , boy , ' ' . Rupert . This is, of course, Quine s position . On his view, it is only in the course of learning other quantifiers, plural markers, and so on, and adjusting to all the contrasts in usage they mark (the process of scrambling up the ' ' intellectual chimney cited above) that the baby works out the meaning of a , ' ' ' ' ' ' ' ' ' ' ' . ' .the another some more all , , many , same,2 etc. , , , 2 Bloom. personal communication, provides the following argument against the Quinian ' interpretation of the Katz et al. data. Among children s very first words are some pronouns (e.g.,

In several studies, my colleaguesand I have attempted to addressQuine' s ' proposal by comparing children s representations of solid physical objects, such as cups, with their representationsof non-solid substances , such as sand . is since adults or gels or creams Our idea that conceptualize the former as kinds of individuals (i .e., in terms of sortals that divide reference), but do not conceptualize the latter in this way, we might be able to find evidence that infants and toddlers respect the quantificational distinction between the two as well. In the first studies, Sola et al. ( 1991) compared 2-year-olds' projection of newly learned :words ostensively defined by reference either to novel solid physical objects (e.g., a brass plumbing T) or novel non-solid substances(e.g., a hair-settin$ gel with grapenuts embeddedin it ) . The objects were made of unfamiliar materials and the non-solid substanceswere presentedformed.into distinctive novel shapes. The child was introduced to the novel entity and ' ' ' ' provided a word for it (e.g. blicket for a novel object; stad for a novel nonsolid substance). The child was then presented two new sets of stimuli and asked to give the experimenter the blicket or the stad. For each object trial , the choicesconsistedof another object of the same shapemade of a different material (e.g., a plastic plumbing T) or three small pieces of the original material (brass) . For each substance trial , the choices consisted of a new substanceformed into the original shape, or three small piecesof the original substance. Figure I shows the design for one trial of each type. There were four object trials and four non-solid substancetrials. Of course, which words were assigned to which entities varied across subjects, but for expository ' ' ' ' clarity I will use blicket as my sample object name and stad as my sample non-solid substancename. ' ' ' he' ' it ' and thesearetreatedfrom theonsetas , ) belongingto the samecategoryas Rupert and ' ' not as dog . Childrendo not usethemwith determinersor modifiers(Bloom 1990 ). The adultlike , denote analysisworks well in accountingfor this finding; pronouns, like proper names individuals. But the Quiniananalysisfails, sincea far broaderrangeof referentsare called' it ' or ' he' than are called'dog' . I agreewith this argument, as I rejectthe Quinian proposalin . However, Quinecould reply that the child hastwo setsof favor of the Sortal First hypothesis words: wordslike ' dog' and wordslike ' Rupert', and a few singularitieslike ' he' , eachtagged with someof the syntacticcontextsin which they appearand eachtaggedwith prototypical . Comingto recognizethat the syntactic referentsand a similarity metric that determinesusage contextsin which ' he' appearsare the sameas those for ' Rupert' might be part of the . scramblingprocess

S. Carey /

the world 8U~ dCEftW .

OBdECr 1' RIAL

RAJ8D Sft MULU8:

T ~

TEST 8' 11MU LI :

. my Dbt

.

S . " . . ..., . ~ .

~

..dad4 JUh ~

. o. Mlabtf .

@8 ~

. o. . . . ?

Fig . 1. Object trials and substancetrials in Sola et al. ( 1991) .

' Sola et al. carried out two analysesto assesswhether children s representations of the referents of the words were influenced by the status of their knowledge of count/ mass syntax. First , they collected production data and assignedeach child a value corresponding to the degreeto which count nouns and mass nouns appeared in selective syntactic frames (e.g., ' a NO U N ' , ' NO UNs ! ' ' too much NO U N ' . Scores ran from 0 to near 1.0. Second , they ) introduced the new words in two different ways. In the neutral syntax condition , no syntactic information as to the count/ mass status of the word was provided ; the words were introduced as ' my blicket , my stad' and ' ' subsequentlyappearedin the context the blicket , the stad . In the informative ' ' syntax condition , the words were introduced as a blicket , some stad , and ' further differentiated syntactically, e.g. another blicket , some more stad' . As figure 2 shows, children at age 2:0 and 2:6 used different basesfor their projection or words for the two different types of entities. They projected ' biicket' to the other whole object the sameshapeas the original referent and ' stad' to the scattered portion of substancethe same texture they projected and color as the original referent. For object trials , children were sensitive to matchesin shapeand number; for non-solid substancetrials , children ignored matchesin shapeand number. Performancewas more adult-like on the object trials, but performance on both types of trials was better than chanceat both ages. Also apparent on figure 2, the syntactic context made no difference. The children were no more likely to interpret ' blicket' as the word for a kind of individual when it was heard in a count noun context. Similarly , hearing

S. Carey / Reconceptualizingthe world % choicesmatchingreferentin shapeandnumber

Ir.a rigid I [] nonsolid .

" 2 60

.

& I

chance .

! :

0 =

e

. 40

i ~ . .

!

"a"

"some " "a" " " "my" " . " "some my" my" my" 2:6 2:0 2:6 2:0 NEUTRAL SYNTAX SYNTAX INFORMATIVE condition

Fig. 2. % trials in which test stimulus chosen matched original referent in shape and number (Sola et al. 1991).

' stad' in a massnoun context made them no more likely to conceptualizestad as a substancethat can appear either in scatteredor singly bounded portions . Further , the child ' s productive control of count/ masssyntax did not influence the pattern of projection : children with differentiation scoresof 0 showed the samepattern as those with differentiation scoresclose to 1. ' We can conclude from these results that an entity s status as a solid physical object (or not ) influences which of its properties are salient in determining what other entities are referred to by the sameword. We can also conclude that this distinction between objects and non-solid substances predates mastery of count/ mass syntax. These data are consistent with the ' ' ' Sortal First hypothesis, for they are consistent with the child s taking blicket ' ' to refer to each individual whole object of a kind , and stad to refer to a kind of substance, conceptualized as a non-individuated entity . But the data are ' also consistent with the following more Quinian interpretation of the child s representationsof the blicket and the stad. ' Babies, being ' body-minded (Quine 1974) could be sensitiveto the perceptual experiencesthat determine objecthood: boundedness, rigidity , coherence through motion . Whenever theseare detected, they could heavily weight such . Shape would features as shape in their representation of these experiences

S. CareyI Reconceptua/;zing the world

thus be a salient feature of the blicket , but not of the stad, for non-solid substancesdo not maintain their shapeswhen manipulated. For non-solid substances , properties such as texture and color might be salient, for these constant over experienceswith substances . In other words, the two -yearstay old could be using ' blicket' to refer to blicketness, and recognize blicketness by shape. The differential patterns of projection do not establish that the toddler is using ' blicket ' to refer to any individual whole object of a certain kind , that the toddler divides the referenceof ' blicket ' . One detail of the data from figure 2 favors the Sortal First over the Quinian interpretation, and that is that toddlers performed more like adults on the object trials than on the substancetrials. Quine' s interpretation of this would have to be ad hoc, perhaps that the baby has had more object experiencethan substanceexperience. But the Sortal First hypothesispredicts this asymmetry. To seethis, supposethe Sortal First hypothesis is true, and ' ' suppose that upon first hearing the word blicket the child assumesthat it refers to each individual object of a certain kind . The choicesfor testing how the child projects ' blicket ' included another si.ngle object, and 3 small objects. Even if the child isn' t exactly sure of which features of the blicket establish its kind , the child can rule out that the 3 small objects are a blicket , for under no interpretation can they be an individual object of the same kind as the original referent. Children should then be at ceiling on the object trials , which ' ' they are. .The substancetrials are another story. If upon first hearing stad , the child takes it to refer to the kind of substanceof the original referent, then scatteredportions have no different status from unitary portions . There is no clue from number of piles which of the choices on the test trials is the stad. If children are not certain what properties of the original sample of stad determine the kind stad, they might do worse on the stad trials. And indeed, they do. The key issue here is the role of number in determining how to project ' blicket' . If the Quinian interpretation of the data is correct, the baby should ' blicket' on the basis of project shapesimilarity , no matter whether the choice that does not match in shapeconsists of one object or three objects. That is, the baby should succeedon an object trial as on figure 3 as well as on an object trial as in figure 1. The Sortal First interpretation predicts that performance on the object trials will fall to the level of performance on the substancetrials if the cue from number is removed (figure 3) . In an object trial such as that on figure 3, ' blicket' is ostensivelydefined as before, but the choicesfor projection are changed: another blicket of a different material (as before) and another whole object of . a different kind made of the same

S. Carey / Reconceptualizingthe world NAMBO B' nMULV8:

1DT rnMULI :

.i~ rt ~ ~

b. . tM MlDbl ?

trialin Sola(1987 ). Fig. 3. Object material as the original referent (instead of the three small objects) . Now the child has no clues from number of objects as to which is the correct choice. Performanceshould fall to the level of the substancetrials , and indeed, this is what happens(Sola 1987) . Apparently , the child uses the information provided by number on the object trials , but not on the substancetrials. We take this as evidencethat the child conceptualizessomeentities as individuals (such as kinds of objects) and conceptualizesother entities as non-individuated (such as kinds of substances ) . These distinct ways of conceptualizing objects and substancespredates ' ' mastery of count/ masssyntax. Toddlers do not merely project blicketness on the basis of shapeof individual piecesof blicketness, as we determine whether somepasta is spaghetti on the basis of the shapeof individual pieces. Instead, ' ' the pattern of projection suggeststoddlers divide referenceof blicket , and take it to refer to any individual of a certain kind .

' ' ' ' ' 6. Toddlers understand i Dgof 8 , someNOUN_ I take the data reviewed in the previous section to show that by age 2:0 children take ' blicket ' to refer to individual objects of a certain kind and ' stad' to refer to non-solid substancesof a kind and that the toddlers' , representationsof blickets and stads have the same quantificational structure ' ' as would adults . ' Blicket is a sortal term. These data disconfirm Quine only on the assumption that the baby did not acquire these representationsfrom

S. Carey / Reconceplualizinglhe world

learning English noun quantifiers. This assumption seemswarranted, given that as a whole toddlers at 2:0 do not produce quantifiers, and given that the pattern of projection was independent of whether the individual subjects produced any noun quantifiers selective for count nouns. A worry , though , is that babies may have better comprehension than production of the quantifiers. We attempted to address that possibility by manipulating the syntactic context in which the word appeared. As mentioned above, the syntactic environment in which the new word appeared had no effect in Sola et alis experiments, even at age 2!j2 when many children did produce quantifiers differentially for what are count and mass nouns in the adult lexicon. The Quinian interpretation of this fact is that quantifiers like ' a' , ' another' , ' some NOUN _ ' , ' some more NOUN _ ' do not yet signal the distinction between individuated and nonindividuated entities, just as the child is not projecting ' blicket' and ' stad' on the basis of that distinction . The Sortal First interpretation : objects are naturally construed as individuals of a kind and non-solid substances are naturally construed as non-individuated entities, even by toddlers, as shown by performance in the neutral syntax condition . Informative ' syntax merely reinforces the child s natural construal of the two types of entities. A study by Sola ( 1992) decided betweenthesetwo interpretations, and also establishedthat our production data did not underestimatetoddlers' interpretation of the quantifiers. Sola taught toddlers words for the objects and substancesin a new condition : contrastive syntax. ' Blicket' was introduced in a mass noun context ; ' stad' in count noun context. That is, when shown a novel solid object, the child was told , ' Here' s some blicket ... Would you like to seesome more blicket?' And when shown a non-solid substancefashioned into a distinctive shape, the child was told , ' Here' s a stad ... Would you like to seeanother stad?' As can be seen from figure 4, at both ages 2 and 2!j2' the pattern of projection was markedly different from that seenin the neutral and informative syntax conditions (figure 2) . At both ages, the syntactic context ' some NOUN _ ' , ' some more NOUN _ ' made children slightly less ' ' likely to construe blicket as referring to an individual whole object of a kind . There was a slight tendency towards interpreting it to mean something like brass. The syntactic context ' a stad' made children significantly less likely to construe the non-solid substance as a non-idividuated entity . Rather, they interpreted the word as meaning something like s-shapedpi /e. Wait , you might say, doesn' t this show that children at theseagesdo know the force of ' a' , ' another' , and so might have learned to represent sortal

S. Carey I Reconceptualizingthe world

number

in

and

referent

choices

shape

%

matching

m rigid 1

1 nonsolidi

10 . . 160 c i .

!

chant J 40

. ~ c ! II Do

2 "

" some

, 6

:

some

.

1 2

0

:

2

CONDm ON CONnl Crl NG SYNTAX . differentfrom d1ar: e

Fig. 4. % trials in which test stimuluschosenmatchedori2inai referent in shape and number in Sole( 1992 ).

concepts in conjunction with bootstrapping the meaning of the quantifiers? No , becauseof one further aspectof the data. At the younger age, sensitivity to conflicting syntax was shown solely by those children who had differentiated count and mass nouns in their production . Those whose differentiation scoreswere low performed just as did toddlers in the informative and neutral conditions, projecting ' blicket' to the other object of the same kind as the ' ' original referent and stad to the other substanceof the same kind as the ' ' original referent. This shows that the interpretation of blicket as a sortal ' ' ' ' predateslearning the meaning of a , another , and presumably underlies the latter achievement, as predicted by the Sortal First hypothesis. ' ' These data tap the very moment children first learn the meaning of a . 's They have only begun the scramble up Quine chimney, and have not had ' ' time to adjust their interpretation of ' a' to many other quantifiers. Yet , a signals an individuated entity of some kind . Together these data provide converging support for the Sortal First hypothesis. The child naturally

S. CareyI Reconceptualizing the world

157

construes physical objects as individuals in distinct kinds , and naturally construes non - solid substances in terms of kind of non - individuated entities . These natural construals support adult - like projection of word meaning (figure 2), and support adult - like interpretation of newly learned quantifiers ' ' ' ' like a , some and plurals .

YGungerinfants ' Altogether the data support the Sortal First hypothesis over Quine s conjecture, but they do not establish when the child first begins to represent sortal concepts. As noted earlier , it is not clear when Piaget would attribute sortal concepts to children , but it is certain that he would deny them to young infants . The argument I have developed so far does not bear on Piaget' s claims about the representational capacities of infants , as it concerns children age 24 months and older . Of course, a demonstration that young infants represent sortal concepts would defeat ' ' ' Quine s conjecture as well as Piaget s characterization of the infants conceptual resources. Studies by Cohen and his colleagues(e.g., Cohen and Younger 1983) show that quite young babies will habituate when shown, for example, a seriesof distinct stuffed dogs, and that they generalize habituation to a new stuffed dog and will dishabituate when shown a stuffed elephant. Similarly , when shown a seriesof distinct stuffed animals, babies of 8 or 9 months habituate, generalize habituation to a new stuffed animal, but dishabituate to a toy truck. Do thesedata not show that babies of that age representconceptssuch as ' dog' and ' animal?' Certainly not. Babies may be sensitive to dog shapes or animal shapes; babies may be habituating to doghood or animalhood. To credit the baby with sortals such as ' dog' , or ' animal' , we must show that such concepts provide the baby with criteria for individuation and identity . My discussion of this question has two steps. First , I argue that babies represent at least one sortal , object. Second, I present some recent data from my lab that suggest that as late as 10 months of age, the baby may have no more specific sortal concepts - not cup, bottle, truck , dog, animal .... Thus, a Quinian interpretation of the above habituation data may well be correct.

S. Carey I Reconceptualizingthe world

8. Principles of individuation : Yo UDger infants

' ' Piaget s characterization of infants cognitive capacitieswas basedon tasks in which the baby must solve some problem, often involving means- end ' analysis, and often involving planning some action. For example, Piaget s conclusions that babies do not represent objects as continuing to exist when out of view were basedon the robust finding that babies under 8 or 9 months ' cannot remove a cover to get a hidden object. The babies failure might be due to their failure to realize the object still exists, as Piaget thought , or equally might be due to their inability to carry out one action (remove a cover) to achievesome other goal (obtain the object). What is neededis some reflection of the baby' s conceptualization of the world that relies on behaviors well within the repertoires even of neonates. Over the past 15 years or so, such a method has been developed and is now very widely used. It relies on babies' ability to control what they attend to. The basic idea is simple. Under most circumstancesbabies will look longer at what is unfamiliar or unexpectedcompared to what is familiar or expected. Researchersuse this fact to diagnosehow the baby representssome situation , especiallywhat the baby considers surprising given his or her current state of physical knowledge. The selectivelooking paradigm has been usedextensively to probe babies' representations of objects, and the data from a subset of thesestudies can be recruited to bear on the question at hand. They establish that by four months of age the baby representsat least one sortal concept the concept of a physical object. The baby has criteria for individuation and for numerical identity of objects. Spelke and her colleagueshave shown that babies establish representations of objects on the basis of criteria which individuate them - an object is a coherent, bounded, entity that maintains its coherenceand boundaries as it moves through space(seeSpelke, 1990, for a review) . The baby predicts the motion of objects according to principles such as that one object cannot pass through the spaceoccupied by another (Spelke et al. 1992, BaillargeoD 1990) . Most relevant to the present discussionare studies showing that babies count objects. These are of two types. In the first , babies are simply presentedwith arrays containing a fixed number of objects, say 2 of them, one after another. For example, two cups, followed by two shoes, two bottles, two hats, two pens, and so on. The pairs of objects are never repeated, so the arrays have nothing ' in common but twoness. The baby s looking is monitored , and after a while, ' s attention to each new the baby , relative to his or her array decreases

S. Carey I Reconceptualizingthe world

original looking time. The baby is getting bored. After looking time has decreased to Y2 its original level, the baby is presented with an array containing one object, or three objects. In both cases, looking time recovers to its original level. The baby notices the difference between two objects, on the one hand, and a single object or three objects, on the other. This result, or one very like it , has been obtained with neonates(Antell and Keating 1983) . In fact, the baby' s capacity to detect similarity in number across distinct arrays serves a methodological wedge into the problem of how babies individuate objects. The baby can be habituated as described above, to two objects, and then presented with an array as in figure 5, consisting of two distinct objects sharing a common boundary. Babies dishabituate to this array, showing that they perceiveit as one object, rather than two. Thesedata support the conclusion, derived from other types of data as well, that babies are not sensitive to shape or texture regularity in individuating objects; they needpositive evidenceof distinct boundaries, such as one object moving with ' respectto the other, or the objects being separatedin space.

..:..~ .111 ,.:~ ;1 ~ : 111111 . ~

Fig. 5. Test~timulu ~ of two adjacentblocksof different size, texture and color .

A second source of evidence that babies count objects derives from data showing that babies can add and subtract. Wynn ( 1992) showed four -month olds events in which a secondobject was added to an array already containing one object. An object was placed on an empty stage while the baby watched and then a screen was raised that covered the object. A hand carrying a second object was shown going behind the screen and returning empty. The screenwas then lowered, revealing either one object (unexpected outcome, even though that was what the baby had last seen) or two objects (expectedoutcome, if the baby knows I + I = 2) . Babies looked longer at the unexpectedoutcome. A further experiment showed that babies expected exactly two objects, rather than simply more than one object. In this study, the expectedoutcome was two objects, as before, but the unexpected outcome was three objects. Again , babies were bored at seeing two there, and looked longer at the

S. Carey / Reconceptualizingthe world

unexpectedoutcome of three objects. Experiments of the same sort demonstrated that babies expected3 - I to be 2, and 2 - I to be I . Whereas these studies were performed to explore the baby' s concept of number, they bear on our question as well. Babies, like anybody, cannot count unlessthey have criteria that establishindividuals to count. Babiesclearly have criteria that establishsmall physical objects as countable individuals.

That babies individuate and count objects does not show that they trace identity of objectsthrough time, that they have the representationalcapacity to distinguishone object seenon different occasionsfrom two numerically distinct similar objects. However, there are now two demonstrationsof but physically ' this capacity in infants age 4 months or younger. Spelke( 1988) showedbabies , screenA to objectsmoving behind and r~ erging from two separatedscreens the left of screenB (figure 6). An object emergedto the left of screenA and returned behind it, and then an object emergedto the right of screenBand returned behind it. At any given time, at most one object was visible, and no object ever appeared in the space between screensA and B. Under these conditions, 4-month- olds inferred there must be two objects, as shown by the fact .that when the screenswere removed, revealing two objects (expected outcome), they looked less than when the screenswere removed revealingone object (unexpectedoutcome). Baillargeon ( 1990) showedinfants two objects at once, one on either side of a screen. The babiesthen used the existenceof two numericallydistinct objectsto make senseof what would be an impossibleevent if only one object were involved. Togetherthesestudiesshow that babiesusetwo spatiotemporalprinciples to individuate and trace identity of objects: one object cannot be in two placesat the same time, and one object cannot go from one p}a(' e to another without tracing a spatiotemporallycontinuous path. In sum, infants have a concept physical object that functions as a sortal ; they have at least one concept that divides reference, that provides criteria for individuation and numerical identity . These criteria are spatiotemporal.

differencebetweenyounginfantsandadults 10. A majorconceptual Adults also look longer at the unexpected events in the experiments described above. Further , they ask how the magic tricks are done. This is

S. CareyI Reconceplualizing lhe world DiscontinuousCondition

8

+- -

- - +

I . . I .

I . . I .

~ ~

~

Steps 2 - 5 repeated

Screens introduced

~

- - +

+- -

Object 1 brought out

Object 1 returned

Object 2 brought out

Object2 returned Screens removed revealing Expected outcome or

Unexpectedoutcome

Fig. 6. Numerical identity trials , after Spelke ( 1988).

becauseadults use spatiotemporal information in just the sameway as do the infants. But adults use other types of information in establishing individuals and tracing their identity through time: property information and membership in kinds more specific than physical object. We use property information - if we see a large red cup on a window still , and later a small green cup there, we infer that two numerically distinct cups are involved, even though we have no spatiotemporal evidenceto that effect. And , as philosophers point out , our identity judgements are relative to sortals more specific than object (Wiggins 1980, Hirsch 1982, Macnamara 1986) . Imagine a junk car, consigned to the crusher. The process of crushing is a spatiotemporally continuous ' , gradual process. Any changes in the car s properties are also continuous ; it changesshapecontinuously, for example. Yet we say that at a certain point the car goes out of existence, and is replaced by a lump of metal and plastic. We trace identity relative to kinds more specific than object, kinds

S. Carey I Reconceplualizingthe world

such as car, person, tab/e. Such concepts (sortals), typically lexicalized as count nouns in languages that have a count / mass distinction , provide additional criteria for individuation and identity to the spatiotemporal criteria that apply to bounded physical objects in general, and to the general ' assumption that an object s properties stay stable over time, or change continuously. When a person, Joe Shmoe, dies, Joe ceasesto exist, even ' though Joe s body still exists. The sortal person provides the criteria for ' ' identity of the entity referred to by the name Joe Shmoe . In collaboration with Fei Xu , I have been exploring the question of whether babies represent any sortals more specific than object, or whether babies can use property/ kind information to individuate and trace identity of objects ( Xu and Carey 1993) . Consider the events depicted in figure 7. An adult witnessing a truck emerge from behind and then reenter a screen and then witnessing an elephant emerge from behind and then reenter the screen would infer that there are at least two objects behind the screen: a truck and an elephant. The adult would make this inference in the absence of any spatiotempotal evidence for two distinct objects, not having seen two at once nor any suggestion of a discontinuous path through space and time. Adults trace ' ' ' ' identity relative to sortals such as truck and elephant and know that trucks do not turn into elephants. Xu and Carey ( 1993) have carried out four experiments based on this design. Ten-month -old babies were shown screensfrom which two objects of different kinds (e.g., a cup and a toy elephant, a ball and a truck ) emerged from opposite sides, one at a time. Each object was shown a total of four times. After this familiarization , the screenwas removed, revealing either two objects (expected outcome) or one object (unexpected outcome). In all four studies, babies looked longer at the expectedoutcome. They could not use the difference between a cup and an elephant to infer that there must be two objects behind the screen. Another group of IO-month -olds was run in a parallel version of this study basedon Spelke's design (figure 6) . That is, babies were shown two identical objects emerging from the two screensa total of four times each, and the timing of the eventswas the samein the one screen/two kinds studies. Babies succeeded , looking longer at the unexpectedoutcome of one object. Apparently , babiescan use spatiotemporal information to individuate objects before they can use kind information . We have ruled out severaluninteresting interpretations of the failure in the property/ kind conditions of these studies. For example, it is not that babies

S. Carey I Reconceplualizillgthe world Different

Condition

I~II~~I UI ~~ III ~~~IIII 1~ llii~~i ~llilllll

- - +

+- -

~

4- -

-...

Screen introduced

~

Object1 broughtout Objecl1 relurnep Object 2 brought out

Object 2 returned

do not notice the difference between the two objects. In one version of the study, babies were allowed to handle each object (one at a time of course, for we didn ' t want to provide spatial information that there were two ) before beginning the events. This made no difference to the results. In another, we compared looking time to the familiarization events when the objects are of different kinds (e.g., a cup and an elephant) to looking times during familiari zation in a condition where the objects emerging from each side of the screen are of the same kind (e.g., two elephants) . Babies habituated much faster in the latter condition . That is, they noticed that the elephant and the cup are different from each other. After habituation, we removed the screen, revealing either one object or two objects. Babies in both conditions (cup/elephant; elephant/ elephant) looked longer at the outcomes of two objects (unexpected in the elephant/ elephant condition ; expected in the elephant/ cup condition ) . The preference for two objects was equal in the two conditions. Thus, although babies notice the difference between the elephant and the cup, they

S. Carey I Reconceptualizingthe world

simply do not use this information to drive the inference that there must be two numerically distinct objects behind the screen. In appears, then, that in one senseQuine was right . Very young infants have not yet constructed concepts that serve as adult -like meanings of works like ' bottle ' , ' ball ' , and ' dog' . How are the babies representing these events? We can think of two possibilities. First , the babies may actually establish a representation of a single individual object (O BJECTi ) moving back and forth behind the screen, attributing to this object the properties of being yellow and duck-shaped at some times and white and spherical at other times. The basis for such a representation could be spatiotemporal : the infants may take the oscillating motion as a single, continuous , path . A second possibility is that the baby is making no commitment at all concerning whether the objects emerging to the left and right of the screen are the same or different. That is, the baby is representing the event as OBJECT emerging from the left of the screen, followed by OBJECT emerging from the right of the screen, and represents these neither as a single object (O BJECTi ) nor as distinct objects (O BJECTi , O B JECTj ) . Supposeyou seea leaf on the sidewalk as you walk to class, and you seea leaf on roughly the same place on the sidewalk as you return from class. That may be the same leaf or it may not ; your conceptual system is capable of drawing that distinction , but you leave the question open. If the infant is leaving the issue open in this case, then why does he/ she appear surprised when the screens are removed and two objects are revealed? On this hypothesis, the longer looking time at two objects is a familiarity effect; the infant has been familiarized with instances of single objects, and thus seeing ' ' two objects is different . After all , babies can be habituated to oneness by being shown a series of objects, one at a time. Even if you were not sure whether that leaf was the same as the one you had seen earlier , if you returned to the classroom later in the day and encountered two leaves on the sidewalk, you would see this state of affairs as different from ones in which you encountered casesof single leaves on the sidewalk. We do not know which possibility is correct. The baby actually may be representing the events as if a duck-shaped object is turning into a ball shaped object (possibility one) or simply may be failing to establish representation of two distinct objects (possibility two ) . The take-home message is the samewhichever possibility is correct ; IO-month -old infants do not use the property/ kind differences between a red metal truck and a gray rubber elephant to infer that there must be two numerically distinct objects involved in the event.

S. Carey/ Reconceptualizing the world

At II months , about half of the babies we test succeed at our task. When babies do succeed, are they doing so on the basis of kind information or property information ? That is, are they representing the events as do adults , as involving a duck and a ball , or are they individuating the objects on the basis of property differences? Further experiments could bear on this question . For example, habituation studies show babies to be sensitive to color changes and size changes, but color and size are not the types of properties that signal kind differences, at least in the adult conceptual system. Would babies of the age succeeding at this task be as likely to infer two objects when shown a blue and red cup, or a big and small cup , emerging from either side of the screen, as when shown a blue cup and a blue elephant of equal sizes emerging from either side of the screen? A difference in success rate favoring the latter pair would be suggestivethat babies, just like adults , come to represent kinds of objects, and individuate objects relative to kinds . These experiments together would provide information about the developmental course of this repres ~ntational capacity . It is significant that babies begin to comprehend and produce object names at about 10 to 12 months of age, the age at which they begin to use the differences between cups and elephants to individuate objects. Again, this pattern of results is consistentwith the Sortal First hypothesis. That is, babies do not seemto learn words for bottlehood ; they begin to learn words such as ' bottle' just when they show evidencefor sortal concepts such as bottle which provide conditions for individuation and numerical identity . Current studies in our lab are exploring the relations between specific words understood and successat individuation basedon the kinds expressedby those words. It is not surprising that babies use spatiotemporal information before kind information to individuate and trace the identity of objects. All physical objects trace spatiotemporally continuous paths; no physical object can be in two placesat the same time. However, what property changesare possible in a persisting object depends upon the kind . An apparent change of relative location of the handle to the body of a ceramic cup signifies a different cup; an apparent change of relative location of a hand to the body of a person does not signify a different person. In sum, these data suggest that babies have at least one sortal concept innately - physical object. Their object concept provides spatiotemporal conditions for individuation and numerical identity . They can use spatiotemporal information to identify individuals in their environment, and can then learn more specific sortals for kinds of theseobjects. Exactly how this is

166

S. Carey uaJizing theworld I Reconcept

is the big question, of course. The presentdata suggeststhat accomplished most of their first yearof life on this accomplishment . they spend II . A few

eRne ' odin! remarks

Where does this leave us vis-a-vis the continuity assumption? The major discontinuity posited by Quine and Piaget does not receive support; there is no reason to think that babies lack the logical resourcesto representsortals, and indeed, object functions as a sortal at least from 4 months on. But if the interpretation of the Xu and Carey data suggestedabove is correct, then an important Quinian discontinuity is supported. Babies may be setting up a representation of an object which sometimesis round , white, and styrofoam and at other times red, metal, and truck -shaped. This is a representational systemvery different from yours and mine. My story is not complete. I do not know if Xu and I are interpreting our data correctly. Also , we have as yet no account of the mechanism by which babies might begin to acquire specific sortals at around 11 months. But as I am convinced important conceptual changes occur later in life (cf. Carey 1991, Carey and Spelke, in press), I would not be shocked to find interesting discontinuities in the conceptual histories of infants, even in arenasso closely implicated in languageas the conceptual underpinnings of count nouns.

References Antell , S. and DiP. Keating, 1983. Perception of numerical invariance in neonates. Child Development 54, 695- 701. ' Baillargeon, R., 1990. Young infants physical knowledge. Paper presented at the American PsychologicalAssociation Convention, Boston. Bloom, P., 1990. Syntactic distinctions. Child Language 17, 343- 355. Bowerman, M ., 1978. The acquisition of word meaning: An investigation into some current

conflicts. In: N. Waterson . New York: . C. Snow (eds.). Developmentof communication Wiley. . Knowledgeacquisition: Enrichmentor conceptualchange ? In: S. Carey. Carey. S.. 1991 R. Gelman(eds.). The epigenesis of mind: Essaysin biology and cognition. Hillsdale. NJ: Erlbaum. . Domain specificknowledgeand conceptualchange . In: L. . in press Carey. S. and E. Spelke Hirschfeld. S. Gelman(eds.). Cultural knowledgeand domain specificity . Cambridge . UK : . CambridgeUniversityPress . Perceptual Cohen. L.B. and B.A. Younger. 1983 in theinfant. In: E.K. Scholnick categorization . 197-200. Hillsdale. NJ: Erlbaum. (ed.). New trendsin conceptualrepresentation

ua/izint. the world S. CareyI Reconcept . . Early lexicaldevelopment . London: CambridgeUniversityPress Dromi, E., 1987 . Theacquisitionof syntacticcategories : Thecaseof thecount/massdistinction. Gordon, P., 1982 Instituteof T~ hnology, Cambridge , Massachusetts , MA. Unpublisheddoctoraldissertation . Evaluatingthe semanticcategorieshypothesis : The caseof the count/mass Gordon, P., 1985 distinction. Cognition20, 209-242. . Hirsch, E., 1982 . The conceptof identity. New Haven, CT: Yale UniversityPress . Cognitive . Earlyword meanings : Thecaseofobj ~ t names Huttenlocher , J. and P. Smiley, 1987 . 19 63 89 , Psychology . What's in a name? A studyof how childrenlearn Katz, N., E. Bakerand J. Macnamara , 1974 . Child Development 45, 469- 473. commonand propernames . Namesfor things: A study of humanlearning. Cambridge Macnamara , MA : MIT , J., 1982 Press . . Macnamara . A borderdispute. Cambridge , J., 1986 , MA : MIT Press . Cambridge Pinker, S., 1984 . Languageleamabilityand languagedevelopment , MA : Harvard . UniversityPress . . Word and object. Cambridge , MA : MIT Press Quine, W.V.O., 1960 W. V . . New York: ColumbiaUniversity .0 . 1969 Onto and other , , Quine essays logicalrelativity . Press . . New York: ColumbiaUniversityPress . The rootsof reference Quine, W.V.0 ., 1974 ' inductionof world . UnpubN.N. 1987 . constraints on 2 olds Onto , , meanings logical year Sola Instituteof Technology , MA. , Cambridge .Ilsheddoctoraldissertation , Massachusetts . Inferences aboutthe meaningsof nouns: The relationshipbetweenperception Sola, N.N., 1992 29- 45. and syntax. CognitiveDevelopment . Ontological categoriesguide young children's , 1991 Sola, N.N., S. Carey and E.S. Spelke : Objecttermsand substance terms. Cognition38, 179-211. inductionsof word meaning . In: L. . Weiskranz(ed.), Thoughtwithout E.S 1988 The of , , Spelke : origins physicalknowledge . , 168- 184. Oxford, UK : Oxford UniversityPress knowledge . CognitiveScience14, 29- 56. . Principlesof objectperception , E.S., 1990 Spelke . . Origins of knowledge Macomber and K. Jacobson E.S. K. J. , 1992 , , , Spelke Breinlinger Review99, 605- 632. Psychological . . Cambridge . Thoughtand language , MA : MIT Press Vygotsky, L.S., 1962 . MA : Harvard Press . and substance . D. 1980 Sameness , , , University Cambridge Wiggins . Addition and subtractionby humaninfants. Nature358, 749- 750. Wynn, K., 1992 . Infant metaphysics : The caseof numericalidentity. MIT Centerfor Xu, F. and S. Carey, 1993 . Science Occasional Paper Cognitive

) 169- 196. North-Holland Lingua92 ( 1994

Explanation , association , and the acquisition of word meaning * FrankC. Keil , CornellUniversity , 223 Uris Hall, Ithaca, NY /4853-760/ , USA Departmentof Psychology A newly emerging view of concept structure, the concepts-in -theories view, suggeststhat adult conceptsare intrinsic mixes of two different sorts of relations: (a) those involving domain-general tabulations of frequenciesand correlations and (b) those involving domain-specific patterns of explanation. Empirical results from early cognitive development suggestthat , by the time first words are acquired, most concepts have this intrinsic mix even though changesin the nature of the mix can produce marked developmental changesin apparent concepts, word meanings, and their use. . The concepts-in -theories view suggests that the sorts of constraints needed to model the representation and acquisition of concepts cannot be based solely on either perceptual or grammatical bases; they must also arise from blases given by specific patterns of explanation, patterns that may depart from standard notions of intuitive theories. These in turn suggest different views of possible constraints on the acquisition of word meaning.

I . Introduction

Categorization is one of the most common and salient of human cognitive activities, and many of the categoriesso fonned appear to be shared among individuals, a pattern that is highlighted by the use of common words to refer to those categories. In experimental psychology, the distinctive aspects of mental life that enable each categorization are usually thought of as concepts. Sharedmental structures are assumedto be constant acrossrepeatedcategorizations of the same set of instances and different for other categorizations. When I think about the category of dogs, a specific mental representation is assumedto be responsiblefor that category, and roughly the same representation for a later categorization of dogs by myself or by another. The . Much thanksto two anonymousreviewersand to Dan Simonsfor extensivecommentson . Muchof the research earlierdraftsof this manuscript reportedon in this paperwassupportedby . NIH grant numberROl-HD23922 - 3841/94/$07.00 @ 1994- ElsevierScienceB.V. All rightsreserved 0024 SSDI 0024- 3841( 93) EO049- D

F.C. Keill Wordmeaning

phenomenonof categorization, coupled with the notion that repeatedcategorizations of the samesort are basedon the samemental representation, is the launching point for most psychological investigations of what conceptsare. Categorization, however, must not be equated with heuristics and other procedures that provide rough and ready identification of instancesabove a modest confidence level. Although one might use hair length as a rough means for identifying human sex at a certain confidence level, a careful and deliberative categorization of humans by sex would make little note of such an attribute . Careful, consideredjudgements of membership may emphasize different aspectsof mental structures than the fastest and loosest means of identifying members of categories. In this paper, these different facets of categorization behaviors are considered together in terms of their implications for models of the acquisition and representation of word meaning. Psychologicalresearchon concepts in the last three decadesstarted largely with the phenomenonof categorization, first with teachingartificial categories , and later with the study of more naturally acquired ones. But psychological viewsof conceptshave undergonetwo dramatic shifts, initially as a consequence of uncoveringmore details on categorizationbehavior, but increasinglyalso as a , measuresthat are starting to consequenceof using other behavioral measures raise questionsabout what conceptsare in the first place. My purpose here is to explore the consequences of the most recent shift for word meanings. I will argue for the following points : ( I ) The currently emerging view, the concepts-in -theories view, must ultimately characterizeadult conceptsas intrinsic mixes of two different sorts of relations: (a) those involving domain-general tabulations of frequencies and correlations, such as done by associative models and many connectionist systems, and (b) those involving domain-specific patterns of explanation, usually of a causal nature. 2 ( ) Empirical results from early cognitive development and first word meanings suggestthat by the time the first words are acquired, most if not all conceptshave this intrinsic mix even though changesin the nature of the mix can produce marked developmental changes in apparent concepts, word meanings, and their use. 3 ( ) The intrinsic mix and its early appearancesuggestsa different kind of ambiguity in word meaning, wherein largely overlapping setsof instances have the same label .applied to them but have different meanings, allow different patterns of induction , provide different categorizationsof critical test cases. Unlike classical lexical ambiguities there is no dramatic change in the class of referents even as there is a discrete change in meaning.

F.C. Keill Wordmeaning

(4) In consideration of thesefirst three points, the sorts of constraints needed to model the representation and acquisition of conceptscannot be based solely on either perceptual or grammatical bases; they must Also arise from blases given by specific patterns of explanation that emphasize different causal roles for the same properties in different domains. These conceptual constraints in turn suggest different views of possible constraints on the acquisition of word meaning.

The emergence of the concepts-in- theories view of concept structure

In the past three decades, models of categorization and concept structure have evolved through three stages. It is useful to briefly describe this history as a way of better understanding the current view. The first view tried to specify concept structure as lists of singly necessary and jointly sufficient features. Conceptssimply consistedof compartmentalized sets of features that were accessedas a cluster by a processing agent. Not surprisingly, this sort of representational theory of concepts was easy to model in a variety of computational systems. In addition , the first view fit nicely with some early feature-based models of word meaning, where again, each word consistedof a set of features that were singly necessaryand jointly sufficient (e.g. Leech 1974) . In psychology, these feature-list views survived far beyond their natural lives by focusing on laboratory learning of arbitrary ' ' ' ' concepts such as bliks being large blue triangles. Such learning tasks uncoveredsome strategiesused to encodearbitrary conjuncts and disjuncts of meaninglessattributes, but provided little insight into the acquisition and representation of natural concepts. The necessary and sufficient features model still has limited appeal becausesome word meaningsseemto logically entail others in a way that suggestsan underlying concept structure with necessityrelations; but several other models of concepts and word meanings can capture such entailments without embracing the idea of concepts being solely comprised in that manner (Chierchia and McConnell-Ginet 1990) . The first view yielded to a family of views collectively known as the probabilistic approach, in which concepts were said to reflect real world correlational structure and feature frequency (Rosch and Mervis 1975) . That approach spawned a huge body of studies arguing that concepts themselves are ill -defined probabilistic representations, consisting of weighted features, samplings of exemplars, or even combinations of dimensional values (Smith and Medin 1981) . Even as these studies collectively embraced probabilistic

F.C. Kei/ / Word meaning

accounts and were united in focusing on such phenomenaas instanceshaving different goodness of membership in a category, they began to diverge on what concepts actually were. Feature-based models" tended to be like the earlier necessaryand sufficient feature view except that they added weights to each feature representingits relative importance in detennining membership. Importance was usually derived from how often a feature co-occurred with members of the category in the real world . Correlations between features were also often included in these representations. One consequenceof these weightings was that a much larger number of features could be included in a representation and thresholds could be set up for excluding features with sufficientlylow weights. Thesethresholdscould also be adjustedas a function of contexts (as in Lakotrs hedges' technically speaking' and ' loosely thinking ' , which were thought to raise and lower the threshold respectively ( Lackoff 1972). Nonetheless, these views continued to compartmentalize concepts as distinct from ,the rest of knowledge. Certain phenomena related to typicality of instances were considered central to understanding conceptsand were shown to be strongly interrelated. These included: intuitions that membership in a category is not all or none but graded such that some instances are better members than others, intuitions that some entities were at the borderline and their membership in a category was indeterminate, and reaction times to identify instances(it takes longer to verify membership of an atypical vs. typical member). All of these responsesseemedto convergeon mental representationsthat encodedfeature frequenciesand correlations. This second view required more subtle refutations than the first. Concepts still seemed to have components that were different from their merely probabilistic parts; and demonstrations appeared showing that people could easily think about typical and atypical instances for completely well-defined categories, such as odd numbers. Such findings suggest that typicality and well-definednessare not mutually exclusive for the sameconcept (Arm strong et al. 1983). Indeed the probabilistic phenomena associated with concepts were sometimes relegated to the realm of ' identification procedures' for picking out instancesof concepts rather than to the concepts themselves. Additional concerns came from demonstrations of illusory correlations in the social and clinical psychology literature. That is, people will see nonexisting feature correlations when they follow from prior beliefs (Chapman and Chapman 1969) . Simple tabulations of feature frequenciesand co-occurrences in the world are not enough. In developmental research as well, conceptsemergedin ways not explainable by mere shiftings of either feature

F.C. Keitt Wordmeaning

frequencies or correlational weights. Thus, one set of studies show how developmental shifts in what are regarded as legitimate members of a kind , such as being a tiger or being an uncle, cannot be plausibly modeled by shifting probabilistic weights on features or changing global criteria of what frequenciesare to count as important (Keil 1989) . Other problems were seenin adult processingstudies in which two concepts with the same highly typical feature nonetheless placed greatly different emphasison those features becauseone played a much more central role in patterns of causal explanation. Thus, bananas and boomerangs were judged to be equally typically curved, but straight bananaswerejudged to be a much better new member of the class of bananas than straight boomerangs in the classof boomerangs(Medin and Shoben 1988) . This particular shapechange seemsto influence what it meansto be a boomerang to a much greater degree than a banana and does so in a way not predictable from feature frequencies or correlations. Problems of this sort with probabilistic models have led to what is known as the concepts-in -theories view (Murphy and Medin 1985) . A closer look at categorization, and increasingly at other phenomenathought to be related to concepts and their structure, has put a central emphasis on notions of explanation, mechanism and cause. People do not simply note feature frequencies and feature correlations ; they have strong intuitions about which frequencies and correlations are reasonable ones to link together in larger structures and which are not. Without theseintuitions , people would make no progressin learning and talking about common categoriesgiven the indefinitely large number of possible correlations and frequencies that can be tabulated from any natural scene. These intuitions seemmuch like intuitive theories of how things in a domain work and why they have the structure they do , hence the concepts-in -theories label. One on the most frustrating drawbacks of the current concepts-in -theories view is the lack of consensuson what theories really are and how they are mentally represented. Theories might be sets of propositions with causal connectivesinterspersedamong various logical connectives, all interconnected by deductive chains. Alternatively , theories might be more like mental models, with image-like notions of mechanism. Or , theories might not need any notions of cause whatsoever, just a more general senseof explanation; indeed some have decried the appearance of causal relations in any wellspecified theory (Hempel 1965, Russell 1924) . These controversies, however, should not blur a simple fact : some notion of explanatory coherence does have a powerful and overarching influence on almost any behavioral measure

F.C. Keill Word meaninR of concept structure. Above and beyond the details of representationalformat lies the common realization that our concepts for dogs, diners, and dandelions emphasizesome features and relations over others becauseof their fit within broader explanatory patterns.

3. Hybrid vigor

The increasedimportance of explanatory systemsin concept structure has suggesteda complementary need for uninterpreted probabilistic tabulations as part of concepts, resulting in a hybrid structure. For all but the most contrived cases, concepts may be intrinsic mixes of both systemsof explanation and atheoretic tabulations of properties. Any detailed account of the theories or explanatory systemsthat embed conceptswill end up confronting the same issue, regardlessof their particular approach. Such systemscannot fully interpret the raw data of experience, yet must rely on it . The limitations of explanatory systemsor theories in organizing properties and concept are best understood by a concrete example. Consider someone who has had a limited exposureto birds ; let us call him Icarus. Becauseof his limited experience, Icarus' s explanatory knowledge specific to birds is modest. In addition to knowing many general things about animals that apply to birds .as well, Icarus only knows explanatory reasons for properties corresponding to two sorts of bird category contrasts: flightless vs. flying birds and predators vs. prey. Thus, Icarus understandswhy flightless birds tend to have much smaller wings compared to body size than flying birds and why their legs tend to be bigger and more muscular; and he understandswhy predators have many properties that help them catch prey and why prey have many features that help them avoid predators. In this way explanatory knowledge helps order the conceptual space and helps influence intuitive inter -birds similarities, inductions about new properties, and categorization. Moreover, this knowledge helps constrain which correlations are first noted for novel instances. There is, however, a problem with birds that occupy the same spot in this two-dimensional spaceout which are easily distinguishable in terms of surface feature probabilities and correlations and which Icarus can in fact distinguish, such as a robin vs. a wren. They occupy the same place in Icarus' s explanatory matrix and yet Icarus stores additional information about them so as to be able to distinguish them. This information may well be largely associative and guided by broad hunches laid down by Icarus' s theory of animals and

F.C. Keill Wordmeaning

175

perhaps some general hunches about birds. Those hunches insure that even the associatively organized features will not include whether a bird was first seenon a tree with an odd vs. even number of branches, or whether it was facing left or right , or whether it was old or young. But , within those general constraints, many other regularities must indeed be stored purely in terms of correlations and frequencies. In this way, associative structures in concepts can persevere. This perseveranceis inevitable, simply becausewe can never have explanations for all regularities that we observe. It is tempting to try to exclude frequency and correlation information from the concept proper and say that they are only part of heuristics and identification procedures having to do with concept use; but such a move fails to explain why our beliefs are often directed specifically towards explaining those correlations; the correlations form an integral part of the system. Only in special caseswhere we have conventionally defined concepts with accidentally correlated features, such as the concept odd number, do the two aspectsbecomemore fully separable. There must be not only explanatory structures that narrow down an indefinitely large number of features and feature relations to a manageablenumber but also mechanismsfor storing information that is outside the ken of current explanation. We could never acquire new explanations if there were not some way of storing information external to current explanations. " These issueshave often led to proposals of a core- periphery distinction in concept structure, wherein typicality information is relegatedto the periphery (see discussions in Arm strong et al. 1983 and Rey 1983) . Although this distinction may imply that the periphery is less important , or even optional , the relation between the two facets of concepts may be more symbiotic. Explanations do not amount to much if they do not have anything to explain, and raw tabulations quickly overwhelm any informati ~n gathering system if it does not partially order that information in terms of explanatory usefulness. A parallel casecan be made for scientific theories. Scientific theories are sometimes presented as a set of tightly connected laws linked through deductive chains (e.g. Hempel and Oppenheim 1948), but in practice, they start with sets of regularities that they seek to explain, regularities that are noticed and remembered often before the fledgling theory has any way of incorporating them. The atheoretical component of concepts should therefore not be relegated to a ' periphery' that is not part of the concept proper . There is no general way to justify such a move despite those cases of a few formal and conventional terms where a clean distinction is possible. However, even as the

F. C. Keill Word meaning

atheoretical component must be part of the concept itself, it remains distinct. There is not an association-to- theory continuum consisting of increasingly strong and reliable correlations and frequencies. Such a continuum could not explain caseswhere high frequenciesand correlations are clearly understood as explanatorily irrelevant to membership in the category, whether it be hair length and gender or economic value and type of chemical element. Although explanatory systems may always need some reserves of correlational and frequency based information for interpretation, not all of the information in such reservesis always found meaningful, even when broad hunches suggest it might be.

4. Theorigi- of the hybrid The idea that most adult conceptshave a hybrid structure has not led to a consensuson how this relation comes about in development. Two sharply contrasting views predominate: ( 1) The earliest concepts are independent of theories and are therefore pure casesof the probabilistic models. The earliest concepts are constructed solely by tabulations of feature frequenciesand correlations presumably with the help of low -level, domain-general perceptual and cognitive blases. Early word meaningsat this time would be similarly basedon such tabulations. (2) Even the earliest conceptsmay be subject to organizational principles that go beyond tabulations of features. Primitive systems of explanation embed concepts even in infancy and certainly by the time of the first words ; or at least young children' s concept-related behaviors are only understandable by imputing to them mental structures like our own explanatory systems. This debate has proved difficult to resolve. Recent studies would seem to favor the explanations-from -the-start alternative as younger and younger children' s judgements are shown to apparently be governed by such explanations . Yet , advocates for the opposite view need merely claim that the acquisition of such explanatory systems occurs very early on in infancy (McClelland , forthcoming), a reply that becomesincreasingly difficult to test as ever younger infants are needed. But the tension between these views has resulted in an important empirical discovery. There has been a rapid downward march of the ages at which concepts are endowed with explanation, from five-year-olds postulated as purely associative beasts ( Vygotsky 1962,

F.C. Kei/ l Wordmeaning

Werner 1948, Quine 1977), to preschoolersbeing in such a state, and now to infants before their first words. Indeed, a strong casecan be made that four month-old infants categorize aspects of their physical and social worlds in ways only understandableby attributing to them domain specific systemsof causal explanation (e.g. Spelke et al. 1993, Leslie, in press) . The debate continues about even earlier origins (e.g. Slater 1993), but there is now little ' doubt that the child s first words must be sensitive to the same sorts of relations that have causeda dramatic shift in how adult conceptsare viewed. In my first attempts to understand how word meaningsmight changewith ' ' development, I posited a characteristic-to -defining shift in the acquisition of word meaning (Keil and Battennan 1984) . Following decades of vaguer claims about shifts from such things as holistic to analytic meanings, accidental to essentialfeatures and the like , one more testable possibility seemedthat early word meanings are very much like the probabilistic views of concepts, ' ' but then shifted to be more defining such that a simple principle characterized meaning. The predominant method was to present scenariosin which an instance had either all the characteristic features associatedwith a category but lacked critical defining ones or in which instances had all the critical defining features but had many highly uncharacteristic ones as well. Being an ~uncle' was asked either of friendly , gift giving , middle-aged men unrelated to ' one' s family or of brothers of one s parents who were still children and ' ' distinctly unfriendly . Being an island was asked of peninsulaswith beaches, palm trees and burled treasures or of cold , forbid ding beachless places ' surrounded by water on all sides. Children s judgements of instancesdid shift with age and on a domain by domain basis as shown in figure I (Kei I1989 ) ; but it becameincreasingly clear that a true characteristic-to -defining shift was not and could not be occurring. Even the youngest children were never simply tabulating up all salient feature frequenciesand correlations. No child thinks that uncles must have glasseseven if all the uncles they happen to have seen wear them. The features selectedby even the earliest word learners were always constrained by somenotions of reasonablenessfor the kind of thing in question. A second problem was that true definitions are rare occurrences in word meanings. Only a handful of words even approach clear simple definitions. Putnam ( 1975), estimated a few hundred in English, and even these are under contention (Lakoff 1987) . Thus, although the developmental changeis robust and easy to produce experimentally, it cannot really indicate a shift from one representational format to another. Rather, it appears to be two related phenomena: ( I ) There are increasing elaborations of explanatory systems

F.C. Kei/ I Wordmeaning

~ + d ",~-~-"+ . I C / u c / + d ~ /,-~-~-"-+ _ c / ~ + d ,K ~ ~ . . + c / ~ -..-".'~ c + / d ,:;.-< ' " ( ' + c / ~ K -

Moralterms

Mealterms

Tool terms

Kinshipterms

Cookingterms

-

-

-c/ +d

-to- definingshift in the acquisitionof word Fig. I . Illustration of an apparentcharacteristic . The y-axesrepresentthe extent to which eachdescribedentity wasjudged to be a meaning . Thex-axesrepresent threegradeschoolclasslevelsandadults. legitimatememberof thecategory The graphsshowthat thereis a developmental shift in whetherthe + characteristic / - defining + defining ones are taken to indicate membersof the descriptionsor the - characteristicl . In addition, the shifts occur at different times on a domain by domain basis. category .) (Adaptedfrom Keil 1989 with age such that explanations play ever more extensive roles in constraining what features and correlations are noticed . Thus , as explanation - based knowledge becomes more and more elaborated , default tabulations of information in associative terms are less common . ( 2) There are shifts in understanding which explanatory system is most relevant to a class of phenomena . A child might realize that ' uncle ' is better understood in terms of set of biological relations that comprise kinship and not in terms of social relations that govern

F. C. Keill Wordmeaning

friendship; the social explanation is discovered to generate more serious mistakes and is abandoned (Keil 1989, 1992) . A similar change of perspectivehas occurred with tenDS for natural kinds. For example, young children assert that zebras who are surgically transformed to look and act like horses are no longer zebras but truly horses. Older children declare the animal to still be a zebra, suggestingadevelop mental shift from a phenomenalsimilarity spaceto a theoretically driven one. After many follow -up studies conducted both by my research group and other groups, a different story emerges(Keil 1989) . Younger children may never be total phenomenalistshelplessly buffeted about by correlations and frequencies. There invariably seemsto be an understanding of deeper relations that allows them to go beyond what would reasonably be considered phenomenal similarity . Thus, even for three-year-olds, tigers can only be changed into lions if you use a mechanism of change that' is reasonably related to being a member of the two kinds. Younger children s beliefs about such mechanismsmay differ , but they have them nonethelessand will use them to override associativeinfonnation . At any age most natural concepts have a mix of associationistic tabulated information and systemsfor interpreting , explaining and guiding the pickup of that information . There can be dramatic developmental change, but not from one kind of representational system to a dramatically different one. Instead,- the predominant change is in how extensively the child is able to interpret and explain the raw data of association. As those explanations become more and more elaborated over time, the child has to fall back less and lesson the associativecomponent to make judgements. ' Following the Icarus example offered above, suppose a child s initial . understanding of birds involved only notions of what properties supported flight and thereby clusters of features corresponding to flightless vs. flying birds but did not understand how to organize the features that clustered around predator vs. prey. Some more general principles about animals may bias the child to associatively store frequenciesand correlations concerning featuressuch as feet shape, beak shape, eye location in head and typical diet ; but these properties may be stored largely in tenDS of associative relations until the predator/prey insight comes to dramatically organize those features and shift similarities accordingly, thereby shifting induction and categorization . Note that the developmental studies also support the essential hybrid structure of concepts. Even as these studies show that the youngest children never rely solely on brute force tabulations of feature frequencies and correlations, they also show that something like an associativecomponent is

F. C. Keil / Word meaning

also present that shrinks with age in each domain in the face of more elaborated explanation. Although the characteristic:-to -defining change in development is not strictly correct, the changesthat do occur support both aspectsof the hybrid .

S. Multiple domains andreferentiallyoverlappingambiguities The concepts-in-theories view suggestsmultiple modes of explanation from the start of acquisition of word meaning. At least two domains have been clearly implicated : (a) a physical-mechanical domain that helps explain the properties of kinds subject to mechanical causality; and (b) a folk -psychological domain that helps explain the properties of kinds subject to beliefdesire accounts of causation. These radically different ways of construing the . world result in sharply contrasting assumptions about what properties are central to membersof a kind . Carey ( 1985) was one of the first to make this proposal, and the evidence for early and universal appearanceof these two domains is now strong and ' widespread, ranging from infants clear intuitions about the mechanics of physical objects to findings that autistic children who have excellent intuitions about physical mechanics can be grossly deficient in their understanding of belief- desire causality. Severalstudies have now shown that high-functioning autistic children can reason about events that do not require belief- desire inferences, while having great difficulty reasoning about events that do require belief- desire inferences (e.g. Baron-Cohen et al. 1985) . One recent study suggeststhat the contrast between the two domains might already be under way in the first few months of life. Infants as young as six months have been shown to have markedly different expectations about the causal interactions among social entities as opposed to non-social ones. If they observe events in which a six foot high cylinder rolls towards another, they expect that only physical contact can produce a movement in a second one. If the first cylinder stops, leaving a gap between it and the second one, and the second one th~n moves off as if launched without contact, the infants (and adults) find the event anomalous. If , however, the same events occur with human actors, the infants find launching of the second person without contact to be no more surprising than launching with contact (Spelke et al. 1993) . If the young child initially has only these two fundamental modes of explanation, most concepts must be somehow assimilated into one of these

F.C. Keill Wordmeaning

two ways of understanding the world . This possibility results in predictions concerning not just the nature of early concepts but also how they become linked to the lexicon. If children are presentedwith partial information about a novel kind , they must.decide whether its known and future properties are to be interpreted in belief/ desire terms or in physical/ mechanical terms and ensuinginductions about properties will vary accordingly. For example, it has been argued that biological kinds and their properties are often interpreted in solely behavioral terms. The behavioral aspectsof a property come to be seen as the basis for it . Young children have been shown to attribute properties such as eating and having babies only to animals that are sufficiently psychologically similar to humans so as to have the behavioral and belief/ desirecorrelates of eating and having babies (e.g. feeling hungry and wanting ' food and being nurturant towards offspring) . Worms don t eat becausethey cannot have feelings of hunger and desires for food in any way like humans (Carey 1985). These sorts of misattributions were found not only forproper ties that have salient behaviors associatedwith them but also for unfamiliar ones described on the spot for the child (such as having a spleen) . Young children seemto assumethat even novel properties are likely to be possessed by other animals to the extent that they are behaviorally similar to the one on which they are taught. Only later in development would an appreciation of biological kinds as such emerge and would new properties be based on induction over sets of functional biological relations. It now seemsthat young children might not be so narrowly restricted in their earliest kinds of understanding, that the explanation part of the hybrid has more diversity than just a mechanics and psychology. For example, a wide range of studies now converge to suggest that preschoolers see living kinds as having their own causal patternings distinct from physical mechanics and social behavior (Keil 1992, Inagaki and Hatano 1993, Hatano et al. 1993, Springer and Keil 1991, Gelman and Gottfried 1993) . They understand that both plants and animals tend to have functional /adaptive explanations associated with them that are not seen elsewhere in the natural world and which are distinct from the sorts of functional explanations used with artifacts. For example, preschoolers show a stronger tendency to explain the properties of living things in terms of the purposes those properties serve for those things than they do for non-living things (Keil 1992). Adaptive , or design, explanations have becomerecognizedas a distinct form of explanation linked to the biological sciences , and it now appearsthat salient to young children available and is the same form of explanation (Woodfield 1976, Wright 1976) .

F.C. Keil / Word meaning

One way to show this distinctive pattern of expectationsconcerning living things is to ask preschoolers which of two explanations they prefer as appropriate for explaining the properties of living vs. non-living things. For ' example, the children might be told the following : Two people are talking about why plants are green. This person says it is becauseit is better for the plants to be green and it helps there to be more plants. This person says it is becausethere are little tiny parts in plants that when mixed together give them a green color. Which reason is a better one?' The same question would then be asked about emeralds. Although both the reductionist and functional explanations are appropriate for the plants, children preferred the functional sorts of explanations for living things while at the same time preferring the reductionist explanation for non-living things (Keil 1992). The contrast with artifacts is more subtle, but involves an understanding that most of the properties of non-domesticated living kinds are ultimately self-serving whereas most of the properties of artifacts are other -serving. This seemingly abstract notion appears to be well within the grasp of preschoolers when they are asked to choose between explanations for functional properties that are for the good of the object vs. the good of another entity ; and they show similar understandings for plants as well as animals, ruling out any need to see the living things as . capable of mental goals or desires. The full set of such early modes of construal may be quite small, possibly as few as half a dozen. The earliest modes of construal seemto originate from notions of broad domains of phenomena, such as physical mechanics, folk psychology, and functional /teleological explanations and not local areas of expertise such as dinosaurs or chess. Other possible basic modes include: moral reasoning, notions of ownership and transfer <;?f ownership, and social power relations. The existenceof other domains besidesmechanics, psychology , and design/teleology is speculative at this point , but remains plausible given the likely universal presenceof such domains and the possibility that they have distinctive fonDS of causal explanation. Alternatively , the basic three might h~ve a special status making them more fundamental than all other schemasfor making senseof the world . A major challenge for future work is to develop criteria that demonstrate how these basic modes of construal are distinct from the thousands of local explanations and mental models that people come to use in everyday tasks. In addition to different roles in processing, the basic modes may be earlier emerging and more invariant over the course of development.

F.C. Keill Word meaning

The early presenceof multiple forms of explanation may lead to a distinct kind of ambiguity for lexical items. The ambiguity arises from caseswhere heavily overlapping sets of individuals are meaningfully interpreted by different forms of explanation. For example, a class of people can often be understood in biological, social, or physical terms, just as properties of c.omputers can be understood .in mechanical, functional , or psychological terms. A computer might be described as failing a task becausetwo routines are seen as ' colliding ' (mechanical), becausethe programs are not designed for the task (functional ), or because the program cannot understand or remember the correct sequences(psychological) . Clearly distinct meanings can be invoked even though the real world referents may often be heavily overlapping. These examplesappear to be different from casesof vagueness or more standard lexical ambiguities where the referents are usually nonoverlapping ' ' ' ' , as with the two meanings of such terms as bat , bank , and ' tank ' . With standard lexical ambiguities, one gets the immediate impression switches meanings, whereaswith referentially as one of a new set of referents overlapping ambiguities, the meaning seemsto switch equally strongly but many of the same referents remain. Referentially overlapping ambiguities suggest an alternative account of many developmental changes in word meanings. Children might not be reworking the feature sets that organize members of a category and discarding the old (a pattern that has the counterintuitive consequenceof suggestingthat adults should have difficulty accessingearlier states of their own vocabularies) but rather realizing the relevance of a different, but already present, system of explanation for interpreting roughly the same properties. In the case of kinship terms, a preschooler might well have notions that there are ways of understanding things in social as well as in biological/ functional ways, but may have assumedearly on that instances of kinship terms were to be understood social/ behaviorally. A change in word meaning, to more classic bloodline relations might reflect not so much a new learning of biological relations, but an insight that a biological/ functional way of explaining properties and judging similarities is the most common one in adult usages of those terms. This speculation gains support from recent studies showing that , although younger children might have different default assumptions about properties for living things, modest contextual cues can result in biologically basedinductions. For example, when children judge that dogs and humans but not birds, snakes or insects eat, it appears as if they ' ' might have encoded eats as referring to social and belief/ desire aspectsof

F. C. KeU / Word meaning

eating. Yet the same child will also judge that eating applies to all and only animals if they are primed to think about terms like eating in biological/ functional terms ( Vera and Keil 1988) . Those studies are relevant because they illustrate how a property can alternately be embedded in two very different systems of explanation. Comparable switches in embedding for kinship terms could cause the sorts of referentially overlapping ambiguities proposed here. Apparent developmentalchangesin word meaning may therefore reflect an awarenessthat a different set of explanatory relations can be superimposed on roughly the same set of instances and properties so as to afford new insights. The different explanatory system may have been present for some time, but is now decided relevant to a class of lexical items in a certain domain. Shifts betweenmodes of explanation occur in adults as well. For example, most of the time we understand the personality traits and behaviors of others in terms of belief/ desire patterns of causation. John is afraid of snakes becausehe had some bad experienceswith them that led him to .believe that snakes are dangerous to him , even if those beliefs are false about most snakes. But we can understand the same traits in biological/ functional terms as well. One would then explain John' s fear of snakes not in terms of his beliefs and desires, but in terms of the ecological niches occupied by humans and their evolutionary predecessorsand of a need to develop an adaptation that causesa built -in fear of snakes. Although explanations of behaviors in biological/ functional terms as opposed to belief/ desire ones may not be the first option taken, once adopted, they lead to different inferences and understandings. If shifting modes of explanation over roughly the samereferents provides a senseof ambiguity, there should be other signs of that ambiguity as well. One example might be a tendency not to switch betweenmeaningswhen a word is shared in a clause. In the sentence' The pilot banked at the intersection of Pine and Locust and the lawyer at Oak and Walnut ' , it is bizarre to have ' bank' being used first as an airplane maneuver and second as fiscal action. ' ' Similarly, in the sentence The rock hit the house and John did too , it is ' awkward to have ' hit be in the mechanical/ billiard -ball sensefor the rock and in the intentional sensefor John. One is more inclined to seeJohn hitting the house in an inanimate senseas a hurtling body not intending the impact. " As a second example, consider the sentence The daffodil neededwater and ' ' ' John did too , where need has the senseof a physiological need in both cases, not as physiological need and explicit desire or wanting in the other. If

F.C. Keill Word meaning

185

this analysis is correct, a child hearing an unfamiliar tenD can safely assume that when a word is shared in usage, the mode of explanation is likely to remain constant. The awkwardness of mixing different sensesseemsless marked for other casesthat are not true ambiguities. For example, with metaphor, one can say ' ' ' ' Phil embraced her body, and Bill her ideas , where embrace clearly has different senses , but little awkwardness results. Similarly , with syncategore' ' matic terms like good , different sensescan be mixed in the sameclause, as in ' 'The meal was good and the fire was too . If thesesorts of intuitions prove to be more general, they suggestthat shifts in explanatory frameworks are not to be confused with a host of other possible context effects that cause more subtle shifts in meaning (Barsalou 1987, Lakoff 1987) . There are dozens of ' ' different nuancesof the word instrument depending on context, as there are ' of the verb ' cut ; but these do not seemto reflect deep shifts in explanatory frameworks.

6. Explanation - based constraints on concepts and word meanings

One possible consequenceof the concepts-in -theories view in concert with multiple early domains and shifting modes of construal is that causal explanatory blasesmight constrain inferencesabout early word meanings in ways that have sometimesbeen assumedto be more in the province of perceptual blases. Thus, several researchershave suggestedthat young children might ' have a ' shape bias in terms of how they approach the meanings of novel nouns, especiallywhen used as count nouns (seeLandau, this volume) . Such a bias might well exist, but its origins may not be simply in the form of a perceptual rule, but rather in terms of what sorts of properties are more central to understanding different sorts of kinds. Fundamental modes of explanation might influence the shapebias from the earliest times of lexical acquisition. Thinking of something simply as a physical object and in terms of its overall paths of motion should actually reduce the shapebias, whereasthinking of it as a biological functional thing, or an artifactual functional thing should have a stronger impact on the shape bias. Thus, the simplest naive physics envisions objects as point massesand predicts their trajectories accordingly with little regard for their shape. Heavy things push lighter things further than the opposite, and objects that hit another head-on have more of an effect than those that just graze them at an angle. A functional approach to objects, however, pays close attention to

F.C. Kei/ I Word meaning

aspectsof shape in making inferencesabout function. Moreover, shape itself might be more profit ably subdivided into overall shape vs. local parts and relative spatial arrangements, with one aspect more important for artifacts and the other for living kinds. There might not be any general shapebias, but rather an understanding of the different degrees of importance of certain aspectsof shapein different explanatory systems. Shape, or more properly its subtypes, might then be seen as one of several sorts of properties whose centrality can vary as a function of the kind of entities considered (seealso Bloom, this volume) . This alternative would gain support if the bias varied in strength as a consequenceof the type of explanatory system invoked by an object or category. Some evidence is already present in cases where adding eyes to inanimate objects changesthe strength of the shape bias at different ages in different ways (e.g. eyes make an uninterpretable object for a younger child becomea living kind for which shape matters a great deal ... whereasmore complex interactions with texture and color are invoked in older children (Landau, this volume . We have begun exploring this issue with both adults and children. With adults, the experimental paradigm involves describing a change in types of properties of typical members of familiar categories and asking about the effectsof such changeson the categorical and lexical status of those members. Four- changesin property types were used: color , size, shape(as specifiedby a tripling of width and a one-third reduction of height), and surface pattern. Thesechangeswere described for randomly selectedmembersof four categories : animals, plants, non-living natural kinds (e.g. rocks), and artifacts. Subjectswere given instancesof a member of a category and asked about the effects of changesof different feature types. For example, subjects might be told about some artifacts that are exactly like ordinary chairs in every way except that all members of this group are normally uniformly shaded bright pink . The subjects would then be asked if it would still be called a chair. A different item might ask about rocks that were just like gold except that they are nonnally bright pink . One group of subjects were asked whether the object with the altered feature should be assigned the standard name, one group was asked how well the altered object would function in the original role, and a third group was asked how unusual a member of the original category the altered member would be. With the exception of functional intuitions about non-living natural kinds, adults have reliable intuitions about such questions and show clear patterns of responses . Functional questions were asked of things like kinds or rocks to keep the design consistent across

F.C. Kei/ I Wordmeaning

all categoriesand to check that in fact they would be lessinfonnative for such kinds. The importance of property types varies strongly as a function of the kind of thing querled. Overall shape and size changeshad a profound impact on functional roles of artifacts as well as on their lexical labels. By contrast changes in color and surface patternings were irrelevant for comparable ratings of artifacts. For non-living natural kinds, such as minerals, the effects were roughly reversed, with color and patterning being seenas essentialto the judgements and shape and size as largely irrelevant. For both plants and animals the results were consistently similar , suggestinga set of expectations about the category of living things in general. Shapeand color were generally seenas most relevant across conditions and size as least relevant, a pattern that was clearly distinct from both artifacts and non-living natural kinds. An example of these different patterns of judgement for naming is shown in figure 2. Above and beyond the differences just described for the data set as a whole, judgements also varied as a function of the kind of question asked. Questionsabout what the altered entity should be called had different profiles across the three category types than did questions about its function or unusualness . This finding illustrates that judgements about preservation of function and typicality are not the sameasjudgements of what a thing should be called.- Such differencesmight seemobvious, but they are often conftated in experimental researchwhere results collected with one type of judgement are compared to those collected with another. Moreover, as seenshortly , the precise nature of their interactions can be informative about the roles of specific properties in particular categories. One final measurement taken in these studies provided an important finding . After all initial ratings were completed, subjects were asked to consider ordinary exemplars of each of the judged categories and rate how much those items normally vary with respect to overall shape, size color , of patterning. Thus, a subject might be asked to judge how much ordinary chairs, or lumps of gold, vary in color. These questions were designed to addressan important issuein the literature : how much the natural variability of features in a category could cause the discounting or emphasis of the importance of featureswhen querled in one of the three ways. A 3-inch disk is more likely to be considered a pizza than a quarter, even though it is more similar to the quarter in physical size, because the minimal variability of currency disallows even modest size changeswhereasthe larger variability of size for pizzas allows for an albeit implausible 3-inch pizza (Rips 1989).

8

F.C. Keill Word meaning

6

. Shape D Size ~ Surface Pattern

Ai

.E c. Ic

5

4.5

8

e

5.5

4

3.5 3

2.5 2 1.5 1 Artifact

Non -

Plant

Animal

Living NK Fig. 2. Resultsof first studyconductedby GruberthandKeil showingadult intuitionsof whether of that classby a newnameshouldbegivento a classof entitiesthat differ from nonnalmembers the mean The axis color. or size in either a represents , y , pattern shape having propertychange ratingsof the extentto which a kind or propertychangewasjudgedto causea needfor a new name(7= definitelya newname, O= not at all).

Perhaps more generally, tabulations of variability are the basis for judgements of the effectsof changing properties: the lower the variability , the more the judged impact of the change. In thesestudies, variability alone did not predict the significant effectsseen for any of the three judgement types. For example, if a subject said that color typically varied more for chairs than for whales, those judgements did not predict other judgements about category membership when the color was changedfor novel instances. However, when the data analysis was performed

F.C. Kei/ I Word meaning

in a rnanner that looked for patterns remaining after all contributions by variability were nurnerically removed, there were differencesacrossjudgement types. For judgernents of function , high variability of a feature had little influence predicting the effect$ of changing that feature. Consequently, when variability was factored out , the differences in property effects across categories were still large and robust. Even when an item' s color or shapevaries greatly in the real world , variability is largely ignored in making judgments about functional influences; instead the deviant instanceitself is examined for . By contrast, judgements about the appropriate specific functional consequences narne were influenced by variability not by allowing predictions of effects, but by eliminating differences in property effects across categories. Thus, although neither variability judgements alone, nor property -type cornbinations alone, could predict judgement of naming, their joint action produced reliable effects. In a related pattern, variability mattered more for non-living natural kinds, for which function is irrelevant, than for artifacts. In surn, in no casesare intuitions about property variability for instances adequate to predict how changes in those properties influence categoryrelated judgrnents about those instances; but in the case of naming judgernents, property variability is involved in interaction with other factors. The assignment of labels may therefore rely more on the heterogeneous aspectsof concepts than other category-related judgements since both variability information for instances and general views about differences in property importance across broad category types are involved. Variability may be information stored largely through associative means, whereas the differential effectsof property types as a function of category seemlinked to the natural explanatory systemsinvoked by such kinds as living things, artifacts, and sentient beings. Any number of simple frequency and correlation tabulation systemscan note how often features vary for instancesin a category and accessthis information through network in which activation strength is a function of those tabulations, hence the notion that associative means of representation would be adequate. By contrast, judgements about which properties are most central to a category and causethe most disruption when negatedseemto arise from notions about the causal structures underlying the property configurations of broad categoriesand how particular properties are much more closely connected to those causal structures than others. Such judgementscannot be captured by storing information about feature frequencies acrossinstancesas they rely on beliefs about mechanismand explanation. Thus, arguments made in earlier sectionsof this paper about the importance of hybrid structure may emerge with a vengeancein the case of mapping

F.C. Kei/ l Wordmeaning

concepts into the lexicon. (For other arguments about the special influences of lexical items seeMarkman 1989and Woodward 1992.) It is one thing to demonstrate that adults seemto be strongly influenced by abstract explanatory belief systemsin how they interpret the importance of properties for different categoriesof things. It is quite another to demonstrate that such systems might be implicated early in life when the lexicon is growing most rapidly . We have begun a seriesof studies addressingthis issue as well. These studies use triads of pictures in which an exemplar with a novel name is shown followed by an object that varies in the shapesof the most salient parts but which preservescolor and surface patterning and by another object that preservesall shapesbut varies in colors and surface patterning. In each case, novel items and names are clearly depicted as being from one of four categories: artifact , plant , animal, or non-living natural kind . The original object is referred to in a manner that is indeterminate betweencount ' ' and mass(e.g. ' this is an animal that is in the hyrax group and this is a rock ' ' ' that is of the malachite group or this is my hyrax ... it is a kind of animal ' ' and This is my malachite ... it is a kind of rock ) . In one preliminary study 16 three- and four -year-olds ignored shape and emphasizedcolor when deciding which of two novel non-living natural kinds shared the same label as an originally labeled one. For the living kinds and the artifacts, shape differencesas depicted by different parts are vastly more important . Color and texture also playa role for living things but not at all for artifacts where, with few exceptions, they may apparently be understood as purely conventional. This insight may get stronger as the children get older and enter elementary school, a pattern that is now under active study. One aspectof the results is shown in figure 3. A secondstudy now under way suggeststhat shapechangesnot implicating changes in salient parts are seen as less important for living kinds than artifacts because of their generally smaller roles in changing functional properties for living things. Different parts compel a senseof category change for li .ying things. We have constructed stimuli sets where a class of novel . things is distorted two ways: ( I ) a computer graphics transformation displaces the image pixels such that the object becomestwisted around its central point and thus the overall shape is dramatically changed whereaslocal parts stay pretty much the sameand in roughly the samerelations to each other ; or (2) the overall shape is not changed, but small properties are substituted for other ones, such as one kind of beak for another on a bird , or one kind of a dial for another on a machine. When preschoolers are shown these images

F. C. Keitt Word meaning V

C

'

~ ! o ! .

9

.

8

.

7

.

6

.

5

.

4

.

3

.

2

0 !

!

.

0 .

3

.

S 0

!

:

~

.

. 0 : V

'

c

0 i .

8 .

~ 0 2

!

. 0 u

0

0

'

~ c 0 ~

0 -

0 ~ 1

.

0 .

B

.

8

0 !

non - liv

animals

plants

artifacts

eo .

.

A

Fig . 3. Results of preliminary preschool study showing that shape is de- emphasizedrelative to color and surface pattern for non-living natural kinds whereas the opposite holds for living kinds and artifacts.

and told that a target image has a certain .novel label and are then asked which of the other two images (the overall shape distortion vs. local parts substitution) should also take the same label, they give different patterns of responsesfor living kinds vs. artifacts. For plants and animals, local part differences from the target are much more important than overall shape differenceswhereasthe opposite is true for the artifacts, as long as the local part changesare not seenas disrupting function . These sorts of expectations could certainly constrain how they learn the meanings of new words. Follow -up studies have used ambiguous displays where the triads are labeled in one caseas a kind of animal, and in other casesas a kind of nonliving natural kind or a machine. In these cases, with precisely the same display, generalizationsare different depending on how the depicted entity is construed. Preliminary studies with preschoolers suggest that when, for example, an entity is labeled a frog , the shapechangeis critical ; but when it is labeled a rock , color and texture becomescritical ; hence expectations of the kind of things observed can guide how properties and similarities are interpreted for exactly the samedisplays. The examples offered here raise the possibility that many constraints on early concept learning and on word learning may flow from our beliefs about the sorts of properties and relations most explanatorily central to fundamental kinds of things. Shape, like any other property , may betheory dependent for its perceived causal powers and central organizing role in a category. Very young children seemto have setsof strong expectationsabout

192

F.C. Keill Wordmeaning

the likely causal power of different properties in organizing categories at a high level of abstraction (e.g. living thing and artifact ) . However, expectations are less influential for individual items, arguing against some simple storage of concrete exemplarsand comparison back to those instances. Perhapsthese effect do not operate at finer levels of categorization. Strong effects of different causal centrality for properties may exist at levels such as living kinds, substancesand artifacts, but not at much finer levels such as birds vs. fish. That is, it may be difficult to find properties applicable to both fish and birds which, when counterfactualized, distinguish among them by having a strong effect in one caseand not in another. Note that the properties cannot be one like ' breathes underwater' since it is already counterfactual for birds. Of course, many properties are causally central to both , but the kinds of properties that are causally most potent may be pretty much the sameacross all animals and reflect general notions about the sorts of causal patterns distinctive to living things.

7. Conclusions I have argued for the following themes: Current views of conceptsrequire that they be acquired and representedin terms of a systemthat has two distinct components: one that stores frequency and correlation information through domain-general procedures, and one that carries domain-specific beliefs about causal relations that help explain property relations. This hybrid structure is well supported by a wide range of experimental studies as well as more principled considerations. The beliefs about causal patternings and property types are linked to broad domains of explanation, such as sentient beings, living kinds and artifacts. Those domains seem different from the many smaller domains of expertise that all of us come to acquire as well. The evidencehere is still accumulating but indications come from distinct sources, such as beliefs about how property changesinfluence category integrity and how ways of understanding classesof things seemto shift in development and acrossdifferent contexts. This hybrid mixture appears to be present throughout much of cognitive development, at least by the time first words are acquired and quite possibly much earlier. The evidence here arises from the failure of domain-general tabulation models on their own to explain young children' s performance on a wide variety of tasks involving concepts. There is less evidencefor just how many of these basic domains might exist in the prelinguistic child. I have

F.C. Kei/ I Wordmeaning

argued that there are good reasons to suspectat least three and speculated that as many as half a dozen more seemplausible. These patterns concerning the acquisition and representation of concepts matter directly for models of lexical representationand acquisition. Both the strategiesused to acquire word meaningsand their underlying representations themselvesare deeply affected by such views about concepts. The hybrid structure suggeststhat word meanings are linked both to frequency-based information and to causal/ explanatory beliefs and that different sorts of lexical tasks may highlight one or the other of these two facets. Empirical studies of the lexicon must therefore keep in mind these two facets and their interactions with specific tasks. Becausethe hybrid structure appears to be already present in the prelinguistic child , both facets influence the child ' s first attempts to learn the meanings of words, hence the claim that there are explanation-based constraints on the acquisition of word meaning. If early representations were only based on tabulations of feature frequencies, then strategiesthat key on some of these features, such as shape, might be the sole means of narrowing down the hypothesis space of what novel words refer to. Instead, even the first guessesabout word meanings may be biased not only by bottom -up counts of feature frequenciesand correlations, but also by top- down abstract expectationsabout what sorts of featureswill be causally central to a domain. Apparent developmental shifts in the internal representations of word meaningsrarely, if ever, are signs of a change in basic underlying representational format . Instead they may representincreasing elaboration of explanatory beliefs that are able to interpret a larger and larger percent of the tabulated information and shifts in which explanatory systemis deemedmost relevant to understanding membersof a category. Younger children may not be showing an inability to represent word meanings in certain formats as much as they are exercisingdifferent default options and having lessdetail in some of their explanatory systems. The hybrid structure and its early appearancehighlights a different kind of ambiguity in word meaning, wherein largely overlapping sets of instances may have the same label applied to them but have different meanings, with different inductions, different categorizations of critical test cases, and a clear senseof ambiguity as opposed to vaguenessor synonymy. It appearsthat the shifts of meaning here are not caused by realizing that a label points to a ' wholly different part of one s conceptual system with different instancesand feature clusters, but rather by realizing that the label may point to a class of things that can be interpreted in terms of more than one explanatory schema.

F.C. Keill Wordmeaning

Seeingconceptsas essentialhybrids of explanation and association has led to the claim that early word learners have explanatory blasesthat guide first guessesabout the sorts of properties that are most strongly causally linked to the central structure of a category. These blasesappear to be high level and abstract yet also powerful in how they guide choices of relevant properties. The acquisition of word meaning is thus suggestednot to rise out of higher and higher order tabulations of regularities among perceptual primitives, but from the start to be strongly influenced by overarching distinct patterns of explanation that are associated with such domains as biological kinds, artifacts, and sentient beings. In addition , the possibility of a handful of such patterns early on allows for much of semantic development to involve switching among these patterns and not acquisition of totally new forms of explanation. These discussionshave carefully skirted questions of what explanations or theories are really like ; but one possibility is that they are not much at all like theories in the classic nomological-deductive sense. Increasingly in the philo sophy of science, it is common to attribute to the working scientist not a set of propositions linked by rules of inference, but something less structured involving notions of causal powers (Salmon 198?) . Not only children, but also adults, often have explanations in the form of notions about the sorts of properties likely to be causally most central to a kind , with only the vaguest of hunchesabout how they are linked together. It may therefore be that the earliest notions of what properties are central to a category are not so much consequencesof a deepertheory, but are much of the theory itself. Lexical items cannot be simply equated with concepts; but neith,er can they ignore the details of concept structure. In this paper I have -argued that a current view of conceptsand concept acquisition has important consequences for how we might think of word meanings and how they are acquired. The notion of hybrid structure, its early origins, the presenceof multiple early modesof construal, and subsequentconstraints suggesta very different model of how lexical items becomemapped onto conceptsand changewith development than would be suggestedby alternative models of concepts. Even if one is solely interested in the lexicon, one cannot avoid making some commitments about concepts themselves. References . Whatsomeconceptsmightnot be. Cognition ArmstrongS.. L. GleitmanandH. Gleitman. 1983 13. 263- 308.

F. C. Keitt Word meaning Baron-Cohen, S., A .M . Leslie and U . Frith , 1985. Does the autistic child have a ' theory of ' mind ? Cognition 21, 37- 46. Barsalou, L . W., 1987. The instability of graded structure: Implications for the nature of concepts. In : U . Neisser (ed.) Concepts and conceptual development: Ecological and intellectual factors in categorization, 101- 140. Cambridge: Cambridge University Press. Bloom, P., 1994. Possible names: The role of syntax-semanticsmappings in the acquisition of

nominals . Lingua92, 297- 329(this volume). . Conceptualchangein childhood. Cambridge S. 1985 . , MA : MIT Press Carey, , L. J. and J.P. 1969 . correlation as an obstacle to the useof valid , , Chapman Chapman Illusory 74, 272- 280. psychodiagnostic signs. Journalof AbnonnalPsychology . The roleof coherence in differentiatinggenuinefrom spurious Cheng,P.W. and Y. Lien, in press causes . In: D. Sperber acrosscultureand development , A. Premack(eds.), Causalunderstandings . Cambridge : CambridgeUniversityPress . -Ginet, 1990 Chierchia . Meaningand grammar: An introduction to , G. and S. McConnell semantics . Cambridge . , MA : MIT Press . The child' s theoryof living things. Presentation Gelman, S. and G. Gottfried, 1993 at the 1993 of the for Research in Child Development . , NewOrleans Meeting Society Hatano, G., R.S. Siegier . Thedevelopment of biological , K. Inagaki, R. Stavyand N. Wax, 1993 : A multinational study. CognitiveDevelopment 8, 47- 62. knowledge . Aspectsof scientificexplanationand other essaysin the philosophyof , C.G., 1965 Hempel . New York: FreePress . science . Studiesin thelogicof explanation . Philosophyof Science , C.G. andP. Oppenheim , 1948 Hempel 15, 135- 175. . Youngchildren's understanding of the mind- bodydistinction. Inagaki, K. andG. Hatano, 1993 Child Development (in press ). . The acquisitionof natural kind and artifact tenDs. In: W. Demopoulos Keil, F.C., 1986 , A. Marras(eds.), Languagelearningand conceptacquisition , 133- 153. Norwood, NJ: Ablex. . Concepts Keil, F.C., 1989 . Cambridge , kindsandcognitivedevelopment , MA : BradfordBooks. . The originsof an autonomousbiology. In: M. A. Gunnar, M. Maratsos(eds.), Keil, F.C., 1992 The MinnesotaSymposium on Child Psychology , Vol. 25: 103- 137. Hillsdale, NJ: Erlbaum. -to-definingshift in the development . A characteristic Keil, F.C. and N. Battennan, 1984 of word . Journal of Verbal and Verbal Behavior , 23, 221- 236. meaning Learning . Hedges : A studyin meaningcriteriaandthe logicof fuzzyconcepts . Journalof Lakoff, G., 1972 PhilosophicalLogic 2, 458- 508. . Women, fire, and dangerousthings: What categoriesrevealabout the mind. Lakoff, G., 1987 . , IL : Universityof ChicagoPress Chicago Landau, B., 1994 . Where's what and what's where: The languageof objectsin space . Lingua92, 259- 296(this volume). Leech . Semantics . Hannondsworth . , G., 1974 , England: Penguin Leslie, A., in press . TOBY and TOM. In: D. Sperber , A. Premack(eds.), Causalunderstandings acrosscultureand development . Cambridge : CambridgeUniversityPress . Markman, E., 1989 . Categorization and namingin children: Problemsof induction. Cambridge , . MA : BradfordBooks/MIT Press . Paralleldistributedprocessing : Implicationsfor cognitivedevelopment . Mclelland, J., forthcoming . Contextandstructurein conceptual combination . Cognitive Medin, D. L. and E.J. Shoben , 1988 20, 158- 190. Psychology

F. C. Keil / Word meaning

. The role of theoriesin conceptualcoherence . Psychological Murphy, G.L. and D. Medin, 1985 Review92, 289- 316. . The meaningof meaning . In: H. Putnam(ed.), Mind, languageand reality, Putnam, H., 1975 . 215- 271. London: CambridgeUniversityPress . Natural kinds. In: S.P. Schwartz(ed.), Naming, necessity , and natural Quine, W.V.O., 1977 . kinds, 155- 175. Ithaca, NY : CornellUniversityPress . Cognition15, 237- 262. . Conceptsand stereotypes Rey, G., 1983 . In: S. Vosnaidu . Similarity, typicality, and categorization , A. Ortony (eds.), Rips, L.J., 1989 . , 21- 59. New York: CambridgeUniversityPress Similarityand analogicalreasoning : Studiesin the internal structureof . Family resemblances Rosch, E. and C.B. Mervis, 1975 7, 573- 605. . CognitivePsychology categories Russell . Logicalatomism. In : J.H. Muirhead(ed.), ContemporaryBritish philosophy , , B., 1924 1stseries , 357- 383. London: Allen and Unwin. . Minneapolis . Four decadesof scientificexplanation Salmon , MN : Universityof , W.C., 1989 . MinnesotaPress at the 1993Meeting . Commentson symposiumon infant cognition. Presentation Slater, A., 1993 . in Child Development of the Society , New Orleans . for Research . Cambridge . Categorizationand concepts Smith, E.E. and Medin, D.L., 1981 , MA : Harvard . UniversityPress . Kind of objectsand causalrelationsin infancy , E., A. Woodwardand A. Phillips, 1993 Spelke acrosscultureand , A. Premack(eds.), Causalunderstandings (tentativetitle). In: D. Sperber . : CambridgeUniversityPress . Cambridge development . Early differentiationof causalmechanisms , K. and F.C. Keil, 1991 appropriateto Springer 62, 767- 781. biologicaland nonbiologicalkinds. Child Development . The development of inductionsabout biologicalkinds. Presentation Vera, A. and F.C. Keil, 1988 . at the Annual Meetingof the Psychonomic , Chicago , IL. Society . Thoughtandlanguage(E. Hartmann, G. Vakar, TransCambridge, MA : Vygotsky, L.S., 1962 . MIT press . Comparativepsychologyof mentaldevelopment Werner, H., 1948 (2nd edition). New York: . InternationalUniversitiesPress . : CambridgeUniversityPress . Teleology . Cambridge Woodfield, A., 1976 . The role of the whole objectassumptionin early word learning. PhiD. Woodward , A., 1992 dissertation , StanfordUniversity. : An etiologicalanalysisof goals and functions. . Teleologicalexplanations Wright, L., 1976 . , CA: Universityof CaliforniaPress Berkeley

Section 4

Categories , words , and language

) 199-227. North-Holland Lingua92 ( 1994

Constraints on word meaning in * early language acquisition Ellen M . Markman , USA , Stanford, CA 94305 , StanfordUniversity of Psychology Department fully acquire the vocabulary of their native languagedespite their Very young children success ' limited information processing abilities. One partial explanation for children s successat the inductive problem word learning presents is that children are constrained in the kinds of hypotheses they consider as potential meanings of novel words. Three such constraints are discussed: ( I ) the whole-object assumptionwhich leadschildren to infer that terms refer to objects as a whole rather than to their parts, substance, color , or other properties; (2) the taxonomic assumptionwhich leads children to extend words to objects or entities of like kind ; and (3) the mutual exclusivity assumptionwhich leadschildren to avoid two labels for the sameobject. Recent evidenceis reviewedsuggestingthat all three constraints are available to babiesby the time of the naming explosion. Given the importance of word learning, children might be expectedto recruit ' whatever sources of information they can to narrow down a word s meaning, including information provided by grammatical form class and the pragmatics of the situation. Word learning cOnstraintsinteract with theseother sourcesof information but are also argued to be an especiallyuseful source of infonnation for children who have not yet masteredgrammatical form class in that constraints should function as an entering wedge into languageacquisition.

1. Introduction

One impressive accomplishment of young children is the degree to which they acquire the vocabulary of their native language. Not only is the rate and ' extent of young children s word learning impressive, it is puzzling as well. ' Given the limitations on children s hypothesis testing, reasoning, memory, and other information processingabilities, their facility for building a lexicon is evenmore striking . This is especiallyso in light of the well-known inductive problem that word learning poses(Quine 1960) . Faced with an infinite set of

.

This work was supported in part by NSF Grant # BNS-9109236. I am very grateful to Andrea Backscheiderfor allowing me to publish the results of our study here. I thank Barbara Landau, Lila Gleitman, and two anonymous reviewersfor their helpful comments.

- 3841/94/507.00 (C) 1994- ElsevierScien~ : B.V. All rights reserved 0024 SSDI0024 - 3841( 93) EO054- B

200

E.M. Markman / Earlylanguage acquisition

possibilities about what a novel word might mean, the speed with which young children acquire word meanings requires explanation. Given the importance of word learning for language acquisition, children might be expected to recruit whatever sources of information they can to narrow down a word ' s meaning. One powerful one, for children old enough to benefit from it , is grammatical form class. As one example, Fisher et ale (this volume) provides dramatic demonstrations of the way in which syntactic information helps constrain the meaning of a novel verb (seealso Fisher et ale 1991, Naigles et ale 1992) . Being able to infer aspectsof the communicative intent of a speaker should provide another source of information about the referent of a novel term. A conscientiousparent or other tutor might further help by arranging the environment in ways to exaggeratethe salienceof the aspect of the situation being labeled in the hope that their child will find it salient too. ,Another potentially powerful source of information young children can use to figure out the meaning of a new word comes from wordlearning constraints. Constraints on word meaning may be particularly critical for babies who have not yet learned enough syntax to rely on ' grammatical form class to limit their hypothesesas to a word s meaning. I will argue that by the time they are ready to acquire vocabulary, children place constraints on possible word meanings, thereby greatly reducing the hypothesis space that needs to be considered. Children would not need to formulate a long list of potential meanings and painstakingly assessthe evidence in support of each. Rather, they could quickly zoom in on some hypothesesthat they are predisposedto prefer. In this paper, I will consider evidence in support of three word -learning constraints: the whole-object, taxonomic, and mutual exclusivity assumptions . I will argue that the most recent evidence supports the claim that children' s early word learning is guided by such constraints. For some different formulations of possible constraints seeBloom (this volume) ; Clark 1991; Golinkoff et al. 1992a). To claim that children' s early word learning is a constrained form of learning is nqt to claim that no other source of information matters. On the contrary, the recent evidence reveals some of the complex and subtle ways that word learning blasesinteract with other sour~ s of information . There is evidence, for example, that word -learning constraints can affect each other. When two or more constraints converge on the same hypothesis the learning will be efficient compared to cases where the constraints conflict and one must override another. Similarly word -learning constraints interact with grammatical form class. In some casesboth thesesourcesof information lead

E.M. MarkmanI Early language acquisition

to the sameconclusion about a novel word , but in other casesthey conflict. Analogous points can be made about communicative or pragmatic sourcesof information and the word learning assumptions. Finally , there is emerging evidence about the ways that the processing demands of a given word learning situation can affect the use of constraints. The evidenceI will review suggeststhat beginning word-learners rely heavily on word -learning assumptions but that these constraints are modulated by other constraints, by ' nonlinguistic context, by children s problem solving and other information processing abilities, and by the pragmatics and syntax of the language children hear.

2. Some preliminary considerado . .

In this paper, I will reviewstudiestestingthe hypothesisthat word-learning constraintsprovide young - children with an essentialsourceof infonnation about the meaning of a novel word. That is, that such constraints provide a partial explanation for the speedwith which new words are acquired. Before turning to the evidence, I will briefly consider a few theoretical issuesin order to sharpen the fonnulation of the hypothesis. 2.1. The developmental hypothesis

One way of fonnulating this hypothesis is to expect word -learning constraints to be available at the start of language acquisition. This would require documenting that at least some constraints are used by babies by the time they are capable even of understanding their first word . Babies acquire their first word often by age one year or less if measuredby production and probably younger if measuredby comprehension. There is reason to believe, however, that the nature of word learning undergoesa marked developmental shift from the time babies are capable of acquiring their first word at say a year and the rapid word learning they becomecapable of by 18 months or 2 years (Bloom et al. 1985, Corrigan 1983, Dromi 1987, Halliday 1975, McShane 1979, Nelson 1973) . The contrast betweenvery early word learning and this vocabulary spurt or naming explosion has been characterized in a number of different ways. The early phase of word learning has been tenDed ' ' prelexical (Nelson and Lucariello 1985) to emphasizethat there are ways in ' which the first ' words' may not be word like ; it has been called nonreferen' ' ' tial (Snyder et al. 1981) and associative (Lock 1980) to distinguish these

E.M . Markman / Early languageacquisition

' ' early words from genuine words that are referential; and it has been termed ' formative' per (Snyder et al. 1981) to suggest that the early words are used to perform some instrumental act rather than functioning as responses words that make reference. Another characteristic of this first phase of word learning is that words are added to the productive lexicon very slowly. A slow accumulation of new words might proceed by some relatively unconstrained associativemechanism. Dozens or even hundreds of trials may be neededfor a baby to learn, for example, to say ' bye-bye' on command. This is in marked contrast to the very fast acquisition seenaround 18 months. This developmental shift at the time of the naming explosion suggestsan alternative formulation of the hypothesis that word -learning constraints are necessaryfor word learning. The more preciseclaim is that some constrained form of learning is necessaryto account for the rapid acquisition of words seen at the time of the naming explosion. Babies could not be acquiring words at the rate of 30 or so a week if they were open-mindedly considering all possible hypotheseseach time they encountered a novel word. Thus the main hypothesis to be evaluated here is that word -learning constraints are available to babies by the time they enter the naming explosion at roughly 18 months of age. 2.2. Word-learning constraintsas default assumptions Before summarizing the evidencethat the word learning of 1Y2-year-olds is a constrained form of learning, I would like to clarify the claim. Some confusion has been generated by the terminology where ' constraints' is interpreted differently by different disciplines, especially linguistics versus ethology. I and other investigators have borrowed the terminology from the ethology of learning (seeWoodward and Markman 1991 for a discussion of this perspective). Within this discipline ' constraints' on learning are formulated as default assumptions - probabilistic blases that provide good first guessesas to a problem an organism must solve (Marler and Terrace 1984, Rozin and Schull 1988, Shettleworth 1984). To take one example from Gould and Marler ( 1984), foraging beesmust learn the color of the flowers that yield a given nectar. Purple is the default value beeshold. It is easier for them to learn about purple flowers than about flowers of any other color. Note that it is by no means impossible for them to learn sources that are other colors. Violations of the default assumption are common. Nevertheless, purple serves as a first guess, and it takes more trials or more evidenceto learn values that differ from the default assumption. Turning to word learning, the hypothesis

, acquisition E.M. MarkmanI Early language

203

is that children are able to make progress in word learning by use of such default assumptions (see Merriman and Bowman 1989 and Woodward and Markman 1991) . These constraints on learning provide good first guesses about the likely meanings of terms. Words that conform to the constraints should therefore be easier to learn than ones that violate them. Some of the controversy surrounding such claims stems from confusion over whether constraints must be absolute or whether they function as default assumptions (Gathercole 1989, Nelson 1988, Tomasello 1992) . 2.3. Are word - learning constraints domain -specific ?

To clarify another potential source of confusion, I ' d like to briefly consider the question of whether the word learning constraints I will consider here- the whole-object, taxonomic and mutual exclusivity assumptions- are specific to languageor available to other domains. Rather than discussthis here, let me refer you to Markman ( 1992), where this issue is considered in detail along with speculationsabout the origins of the constraints. My conclusion is that there is no reasonat all to believe that thesethree constraints are limited only to word learning and good reason to think they are available to some other domains. This is not to say that they are domain-general in that important domains are organized by very different principles. Given that analogous constraints are available to some fundamental domains, I suggestedthat word-learning constraints have been recruited from existing abilities that children possessrather than evolving as special purpose mechanismsonly for word learning (Markman 1992). Such speculations aside, these assumptions function as word -learning constraints, not in the sensethat they are domainspecific, but in the sensethat they help solve the inductive problem that word learning poses.

3. The taxonomic assumption

The first assumption I will consider is the taxonomic assumption which statesthat terms refer to entities of the samekind . An object term would refer to objects of the samekind , a color term to colors of the samekind , an action term to actions of the same kind , and so on. Although to adults this might seemself-evident, there was reasonto question whether young children would interpret terms in this way. First , taxonomic relations are not the only relations available to children. A label learned for a given object, for example,

204

E.M . Markman I Early languageacquisition

could refer to objects of the same kind , but it could in principle refer to the object and its spatial location , or the object and its owner, or the object and a salient part , etc. This is not , moreover, just a theoretical possibility . Many studies of classification have found that not only do children attend to such ' thematic' relations between objects, they often find them more salient or relations taxonomic than per se (seeMarkman 1989). When shown interesting to find another one, children might pick a man told and a car, for example, becausehe drives a car rather than a truck , another vehicle. Given such ' findings from children s classification, the question arises as to how children avoid such thematic interpretations of a novel word. Hutchinson and I proposed that children expect novel terms to refer to objects of like kind (Markman and Hutchinson 1984) . Hearing a novel label, then, should cause children to seek out taxonomic relations even in cases where thematic relations would otherwise be more salient. There is now a substantial body of evidencethat children from about 21J2or 3 expect labels to refer to things of like kind (Baldwin 1989, in press; Hutchinson 1984, Markman and Hutchin son 1984, Waxman and Gelman 1986, Waxman and Kosowski 1990) . I turn now to consider whether the taxonomic assumption is available to babies by the time of the naming explosion. 3.1. ~he taxonomic assumptionat the time of the naming explosion Bauer and Mandler ( 1989) first addressedthis issue in their study of the , they categorization abilities in very young children. As part of their study ' asked whether labeling would increase 16- 31-month -old children s tendency to sort taxonomically . Unexpectedly, however, even the youngest children in their study were sorting taxonomically from the start. That is, even with no labels children were sorting taxonomically about 75% of the time. Labeling did not increasethis already high level of performance. Bauer and Mandler ( 1989) have thus convincingly demonstrated that quite young children are capable of sorting taxonomically . They also argue that there may not be any general thematic preference. Because of the already high rate of sorting taxonomically, however, they were unable to test whether children of this age adhere to the taxonomic assumption. That is, it is still important to know whether there are situations in which very young children show thematic preferencesand, if so, whether hearing a label will cause them to shift to taxonomic sorting. Backscheiderand I addressedthis by changing aspectsof ' Bauer and Mandler s procedure that resulted in children responding taxonomically from the start.

E.M. Markman I Earlylanguage acquisition

205

One reason why Bauer and Mandler ( 1989) achieved such a high rate of taxonomic responding in their young children is that they used areinforcement procedure whereby they briefly pretrained children to selecttaxonomicand maintained this selectivereinforcementof taxonomic choicesthroughout ally the testing procedure. The selective reinforcement clearly mattered becausein a control study Bauer and Mandler achievedan equally high rate of thematic responding by selectively reinforcing thematic rather than taxonomic responses . Since they demonstrated that selective reinforcement is a to influence children' s responses powerful way , Backscheiderand I avoided selectivereinforcement (as did all earlier studies with older children). Other differencesbetween our procedure and Bauer and Mandler ' s ( 1989) are that we used pictures instead of objects, we used items whose thematic relations we thought would be better known to 18-month -olds, and we used an up/ down placemen,t of thematic and taxonomic choices instead of left/right . Finally, Bauer and Mandler counterbalanced the position of thematic and taxonomic choices by alternating the left /right placement from trial to trial . This alternation is unfortunate especially when coupled with a reinforcement procedure. Children may have simply learned the correct answer changes sidesevery trial . We counterbalancedside but not by alternating from trial to trial . The major difference between the studies, however, is that we did not differentially reinforce taxonomic responding. Thirty -three children participated in our study. They ranged in age from 18 to 25 months with a mean age of 21.5 months. The experimental materials consisted of ten triads, each containing one target picture, one picture thematically related to the target, and one picture taxonomically related to the target picture. We selectedthematic relations that we thought would be highly familiar to even very young children. The taxonomic match belonged to the same basic level category as the target but , where possible, came from distinctive subordinate categories. The ten triads are listed in table 1. The experimental questions were preceded by a set of four warm-up questions designed to clarify the instructions and procedure but not to differentially reinforce either taxonomic or thematic responding. The warmup questions and the experimental questions were oddity tasks, where children viewed a target picture and then selectedone of two choice pictures. To begin, children were introduced to a frog hand-puppet who manipulated the pictures. They were asked to help the frog find the pictures it wanted. The children were told that the frog had just gotten a new house, and wanted things to put in its house. In order to ensure that children gave unambiguous , they were shown how to place the pictures they selectedin the frog responses

206

E.M . Markman I Early languageacquisition

Table I 's study Experimentalitemsfrom Markmanand Backscheider Choices

Target

Taxonomic

Thematic

Sitting baby Bottle Chair Rabbit cup Foot Glasses Blue scoop shovel

Lying baby Bottle Chair

Stroller

Spoon Toilet paper Mitten

Spoon Toilet paper Mitten

Baby Sitter Pitcher Shoe

Cup Foot Glasses Red shovel

' puppet s mouth . There

were

two

Eyes Pail Cereal in bowl Toilet Hand

experimental

conditions

, the No

Label

Conditionand the Novel Labelcondition. 3.1.1. Warm-up questions For the wanD-up questions, children were shown a target picture and then had to select one of two pictures, one of which was an unrelated distractor. The other picture, the correct choice, was thematically related to the target for half of the trials and taxonomically related for the other half. The items used for the wanD-up questions are listed in table 2. Table2 ' Warm-up itemsfrom Markmanand Backscheiders study Target

Choices . Taxonomic

match

Thematic match

Distractor

Hat

Hat

Head

Car interior

Watch

Watch

Wrist

Shopping cart Hair

Shopping cart Hair

Bag of groceries Brush

Road Pillow Pot

.

In a single trial the child saw either the taxonomic match or the thematic match , not both . The distractor was always present .

The experimenterplacedthe targetpicture(for examplea hat) on the table ' ' and askedthe child to ' look at this picture' or seethis , etc. Then the two

E.M . Markman /

~. acquisition Ear'v jranguag

207

choice pictures were placed on a magnetic board, one above the other. (Pilot work suggestedthat a vertical alignment of the pictures produced fewer responseblasesbased on position than did a horizontal alignment.) The top/ bottom position of the random distractor and correct picture was counterbalanced . The target picture was held next to the top picture while the ' ' experimenter asked Is this another one? and then held next to the bottom ' picture while the experimenter asked or is this another one?' When children selectedthe wrong picture they were told that that was not the one that the frog wanted and were encouragedto make another selection. For the warmup questions, the frog would not accept the wrong picture - it would keep its mouth closed and shake its head, etc. When children selected the correct choice the frog opened its mouth to take the picture and enthusiastically thanked them, saying ' Yeah! Yeah! That is just the one I wanted. Thank ' you , kissed the child , etc. The experimenter then explained why that was the picture the frog wanted, describing either the taxonomic relation betweenthe target and the picture or the thematic relation as appropriate. For example for the thematic pair hat/ head, the frog said ' That ' s great! A hat goes on a head. Do you ever put a hat on your head? A hat goes on a head, they go ' ' ' together . For the taxonomic pair hat/ hat, the frog said That s great! Now I ' have two hats. Look , these are the same, two hats . Thus, the warm-up questions encouraged, explained, and reinforced taxonomic and thematic responsesequally. The procedure for the two experimental conditions was similar to that of the warm-up questions with two exceptions. First , in the procedure proper there was no unrelated distractor. One of the choice pictures was thematically related to the target and the other was taxonomically related to it , as shown in table I . Second, the frog acceptedany choice the child made, that is, there was no selectivereinforcement during the experiment proper. 3.1.2. No label condition To begin each trial , the puppet placed the target picture on the table and said ' Now let' s look at this picture' or ' Here' s a new picture' . Then the two choice pictures were placed on the magnetic board, one above the other. Whether the thematic choice was placed on the top or on the bottom was counterbalancedsuch that for each child half of the time it appeared on the top and half the time on the bottom . The order of presentation of the items was randomly determined for each child. Once the child saw the target ' ' picture, the puppet asked the child to find another one . It held the target ' picture next to each of the choice pictures, asking Is this another one, or is

E.M . Markman I Early languageDCQuisiliOIl

this another one?' The frog then held the target between the two pictures while the child selectedone of them and placed it in the frog puppet' s mouth. The experimenter took the target out of the frog ' s mouth as the child moved his or her choice over to the frog . Children were enthusiastically thanked on every trial , regardlessof which picture they chose. 3.1.3. Novel label condition The materials and procedure for the Novel Label Condition were identical to those of the No Label Condition except that children in this condition were told that the puppet would sometimes speak in puppet talk and the puppet gave each target picture a novel label. For example, the puppet would show the child the target picture saying ' Now I am going to show you a sud. ' Look at this, it is a sud. Can you find another sudl When children were ' making their choices the puppet said, e.g., Is this another sud or is this another sud1' Ten nonsensesyllables were used as the novel labels and were randomly assignedto the targets for each child. The first question this study addressedwas whether children as young as 18 to 24 months will show thematic preferencesat all. In marked contrast to Bauer and Mandler ' s ( 1989) findings, these young children did show a thematic bias. When children in the No Label Condition were asked to choose between an object that was from the samecategory as the target and one that was thematically related to it , they chose the taxonomic match only 32% of the time, which was significantly less than chance, t ( 15) = 4.78, p < O.OOI. That is, they chose thematically 68% of the time. Thus, when children are not selectivelyreinforced for choosing taxonomic items and when the thematic relations are geared towards what would be well-known to even 18-month -olds, quite young children will reveal the thematic blasesoften seen ' in older children. This does not contradict Bauer and Mandler s claim that these young children are capable of grouping the objects on the basis of common categories, but it does establish that when the thematic relations are well-known and when children are not reinforced for one type of responseor another they prefer to organize objects thematically. Given that for theseitems, theseyoung children do show thematic blases, it then makes senseto ask whether hearing an object labeled will help them override their thematic preferencein favor of taxonomic relations. As predicted , there was a highly significant effect of condition , with children in the Novel Label condition picking taxonomically fully 77% of the time compared to 32% of the time for children in the No Label Condition , F( I , 31) = 66.58, P < 0.000I . Children hearing an object labeled selected a taxonomic match

E.M. Markman/ Early language acquisition

209

well over what would be expected by chance, 1( 16) = 6.75, p < O.OOOI . Thus, these young children do adhere to the taxonomic assumption. They extend object labels to other objects of like kind rather than to objects that are thematically related. Wasow and I have conducted subsequent studies to detennine if these results would replicate with babies 16 to 18 months old. Several changesin procedure were made to accommodate to the younger children. We used objects instead of pictures and reduced the number of questions that children were asked. The experimental triads that were used are presentedin table 3. The procedure was otherwise quite similar to that of Backscheider and Markman , including use of wanD-up questions. Thirty -two 16-month -olds and 32 18-month-olds participated in this study, half of each age group in eachcondition . As before 18-month -olds honored the taxonomic assumption. First , when in the No Label condition , 18-month -olds showed a clear thematic bias selecting the taxonomic choice only 28% of the time, which is below chance, 1( 15) = 10.97, p < O.OOOI . When the target object was given a novel label, the 18-month -olds significantly increased their taxonomic selections , now picking a member of the same category 54% of the time, 1(30) = 3.34, p < 0.005. Table3 Itemsfrom Markmanand Wasow's studywith familiar objects

Target

Choices Taxonomic

Thematic

Felt doll hat Plasticdoll hat Doll head Whitedoll-sizesock Yellow toddler-sizesock Doll-sizefoot " Smallerdoll (approx. 3" ) Babydoll (approx. 5 ) Doll-sizebottle Red& whiteadult-sizetoothbrushPink child-sizetoothbrush Setof plasticteeth Pink plasticdollhousetable White woodendollhousetable Pink plasticdollhousechair Yellowplasticpail Redplasticpail Yellow& white~plasticshovel

We did not find this labeling effect with 16-month -olds, however. The 16month-olds did find the thematic relations salient, picking taxonomically only 34% of the time in the No Label condition , which was less than chance, 1( 15) = 3.76, p < O.OO2 . But labeling had no effect on their selections, with babies selecting taxonomically only 39% of the time in the Novel Label condition . One possible reason for our failure to replicate the labeling effect with 16month-olds is that in this procedure we provide a novel label for a familiar

210

E.M. MarkmanI Early language acquisition

object and then look to seehow the child interpreted the label. As I will argue later, children find it more difficult to learn second labels for objects because this violates mutual exclusivity, another word -learning constraint. Other studies mimimized the conflict betweenmutual exclusivity and the taxonomic assumptions by telling children that the novel word was a word in puppet language or a foreign language (Markman and Hutchinson 1984, Waxman and Gelman 1986) . Although older children have been shown to treat mutual exclusivity as applying within a single language, not acrosslanguages(Au and Glusman 1990), 16-month -olds would not have understood a discussion of puppet language. To test the possibility that the conflict between mutual exclusivity and the taxonomic assumption was confusing the 16-month -olds, Wasow and I ran another version of the study, this time using novel objects. Thirty -two 16-month-old babies were introduced to novel triads of toys where two of the toys were similar in appearanceand two were related by some thematic relation we demonstrated. For example, two honey dippers of different colors were novel taxonomic pairs and one honey dipper was scraped against a plastic grid to demonstrate the thematic relation. Babies were then shown a target, e.g., a honey dipper, heard it labeled or not , and were then depending on the condition , asked to find another one or another X (where X was the novel label) . The full set of novel items used is described in table 4. Table 4 ' Tern~ frnm Markmal1 and Wasow' s study with novel objects

Target

Choices Taxonomic

Thematic

Smallgreenspongecircle

Large yellow spongecircle

Smallorangeblock

Largeblueblock

Red ' person' peg Largered propellor

Yellow ' person' peg Smallyellowpropellor

Wooden car with round hole Red helicopter (with post to hold propellor)

Yellowhoneydipper DecoratedPVC pipe

Greenhoneydipper Plain PVCpipe

Yellow plasticgrid Decoratedfunnel

Plastic' tub' containerwith squarecut out in lid

E.M . Markman / Early languageacquisition

Once again we failed to find an effect of labeling with 16-month -olds. Babies selectedthe taxonomic choice 40% of the time when the object was unlabeled and 44% of the time when it was labeled. Our avoiding second labels for objects did not improve babies' performance in the labeled condition . One possibility, of course, is that 16-month -olds lack the taxonomic assumption. Another is that the oddity task we have been using is insensitive to the knowledge babies have. To reveal use of the taxonomic assumption in this task, babies must inhibit a dominant response. We deliberately selected thematic choices that would be preferred by babies this age. Being asked for "a dax' for example may not be compelling enough to prevent babies from ' ' what doing they most prefer, even if they do interpret dax as referring to objects of like kind . Thus, our failure to find the labeling effect with the 16-month -olds might be partially acCounted for by the requirements of the oddity task. The taxonomic assumption may not be strong enough to allow young children to inhibit reaching for a preferred object. Support for this interpretation comes from work from Waxman and her colleagues, who found that labels highlight categorical structure for even 12and 13-month-olds (Waxman, this volume; Waxman and Helm 1991, Mar kow and Waxman 1992). Instead of an oddity task where children would need to .inhibit a dominant thematic response to reveal knowledge of the taxonomic assumption, Waxman and Helm ( 1991) and Markow and Waxman ( 1992) used a manual habituation procedure which assessedwhether babies could notice and distinguish two different categories of objects. The ' hypothesis being tested is that words should enhance babies tendency to notice object categories. The prediction was that labeling objects should ( I ) increase the rate of habituation to category exemplars and (2) increase the degree of dishabituation to an object from a different category. In this procedure, babies are first familiarized with four toys from a given category (e.g., four cars or four animals) . Children were presentedwith each toy, one at a time for 30 secondseach. On analogy with visual habituation tasks, a decrementin the time spent exploring the last toy compared with the first was used as an index of habituation . After the familiarization trials , babies were presentedwith a new exemplar from the old category and an object from the novel category. Here, again on analogy with standard habituation tasks, babies who had best formed the category would be expected to show a greater interest in the object from the novel category than in the object from the original category. For some of the babies the experimenter labeled the

E.M . Markman / Early languageacquisition

' ' object during the familiarization phrase saying, e.g., Look at the car . For some babies the experimenterdrew attention to the object without labeling it ' ' saying, e.g., Look at this . As predicted, labeling the objects improved babies' ability to categorizeas measuredboth by rate of habituation to within category exemplarsand by dishabituation to a novel category (Waxman and Helm 1991, Markow and Waxman 1992). Moreover, Markow and Waxman ' discoveredthat at 12 months labeling objects improved babies categorization even when adjectivesrather than nouns were used to refer to the objects. This finding is of interest becauseit supports the idea that babies first treat words as referring to objects of like kind rather than treating only nouns that way. One-year-olds, presumably with minimal knowledge of grammatical form class, were better able to categorize objects if they were labeled regardlessof whether a noun or adjective was used as the label. ' At this point it is of interest to consider briefly Bloom s (this volume) alternative framework for constraints on word meaning. Bloom views the postulation of particular constraints such as the taxonomic and whole-object assumptions as unmotivated and corrects this in his model with an elegant systemthat generatesconstraints to conform to universal grammar-cognition ' mappings. Bloom s systematicanalysis provides a clear statement of what the endpoint of development should be. Where we disagreeis in what to attribute to the very early phasesof language learning - the time before children have mastered grammatical form class. On the view I have proposed very young children will break into the systemby assumingthat a word refers to a whole object (the whole object assumption) and is extended to things of like kind ' (the taxonomic assumption). Bloom s account, which is bolstered by counterexample to my formulation , is that children map count nouns on to kinds of individuals rather than kinds of objects. Yet , this may describe adevelop mental achievement rather than the initial state. Babies may treat terms as referring to objects but with time note the regularity with which count nouns refer to objects. Once this correlation is establishedchildren may then be able ' ' to metaphorically extend object-like or individual status to non-objects that ' ' ' ' are referred to with count nouns such as a forest or a lecture . Second, on ' my account as well as Waxman s, young children rely on word -learning constraints before they have mastered grammatical form class. Very young children should err by treating adjectives, for example, as referring to kinds of objects. The work just described (Markow and Waxman 1992) is one source of evidencethat before children have masteredthe noun/adjective contrast in ' English, adjectivesas well as nouns enhanceone-year-old s attention to kinds of objects.

E.M. Markman I Earlylanguage acquisition

213

' Labeling has been found to enhance even younger babies attention to objects. Baldwin and I found that babies from 10- 14 months old attended more to toys that had been labeled compared to ones that had not ( Baldwin and Markman 1989) . In a second study, we compared labeling to another ' powerful means of directing babies attention , namely pointing . Although labeling did not increasebabies attention over pointing at the time pointing occurred, during a subsequentplay period babies attended more to toys that had been labeled than to those that had not . Thus, labeling may sustain babies' interest in objects beyond the time the labeling occurs. By helping babies sustain their attention to relevant objects, this labeling effect may help babies notice and remember the word -object correspondences. It is not yet known whether this labeling effect is caused by labels per se or whether other nonlinguistic sounds would have the same effect (see Baldwin and Markman 1989 for other possible explanations as well) . So far , the existing data are inconsistent. Waxman and Balaban ( 1992) have found that words but not tones increased attention to object categories in 9-month -olds. On the other hand, Roberts and Jacob ( 1991) found that music as well as words facilitated object categorization in infants . Whether the effect is specific to words or not , words do heighten babies' attention to objects and object categories. Taken together, then, the experimental findings document that babies from 12- 18 m"Onths(and maybe even younger) use the taxonomic assumption to help them determine the appropriate referents of a word. This conclusion from the experimental data is supported by the results of Huttenlocher and ' Smiley s ( 1987) study of naturalistic data. They generateda set of criteria to distinguish thematic (complexive) extensionsof words from taxonomic extensions , arguing that not every use of a word by a young child should be taken as a simple label of an object. They followed severalchildren from the time of their first word (around 13 months for most of the children) and periodically recorded both the words children produced and details of the situation and context in which utterances occurred. For example, a child who reaches towards a cookie jar saying ' cookie, cookie' with an insistent request intonation , should not be interpreted as labeling the cookie jar as a cookie. Instead, the child is most likely requesting a cookie. Using coding categories that ' capture reasonable interpretations of children s utterances, they found that from the start children use words to refer to objects of like kind . Thus, evidence from experimental and naturalistic studies alike indicate that the taxonomic assumption is available to babies who have not yet undergone the naming explosion.

214

E.M. Markman/ Early language acquisition

Although the taxonornic assurnption goesa long way in reducing the kinds of hypothesesbabies need to consider in figuring out the rneaning of a new word , it by no rneans solves the problern. Babies rnay be led to treat novel words (or nouns) as referring to kinds of objects but that still leavesopen the question of which kind . This problern has been extensively addressed by Waxrnan and her colleaguesand is reviewed in Waxrnan (this volurne) . To ' briefly surnrnarize, labels facilitate young children s attention to objects at sorne hierarchical levels but rnight actually interfere with their ability to categorizeat other levels. Young children hearing a noun expect it to refer to categories either at a basic or superordinate level of categorization, but do not expect it to refer to the distinctions rnade at subordinate levels of categorization. Rather, they expect this level of detail to be indicated by ' adjectives. Waxrnan s work reveals the interaction between use of wordlearning constraints, grarnrnatical form class, and conceptual structure. Not only does the taxonornic assurnption leave open the question of which kind of object is being referred to , it even leavesopen which object is being referred to. As Baldwin ( 1991) pointed out , in normal environrnents that babies find thernselvesin , there are often rnany candidate objects around that could serveas potential referents for a new word. Even if babies are prepared to treat a novel word as referring to , say, a basic level category, they rnust figure out which one. One possibility is that babies rnay treat a novel word as refelTing to whatever novel object they are attending to. Whenever babiesand adults are focussing on the sarne object, this would lead babies to correctly identify the referent of a novel term. On those occasions, however, when the adult was in fact labeling an object other than the one the baby was focused on, the baby would wrongly interpret the new word and would rnake a rnapping error. Such rnapping errors appear to be very rare, although there is evidence that babies learn words rnore readily when parents tend to label what their baby is attending to (Tornasello and Farrar 1986) . Baldwin ( 1991) suggested that babies could avoid rnapping errors if they rnonitored the ' speakers focus of attention. If babies recognize that the object they are interested in is not the one the speakeris attending to , then they would know not to treat the word as a label for their object. Baldwin tested whether ' babies use information about a speakers focus of attention in inferring the referent of a novel word or whether they rnapped a novel word to whatever novel object they were interested in. To test this, she provided novel labels to ' ' 16- 19-rnonth-old babies in one of two conditions. In the discrepant label condition , babies were given a toy to play with . Once the experirnenter was assured that the baby was exarnining the toy the experirnenter provided a

E.M . Markman I Early languageacquisition

215

novel label saying, e.g., ' It ' s a toma' . Instead of looking at the baby or the ' ' toy , however, the experimentergazedinto an opaque bucket as she said It s a ' toma . Thus, the baby heard the novel label while looking at a novel toy but while the speaker looked into a bucket that contained a second toy . In the ' follow -in ' labeling condition , the experimenter looked at the visible toy while she provided the novel label. In the follow -in labeling condition , babies of both ages treated the novel word as referring to the visible toy . Of more interest is what happenedin the discrepant conditions. The results were quite clear: at neither age did babies treat the novel word as referring to the visible toy even though they were looking at the toy at the time they heard the label. By monitoring the eye-gaze, posture, direction of voice, or some other cues to the speaker's focus of attention babies avoided making a mapping error. The 16- 17-month -olds avoided errors by simply failing to learn the new word. The 18- 19-month -olds not only avoided errors, but were able to infer that the new word referred to the object that was hidden in the bucket at the time of labeling. Baldwin' s results reveal how babies' use of word-learning constraints is coordinated with their construal of the communicative intent of the speaker.

4. The wbole- objeet assumption

The whole-object assumption states that babies should treat a novel label as referring to an object rather than a part , substance, color , movement, etc. There is somecontroversy as to whether young children treat a novel label as referring to the shapeof an object rather than the object per se. Some studies have found that young children will extend an object label to things of the sameshaperather than those of the samecolor , or texture (Au 1989, Au and Markman 1987, Baldwin 1989, Landau et al. 1988, Ward et al. 1989) . One interpretation of these findings, however, is that children are treating the terms as referring to whole objects of like kind , and that shape is a ' particularly reliable clue to an object s taxonomic category. Sola et al. ( 1991) reinforce the conclusion that it is the object, not the shape, that children are responding to in that they do not treat words as referring to the shape of non-solid substances even when different substances were arranged into common shapes. Baldwin (in press) has found that , on the one hand, children will extend a label on the basis of shapewhen there is no obvious taxonomic category, but , on the other hand, children will extend a label on the basis of

E.M. MarkmanI Early language acquisition

taxonomic category even when the objects in question do not have the same shape. Moreover, Landau (this volume) along with Becker and Ward ( 1991) have found that preschoolers treat novel terms as labels for objects rather than shape in that they treat a term for a worm -like animal as referring to another worm twisted into a different configuration . (Landau, however, argues for a different interpretation of these results. In particular she argues that children are still responding on the basis of shapeibut shape now is defined as the possible shape transformations an object of a given shape can undergo.) In addition , there are developmental differences in the way this whole-object bias interacts with grammatical form class (Landau et al., in press; Landau, this volume) . Older children and adults will treat a novel noun as referring to the kind of object but treat an adjective as referring to a property such as texture. But the youngest children treated even novel adjectives as referring to kinds of objects. Here again we see that when knowledge of grammatical form classis weak, word -learning constraints such ' as the whole-object assumption dominate the child s interpretation of a novel term. Taken together, these findings suggestthat children attempt to honor the whole object and taxonomic assumptions per se, rather than treating words as shapeterms. Further support for the whole-object assumption comes from studies that have documented that children interpret a novel term as a label for an object and not its part (Markman and Wachtel 1988, Mervis and Long 1987) and for an object over its substance(Markman and Wachtel 1988, Sola et al. 1991) . 4.1. The whole-object assumptionat the time of the naming explosion Although the whole-object assumption has received experimental support, only Mervis and Long ( 1987) examined babies around the age of the naming ' explosion. Moreover, none of the studies (except perhaps Landau s, this volume) provided a strong test of the whole-object assumption. Babies might map words onto objects not becauseof a whole object assumption per se, but simply becauseobjects are salient and babies might map novel words to whatever is most salient at the time of labeling. Thus a stringent test of the whole-object assumption requires examining whether babies treat novel words as referring to objects even when an object is not the most salient aspectof the environment. Woodward ( 1992) has provided such a test. Woodward ( 1992) had babies view two video monitors. On one screen babies viewed a dynamic substancein motion , such as flowing lava. On the

E.M. Markman/ Early /anxuaR .e acquisition

217

other screenthe babies viewed a static novel object. When the screenswere turned on and babies allowed to watch freely, they clearly preferred the swirling substancesto the static objects. Thus, Woodward created a situation where the object was lesssalient than the substancein motion . On some trials babies heard a label that could be interpreted as referring either to an object or substance. The prediction from the whole-object assumption is that hearing a label should cause babies to shift attention more to the whole object. This hypothesis was confirmed for 18-month-olds (though lessclearly for the 24-month -olds) . Around the time of the naming explosion, then, babies treat novel words as referring to objects per se, rather than to . whatever is most salient. Testing a somewhat different hypothesis, Echois also has evidence suggesting that well before the time of the naming explosion babies honor the whole-object assumption, but that younger babies (8- to 10-month -olds) may not. Echois ( 1990, 1991) asked whether young babies might map labels to whatever in the environment is consistent rather than to objects per se. She used an habituationjdishabituation procedure where babies heard labels either in the presenceof consistent objects with varying motions or consistent motions with varying objects. Trends in the pattern of habituation and dishabituation led Echois to speculatethat there is a developmentalshift from 8- 10 months to 13- 15 months in how babies are affected by labeling. The younger babies appeared to focus on what was consistent when they heard a label while the older ones focused on objects per se. The older babies were only 13 to 15 months old , however. Woodward ' s visual preferencestudy and Echols' habituation studies both suggest that the whole-object assumption is in place by the time of the ' naming explosion. By focussing children s attention on objects as the most likely referent of a novel word , the whole-object assumption greatly reduces the number of hypotheseschildren need to consider for the meaning of a novel term. The whole-object assumption thus promotes the rapid learning of object labels. While the whole-object assumption can account for the speed with which children acquire names for objects, it poses an obstacle for , colors, texture, and other properties of learning terms for parts, substances objects. With only the whole-object assumption to guide their interpretation of novel words, babies would be limited to learning only object labels. One of the functions of the mutual exclusivity assumption discussednext is that it can override the whole-object assumption, freeing children to learn a greater variety of terms.

E.M. MarkmanI Early language acquisition 5. The mutual exclaivity

assumption

The mutual exclusivity assumption leads children to prefer only one label for an object. This assumption is stronger than a related one of contrast (Clark 1987, 1990, 1991), which is that any two words contrast in meaning. ' ' ' ' Although words such as car and vehicle contrast in meaning with one being a superordinate of the other, they violate mutual exclusivity in that one ' ' ' ' object can be called both car and vehicle . In some cases, mutual exclusivity comesinto conflict with the whole-object assumption. Supposea child hears a ' ' novel word , say finial in the presenceof a novel object, a pagoda. On the ' ' whole-object assumption, the child should (mistakenly) treat finial as referring to the pagoda. Supposeinstead that another child who has learned that a ' ' ' ' pagoda is called pagoda also hears the novel word finial in the presenceof the pagoda. Here the whole-object and mutual exclusivity assumptions are brought into conflict. As before, the whole-object assumption would lead ' ' children to interpret finial as referring to the pagoda. In contrast, children ' ' ' ' who know a pagoda is called a pagoda are led to avoid treating finial as ' ' yet another word for the same object. By rejecting finial as a second label children would then be motivated to find some other referent for the term. , Thus, mutual exclusivity can motivate children to find salient parts, substances and properties of objects as referents for novel terms. As predicted by mutual exclusivity, preschool children are in fact better able to .learn novel terms for parts and substancesif taught on objects with known labels (Markman and Wachtel 1988) . More generally children should be better able to learn a variety of property terms when taught on objects with known labels. Mutual exclusivity can then help override the whole object assumption allowing as word the novel to children something other than an object label. interpret Mutual exclusivity may also enable children to override the taxonomic assumption to enable them to better learn proper names (Hall 1991) . While the taxonomic assumption leads children to interpret new terms as referring to objects of like kind , proper names refer to particular individuals , not kinds. Hall ( 1991) predicted that children should be better able to learn proper namesfor familiar objects with known labels than for objects without known labels. To test this Hall ( 1991) relied on the logic first put forward by Katz et ale( 1974) . Katz et ale taught children a new label for an object. They then looked to see whether children restricted the new term to the labeled object or whether they extendedit to a highly similar secondobject. Choosing the original object above chance was taken as evidencechildren treated the term as a proper name; indifference between the two objects was taken as

E.M . Markman / Early languageacquisition

219

evidence they treated the term as a common name. Katz et al. ( 1974) manipulated two sourcesof information about whether something might be a proper name or not. The first was grammatical form class, in particular , ' whether the novel term took an article or not , e.g., ' a dax' vs. ' Dax . The second was the type of object that was labeled, in particular , animate-like things which are appropriate referents of proper names, in this case dolls, versus inanimate objects unlikely to be treated as unique individuals , in this caseblocks. Their results for girls suggestedthat when both the grammatical form class and the conceptual domain were appropriate, girls treated the novel term as a proper name. That is, when a doll was called ' Dax ' , girls interpreted Dax as a name for that doll and not the other one. When the doll was called ' a dax' , girls treated the term as applying to both dolls. No matter whether the block was called by common or proper name, girls treated the new word as referring to both blocks. A partial replication of this stud)' suggestedthat 18-month -old girls would also distinguish betweenproper and common names for dolls. These results did not hold up for the boys, however. Two -year-old boys were at chance for all object selections, even when hearing a proper name for a doll . Gelman and Taylor ( 1984) replicated the Katz et al. ( 1974) study with some improvements and modifications. They addressedthe concern that evidence for treating a term as a common noun was undifferentiated from chance performance. By adding other distractors to the responseset, they avoided this problem. Gelman and Taylor ( 1984) also used unfamiliar rather than the familiar objects that Katz et al. had used. Testing somewhat older children (2- 3-year-olds), Gelman and Taylor found that boys as well as girls treated terms as common nouns except in the casewhere proper nameswere used to refer to animate-like objects (stuffed animals) . To return to Hall ( 1991), then, the question is whether, along with grammatical form class and conceptual domain, mutual exclusivity is used to help children learn proper names. By overcoming the tendency to treat a novel label as a second term referring to things of like kind , mutual exclusivity could promote the interpretation of a new term as a proper name. Second labels for objects should be more likely than first labels to be construed as proper names. To test this Hall ran a partial replication of Gelman and Taylor ( 1984), running only the condition that resulted in proper name interpretations in the prior work . That is, two - and three-year olds were taught proper names for animate objects. They were taught in one of two conditions: either the animate objects had known names or they did not. Overall, Hall ( 1991) again replicated Gelman and Taylor ( 1984) and Katz et

E.M. MarkmanI Early language acquisition

al. ( 1974) with children treating the novel term as referring to the individual object labeled at above chance levels. Moreover, as predicted, children were more likely to treat the novel term as a proper name when they already knew another common name for the object. Combining the results of Hall ( 1991) with those of Gelman and Taylor ( 1991), we can conclude that young children recruit information from three sourcesto determine how to interpret a novel term: When the conceptual domain is appropriate, when the syntactic information is consistent with a proper name interpretation , and when mutual exclusivity can help override the taxonomic assumption, young children are most likely to treat the term as a proper name. These studies with parts and substancesand proper and common names lead to the general conclusion that some word -learning constraints can be used to moderate or override others. Mutual exclusivity can override the whole-object assumption leading children to learn terms for parts, substances, and other attributes of objects. It can also override the taxonomic assumption , thereby helping children interpret a proper name as a term for a unique individual . ' 5.1. Mutual exclusivity as a guide to a word s referent Another advantage of mutual exclusivity is that it can provide an indirect means of inferring the meaning of a novel word. Suppose a child hears ' ' ' someoneuse a novel object label, for example Look at the gadget or Please ' bring me the gadget . The child looks around and seesone or more objects with known names - say, a ball , and a novel object, say, a garlic press. By ' ' ' mutual exclusivity the child should reason that gadget can t refer to the ball ' ' becauseit is a ball , so it must refer to the garlic press, the only novel object around. In this case the child can fulfill both the whole-object and mutual exclusivity assumptions and use them to infer the meaning of a novel word . Adults and children alike use mutual exclusivity to figure out which object a novel label refers to without anyone explicitly pointing to or otherwise indicating the object (Au and Glusman 1990, Dockrell and Campbell 1986, Golinkotr et al. 1992b, Hutchinson 1986, Markman and Wachtel 1988, Merriman and Bowman 1989). Golinkotr et al. ( 1992b) went a step further and documented not only that mutual exclusivity provides children with an indirect means of inferring the meaning of a novel term but that that term then functions as a familiar word -year-olds used mutual in the child ' s lexicon. They showed that after 21J2 exclusivity to infer the referent of a novel term, they treated the referent as an

E.M . Markman I Early languageacquisition

object with a known label which in turn prevented the children from accepting another novel label for that object. Two -year-olds' successat using mutual exclusivity to infer the referent of a novel term belies the inferential requirements of this task. The logic of this ' problem is of the form of a disjunctive syllogism: The novel word must refer to either object A or to object B. It is not A (by mutual exclusivity), therefore, it must be B' . Hutchinson ( 1986) has found , in fact, that this logic of the problem may pose a problem for developmentally delayed children even when they are matched in mental age to normal children. In her task, normal and developmentally delayed children were given pairs of objects where one object had a known name and the other did not. The normal children were 2, 2Y2 and 3 years of age and, as expected, at all age groups they chose the unfamiliar object at above chance levels as the referent of a novel label. The performance for the developmentally delayed children showed a marked developmental difference with children matched in mental age to the 2Y2- and 3-year-olds selecting the novel object at above change levels, but with children matched in mental age to the 20- 24- month-olds failing to do so. Thesechildren did not select the novel object as the referent of the novel word even though normally developing children of 20- 24 months did. It was not until retarded children reacheda mental age of 28 months that they seemed capable of using mutual exclusivity to infer the appropriate referent of a novel term. One possible reason for this delay is that the logic of the problem is particularly slow to develop in the delayed children. 5.2. Mutual exclusivity at the time of the naming explosion There are two concerns that have been raised about whether children around the age of the naming explosion can use mutual exclusivity to infer the meaning of a novel term. The first is that this use of mutual exclusivity has been documented only in children from two years of age and up . Studies with younger children are needed. Second, an alternative explanation for these results has been proposed by Merriman and Bowman 1989 . ( ) They argued that children might map a novel word to a novel object in these studies becausethey are predisposed to fill lexical gaps - to find first labels for objects. Thus , when children encounter an object for which they do not know a label , they should seek out its name. Children might be mapping the novel label to the novel object becausethey desire a name for the novel object, not because they reject a second label for the familiar object .

E.M. Markman/ Early language acquisition

Wasow and I have addressed both of these concerns in a recent set of studies(Markman and Wasow, in preparation) . We asked whether children at around the age of the naming explosion can use mutual exclusivity unaided by a bias to fill lexical gaps to infer an appropriate referent of a term. The lexical gap hypothesis states that in the presenceof a novel object, children will seekits name. With no novel object visible, no lexical gap can be created. We ruled out the possibility of children using a lexical gap strategy by not having a novel object visible at the time of labeling. At the time they heard the novel label children saw only a familiar object and a bucket which served as a possible location for objects. The prediction from mutual exclusivity is that upon hearing a novel label children should reject it as a secondlabel for the visible familiar object and search for an appropriate referent. This search could consist of looking around the room , or on the floor , etc. or by wanting to seewhat is in the bucket. We found that babies as young as 15 months of age can use mutual exclusivity to reject a second label for an object and to motivate them to search for another potential referent. In this case, not only do the babies need to use the disjunctive logic to reject the label as a label for object A , they infer it applies to some unknown object B that is not visible at the time. It is remarkable that this very demanding test of the use of mutual exclusivity is passedby such young babies. Another less demanding test of the use of mutual exclusivity in young children is to seewhether mutual exclusivity simply causeschildren to reject second labels for objects, without requiring the children to make any inferences beyond that. The prediction is simply that second labels for objects should be more difficult to learn than first labels. Although there are a number of studies that document young children are capable of learning second labels for objects (Banigan and Mervis 1988, Mervis et al. 1991, Taylor and Gelman 1989, Tomasello et al. 1988, Waxman and Senghas1990) all except Mervis et al. ( 1991) were conducted for other purposesand none of thesecompared the learning to first labels. The prediction is not that second labels should be impossible for young children to learn, but rather that they should be harder becausechildren prefer not to have secondlabels for things. Liittschwager and I tested this hypothesis with 18- and 24-month -olds (Liitschwager and Markman 1991) . Children were briefly taught either a first label for a novel object or a second label for an object with a known first label. The results for the 18-month -olds were as expected- they success fully learned a new first label for an object but failed to learn a secondlabel. The results for the 24-month -olds were more complicated and reveal another subtlety to the way constraints on word learning work . Not only did the 24-

E.M . Markman I Early languageacquisition

month-olds success fully learn a new first label after brief training , but they learned a second label just as well. These young children clearly violated mutual exclusivity to acquire this second label. We then ran a second study with 24-month-olds to see what would happen if the task were made somewhat more demanding. Now children were required to learn two new labels rather than just one - either two new first labels or two new second labels. In contrast to the earlier results, the two -year-olds now showed evidenceof using mutual exclusivity. They succeededat learning two new first labels but failed to learn the second labels for objects. This suggestsanother way in which word -learning constraints function as default assumptions: children may rely on the default assumptionsmore heavily when the demands of the task increase.

6. Conclusions

Word -learning constraints were presente~ as necessaryto help children cope with the inductive problem involved in learning a novel word . An unconstrained, unbiased learning mechanismwould be forced to consider too many hypothesesand would be unable to converge on a candidate meaning in a reasonable amount of time. Some constraints on hypothesess are required to explain the fast learning actually seenat around 18 months to two years of age. I reviewed evidence for three specific constraints - the wholeobject, taxonomic, and mutual exclusivity assumptions. Recent work reveals that all three assumptionsare available to babies by the time of the naming explosion. Even for young children, however, other sources of information about a word ' s meaning may be available. When several sourcesof information , such as grammatical form class, eye gaze of the speaker, and all three word-learning constraints converge, word learning should be especially efficient . Yet there remain many interesting issues to be resolved about the complex and subtle interplay of word -learning constraints with each other and with other sourcesof information about a word ' s meaning.

References , manuscript, Brown . Children's useofinfonnation in word learning.Unpublished Au, T.K., 1989 University.

224

E.M . Markman / Early languageacquisition

Au, T.K. and M. Glusman . Theprincipleof mutualexclusivityin word learning: To honor , 1990 or not to honor? Child Development 61, 1474 - 1490 . Au, T.K. and EM . Markman, 1987 . Acquiringword meaningsvia linguisticcontrast. Cognitive 2, 217- 236. Development Baldwin, D.A., 1989 . Prioritiesin children's expectations aboutobjectlabelreference : Form over - 1306 color. Child Development 60, 1291 . Baldwin, D.A., 1991 . Infants' contributionto the achievement of joint reference . Child Development 62, 875-890. Baldwin, D.A., in press . Clarifyingthe role of shapein children's taxonomicassumption . Journal of ExperimentalChild Psychology . Baldwin, D.A. andE.M. Markman, 1989 . Mappingout word-objectrelations: A first step. Child 60 381 398 . Development , . Roleof adult input in youngchildren's categoryevolution, , R. L. andC. B. Mervis, 1988 Banigan II : An experimental study. Child Language15, 493- 505. Bauer, P.J. andJiM . Mandler, 1989 . Taxonomies and triads: Conceptualorganizationin one-totwo-yearolds. CognitivePsychology 21, 156- 184. Becker . Children's useof shapein extendingnovellabelsto animate , A.H. and B.T. Ward, 1991 : versus . CognitiveDevelopment 6, 3- 16. objects Identity posturalchange Bloom, B., 1994 . Possiblenames: The role of syntax-semanticsmappingsin the acquisitionof . Lingua92, 297- 329(this volume). nominals Bloom, L., K. Lifter andJ. Broughton, 1985 . Theconvergence of earlycognitionand languagein the secondyearof life: Problemsin conceptualization and measurement . In: M. Barrett(ed.), Children's single -word speech , 149- 180. New York: Wiley. . The principle of contrast: A constraint on languageacquisition Clark, E.V., 1987 . In : B. MacWhinney(ed.), The 20th Annual CarnegieSymposiumon Cognition, 1- 33. Hillsdale, NJ; Erlbaum. Clark, E.V., 1990 . On the pragmaticsof contrast. Journalof Child Language17, 417- 431. Clark, E.V., 1991.Acquisitionalprinciplesin lexicaldevelopment . In: S.A. Gelman, J.P. Byrnes on languageand thought: Interrelationsin development (eds.), Perspectives , 31- 71. Cambridge . Press University . The development of representational skills. In: K.W. Fischer(ed.), Levels Corrigan, R., 1983 -Bass and transitionsin children's development . , 51- 64. SanFrancisco , CA: Jossey Dockrell, J. and R. Campbell . 1986 Lexical in the , acquisitionstrategies preschoolchild. In: S. Kuczaj, M. Barrett(eds.), The development of word meaning . , 121- 154. Berlin: Springer Dromi, E., 1987 . Early lexicaldevelopment . New York: CambridgeUniversityPress . Echois, C.H., 1990 . An influenceof labelingon infant' s attentionto objectsand consistency : . Paperpresented at the InternationalConference on Implicationsfor word-referentmappings Infant Studies . , Montreal, Quebec , April 1990 . Infant' s attentionto objectsand consistency Echois, C.H., 1991 in linguisticand non-linguistic context. Paper presentedat the Biennial Meeting of the Societyfor Researchin Child . , Seattle , WA , April 1991 Development Fisher, C., L. R. Gleitmanand H. Gleitman, H. 1991 . On the semanticcontentof subcategorization frames.CognitivePsychology23, 331- 392. Fisher, C., D.G. Hall, S. Rakowitzand L. Gleitman, 1994 . Whenit is betterto receivethan to give: Syntacticand conceptualconstraintson vocabularygrowth. Lingua 92, 333- 375 (this volume).

E.M . Markman I Early languageacquisition

225

Gathercole, V.C., 1989 . Contrast: A semantic constraint ? Journalof Child Language16, 685- 702. Gelman, S.A. and M. Taylor, 1984 . How two-year-old childreninterpretproperand common namesfor unfamiliarobjects . Child Development 55, 1535 - 1540 . Golinkoft', R.M., C.B. Mervis and K. Hirsh-Pasek . Early object labels: The casefor , 1992a lexicalprinciples . Unpublishedmanuscript . Golinkoft', R.M., K. Hirsh-Pasek . Young childrenand , L.M. Baileyand N.R. Wenger , 1992b adultsuselexicalprinciplesto learnnewnouns. Developmental Psychology28, 84- 89. Gould, J.L. and P. Marler, 1984 . Ethologyand the natural history of learning. In: P. Marler, H.S. Terrace(eds.), The biologyof learning,47- 74. Berlin: Springer . Hall, D.G., 1991 . Acquiring proper nounsfor familiar and unfamiliaranimateobjects: Two' - 1154 . Child Development 62, 1142 . year-olds word-learningblases . Learninghow to mean. In : E.H. Lenneberg Halliday, M.A.K., 1975 , E. Lenneberg(eds.), Foundationsof languagedevelopment : A multidisciplinaryapproach , Vol. 1, 239- 265. New York: AcademicPress . Hutchinson . Constraintson children's implicit hypotheses , J.E., 1984 about word meanings . , StanfordUniversity, Stanford, CA. Unpublisheddoctoraldissertation Hutchinson . Children's sensitivityto the contrastiveuseof objectcategoryterms. , J.E., 1986 ' 25, 49- 56. Papersand Reportson Child LanguageDevelopment Huttenlocher . Earlyword meanings : Thecasefor objectnames , J. andP. Smiley, 1987 . Cognitive Psychology19, 63- 89. . What' s a name? On thechild' s acquisitionof proper Katz. N., E. BakerandJ. Macnamara , 1974 andcommonnouns. Child Development 45, 469- 473. Landau, B., 1994 . Where's what and what's where: The languageof objectsin space . Lingua92, 259- 296(this volume). Landau, K.B., LiB. Smith and SiS. Jones . The importanceof shapein early lexical , 1988 3, 299- 321. learning.CognitiveDevelopment Landau, B., L. B. Smithand S. Jones . Syntacticcontextand the shapebiasin children's , in press andadultslexicallearning. Journalof Memoryand Language . . Mutual exclusivityas a default assumptionin , J.C. and E.M. Markman, 1991 Liittschwager secondlabellearning. Paperpresented at the biennialmeetingsof the Societyfor Research in Child Development . , Seattle , April 1991 Lock, A., 1980 . The guidedreinventionof language . London: AcademicPress . Markman, EM., 1989 . Categorizationand namingin children: Problemsof induction. Cambridge , MA : MIT Press , BradfordBooks. Markman, E.M., 1992 . Constraintson word learning: Speculations about their nature, origins, and domainspecificity . In : M.R. Gunnar, M.P. Maratsos(eds.), Modularity and constraints in languageand cognition: The MinnesotaSymposiaon Child Psychology , 59- 101. Hillsdale, NJ: Erlbaum. Markman, E.M. and J.E. Hutchinson . Children's sensitivityto constraintson word , 1984 : Taxonomic vs. thematic relations . meaning CognitivePsychology16, 1- 27. Markman, E.M. and GiF. Wachtel, 1988 . Children's useof mutualexclusivityto constrainthe of words. CognitivePsychology 20, 121- 157. meanings Markman, E.M. and J. Wasow, in preparation . Very youngchilren's useof mutualexclusivity . Markow, D.B. and SiR. Waxman . The influenceof labels on 12-month-olds' object , 1992 at the InternationalConference on Infant Studies categoryformation. Paperpresented , Miami, Florida, May 1992 .

226

E.M . Markman / Early /anRUQY.E~acquisition

Marler, P. and HiS. Terrace(eds.), 1984 . The biologyof learning. Berlin: Springer . McShane . The development of naming. Linguistics17, 879-905. , J., 1979 Merriman, W.E. and L.L. Bowman . The mutual exclusivitybias in children's word , 1989 . of the for Researchin Child Development54( 3/4), Serial learning Monographs Society No. 220. Mervis, C.B., and L.M. Long, 1987 . Words refer to wholeobjects: Young children's interpretation of the referentof a novelword. Paperpresented at the biennialmeetingof the Society for Research in Child Development , Baltimore, MD. Mervis, C.B., RiM . Golinkoff and J. Bertrand, 1991 . A refutationof the principleof mutual . Paperpresented at the Societyfor Research in Child Development , Seattle , WA. exclusivity . Childrenacquireword meaningcomponents , L.G., H. Gleitmanand L.R. Gleitman, 1992 Naigles from syntacticevidence . In: E. Dromi (ed.), Languageand cognition: A developmental . Norwood, NJ: Ablex. perspective . Structureand strategyin learningto talk. Monographsof the Societyfor Nelson, K., 1973 Research in Child Development 38( 1/ 2), SerialNo. 149. Nelson, K., 1988 . Constraintson word learning? CognitiveDevelopment 3, 221- 246. Nelson, K. and J. Lucariello, 1985 . The development of meaningin first words. In: M. Barrett ' - word speech : Wiley. , 59- 86. Chichester (ed.), Childrens single W. V.O. . Word and object. Cambridge . , 1960 , MA : MIT Press Quine, Roberts . Linguistic versusattentional influenceson nonlinguistic , K. and M. Jacob, 1991 in 15-month-old infants. CognitiveDevelopment 6, 355- 375. categorization Rozin, P. and J. Schull, 1988 . The adaptive -evolutionarypoint of view in experimentalpsychology and , Vol. I. In: R.C. Atkinson, R.J. Herrnstein , G. Lindsey, R.D. Luce(eds.), Perception motivation(2nd ed.), 503- 546. New York: Wiley. Shettleworth . Naturalhistoryandevolutionof learningin nonhumanmammals . In: P. , S.J., 1984 Marler, 8 .S. Terrace(eds.), The biologyof learning, 41~ 33. New York: Springer . . Contentandcontextin earlylexicaldevelopment . , L.S., F. Batesand I. Bretherton , 1981 Snyder Journalof Child Language8, 565- 582. . Ontological categoriesguide young children's , 1991 Sola, N.N., S. Carey and E.S. Spelke inductionsof word meaning:Obj~ t termsand substance terms. Cognition38, 179-211. . Incorporatingnew words into the lexicon: Preliminary Taylor, M. and S.A. Gelman, 1989 59, 411- 419. evidencefor languagehierarchiesin two-year- old children. Child Development . The socialbasesof languageacquisition . SocialDevelopment1, 67- 87. Tomasello , M., 1992 . Object permanence and relational words: A lexical Tomasello , M. and M.J. Farrar, 1986 trainingstudy. Journalof Child Language13, 495- 505. . Theeft'~ t of previouslylearnedwordson Tomasello , M., S. Mannieand L. Werdenschlag , 1988 . Journalof Child Language15, 505- 515. the child' s acquisitionof wordsfor similarreferents Ward, T.BE . Vela, M.L. Peery,S. Lewis, N.K. Bauerand K. Klint , 1989.Whatmakesa vibble . Child Development 60, 214- 224. a vibble: A developmental studyof categorygeneralization . The development of an appreciationof specificlinkagesbetweenlinguistic Waxman , S.R., 1994 . Lingua92, 229- 257(this volume). andconceptualorganization . The influenceof words vs. tones on 9-month-old Waxman , 1992 , S.R. and M.T. Balaban . Paper presentedat the InternationalConferenceon Infant infants' object categorization . Studies , Miami, Florida, May 1992 's useof . Preschooler relationsin classifica Waxman , S.R. and R. Gelman, 1986 superordinate tion. CognitiveDevelopmentI , 139- 156.

E.M . Markman I Early languageacquisition

227

Waxman . Nounshighiijbt categoryrelationsin 13-month-old infants. , S.R. and L. Helm, 1991 Posterpresented at the Societyfor Research in Child Development . , Seattle , WA, April 1991 Waxman . Nouns mark categoryrelations: Toddlers' and , S.R. and T.O. Kosowski, 1990 ' - 1473 . Child Development 61, 1461 . preschoolersword-learningblases Waxman 1990 . Relations word , S.R. and A. Senghas , among meaningsin early lexicaldevelopment . Paperpresented at the InternationalConference for InfancyStudies , Montreal, Canada , . April 1990 Woodward . The role of the whole object assumptionin early word learning. , A. L., 1992 , StanfordUniversity, Stanford, CA. Unpublisheddoctoraldissertation Woodward . Constraintson learningas default assumptions : , A.L. and E.M. Markman, 1991 ' s ' The mutual ' Commentson Merriman & Bowman exclusivitybias in childrens word ' Review14, 57- 77. learning. Developmental

) 229- 257. North-Holland Lingua92 ( 1994

229

Thedevelopment of an appreciation of specificlinkagesbetween linguisticand * organization conceptual Sandra R. Waxman Road. Evanston . Northwestern . 2029Sheridan . 1L 60208 , USA Department of Psychology University Evidencefrom infants and toddlers, and from preschoolchildren learningeither English, French, or Spanishas their first languageare summarizedto revealthe emergence of specific . The data suggestthat infantsbeginthe linkagesbetweenlinguisticand conceptualdevelopment processof word learningwith a generalexpectationthat words (independentof the linguistic form) referto objectsand objectcategories . This initial, rudimentarylinkagegivesway to more ) and particular specificpairingsbetweenparticular linguisticforms (e.g., nounsvs. adjectives ). Thesemorespecificlinkages typesof meaning(e.g., objectcategoriesvs. propertiesof objects . maydependupon languageexperience

. ' 'WhenI makea word do a lot of work like that I , alwayspay it extra (LewisCarroll 1895 ) I . Introduction Humans are uniquely endowed with the capacity to build complex , flexible , ' ' and creative linguistic and conceptual systems. Infants and toddlers remarkable achievements in each of these arenas have engaged researchers for decades. Yet in recent years , it is the relation between linguistic and conceptual development that has come to occupy center stage. Some of the most exciting current work has been designed to explore the relation between early ' linguistic and conceptual development in the young child s acquisition of the lexicon . This new , integrative approach has brought into sharp focus a fascinating puzzle . We know that infants acquire their native language naturally at a . This work wassupportedin part by grant # HD30410from NIH . Thanksto Marie Balaban , D. GeoffreyHall, ElizabethShipley , and an anonymousreviewerfor their commentson a previousversion. - 3841/94/ $07.00 ~ 1994- ElsevierScienceB.V. All rightsreserved 0024 SSDI0024- 3841( 93) EO041- S

230

andconceptual S.R. Woxman I Linguistic organization

remarkable pace (Carey 1978, Dromi 1987, Gopnik and Meltzoff 1987, Nelson 1983) . We also know that even before learning the words to express them, infants appreciate many different kinds of conceptual relations among objects, including category relations, thematic relations, causal relations, and event-related or associative groupings (Bornstein 1984, Leslie and Keeble 1987, Mandler et al. 1987, Younger and Cohen 1986) . These early linguistic and conceptual achievementsset the stagefor what has beendescribedas ' the inductive problem of word learning' (Goodman 1983, Quine 1960, Carey 1990). The problem is that , in principle, the richnessand flexibility of infants' conceptual abilities should complicate the task of mapping new words to their meanings. To understand why this is the case, consider a typical word learning scenario, in which an adult introduces a child to a novel object (say, a ' ' flamingo) and offers a novel label ( a flamingo ) . Let us assumethat both the child and adult are focusing attention on the same object or scene( Baldwin and Markman 1989, Tomasello 1988) . If children have the conceptual ability to appreciateso many different kinds of relations involving that object, and if ' each of these is a potential candidate for the new word s meaning, then how do infants select from among these many possible meanings when determining what the new word is intended to convey? How do infants so rapidly learn that a given word (e.g., flamingo) may apply to a particular whole object and may be extended to other members of that object category (e.g., other flamingos)~but not to salient properties of the object (e.g., its long neck or unusual color), to salient actions in which it is engaged(e.g., feeding its young), or to salient thematic relations (e.g., a flamingo and sand) ? If children had to rule out theseand countlessother logically possiblecandidate meanings, word learning would be a formidable task indeed.

1.1. Solvingthe inductiveproblem Yet despite the logical difficulty of the task, young children rapidly and success fully map novel words and meanings. This observation has led several researchers , working from several different paradigms, to suggestthat children come to the task of word learning equipped with certain implicit blases or expectationswhich lead them to favor some types of conceptual relations over others when ascribing meaning to a new word (Chomsky 1986, Landau and Gleitman 1985, Pinker 1984, Markman 1989, Waxman 1990, 1991) . The claim is that these expectations reduce the logical difficulty of word learning

S.R. Waxman I Linguistic and conceptualorganization

by narrowing the range of candidate meanings a child will consider for any given new word. Several such implicit blaseshave been proposed. For example, researchin several different laboratories has converged on the finding that children expect that the first word applied to a novel object will refer to the whole object and other members of its basic-level kind , rather than to its parts or other salient aspects(Markman and Wachtel 1988, Taylor and Gelman 1988, Hall et al. 1993, Markman 1989) . Further evidencerevealsthat children, like adults, expect that different words will contrast in meaning (Clark 1987, Golinkoff et al. 1992, Markman 1984, 1989; Merriman and Bowman 1989, Waxman and Senghas1992). A third type of bias or predisposition will serve as the focal point of this article. There is now considerableevidencethat children use the grammatical form of a novel word (e.g., count noun, proper noun, adjective) as a guide to determining its meaning (Brown 1957, Katz et al. 1974, Landau and Gleit man 1985, Naigles 1990, Gleitman et al. 1987, Hall and Waxman 1993) . For example, by two to three years of age, English-speaking children expect objects and object categories(e.g., flamingos, birds, animals) to be marked by count nouns (Markman and Hutchinson 1984, Waxman and Gelman 1986, Waxman and Kosowski 1990, Waxman and Hall 1993) ; they expect substances (e.g., wood ; gel) to be marked by mass nouns (Dickinson 1988, Sola et al. 1991); and they expect object properties (e.g., size, color ) to be marked by modifiers (Gelman and Markman 1985, Hall et al. 1993, Taylor and Gelman 1988, Waxman 1990) . Notice that any linkage between grammatical form class and meaning requires that the word learner has (a) the linguistic capacity to distinguish among the relevant syntactic categories (e.g., count noun vs. mass noun vs. adjective) in her language, and (b) the conceptual or perceptual ability to appreciate the various kinds of relations among objects. In sum, recent work has establishedthat young children appreciatelinkages between particular types of words (e.g., count nouns vs.. adjectives) and particular types of conceptual relations. These linkages help to explain how children so rapidly map novel words appropriately to their meanings. For , although children appreciatemyriad kinds of conceptual relations, only some of these relations become lexicalized. Children do not sample randomly among thesemany possible relations when determining the meaning of a new word. Instead, particular kinds of conceptual relations are favored in the context of learning particular kinds of words.

232

S.R. WaxmanI Linguisticandconceptual organization

1.2. Overview of the article

In this article, I take as a starting point the general hypothesis that (a) young children are sensitiveto preciserelations linking linguistic with conceptual development, and that (b) these linkages promote rapid lexical acquisition and foster the establishmentof powerful systemsof conceptual organization . I begin by reviewing the evidence in support of this position with preschool-aged children, focusing primarily on the role of linguistic information ' (e.g., count nouns vs. adjectives) in the young child s ability to form categories of object kinds at various hierarchical levels (e.g. flamingo, bird , animal) . Seefigure I . Next , I turn to theoretical questions concerning the origin or developmental status of these linkages between word learning and conceptual organization, asking how and when they emerge in the developing child , This discussion highlights some limitations in the existing literature and underscores the importance of two distinct , but complementary research approaches. First , becauseso much of the existing literature is devoted to primarily preschool children who have already made significant linguistic advances, we are left with a very limited understanding of how the child acquires these important linkages early in development. Therefore, the goal of the first approach is to chart the emergenceof these links in preverbal infants and in toddlers by examining the influence of language on their categorization abilities. Second, becausethe existing researchhas been basedalmost exclusively on English-speaking subjects, it is unclear whether these linkages are universal features of human development or specific to English. Therefore, the goal of the secondapproach is to seekevidencefor theselinkages in children learning languagesother than English. Although these developmental and cross-linguistic research programs are still very much in progress, our initial results converge to suggest that the linkage between count nouns and object categories is evident even at the onset of language acquisition and may be a universal phenomenon. In contrast, the data suggesta very different developent course for the linkage between adjectives and properties of objects. This linkage appears to emerge later in development and to vary across languages.

S.R. WaxmanI Linguisticandconceptual organization

SQq1

~ J [ [ ~ e e = A . 0 O [ '~ 8J~ ' i = ~ s 1 s ~ ob ~

";.0 -e ~

~ ~ ~ ~ ~ ~ ~ ~ ~

R =0 e 0 oWE u d ~ .ce Q 0B .9 CE ~ ~ := 8 . .. ~ ~ ~

.. ~ ~ .. ~ ~ ~ ~

233

234

S.R. Waxman / Linguisticand conceptualorganization

2. Describing the phenomenon: The influence of linguistic form class infor mation on object categorization in preschool children

An essential task in early development is to form categories that capture the commonalities among objects and to learn words that describe these categories. A considerableamount of scholarly attention has been devoted to examining the establishment of categories within hierarchical systems of organization. Hierarchical systems are exceptional for their efficiency in organizing existing information and for their power in generating new information ; it therefore stands to reason that developmentalistswould seek to understand when such systemsare available to the young child. Empirical work with both adults and children has singled out one particular hierarchical level - referred to as the basic level - as being most salient psychologicalliy (Rosch et ale 1976) . The basic level, which occupies a midlevel position within a hierarchy, has been shown to have a privileged status on a range of psychological tasks. (See Rosch et ale 1976 for details.) Although it has been difficult to account formally for this privileged psychological status, the construct of a basic level has proven useful as a summary description or heuristic in research with adults and children alike (cf., Gleitman et ale 1987) . Although Mandler and her colleagues have argued against this position in favor of the view that infants initially conceptualize object~ at a level more abstract than the basic level (Mandler 1988, Mandler and Bauer 1988, Mandler et ale 1991), the weight of the evidence overwhelming favors the developmental primacy of the basic level. For example, one of the most robust findings in the developmental literature is that preschool children succeedin classifying and labeling objects at the basic level long before they do so at other hierarchical levels (Anglin 1977, Brown 1958, Mervis 1987, Mervis and Crisafi 1982, Rosch et al. 1976) . However, because the inductive and organizational power of hierarchical systemsderives from relations among categoriesat various levels of abstraction ' , developmentalistshave also been concerned with children s acquisition of categories beyond the basic level. Although preschool children have considerabledifficulty forming superordinate and subordinate level categories under most circumstances, their performance improves dramatically when they are introduced to novel words in the context of categorization tasks. Indeed, it is at the non-basic levels that the interplay between word learning and conceptual organization has become especially evident. To observe this ' interplay, we have compared children s ability to form object categories at various hierarchical levels with , and without , novel labels.

S. R. Waxman/ Linguistic and conceptualorganization

235

In an early study, we examined the impact of introducing novel nouns in a superordinate level categorization task ( Waxman and Gelman 1986) . The ' ' experimenter introduced preschool children to three very picky puppets and then displayed three typical members (e.g., a dog, a horse, a cat) of a superordinatecategory (e.g., animal) to indicate the type of thing each puppet would like. She then asked children to sort additional items for each puppet. Children in the Instance condition , who sorted the additional pictures (various membersof the classesanimals, clothing , and food ) with no further instructions, perfonned only slightly better than would be expected by chance. This is consistent with traditional reports that children have difficulty establishing superordinate relations (Inhelder and Piaget 1964, Rosch et ale 1976). In contrast, children in the Novel Label condition , who encountered the sametypical instances, but were also introduced to a novel Japaneselabel for each superordinate class (e.g., ' These are the dobutsus, these are the gohans) formed superordinate classesvery success fully . Simply introducing them to novel labels led these children to classify as success fully as other children who had been given familiar English superordinate labels for the classes (e.g., ' These are animals, these are clothes' ) . Clearly, novel count nouns effectively oriented preschool children toward object categories and licensedthe induction of superordinate level categories. Data from Markman and Hutchinson. ( 1984) have revealed that count nouns also highlight basic level object categoriesfor 3- and 4- year-old children. This result has linked one particular linguistic fonn class - count nouns to object categories at the basic and superordinate levels. This intriguing finding raised two important questions, both of which concern the specificity of the linkage: First , do novel count nouns draw attention to object categories at all hierarchical levels, or is this effect specific to the basic and superordinate levels? Second, are object categorieshighlighted in the context of word learning in general, or is this focus specific to learning novel nouns? To address these questions, I systematically compared the effect of introducing either novel nouns or novel adjectivesin a multiple -level classification task (Waxman 1990) . Each child in this study classified pictures of objects from three contrastive classesat three different hierarchical levels (subordinate , basic and superordinate) within the two different natural object hierarchies (animals and food ) depicted in figure 1. As in Waxman and Gelman ' ' ( 1986), the experimenter introduced three very picky puppets and revealed three typical membersof each class to indicate the type of thing each puppet would like. Children in the No Word condition sorted with no further clues. Children in the Novel Noun condition were introduced to a novel noun in

S.R. Waxman / Linguistic and conceptualorganization

conjunction with the photographs from each class (e.g., These are the akas; these are the dobus) . Children in the Novel Adjective condition also heard novel words, but the words were presented within an adjectival syntactic context (e.g., ' These are the ok-ish ones, theseare the ones that are doh-ish' ) . The children in this experiment were very sensitiveto the linguistic context in which the novel words were introduced. Novel nouns facilitated object 1 categorization at the superordinate, but not the subordinate level. In the Novel Adjective condition , this pattern was completely reversed. Unlike nouns, novel adjectives supported the formation of subordinate level object categories, but exerted no demonstrableeffect at either the basic or superordinate levels. Thus, each of these different linguistic forms facilitated object categorization at particular hierarchical levels. An interestingparallel to this phenomenonin children has beendocumented acrossa wide variety of adult languages , both spokenand signed. According to data count nouns , typically mark objectsand object categoriesat ethnobiological the basicand superordinatelevelswhile adjectivestend to mark subordinatelevel distinctions (Berlin et aI. 1973, Newport and Bellugi 1978). Although these correlationsbetweenlinguistic form and object categoriesat particular hierarchical levelsare not perfect, they do suggestthat a relation betweennaming and object categorizationmay exist throughout the lifespan. (SeeWaxman, 1991, for a more thorough discussionof this literature and its relevanceto acquisition.) The developmental finding that novel nouns and adjectiveseach produced systematic, but distinct , patterns of results at distinct hierarchical levels revealsthat three-year-olds are not only sensitive to the distinctions between these two linguistic forms, but also consider linguistic form as relevant to establishing meaning. This important finding constitutes strong support for the hypothesis that by three years of age, children appreciate powerful and preciselinkages betweenword learning and conceptual organization. Notice , however, that the data from preschool-aged children cannot address crucial questions concerning the development of these linkages in infants and toddlers. (See Nelson, 1988, for an extended discussion of this point .) Neither does the existing evidence address questions concerning the universality of such linkages across languages. These questions become 1 In fact, although novel nouns facilitated classification at the superordinate level, they made classification at the subordinate level more difficult . Children in the Novel Noun condition classified less success fully at the subordinate level than did their agemates in the No Word condition . This very interesting result has spumed a whole independent line of research (see ' Waxman et al. 1991) which suggeststhat children s interpretations of novel words are mediated infonnation . their lexical and conceptual by existing

S.R. Waxman I Linguistic and conceptualorganization

237

especiallyengagingwhen they are consideredin light of the nonnative pattern of early lexical acquisition. (See Nelson, 1988, for a review of this pattern; also seeGopnik and Choi , 1990, for a suggestionthat this pattern may not obtain in the acquisition of Korean.) The milestones of early lexical acquisition have beenwell-documented. Infants typically produce their first words at approximately 12 months of age and continue to add new words to their productive vocabularies at a gradual pace. However, at approximately 17- 20 months, both the pace and character of lexical acquisition changesdramatically . Infants exhibit a sudden burst in vocabulary development (Benedict 1979, Carey and Bartlett 1978, Goldfield and Reznick 1990) . Becausemost of the words acquired at this period and at this pace are basic level count nouns (Dromi 1987, McShane 1980, Gentner 1982), this period has been dubbed the naming explosion. The naming explosion draws to a close as infants begin to produce combinatorial speech, typically around their secondbirthdays. Clearly, any thorough account of the early development of an appreciation of linkages between word learning and conceptual organization must be compatible with thesemilestonesin lexical development. Bearing this in mind , three broad alternative accountsconcerning the developmentof this appreciation warrant consideration. 3. Three alternative accounts of the development of aD appreciation of Unkages between Hnauisti (: form and conceptual organization

The first alternative account posits that these linkages are learned entirely on the basis of infants ' experience with human language. On this account, infants embark upon the process of lexical development with no a priori expectations concerning linkages between word meaning and conceptual organization . Instead, they learn their first words in an unconstrained fashion, slowly establishing the mappings betweenwords and their meanings. Later, once they have made a sufficient number of word -to -meaning mappings, infants may come to notice a correlation between particular linguistic forms (e.g., nouns, adjectives) and particular types of meaning (e.g., object categories, object properties) . They may then exploit this correlation in future word learning (seeNelson, 1988, for a fuller discussionof this account) . This first account is plausible becauseprior to the naming explosion, lexical acquisition is indeed comparatively slow; later, perhaps once infants come to notice the linkages betweenform class and meaning, their rate of acquisition increasesexponentially. It is possible that it is only at the onset of the naming explosion that infants have accumulated enough word -to -meaning mappings

238

S. R. Waxman I Linguistic and conceptualorganization

to make the appropriate induction regarding the relation between linguistic form and meaning. If this account is correct, then infants who have yet to commencethe naming explosion should evidenceno labeling effects. Instead, novel words should influence object categorization for infants only after the onset of the naming explosion. The secondaccount posits that the specific linkages that we have observed in preschool children are available even at the very onset of lexical acquisition . This alternative requires that preverbal infants expect (a) that there are distinct linguistic forms and (b) that these distinctions are relevant to establishing meaning (c.f ., Pinker 1984, Grimshaw 1981) . If this account is correct, then novel words should influence preverbal infants in just the same way as they influence older infants and preschool children. That is, even infants who have yet to commence the naming explosion should expect that object categorieswill be marked by count nouns and that object properties will be marked by modifiers. The third alternative account strikes a balancebetweenthose outlined above. On this account, infants begin the processof lexical acquisition equippedwith a generalrudimentary expectation that will becomefurther refined overdevelopment and with experiencewith the particular language to which they are exposed. On this account, infants will interpret words (independent of their linguistic form ) as referring to objects and object categories. This alternative is plausible becauseprior to about two yearsof age, infants do not yet distinguish among linguistic form classesin their languageproduction or comprehension (Bloom 1990, Gordon 1985, McPherson 1991, Prasada 1993, Valian 1986). Later, at around 2 years of age, when infants do begin to distinguish among the linguistic forms, so do they discover the finer correlations made in their native languagebetweenparticular linguistic forms and meaning. There are actually two variants of this account. One possibility is that infants begin with an abstract expectation that particular linguistic forms will mark particular kinds of conceptual relations; however, becausethey have not yet learned how these linguistic distinctions are marked in their own language, they (mistakenly) interpret adjectives as they do nouns. Another possibility is that infants begin with an expectation that words in general will mark object kinds ; they only later learn that this linkage is true for count nouns, but not for other grammatical categories(e.g., adjectives) . These two variants are quite difficult to disentangle empirically . For in either case, the patterns of performance should be the same: prior to the onset of the naming explosion, infants should interpret all novel words, independent of their linguistic form , as referring to objects and object

S.R. Waxman/ Linguisticandconceptual organization

239

categories. This pattern would suggestthat infants embark upon the process of lexical acquisition with a rudimentary linkage between words and object categories that will become increasingly specific as a function of their experiencewith the particular syntactic distinctions drawn in their language. To adjudicate among these broad alternative accounts, I have initiated a detailed examination of the influence of words of various linguistic forms on infants' and toddler ' s object categorization. I have also begun to examine young children learning languagesother than English. To foreshadow, the results of thesetwo complementary setsof experiments converge to provide initial support for the third alternative account. Infants at 12 months begin the processof lexical acquisition with a general expectation that words (independent of their linguistic form class) will refer to objects and object categories. This initial , rudimentary linkage becomes increasingly specific in the second year of life , perhaps as a function of their own languageexperience.

toddlersand in preverbal infuts The overarching goal of this seriesof studies is to elucidate the manner in which the linkages between word learning and conceptual organization unfold in infants and toddlers. The studies are designed to go beyond the well-established finding that infants and toddlers form object categories. Instead, each study is designedto ascertain whether and how the introduction of novel words from different linguistic form classesinfluences their object categorization. Before describing the studies themselves, one important issue bears mention . It is clear that the distinctive intonational contours characteristic of ' motherese are ' especially effective in arousing and sustaining infants' attention (Fernald 1992) . Several previous studies have explored the influence of labels on infants' attention to objects by comparing performance in a label condition with performance in a silent condition (e.g., Baldwin and Markman 1989). However, in such designs, it is unclear whether the effect observedin a label condition is due to labels, per se, or to infant -directed human language, in general. Therefore, we use the mothereseregister to capture the attention of infants in all conditions. Even subjects in a No Word condition are introduced to the objects (or pictures of objects) with infant-directed speech. Then, against this ' baseline', we examine the effect of introducing words from various linguistic form classes. In this way, we are able to determine whether

240

S.R. Waxman / Linguistic and conceptualorganization

of labels, per se, or to the 4.1. Evidencefrom toddlers: Forced choiceprocedures

Two -year-old children are at an important developmental crossroad. They have just completed the naming explosion and have entered a phaseof rapid syntactic and semantic development. To examine the influence of linguistic form class on object categorization during this very active period of development ' , we compared 2-, 3- and 4- year-olds performance in a match-to-sample task. Children read through a picture book with an experimenter. On each page, there were 5 pictures: a target (e.g., a cow), two taxonomic alternatives (objects from the same superordinate class as ~he target, e.g., a fox and a zebra), and two thematic alternatives (objects that were thematically related to the target, e.g., a barn and milk ) ( Waxmanand Kosowski 1990) . Children participated in one of three conditions. In the No Word condition , the experimenter pointed to the target and said, ' See this? Can you find ' another one?' In the Novel Noun condition , she said, for example, See this ' fopin ? Can you find anotherfopin ? In the Novel Adjective condition , she said, ' for example, ' Seethis fopish one? Can you find another one that is fopish ? The child and experimenter read through the book two times. On the second reading, the experimenter reminded the children of their first choices and asked them to selectanother from the remaining (3) alternatives. In this way, we were able to examine the conditions under which children make consistently taxonomic choices. We reasonedthat if children are sensitive to a specific link between nouns and superordinate relations, then children in the Novel Noun condition should be more likely than children in the Novel Adjective and No Word conditions to selectthe superordinate category memberson a page, and not the thematic alternatives. The results with the 3- and 4- year-olds supported this prediction entirely. Only in the Novel Noun condition did children consistently select taxonomic alternatives. In both the Novel Adjective and No Word conditions, children performed at chance. Thus, superordinate relations gained priority only in the context of novel nouns, not in the context of word learning in general. Moreover, the effect of the novel noun was powerful enough to guide both a first and secondset of choices, even in the presenceof a clear thematic alternative. Two -year-olds' performance was very similar to that of the older preschoolers . Two- year-olds in the Novel Noun condition were more likely than

S. R. Waxman / Linguistic and conceptualorganization

those in the Novel Adjective and No Word conditions to select superordinate category members. However, when comparing performance in each condition to chance, one slight developmental difference emerged: 2-year-olds in the Novel Noun condition selectedtaxonomic alternatives more often than would be expected by chance; those in the No Word condition selectedtaxonomic alternatives less often than would be expectedby chance; those in the Novel Adjectivecondition were intermediate. As predicted, 2-year-olds in this condition selectedtaxonomic alternatives less often than did children in the Novel Noun condition ; however their mean rate of taxonomic selectionswas greater than would be predicted by chance. This difference in the 2-year-olds' interpretation of novel adjectives, though it is a slim one, provides an important clue into the development of an appreciation of linkages between word learning and conceptual relations. Although the linkage between nouns and object categories is clearly evident by two years of age, toddlers at this age also revealed some inclination to interpret adjectivesin a similar fashion. This suggeststhat 2-year-old children may overextend the linkage between count nouns and object categories to include new words from other linguistic form classes. This possibility is consistent with the hypothesis that infants embark upon the task of word learning with an assumption that words (not specifically nouns) highlight object categories. If this is the case, then the tendency to interpret adjectives, like nouRs, as referring to object categoriesshould be even more pronounced in younger subjects. 4.2. Evidencefrom infants: Novelty-preferenceprocedures With this question in mind , we designeda procedure to examine the impact of novel words on object categorization in 12- and 13-month -old infants (Markow and Waxman 1992, 1993) . Infants at this age are acutely interested in human language, but produce very little , if any, of their own. To accommodate the very active nature of the infants, we developed an object manipulation task, analogous to standard novelty-preferenceprocedures (see also Ruff 1986, Ross 1980, Oakes et al. 1991) . In thefamiliarization phase, the experimenteroffered the child four toys from a given category (e.g., four different animals) one at a time, in random order, for 30 secseach. This was immediately followed by the test phasein which the experimenterpresentedboth (a) a new member of the familiar category (e.g., another animal) and (b) an object from a novel contrasting category (e.g., a fruit ). In both phases, infants manipulated the objects freely. Each infant

242

S. R. Woxman I Linguistic and conceptualorganization

completed this procedure four times, with four different sets of objects: 2 basic level sets(cows vs. horses; cars vs. planes) and 2 superordinate level sets (animals vs. vehicles; tools vs. animals) . Infants were assignedto one of three conditions, which differed only in the ' experimenters comments during the Familiarization phase. See figure 2. In the Novel Noun condition , the experimenter labeled objects during the fami ' liarization phase (e.g., ' See the auto ) . In the Novel Adjective condition , she ' ' introduced the novel word in an adjectival context (e.g., Seethe aut-ish one ) . In the No Word condition , she drew attention to each object but offered no ' label (e.g., ' See this ) . The test phase was identical for infants in all three conditions. The experimenter introduced the test pair (e.g., cow vs. horse), ' ' saying, Seewhat I have . No object labels were introduced in the test phase. Becauseinfants at this age do not yet distinguish among linguistic form classes such as noun and adjective in their own language production or comprehension(Bloom 1990, Gordon 1985, McPherson 1991, Prasada 1993, Valian 1986), it is unlikely that they would consider linguistic form as relevant to establishing meaning. We therefore hypothesized that for infants at this developmental moment, object categories would be highlighted in word learning in general, not by nouns in particular . We predicted that infants in both the Novel Noun and Novel Adjective conditions would categorize more readily than would infants in the No Word condition . More specifically, we predicted that infants hearing novel words (be they nouns or adjectives) would show ( 1) a greater decreasein attention to the objects over the familiarization phase, and (2) a stronger preferencefor the novel object in the test phase than should infants in the No Word condition . The results of the experiment were consistent with these predictions. Consider first the data from the familiarization phase, depicted in figure 3. We calculated individual contrast scores to test the prediction that infants would show a linear decreasein attention across the four familiarization trials. At the basic level (figure 3a), infants in all three conditions showed this linear trend. This is consistent with arguments concerning the primacy of the basic level. However, on the superordinate level trials (figure 3b), only infants hearing novel words (in the Novel Noun and Novel Adjective conditions) showed a decreasein attention. During the test trials , the effects of novel nouns and adjectives were also quite comparable. Figure 4 displays the proportion of attention the infants devoted to the novel test object. At the basic level, infants in both the Novel Noun and Novel Adjective conditions showed a reliable novelty preference; those in the No Word condition showed no such preference. At the super-

S. R. Waxman / Linguistic and conceptualorganization FamlUarizatioa Phase P8DIi1i1ri8ti- Trial7. (ADimalVI. Fruit)

Pm Dili8iz8ticm Tri8l1 (ADiI D8lVI. Fruit)

".u-ni8h "S:V "S:1118 "..mkh 1118 , , : ee , NOUN :I1VB ee ? NOUN r 1118 1118 r S ee ADJFL " S ee rAD J B Cn B '" ,'" NO WORD :"S-ee NO WORD :"S-ee Pamililriz8timl Trial3 (AnimllVI. Fruit)

"Seewb8tI haver

Test Phase

F8I Dili8rizali - Trial4 I VI. Pruit) (ADim8

-WO '_ -dI8 :NO NOU S ee 7 -8h r"-i7 ADJ :dI8 S ee T. Trial I VI. Pnit) (ADim8

-SeewM I baver

Fig. 2.

S.R. Waxman I Linguistic and conceptualorganization - BASICLEVEL FAMlLI ARI ZATl O~ PHASE

LEVEL FA~ ILIARIZATIO:\: PHASE- SUPERORDINATE

(

~

oo

30

1

-

Q

-

0 -

-

.

-

4

NOUN ADJK " nVE NOWORD

1 1 'U :!:::.25 II ~ ~ ~ -~ ;IIz:20 ~ < IS@

-

0

-

-

0

-

.

NOUN ADJ~ NOWORD

4

~ TRIAL FAMILIARIZATIO

FAMILIARIZATIO~ TRIAL

Fig. 3. ordinate level, only infants in the Novel Noun condition showed this preference. These are very striking results, for they reveal a nascent appreciation of a linkage between words and object catagories in infants who have yet to commencethe naming explosion. This finding weakensconsiderably the first alternative account - that prior to the naming explosion, infants fail to appreciate any linkages betweenword learning and conceptual organization. ' Clearly, novel words do focus infants attention on object categories. These results also weaken the second alternative account - that infants embark upon the processof word learning with a fully developedappreciation of the specificlinkages betweentypes of words (e.g., nouns and adjectives) and types of meaning. Instead, these data support the third view - that 12- and 13month -old infants begin the processof word learning with a general expectation that words, be they nouns or adjectives, will refer to object categories. How do the piecesof evidencefrom the infants fit together with the data from the 2-year-olds? Together, the data suggestthat from the earliest stages ' of lexical acquisition, count nouns focus infants attention on object categories . Indeed, we have also obtained converging evidenceon this point with 16- and 2O-month -old subjects, using an entirely different method (Waxman

245

S. R. Waxman I Linguistic and conceptualorganization

Tf:ST TRIALS 0

.

70 . . 65 . . .

~

~

ADJB: ' nvE

0

NOWORD

.

0

.

. 60 .

OOS .

O

*

<

.

5

.

5

0 0

*

p * OS.

O <

p 0

BASIC

SUPERORDINATE

LEVEL Fig.4. and Hall 1993). However, initially , this focus is not specific to count nouns. At 12 and 13 months, both nouns and adjectives focus infants' attention on ' object categories. The infants ability to distinguish between linguistic form classesand to use these distinctions as a guide to establishing word meaning must undergo important developmental change during the second year. By the time they are approximately two years of age, infants begin to teaseapart the syntactic form classes; by two and a half years, we begin to get evidence that they treat nouns and adjectivesdifferently with respect to object categorization (Waxman and Kosowski 1990, Taylor and Gelman 1988) . Put differently , the data suggestthat .the affinity betweencount nouns and object categoriesis evident even in preverbal infants, but the specificity of this

246

S. R. Waxman / Linguistic and conceptualorganization

affinity increasesover development. This pattern fits nicely with some anecdotal evidence concerning early word learning. One interesting observation has been made by severalresearchers: Prior to the naming explosion, infants seemto interpret most words, independent of syntactic form , as referring to objects and categoriesof objects. This is illustrated by the oft -cited anecdote ' regarding infants initial interpretation of adjectives like hot. In the earliest ' ' ' ' stagesof lexical acquisition, when children hear, Don t touch that. It s hot , they often interpret hot as referring to an object (e.g. a stove), rather than to a salient property of that object. Indeed, all of the data documenting that children use syntactic form to affix meaning to a new word comes from children who have at least embarked upon the naming explosion (Hall 1992, Hall et ale 1993, Katz et ale 1974, Markman and Wachtel 1988, Sola et ale 1991, Waxman 1990, Waxman and Kosowski 1990) . Basedon the data reviewed thus far , I have suggestedthat the appreciation of a linkage between count nouns and object categories undergoes no developmental change: it appears to emerge early, requiring little , if any, experiencewith the language. In contrast, an appreciation of specific linkage between other grammatical categories (e.g., adjectives, mass nouns, verbs) and meaning emergeslater in development and may depend upon language experience. Notice , however, that this suggestion is based almost exclusively on English-speaking subjects. This is a serious limitation , for it is important to determine whether the patterns observedin our English-speaking samplesare universal to human development. (SeeSlobin, 1985, for excellentdiscussionsof the necessityof cross-linguistic work in establishingtheories of acquisition.)

S. Cross - linguisticdevelopmental studies:FrenchandSpanish Becausecross-linguistic evidenceis essentialin piecing together the nature and development of an appreciation of linkages between linguistic and conceptual development, I have begun to examine these linkages in young children learning languages other than English. The studies have been designedto pinpoint universals, and at the same time, document any differences in children' s appreciation of the relation between linguistic form and meaning. If the hypotheses proposed here are correct, then the effect of introducing novel nouns should be universal: In all languages, count nouns should highlight object categories. In contrast, the effect of introducing novel adjectivesmay vary, depending upon the particular languagebeing acquired.

S. R. WaxmanI Linguisticandconceptual organization

247

To date, our sample includes unilingual speakersof two different language communities. The French-speaking preschool children came from Montreal , Canada. All of the children in this sample were members of families for whom French was the language spoken at home. Moreover, these children were enrolled in French-speaking preschool programs. The Spanish-speaking children came from Buenos Aires, Argentina . Despite the similarities among these two Indo -European languages, there are variations in their grammars that bear on the questions at hand. For example, in Spanish and French, as opposed to English, each object or class of objects has associatedwith it a grammatical gender. Therefore, the words (e.g., nouns, adjectives, determiners ) which refer to these carry gender markings as well. One possibility is that the gender markings associatedwith the various terms would influence the children' s interpretations of the novel words. Briefly stated, we found that this was not the case(Waxman et al., in preparation) . Another difference in theselanguageswas of greater potential relevance. In Spanishand French, nouns are typically dropped if the grammatical subject is recoverable from context. If I have six mugs before me, in English, I ' ' distinguish them linguistically by pairing the noun mug with an adjective ' ' ' ' (e.g., the big mug or the big one ) . In Spanish, such constructions are ungrammatical. Instead, the noun is dropped, leaving the determiner and ' ' adjective (e.g., el grande ) to refer to the intended mug. This construction is also common (although not obligatory) in French, where one might ask for ' la ' petite to refer to the smallest mug. In such instances, adjectives have referential status and convey nominal information . This grammatical difference in the referential status of adjectives may have consequencesfor chil' dren s interpretations of novel words. Perhaps in Spanish, novel adjectives, like nouns, will highlight category relations. Perhapsin Spanish, the influence of novel adjectivesis lessdistinct from that of novel nouns. To addressthis hypothesis, ,we adapted the five-item forced-choice method (Waxman and Kosowski 1990) to test 2- to 4- year-old unilingual speakers of French and Spanish (Waxman et al., in preparation) . For French-speaking preschoolers, the results were identical to those obtained in English : Children in the Novel Adjective and No Word conditions demonstrated no particular preferences; only those in the Novel Noun condition chose predominantly taxonomically related items. These data support the view that the specific linkage between nouns and object categoriesin English is evident in French as well. However, our results with the Spanish-speaking children were different : Like English- and French-speaking children, Spanish-speaking preschoolers

S.R. WaxmanI linguistic andconceptual organization

in the Novel Noun condition exhibited a strong preferencefor taxonomically related items; those in the No Word condition showed no particular preference for taxonomic, thematic, or gender-related matches. The essentially random performance in this condition replicates the results from our other two language samples. However, unlike their English- and French-speaking counterparts, Spanish-speaking children in the Novel Adjective conditions did display a systematic inclination toward the taxonomically related items. In ' Spanish, then, adjectives also seem to focus young children s attention on superordinate category relations. This finding has now been replicated twice with two independent groups of Argentine preschool children ( Waxman et al., in preparation) . - This observed difference in Spanish-speaking children' s interpretation of novel adjectives cannot be attributed to any procedural differences between the Spanish and English protocols, for the proceduresemployed were identical in all languages. Neither can the differences be attributed to the stimuli themselves, for when we testeda group of English-speakingchildren using the picture book designedfor the Spanish-speakers, the data were identical to the original English findings (Waxman and Kosowski 1990). This difference, then, may indeed be due to cross-linguistic differences in the referential status of adjectives. In English, adjectives do not (as a rule) convey object reference. Although 2-year-old English-speakersare somewhat inclined to interpret adjectives as referring to objects and classesof objects, this is not the casefor 3- and 4-year-olds. In Spanish, where adjectivesdo, in fact must, convey nominal information , experience with the language may lead to a different outcome. Here, even 3- and 4-year-olds often interpret adjectivesas referring to objects and classesof objects. Thus, the role of adjectivesappearsdifferent in Spanishthan in French or in English. And it appears to differ in a predictable way, given the grammar of the adult languages. There are severalpossibleexplanations for this difference. First, it is possiblethat in Spanish, the grammatical distinction betweennouns and adjectivesdevelopsover a more protracted period. It is also possible that the grammatical distinction between these linguistic forms is made early, but that the appreciation of specificlinkages betweenlinguistic form and meaning develops over a more protracted period in Spanish. Additional research is to examinethesepossibleexplanations. .currently underway Let us now integrate thesefindings from the French- and Spanish-speaking children with those from children, toddlers, and infants learning English. The results of these complementary lines of research converge to provide initial support for the hypothesis that from the earliest stagesof lexical acquisition,

S.R. Waxman / Linguistic and conceptualorganization

249

infants expect that words (independent of their grammatical form ) will refer to objects and categories of objects. Later , this general linkage gives way to more specific pairings between particular grammatical forms and particular types of meaning. The affinity between count nouns and object categories emergesearly and is evident in all three languageswe have examined to date. In contrast, the more specific linkages for adjectives emerge later and may vary, depending upon the language being acquired. This account of the child ' s emerging appreciation of linkages between linguistic and conceptual organization is consistent with other major milestones in lexical acquisition. It gains further plausibility by virtue of the fact that it is also consistentwith cross-linguistic evidenceconcerning the linguistic categoriesnoun and predicate (including, e.g., adjectives, verbs) .

6. Cross- linguistic analyses of nouns and memben , of the predicate system

The cross-linguistic consistency of the gra~ matical category noun and the variability of the category adjective, which has been documented by linguists (c.f., Dixon 1982), is quite relevant to issues of acquisition. The syntactic category noun has a stable and uniform function across human languages. Count nouns refer primarily to objects and classesof objects. Furthermore, unlike words from other form classes, nouns supply principles of individuation and identity for their referents. (See Macnamara, 1986, for thorough discussionsof these principles; see Hall 1992, Hall and Waxman, 1993, for evidencethat preschool children, like adults, expect that count nouns supply these principles.) Additionally , early lexical acquisition consists predominantly of nouns (Gentner 1982, Nelson 1973) . In contrast to the cross-linguistic stability of the classnoun, membersof the predicate system (e.g., adjectives, verbs) have a more fluid status. There is considerablecross-linguistic variation as to what information is conveyed as part of one predicate and what is conveyedas part of another (Gentner 1982, Talmy 1985). Furthermore, there is cross-linguistic variability in the evolution of particular predicates. Dixon ' s ( 1982: 1- 3) discussion of the distinction betweennouns and adjectivesis quite revealing. ' It is an empirical fact that there is always a major class that is aptly termed Noun ; there is never any doubt as to the applicability of this traditional label, and never any question as to which class should be called Noun ... However, not all languageshave the major word class Adjective. Either they have no Adjective class at all , or else there is a small non-productive minor class that can be called Adjective. In either of thesecasesit is interesting to ask how the

250

S.R. Waxman I Linguistic and conceptualorganization

all adjectival ) ... express language getsalongwithout a full Adjectiveclass... Some(languages conceptsthroughintransitiveverbs... othersexpresssomethroughnounsand somethrough .' verbs... and othersinvokefurther means

Thus, the syntactic category adjectivediffers widely acrosslanguages. Some languages(like English and the Australian language Dyirbal ) have extensive and elaborate adjective systems; others (like Igbo and the Bantu languages) have very few adjectives (Dixon 1982) . Further , adjectives and other predicates appear to be acquired later than nouns. Moreover, there is question as to whether there is anything analogous to the naming explosion for the acquisition of predicates (Gopnik 1988) . Finally , members of the predicate systemare both semantically and syntactically dependent upon nouns.

of an appreciationof linkagesbetweenlinguistic and 7. The development rel _tioM conceptual Taken together, the cross-linguistic analyses of the noun and predicate systems converge with the developmental data to suggest the following account of the development of an appreciation of linkages between linguistic and conceptual relations. ' Early in infancy, infants visual attention is augmented by what appear to be very general sensory and/ or perceptual factors, rather than by specifically linguistic ones. Throughout infancy, the distinctive intonational contours characteristic of mothereseare especially effective in arousing and sustaining ' infants' attention (Fernald 1992) . Moreover, in the first six months, infants visual attention is also heightenedwhen objects are presentedin conjunction with moderate auditory stimulation (Kaplan et al. 1991, Mendelson and Haith 1976, Paden, 1975, Self 1975) . Initially , then, general auditory factors ' (rather than specifically linguistic ones) appear to intensify infants general visual interest (rather than their interest in objects or categoriesof objects) . In the latter half of their first year, a more specific pairing becomesevident as infants begin to single out words from other, more general sources of auditory input . By 9- to 12-months of age, infants focus more on objects and categories of objects in the presenceof novel words than in their absence (Baldwin and Markman 1989, Echois 1992, Waxman and Helm 1991, Mar kow and Waxman 1992, 1993) . Indeed, we find evidence that 9 month old infants establish object categoriesmore readily when the objects are accompa' nied by a label (e.g., ' a bird ) than when they are accompaniedby a sine wave

S. R. Waxman / Linguistic and conceptualorganization

tone (Waxman and Balaban 1992). Thus, by 9 months of age, labels facilitate categorization of objects. Moreover, this labeling effect appears to be tied to language, rather than to auditory stimulation , in general. (But see Roberts and Jacob 1991, for a different view.) However, at this point in development, the data do not support the claim that infants make systematic distinctions among words from various form classes. The data from our laboratory reveal that at 12 months, infants tend to interpret most words, independent of their syntactic status, as referring to objects or categoriesof objects. Therefore, prior to the onset of the naming explosion, there appears to be a general (and possibly universal) linkage betweenwords (not specifically count nouns) and object categories. As the naming explosion draws to a close, and as infants begin to distinguish among the linguistic form classes(e.g., count nouns, mass nouns, adjectives, verbs) in their own language production and comprehension ( Bloom 1990, Gordon 1985, McPherson 1991, Prasada 1993, Valian 1986), infants probably begin to consider syntactic form class as relevant to determining a novel word ' s meaning. By two to three years of age, children begin to reveal an appreciation of specific linkages between particular linguistic forms and particular types of meaning. For example, English-speakingchildren expect that object categorieswill be marked linguistically by count nouns (Brown 1957, Markman and Wachtel 1988, Wa~man 1990, Waxman and Kosowski 1990, Taylor and Gelman 1989, Waxman and Senghas1991), that substanceswill be marked by mass nouns (Dickins.on 1988, Sola et ale 1991), that individuals will be marked by proper nouns (Katz et ale 1974, Gelman and Taylor 1984, Hall 1992), and that various properties (e.g., size, color , temperament) will be marked by modifiers (Hall et ale 1993, Markman and Wachtel 1988, Waxman 1990, Taylor and Gelman 1988). I t is interesting to note that even at this point , when children are clearly capable of using syntactic information as a cue to meaning, they do not do so invariably . Instead, the tendency to use syntactic information is modified considerably by the child ' s existing lexical and conceptual knowledge (Au 1990, Banigan and Mervis 1988, Callanan 1985, Chi 1983, Mervis 1984, Mervis and Mervis 1988, Waxman et. ale 1991; Hall et ale 1993) . Children' s interpretation of a novel word depends, at least in part, upon whether or not they already have an existing label for the referent object. If the object is familiar (that is, if children have already acquired a count noun label for the object), then they use syntactic information as a guide in interpreting the meaning of subsequentwords applied to that object. For example, if a child is

S.R. Waxman / linguistic and conceptualorganization

taught a new noun for a familiar object (e.g., a dog), the child exhibits a strong tendencyto interpret the word as referring to an object category that is subordinate to (e.g., collie), superordinate to (e.g., mammal), or overlapping with (e.g., household pet) the familiar basic level category (Taylor and Gelman 1989, Waxman and Senghas 1992) . For a new adjective, children tend to interpret the word as referring to a salient property , substanceor part of the object (Hall et al. 1993, Markman and Wachtel 1988) . However, if the object is unfamiliar (that is, if children have not yet acquired a count noun for the object), they tend to rely upon an earlier pattern of behavior; they tend to interpret any word applied to that object (be it a count noun, proper noun, or adjective), as referring to an object category, typically at the basic level (Hall 1992; Hall et al. 1993, Markman and Wachtel 1988) . Thus, children are attentive to syntactic form in ascribing meaning only after a count noun has beenassignedto that object and to other membersor its kind .

8. Summaryandconclusion In summary, an appreciation of the linkage between nouns and object categories is likely to be a universal phenomenon which guides human development from the very onset of lexical acquisi.tion . In contrast, a distinct role for other form classes(such as adjectives) appears to emergelater, may rely upon an existing base of linguistic and conceptual knowledge, and may vary according to the specificsof the languagebeing acquired. What do theseemerging linkages betweenlinguistic and conceptual organization mean for the developing child ? In the first few years of life , children encounter a virtually continuous stream of new sounds, new objects, and new events. The linkages described in this article help infants to organize these encounters rapidly into efficient and coherent systems. When ascribing meaning to a novel word , infants and children do not sample randomly among all the possible relations and meanings that might logically be considered. Instead, in the context of word learning, they pay special attention to particular types of meanings. Although the linkages may initially be are acutely sensitiveto linguistic quite general, by 2 to 3 years of age, children ' form and use it to arrive at a novel word s meaning. The existing evidencesuggeststhat a nascentlinkage is in place at the onset of lexical acquisition, that it may serve as a general guide to lexical acquisition , and that it will becomeincreasingly specific over the course of develop

andconceptual S.R. Waxman I Linguistic organization

253

ment. Of course, linkages like the ones described here cannot tell the entire developmental story, for children do not learn meaning on the basis of syntactic context alone. Additional researchwith preverbal infants and with children learning diverse languages will further clarify when these various linkages betweenlinguistic and conceptual developmentemerge, how they are modified by linguistic input , and how they are modulated within the context of the child ' s existing fund of knowledge.

References . New York : Norton. Anglin, J.M., 1977. Word, object, and conceptualdevelopment Au, T.K., 1990.Children's useof informationin word learning. Journalof Child Language17, 393- 416. Baldwin, D.A. and E.M. Markman, 1989. Establishingword-object relations: A first step. Child Development60, 381- 398. ' , R.L. and C.B. Mervis, 1988. Role of adult input in young childrens category Banigan evolution. II : An experimentalstudy. Journalof Child LanguageIS, 493 504. Bauer, P.J. and J. M. Mandler, 1989.Taxonomiesand triads: Conceptualorganizationin oneand two-year-olds. CognitivePsychology4, 100- 110. : Comprehensionand production. Journal of Benedict , H., 1979. Early lexical development Child Language6, 183- 200. Berlin, B.. .D. Breedloveand P. Raven, 1973. Generalprinciplesof classificationand nomenclature in folk biology. AmericanAnthropologist75, 214- 242. . Journalof Child Language17, 343Bloom, P., 1990. Syntacticdistinctionsin child language 355. . A descriptivetaxonomyof psychological Bornstein . M.H.. 1984 categoriesusedby infants. In: C. Sophian(ed.), Originsof cognitiveskills, 303- 338. Hillsdale, NJ: Erlbaum. . Linguisticdeterminismand the part of speech . Journalof Abnormaland Social Brown. R., 1957 55, 1- 5. Psychology . Wordsand things. Glencoe . Brown. R., 1958 , IL : The FreePress . A studyof thinking, New York: Wiley. Bruner, J.S., J.J. Goodnowand G.A. Austin, 1956 . How parentslabelobjectsfor youngchildren: The role of input in the Callanan , M.A., 1985 . Child Development 56, 508- 523. acquisitionof categoryhierarchies . as learner . In: M. Halle S. 1978 The child word , , , J. Bresnan , G.A. Miller (eds.), Carey . , MA : MIT Press Linguistictheoryand psychological reality. Cambridge . On somerelationsbetweenthedescriptionand the explanationof developmental Carey, S., 1990 : Interdisciplinary . In: G. Butterworth, P. Bryant (eds.), Causesof development change . , 135- 157. New York: HarvesterWheatsheaf perspectives . Acquiring a singlenew word. Papersand Reportson Child Carey, S. and E. Bartlett, 1978 of Linguistics , StanfordUniversity) IS, 17- 29. (Department LanguageDevelopment -derivedcategorizationin youngchildren. In: D.R. Rogers . Knowledge Chi, M.T.H.. 1983 . J.A. . Sloboda(eds.), The acquisitionof symbolicskill, 327- 332. New York: PlenumPress . . Knowledgeof language : Its nature, origin, and use. Westport, CT: Praeger , N., 1986 Chomsky

254

S. R. Waxman I Linguistic and conceptualorK,anizatioll

. In: Clark, E.V., 1987 . The principle of contrast: A constraint on languageacquisition B. McWhinney(ed.), Mechanisms of languageacquisition:The 20thannualCarnegieSymposium on Cognition, 1- 33. Hillsdale, NJ: Erlbaum. . Learningnamesfor materials : Factorsconstrainingand limiting hypotheses Dickinson, D.K., 1988 . CognitiveDevelopment 3, 15- 35. aboutword meaning . Wherehaveall the adjectivesgone? Berlin: Mouton. Dixon, R.M.W., 1982 . . Early lexicaldevelopment . Cambridge : CambridgeUniversityPress Dromi, E., 1987 Echois, C., 1992 . Developmental changesin attentionto labeledeventsduring the transitionto on InfancyStudies . Paperpresented at the InternationalConference , Miami, FL. language : An . Humanmaternalvocalisations to infantsas biologicallyrelevantsignals Fernald, A., 1992 . In: J.H. Barkow, L. Cosmides , J. Tooby (eds.), The adaptedmind: evolutionaryperspective and the generation of culture, 391- 428. NewYork: OxfordUniversity Evolutionarypsychology Press . . Implicit contrastin adjectivesvs. nouns: Implications Gelman, S.A. and E.M. Markman, 1985 . Journalof Child Language12, 125- 143. for word-learningin preschoolers . How two-year-old childreninterpretproperand common Gelman, S.A. and M. Taylor, 1984 . 55, 1535 - 1540 nounsfor unfamiliarobjects.Child Development . Why nounsare learnedbeforeverbs: Linguistic relativity versusnatural Gentner, D., 1985 : Language , thought, andculture, 301development partitioning. In: S. Kuczaj(ed.), Language 334. Hillsdale, NJ: Erlbaum. . Wherelearningbegins: Initial Gleitman, L.R., H. Gleitman, B. Landauand E. Wanner, 1987 for languagelearning. In : F. Newmeyer(ed.), The CambridgeLinguistic representations . , Vol. III , 150-193. New York: CambridgeUniversityPress Survey . Early lexical acquisition: Rate, content, and the Goldfield, B.A. and J.S. Remick, 1990 vocabularyspurt. Journalof Child Language17, 171- 183. Golinkoff, R., K. Hirsh-Pasek , L.M. Baileyand R.N. Wenger , underreview.Youngchildrenand adultsuselexicalprinciplesto learnnewnouns. . . Fact, fiction, and forecast . Cambridge Goodman , MA : HarvardUniversityPress , N., 1983 . Three types of early word: The emergence of social words, namesand Gopnik, A., 1988 -relationalwordsin the one-word stageand their relationto cognitivedevelopment . cognitive First Language8, 49- 70. ? A cross. Do linguisticdifferences lead to cognitivedifferences Gopnik, A. and S. Choi, 1990 . First Language10, 199215. and cognitivedevelopment linguisticstudyof semantics of categorization in the secondyearand its Gopnik, A. and A. Meltzoff, 1987.The development - 1531 . . Child Development 58, 1523 relationto othercognitiveand linguisticdevelopments : The caseof the count/mass Gordon, P., 1985. Evaluatingthe semanticcategorieshypothesis distinction. Cognition20, 209- 242. Grimshaw, J., 1981. Form, function, and the languageacquisitiondevice. In : C. L. Baker, J. McCarthy (eds.), The logical problem of languageacquisition. Cambridge , MA : MIT . Press . Acquiring proper namesfor familiar and unfamiliaranimateobjects: TwoHall, D.G., 1991 ' - 1154 . . Child Development 62(5), 1142 yearolds word-learningblases . Basiclevelkindsand individuation. Unpublishedmanuscript Hall, D.G., 1992 , MRC Cognitive Unit, London. Development . In: J. Macnamara Hall, D.G., in press. How childrenlearncommonnounsand propernames , . G. Reyes(eds.), The logicalfoundationsof cognition. Oxford: Oxford UniversityPress

S.R. WaxmanI Linguisticandconceptual organization

2SS

Hall, D.G. andSiR. Waxman, 1993 . Assumptions aboutwordmeaning:Individuationandbasic levelkinds. Child Development 64, 1550 - 1570. Hall, D.G., SiR. Waxmanand W.R. Hurwitz, 1993 . The development of sensitivityto syntactic cues: Evidencefrom preschoolers learningpropertytermsfor familiar and unfAmiliArobjects. - 1664 Child Development 64, 1651 . Huttenlocher . The semanticorganizationof somesimplenounsand verbs. , J. and F. Lui, 1979 Journalof VerbalLearningand VerbalBehavior18, 141- 162. Inhelder, B. and J. Piaget, 1964 . The earlygrowthof logic in the child. New York: Newton. and L. Jenkins . Cross -modal facilitation of infant Kaplan, T., K. Fox, D. Scheuneman , 1991 visualfixation: Temporaland intensityeffects . Infant Behaviorand Development14, 83- 109. Katz, N., E. Bakerand J. Macnamara . What's in a name? A studyof how childrenlearn , 1974 commonand propernames . Child Development 45, 469- 473. Landau, B. and L.R. Gleitman, 1985 . Languageand experience : Evidencefrom the blind child. : HarvardUniversityPress . Cambridge Landau, B., L.B. Smithand S.S. Jones . The importanceof shapein earlylexicallearning. , 1988 3, 299- 321. CognitiveDevelopment Leslie, A.M. and S. Keeble, 1987 . Do six-month old infantsperceivecausality? Cognition25, 265-288. Macnamara . A border dispute: The placeof logic in psychology . Cambridge : MIT , J., 1986 . Press Mandler, J.M., 1988 . How to build a baby: On thedevelopment of an accessible representational . CognitiveDevelopment 3, 113- 136. system Mandler . The cradle of categorization : Is the basiclevel basic? , J.M. and P.J. Bauer, 1988 . 3, 237- 264. CognitiveDevelopment Mandler, J.M., P.J. Bauerand L. McDonough . Separatingthe sheepfrom the goats: , 1991 . CognitivePsychology23(2), 263- 299. Differentiatingglobalcategories Mandler, J.MR . Fivushand J.S. Reznick . The developmentof contextualcategories . , 1987 2, 339- 354. CognitiveDevelopment Markman, E. M., 1984 . The acquisitionand hierarchicalorganizationof categories by children. In: C. Sophian(ed.), Originsof cognitiveskills, 371- 406. Hillsdale, NJ: Erlbaum. Markman, EllenM., 1989 . Categorization and namingin children. Cambridge . , MA : MIT Press Markman, E. and J.E. Hutchinson . Children's sensitivityto constraintson word meaning: , 1984 Taxonomicvs. thematicrelations.CognitivePsychology16, 1- 27. Markman, E.M. and GiF. Wachtel, 1988 . Children's useof mutualexclusivityto constrainthe of words . 20, 121- 157. meanings CognitivePsychology Markow, D.B. and SiR. Waxman . The inftuenceof labelson 12-month-olds' category , 1992 formation. Paperpresented at the Eighth InternationalConference on Infant Studies , Miami, FL. Markow, D.B. and SiR. Waxman, 1993 . The impactof introducingnovel noUDS versusnovel ' . Paoerpresented at the Meetingof the Societyfor adjectivesin 12-month-olds categorization Research in Child Development , New Orleans , LA. McPherson . A little goesa long way: Evidencefor a perceptualbasisof learningthe , L., 1991 nouncategories COUNT and MASS. Journalof Child Language18, 315- 338. McShane . Learningto talk. Cambridge : CambridgeUniversityPress . , J., 1980 Mendelson . The relationbetweenauditionand visionin the human , M.J. and M.M. Haith, 1976 newborn.Monographsof the Societyfor Research in Child Development 41(4), 1- 72.

256

S. R. Waxman I Linguistic and conceptualorganization

's word biasin children . The mutualexclusivity Merriman , 1989 , W.E. and L.L. Bowman in ChildDevelopment of the Societyfor Research . Monographs , SerialNo. 220, learning 54(3/4), 1- 23. of motherandchild. In: C. : Thecontributions . Earlylexicaldevelopment Mervis , C.B., 1984 . skills, 339-370.Hillsdale , NJ: Erlbaum (ed.), Originsof cognitive Sophian . In: U. Neisser andearlylexicaldevelopment . Child-basiccategories Mervis (ed.), , C.B., 1987 factors in and intellectual : and , categorization Ecological development Concepts conceptual . Press : Cambridge 201-233.Cambridge University -, basic -, andsuperordina of subordinate . Orderof acquisition Mervis , 1982 , C.B. andM.A. Crisafi 53, 258-266. . ChildDevelopment -levelcategories 's : evolution . Roleof adultinputin youngchildren Mervis , 1988 category , C.G. andC.A. Mervis 272 . 15 2 257 I. An observational study.Journalof ChildLanguage( ), 17, . Journalof ChildLanguage . Childrenusesyntaxto learnverbmeanings , L., 1990 Naigles - 374. 357 of the Societyfor andstrategyin learningto talk. Monographs . Structure Nelson , K., 1973 . in ChildDevelopment Research , 38(1/2), 1- 136 , SerialNo. 149 . In: T. Seller . Theconceptual basisfor language Nelson , W. Wannanmacher (eds), , K., 1983 - 190 . . Berlin: Springer of wordmeaning andthedevelopment , 173 development Concept 3, 221-246. ? Cognitive . Constraints onwordlearning Nelson Development , K., 1988 -gestural levelsin a visual of category . Linguistic , 1978 , E.L. andU. Bellugi expression Newport eds . E. Rosch B.B. . In: : A floweris a floweris a flower , Lloyd( ), Cognitionand language Erlbaum . . Hillsdale NJ: 49 -71 , , catagorization ' : Habituation and . Infants Oakes , 1991 objectexamining , L.M, K.L. MadoleandLiB. Cohen . Cognitive categorization Development 6(4) 377- 392. variations of auditory stimulation(music) and interspersed effects of L. 1975 . The Paden , , stimulusprocedureson visualattendingbehaviorin infants. Monographsof the Societyfor in Child Development Research , 39( I 58), 29- 41. : Evidencefrom the acquisitionof . On the autonomyof languageand gesture Petitto, L.A., 1987 . Cognition27( 1), I - 52. personalpronounsin AmericanSignLanguage . Cambridge . Languageleamabilityand languagedevelopment Pinker, S., 1984 , MA : Harvard Press . University ' . Prasada , S., 1993. Young childrens linguisticand non-linguisticknowledgeof solid substances 8, 83- 104. CognitiveDevelopment . . Word and object. Cambridge , MA : MIT Press Quine, W.V., 1960 on nonlinguisticcategorization . Linguisticvs. attentionalinfluences Roberts , K. and M. Jacob, 1991 6(4), 355- 375. in lS-month-old infants. CognitiveDevelopment -Braem, 1976 . Basicobjectsin Rosch, E., C.B. Mervis, W.D. Gray, D.M. Johnsonand P. Boyes . CognitivePsychology8, 382- 439. naturalcategories in 1- to 2-year-olds. Developmental . Categorization Ross, G.S., 1980 Psychology16, 391- 396. . Componentsof attention during infants' manipulativeexploration. Child Ruff, H.A., 1986 57( 1), 105- 114. Development ' stimulation. Self, P.A., 1975. Control of infants visualattendingby auditory and interspersed 28. 158 16 in Child Research for of the 39 ( ), Development Society Monographs . Hillsdale, NJ: Erlbaum. Siobin, D.I., 1985.The cross1inguistic studyof language ' . Ontologicalcategories , 1991 guideyoungchildrens inductions Sola, N., S. Careyand E. Spelke tenDS. Cognition38, 179- 211. aboutword meaning:Objecttermsand substance

S. R. Waxman I Linguistic and conceptualorganization

. Lexicalizationpatterns: Semanticstructurein lexicalforms. In: T. Shopen(ed.), Talmy, L., 1985 : CambridgeUniversity , Vol. 3: 57- 149. Cambridge Languagetypologyand syntacticdescription Press . . Adjectivesand nouns: Children's strategies for learningnew Taylor, M. and S.A. Gelman, 1988 words. Child Development 59, 411- 419. . Incorporatingnew words into the lexicon: Preliminary Taylor, M. and S.A. Gelman, 1989 evidencefor languagehierarchies in two-year-old children. Child Development 60, 625- 636. . The role of joint attention in early languagedevelopment Tomasello . Language , M., 1988 II , 69- 88. Sciences Valian, V.V., 1986 . Syntacticcategoriesin the speechof young children. Developmental 22, 562- 579. Psychology . Thoughtand language . Cambridge : MIT Press . Vygotsky, L., 1962 . Linguisticblasesand the establishment Waxman of conceptualhierarchies : Evidence , S.R., 1990 from preschoolchildren. CognitiveDevelopment 5, 123- 150. . Convergences Waxman betweensemanticand conceptualorganizationin the , S.R., 1991 on languageand thought, preschoolyears. In: S.A. Gelman, J.P. Byrnes(eds.), Perspectives 107- 145. Cambridge : CambridgeUniversityPress . ' Waxman , S.R. and M.T. Balaban , 1992.The influenceof wordsvs. toneson infants categorization . Paperpresented at the EighthInternationalConference on Infant Studies Miami , , FL. ' useof Waxman . Preschoolers relationsin classification , S.R. and R. Gelman, 1986 superordinate . CognitiveDevelopmentI , 139- 156. and language Waxman . The developmentof a linkagebetweencount nounsand , S.R. and DiG. Hall, 1993 : Evidencefrom 15 to 21 month old infants. Child Development64, 1224 objectcategories 1241 . Waxman . Nounshighlightcategoryrelationsin 13-month-olds. Paper , S.R., and L. Helm, 1991 , Seattle , presentedat the annualmeetingof the Societyfor Researchin Child Development WA. Waxman . Nouns mark categoryrelations: Toddlers' and , S.R. and TiD . Kosowski, 1990 ' word- 1473 . Child Development 61, 1461 . preschoolers learningblases Waxman . Relationsamongword meaningsin early lexicaldevelopment , S.R. and A. Senghas , 1992 . Developmental 28(5), 862- 873. Psychology Waxman . Establishingnewsubcategories son, 1991 : The role , S.REiF . Shipleyand BS hepper of categorylabelsand existingknowledge . Child Development 62. 127- 138. Waxman . Word-learningblasesin , S.R., A. Senghas , D. Rossand L. Benveniste , in preparation -speakingchildren. French- and Spanish . Infant perceptionof correlationsamongattributes.Child Younger, B.A. and LB . Cohen, 1983 54, 858- 867. Development ' . Developmental Younger, B.A. and LB . Cohen1986 changesin infants perceptionof correlations amongattributes.Child Development57(3), 804- 815.

) 259- 296. North-Holland Lingua92 ( 1994

259

Where's what and what's where: The languageof objectsin space* Barbara Landau , Universityof California, Irvine, CA 92717 , USA Department of CognitiveScience Many traditional views of languagelearning emphasizethe importanceof non-linguistic ' knowledgein constrainingchildrens interpretationsof the utterancesthey hear as they learn form- meaningcorrespondences . In the presentpaper, I proposethat the structureof the human spatial-cognitivesystemmay imposemassivelimitations on how the languageof objectsand , the separationand placescan be representedduring first languagelearning. Specifically structuresof the 'what' and ' where' systems(as documentedby neuropsychologists respective ) show striking correlationwith the propertiesof geometricrepresentations underlyingobject namesand placeexpressions (as proposedby Landauand Jackendoff1993 ). Spatialrepresentations underlyingobjectnamesseemto be linked to detailedobjectshape(along with possible relevantshapetransformations ), while the representations underlyingplace-functions(spatial in seem to be linked to prepositions English) objectdescriptions preservingprincipalaxes,but no detailedshape . Evidenceis reviewedthat thesedistinctrepresentations are reflectedin children's . Additional , production, and learningof namesfor objectsand places early comprehension evidencefrom cross-linguisticstudiesof spatialtermsappearsto be consistentwith the account remain. As a whole, the linguisticand psychological evidencesupport given, althoughchallenges the notion that non-linguisticrepresentational systemsmay playa part in shapingsomeof the . characterof languages

I . Objects , places, and language leaming For anyone who has ever watched as an infant begins to speak, it is obvious that the observational context for learning serves as a critical support . Common sense suggests that language learning proceeds as rapidly as it does in part because children can map the utterances they hear onto . . This researchwas supportedby Socialand BehavioralSciences ResearchGrant # 12-214 from the Marchof DimesBirth DefectsFoundation - OI, and the , NIH Grant # I -ROI-HD2867S Institute for Researchin CognitiveScience , whereLandauwas a Universityof Pennsylvania - 93. Thanksgo to Lila and HenryGleitmanfor numerouscomments visitingscientistduring 1992 on a previousversionof this paperand for suggestingits title; and to D.Stecker , E.Lynch, D.Beckand J.Chosakfor assistance in the work describedherein. 0024 - 3841/94/$07.00 ~ 1994- Elsevier. ScienceB.V. All rights reserved - 3841( 93) EO043- 7 SSD10024

' ' B. Landau I Wheres what and what s where

what they perceive and know. Of chief importance is their perception and knowledge of objects, the motions they undergo, and the locations they occupy. But how does perceiving or knowing about objects help in learning language? As Quine ( 1960) argued, the novice analyst hearing an utterance might be free to conjecture any of an infinite number of hypothesesabout its meaning, each compatible with what he observes. Such rampant indetenninacy would , of course, prevent learners from converging on the correct meaning of a word. However, human infants do not seemto find themselves mired in such an impossible learning situation ; quite the contrary, they manage to map many words onto coherent meanings in a relatively short period of time. Accounting for this phenomenon in tenDS of constraints on learning - both linguistic and non-linguistic - is currently a major challenge for both linguists and psychologists. One sourCeof constraints is the human spatial- cognitive system. Recent non-linguistic investigations have shown that the spatial systemcontains two ' separate components: one dedicated to representing objects (the what sys' ' tem' ) and the other to representinglocations (the where system ; Ungerleider and Mishkin 1982) . The two components exhibit quite different properties, perhaps not surprising if one evolved to represent objects, the other, places. What is intriguing about each component, however, is that broad properties of each appear to converge with the differences in the ways that languages expressobjects and locations (Landau and Jackendoff 1993) . Theseconvergencescould impose significant constraints on learning. For if the design of the ' what' and ' where' system engagesonly certain geometric representations, then only thesewill be available for mapping onto language. Learners possessingsuch a representational system would not be subject to ' Quine s radical indeterminacy; rather, they would approach language learning equipped with strong blases to consider only certain properties of the world as ' relevant' to naming objects vs. places. In arguing for such constraints, severallines of evidenceare required. First , there must be differences in the way that languagesuniversally encode the two onto logical categories of object and place. Second, these linguistic universals should show properties that are similar to those found in nonlinguistic ' ' ' ' representations of what and where . Finally , there must be evidence that children learning names for objects and places generalizeon the basis of different geometric properties in the two cases- that , essentially, the problem of learning names for objects and places is bifurcated, with each domain drawing on quite different kinds of representations.

's whatandwhat's where B. Landau I Where

261

In the sectionsthat follow , I first provide evidencefor such a bifurcation in young children. This phenomenonwill set the stagefor asking in detail about ' ' the nature of ' what and ' where representationsin languagesand language learning. I then explore in detail the spatial representationsunderlying object naming and those underlying the language of places. Finally , I discuss questions about the possible contributions of spatial representation vs. language itself to thesedifferent representations. Before turning to the findings, however, the reader might wonder how learners know in the first place that a particular part of an utterance expresses the name of an object vs. a place. Our researchis neutral on how learners first come to know the mappingbetweenforms and these broad ontological distinctions. In fact, our empirical research begins at the point (age 21f2years) when young children can already use syntactic context to create a coarsedivision of the hypothesisspace. By this time, they can use the syntactic and morphological properties of count and mass nouns, adjectives, and verbs to disambiguate among various readings in English. For example, they can use count/ mass syntax to distinguish between object and substance readings (Brown 1957; P.Bloom, this volume); noun/adjective syntax to distinguish betweenobject and property readings (Landau et al. 1992a); and number of verb argumentsto disambiguatebetweenagentive vs. non-agentive readings (Gleitman 1990, Fisher et al. this volume) . In our research, we use children' s knowledge of syntax and morphology as a wedge in asking about the representationaldifferencesbetween object and place. I assume that such broad mappings between form and meaning are achievedquite early in languagelearning. For example, prelinguistic children can represent major onto logical categories, such as OBJECT, PLACE , and ACTION : Infants know that objects continue to exist over spatial and temporal displacements (see Carey, this volume; Spelke 1990) and that ' objects occupy spatial locations which are constant over one s own movements (Landau and Spelke 1988, McKenzie et al. 1984, Rieser 1979) . Objects, ' static locations, and trajectories are mapped to languagein the child s earliest vocabulary, including words like in, on, up and down (Brown 1973, L . Bloom 1973). Children also can representlexical and phrasal categoriessuch as N , NP , P, PP, V , VP. These categories might be identified in the speech stream by correspondencesbetween certain properties of the wave form ' and major ' lexical, phrasal, or clausal units (i.e. by prosodic bootstrapping , see Gleit man et al. 1988, Kelly , this volume) . Another possibility is that nouns are discovered first , when the child hears a single word in isolation (usually an

262

's whatandwhat's where B. Landau / Where

object name) together with the sight of an object (Gillette and Gleitman, forthcoming) . Given only a small group of nouns, the remaining categoriesin the syntactic configuration may be deduciblefrom internal linguistic principles (Grimshaw, this volume). In either case, it seemsreasonable to assumethat children also approach language learning with certain expectations about the canonical mappings betweenthe major onto logical categoriesand the syntactic categories(Grim shaw 1981) . For example, OBJECT is mapped canonically onto N , and PLACE - more properly , spatial-locative-functions - onto Preposition, PP, or other argument-taking categories.! Once the learner has come to know how objects and placesare syntactically expressed in his or her language (in English, by count nouns and PPs respectively), he or shecan infer that a novel lexical item falls into the domain of either object or place. Now , from the different structural properties of the ' what' and ' where' system, the relevant geometric generalizations should automatically fall out. Given this view, the syntactic properties of a novel word can provide the learner with quite coarse semantic distinctions (e.g. the major onto logical ' ' ) . The remainder of the learning of a categoriesand broad semantic classes word must be discoveredfrom sourcesoutside of syntax2 (seeGrimshaw, this 1 The encodingof spatiallocation functionsvariesacrosslanguages : Englishencodesthem canonicallyasprepositionsand PPs, whileotherlanguages incorporatethe locationinto the verb .) In this paper, I focuson Englishprepositions , whichhavebeenanalyzedin detail ( Talmy1985 -Laird 1976 , Herskovits (Miller and Johnson by linguists, computerscientistsand psychologists in 1983 1983 . However the 1986 Jackendoff , , Talmy , ) principlesproposed this paperaremeantto of their be universallyrepresented in expressions of spatiallocation(cf. Talmy 1983 ), regardless . The strengthof analyzingEnglishspatialprepositionslies in the fact that syntacticrealization - unlike, e.g., manyspatialverbs- encodeonlyobjectlocationfunctions,henceare theseelements . clearexampleof how languages an especially representplaces 2 Thesesemantic -syntacticmappingswill be richerin somedomainsthan in others, and we can differences in theextentto whichlearnerswill beableto usesyntacticcontext expectcorresponding to learndetailsabout the meaningsof newwords. For example , in the caseof verbs, we might -relatedwords(Pinker 1989 , carvingout setsof meaning expectrelativelyrich correspondences to can be LevinandHovav1991 and such ) correspondences expected aid thelearningof newverbs , the , Fisheret aI., this volume.) In contrast , Fisheret al. 1991 (Landauand Gleitman 1985 count between interpretationof novelnounsshouldbehelpedrelativelylittle beyonddistinguishing . For example vs. massreadings , a learnerwould not be able to tell, from the syntacticcontext ' an ' individuated alone, muchmorethan that theword describes entity(e.g. a concreteobject, see . In English seems true for P. Bloom, this volume.) The samesparsemapping , spatialprepositions this 1983 . If the learner understands onto PPs Jackendoff ) mapping ( placesmap canonically a place,but it will only be through principle,shemightbeableto interpreta newPPasdescribing

B. Landau / Where's what and what's where

263

volume; Fisher et al., this volume; Pinker, this volume) . Our interest is in how the spatial-cognitive system might further pare down the hypothesis spacefollowing the broad division into object vs. place.

2. Namesfor object vs. place: Two different kinds of representation Given that words for objects and places are among the earliest acquired, one might wonder whether a general, all -purpose representationunderlies the earliest learning in both domains. Perhaps human learners are designed to detectjust the samesetsof properties whether attending to an object (and its name) or the object' s location (and its name) . This kind of representation would keep track of a compound - an object' s identity together with its location. Such a relatively unbiased device could definitely learn to represent objects and places separately (see Rueckl et al. 1988 for a demonstration using connectionist modelling) . The question, however, is whether this is the way that young learners do represent objects and their locations under different form-classinstructions, i.e., when they are being told the name of a novel object vs. a novel pla~ . In a seriesof experiments, we asked whether young children could draw on qualitatively different geometric representations acorss the two domains of object and place (Landau and Stecker 1990) . In order to induce these two kinds of representation, we showed 3-year-olds and adults a novel object ' ' being placed on top of a box (the standard location), and simultaneously presentedthem with a novel word occurring in one of two different syntactic contexts. A count noun context was meant to induce a representation of the object itself, while a preposition context was meant to induce a representation of the place where the object was located. In the count noun context, subjectsheard ' Seethis? This is a corp' . In the ' ' preposition context, subjects heard See this? This is acorp my box . The entire display then was set aside, and subjects saw each of three different shapedobjects - one identical to the standard plus two shapechanges- being placed in each of five different locations on and around a second box (see

264

B. Landau I Where's what and what's where

figure I for objects and standard location) . Each time subjects observed an ' ' object being placed on the secondbox, they were asked either Is this a corp? ' ' or Is this acorp your box? (matching their initial instructions) . We asked whether subjectswould attend to different properties of the array in the two contexts.

.r

uu Thisis a corp. This is acorpmy box.

No I

~ I

Yes

Yes

No

Yes

I

I

' ' Fig . 1. Objects and array used in spatial preposition experiment. In this experiment, a standard ' ' object was placed in a standard position (as shown) on the upper surfaceof a box. Subjectswere " told either " This is a corp" (count noun condition ) or "This is acorp my box ( preposition condition .) Then each of the objects was placed in each of several positions on and around the " " box, and subjects were asked either " Is this a corp?" or Is this acorp your box? (with the test showed that in the count noun instruction . The ) responses question matching the initial condition , subjects acceptedonly the standard shape and rejected shapesthat differed from the standard. In contrast, in the preposition condition , subjects accepted all shapes, apparently considering object shapeirrelevant to decisionsabout location.

' The results showed that both children and adults attended to the object s shapein the count noun condition , accepting the original shape but rejecting all shape changes. In the preposition condition , however, subjects ignored shape, accepting all three shapesequally. In complementary fashion, subjects ' ignored the object s position in the count noun context, but attended to it closely in the preposition condition . Children accepted all positions on the upper surface of the box, while adults were more conservative, primarily accepting the standard position .

' B. Landau I Wheres what and what's where

This pattern of results shows a rather extreme dissociation - presenceof detailed shape vs. no shape information at all - in 3-year-olds who are generalizing a novel preposition. But such an extreme dissociation is not a necessaryproperty of the geometriesof object vs. location tenDs. There are terms in English and other languageswhich preserverestricted elementsof an ' object s shape. For example, the terms acrossand along both require that the ' ' figure object be roughly linear , that is, that the object have a principal axis that can be located with respect to the axis of the reference object (Talmy 1983; seesection 4) . Therefore, in a subsequentexperiment, we asked whether children could represent objects in this intennediate fashion: as a geometric kind that does not represent exact shape, but preservesjust one critical ' component - the object s principal axis. We now used the same method but changed the objects' shapes and locations (see figure 2 for shapes and ' standard' location). The standard " object was now a 7 straight rod , and the test objects included a replica of the standard, a squiggly rod that was the same extent as the standard, and a 2" x 2" x I " block. The standardwas placedperpendicularto the box's main axis when we introduced it , and the test positions included this sameposition, one ' slightly to the left, one parallel to the box s main axis, and one diagonal to it . The overall results showed the same complementary pattern for attention to shapevs. position as we had found in the first experiment: Subjectsin the count noun context attended to shape but not position , while subjectsin the preposition context did the converse. However, it is the details of the shape results that are of particular interest, for we observed two different kinds of shaperepresentationsfor objects in the two syntactic contexts. In the count noun context, children and adults preservedthe preciseshape of the figure object when generalizing: They accepted the standard and rejected both the squiggle and the block. In the preposition condition , some 3-year-olds ignored shape altogether as they had in the previous experiment. But other 3-year-olds, and most 5-year-olds and adults acceptedthe standard and the squigglewhile rejecting the block. That is, they acceptedboth objects that had intersectedthe box, even though thesewere quite different in precise shapeand hencehad not both been a~ pted as instancesof the same named object category in the count noun condition . What thesechildren did was to treat both the standard and the squiggle alike as linear objects in the preposition condition , attending to the one property of shape- the extent of its main axis - that was apparently important in describing its location. In sum, we found that when young children were asked to generalize a name for a novel object, they engageda full description of the object' s shape.

' ' B. Landau / Wheres what and what s where

-

266

. Thisis a corp. This is acorpmy box.

Fig . 2. Objects and array used in ~ nd spatial preposition experiment. This experiment was ' ' identical to the first , except for the shapesof the standard and test objects, and the positions that were tested (see text.) As in the first experiment, subjects in the count noun condition ' ' attended to shape, aa: epting only the standard (the straight rod.) In the preposition condition , ' linear' two the objects objects whose principal axis could span the box and subjecg a<x:epted test the schematized condition in the that This the block. preposition subjects suggests rejected objects, focussingonly on the existenceof a principal axis of a pa~ cular extent. The results of the two experimentsshow that detailed object shapematters for the extension of object count nouns, but that schematicobject shapematters for the extension of spatial prepositions.

In contrast, when asked to generalizea new place term, they either ignored ' ' ' the object s shape altogether - representing it essentially as a blob - or did something in between, representing just one component of shape as if the ' ' target object were a linear thing . Why did young children generalizeso differently in the context of a count noun vs. a preposition? Clearly, we cannot appeal to an 1;inbiased Quinean ' ' similarity spacewhich dictateswhat will be salient in any scene, for thesewere identical in the two situations. The only differencebetweenthe conditions was the syntactic context in which the novel word was introduced. Children must have assumedthat the novel count noun referred to the object being placed on the box - in this case, an object of a specific shape (to be explored in section 3) . And they must have assumedthat the novel preposition referred to the place where the object was located. Syntactic context thus directed

' ' B. Landau I Wheres what and what s where

267

children' s attention to two different ontological categories- object and place from which followed quite different kinds of geometric representations. These findings raise many questions. As for objects: Do children always generalizenovel object nameson the basis of shape and if so, what does this ' suggestabout the role of object perception in constraining the child s initial hypothesesabout object names? As for places: Is the lack of shape information in locational terms a universal feature of languages and is failure to attend to shape for place terms a universal feature of language learners? Finally , do these differences between object and place representation reflect facts about the way languagesare organized? Or , as I ' ve already hinted, do they partly reflect the way the human spatial-cognitive system operates independently of language?

3. How named objects are represented Count nouns such as dog , apple , chair name categories of objects - sets of objects that cohere under some criteria that can be used to generalize the word to new instances. The nature of these criteria is as yet unclear , but often has been assumed to include some set of perceptual or conceptual features united by an underlying theory (Carey 1982, Arm strong et al . 1983, Murphy and Medin 1985) . Such theories would specify not only which object properties are critical for different named categories , but why the same property can assume different weighting for different categories - for example , why feathers are more critical to the description of birds than to the description of hats.3 3 Therecentemphasis on the importanceof suchunderlyingtheoriesin conceptacquisitionand -driven' word learningoftenhasbeenviewedasan alternativeto the moretraditional' perception . For example , it has been argued that young children use their knowledgeof approach ontological categoriesto constraindecisionsabout word meaning , therebyallowing them to ' ' overrideperceptualinformation(Kei11989 , Solaet al. 1991 ). This either- or approachto word , however , for althoughsuchknowledgecan be important in certain learningseemsmisguided kindsof decisionsaboutcategorymembership , perceptualinfonnationalsois likely to play an , Smith , Landauet aI. 1992 importantrole, especiallyamongyoungchildren(Landauet aI. 1988 1989 .) the samekinds of rich theoreticalsystemsas For example , youngchildrenmay not possess .) Careyhas arguedthat children's theoriesof ' living thing' are adults(Carey 1985 , Keil 1989 . Evenif shallowerand qualitativelydifferentfrom thosecontainedin maturebiologicaltheories childrendo havetheoriesof objectkind that enterinto their classification (henceobjectnaming), of objectsis boundto be a critical factor in initial objectsorts, as it is for perceptualappearance adults in many contexts(Bosterand Johnson1989 , , Murphy and Wright 1984 , Landau 1982

268

B. Landau I Where's what and what's where

Common sensesuggeststhat object shape might be particularly critical in the extensionof object names. For adults who have full -blown conceptsand a mature lexicon, the representationsof objects named by count nouns such as chair, apple, and dog must in part be linked to the shapesof these objects (Jackendoff 1987, Jackendoff and Landau 1991) . Object shape remains constant over space and time, is an excellent predictor of object kind , and shape perception is undoubtedly a part of human native endowment. Shape has been shown to be a critical property of object representation at the basic level (Rosch et ale 1976); it is the principal basis for children' s spontaneous over-generalizations (Clark 1973); and it may be a privileged property in extending object namesto depictions in representationalart. For example, the well-known sculptures by Claes von Oldenburg include ' The Clothespin' (a 45-foot metal object) and ' Good Humours' (floppy, furry objects); in both cases, similarity to the real thing is solely in their shape.4 In these cases, ' similarity of shape licensesthe use of the real object s name, despite the fact that the representational object is not really an X (seeJackendoff, 1992, for related discussion) . Yet shapeis definitely a perceptual property . If object names are extended on the basis of perceptual properties only as weighted by theories, then what is the status of shape? Is shape somehow ' special' as a brute-force preferred property? Or are there always effectsof higher-level knowledgein determining whether and to what extent shape is important ? In our studies of object naming, we have discovered that the answers to thesequestionsare complex. As the following evidenceshows, shapeis indeed ' ' special as a property often critical to the extension of object names among young children. Considering the reasonsfor this suggeststhat constant object shapemay be intimately related to our notions of object kind. However, the results also show that leamer' s use of object shaperecruits a systemof shape perception that is itself quite complex and subtle. Moreover, the system can

L .B.Smith 1989, Smith and Medin 1982.) Most important , the existence of theories does not logically preclude the use of perceptual information in word learning. Presumably, our theories have evolved in part to make senseof what we observe. ' 4 Jackendoff 1983, 1992 has ( ) suggestedthat an object s name can be extendedto many possible referentsX , including ' depiction of X ' . This would account for the naturalnessof referring to the ' ' ' sculpture as clothespin even though we all know it isn t really one. Jackendoff suggeststhat the distinction betweenreal referentsand depictions may be critical for binding theory. I thank Paul Bloom for pointing out the relevanceof Jackendoff' s treatment, and Robert Remezfor providing an extensivelist of Oldenburg' s work that exemplifies this problem.

' ' B. Landau I Wheres what and what s where

269

3.1. Rigid objectsand theshapebias When a child hears an object labelled with a novel count noun, he facesat least a circumscribed version of Quine' s problem : He must decide which of the properties he perceivesare relevant to generalizing the new word. Which ' ' properties of the whole object (Markman , this volume) will the learner select as the basis for generalization? In a first seriesof experiments, Landau et al. ( 1988) investigated 2- and 3' ' year-olds and adults representation of concrete 3-dimensional objects. We askedwhether children and adults would tend to generalizea new count noun ' on the basis of an object s shape, its size, or its texture, and whether any such blaseswere due to generalizedperceptual preferencesrather than the task of learning a new word. We showed subjects a single object - for example, a 2" blue wooden Ushaped ' ' thing - and labelled it by saying, Seethis? This is a dax . This object the standard - was placed aside but still in view and subjects were shown a series of objects, one at a time, which were either size, texture, or shape changesfrom the standard. These changes ranged from relatively minor or moderate changestested in a first experiment (e.g., ~2" larger for the minor sizechange or a slightly bent leg for the shapechange) to either moderate or " quite extremechangestested in a secondexperiment (e.g., 24 for the extreme sizechangeand a completely rotated leg for the shapechange; seefigure 3 for ' ' ! examples ) . For each object, subjectswere asked Is this a dax? The results are shown schematically in figure 3. Subjects of all ages acceptedthe size and texture changesat ceiling levels, but rejected the shape changes. For the minor and moderate magnitudes of change in the first experiment, adults acceptedall size and texture changesbut rejectedall shape changes. Two -and three-year-olds showed the same pattern but probabilisti cally rather than categorically. 5 Note that varyingthe magnitudeof the changefor eachdimensionis critical to determining . of the change itself that childrenattendto, or simplythe magnitude whetherit is the dimension For example , a child might rejecta givenshapechangewhileacceptinga givensizechangejust becausethe shapechangeis quite radical but the size changeis rather minor. Unlessthe , wecannotdissociatethesetwo possibilities(see magnitudeof changesis variedoverdimensions Landauet al., 1988 , for discussion ).

' ' B. Landau / Wheres what and what s where

270

The Bias Shape (Landa ) ~ Smith ,and ,1988 Jones YES

[1:0

~

!

!

~.x.:. U ClIck - Wile

CaIh Us,..

St8Dd8 Id ) (2" woodeD

~ nn 0 ?

b

N

[ J

~ :

:

~ .

.

L

Fig . 3. The shapebias in early lexical learning. When children or adults bear a novel object (the ' standard' named with a novel count noun, they tend to generalizethat noun on the basis of the ) 's texture. on its size or and not , object shape

For the moderate and extreme magnitudes of change in the second experiment, adults again acceptedall size and texture changesand rejectedall ' shapechanges. The adults pattern here was quite rigid : They rejectedeven a very minor shape change while accepting at ceiling a size change that was more than 10 times the size of the standard. Three-year- olds showedthe same pattern as adults, again in a weaker form , while the two year olds did not show any particular preferences, accepting all objects at a relatively high ' ' level. This pattern of results for the 2-year-olds was due in part to the yes/ no ' ' ' ' procedure, which often elicits blanket yes or no in this age group . Using a forced-choice procedure, 2-year-olds did tend to accept size and texture changesand reject shape changes,. as did the 3-year- olds (see Landau et aI., . 1988 . , Experiment 2 forced-choice procedure) Thus we found that children and adults alike extend the reference of a count noun to new objects on the basis of shape. But they reject extension to objects with even quite minor shapechanges. This pattern which we called

B. Landau / Where's what and what's where

the ' shape bias' - appears by age 2 in a relatively fragile form , is stable by age 3, and reachesadult strength by age 5 ( Landau et ale 1992a) . Does the shapebias reflect a general, indiscriminateperceptualpreferencefor grouping same-shapedobjectstogether? One might supposethat children generalize a count noun on the basisof shapejust becausethe objectswith the same shapeas the standardlook more like the standardthan objectswith the samesize or the sametexture. Evidencefor this interpretation would be a shapebias across a variety of object matching tasks, whether they involve an object' s nameor not. To assessthis possibility, Landau et ale ( 1988) presented2- and 3-year-olds with the same standard as before, but did not name the object. Children' s attention was drawn to the standard by pointing and saying ' Seethis?' Then the object was placed aside as before, and the test objects were brought out in ' ' pairs. Children were now asked Which one goes with this? (requesting a ' ' match to the standard) . The expression goes with is, of course, ambiguous; that is, one object could ' go with ' another on many different bases. Previous experiments have shown that children who are asked this question might group on a number of different bases, including various dimensions of perceptual similarity , magnitude of similarity , or thematic relationship with another object (Smith 1989, Markman and Hutchinson 1984, Inhelder and Piaget 1964). The question was whether children would still group on the basis of shape given that there are many possible bases for similarity grouping. Although children had shown a shape bias in the count noun condition , this time they showed either a weak shape bias or none at all. Two -year-olds were as likely to pick the most extreme size change (which had the same shapeand texture as the standard) as the most extreme shapechange (which was identical to the standard in size and texture) . Three-year-olds showed a weak shapebias, nowhere near as strong as in the count noun condition . Thus, by age 3, the shapebias is elicited reliably and strongly in the context of object naming. During the early years of languagelearning, the perceptual ' ' space has already become biased - stretched towards emphasizing different dimensions in different contexts. The context of object naming draws on a perceptual spacewhere similarity of object shapeis critical . Further researchhas confirmed that the shape bias is extremely robust in the context of object naming, that is, under count noun instructions. It appears in 3-year-olds and adults when the objects are three-dimensional (Landau et ale 1988, Au and Markman 1987), or two-dimensional ( Landau et ale 1992b, 1993; Baldwin 1993) ; when the object' s texture or coloration is so salient that children are drawn to those dimensions if asked for non-linguistic

272

's whatandwhat's where B. Landau I Where

similarity judgments (Smith et al. 1992); and when cues to animacy (eyes) ' also pull children s attention to the texture or substanceof objects (Jones et al. 1991) . A shapebias has been replicated in other labs and with school age children (Becker and Ward 1991). In contrast, the shape bias is weakenedor completely suppressedin other syntactic contexts, even though the domain in question still concerns objects (Landau et al. 1992a, Smith et al. 1992) . For example, we found that when ' objects were introduced as membersof a more inclusive category (i.e., This is ' a kind of dax ), young children and adults showed only a weak preferencefor same shape, and no strong preferencefor either size or texture. This makes sense, as membersof more inclusive object categoriesare often lesssimilar in shape (and other dimensions) than members of their subsets. (For example, members of the animal category are less similar to each other than members of the dog category; and dogs are lesssimilar to each other than Dalmatians) . We also found that when objects were introduced to learners using an ' ' adjectival context ( This is a daxy one), they showed no preferenceat all for same shape, but rather, a strong preferencefor samesurface texture - which is encoded by adjectivesin English and other languages(Schachter 1985; see also Waxman, this volume). The results of both syntactic manipulations confirm the notion that the shape bias is confined to the count noun context. This provides further ' evidencefor the importance of syntactic context in directing young children s interpretations of novel words to different parts of the object representation system. 3.2. Artifacts , natural kinds, and shape.. Influencesof perceptionand knowledge

The existence of a shape bias should certainly aid the child in learning new object names and generalizing them to new instances. However , if the bias were too rigid , children would fail to generalize in cases where they should. Although many artifacts are rigid , others regularly undergo transformation of shape - changes in their configuration - because they are jointed (e.g. tools , vehicles) . Still other objects, especially natural kinds , 6 undergo changes of configuration as they move, grow , are squashed, etc. 6 I omit from this discussionthe importanceof perspective transformations , to which young to shape sensitive infants are that shows . Evidence sensitive are learners surely language name a novel will 2 olds that . I assume 1983 and Banks see generalize year ) Salapatek ( constancy ' ' to different perspectivetransformationsof an object (although not necessarilyaccidental ). , i.e. thosethat obtain from unusualviewingperspectives perspectives

B. Landau / Where's what and what's where

273

Does a bias for same shape mean that shape is preserved in a static . ' snapshot -like fashion , corresponding to the configuration shown by an object in just one pose? Or is the bias for same shape responsive to the different kinds of shape transformations a given object might normally undergo? The objects used by Landau et al. ( 1988) all were made of wood and most had hard straight edges and sharp angles, suggesting the rigidity characteristic of artifacts . Thus the strong shape bias among children might have been due to their understanding that objects possessingthese properties could not undergo shape transformations . Such an understanding could stem from perceptually -based or knowledge-based causes. Perceptual processes might specify the kinds of transformations normally expected from objects possessing properties such as straight edges and sharp angles. Knowledge about different object kinds might also affect judgments ; for example, knowing that artifacts normally do not undergo massive shape transformations might lead the child to reject shape changes. And these perception -driven and knowledge-based causes might well interact . To determine whether perceptual causescould modulate children' s judgments about named objects, Landau et al. ( 1992b, 1993) first sought to vary suggestedmalleability by purely perceptual means. Using the same overall method .as in previous studies, we first created outline drawings of the original shapes, including the standard plus the three shapechangesand three new size changes(seefigure 4A for the shape changes). Three-year-olds and adults who were tested using thesestimuli showed a strong shapebias, just as before. In a second series, we tried to suggestmalleability by drawing the same shapes but with curved contours (figure 48 ) . Contour curvature generally correlates with malleability in the physical world , and the question was whether this change would suggesta greater potential for transformation to the children. In a third series, we added another cue to malleability by imposing wrinkles on the curved contours (figure 4C) . In the fourth and final series, we added a single perceptual property that powerfully suggestsa particular onto logical kind . Following Jones et al. ( 1991), we drew eyeson the wrinkled curved drawings (figure 4D ) . Although eyes are a perceptual property , they clearly are powerful in suggesting animate beings. In fact, Jones et al. found that when eyeswere placed on the original wooden objects, 3-year-olds generalized on the basis of same shape

' ' B. Landau I Wheres what and what s where

274 Staadard

! Shape

Sbape2

S Shape

~ ~ ~ QD Q :D~ ~ aD~ ~

C1: O

M

Fig. 4A-D. Drawingsusedin studiesof rigidity, malleability, objectkind, and naming. The first ' the ' standard objectthat wasnamedwith a countnoun. Subjectsin objectin eachrow represents saweitherA , B, C, or D. Testitemsincludedthe corresponding eachof four experiments shape straightedgesand sharpcomers. Thosein changesin eachrow. Objectsin the first row possess the second , , third, and fourth rows vary the basicshapeso as to suggestmalleabilityby curves ' . Subjects wrinkles, andeyes , respectively belongin the named judgmentsof whichshapechanges malleability. ablyin ~ rd with suggested categoryvariedconsider ' ' ' and texture , as if an animate object s stuff mattered in generalizing its

name.7

7 This findingsuggests that youngchildrenaresensitiveto the importanceof texture(or ' stuff' ) of that childrenare as rigid in their preservation of whichanimatesare made, but alsosuggests shapefor animatesas for artifacts. Thedifferencein findingsbetweenJoneset aI. and Landauet the importantrole of materialas it interactswith objectshapein determining aI. ( 1993 ) suggests relativemalleability.

B. Landau I Where's what and what's where

Subjectsshowed a strong shape bias for the drawn versions of the original stimuli (4A ) - the ones with straight edges and sharp comers - but this restriction to the original shape was progressively weakened with the other manipulations. With the curved contour st subjects accepted some shape changesand with the wrinkled and curved contour st they accepted slightly more. The most powerful change occurred with the combination of wrinkled and curved contours topped by eyes; in this casetsubjectsacceptedall of the shape changes. There were no age differencesin the overall pattern : 3-yearolds showed the samedistribution of choices as adults. Thesefindings make two points. Firstt they show that the shapebias is not a bias for accepting only objects having precisely the same static configuration. Children as young as 3 yearsold appreciatethat an objecttsshapeis definedwith respectto the range of shapetransformationsit can undergo- over transforma' tiont the same shapetwill undergo changesin its configuration. Although the rangeof acceptableconfigurationsbroadenswith suggestedmalleabilityt it still is constrainedwithin a clearly delimited range of possibilities. We think it unlikely that children would have accepteda radical topological transformation (e.g. an irregulart roundish shape) as an instanceof a dax. The secondpoint is that perceptualinformation (suchas curvinessor wrinkles or eyes) can playa rich role in engagingchildrents and adultst knowledge of different object kinds and their potential for shapetransformation. This information may concern the objecttsshapeitself (i.e.t its contours) or it may concern other properties that signal certain categories(e.g.t eyes). Although 3-year-olds do not have mature biological theoriesof conceptssuch as ' animaltt ' mammaltt or ' primate' (Johnsonet al. 1992)t perceptualinformation can engageknowledge of different object kindst allowing young children to make predictions about what an object will look like over spaceand time. Thesepredictionsserveas an important foundation for the extensionof object names. 3.3 Summary : Object shape and object name

The foregoing evidencesuggeststhat shape - properly defined - is critical in young children' s extensions of novel object names. Sensitivity to ' same ' shape is clearly complex, with the range of possible shape transformations varying as a function of properties such as malleability and animacy; these properties can be suggested by perceptual information . However, other properties such as size, texture, or material do not appear to play the same role in extension of object names. What underlies the critical role of shapein object naming?

276

B. Landau I Where's what and what's where

One factor is that shape appears to be a privileged dimension of object representation in the human spatial-cognitive system: Objects are recognized largely on the basis of their shape (see, e.g. Biederman 1987) . Shape makes sense as a privileged dimension, because of the spatial constraints under which objects (as opposed to substances ) operate. Normally , under internal or external motion , objects remain rigid (locally , in the case of jointed or semi-malleable objects; globally in the case of non-jointed objects). Infants are sensitive to the intermodal consequencesof rigid vs. non-rigid motions ' (Gibson and Walker 1984) and sensitivity to rigidity appearsto guide infants 8 perception of object unity (Spelke 1990). One important consequenceof rigidity is that an object' s shape remains constant over motion through space: Objects typically do not deform randomly as they move. In contrast, substancesdo not obey such a constraint ; to the extent that they are arranged into a shape when static, this shape changes radically over motion through space, both locally and globally . Indeed, the special status of shape does not generalizeto substances: When we observe paint or sand, shape is not salient at all ; rather, it is the ' stuff ' that is most immediately apprehended.9 The distinction between objects and substancesexists quite early in development, certainly prior to productive control of count/ mass syntax (Sola et al. 1991; seeCarey, this volume); and this no doubt servesas one important constraint on word learning. But how does such a shape-based object recognition system become linked to object naming - and to a shape bias - in the child ? One possibility is that children assume(natively) that object namesare linked to object kinds; when ' ' they are told This is an X , children may assumethat the name of that thing refers to the kind of object it is (seealso Markman , this volume ; Carey, this volume) . Given the fact that object kinds in the world often are differ~ntiated ' ' by object shape, the child will often hear This is an X referring to some kind that is highly correlated with the shape of the object they are observing. In such event, a bias to represent objects in terms of their shapeswill lead the ' 8 To be specific, Spelke ( 1990) proposes that infants object perception is guided by four principles: cohesion, boundedness, rigidity , and no action at a distance. Together, theseprinciples specify which surfacesbelong to the same object and which belong to separateobjects. Spelke further proposes that theseprinciples of object perception are deeply related to basic constraints on the motions of physical bodies: we perceiveobjects according to certain perceptual principles such as rigidity becausebodies move as they do , i.e. they do not deform as they move. 9 We obviously can construe what we seein terms of shape, but then we would describewhat we ' ' see as a ' splotch of paint or a ' pile of sand , drawing attention to the individuated entity . ' or ' ' seeP. Bloom this volume for discussion of related issues. ' , , ) pile ( splotch

B. Landau I Where's what and what's where

277

child to link two pairs of relationships: object name and object kind on the one hand, object name and object shape on the other hand. Such a linking could occur on just one or two exposures(in effect, a kind of ' triggering' in Fodor ' s, 1981, sense), although the evidencereviewedabove does suggestthat the shape bias becomes sharper over time. In any case, once this link is established, children hearing an object' s name would generalizethat name on the basis of object shape. The reliance on object shape in extending object names does not entail that shape is definitional for object kind , nor that it exhausts the possible cues to kind . Obviously , objects sharing the same shape are sometimes members of different kinds (e.g. whales and fish), which means that sometimes a shape bias will lead to incorrect extension of an object name. However , it is significant that in these cases, children ' s errors may be longlived and require specific tutoring before they are corrected. Shape may " remain at the top of a preference hierarchy even for adults , although additional computational resources and brute - force knowledge of the world will allow adults to use other sources of information more easily than young children , thus leading to more complex modulations of any shape bias. The strong reliance on object shape in object naming points to an intricate ' ' system of representation which is specializedfor the whats : object recognition , object naming. This specialization does not appear to be engagedin the systemunderlying the languageof places.

4. How named places are represented

Like names for objects, names for places emerge quite early in language learning: Among the earliest words acquired by children learning English are up and down, in and out. By the age of two and a half , children have a respectablestock of spatial prepositions and by age four , they have learned even some of the most complex English prepositions, such as in front of and behind (Johnston and Slobin 1978, Tanz 1980). The early acquisition of spatial terms equivalent to English prepositions has been documented among children learning languages other than English (Gentner 1982, Bowerman 1991, Johnston and Slobin 1978, Choi and Bowerman 1991), as well as among children whose learning experienceis clearly different from normal those born deaf (Supalla 1988, Feldman et al. 1978) or blind ( Landau and Gleitman 1985, Mulford 1985) .

278

' ' B. Landau I Wheres what and what s where

Given that spatial locational terms are acquired early and with little difficulty , the question arises how learners come to know just what spatial properties of the world are relevant to these terms. Having just seen that object shape is a privileged property for learning object names, one might supposethat learnersshow a shapebias even when learning namesfor places. For example, lerners might assume that something can be in only those objects that are shapedlike a box or bottle. Yet cross-linguistic study of place expressionsrevealsthat locationallanguage does not , in general, preservethe details of object shape. What , then, is the structure of spatiallocational terms that renders them so easy to learn? And how precisely do young learners acquire this structure? 4.1. The structure of place expressions

In English, the standard expression for place is the prepositional phrase PP ( ) . The central element in the place expressionis the spatial preposition words like on, in, off, and around. Sincean object' s location is often defined in terms of other objects, the spatial preposition usually is joined by at least ' two NP arguments, one representing the object being located (the ' figure according to Talmy 1983) and the other representingthe referenceobject (the ' ' ' ' ground ) . For example, in the sentence The cat is on the mat , the preposition Qn is a place-function that defines a region (the surface) of the reference object (the mat) where the figure object (the cat) is located. Each preposition has a characteristic geometry that specifiesthe relevant ' region of the referenceobject (e.g. within the object s boundaries for in ; in ' contact with its surface for on) . The preposition s geometry also specifiesthe kinds of geometric entities that can play the role of figure and reference object (Talmy 1983, Jackendoff 1983, Miller and Johnson-Laird 1976, Herskovits 1986) . For example, the geometry of in roughly specifiesa region of ' containment' . This region then imposes selection restrictions on the arguments of the preposition : In order to qualify as an appropriate reference ' object for the term in, an object must be potentially construable as container' like , e.g. in a bowl , in a carton , or even in a tree, where the tree is an * enclosing volume; but not in a point (since a point , by definition , cannot contain anything) . There are no geometric restrictions at all on the figure object for this term - that is, any object of any geometric classcan be located in a suitable ' container' . Note that the ' suit ability ' of a reference object is almost completely dependenton coercion by the preposition: Things that are not really containers

's whatandwhat's where B. Landau I Where

279

can easily be suitable arguments for in. This fact has led many to emphasize that spatial language draws on ' schematic' representations - that what matters is one' s mental representation of an object, and not necessarilyits physical properties (Talmy 1983, Herskovits 1986). Given the geometric restrictions on prepositions, violations (usually, grossly unsuitable arguments) result in anomaly. For example, since in requires an enclosingareaor volume, a sentencesuchas ( la ) is permissiblewhile ( 1b) is not.

( I a) The fly is in the circle. ( I b) *The fly is in the point. As a more complicated case, the English words across, along, and around seem to require that the figure object be construable as having a principal linear axis . For example , ( 2a) The snake lay across (along) the road . is fine , because the snake has a clear principal axis which can either intersect ( for across) or lie parallel to (for along) the road . However ,

(2b) * The ball lay acrossthe road. is anomalous, except on the reading that the ball lies on some path leading from the observer ' s viewpoint to some point on the other side of the road .

(2c) ?The ba111ay alongthe road. similarly is somewhat odd, although it can be repaired easily by substituting ' alongsideoj, an expressionthat picks out a small section of the road s edgeas the referenceobject. Now the ball can be construed as having an axis that lies parallel to this section. Cross-linguistic studies of spatial meanings have revealed universals in the geometries underlying regions, figure and reference objects (Talmy 1983, Jackendoffand Landau 1991, Landau and Jackendoff 1993) . Most important for the present discussion are the geometries of the figure and reference object. Unlike the representationsunderlying object names, the figure and reference objects tend to be represented in quite sparse geometric terms. Across , they typically are representedas points, areas, surfaces, or volumes, languages

280

' ' B. Landau I Wheres what and what s where

at most preserving quite restricted aspectsof object shape such as the three orthogonal sets of axescorresponding to up/ down, front / back, and right/ left (Talmy 1983, Landau and Jackendoff 1993) . Objects are not representedby precise metric descriptions of shape, even though shape is critical to the representation of objects as category members (but see section 4.4 for some possible exceptions) . For example, as we noted above, the preposition in must be used with a ' ' referenceobject that can be potentially construed as a container , e.g. in a bowl , in a carton, or even in a tree, where the tree is an enclosing volume. Although what is selectedby the preposition is a geometric volumecommon to all three acceptable objects, there clearly are enormous shape differences among bowls, cartons, and trees which are critical in assigning these objects to different named categories. Similarly, on selectsfor a referenceobject that can be thought of as a surface, e.g. on my nose, on the field, on the earth, but not . on the hole. Again , each of thesecan function as a surface, though their individual shapeparametersdiffer greatly. Other than volume and surface, the main elements of shape that are ' preservedconcern the object s three setsof axes. Some terms require that the reference object be construed as having at least one directed axis, either vertical or horizontal (up, down, above, below) . Others call for the reference object to be construed as having two orthogonal axes in the horizontal plane (in front of, behind, beside). Both of these sets of terms leave the geometry of the figure object completely unspecified, while the referenceobject can vary indefinitely in shape as long as its axes are preserved. Terms like acrossand along are among the most geometrically complex prepositions in English, as they require a linear axis for both figure and referenceobjects. Drawing on thesegeometries, only a relatively small set of region types are expressed. At the most general level, all prepositions appear to describe regions that preservenon-metric properties; for example, there is no spatial preposition that describesregions extendingjust 48 degreesto the right of an " object or 2 around its border. Rather, prepositions representproperties such as containment (in, with the referenceobject a volume), support or attachment (on, with the reference object a surface), intersection (across, with reference object an axis), co-linearity (along, the reference object again an axis), and relative proximity (near, in front of, the reference object either a point or a set of axes) . Landau and Jackendoff ( 1993) suggest that these properties may factor into several levels of direction and distance. The direction component relies on analysis of the reference object into the three sets of orthogonal axes

B. Landau I Where's what and what's where

describedabove (up/ down, front / behind, right/ left ) . The distance component chunks continuous space into a small set of discrete categories which vary somewhat from language to language. In English there are four such levels: (a) internal to the reference object (e.g. in, inside), (b) in contact with the object (e.g., on), (c) proximal to it (e.g. near, in front oj) , (d) distal from it (e.g. far and beyond) . Thus, the geometric system for representingfigure and referenceobjects is rather tight and rather ' sparse'. What do these restrictive geometric descriptions portend for the languagelearner? On the one hand, strict limitations in the range of object geometriesmight be a positive force in learning: If the representations are limited , there is less to learn. On the other hand, the object geometriesthat are relevant to prepositional meanings might create a serious problem for the learner, for these geometries are qualitatively different from those used when naming objects. Although detailed object shapeis relevant when learning an object' s name, most or all of an object' s shape is irrelevant when learning the meaning of a spatial preposition. 4.2. Lack % bject shape in spatial term learning

Given the salience of object shape in object naming, one might suppose that children would require considerable time to learn to ignore object shape in the context of prepositions. For example, children might find it easiestto learn terms drawing on richer shape representations (like across or along) than sparser ones. Or , children' s earliest use of terms with sparse shape descriptions might occur only with referenceobjects of a certain shape: The word in might be used only with objects that are canonical ' containers' - fully closing, opaque-sided objects such as boxes and drawers. To my knowledge, there are no reports describing such patterns among young children learning spatial prepositions. What evidence does exist on early usage suggeststhe opposite. For example, experimental studies on the cross-linguistic order of acquisition of spatial prepositions suggestthat terms requiring objects with very little geometric restriction (e.g. near, on) are acquired before those requiring objects with more complex restrictions (e.g. in front of, across, seeJohnston and Slobin 1978) . Similarly, evidence from spontaneous speechsuggeststhat children represent objects schematically- that is, preserving little shapeinformation - when using them as the arguments of prepositions. Landau et al. ( 1990) analyzed ' young children s spontaneoususes of prepositional phrases headed by in or on, asking what kinds of objects served as the reference objects for these

282

' ' B. Landau / Wheres what and what s where

prepositions. The children ranged from 21 to 40 months with mean length of utterance (in morphemes) ranging from 1.0-2.8 (younger group) or 3.0- 5.5 (older group) . We found that not even the youngest children confined their production to ' ' reference objects considered canonical containers (for in) or horizontal surfaces providing gravitational support (for on) . For PPs headed by in, children in the younger group produced referenceobjeCtsthat were enclosures of sometype ; but theseincluded ones that were solid-sided and fully closeable (e.g., house, closet, plane, bottle) , partial volumes open on at least one side (e.g., cup, basket), bounded volumes with non solid sides(e.g. crib, cage), and more sides (e.g. chair, stroller) . or two on partially bounded volumes open Older children also used referenceobjects conceivedas solid objects affording embedding (e.g. a tick in the ear), homogeneoussubstancesconceived as a mass(e.g. in . the woods, rain, hair), and apertures (e.g., window, door) . The same generality held for referenceobjects used with on. The children used referenceobjects that included 2-dimensional horizontal planar surfaces (e.g. paper, fl ()or), horizontal surfacesthat were parts of 3 dimensional objects (e.g. train , bike, chair), and outer surfacesof objects that were not necessarily horizontal (e.g. baby, dolly, fingers, face) . All of thesecould furnish a basefor support, but not only gravitational support. Such patterns of generalization suggestthat young children learning their . first prepositions can represent many objects properly as affording either ' containment' in a ( highly abstract sense) or support/ attachment. from cross-linguistic studies shows that young children evidence Converging learning other languagesalso can represent object classesthat abstract away from exact shape as in English; but they can also schematizeobjects along dimensionsthat are somewhat different from those relevant to English spatial prepositions. Bowerman and Choi (reported in Bowerman 1991) have noted that while English has separateterms for putting objects in, on, or together, Korean collapses these together under the verb kkita , and opposes them to relationships English classifies as taking out, off, or apart, collapsing these under the verb ppayta. While these verbs collapse distinctions that English ' makes, they also cross-cut the English spatial terms with a dimension of tight ' fit / loose fit . For example, while English distinguishes putting a ring onfinger from putting a button in pocket from putting beadstogether, Korean collapses them together as joining relations of tight fit . Other verbs are used for relationships of loose fit , such as cup on table, apple in bowl, or tables together. Bowerman and Choi have reported striking evidence that children learning Korean appear to observe this distinction in their earliest speech,

B. Landau / Where's what and what's where

suggestingthat they can represent objects in tenDs of certain force dynamic notions (i.e., degree of fit ) as well as object geometries such as volume or supporting surface. ' Finally , recall Landau and Steckers ( 1990) findings (seesection 2) . Threeyear-olds who were shown an object being placed on the top right -hand corner of a box and told ' This is acorp my box' generalizedthe tenD acorp to objects of very different shapesas long as they occupied a location on the ' ' upper surface of the box. In contrast, children who were told This is a corp generalizedto objects of the sameshape, regardlessof position . The results of the preposition condition concur with the findings from spontaneousspeech in showing that young children ignore detailed object shapewhen acquiring a new preposition. As a whole, these results suggest that there may not be much learning required by the child in order to ignore the full shape specification of an object. Children might be able to representobjects schematically (in tenDS of certain limited geometric properties) but also more fully (in tenDS of a complete geometric description of shape, for the purposes of object naming) . Critically , however, the schematic representations are engaged only when learning place terms.

4.3. Axesin spatialtermlearning If children ignore object shape when learning place tenDs, when and how do they come to representan object in tenDs of its axes, as would be critical for learning tenDs such as up, down, top, bottom, in front of, behind, and across? Much of the literature on acquisition order for spatial prepositions (and comparable tenDs across languages) suggeststhat the axial system is acquired rather late. For example, Johnston and globin ( 1985) reported that terms such as in front of and behind are mastered substantially later than terms such as in or on. However, the conditions for application of in front of and behindare rather complex, involving more than an axial representationof the referenceobject. For example, one must understand the particular pattern of deictic usageof one' s language- for instance, which portion of the object' s axis is mapped onto each tenD (see, e.g. H . Clark 1973, Herskovits 1986) . If we confine ourselves to the question of when children are able to representobjects in tenDS of their axes - regardlessof whether the particular mapping for individual spatial tenDS is correct - we find that children from age 2 on can represent the axial system. One piece of evidence comes from Tanz ( 1980), who asked children from 2Y2 to 5 years of age to place objects

284

' ' B. Landau I Wheres what and what s where

' in front of ' ' in back of ' and ' at the side of ' reference , objects. Tanz found a the errors, 96% of the 4. However errors , despite through age great many ' ' placementswere made at the cardinal directions: in line with the two axes that divide the referenceobject, front from back and side from side. Additional evidencecomes from some current work in our laboratory on ' the nature of young children s regions mapped to in front of In three studies, we have shown 3-year-olds, 5-year-olds, and adults an array with a single ' reference object, and have asked them to judge when a small object is in front of ' it . In one study, the reference object is V -shaped - essentially, a ' ' largish version of the dax we have used in our object naming studies. This object has a clear vertical axis, and when it is placed horizontally on a table, it has a clear principal axis that runs from front to back, at least for adults. In another study, the referenceobject is round , lacking clear axes. In the third study, the referenceobject is still round , but now is adorned with plastic eyes ' and a small tail , rendering clear the location of this object s principal axis. ' ' Adults , not surprisingly, tend to judge one object in front of the reference object when it falls within a quadrant surrounding the half-axis closest to the viewer. Five-year-olds do the same. Three-year-olds - who might be least ' likely to understand an object s axial system - perform randomly when judging what is in front of the round unmarked object. However, they follow the samepattern as older children and adults when judging what is in front of either the V -shapedobject or the round object with eyesand tail . When they ' perform differently , they follow the reference object s projected axis more closely than either five-year-olds or adults. For example, some of them judge that only objects falling exactly on the relevant half axis are in front of the referenceobject. A last piece of evidencefor early representation of object axescomes from an experiment by Landau ( 1991) on the understanding of spatial part terms (top, bottom, front , back, and side) in 3- to 4- year-olds, one of whom was congenitally blind . Children were shown novel objects, and were specifically told that one region (indicated by the experimenter) was either the top , bottom , front , back, or side of the object. Children now had to identify the regions corresponding to the remaining four terms. Correct responding would have required the children to identify the three major axes, and anchor them with the single term provided by the experimenter. In order to determine whether the terms corresponded to only the canonical position of an object, children were shown regions in each of two conditions : In the Canonical condition , the region was designated as if the object was upright - for example, top was the uppermost region, bottom the

B. Landau I Where's what and what's where

lowermost. In the non-canonical condition , the regions were designatedas if the object was not upright - for example, now top was the lowermost region or the region corresponding to the canonical side. Children performedquite well on this task, identifying the querled regionswell above chance. The blind child performed similarly to the sighted children who viewed the objects. The children performed above chancein both the canonical and non-canonicalconditions, indicating that they could parsethe object into its three orthogonal axes, and that the regionscorrespondingto eachhalf-axis were definedin an object-centeredfashion. That is, the top was at the oppositeend of the bottom, regardlessof the overall orientation of the object. In sum, children from about age 2 appear to be capable of representingan object in terms of its axes. This representation- sparseas it is - may contain the richestgeometricpropertiesnecessaryfor learning placeterms. Although the axes are not physically ' present, young children possessobject representationsthat contain axes; and theseaxescan be engagedduring the learning of place terms. 4.4. Some challenges : How much shape, and when?

Although the foregoing description of spatial terms relies largely on analysis of English spatial terms, a number of investigators have concluded that the geometric properties on which I have focussed also capture place terms in other languages(Talmy 1983, Landau and Jackendoff 1993). There is, however, some disagreementon this issue, and this paper would not be complete without mentioning some dissenting views that raise interesting empirical questions. A number of investigators have suggestedthat there exist languageswhose place terms incorporate much richer object shape than I have described. Specifically, Bowerman ( 1991), Levinson ( 1992) and Brown ( 1993) argue that Tzeltal (and related Mayan languages) exhibit a large class of predicates (several hundred) that include shape as a crucial feature. According to Brown, Tzeltal has but one preposition (roughly meaning ' at' ) . Most detail of the relationship between figure and reference object is expressed by two systems. One draws on body-part terms using a possessiveconstruction: The figure object is located with respect to a named part of the referenceobject, for example, ' The frying pan is at the waterpot' s butt ' , where the ' butt ' is the bottom of the waterpot (independent of its orientation ) This system also is used in English: We can say, for example, ' Mary was at the / oot of the table' or ' The ball was at the bottom of the tree' . Although the number of usable body parts is larger in Tzeltal than in English, neither caseseemsto genuinely

286

' ' B. Landau I Wheres what and what s where

' depend on analysis of the object s shape: The butt of the waterpot is where it ' is in virtue of the object s axial system, not its rotund shape. The second system uses locational predicates that appear to describe the location of objects with particular shapes. For example, according to Brown, ' ' pachal describesa wide-mouthed container canonically sitting , waxal describes ' ' a tall oblong-shapedobject canonically standing , lechel describesa wide flat object lying flat , pakal describesa blob-type object with a clear flat surface ' ' lying face-down . Again , English does share some of these restrictions in its verb system: Stand usually is restricted to objects in their vertical orientation , lie to objects in their horizontal orientation , and numerous denominal verbs (to bottle, to house, to jar ) incorporate specific object categories. Is the Tzeltal system an exception to the sparserepresentation of shape in place naming? It is difficult to tell , since the items in question are verbs, lexical items that can incorporate all kinds of information . But even Tzeltal locational verbs do not appear to incorporate object-shapedistinctions as fine as the vocabulary of object nouns in that language or any other. In fact, the Tzeltal shape distinctions seem to fit with the kinds of distinctions made by classifier systems in other languages (Allan 1973.) Shape classifiers group ' ' ' ' ' ' objects in terms of such parametersas lump , flat , and long thin thing . But classifiersdo not group objects in terms of specific basic-level (shape-based) object categories- this is what object namesare for . It remains an interesting empirical question whether the children learning Tzeltal readily incorporate the so-called richer elementsof shape in their place expressions.

5. What and Where: Spatial languageand spatial cognition Landau and Jackendoff ( 1993; Jackendoffand Landau 1991) have proposed two complementary hypothesesfor the differences in the object geometries underlying the language of object vs. place. One hypothesis focusses on general design features of language, highlighting the fact that vocabulary must filter out detail in some fashion. The second hypothesis focusseson the remaining question of why object representationsare filtered so differently in the two domains of object and place. 5.1. The Design of LanguageHypothesis: Vocabularymustfilter out detail By design, languagesmust filter out a great deal of detail in the translation from perception to word . This is just a consequenceof mapping from the

' B. Landau I Where's what and what s where

287

infinite detail in the observable world to the finite vocabulary. Of course, critical in this filtering is removing just the right kind of detail - and this detail differs for named object vs. located objects. In the case of objects, count nouns group together sets of objects that cohere in a number of respects. Samenessin shape (properly defined) seems to be an important dimension for such categories, while color or texture is not. However, even members of a category named by a basic level noun, though similar in shape, are not identical in shape: Members of the category dog include things as diverse as Dalmatians and Pekinese, Great Danes and Chihuahuas.1OMany of these distinctions are captured at the subordinate level ( Waxman , this volume), but even there, category namescover a multitude of variation . Thus, language filters out a good deal of the perceptual variability among objects, as it must if words are to repres(;nt categories. Although we can perceiveand remember the differencesamong membersof a named category, the design of language does not allow representation in its basic vocabulary for all these differences. Our perceptual rep.resentations of objects seem to preservea good deal more detail than languageencodes. A picture is worth a thousand words. This filtering is even more dramatic in the caseof prepositions. Recall that cross-linguistic studies have shown that a very small number of geometric properties are preserved in the representation of figure and ground objects (Talmy 1983). In English, the geometriesboil down to treating the figure as a ' ' ' ' ' ' ' ' ' blob' or ' line' and the reference object as a point , volume , surface , line or set of axes (as well as a few others, see Landau and Jackendoff 1993) . As demonstrated earlier, this sparse treatment of objects in the spatial domain does not causedifficulties for young children learning spatial prepositions or spatial verbs (see Bowerman 1991) . The filtering of object properties in the place domain is more extreme than it is in the object domain. Could the extreme filtering for prepositions be due to the fact that prepositions are closedclass(function words) ? That is, all closed classitems by virtue of their small number - must categorize perceptual and conceptual spaceto a great degree. Perhaps the need for stretching a very small number of words to cover a large number of potential distinctions inevitably gives rise to such drastic filtering . 10 We find it interesting that children can be resistant to labelling the more unusual looking of thesewith the word dog. The daughter of a friend insisted for a long time that a Pekinesewas a cal, presumably in virtue of its flat face.

' ' B. Landau / Wheres what and what s where

This is unlikely to be the whole story. The fundamental question is not why spatial prepositions filter out detail at all , for all linguistic categoriesfilter as they must, given the fact that their memberships are finite. Rather, the fundamental question is why spatial prepositionsfilter as they do, preserving only certain shape-based qualities. If the requirement were simply to filter without further specification - spatial prepositions could representobjects by gross variation in size, brightness, texture, or any of the other infinite possible properties. A telling comparison involves the noun classifiersfound in many languages such as Japanese, Chinese, and American Sign Language. Noun classifiersare ' ' morphemesthat classify or describethe noun they accompany, and in many languagesthey are obligatory in certain contexts (such as counting) . They are closed class, are relatively few in number, and expressjust a few distinctions among objects. Universally, these distinctions include properties such as ' ' shapeand size of objects (Allan 1973), for example, long and straight thing , ' ' flat ' . These ' small round thing , thing properties appear quite similar to the kinds of geometric properties described in conjunction with figure and referenceobject and might seemto suggestthat it is the closed class nature of the prepositions - and not their pertinence to location - that forces them to draw on the properties they do. However, there are a number of differencesbetweenthe properties engaged by noun classifiers vs. spatial prepositions. First , classifiers do not describe only object geometries. They also often mark animacy, texture or substance, physical integrity (e.g., broken), arrangement (e.g. pleats, loops), and other properties that do not seem to appear in expressionsof pure location (see Allan 1973) . Second, they can be much more specific than preposition geometries, for example, marking certain function -related categories (e.g. in Mandarin Chinese, ' handled things' ; in ASL , ' tools' ), basic level or subordinate ' ' ' ' categories(e.g. in Mandarin , book , rifle ), and even unique"honorific object ' ' categoriessuch as horse (Erbaugh 1986). This suggeststhat there is actually a fair amount of latitude in how closed classelementscan abstract away from object representations. The particular geometric object properties preservedin the spatial preposition meaningsdo not follow necessarily(or solely) from the ' prepositions status as closed class elements. So, although languagesmust filter out detail from spatial and perceptual representations, such filtering cannot by itself account for why objects being named are representedin terms of their detailed shapeswhile objects playing the role of figure or referenceobject are representedin terms of geometries such as point , blob , volume, or axes.

B. Landau I Where's what and what's where

5.2. The Designof Spatial Representations : Languages draw on Hypothesis spatial representations of object and place

The great differences in how objects are representedfor the purposes of naming vs. location might reflect differences in the non-linguistic spatial representationsof object and place. If so, then these representationsmight impose a fundamental constraint on the words representingobject and place, and might also impose a critical constraint on the learner who must sort out the correct geometric descriptions for the two domains. A variety of non-linguistic evidencesuggeststhat there are indeed interesting constraints on the spatial representationsof objects in the two domains of object and place. Studies of object recognition emphasizethe important role played by object shape(e.g. Biederman 1987) and correctly disregard location as a defining element in recognizing an object (except for con~ext effects, which suggestbetter retrieval in familiar contexts) . Studies of spatial cognition have traditionally emphasizedthe important role played by elementsthat ' stand for ' objects in the environment, and the characteristics of the regions within which these objects are located. For example, in his classic study of ' people s understanding of spatial layouts in three major U .S. cities, Kevin ' ' Lynch ( 1960) proposesthat our representationsof layout contain nodes and ' landmarks' ' ' (representing critical objects such as buildings), and paths , ' and ' ' ' districts (representing the spatial regions and links among edges ' . objects) Peoples representation of location does not include detail on the particular objects being related; rather, those objects are represented by abstract elementswhich are often shape-less, or at best, a line or intersection (seeLandau and Jackendoff, 1993, for discussionand related literature). The idea that there are separatesystemsfor representingobjects and places has receivedsupport from other literatures as well. Specifically, neurological and psychological evidence converges on the existence of two qualitatively different kinds of spatial representation- the 'what' system, devotedprimarily to ' ' objects and their identification and the where system devoted to locating objects in space relative to each other and the observer (Ungerleider and Mishkin 1982; Farah 1988, for a review) . These systemsrepresentthe spatial world in different ways, preserving distinct properties for objects vs. places. Evidence for the two systems has been found in studies of lesioning in animals as well as brain damage in humans. The animal evidence shows selectivedeficit of object identification following damage to one area of the brain together with selectivedeficit of object localization following damageto a different area. As an example, work reviewed by Ungerleider and Mishkin

290

B. Landau I Where's what and what's where

shows that damage to the inferior temporal cortex often impairs an animal' s ability to identify an object (often by its shape) in order to obtain food. In contrast, damage to the posterior parietal cortex appears to impair the animal' s ability to identify the location an object occupies, again, in order to obtain food. In the former case, object recognition is impaired while object location is spared; in the latter case, the ability to locate an object is impaired while object recognition is spared. This general dissociation of ' what ' and ' where' has also received support from human psychophysical studies and from studies of brain damage in humans. The psychophysical studies indicate that two separate streams of visual processing exist from lower visual levels up to higher levels. These streams appear to segregatethose properties relevant to object identification (primarily shapeand color ) from those relevant to locating objects (primarily motion , depth, and location ; see Living stone and Hubel 1989) . The human clinical evidencealso showsdissociationof object and place through individual casesin which brain damagecan selectivelyimpair either object recognitionwhile sparing object localization or the converse(Farah et al. 1988) . What do these findings imply for our studies of spatial language? We conjecture that the two kinds of object representation we have uncovered in studies of language learning may be rooted in the larger design of the spatial representational system. The reason that detailed shape matters for object naming is that the count nouns naming objects draw on representationsin the ' what' system. And the reason that sparse shape - only point , line, volume, etc. matter for object location is that spatial prepositions draw on representations in the ' where' system, a systemthat does not representobject shapein detail. We speculatethat the relatively simple shapespecificationsobservedin the preposition system reveal the extent of detail possible in object descriptions within the ' where' system. That is, the degreeof filtering seenfor objects when they play the role of figure or ground is rooted in the fact that the spatiallocational systemsimply does not , by itself, representobject shapesin detailed fashion. This hypothesis is compatible with the view put forth by Talmy ( 1983) in his explanation of why prepositions - and other closed class elements- have the sparse structure that they do. Talmy proposes that the structure of the closed class reflects much of the fundamental structure of cognition ; for example, that the cognitive system itself represents location in relatively qualitative (non-metric) terms. The present proposal further points to a reason why certain members of the closed class - spatial prepositions abstract away from shape as they do. We propose that , in part , they do

B. Landau / Where's what and what's where

reflectthe organizationof spatialcognition, and we haveprovidedconverging evidencefrom non-linguisticstudiesthat this characterizationof the spatial locational systemis correct. It is an empirical question whether similar canbe found within otherclosedclasssets- suchasmarkers correspondences of time, aspect , number, etc. This view suggestsa reasonfor the fact that children find it easy to ' preserveshapewhen learning an objects name, but preserveshapeonly whenlearninga term for an object's place. Theydo so because schematically objectand placeare represented separatelyand distinctly, engagingdifferent schematizations of the world.

6. Conclusions : Observation , object , and word in first language learning

This lengthy foray into the representation and acquisition of object and place terms seemsto have taken us far from the question set out in the beginning: Given the potentially massiveproblems of indeterminacy derived from learning words ' from observation' , how do children manage to map many words onto meanings in a relatively short period of time? The work summarized in this paper suggests that the spatial-cognitive system may impose significant limitations on the kinds of hypotheseschildren will entertain about the meaning of a new word for an object vs. a place. The limitations that I have emphasizedare domain -specific constraints on the kinds of geometric representations that can be engaged. When learning a novel word representing an object, learners will view shape as important , but not location . When learning a novel word representing an object' s place, learners will view location as important , and critically , they will selectively view only certain geometric object properties as r~levant. Some of these limitations will be set in place during the course of early language learning for children learning Korean must focus on somewhat different properties than children learning English. However, on a more general level, the fundamental limitations may precede language learning and serve as a critical boost into the system. Such early predispositions are probably universal, perhaps reflecting the fact that languagesare designed to be set in correspondencewith our spatial representations. Presumably, rich specifications of object shape will be critical in object naming for children learning all languages. Similarly , schematic specifications of object shape will be critical in learning locational expressions, regardless of the language to be learned.

292

' ' B. Landau / Wheres what and what s where

At the most general level, this view emphasizesthe role of observational context , for what the child observes under different linguistic conditions does seemto guide his inferences about the meaning of a new word . But in detail , the evidenceand arguments presented in this paper should serve as a sobering reminder that to note the importance of observational context is not to solve the problems of exactly how this context is engaged during language learning . Our challenge in solving that problem is to discover in detail how learners do represent the world for the purposes of language learning . The study of spatial language appears to provide fertile ground for meeting these challenges.

References . Language53(2), 285- 311. . Classifiers Allan, K., 1977 . What someconceptsmight not be. ArmstrongS ., L.R. Gleitman and H. Gleitman, 1983 Cognition13(3), 263- 308. . Acquiring word meaningsvia linguistic contrast. Cognitive Au, T. and E. Markman, 1987 . 2 217 236 , Development . Journalof . Clarifyingthe role of shapein children's taxonomicassumption Baldwin, D., 1992 ExperimentalChild Psychology54, 392- 416. . In : M. Haith, J. Campos(eds.), Banks, M. and P. Salapatek , 1983. Infant visual perception Infa~cy and developmental , Vol. 2 of P.H. Mussen(ed.), Handbookof child psychobiology , 435- 572. New York: Wiley. psychology . . Children's useof shapeand texturewith objectsand substances T. Ward, 1991 A. and Becker , . in Child Development at the Societyfor Research , Seattle , Washington Paperpresented -by- components . : A theory of human imageunderstanding . Recognition Biederman , I., 1987 . 115 147 2 Review 94( ), Psychological . Oneword at a time. The Hague: Mouton. Bloom, L., 1973 . The role of syntax-semanticmappingsin the acquisitionof . Possiblenames Bloom, P., 1994 . Lingua92, 297- 329(this volume). nominals : Cognitive vs. . The origins of children's spatial semanticcategories Bowerman , M., 1991 . In: J.J. Gumperz , S.C. Levinson(eds.), Rethinkinglinguisticrelativity. linguisticdeterminants . , MA : CambridgeUniversityPress Cambridge . 25th Annual . The role of shapein the acquisitionof Tzeltal(Mayan) locatives Brown, P., 1993 Forum, StanfordUniversity, April . Child LanguageResearch . Journalof Abnormaland Social . Linguisticdeterminismand parts of speech Brown, R., 1957 55, 1- 5. Psychology . . Cambridge . A first language Brown, R., 1973 , MA : HarvardUniversityPress L.R. Gleitman(eds.), In: E. Wanner the art. The state of : . , 1982 Semantic S. development Carey, , . : The stateof the art, 347- 389. NewYork: CambridgeUniversityPress acquisition Language . MIT Press Books MA : Bradford . in childhood . Conceptualchange , / Cambridge Carey, S., 1985 ? 92 the world to the child , a 1994 . Does S. Lingua reconceptualize , language require learning Carey, 143- 167(this volume).

B. Landau 1 Where's what and what's where

Clark. E.V.. 1973 . What's in a word? On thechild' s acquisitionof semantics in his first language . In: T. E. Moore (ed.), Cognitivedevelopmentand the acquisitionof language , 65- 110. New York: AcademicPress . Clark. H., 1973 . Space , time, semantics , and the child. In: T.E. Moore(ed.), Cognitivedevelopment and the acquisitionof language . , 27- 64. New York: AcademicPress Choi, S. and M. Bowerman . Learningto express , 1991 motioneventsin Englishand Korean: The -specificlexicalizationpatterns.Cognition41, 83- 122. influenceof language . Takingstock: Thedevelopment , M., 1986 of Chinesenounclassifiers Erbaugh historicallyandin : youngchildren. In: C. Craig (ed.), Noun classesand categorization , 399- 436. Amsterdam . Benjamins Farah. M., 1988 . Is visualmemoryreally visual? Overlookedevidencefrom neuropsychology . Review95(3), 307- 317. Psychological Farah. M.. K. Hammond . Visualand spatialmentalimagery: , D. Levineand R. Calvanio, 1988 Dissociable . CognitivePsychology 20, 439- 462. systemsof representation Feldman , H.. S. Goldin-Meadowand L.R. Gleitman, 1978 . BeyondHerodotus: The creationof languageby linguisticallydepriveddeaf children. In: A. Locke (ed.), Action, symbol, and : The emergence of language . New York: AcademicPress . gesture Fisher, C., H. Gleitmanand L.R. Gleitman, 1991 . On the semanticcontentofsubcategorization frames . CognitivePsychology 23(3), 331- 392. Fisher. C.. D.G. Hall, S. Rakowitzand L. Gleitman, 1994 . Whenit is betterto receivethan to give: Syntacticand conceptualconstraintson vocabularygrowth. Lingua 92, 333- 375 (this volume). Fodor. J., 1981 . The presentstatusof the innateness . Representations . Cambridge controversy , MA : MIT Press . Gentner. D.. 1982 . Why nouns are learnedbefore verbs: Linguistic relativity vs. natural : Language partitioRing. In: S. Kuczaj (ed.), Languagedevelopment , culture, and cognition 301- 334. Hillsdale, NJ: Erlbaum. Gibson. E and A. Walker, 1984 . Developmentof knowledgeof visual-tactual affordancesof . Child Development substance 55, 453- 460. Gilette. J. and L.R. Gleitman, forthcoming . Observationand nounverb learning. Gleitman, L.R., 1990 . The structuralsourcesof verbmeanings . LanguageAcquisitionI , 3- 55. Gleitman, L.R., H. Gleitmanand E. Wanner, 1988 . Wherelearningbegins: Initial representations for languagelearning. In: F. Newmeyer(ed.), Linguistics:The CambridgeSurvey. New York: CambridgeUniversityPress . Grimshaw, J.. 1981. Form, function, and the languageacquisition device. In : C. Baker, J. McCarthy(eds.), The logicalproblemof languageacquisition , 183-210. Cambridge , MA : MIT Press . Grimshaw . Lexicalreconciliation . Lingua92, 411- 429(this volume). . J., 1994 Herskovits . Languageand spatialcognition: An interdisciplinarystudyof the prepositions , A.. 1986 in English. Cambridge : CambridgeUniversityPress . Inhelder,B. andJ. Piaget, 1964 . Theearlygrowthof logicin thechild. NewYork: Norton, 1969 . Jackendoff . Semantics and Cognition. Cambridge . . R., 1983 , MA : MIT Press Jackendoff . On beyondzebra: The relation of linguistic and visual information. , R., 1987 Cognition26, 89- 114. Jackendoff . Mme. Tussaudmeetsbinding theory. Natural Languageand Linguistic . R., 1992 Theory 10, 1- 31.

294

' ' B. Landau I Wheres what and what s where

. Spatiallanguageand spatialcognition. In: D.J. Napoli Jackendoff , R. and B. Landau, 1991 (ed.), A SwarthmoreFestschriftfor Lila Gleitman, 145- 170. Hillsdale, NJ: Erlbaum. in English, Serboof locativeexpressions . Thedevelopment Johnston , J.R. and D.I. Siobin, 1978 Croatian, and Turkish. Journalof Child Language6, 529- 545. within thestructureof . Developmental Johnson changes , K.E., C.B. MervisandJ.S. Boster, 1992 28( 1), 74- 83. the mammaldomain. Developmental Psychology . Objectspropertiesand knowledgein early lexical JonesS., L. Smith and B. Landau, 1991 62, 499- 516. learning.Child Development . . Cambridge . Concepts Keil, F., 1989 , MA : MIT Press , kinds, and conceptualdevelopment : to domain tasks abilities . Domain 1994 M.H. and S. Martin , specific applied general Kelly, . Lingua92, 105- 140(this , cognition, and language Sensitivityto probabilitiesin perception volume). . Will the real grandmotherpleasestandup? The psychological Landau, B., 1982 reality of dual Research11( 1), 47- 62. . Journalof Psycholinguistic meaningrepresentations Landau, B., 1991. Spatial representationsof objectsin the blind child. Cognition 38, 145178. : Evidencefrom the blind child. . Languageand experience Landau, B. and. L. Gleitman, 1985 . , MA : HarvardUniversityPress Cambridge . ' What' and ' Where' in spatial languageand spatial Landau, B. and R. Jackendoff , 1993 16, 217- 266. cognition. Behavioraland Brain Sciences . Geometricalcomplexityand objectsearchin infancy. Developmental Landau, B. and E. Spelke , 1988 4 521. 512 24( ), Psychology in : Geometricand syntacticrepresentation . Objectsand places Landau, B. and D. Stecker , 1990 5, 287- 312. earlylexicallearning. CognitiveDevelopment . The importanceof shapein early lexicallearning. Landau, B., L. B. Smith and S. Jones , 1988 3, 299- 321. CognitiveDevelopment . Syntacticcontextand the shapebiasin children's and Landau, B., L. Smithand S. Jones , 1992a ' . Journalof Memoryand Language31, 807- 825. adults lexicalleaming . Geometricconceptsin youngchildren's usesof in Landau, B., L. Sorichand D. Stecker , 1990 on Infant Studies and on. InternationalConference , Montreal. Landau, B., M. Leyton, E. Lynchand C. Moore, I 992b. Rigidity, malleability, objectkind, and SocietyMeeting, St. Louis, April . objectnaming. Psychonomics . Perception Landau, B., M. Leyton, E. Lynch and C. Moore, 1993 , object kind, and naming. . Manuscript . Wipingthe slateclean: A lexicalsemanticexploration.Cognition Levin, B. and M. Hovav, 1991 41, 123- 152. . Vision, shape Levinson , and linguisticdescription: Tzeltalbody-part terminologyand , S., 1992 . Working paper No. 12, CognitiveAnthropologyResearchGroup, Max object description . PlanckInstitutefor Psycholinguistics . Segregationof form, color, movement , and depth: Livingstone, M. and D. Hubel, 1989 . Science240, 740- 749. , and perception Anatomy, psychology . . The imageof the city. Cambridge , MA : MIT Press Lynch, K., 1960 . Lingua92, . Constraintson word meaningin early languageacquisition Markman, E.M., 1994 199-277(this volume). ' Markman, E. and J. Hutchinson , 1984. Childrens sensitivityto constraintson word meaning Taxonomicversusthematicrelations.CognitivePsychology16( 1), 1- 27.

' B. Landau / Wheres what and what's where

McKenzie . Localizationof eventsin space : Young infants , B.E., R.H. Day and E. Ihsen, 1984 are not alwaysegocentric . BritishJournalof Developmental 2. 1- 9. Psychology -Laird. 1976 Miller, G. and P. Johnson . Languageand perception . Cambridge . MA : Harvard Press . University Mulford, R.. 1985 . First words of the blind child. In: M.D. Smith. J.L. Locke (eds.). The ' . New York: Academic emergentlexicon: The child s developmentof linguisticvocabulary Press . . The role of theoriesin conceptualcoherence . Psychological Murphy, G. and D. Medin. 1985 Review92, 289- 316. . Changesin conceptualstructurewith expertise : Differences Murphy. G.L. and J.C. Wright, 1984 betweenreal world expertsand novices . Journalof ExperimentalPsychology : Learning. Memory. and Cognition10. 144- 155. . Learnabilityand cognition: The acquisitionof argumentstructure.Cambridge Pinker, S., 1989 . MA : MIT Press . Pinker, S., 1994 . How coulda child useverbsyntaxto learnverbsemantics ? Lingua92. 377- 410 (this volume). . Word and object. Cambridge . . MA : MIT Press Quine, W., 1960 Rieser . Spatialorientationof six-month-old infants. Child Development50. 1078 , J.J.. 1979 . 1087 -Braem. 1976 . Basicobjectsin Rosch , E., C. B. Mervis. W.O. Gray, D.M. Johnsonand P. Boyes naturalcategories . CognitivePsychology8. 382- 439. . Why are 'what' and ' where' processed Rueckl, J., K. Caveand S. Kosselyn . 1988 by separate ? A computationalinvestigation . Journalof CognitiveNeuroscience I, corticalvisualsystems 171- 186. . Parts of speech . In: T. Shopen(ed.). Languagetypology and syntactic Schachter . P.. 1985 . . Vol. I : Clausestructure. 3- 61. Cambridge . MA : CambridgeUniversityPress description . A model of perceptualclassificationin childrenand adults. Psychological Smith, L.B., 1989 Review96. 125- 144. . Cambridge . Categoriesand concepts Smith, E. and D. Medin. 1981 . MA : Harvard University . Press . Countnouns. adjectives Smith. L. B., S. Jonesand B. Landau. 1992 . andperceptualpropertiesin . Developmental 28, 273- 286. novelword interpretations Psychology ' . Ontologicalcategories , 1991 guideyoungchildrens inductions Sola, N., S. Careyand E. Spelke terms. Cognition38(2). 179-211. of word meaning:Objecttermsand substance . Principlesof objectperception . CognitiveScience14. 29- 56. , E., 1990 Spelke . . Structureand acquisitionof verbsof motion in AmericanSign Language Supulla, E., 1988 . MA : BradfordBooks. Cambridge . In: H. Pick. L. Acredolo (eds.). Spatial . How languagestructuresspace Talmy, L.. 1983 . orientation: Theory, research , and application. New York: PlenumPress . Lexicalizationpatterns: Semanticstructurein lexicalforms. In: T. Shopen(ed.). Talmy, L., 1985 and the lexicon. . Vol. 3: Grammaticalcategories Languagetypologyand syntacticdescription . : CambridgeUniversityPress 57- 149. Cambridge : CambridgeUniversity . Studiesin the acquisitionof deictic terms. Cambridge Tanz, C., 1980 Press . . Two corti~ l visualsystems . In: D.J. Ingle. M.A. Goodale . , L.G. andM. Mishkin. 1982 Ungerleider . Cambridge . R.J.W. Mansfield(eds.). Analysisof visualbehavior . MA: MIT Press

296

's whatandwhat's where B. LandauI Where

of an appreciationof specificlinkagesbetween linguistic . The development Waxman , S.R., 1994 . Lingua92, 229- 257(this volume). andconceptualorganization

) 297- 329. North-Holland Lingua92 ( 1994

297

Possible names: The role of syntax - semantics mappings in the acquisition of nominals * Paul Bloom Departmentof Psychology, University of Arizona, Tucson, AZ 85721, USA Many scholars have posited constraints on how children construe the meaningsof new words. These include the restriction that new words refer to kinds of whole objects ( Markman and Hutchinson 1984), that words describing solid objects refer to individuated objects while words describing non-solid substancesrefer to portions of substance(Sola et al. 1991), and that count nouns that name objects are generalizedon the basis of shape (Landau et aI. 1988). There are theoretical and empirical problems with theseproposals, however. Most importantly , they fail to explain the fact that children rapidly acquire words that violate these constraints, such as , and names for non-material entities. The pronouns and proper names, names. for substances theory defended here is that children and adults possessmappings from grammatical categories ' ' ' ' ' ' ( count noun , mass noun , and noun phrase ) to abstract semantic categories; thesemappings serveto CQnstraininferencesabout word meaning. Evidencefrom developmental psychology and linguistic theory is presentedthat suggeststhat even very young children possesssuch mappings and use them in the course of lexical development. Further issues- such as the possibility of developmental change, the precise nature of these semantic categories, and how children learn words prior to the acquisition of syntax - are also discussed. It is concluded that although these ' syntax-semanticsmappings are not by themselvessufficient to explain children s successat word crucial role in lexical . As such a learning, they playa , only theory that posits a deep development relationship betweensyntax and semanticscan explain the acquisition and representationof word meanings.

'

Language ... must have its perfectly exclusivepigeon-holes and will tolerate no flying vagrants. Any concept that asks for expressionmust submit to the classificatory rules of the game ... It is almost as though at some period in the past . I amgratefulto FeliceBedford, Susan Carey, Frank Keil, EllenMarkman, JanetNicol, Mary Peterson . NancySola. an anonymousreviewer , andespeciallyKarenWynnfor their veryhelpful commentson earlierversionsof this paper. This work was supportedby a NIH Biomedical Research SupportGrant. 0024 - 3841 / 94/$07.00 @ 1994- ElsevierScienceB.V. All rightsreserved SSDI0024- 3841( 93) EO042- 6

298

P. Bloom I Possiblenames

the unconsciousmind of the race had made a hasty inventory of experience , committed itself to a premature classification that allowed of no revision, and saddled the inheritors of its language with a sciencethat they no longer quite believed in nor had the strength to overthrow. Dogma, rigidly prescribed by tradition , stiffens into fonnalism. Linguistic categories make up a system of surviving dogma - dogma of the unconscious.' Edward Sapir, Language( 1921: 99- 100)

Introduction

One of the deepestmysteries in the study of languagedevelopment is how ' children learn' the meanings of words. A child s vocabulary grows at an extraordinary rate - one estimate is that children acquire about nine new words a day from the age of 18 months to six years (Carey 1978) - and there is little understanding of how this processtakes place. This ignorance is due at least in part to the fact that there is no consensuson what it is for someone ' to possess' the meaning of a word (for discussion, see Carey 1982, Lakoff 1987, Premack 1990) . In general, no theory of acquisition can be complete without some understanding of the nature of what must be acquired. A further difficulty concerns the nature of the learning problem itself. Word learning is standardly viewed as an inductive processand, as Goodman ( 1983) has stressed, there is an infinity of logically possible generalizations that one can make on the basis of a finite set of instances. To take a specific ' example, consider an adult pointing to Fido and saying to a child Look at ' the dog . Imagine that somehow the child is capable of determining what the word is intended to describe, i .e., it describes Fido , not the child , or the finger, or the act of pointing , etc. Imagine further that the child can segment ' ' the utterance into words and can determine that the relevant word is dog , ' not ' look ' , ' at' , or ' the . Still , there are countless possible meanings of this novel word. It could refer to the basic-level kind (dogs), but it could also refer to a subordinate kind (poodle), or a superordinate kind (animal), or to the individual (Fido ) . It could refer to the color of the entity being pointed to (brown), to its shape(oblong), or its size (large) . It could refer to a part of the dog (tail ) . It could refer to the front half of the dog, to dogs until the year 2000 and then to pigs, to all dogs and all pencils, to all dogs and also to Richard Nixon . To modify an example from Quine ( 1960), it could even refer to undetacheddog parts.

P. Bloom / Possiblenames

299

From a logical standpoint, all of theseexamplesare possible, since they are all consistent with the ostensive act. From a psychological standpoint, however, some of these possibilities are ludicrous - no child would ever construe the word ' dog' as referring to just the front half of Fido , or to the ' ' category of dogs and pencils . Any theory of word learning must explain why some word meanings are more natural than others, and how children determine which of the set of natural meanings that a word could have actually corresponds to its meaning in a given language. More generally, any succesfulinductive procedure requires that hypotheses be somehow ordered or ranked (Fodor 1975, Goodman 1983) and one possibility is that there exist constraints that rule out (or bias against) entire classesof hypotheses. In the domain of languagedevelopment, Markman and her colleagueshave presentedthe following two constraints, which are argued to be special to the domain of word learning (e.g., Markman and Hutchinson 1984, seeMarkman 1990for a review) : Whole Object constraint: ' ... a novel label is likely to refer to the whole object and not to its parts, substance, or other properties.' (Markman 1990: 59) Taxonomic constraint: ' ... labels refer to objects of the same kind rather than to objects that are thematically related' ( Markman 1990: 59) . Thematically related entities include those that fall into ' spatial , causal, temporal , or other relations ' such as a dog and its bone, a dog and the tree that it is under , a dog and the person who is petting it , and so on. Although children are sensitive to these sorts of relations in non -linguistic tasks (for instance, they will put a dog and a bone together when asked to sort objects into different piles), this constraint forces them to attend to taxonomies ' ' (such as the kind dog ) when faced with the task of inferring the meaning of a new word . Another proposal, advanced by Sola et ale ( 1991: 182- 183), is that the following two proceduresapply in the processof word learning : Procedure1: Step I : Test to seeif the speakercould be talking about a solid object; if yes, Step 2: Conclude that the word refers to individual whole objects of the sametype as the referent.

P. Bloom / Possiblenames

300

Procedure 2 .. Step I : Test to see if the speaker could be talking about a non solid substance ; if yes, Step 2: Conclude that the word refers to ,portions of substance of the same

type as the referent.

) posit the following bias in lexical Finally, Landau et al. ( 1988, 1992 acquisition :

Shapebias; ' ... the biasto of a novelcountnoun group objectsby shapein thepresence ...' (Landauet al. 1992: 87) There is by now a large body of evidenceshowing that 2- and 3-year-olds behavein accordancewith theseposited constraints. When taught a ~ 9rd for a novel object, children tend to categorize the word as referring to other whole objects of the same kind ; they will not extend the word to entities ' ' sharing a thematic relation (Markman and Hutchinson 1984, Waxman and Gelman 1986) and will not initially interpret it as referring to a part of the object, a property of the object, or the stuff that the object is made of (Baldwin 1989, Clark 1973, Macnamara 1982, Markman and Wachtel 1988, Sola et al. 1991, Taylor and Gelman 1988) . In contrast, when taught a word for a novel non-solid substance, they will tend to generalizethe word on the basis of the kind of substance(perhaps using texture and color as cues), and ignore properties such as shape and size (Sola 1987, 1992; Sola et al. 1991) . Finally , when taught count nouns that describe objects, children will tend to generalize these nouns on the basis of shape, not color , size, or texture (Baldwin 1989, Landau et al. 1988) . Nevertheless, there are reasonsto doubt that these precise constraints are present in the minds of young children. For one thing , they are false of adult language- all languageshave words that do not refer to taxonomies, words that do not refer to whole objects, and count nouns that name kinds of objects that do not share a common shape. Below it is argued that these counterexamplesare also present in the language of very young children. A further concern is that the sole motivation for positing these constraints is their role in solving the word learning problem. It would be preferable to derive these constraints from deeper properties of language and cognition , instead of having to simply stipulate them. In this paper, I present a theory of where constraints on word meaning

P. Bloom / Possiblenames

comefrom. It is oftenmaintained thatchildrenpossess between mappings whichfacilitategrammatical andlexicaldevelopment syntaxandsemantics (e.g., BloomI990a ,b; I994a ,b; Brown1957 , Carey1982 , Gleitman1990 , Grimshaw 1981 andGleitman 1985 etal. , Katzetal. 1974 , Landau , Landau 1988 1982 andReyes1990 , 1992 ; Macnamara , 1986 ; Macnamara , Naigles 1990 Pinker 1984 1989 and Gelman 1988 , , ; Taylor , Waxman1990 ). It is -semantics herethattheexistence of certainsyntax in the argued mappings domainof nominals makesit unnecessary to positspecialword-learning constraints andthata theorybased on suchmappings allowsfor a better ! of how children learn . words explanation young 2. Problems with proposed constraints 2.1. Description of adult language

As Markman ( 1990) notes, the whole object constraint is false at least for the lexicons of older children and adults. There exist words that refer to ' ' ' ' ' ' properties ( happy ), to actions ( hit ), to spatial relations ( under ), to substances ' water' and so on. In general, adjectives, verbs, prepositions, and ( ) massnouns do not refer to whole objects. One apparent solution would be to redescribe the constraint as only applying to count nouns, and thus exclude examplesfrom other grammatical categories. But the generalization still does not hold. Some count nouns describewhole objects (' dog' ), but most do not. Within the domain of count nouns that describe material entities, some refer to collections of objects ' ' ' ' ' ' ' ' ' ' ( forest , bikini , flock ), while others refer to parts of objects ( finger , foot , 1 Otherconstraintson word learningthat havebeenproposedincludemutualexclusivity , which is that wordscannothaveoverlappingextensions (MarkmanandWachtel1988 ), andtheprinciple fonDSin language . As with of contrast(Clark 1987 ), whichstatesthat thereare no synonymous theconstraintsdiscussed above,mutualexclusivityis clearlyfalseof adult language(asit excludes ' ' ' ' debateover how well it pairs of words suchas dog and animal) and thereis considerable . appliesto the languageof youngchildren(seeAu andGlusman1990 , Gathercole1987 , Merviset al. 1991 , Nelson1988 ). The principleof contrastmay be correctfor adult languagebut it is not . the child' s inferencesand must work in aa: ord with other by itself sufficientto constrain . Neitherof theseconstraintscan becapturedin tenDS of the sortsof syntax-semantics principles here, but it is possible , as suggested mappingsdiscussed by both Clark ( 1990 ) and Gathercole andcognition. In ( 1987 ), that theseconstraintsmightalsoreduceto deeperpropertiesof language particular, a prohibition on certainfonDS of lexicaloverlapmay be derivablefrom children's . , e.g., of Griceanmaximsof communication knowledgeof pragmaticprinciples

302

P. Bloom I Possiblenames

' handle' . In fact some nouns refer to , ) ( parts of objects that can neverappear ' ' ' ' as discrete entities, such as surface or coating .) The largest class of ' ' ' ' ' ' ' ' exceptions are count nouns like nap , idea , race , and dream , which do not refer to material entities at all. Given that the shape bias is intended to apply only to count nouns that name objects, words like ' nap' do not violate this bias. But other nouns do , ' ' ' such as names for collections (' army' , ' family ), superordinates ( animal , ' ' ' ' ' ' weapon ) and relationship terms ( brother , friend ) . In fact, it is far from clear that shape is the crucial dimension even for count nouns that refer to basic-level discretewhole objects, which are those consideredby Landau et al. ( 1988). As Sola et al. ( 1991, 1992) argue, even for young children, shapeis not ' ' . ' criterial in determining the extension of count nouns like dog or skunk ; something can be thought of as a dog even if it is shaped like a cat (Keil 1989). What Landau et al. have found is that children prefer to generalize novel nouns referring to discrete material objects on the basis of shape as opposed to size or texture. But this is a far weaker conclusion than the claim ' that count nouns, or even count nouns referring to objects, correspond to ' categorieswhose membershave similar shapes ( Landau et al. 1988: 316) . On ' ' ' ' the contrary , there is only a small number of nouns, such as square , globe , ' ' and pyramid , where shape is an essentialproperty of how they are used. The taxonomic constraint is also false for adults. While it is true that words cannot refer to chains of thematically related categories, there are nominals that do not refer to taxonomies; these include pronouns and proper names ' ' ' ' ' ' ( he , Fred , Canada ), which refer to particular individuals, and do not generalizeto other entities. Of all of the constraints proposed above, only those advancedby Sola et al. ( 1991) are largely correct with regard to adult language - but (once again) ' ' ' ' only when we replace word with noun (as they themselvessuggest, p . 203). It is clearly false that any word used to describea solid object can be extended to objects of the same type, since the word could be describing a property of ' ' the object (' red' ), the state of the object ( resting ) , and so on, and the same observation applies to the claim that any word U'sedto describe a non-solid substancecan be extended to that type of substance. But although these , there are still procedureswork better when their domain is restricted to nouns ' ' ' ' caseswhere they fail. Somemassnouns, like wood and metal , describesolid objects but are not extendedto objects of the sametype; they are extendedto objects composedof the same material. The opposite sort of counterexample ' ' ' concernscount nouns like ' pile and puddle ; they describesubstances , but are extendedon the basisof configuration, not substancekind.

P. Bloom / Possiblenames

2.2. Descriptionof child language The same counterexamplesfound in adult language also show up in the spontaneousspeechof young children. As Nelson ( 1988, 1990) has stressed, although nouns (or nominals) may be the largest single class of words in the beginning vocabulary of some young children (Gentner 1982), it is false that all , or even most, of children' s first words are nouns. Children ' s early lexicons include expressionslike ' more' , ' bye' , ' hit ' , ' want' , ' up' , and ' no' and these ' clearly refute any claim that children s first words must refer to kinds of whole objects (seealso Benedict 1979, Gopnik and Choi 1990) . What about the more specific claim that all of the nomina/s used by young children refer to kinds of whole objects? This still leavesus with constraints that are descriptively inadequate, however. One problem concerns words like ' milk ' and ' water' , which refer to kinds, but to kinds of substances , not objects. A second problem concerns pronouns and proper names, which do not refer to kinds. Given thesecounterexamples, one could further revise the constraint proposal as follows : Constraint hypothesis: In the lexicons of young children, count nouns refer to kinds of whole objects In some discussionsof the taxonomic and whole object constraints, this is alluded to as the most descriptively adequateversion of the constraint theory and Markman ( 1989), although usually defining the constraints as applying to ~labels' , sometimesdescribesit in this manner, as limited to count nouns. This revision still has difficulties, however. Even two -year-olds possess names for collections like ~forest' and ~family ' , despite the fact that these count nouns violate the whole object constraint (Bloom and Kelemen, under review; Callanan and Markman 1982). Nelson ( 1990: 335) gives as a further ~ example the many abstract social and cultural conceptsthat are incorporated into the language and presented in passing to children, who pick them up ' seemingly without effort and she lists some words found in the speech of 20-month-old children ; theseinclude count nouns such as: ' bath' , ' breakfast' , ~friend' ' week' and ' uncle' . Finally , Nelson et al. ( 1993) analyzed the speech , , of 45 20-month-olds and found that , although thesechildren possessedmore nouns than any other category, only about half of these nouns referred to basic-level classesof objects; many referred to locations (' beach'), temporal entities (~day' ), and events (~party ' ) .

304

P. Bloom I Possiblenames

The rapid acquisition of such words is clearly problematic for the whole object constraint and the taxonomic constraint. It also poses a difficulty for the shape bias. As noted by Sola et al. ( 1991, 1992), it would be surprising ' ' ' ' indeed if children extended words like uncle and clock on the basis of shape; if they did , they would be unable to use thesewords in any way similar to adults. Finally consider the Sola et al. ( 1991) procedures, which state that words describing objects refer to kinds of individual whole objects and words describing substancesrefer to portions of substance. Revising their proposal so that it applies only to nouns, this predicts that children should be unable ' to acquire solid substancenames(' wood ) and namesfor bounded substances ' . But in fact even 2' year-olds can learn solid substancesnames ( puddle ) (Prasada 1993), and there is evidence from Sola ( 1992) suggestingthat they can also acquire namesfor bounded entities (seesection 4.2 for discussion) . It ' is worth noting, however, that these are not children s first guessesas to a ' word s meaning. For instance, when a word is used to describe a novel ' bounded entity , children s first interpretation of its meaning is that it refers to the kind of object, not the stuff that the object is made out of , and this holds regardlessof the syntax in which the word is presented (Au and Markman 1987, Dickinson 1988, Markman and Wachtel 1988, Sola 1987) . Nevertheless, ' the fact that words like ' wood are acquired at all militates against the claim that children possessthe proceduresposited by Sola et al. ( 1991). None of the counterexamplesdiscussedabove necessarilyrefutes the hypothesis that these constraints exist. One possibility, discussed in detail by Markman ( 1989), is that although they are present at the start of lexical ' ' development, the constraints can be abandoned or overridden in the course of language development (possibly as the result of the application of other constraints; see footnote 1) . More generally, they can be viewed as default conditions which only apply in the absenceof certain countervailing circumstances , and such circumstances may be present at any stage of lexical development. . Under this interpretation of what constraints are, however, it is unclear whether any degree of understanding on the part of young children could refute theseproposals. For instance, pronouns and proper namesare acquired very early, often before children have learned any other noun that describes ' ' ' ' ' ' people (such as person , man , and woman ) . As such, constraints such as' ' Mutual Exclusivity cannot block the child from interpreting words like Fred ' ' and she incorrectly, as namesfor kinds of whole objects. In fact, thesesorts of errors do not occur (see Macnamara 1982) - but a reasonablereply by a

P. Bloom / Possiblenames

305

constraint -theorist is that the special status of what pronouns and proper names refer to (i .e., people) causes children to override the taxonomic assumption in such cases. In general, the fact that these constraints are posited as default conditions makes it important to provide some theory of what other rules and constraints can override them. Without such a theory , the constraint proposal runs the risk of begging all of the difficult questions. In sum, the theories of Markman , Landau et al., and Sola et al. cannot account for most of the words that children acquire. But the very same induction problem that exists for the acquisition of a word such as ' dog' also exists for words such as ' Fred' , ' water' , and ' forest' , and thus the very same arguments for the necessityof constraints also apply. The goal of a theory of lexical development is to account for the acquisition of all words, not just names for kinds of objects, and this motivates an effort to explain the acquisition of object names in the context of a more general theory of word learning. 2.3. Learning issues

Where do constraints on word learning come from ? In this regard, it is worth echoing Nelson' s ( 1988) complaint that some of the theorists who posit these constraints are vague as to whether or not they are presumed to be unlearned. There is a strong argument, however, that the only claim consistent with the idea of such constraints is that at least some of them are present prior to the onset of word learning. The motivation for positing constraints in the first place is to explain how children solve the induction problem and learn words. From this perspective, it would be contradictory to claim that (for example) children learn that words describe kinds of whole objects, as this would requir~ that they first learn the meaningsof some words and then notice that they tend to refer to whole objects. This would imply that children are able to acquire words without this constraint, and thus one could not appeal to its existenceas an explanation of how children initially solve the word learning problem. Consider also the specific proposal that children induce that members of certain grammatical categories tend to share certain meaningful properties, and that this is the origin of some of the constraints. Landau et ale ( 1988: 317) suggest that ' the development of a same-shape preference in children may originate in language learning, specifically in the process of learning count nouns. Many of the words acquired by early language learners do in

P. Bloom I Possiblenames

fact partition the world according to the shapes of the objects in it ... young children very quickly realize this , abstracting a rule from their early word learning experience that says shape is the critical factor in decisions about the extensions of these nouns. Then they use this rule when encountering new nouns and new classes, thus immensely simplifying and ' speeding up the mapping of the one onto the other . While possible, this assumes that children are able to learn the correct meanings of count nouns (and can thus infer that they tend to refer to shape) prior to the onset of the shape bias. Thus although this bias might facilitate word learning later on , we are still left with the problem of what constrains children ' s inferences in the first place. If the constraints are unlearned, what is their precise nature? Do they constitute a subpart of a distinct language acquisition mechanism that exists solely to facilitate the learning of words? This is certainly conceivable, but it would be preferable to motivate these constraints on word learning in terms ' of more general properties of children s linguistic and cognitive competence. One specific proposal, advanced below, is that theseconstraints emergefrom ' ' other properties of children s knowledge; in particular , from children s grasp of syntax-semanticsmappings.

3. Syntax - semantics mappings as a theory of constraints on word learning

Below, I discusscertain mappings betweensyntax and semanticsand argue that they effectively subsumethe constraints reviewedabove. These mappings have two advantagesover the constraints: (i ) they have independentlinguistic and psychological support and (ii ) they provide a framework to explain the ' ' ' ' ' ' acquisition of all nominals, including those such as dog , water , wood , ' forest' and ' Fred' and also , , including those count nouns that are not names for material entities.

3.1. Lexical categoriesvs. phrases Markman ( 1990) briefly discusses a theory of the origin of the taxonomic constraint, which is based on Di Sciullo and Williams ( 1987) and states that ' ' words are generic in meaning in a way that phrasesare not ; words do not make reference to specific things, times, or places. She suggeststhat the taxonomic constraint may be a consequenceof this property of words. This insight seemsfundamentally correct, but the explanation needsto be modified

P. Bloom I Possiblenames

307

in order to account for words like ' Fred' and ' she' which refer to individuals and are not ' generic' in the sensediscussedby Di Sciullo and Williams . An alternative is that the relevant distinction is grammatical, and related to how languages use syntactic categories to express meaningful notions and relations ( Bloom 1990a, Jackendoff 1983) . In particular , it is not words that have generic reference- it is categoriessuch as nouns and verbs. Nouns like ' ' ' ' dog and water are generic in the sensethat they can be extended to an indefinite number of novel instances(an infinity of different dogs, an infinity of different portions of water). Put differently, they refer to kinds, not to particular individuals or entities. In contrast , noun phrases (NPs) like ' the big dog' can be conceptualized as referring to individuals , and not to kinds. The standard examples of this are when nouns combine with quantifiers to become NPs. Thus ' a dog' can ' ' pick out a particular individual that happens to be a dog, those big dogs picks out those dogs that have the property of being big , and so on. (See Parsons, 1990, for a discussion of how a similar analysis can apply to verbs and V Ps.) The distinction betweenwords and nouns is crucial here, since some words are NPs, not nouns. This allows us to explain the peculiar status of pronouns and proper names. From the standpoint of grammar, they are lexical NPs (seeBloom 1990b). With regard to their role in syntactic structure, words like ' "Fred' and "he' behave like ' phrasal NPs such as the dog and thus cannot appear with adjectivesand determiners. Under the hypothesis that pronouns and proper namesare NPs, we can posit the following mappings: Mapping 1.. NPs refer to individuals. Mapping 2.. Count nouns and mass nouns refer to kinds. It is likely , however that Mapping I is too strong; there are NPs that appear not to refer at all. In a language like English, these include expletive ' ' ' ' ' ' ' pronouns, such as it as in it is raining or there as in there is trouble ' , One brewing theory of why such semantically empty NPs exist is becauseof a requirement in English that all tensed sentencesmust have overt subjects (Chomsky 1981), even if thesesubjectsplay no meaningful role. In languages like Italian , where overt subjectsare not necessary , one can say the equivalent of ' is raining' , while in English, it is necessaryto add the meaninglessNP ' it ' in order to satisfy the grammatical requirement (but see Bolinger, 1973, for

P. Bloom / Possiblenames

evidence that even expletives have some semantic content) . This class of counterexamples(and many others; see Bloom, under review) suggeststhat the mapping from NPs to individuals does not hold in all cases. These exceptions must somehow be learnable by children through something other than the mapping posited above. One possibility, first advanced by Nishigauchi and Roeper ( 1987), is that children must initially acquirea given NP by using the syntax-semanticsmapping ' ' (e.g., referential it ), and only after having done so, can they understand the sameword or string of words in a semanticallyempty context (e.g., expletive' it ' ). This makesthe prediction that acrossdifferent languages , all words and phrases that are NP expletivesmust also be NPs of the semanticallywell-behavedtype ( NPsthat refer), becauseotherwisechildren would not be able to acquire them. There is someevidencethat this is the case( Nishigauchiand Roeper 1987), and one could make the further prediction that children can only categorizea string of words within an idiom as an NP (e.g., ' the bucket' as in ' kick the bucket') if they are already capable of construing that string of words as having some referentialmeaningwhen it is outside of the idiom. A theory along theselines is discussedin Bloom (under review); for the purposeshere, it should be noted that a completeaccountof lexical acquisitionmust explain how children acquirethese sorts of non-referentialNPs.2 3 .2 . Count nouns VS. mass nouns

If we restrict their domain to count nouns, we can collapse the whole object constraint, the taxonomic constraint , and the shapebias as follows : Count nouns refer to kinds of whole objects (Markman and Hutchinson 1984) and children are biased to extend them on the basisof shape( Landau

et ale1988 ). This generalization connects with a sizable literature that attempts to discover the semantic basis of the grammatical count/ mass distinction (e.g., 2 A different sort of puzzle concernsNPs that apparently refer to kinds, as with the subject NPs ' ' ' ' ' ' ' ' dogs and water in the sentencesDogs are friendly animals and Water is good to drink . These NPs can be construed as referring not to actual dogs or actual portions of water, but to abstract kinds - to the speciesDOG and the substanceWATER . Under one analysis, they serveas proper namesfor these kinds, and thus refer to kinds in a very different way than do nouns (Carlson 1977). In any case, the acquisition and representationof theseNPs constitute a further domain of study from the standpoint of syntax-semanticsmappings.

P. Bloom / Possiblenames

309

Bach 1986, Bloom I 990a, I 994a; Bloomfield 1933, Gordon 1985, 1988; Jackendoff 1991, Langacker 1987, Levy 1988, McCawley 1975, McPherson 1991, Mufwene 1984, Quine 1960, Ware 1979, Weinrich 1966, Whorf 1956, Wierzbicka 1985, seeGathercole 1986for a review) . In general, it is clear that ' ' ' ' objects tend to be described by count nouns ( a dog , five tables ) and nonsolid ' substancestend to be described by mass nouns ( much water' , ' less ' sand ). This holds for languagesother than English; acrossdifferent language families, names for entities like dogs and tables are always count nouns, and names for entities like water and sand are always mass nouns (Markman 1985, M ufwene 1984, Wierzbicka 1985). This pattern is unlikely to be a coincidence, and it might lead one to the hypothesis that entities describedby count nouns have perceptually salient boundaries, and thus are countable, while mass nouns describeeverything else. Nevertheless, the sameobjections discussedearlier against the whole object constraint have also beenmade against this theory of count/ masssyntax (e.g., by Ware 1979) . One objection is that many count nouns do not describe ' ' whole objects; there exist abstract words like ' nap and 'joke , as well as collective nouns like ' forest' and ' anny ' . If one is to retain the notion that the grammatical count/ mass contrast maps systematically onto some cognitive division , the cognition side of the mapping must be considerably more ' ' abstract than ' whole object and ' non-solid substance. As a result of theseconsiderations, many scholars have proposed that the grammatical count/ massdistinction maps onto the semanticcontrast between nouns that refer to kinds of individuals vs. those that refer to kinds of nonindividuated ' entities, which we can view as ' portions (seeBach 1986, Bloom 1990a, 1994a; Bloom and Keleman, under review; Gordon 1982, 1985, 1988; Jackendoff 1983, 1990, 1991; Langacker 1987, 1990; Macnamara 1986, Macnamara and Reyes 1990). The cognitive notion of inviduals is related to , and is roughly properties such as countability , indivisibility , and boundedness ' . Within the domain of material entities ' discretebounded to , entity equivalent this usually corresponds to whole objects, though it can also correspond to entities such as a forest, a puddle, and a pile. Outside of the domain of material ' ' ' ' entities, an event that takes a bounded interval of time ( a race , a conference) ' ' ' can be construed as an individual , as can a mental state ( a headache, a ' ' ' ' ' nightmare ) or a period of time ( a day , an hour ). (Some speculationsabout ' ' the precisenotion of individual will be discussedin section 6.2.) What support is there for the claim that count/ mass syntax actually corresponds to these aspectsof abstract cognition ? One source of support is from linguistic analyses (e.g., Barwise and Cooper 1981, Jackendoff 1991,

310

P. Bloom I Possiblenames

Langacker 1987). Unless one assumesthat there is some consistent semantic property holding acrossboth material count nouns and abstract count nouns, it is difficult to provide a consistent theory of quantification. In many ' ' important regards, an NP such as a dog is semanticallyequivalent to an NP ' ' suchas a nightmare , and one way to capture this is by describing' a' as having the semanticrole of combining with nouns that refer to kinds of individuals to form an NP that can denote a single individual. Nouns such as ' dog' and ' ' ' ' ' ' nightmare - but not nouns such as water and advice - refer to kinds of individuals and, thus can be usedwith count noun syntax. There also exists empirical evidenceconcerning the productive use of these syntax-semanticsmappings in adults. In one study, adults were taught novel words referring to sensationsor sounds(Bloom 1994a). The syntax of the word was kept neutral with regard to its count/ massstatus; what varied was whether the new wor~ was describedas referring to something that occurs in discrete units of time (temporal individuals) or to somethingoccurring over continuous periods of time (temporal portions). This had the predicted effect on adult categorizationof the new word: Names for temporal individuals tended to be categorizedas count nouns, while names for temporal portions tended to be categorizedas massnouns. We can now posit the following three mappings:

Mapping1: NPs refer to individuals Mapping2: Count nounsrefer to kinds of individuals Mapping3: Massnounsrefer to kinds of portions Before turning to the question of precisely how these three mappings enablechildren to acquire new words - and how they fare relative to the sorts of hypothesesadvancedby Markman and others - it is necessaryto consider whether young children actually possessthis understanding of the relationship between syntax and semantics.

4. Syntax-semanticsmappingsin young children 4.1. Argumentsagainst early competence What is the evidence regarding children' s knowledge? With regard to the count/ mass distinction , it is often argued that children are insensitive to the

P. Bloom / Possiblenames

semanticbasis of this distinction until a relatively late age. For instance, Levy ( 1988: 186) reviews the work of Gordon and Gathercole and concludes as follows : ' Thus Gathercole' s conclusions are in , complete agreement with the conclusions reached by Gordon ( 1985); namely, that children first learn the linguistic distinction as a morphosyntactic ' rather than a semantic distinction .

Others have reached similar conclusions . Thus Schlesinger ( 1988: 147), in his discussion of domains in which his semantic assimilation theory does not apply , states: ' ... Gordon 1985 and Gathercole 1985 have shown that the count- massdistinction is ( ) ( ) distinction. The acquiredthroughformal cluesrather than via the semanticobject-substance correlationbetweenthesetwo reasonseems to be that, in English,thereis not a veryconsistent .' distinctions

' These findings have been taken as strong evidence for a distributional ' of theory language development. Levy ( 1988) argues, following Karmiloff Smith ( 1979), that children view the acquisition of grammar as a formal ' ' puzzle, a problem space per se , and semantics is irr .elevant. This is also ' Gathercole s ( 1985) conclusion, but Gordon ( 1985, 1988) proposes a quite different view, maintaining that count/ masssyntax is basedon quantificational semantics from the very start. What young children lack, according to Gordon , is knowledge of how this semantic contrast maps onto perception. That is, they understand that the contrast between count nouns and mass nouns corresponds to the distinction between words that refer to kinds of individuals vs. words that refer to kinds of portions - but they lack an understanding that physical objects are canonical individuals and non-solid substancesare canonical portions . If this were correct, then these mappings would be uselessas a source of constraint in word learning. The specific studies of Gathercole and Gordon are critically reviewed in considerabledetail in Bloom ( 1990a) ; it will suffice here to raise a conceptual ' s understanding point . All of the experiments purportedly showing that children ' of count/ mass is not semantic involve studying children s sensitivity to linguistic cues. Thus one finding is that if 3- and 4- year-olds hear (e.g.,) ' ' ' This is a blicket ' they tend to grammatically categorize blicket as a count ' ' noun, while if they hear This is some blicket , they tend to grammatically ' ' syntactic categorize blicket as a mass noun. Further , children will give these ' ' " cues priority over referential cues. If they hear This is a blicket they are

312

P. Bloom I Possiblenames

likely to interpret the word as a count noun regardlessof whether they are being shown an object or a substance(Gordon 1985) . One interpretatio~ of this result is that children' s understanding of count/ mass is not semantic. Instead children possesssome generalization of the form : ' Everything following the word " a" is a count noun' , and this is distinct from any semantic understanding, which has to be learned at some later point . This assumesa dichotomy between' linguistic cues' and ' semantic cues', where the latter is restricted to information that children receive through perception of the external world . An alternative, however, is to reject this dichotomy altogether. Semantic information can be conveyed through ' ' ' ' language; when children hear a blicket and categorize blicket as a count noun, they may be drawing a semanticinference. Specifically, children might encodethe determiner ' a' as having the semanticpotential of interacting with a noun that 'refers to a kind of individual to pick out a single individual becausethis is what it means- and it follows from this that any noun that follows ' a' must refer to a kind of individual and thus must be a count noun. In fact, linguistic cues are a more reliable cue to the semantic' status of a novel word than perceptual cues are. This is becausea given percept can be construed in different ways; if a solid object is describedas ' blicket' , it is quite ' ' possiblethat blicket is actually a massnoun, becauseit could refer to the stuff that the object is made out of. But linguistic cuesare flawless; every noun that can co-occur with a quantifier that has the semanticrole of individuation has to be a count noun. Given this, the child' s early sensitivity to linguistic information actually supports a semantictheory; it does not refute it . 4.2. Argumentsfor

early competence

What positive evidence exists that children possessthe requisite syntaxsemantics mappings? Gordon ( 1982t 1985t 1988) provides a learnability argument: Children must be capable of using semantic information when acquiring the grammatical count/ mass distinctiont becausea non-semantic distributional analysis would have to sift through several billion possibilitiest and children have productive command of count/ mass syntax by about the age of 2-and-a-half. More generallyt the argument is that some semantic categorization would have to be done by children in order for them to so rapidly converge on the correct linguistic generalizationst becauseno other source of information servesto distinguish count nouns from mass nouns in the input they receive(seealso Bloom 1994a) .

P. Bloom / Possiblenames

Children' s errors provide further evidencethat they are exploiting syntaxsemanticsmappings. Words that are mass nouns in English but which refer to entities that can be construed as discrete objects, like ' money' , ' furniture ' , and ' bacon' are , occasionally misencodedas count nouns, e.g., young children will sometimes say things such as ' a money' (Bloom 1994a). These errors are ' ' significantly more frequent than errors with more canonical mass nouns, ' ' ' ' such as water and milk , which refer to substances . This suggeststhat the of new words as either count or mass is facilitated (and categorization sometimeshampered by) children' s use of mappings from syntax toseman tics. A further source of support is experimental. In a classic study, Brown 1957 ( ) showed 3- to 5-year-olds sets of pictures, one that depicted an object, another that depicted a substance, and told them to either ' show me a sib' or ~show me sib' . Children tended to point differently as a function of the syntax; when given a count noun they would tend to point to the object; when given a mass noun, they would tend to point to the substance. More recently, Sola ( 1992) found a sensi~ivity to syntax-semantics mappings as soon as children start to productively use count/ mass syntax in their spontaneous speech. When these 2-year-olds are taught a mass noun that describesa pile of stuff, they tend to construe it as a name for that kind of stuff (i.e., as having a similar meaning to ' clay' ), but when taught a count noun that describesa pile of stuff, many appear to construe it as referring not to the stuff itself, but to the bounded pile (i.e., as having a similar meaning to words like ' puddle' or ' pile' ) . Interestingly, this effect of syntax was limited to the stuff-condition : when children were taught count nouns and mass nouns describing a novel object, few of the children construed the mass noun as referring to the stuff that the object was made of (i.e., they would not construe it as having the same meaning as words like ' wood' or ' metal' ) . Regardlessof the count/ massstatus of the noun, they would tend to interpret it as a name for that kind of object. An explanation for this asymmetry will be discussedin section 5. There is lessevidencethat young children can extend the semantic implications of count/ mass syntax to non-material entities, but there is one relevant study (Bloom 1994a) . Here, 3- and 4- year-olds were taught names for perceptually ambiguous stimuli , which could be construed as either a set of individuals or as an unindividuated portion . In one condition , the stimulus was food , either lentils or colored pieces of spaghetti, and was the sort of entity that could be easily named with either a count noun or a massnoun. In another condition , the stimulus was a string of bell sounds from a tape-

314

P. Bloom I Possiblenames

recorder, presented one after the other at a very fast rate, which could be construed either as a set of discrete bells or as an undifferentiated noise and therefore could also be describedwith either a count noun or a mass noun. ' ' ' ' All children were presentedwith both the food and the bell stimuli. One ' ' group was told : These are feps there really are a lot of feps here (count ' noun condition ); the other group was told : this is fep - there really is a lot of ' were taught the word fep here (massnoun condition ) . Then the children who ' ' as a count noun were told to give the puppet a fep for the food condition ' and, in the sound condition , were given a stick and a bell and asked to make ' a fep . The children who were taught the word as a mass noun were told to ' ' give the puppet fep in the food condition or , in the sound condition , to ' ' make fep with the stick and the bell. If children are sensitiveto the semantic in the count properties of count/ mass syntax, they should act differently ' ' condition than in the mass condition . When asked for a fep , they should ' ' tend to give one object or make one sound, and when asked for fep , they should tend to give a handful of objects or make a lot of sounds. These were the results obtained: Both 3- and 4-year-olds performed significantly above chance on both the food and sound conditions. This finding provides further support for the hypothesis that there is a semantic basisto count/ masssyntax evenfor non-material entities, and indicates that an understanding of mappings betweensyntax and semanticsis present in 3 and 4- year olds. Sensitivity to the semanticsof the noun/ NP contrast is evident at an even earlier age than is an understanding of the semantic basis of count/ mass. The mapping hypothesis is that young children should understand that the grammatical contrast betweenwords that are nouns and words that are NPs correspondsto the contrast betweenwords that refer to kinds and words that refer to individuals. In a classic study by Katz et al. ( 1974), the experimenter object and saying either taught young children new words by pointing to an ' ' count noun context or ' This is ' This is a wug ( NP context). Even some ) wug ( to this 17 months olds were sensitive grammatical difference; when the word was presentedas a noun they tended to construe the word as the name for a kind , but when it was presented as an NP , they .tended to construe it as a name for a particular individual (see Gelman and Taylor , 1984, for a replication with slightly older children) . The findings that children younger than two treat nouns and NPs differently with regard to how they interact with determiners and adjectives, and that they categorize pronouns and proper namesas NPs (Bloom 1990b) constitute further evidencethat children possesssome grasp of syntax-semanticsmappings.

P. Bloom / Possiblenames

315

S. Subsuming the co. - traints

At this point , we can return to the constraints discussedin section I and examine the extent to which they can be replaced by the hypothesizedsyntaxsemanticsmappings. The taxonomic constraint is subsumed by Mappings 2 and 3, which specify that both count nouns and mass nouns refer to ' kinds' and thus have generic referencein the sensediscussedby Markman . Proper names and pronouns, as they fall into the grammatical class of NPs, are in the domain of Mapping I and thus refer to particular individuals. The whole object constraint - that novel words refer to whole objects- can be argued to result from the interaction betweenMapping 2 (which statesthat count nouns refer to kinds of individuals) and a non-linguistic bias to construe discretephysical objects (DPOs) as individuals (Shipley and Shepper son1990). That is, independentof grammar, children are very strongly biased to encode objects as individuals. When shown a display of objects and asked to count kinds (' How many kinds of animals?' ) or properties (' How many colors?' ), they will show a strong tendency to ignore the question and to only count the discrete objects (Shipley and Shepperson 1990). Note that children are quite able to count individuals that are not objects, such as sounds and actions (Wynn 1990). What the Shipley and Shepperson study shows, however, is that in the material domain, objects stand out as individuals. This is apparently a cognitive phenomenon, having to do with the nature of the category ' individual ' ; it does not directly result from the syntax-semantics mappings. But when the DPO interacts with Mapping 2, it leads to a tendency to construe count nouns as names for whole objects. This relates to ' young children s unwillingness to initially interpret a noun, either count or mass, that refers to a novel bounded entity as naming the stuff that the entity is made of (e.g., Dickinson 1988) . In other words, the DPO causeschildren to strongly favor interpreting a word that describes an object as having a ' ' ' ' meaning like desk (and not wood ), even if the word is presentedwith mass noun syntax. Although children show some evidence for a parallel bias to construe words describing substancesas kinds of portions, this is not as powerful and will not override syntactic cues to the contrary . Thus 2-yearolds are quite willing to interpret a count noun that describesa substanceas ' ' referring to that kind of bounded entity (as having a meaning like puddle ), even though their default assumption is that a word describing a substance will name the kind of portion (as having a meaning like ' water' ) (Sola 1992) . The shape bias posited by Landau et al. ( 1988) can be recast as a claim about how children think about objects; at least for some sorts of objects,

316

P. Bloom I Possiblenames

they tend to initially view the appropriate dimensionfor generalizationsas being madeon the basisof shape, and this is viewedas a better basisfor extendingthe noun usagethan propertiessuch as color, size, or texture. (Recall that this does not apply for namesfor substances , where propertiessuch as texture and color are more relevant; Sola et al. 1991.) The appealto count nouns is relevantonly insofar as count nouns are the only linguistic category that specificallypick out whole objectsin the material domain, and thus any bias to favor shapeis most likely to apply for this set of nouns. But the nature of the bias has to do with children's understandingof object kinds, not of count nouns.3 Somesupport for this interpretation comesfrom the finding that the bias towards shapeappearsto ' shift in the course of development, presumably as the result of the child s expandingunderstandingof how different categoriesof objectsmight begener . Macario 1991 alized in different ways (Beckerand Ward 1991, ) Finally , consider the proceduresof Sola et al. ( 1991) . A solid object is likely to be construed as an individual , and thus a noun (but not an NP , adjective, or verb) that describessuch an object is likely to be a count noun and refer to that kind of object. Similarly, a substance is likely to be construed as a be a portion and thus any noun that describessuch a substanceis likely to ' ' massnoun and refer to that kind of portion . Exceptional casessuch as wood ' and ' puddle are fully consistent with Mappings 2 and 3, but , as discussed above, the mapping from mass nouns to kinds of portions may be difficult , as it for children to exploit in caseswhere the portions are solid substances Put as individuals. runs afoul of the general cognitive bias to treat objects ' ' differently, to learn a word like wood the child must construe an object as a unit of stuff, rather than as a single individual , and this violates the bias to construe discrete physical objects as individuals.

~ 6. Openquestio 6.1. Is there developmental change ? The claim above is that 1- and 2- year - olds possess unlearned

mappIngs from count nounsto individualsand massnounsto non-individuatedentities, 3 Landau et al. ( 1988) argue that children' s different responseson linguistic and non-linguistic tasks (they treat shape as more relevant for the former) suggeststhat the shape bias is special to language. As suggestedabove, however, an alternative is that the use of the count noun informs the children that the task has to do with objects (where shapeis very relevant); when the noun is not present the children might just as well assumethat the task has to do with properties (where

P. Bloom / Possiblenames

317

as well as the bias to construe whole objects as individuals. Is it necessaryto posit this sort of abstract knowledge, or might children start off with a simpler representation, linking up the count/ mass distinction directly to the contrast between bounded and unbounded physical entities, and only later developing the more abstract adult understanding? It appears that even 2' ' year-olds possessa cognitive understanding of count noun that includes bounded substancesand is thus not limited to the category of ' whole object' (Sola 1992), but the question remains of whether this understanding is initially restricted to the material domain, or whether it can extend to sounds, events, collections, and so on. In the absenceof any evidencefor developmental change, one might argue that lack of a child - adult difference should be viewed as the null hypothesis in psychology (Fodor 1975, Macnamara 1982) - an argument that gains force from the fact that we have as yet no understanding of how a cognitive notion can become ' more abstract' . But casesof representational change do appear to exist (seeCarey 1986, 1988), and so it remains an open question whether syntax-semantics mappings are yet another domain where children differ from adults. No decisive evidence exists at this point , but there are three sources of evidence suggesting that the abstract adult-like understanding is present in very young children. First , some evidenceconcerning early possessionof the notion of ,oindividual' emergesfrom the researchof Starkey et al. ( 1990), who found that 6- to 8-month-olds possessa unified concept containing both whole objects and temporally bounded sounds. In one study, infants were exposedto either two sounds or three sounds. Immediately following this, two pictures were simultaneously shown to the infants, one with two objects and one with three objects. The subjects tended to look longer at the picture which showed the same number of objects as there were sounds, providing some evidencethat infants possessnotions of ' two individuals ' and of ' three individuals' , where ' individual ' es both sounds and objects. Along the same lines, encompass 1990 discovered that almost immediately after children are able to ) Wynn ( use the linguistic counting system to count objects, they can also use it to count sounds and events. These studies suggestthat , quite independently of ' ' syntax, children do havethe appropriateabstractsemanticnotion of individual .

318

P. Bloom I Possiblenames

Second, as noted above, children appear to be capable, even at a very early to non-material age, of productively and appropriately using words that refer ' ' events ' bath' ' ' ' ' minute entities, such as temporal intervals ( day , , nap ) and ( ) ' ' ' ' abstract entities ( story , joke ) . If it turns out that (i ) they encode these ' ' words as falling into the grammatical category of count noun and (ii ) they understand them in the same way that adults do, then this would show that the abstract understanding of the count/ massdistinction is present in 2-yearolds. But there is no strong support at present for either of these claims, at least not for children younger than three. Finally , there is the experimental study noted above (Bloom 1994a), which showed that 3- and 4- year-olds are sensitiveto the application of quantificational syntax in a domain of non-material entities (sounds) . Once again, however, evidence for this sort of capacity on the part of younger children does not yet exist. At this point , then, it remains possible that the semantic ' ' category individual emergesfrom some sort of simpler representation, such ' ' as bounded physical entity . ' ' 6.2. What is the nature of kind of individual ?

This brings us to the second concern. Without a substantive theory of the ' ' precisenature of kind of individual , the sort of account proposed here runs the risk of being empty. Despite the central role of this notion in semantic theories of quantification and reference, we have as yet little understanding of how it links up with perception and non-linguistic cognition , and how it servesto constrain the extent of possible word meanings. If we assumethat count nouns refer to kinds of individuals , it is apparent that the reference of such nouns includes bounded substances , periods of social constructs abstract and of collections states mental events time, , , objects, have in referents all of these . There are severalhypothesesabout what common, and thus what the core of this semanticnotion is. Some suggestions include boundedness, having a single functional role, and proximity or connectednessof parts (see e.g., Hirsch 1982, Jackendoff 1991, Langacker 1987, 1990) . ' ' Consider, for instance, count nouns that name collections, such as forest , ' ' ' ' family and army . Although they describe material entities, they violate the generalization that a count noun refers to a kind of object. As such, they are problematic for theories that posit a privileged link between words (or count nouns) and kinds of whole objects. The premise of the mapping theory .sketched out above is that although there exist semantic constraints on what

P. Bloom/ Possible names

319

can be a possiblecount noun, theseare basedon the semanticcategory ' kind of individual ' , not ' kind of object' . One could thus describethe acquisition of words such as ' forest' by assumingthat a collection of trees is construable by children as being a single individual and therefore a word that refers to that kind of individual is learnable as a count noun. But what is it about forests that makes them construable by children and adults as individuals? Why do children readily construe a group of trees as a possible individual (' a forest' ), and yet do not construe all of the leavesof a tree as a single individual (see Chomsky and Walker 1978) ? One tentative proposal is as follows : Hypothesisabout possibleindividuals: Something (e.g., an object or set of objects) can be encoded as an ' individual ' if we can construe it as playing an independent causal role in some conceptual domain. The intuition underlying this is as Folio~ s: We view something as an individual only if doing so allows us to better understand and predict the causal relationships that hold within a given conceptual domain . The strongest example is the case of bounded objects; these are the canonical ' ' example of individuals within the physical domain and are highly privileged in .the course of development. Infants are predisposed to analyze their chaotic sensory input into a world of distinct bounded objects that persist over time and space (e.g. Spelke 1988) . This mode of interpretation is likely to have evolved becauseconstruing the environment in these tenDS is the best way to make senseof what is going on , and allows for the most predictive power . Any primate that lacked this conceptual scheme would fail to respond to the world in a timely and effective manner - it would not be able to track prey and avoid predators, for example - and would not survive. The bias to construe objects as individuals is the result of evolution, not learning, and this may hold as well for individuation within other domains, such as social cognition or naive theories of mind. For instance, humans and other primates might be predisposed to classify certain social groups as individuals for the purpose of inference and prediction ; as such, notions like ' ' ' ' family and social group might be innate (Hirschfeld 1987, Jackendoff 1992). But for most domains, we have to discover the relevant individuals in the course of understanding the nature of the domain. One would have a difficult time understanding geography or politics , for instance, without the

320

P. Bloom I Possiblenames

' ' ' ' ability to understandentities like Canada and France as discreteindividuals that causally interact with each other. ' Returning to names for collections, it is clear that one s conceptual framework also determines the specific conditions under which a set of objects can be construed as forming a collection. The adult understanding of ' forest' is of a group of trees growing together in a certain environment - for the word to apply, there have to be a sufficient number of treesand they must be bunched together, but they need not form a precise shape. Nouns such as ' ' family have an even looser spatial restriction: they can apply even if the elementsthat make up the collection bear no non-trivial spatial relationship ' at all ; it is sensibleto say ' that family is scatteredaround the world (compare ' ' the semantic oddnessof that forest is scatteredaround the world .) But now ' consider Donald Judd' s sculpture ' untitled ( 1928), composed of 10 Plexiglas piecesmounted vertically on a wall , separated from one another by exactly the same distance. In this case, the precise configuration does matter ; the intuition would be that two rows of five Plexiglas piecesstacked on a table would be different from ' untitled ' and Judd might reasonably view this modification as a destruction of his artwork . If the notion of ' individual ' can be related to notions of intentionality and social interaction, it follows that, in principle, any set of objects can be construedas a singleindividual and thus namedwith a count noun. There is no count noun in English referring to a single shoe and a single glove (i.e., such ' that exactly one shoeand one glove would be ' one fizzbit ) but if such a pair of objects was exactly what one neededin order to participate in some kind of religious ceremony, such a name would probably be learnable by someone trying to make senseof that ceremony. One intriguing example of this again ' involves artwork ; a proper name (e.g., 'January Angst ) might describe six concrete columns surrounded by broken glass. Although this set of objects is ' an individual solely by virtue of the artist s intention, this fact is sufficient for adults (and possibly children) to acquire and understand this new name. Current research addresses these issues by exploring the circumstances under which people will give a collective interpretation for a set of discrete objects. One methodology used is to show adults a set of four identical ' ' ' objects and tell them either this is a fendIe (count syntax) or the new ' word for this is : fendle (neutral syntax) . Thus the novel word could either be a collective noun , like ' forest ' , and refer to all four objects, or it could be an object name, like ' tree' , and refer to a single object . Then the adults are shown other displays with the same kind of objects used in the training phrase, such as a display with one object and a display with eight

P. Bloom / Possiblenames

objects, and asked to describe these using the new word . Their responses indicate whether they think the word refers to a collection or to an individual ' ' object. For instance, if they interpret fendle as a collective noun, they should ' describe eight objects as either one fendle' or ' two fendles' (depending on how crucial numerosity is in their collective interpretation) and should describeone object as ' part of a fendle' . If they interpret it as an object name, ' ' ' they should describe eight objects as eight fendles and one object as one ' fendle . In a pilot study with a group of 36 adults, we tested the effects of syntax ' ' (singular count vs. neutral) and intentional integrity on their interpretation of novel words. This second manipulation went as follows : For half of the trials, the objects were placed in front of the subject slowly and carefully ; for the other half , the objects were casually dumped in front of the subject. The prediction was that this simple manipulation would have an effect on whether the subjects construe the novel word as a collective noun. The mere act of purposefully and intentionally setting out the stimuli in a given configuration should be sufficient to emphasizeto the subject that the set itself is relevant as a single individual . This prediction was confirmed: When the novel word was presentedas a singular count noun, there was a bias towards interpreting it as referring to the entire collection. The bias increased(though not significantly so) when the objects were placed in front of the subject, rather than dumped. When the novel word was presentedwithout syntactic support, however, there was an actual switch in the favored interpretation : In the dumping condition , almost all subjectsconstrued the word as a name for a single object (like ' tree' ), but for the placing condition , the collection interpretation (like ' forest' ) was strongly favored. Further research will explore whether young children can also acquire a collective noun through this sort of intentional cue, and will also focus on the ' ' precisenature of adults and children s construal of the new word. The hope is that by studying the acquisition and understanding of nominals referring to sets of discrete objects - including collective nouns and names for artwork we will gain some insight into the nature and development of the notion ' individual ' and how it relates to cognition and perception. 6.3. How can thesemappingsapply prior to the acquisition % vert syntax? The proposal here is that the constraints children use when acquiring words are the result of their understanding of the mappings from syntactic

322

P. Bloom I Possiblenames

categoriesto categoriesof cognition . But productive command of count/ mass syntax comes in at roughly the age of 2-and-a-half and children have begun to learn words over a year prior to this. To make matters worse, there exist languageswhich do not appear to exploit the count/ massdistinction in either syntax or morphology (e.g., Mandarin Chinese), and yet children acquiring theselanguageshave no difficulty learning words. Also , many scholars have argued that children use properties of word meaning to determine the syntactic categories that their very first words belong to ( Bloom 1990a, 1994a, under review; Grimshaw 1981, Macnamara 1982, 1986; Pinker 1984), a proposal that has been dubbed ' semantic ' bootstrapping . Thus these syntax-semantics mappings might work in both directions, to facilitate both lexical and syntactic development. But if this is correct, then very young children must have the capacity to learn at least some aspectsof word meaning in the absenceof syntax. Finally , we know that syntax is not essential for adults. One can learn a word like ' pencil' perfectly well without hearing it used with count syntax ' ' ' ' (e.g., without it being preceded by a quantifier like a or many ) . This proves that overt syntax cannot be essential for word learning . Nevertheles, even where there is no specific syntactic information , the existenceof the mappings still servesto narrow down the possible construals of what a word can mean. This is becauseany novel word must belong to a syntactic category, and as such it must fall into one of a limited set of semantic classes. For instance, there are no words that refer to chains of thematically-related entities - no natural language could have a word that refers both to dogs and to everything that dogs standardly interact with - and the reason for this could be that there is no syntactic category that encodes such a notion . More generally, there may be no constraints on word meaning per se; there might only be constraints on possiblecount nouns, possiblemass nouns, possible intransitive verbs, and so on. But since any word has to be either a count noun or a mass noun or an intransitive verb and so on, any word is thereby constrained as to what it can mean. For example, Mappings 2 and 3 limit possible word meanings even before children can distinguish the grammatical markings of count nouns from those of mass nouns, becauseonce children know that the word is a noun, they know it must be either count or mass, and the mappings limit its meaning to two semantic classes- either it refers to a kind of individual or it refers to a kind of portion . Of course, the mappings are lesseffective at this point , since any ambiguities (caseswhere children cannot tell whether the adult intends t .o describe an object or a substance) cannot be resolved through attention to

P. Bloom / Possiblenames

overt grammatical cues. This may be one explanation for the well-known finding that 3-year-olds show a greater sensitivity to the constraints than 2year-olds (e.g., Landau et al. 1988, Markman 1989, Nelson 1988) - children can use the constraints to their fullest only once they have some facility with the grammatical and morphological structure of their language. The view that thesemappings provide constraints on children' s word meanings evenprior to the acquisition of surfacesyntax may be counterintuitive, but there is some support for it . In an extended analysis of the production and comprehensionabilities of children in the one-word stage, Huttenlocher and ' Smiley ( 1987: 84) state, Taken as a group, the object words in the single-word period form a broad semanticclasswhich contrasts with other semanticclasses emerging at the same time. That is, the pattern of usage of object words contrasts with that of words for events ..., words for persons ..., words for ' temporary states, greetings, and negation, and so on . They go on to suggest that this early demarcation of words into these classesprovides a semantic foundation for the later acquisition of syntactic categories. A more radical interpretation is that thesechildren have already classifiedthesewords into the relevant grammatical classes , and all they have left to do is learn how (or if ) their languageexpresses theseclassesin the grammar and morphology. (Does their languagemark the contrast betweencount nouns and massnouns? Is the morphology different for verbs than for adjectives?) Once they have acquired these surface expressionsof the linguistic categories, children can use the mappings to further facilitate the acquisition of word meanings.

7. Limitationsof syntax-semantics mappings One might be tempted to view the constraints that emerge from these syntax-semanticsmappings as constituting a solution to the problem of word ' ' learning. For instance, once the child knows that dog is a count noun, she can infer that the word refers to a kind of individual , and this excludes innumerable other possible interpretations. If knowing the grammatical class that a word belongs to is tantamount to knowing what it means, then these mappings could effectively solve the induction problem. Unfortunately , there are two main reasonswhy this optimism is misguided. First , we have been assuming throughout that the child is able to infer what the novel word is being used to describe, leaving only the question of how sheencodesthe meaning and appropriately extendsit to novel instances. As Gleitman ( 1990) has argued, however, there are many caseswhere children

324

P. Bloom I Possiblenames

are exposed to words in the absenceof the entities that they describe. An adult , for example, might point to a bowl of cereal and say ' Do you want milk with that ?' even when no milk is present. Consider also words for nonmaterial entities, like ' nap' and 'joke ' , where the notion of inferring what an adult is pointing to or looking at does not apply. One could also note the successof blind children at learning words ( Landau and Gleitman 1985) in order to appreciate the mystery here. A complete theory must explain how children grasp the adult ' s intention to refer - how they somehow make the correct guessas to what adults are talking about when they use novel names (for researchalong these lines, seeBaldwin 1991) . More generally, no theory of the acquisition of words can be complete without a prior theory of how children can pick up the intended reference of language-users (Macnamara 1982) . Second, even with the aid of grammar-cognition mappings, there is still an infinity of possible meanings of the new word and children are stuck with ' ' sorting them out. The count noun dog could refer to dogs, but it could also refer to dogs and pencils, to dogs until the year 2000 and then to cats, and so on. Knowing that a given word refers to a kind of individual is only a small part of the word learning puzzle; children must also determine which kind of individual the word refers to , and it is here that the induction problem runs ' ' deep, particularly given how the notion of kind of individual interacts with conceptual systems such as social cognition (see section 6.2) . Crucially , an explanation of how children learn words involves a theory of psychologically ' ' ' ' ' possible kinds (one that includes dogs and tails , but excludes dogs and ' pencils ) . In sum, while syntax-semantics mappings may be part of the solution to how children learn new words, they are not sufficient. Not only does a complete account of the acquisition of word meaning require an explanation of how people understand the intended referenceof others, it also requires a theory of conceptual representation.

8. Concludingcomments If one were to follow the standard course in the study of word learning, ' ' ' ' and only consider the acquisition of words like dog and cup , it would be hard to empirically distinguish the claim that children possessspecial word learning constraints from the alternative that they apply syntax-semantics mappings. Both theories avoid the same hard questions how do children determine what a new word is meant to describe, and what constitutes a

P. Bloom I Possiblenames

325

psychologically possible kind - and both can capture the same simple phenomenon; if a child hears a single object describedwith a word , he or she will tend to take the word as referring to that kind of object. Within this domain, the advantage of the syntax-semantics mapping theory is solely theoretical; it only posits aspectsof children' s psychology (a mapping from count nouns to kinds of individuals and a bias to view discrete physical objects as individuals) that have independent empirical support and have been previously proposed in adults for reasonsthat have nothing to do with lexical acquisition. This is preferable to having to posit special unlearned constraints that exist solely to help children acquire words and which have no other motivation or support. The empirical differences between the two theories become more obvious when we consider words that do not describe whole objects. The constraints advancedby Ma ,rkman ( 1990), Sola et al. ( 1991), and Landau et al. ( 1988) do not apply to words like ' Fred' , ' she' , ' map' , ' foot ' , and ' forest' - and these sorts of words are present in the speech of 1- and 2-year-old children. By shifting the focus to mappings between grammar and abstract cognition, we have a framework in which to deal with the acquisition of pronouns and , words for material entities that are not proper names, words for substances whole objects (like parts and collections), and words for abstract entities. With the notable exception of research on the development of verb " meaning (e.g., Gleitman 1990, Pinker 1989) , most scholars have viewed word learning as an independent issue from the nature and development of grammatical knowledge. It is also often assumed that the theoretically interesting casesof word learning are limited to the acquisition of words for material entities, usually names for whole objects. ~ is article has presented reasons for abandoning both of these assumptions, and for exploring how ' ' mappingsbetweensyntactic categorieslike count noUJ) and abstract semantic ' ' categorieslike kind of individual facilitate the acquisition of word meaning.

References . Theprincipleof mutualexclusivityin word learning: To honor Au, T.K. and M. Glusman , 1990 - 1490 . 61, 1474 or not to honor? Child Development . Acquiringword meaningsvia linguisticcontrast. Cognitive Au, T.K. and E.M. Markman, 1987 2, 217- 236. Development . The algebraof events . Linguisticsand Philosophy9, 5- 16. Bach, E., 1986 : Form over . Prioritiesin children's expectations aboutobjectlabelreference Baldwin, D.A., 1989 - 1306 . 60, 1291 color. Child Development

326

P. Bloom I Possiblenames

' contribution Baldwin . Infants to theachievement of joint reference . ChildDevelopment , D.A., 1991 62, 875-890. Barwise . Generalized andnaturallanguage . Linguistics and , J. andR.Cooper , 1981 quantifiers 4, 159 - 219. Philosophy 's useof Becker . Children in extending newlabelsto animate , A.H. andT.B. Ward, 1991 shape : Identityversus . Cognitive 6, 3- 16. objects posturalchange Development Benedict . Earlylexicaldevelopment : Comprehension andproduction . Journalof Child , H., 1979 - 200. 6, 183 Language . Semantic stnK :tureandlanguage . Unpublished doctoraldissertation Bloom , P., 199Oa , development MIT. Bloom . Syntactic distinctions in childlanguage . Journalof ChildLanguage 17, 343 , P., 1990b 355. . Semanticcompetence Bloom, P., 1994a as an explanationfor sometransitionsin language . In: Y. Levy(ed.) Otherchildren, otherlanguages : Theoreticalissuesin language development , 41- 75. Hillsdale, NJ: Erlbaum. development Bloom, P., 1994b . R~ t controversies in the studyof language . In: M.A. Gemsbacher acquisition . SanDiego, CA: AcademicPress . (ed.), Handbo,ok of psycholinguistics -form mappingsand their role in the acquisitionof nominals Bloom, P., underreview. Meaning . Language. Bloom, P. and D . Keleman, under review. Syntactic cuesand the acquisition of collective nouns. Coanition .

. Language . New York: Holt. Bloomfield , L., 1993 . Ambientit is meaningfultoo. Journalof Linguistics9, 261- 270. BolingerD ., 1973 . Linguisticdeterminism Brown, R., 1957 and the part of speech . Journalof Abnormaland Social 55 Psychology , 1- 5. Callanan . Principlesof organizationin youngchildren's natural , M.A. and E.M. Markman, 1982 . . 53, 1093 - 1101 languagehierarchiesChild Development S. 1978 . The child as word learner . In: M. Halle J. Bresnan , , , , A. Miller (eds.), Linguistic Carey . , MA : MIT Press theoryand psychological reality, 264-293. Cambridge . Semantic : The stateof the art. In: E. Wanner , L.R. Gleitman(eds.), Carey, S., 1982 development : . The state of the art 347 389 New York: . , Language acquisition CambridgeUniversityPress . Conceptualchangein childhood. Cambridge . , MA : MIT Press Carey, S., 1986 . Conceptualdifferences betweenchildrenandadults. Mind and Language3, 167Carey, S., 1988 181. . Reference to kindsin English.Unpublisheddoctoraldissertation Carlson., G., 1977 , Department of Linguistics . Universityof Massachusetts , Amherst. . Lectureson governmentand binding. Foris: Dordrecht. , N., 1981 Chomsky . The linguisticand psycholinguistic . In: E. Walker , N. and E. Walker, 1978 Chomsky background . , MA : MIT Press , 15-26. Cambridge (ed.), Explorationsin the biologyof language in his first language . . What's in a word? On the child' s acquisitionof semantics Clark, E.V., 1973 In: T.E. Moore (ed.), Cognitivedevelopment and the acquisitionof language , 65- 110. New York: Academic . . In : B. Mac. The principleof contrast: A constrainton languageacquisition Clark, E.V., 1987 of languageacquisition , 1- 33. Hillsdale, NJ: Erlbaum. Whinney(ed.), Mechanisms . On the pragmaticsof contrast. Journalof Child Language17, 417- 432. Clark, E.V., 199Q . . On the definitionof word. Cambridge Di Sciullo, A. and E. Williams, 1987 , MA : MIT Press

P. Bloom I Possiblenames

327

. Learningnamesfor materials: Factorsconstrainingand limiting hypotheses Dickinson, D.K., 1988 aboutword meaning . CognitiveDevelopment 3, 15- 35. . The languageof thought. New York: Crowell. Fodor, J., 1975 ' : The . ' He hastoo manyhardquestions Gathercole, V.C., 1985 acquisitionof thelinguisticmass countdistinctionin muchand many. Journalof Child Language12, 395- 415. Gathercole, V.C., 1986 . Evaluatingcompetingtheorieswith child languagedata: Thecaseof the count-massdistinction. Linguisticsand Philosophy6, 151- 190. . The contrastivehypothesisfor the acquisitionof word meaning: A Gathercole, V.C., 1987 reconsideration of the theory. Journalof Child Language14, 493- 531. . How two-year-old childreninterpretproper and common Gelman, S.A. and M. Taylor, 1984 . Child Development55, 1535 namesfor unfamiliarobjects . - 1540 . Why nounsare learnedbeforeverbs: Linguisticrelativity versusnatural Gentner, D., 1982 , Vol. II : Language , thought, and partitioning. In: S.A. Kuczaj(ed.), Languagedevelopment culture, 301- 334. Hills4ale, NJ: Erlbaum. . The structuralsourcesof word meaning Gleitman, L.R., 1990 . LanguageAcquisitionI , 3- 55. . Fact, fiction, and forecast . Cambridge Goodman . , N., 1983 , MA : HarvardUniversityPress . Do linguisticdifferences lead to cognitivedifferences ? A crossGopnik, A. and S. Choi, 1990 . First Language10, 199-215. linguisticstudyof semanticand cognitivedevelopment . Theacquisitionof syntacticcategories : Thecaseof thecount/massdistinction. Gordon, P., 1982 , MIT . Unpublisheddoctoraldissertation . Evaluatingthe semanticcategorieshypothesis : The caseof the count/mass Gordon, P., 1985 distinction. Cognition20, 209- 242. . Count/masscategoryacquisition: Distributional distinctionsin children's Gordon, P., 1988 . Journalof Child Language15, 109- 128. speech . Fonn, function, and the languageacquisitiondevice . In: C.L. Baker, Grimshaw , J., 1981 J. McCarthy(eds.), The logicalproblemof languageacquisition , 183-210. Cambridge , MA : MIT Press . . Theconceptof identity. New York, NY : Oxford UniversityPress . Hirsch. E., 1982 : Thecaseof objectnames . Cognitive . Early word meanings Huttenlocher , J. and P. Smiley, 1987 Psychology19, 63- 89. . . Semantics and cognition. Cambridge Jackendoff , R., 1983 , MA : MIT Press . Semanticstructures . Cambridge . Jackendoff , R., 1990 , MA : MIT Press . Partsand boundaries . Cognition41, 9- 45. Jackendoff , R., 1991 . Is therea facultyof socialcognition? In: R. Jackendoff(ed.), Languages of Jackendoff , R., 1992 . the mind: Essayson mentalrepresentation , 69- 81. Cambridge , MA : MIT Press . A functionalapproachto language . New York: Cambridge Karmiloff-Smith, A., 1979 acquisition . UniversityPress . What' s in a name? A study of how childrenlearn Katz, N. E. Bakerand J. Macnamara , 1974 . Child Development 45, 469- 473. commonand propernames . Cambridge . . Concepts Keil, F.C., 1989 . kinds, and cognitivedevelopment , MA : MIT Press . Women, fire, and dangerousthings: What categoriesrevealabout the mind. Lakoff, G., 1987 . , IL : ChicagoUniversityPress Chicago . Cambridge . Languageand experience Landau, B. and L.R. Gleitman, 1985 , MA : Harvard . UniversityPress . 'What' . and ' where' in spatial languageand spatial Landau, B. and R. Jackendoff , 1993 16, 217- 238. cognition. Behavioraland Brain Sciences

328

P. Bloom I Possiblenames

. Perception , ontology, and namingin youngchildren: Landau, B, S. Jonesand LB . Smith, 1992 43, 85-91. . and on , , Cognition Spelke Commentary Sola Carey Landau, B., L. B. Smith and S. Jones , 1988. The importanceof shapein early lexicallearning. 3, 299- 321. CognitiveDevelopment and verbs. Language63, 53-94. . Nouns R.W. 1987 , , Langacker . Foundationof cognitivegrammar , VolumeII : Descriptiveapplication. , R.W., 1990 Langacker . UCSD , Unpublishedmanuscript : Evidencefrom studiesof the . On the early learningof grammaticalsystems Levy, Y., 1988 . Journalof Child Language15, 179- 186. acquisitionof genderand countability : Foodsandcanonicallycolored . Youngchildren's useof color in classification Macario, J., 1991 46. 6 17 . , Development objectsCognitive , MA : MIT . Namesfor things: A study of humanlearning. Cambridge Macnamara , J., 1982 . Press . Cambridge , MA : MIT . A borderdispute: Theplaceof logicin psychology Macnamara , J., 1986 Press . . The learningof propernamesandcountnouns: Foundational Macnamara , 1990 , J. and G. Reyes . . Unpublishedmanuscript and empiricalissues can be massnouns. Cognition19, 31- 53. Markman, E.M., 1985.Why superordinate categories . Cambridge of induction , and namingin children: Problems . Categorization Markman, E.M., 1989 . MA : MIT Press . CognitiveScience14, 57. Constraintschildrenplaceon word meanings Markman, E.M., 1990 77. . Children's sensitivityto constraintsin word Markman, E.M. and J.E. Hutchinson , 1984 . CognitivePsychology16, 1- 27. relations thematic : Taxonomicversus meaning . Children's useof mutualexclusivityto constrainthe Markman, E.M. and GiF. Wachtel, 1988 20, 121- 157. m~ ning of words. CognitivePsychology -massdistinction. Proceedings of thefirst annual count the and . , J., 1975Lexicography McCawley . of the BerkeleyLinguisticsSociety conference ' ' McPherson , L., 1991. A little goesa long way: Evidencefor a perceptualbasisof learningfor COUNT and MASS. Journalof Child Language18, 315- 338. the nouncategories : A . Young children learn synonyms Mervis, C.B., RiM . Golinkoff and J. Bertrand, 1991 at the BiennialMeetingfor the Societyfor . Posterpresented refutationof mutualexclusivity . in Child Development , WA , April, 1991 Research , Seattle on . Thecount/massdistinctionandthe Englishlexicon. 1984CLS Parasession Mufwene , S., 1984 . LexicalSemantics . Journalof Child Language17, , L., 1990. Childrenusesyntaxto learnverb meanings Naigles 357- 374. 3, 221- 246. ? CognitiveDevelopment . Constraintson word meaning Nelson, K., 1988 '. ' s 'Constraintsand Behrend on CognitiveDevelopment . Comment development K. 1990 Nelson, , 5, 331- 339. , , explanations . Nounsin early lexicons: Evidence Nelson, K., J. Hampsonand L.L. Shaw, 1993 61 84. 20 of Child . Journal , extensions and Language . and the growthof emptycategories . Deductiveparameters , T. and T. Roeper, 1987 Nishigauchi . 121 91 and , languageacquisition In : T. Roeper, E. Williams (eds.), Parametersetting Dordrecht: Reidel. . of English. Cambridge , MA : MIT Press . Eventsin the semantics Parsons , T., 1990

P. Bloom I Possiblenames

. Languageleamabilityand languagedevelopment . Cambridge Pinker, S., 1984 , MA : Harvard . UniversityPress . Leamabilityand cognition. Cambridge . Pinker, S., 1989 , MA : MIT Press . Learningnamesfor solid substances : Quantifyingsolid entitiesas portions. Prasada , S., 1993 8, 83- 104. CognitiveDevelopment . Words: What are they, and do animalshavethem? Cognition37, 197-212. Premack , D., 1990 . . Word and object. Cambridge , MA : MIT Press Quine, W.V.O., 1960 . Language . New York: Harcourt, Brace. Sapir, E., 1921 . The origin of relational categories . In: Y. Levy, I.M. Schlesinger , I.M., 1988 , Schlesinger M.D.S. Braine(eds.), Categories of process esin languageacquisition , 121- 178. Hillsdale, NJ: Erlbaum. . Cognition son, 1990 . Countableentities: Developmental , E.F. and BS hepper changes Shipley 34, 109- 136. . Unpublished . Ontologicalconstraints on 2-year-olds' inductionof wordmeanings Sola, N.N., 1987 doctoraldissertation , MIT . . Inferences aboutthe meaningsof nouns: The relationshipbetweenperception Sola, N.N., 1992 7, 29- 45. and syntax. CognitiveDevelopment . Ontological categoriesguide young children's , 1991 Sola, N.N., S. Carey and E.S. Spelke : Objecttermsand substance tenDs. Cognition38, 179-211. inductionsof word meaning . . Cognition N.N. S. and E.S. 1992 , , , , Perceptionontology, and word meaning Sola Carey Spelke 45, 101- 107. of objectsin . Whereperceptionendsand thinking begins: The apprehension , E.S., 1988 Spelke 197 . . A. Minnesota on Child 234 Hillsdale In: Yonas ed. , , NJ: ), ( Symposia Psychology infancy Erlbaum. . Numericalabstractionby humaninfants. Cognition , P., E.S. Spelkeand R. Gelman , 1990 Starkey 36, 97- 127. . Adjectivesand nouns: Children's strategiesfor learningnew Taylor, M. and S. Gelman, 1988 59, 411- 419. words. Child Development . In: F. Pelletier(ed.), MasstenDS: Somephilosophical Ware, R., 1979 . Somebits and pieces , 15- 29. Dordrecht: Reidel. problems : Evidence of conceptualhierarchies . S., 1990 . Linguisticblasesand the establishment Waxman 5, 123- 150. from preschoolchildren. CognitiveDevelopment ' useof . Preschoolers relationsin classifications Waxman , S. and R. Gelman, 1986 superordinate . CognitiveDevelopmentI , 139- 156. and language . Explorationsin semantictheory. In : T. Sebeok(ed.), Current trendsin Weinreich , U., 1966 , Vol 3: 395- 477. Mouton: The Hague. linguistics . . Language Whorf, B., 1956 , MA : MIT Press , thought, and reality. Cambridge . In: J. Halman(ed.), lconicity . Oatsand wheat: The fallacyof arbitrariness Wierzbicka , A., 1985 . : Benjamins in syntax, 311- 342. Amsterdam . Children's understanding of counting. Cognition36, 155- 193. Wynn, K., 1990

Section 5

The caseof verbs

-Holland - 375 . North Lingua92(1994 ) 333

When it is better to receive than to give : Syntactic and conceptual constraints on * vocabulary growth C and Lila . .. Cynthia Fisher ' , D . Geoffrey Hallb , Susan Rakowitz d Gleitman . Department , /L 61820 , USA , Champaign , University , 603E. DanielStreet of Illinois of Psychology b Medical H 4 Taviton Street London WC1 OBT Unit Research Council , , UK , Cognitive , Development CDepartment USA CT06430 , , Fairfield , , FairfieldUniversity of Psychology d Department , Philadelphia , PA , 3815WalnutStreet , University of Pennsylvania of Psychology 19104 , USA We askhow childrensolvethe mappingproblemfor verbacquistion: how they pair concepts . Thereis evidencethat nounsbut not verbs with their phonologicalrealizationsin their language ' with a can be acquiredby pairingeachsound(e.g., ' elephant ) conceptinferredfrom the world in which that soundoccurs. Verb meaningsposeproblemsfor this word-world circumstances , motivatinga modelof verbmappingmediatedby attentionto the syntactic mappingprocedure structuresin which verbsoccur (Landauand Gleitman 1985 , Gleitman 1990 ). We presentan influence the bias to interpret between a the interaction ( conceptual experimentexamining observedsituationsas involving a causal agent) and syntacticinfluences , as thesejointly . Children were shown scenes contributeto children's conjecturesabout new verb meanings or e. . and as to two , gelling chasingandfleeing) and were interpretations( g giving ambiguous askedto guessthe meaningof novel verbsusedto describethe scenes , presentedin varying . Both conceptualand syntacticconstraintsinfluencedchildren's responses , but syntacticcontexts with . This collatoral bias the overwhelmed infonnation , finding conceptual largely syntactic . evidence , supportsa syntax-mediatedprocedurefor verbacquisition

. Thanksare due to Stevenand Marcia Roth for funding that supportedthe experimental work, througha grant to Lila Gleitman. Preparationof this paperwasalsosupportedby NSF , Henry , Judy DeLoache grant DBC 9113580to C. Fisher. We wishto thank ReneeBaillargeon Gleitman, JaneGrimshaw , Ellen , AravindJoshi, MichaelKelly, BarbaraLandau, Anne Lederer Markman, Kevin Miller, Letty Naigies. ElissaNewport, Brian Ross, and SandyWaxmanfor helpfulcommentson earlierdrafts. NoamChomskyis alsothankedfor continuingcommentary on this line of work. Finally, we wishto thank the children, staff, and parentsof the SaintFaith . NurserySchoolin Philadelphiafor their cooperation . . Corresponding author. - 3841 0024 /94/$07.00 @ 1994- ElsevierScienceB.V. All rights reserved -H SSDI0024- 3841( 93) EOOSO

334

C. Fisher et at. I Syntactic constraints

1. Introduction

Every standard text in psychology (or in education or linguistics for that matter) assertsthat children aged 18 months to 6 years acquire 5 to 10 new words a day. How do they manage to do so? We concentrate here on a single aspect of the word -learning problem : Granted that children can hypothesizesome appropriate set of concepts, how do they decide which sound segmentcorresponds to each such concept? For instance, granted that they can entertain the concepts ' elephant' and ' give' , how do they come to select the sound /elephant/ for elephants and /give/ for giving ? This aspect of acquisition is called the mapping problem for the lexicon.1 Solution of the mapping problem has traditionally been assigned to a word -towor ~d pairing procedure in which the learner lines up the utterance of a word with the co-occurring extralinguistic contexts. Thus elephantcomes to mean ' elephant' just becauseit is standardly uttered by caregivers in the presenceof elephants. Gillette and Gleitman (forthcoming) have begun to document just how well the word-to -world pairing procedure works in practice for simple nouns. In these manipulations, adult subjects watch a video, five or ten minutes in length, of mothers and their young children (MLU < 2) at play, but with the audio turned off. These lengthy situational segmentsallow the subjects to pick up whatever clues are available from the pragmatic concomitants of the speechevent. They are told that , at the instant some particular noun is being uttered by the mother, they will hear a beep, their task being to guesswhat noun that was. For the nouns most frequently used in these mother/ child interchanges, the subjects are almost at ceiling. Usually, even a single scene/ beep pair is enough for the subject to identify the noun the mother was uttering . These findings imply two things. The first concerns the input situation itself: Evidently, mothers of very young children usually say nouns just when the objects that these label are the focus of conversation and are being manipulated by the participants. This makes their recovery from context easy (seeBruner, 1975, and Slobin, 1975, for prior evidenceof this here-and-now ' ' property of maternal speech to children) . The second concerns natural interpretations of situational information : The observer seems efficient at ' " ' " 1 In our notation , Islashesl represent sound, single quotes the concept, double quotes the utterance, and italics the word as an abstract object.

C. Fisheret aJ. I Syntacticconstraints

guessing the level of specificity at which the speaker is making reference elephants rather than animals or puppets - despite the fact that all of these interpretations fit the observed scenesequally well (pace Quine 1960; for evidence from child word learning , see Hall 1993; Hall and Waxman 1993) . So far so good : One can learn that the word for ' elephant' is /elephant/ (or /beepf) becauseit is said in the presenceof elephants. However, when we turn to the acquisition of lexical categories other than the noun, this promising story appears to fall apart. Subjects cannot correctly guess which verb the mother is saying under the same circumstances- obseryation of the mother/ child sceneswithout audio other than the beep. Though the subjects do choose as their guessesthe most common maternal verbs of all (e.g., come and put as opposed to arrive and situate), they fail to selectthe one that the mother was actually uttering at the sound of the beep. Their successrate is between.0 and 7% , depending on details of the manipulation . Why is the observed sceneso decisive for nouns and so uninformative for verbs? One factor proposed by Gentner ( 1978, 1982) has to do with the concepts that these lexical classesstandardly encode, namely the difference between object-reference concepts and relational concepts (see also Nelson 1974). The referenceof many nouns can apparently be extracted by appeal to principles of object perception and pragmatic inference, but even the homeliest verb meanings express relations among such concepts. Which such relation the speaker has in mind to convey is rarely accessiblefrom observation alone. Moreover, the nouns are frequently used in deictic-ostensive contexts to young learners: " This is a ball " ( Ninio 1980, Bruner 1983), while verbs are much rarer in such contexts as " This is hopping" . Another important factor is that the verbs are not uttered even to young learners in a tight time-lock with the events (Tomasello and Kruger 1992, Lederer et al. 1991) . Even when the events and verb utterances are relatively close in time, their seriation differs, a problem we have called interleaving. For instance, consider a scenein which the child is pushing a car, and then upon request from his mother .carries it over to show to his grandmother, who beams. The serial order of events here is push, go, show, beam. But the mother actually says " Go show Granny what you' re doing, she'll think you " push the car so well . Little problem arises for these adult subjects (or , we presume, for children) in getting the gist of the conversation, but the gist is very far from explicit identification of the verbs. The problem posed for identification is that the number and order of verb utterance (go, show, do, think, push) do not line up with the event sequence . Notice, as well, that

336

C. Fisher et alI Syntactic constraints

certain verbs commonly used by the caregivers are so general (do) or so ' abstract (think) as to be difficult to relate at all to what s actually going on (in 2 this latter example, beaming) . All thesecomplexities bear on an extremely robust finding in the language learning literature : Verbs are very rare in the first spoken (or comprehended) 50 words of child vocabularies, rather most items are nouns with a scattering " " of social items ( bye-bye ) and spatial prepositions (Goldin -Meadow et ale 1976, Nelson 1974, Bowerman 1976, Dromi 1987). This striking dominance of nouns (above their type frequency in maternal speech) persists until the third year of life (Gentner 1982) . We will contend here that , owing to the kinds of problems just sketched, verbs must be learned by a procedure that differs from the early nounlearning procedure that pairs isolated words (or beeps) to their real world contingencies. According to our hypothesis, verb learners recruit evidence from the syntactic structure in which new verbs appear, and pair this structural evidence with the information present in the scene. Thus we postulate a sentence-to -world mapping procedure for verbs rather than the word -to -world procedure that is satisfactory for explaining first nouns (for earlier statementsof this position , see Landau and Gleitman 1985, Fisher et ale 1991, Gleitman 1990). This would begin to explain the noun-before-verb developmental findings : It takes time to acquire structural knowledge, and nouns but not verbs can be acquired efficiently in the absence of such knowledge. Moreover, knowledge of the noun meaningsis, as we shall see, a prerequisite to extracting the verb meanings. ' es children s use of situational We will present here an experiment that assess and syntactic evidence for solving the mapping problem. But before doing so, we want to describeinformally the ideas behind the syntax sensitive learning procedure we have in mind. Fuller discussion is reserved until the experimental findings have been presented.

2 Of course it is easy to think of nouns that are similarly ' abstract ' , such as liberty , so relative ease of learning via extralinguistic observation is not theoretically identifiable with the noun / verb distinction . But it is as a practical matter : Abstract verbs are common in usage to children ( 5 of the most frequent verbs in maternal use to children under two years refer to mental states and acts, want, like , think , know, and see) but all the most frequent nouns in our corpus refer to visible object classesor names, e.g., Mommy . The more important point is that subjects cannot reconstruct even the maternal verbs that refer to observable actions (go, eat, catch, etc.) by watching the scenein the presenceof evidence (the beep) of just when they were uttered.

C. Fisher et a/. I Syntactic constraints

2. Syntactic supports for verb learning

As just discussed, observation of the pragmatics of scenes appears to underdetermine verb construals. The latitude for interpretation seemsmuch too broad if we grant (as we must) formidable perceptual, conceptual, and pragmatic interpretive capacitiesto youngsters. Though it is clear (seePinker, this volume) that there are severeconstraints on what can be lexicalized as a verb, and on just how humans conceptualize a scenein view, within these limits the hypothesis spaceis still vast. It is often and truly said that a picture ' is worth a thousand words, but evidently that s just the problem for verb vocabulary acquisition. One plausible way to distentangle verbs in ambiguous circumstancesis to note which known nouns occurred in construction with the verb ( Wasthe patient mentioned?) and the structural positions of these nouns ( Which one was the subject of the sentence?) . In principle this information can be useful becausethe surface-structural properties of sentencesare well correlated with " (in fact, are projections from ) certain aspectsof their semantics. To use such clues, the learning device must analyze the structure within which the novel verb is being heard. We have termed such a structure-dependent learning 3 procedure syntactic bootstrapping. This approach is similar to prior proposals for solving the mapping problem-in many regards. It posits that learners inspect ongoing events for clues to the verb meanings, armed with sophisticated perceptual, conceptual, and pragmatic knowledge.4 In addition , and in accord with the known facts ' 3 Pinker(this volume) hasobjectedto this terminology , but whats in a name? We meanby it that the learner is presumedto convergeon verb meaningby joint use of structural and . Semantic situationalevidence ), ultimatelyconcerns bootstrap ping, a termcoinedby Pinker( 1984 -structure grammar. The first steps in this procedureextract word acquisitionof a phrase for their use- by word-to-world pairing. If solelyby observingthe contingencies interpretations thiscanbedonefor verbs(whichwecontendit cannot) aswell asfor nouns(whichwecontendit basedon can), thenthe nextstepin Pinker's procedureis to assignthe itemsto lexicalcategories theacquiredmeanings (e.g., a thing is likely to be labeledby a nounwhilean actionis likely to be that holdscross-linguistically; Grimshaw1981 labeledby a verb, a generalization ). So the terms ping are not really parallel. The questionremainswhetherthe syntacticand semanticbootstrap , namely, whetherthe verb ping can be satisfied backgroundassumptionof semanticbootstrap can be acquiredsolelyfrom extralinguisticevidence , prior to acquisitionof the phrase meanings . structure ' 4 Pinker ( 1989 , this volume) assertsthat our position has limited the infant s conceptual ' ' repertoireto sensoryproperties a la Locke. Quite the contrary. It is the latitude of the hypothesisspacein the conceptuallywell- endowedinfant that opensthe door to multiple , if learnerswereas ). Of course (Chomsky1957 interpretationsof singlewords and sentences

C. Fisherel al. / Syntact; constraints

about languagelearning, it posits that verb learning occurs in the presenceof a priorly learned vocabulary of nominal items (Lenneberg 1967, Gentner 1978, 1982) . The innovation has to do with the way learnersare posited to representthe linguistic input that is to be paired with the extralinguistic input : as a parse tree within which the novel verb occurs. The structured sentencerepresentation can help in acquiring the novel verb just becauseit is revealing of the ' argument-taking aspects of that verb s interpretation . If phrase-structural knowledge of the exposure language facilitates verb learning, then the developmenta priority of nouns begins to be understandable; and, so does the of explosion verb vocabulary acquisition simultaneouswith the appearanceof rudimentary sentencesin speech(Lenneberg 1967) . In the experiment that follows we examine an example of this problem and its proposed solution : There are many meaningfully distinct paired verbs that occur in virtually all and only the samereal-world contexts, for example, give and receive, or chase and flee, lead and follow . When John gives a ball to Mary , Mary receives the ball from John. Movie directors make an art of ' distinguishing such notions visually. They can zoom in on the recipient s grateful mien, the giver out of focus or off the screencompletely. Using the word receive rather than give is a linguistic way of making the same distinction. But only for a listener who understandstheir meanings. Without a zoom lens, how is a learner to acquire the distinction in the first placeiS If the learner considers the novel verb use within a syntactic structure, and requires an interpretation that is congruent both with the scene and the Locke proposed , they would fail to acquirelanguagefor quite differentreasons ; namely, that almostno wordsreferto sensorypropertiesor combinationsof these(Fodor 1981 , Armstronget at. 1983 ). 5 For all theseperspective - changingverbpairs, distinguishing environmental conditionsarenot to say Thepeoplefled thecity , but arevery rare. For instance , it is reasonable reallynonexistent but not so reasonableto say The city chasedthe people(examplefrom Pinker, personal . Thequestionis whetherthese communication ). Soin principleonecanfleewithout beingchased ' raredissociatingenvironments playa role in the child s differentiationof thesepairedverbs. We know that closeto a third of verb usesto youngchildrenare in the absence of their referents" " not aboutthe here-and-now (Beckwithet al. 1989 as in is to ), Granny coming visit nextweek , whichoccursin the absence of visiblecoming(or visibleGranny). This meansthat the learning -to-world conjectures devicemustbe quite tolerantin evaluatingscene . /Come/ mustbe mapped onto ' come' thoughit is oftensaidwhennothingis coming, and oftennot saidwhensomethingis /usagemiscoming. No learningprocedurewilling to discountthe large percentageof scene es for give/receiveor chaselftee as matchesfor comecould treat the vanishinglyrare mismatch anythingbut noise.

C. Fisheret alI Syntacticconstraints structure , there is a solution

to the mapping

problem

339 for these verbs .

Considera listenerhearingoneof thesesentences : ( I ) Look, biffing! (2) The rabbit is biffing the ball to the elephant. (3) The elephantis biffing the ball from the rabbit. while watching a rabbit give and an elephant receivea ball. As we will show, if the listener has no accessto the syntactic framework, as in ( I ), she will probably interpret / biff / as related in meaning to English give. Hearing sentence(2) bolsters this choice. But a learner who inspects sentence (3) favors receive. There are two clues to this choice in sentences(2) and (3) . First is the to/ from distinction , which indicated which entity is source and which is goal of the moving ball. Second is the placement of rabbit and elephant within the structure, for whatever entity showed up as the subject of the sentencehas been selected, in the utterance, as the one that the sentenceis ' about' - the entity of whom the act is predicated. The notional interpretation of / biff / must be one that still fits the sceneobservedbut castsit in a different light : If the subject of the predication was " rabbit " , then the act was giving; if it was '. " , for a listener sensitive to elephant , then the act was receiving. In essence the full sentence, the interpretation of the observed scene will have been affected by the linguistic observation that accompaniesit .6 The difficulty of the mapping problem is not restricted to the perspectivechanging verbs that we have just discussed. Consider a learner observing a scenein which a rabbit pushesa duck, who falls ; and hearing one of these three sentences:

6 Presaging later discussion, note that a discovery procedure that implicates semantic deductions from surface structure must confront the fact that the relation between surface syntax and argument structure , even within a single language, is complex at best and can be misleading in some cases. Consider the case of gel, a near relative of the two verbs ( give and receive) just discussed. Gel is subject to two interpretations . When we say " Emmanuel got a book from the library " , the subject (Emmanuel) is certainly the causal agent in the book ' s " " moving out of the library . But when we say Emmanuel got the flu from Aaron , Aaron was the intended causal agent, assuming that Emmanuel wanted no part of the flu . Thus surface position of the nominals does not uniformly reflect distinctions in their thematic roles. " Moreover, gel can appear in two -argument sentencessuch as " Emmanuel got the flu in which its transfer-of -possessionsenseis masked (if intended) and may not be intended in the first place. We will return to these issues.

340

C. Fisheret al. / Syntacticconstraints

(4) Look, ziking! (5) The rabbit is ziking the duck. (6) The duck is ziking. ' ' ' ' Hearing (4) should not support the selection of push -like vs. fall -like ' ' ' ' interpretations . But ( 5) must be push and (6) must be fall . This time it is the number of noun phrases (in separate argument positions) in the sentencesthat bears on the interpretation of the verb item ; the intransitive sentence(6) simply will not support the causal property of push. For this latter pair (as opposed to give/ receive), cross-situational observation is available as an alternate route for acquiring the distinction between them, requiring no attention to syntax : Eventually there will be falling sceneswithout a pusher, allowing their disentanglement (Pinker 1984) . All the same, in the real case learners may recruit syntactic cues to facilitate the choice.

3. Salient interpretations of the action

Though we have just conjectured that linguistic observation will affect listeners' interpretation of scenes, we also mentioned in passing that syntax all aside - there are bound to be blases in how any event is most naturally represented. Thus in a ball -exchange by rabbit and elephant, its giving turns out to be more salient than its receipt. This differential salience of two interpretations of a scene is the second topic that will be taken up experimentally here. Specifically, we will consider a distinction in plausibility that we term the ' ' agency bias . Even infants appear to be inclined to interpret action scenes as involving a causal agent and an affected entity , or patient (Michotte 1963, Fritz and Suci 1981, Leslie 1982, Leslie and Keeble 1987, Mandler 1991) . reference to the This is in preference to an interpretation which excludes ' is feeding an in which a rabbit a scene agent. As an example, consider ' ' ' ' elephant, who eats. The notions feed and eat appear to be equally likely interpretations of a novel verb then uttered. Yet as we will demonstrate, observers usually interpret the novel verb in such a scenario as expressing ' feed' a causal act rather than ' eat' the noncausal ,( ) ( option ) . choice between give and well for the By hypothesis, this bias operates as receive even tho Qgh both these verbs occur in sentencesthat mention the agent, i .e., in sentence (2) the rabbit is agent of giving and in sentence

constraints C. Fisheret al. / Syntactic

341

(3) the elephant is agent of receiving. But there may be a preference between them all the same. In most give/ receive scenarios, the giver seems more volitional and thus is the plausible candidate for the agent role - the cause, first -mover, or instigator (Dowty 1991) . Our experiment will examine the joint effects of preferences in event representation (the agency bias) and syntactic deduction. Sometimes these ' factors work together to reveal the verb s meaning. For example, causal agent is linked to subject position in the sentencein all known languages (Clark and Begun 1971, Grimshaw 1981, Bates and MacWhinney 1982, Pinker 1984, Givon 1986, Schlesinger 1988, Dowty 1991) . Thus in sentence ' (5) the listener s event bias will mesh with syntactic deduction. In contrast , sentence (6) pits the two evidentiary sources against each other , for the ' ' preferred causal interpretation ( push ) is in this case incompatible with the 7 intransitive syntax.

Experiment

We have proposed that perspective-changing verb pairs like give/ receive and chase/flee (which are legion in the verb lexicon) pose a special problem for an observational word -mapping schemeand hence offer a useful testing ground for determining whether learners might be sensitive to other kinds of evidence. Specifically, we asked whether young children show sensitivity to syntactic structure in disentangling the sensesof such pairs of verbs, as well as other pairs which might be learned via cross-situational observation. To find out , we taught 3- and 4-year-olds novel (nonsense) verbs by using them to describe action scenes. These scenesdepicted single events which could be interpreted in two complementary ways. The manipulations of interest concerned the linguistic context in which the novel verb was presented. A child whose attention is directed to a 7 For adults, feed can occur intransitively , e.g., The callie arefeeding. But as predicted, it then is synonymous with eat. The wary reader will have noticed, as well , that eat often occurs transitively , as in The cattle eat thefodder , but does not mean that the cattle causethe fodder to eat. In short , eat can drop its object while feed can omit its causative subject. Then the two verbs share their licensed syntactic environments, as both can be both transitive and intransitive . It is the positioning of nominals in the structures, as mapped against the scene in view, that reveals the difference in their argument structures (see Levin and Rappaport , this volume, for a discussion of theseverb types) . The mapping problem can no more be solved by attention to syntax alone than by attention to observation alone. It is the joint operation of the two evidentiary sources that does the work .

342

C. Fisher et a/. I Syntactic constraints

relevant sceneshould be more likely to interpret a novel verb then heard as notionally resembling give if it is presented in sentence (2) than if it is presentedin sentence(3) . Symmetrically, taking a novel verb to resembleget or receiveshould be more likely upon hearing sentence(3) than upon hearing sentence(2) . To test these predictions, we needed some way of assessinghow our subjectsinterpreted novel verbs. The method used was a straightforward one: We simply asked the children what they thought the words meant. This method has two main advantages. The first is its very straightforwardness. When children can provide a paraphraseof a novel word , it is unnecessaryto attempt to infer from some less obviously relevant aspect of their behavior what they consider the novel word to mean. The second advantage is more central to our question: The essenceof the problem we addressin this work is that there are pairs of verbs that will be difficult or impossible to differentiate from observational evidence alone. By our own argument there will be no event we could show subjects that would isolate giving from receiving or chasing from fleeing. Thus we could not assesswhat children learned about the novel verbs by, for example, asking them to choose among pictured events(as in Brown 1957). The paraphrasemethod allowed us to examinejust those casesthat we have argued pose the knottiest problem for verb mapping. This method has one disadvantage as well, in that two-year-olds learning first "Verbscannot be induced to provide glossesor paraphrasesfor made-up words, so our youngest subjects are three-year-olds. As we will show, these children are capable of answering the question " What does biffing mean1" in revealing ways. No doubt can arise as to the relevance of three- and four year-olds to the question posed, for the bulk of the basic verb vocabulary is acquired by theseage groups.

4.1. Method Video-taped sceneswere shown to preschoolersand to adult controls. Each scenewas described by the experimenter with a sentencethat contained a nonsenseverb. The subjects' task was to paraphrase the verb. 4.1.1. Subjects The child subjects were twenty-four 3-year-old children (mean age 3;8, range 3; 1- 4;0), and thirty 4- year-old children (mean age 4;8, range 4;3- 5;0) . Nine children (five 3-year-olds and four 4- year-olds) were replaced in the

C. Fisheret at. / Syntacticconstraints

343

design for failure to respond (see the Procedure section below) . Eighteen adults were included to provide a baselinemeasureof competent performance in this task. A third of the subjectsin each age group were randomly assigned to each of three introducing context conditions (seeProcedure, below) . 4.1.2. Stimuli Six brief motion sceneswith puppet actors were video-taped. The scenes were designedto be naturally describablewith two English verbs that differed in their semantic and syntactic properties. One of the sentencecontexts that could accompany each scenewas arbitrarily called the ' X ' context, and the other was called the ' V ' context. Descriptions of the scenesand thesesentence contexts are shown in table I . Table I Scenes /sentence pairs Scene

Sentences

I : A rabbit is feedingan elephant with a spoon. 2: A rabbit comesup and pushes a monkeyoff a box. 3: A rabbit runsacrossthe screen , followedby a skunk. 4: A monk :ey is riding piggy-back on a rabbit. An elephanthandsa ball to a rabbit.

X: Y: X: Y: X: Y: X: Y: X:

6: A rabbit putsa blanketovera monkey.

The elephantis -- ing. (eat) . (feed) The bunnyis - -ing the elephant The bunnyis - -ing the monkey. (push) The monkeyis - -ing. (fall) The bunnyis ---ing the skunk. (flee) The skunkis ---ing the bunny. (chase ) The monkeyis - -ing the bunny. (ride) The bunnyis ---ing the monkey. (carry) The elephant is -- ing the ball to the bunny. (give) Y : The bunny is ---ing the ball from the elephant . (take) X : The bunny is ---ing the blanketonto the monkey. (put) Y : The bunny is -- ing the monkeywith the blanket. (cover)

For the first two scenesin table 1 (feed/eat and push/ fal/), the syntax of the two sentencesdiffers in the number of noun phrases, i.e., transitive feed expresses the causal relationship w~ile intransitive .eat does not. For the next two scenes(chaselfleeand .carry/ ride), the number of noun phrases is equal but the order of the nouns encodes two perspectives on the event and, 8 consequently, who is the agent. 8 Notice tha~ride/carry and choselflee don' t differ syntactically , i.c., in their subcatcgorlzation . Moreover, both membersof thesepairshave frames; both appearin simpletransitivesentences

344

C. Fisher et alI

Syntactic constraints

In the last two scenes(give/ receiveand put/ cover), the sentencesfor the two standard choicesdiffer in the order of NP argumentsin the sentenceas well as in the preposition used to mark the indirect object (to vs. from and onto vs. with) . These last two casescan be subdivided into one pair relevant to the choice of agent (give/ receive), thus subject (elephant vs. rabbit ), and one pair relevant to the choice of goal and located object (put/ cover) , thus direct object (blanket vs. monkey) . Thus the stimuli overall can provide some indication of the kinds of syntactic-semantic linkages that young children can recruit for verb mapping. Each sentence structure was randomly paired with one of six nonsense " syllables for each subject (zike, blick , pilk , dack, moke, node) . All sentences were presentedwith the verb in the progressiveform (blicking) to maximiz@ intelligibility and pragmatic felicity as descriptions of ongoing actions. 4.1.3. Procedure Two experimenters tested each child individually ; one showed the videotapes and uttered the stimulus sentences , while the other recorded the ' were also audio -taped, to allow later subjects responses. The sessions checking of the accuracy of the recording experimenter. " A puppet (Mac) was introduced to the child who was then told Mac ' doesnt speak English very well, so sometimeshe usespuppet words. Can you " help us figure out what the puppet words mean? . Assent received, the child was then given a practice trial in which Mac said " Look ! The elephant is " zorping ! as an experimenter made a hand-held elephant puppet laugh. The " child was then asked, " What does zorping mean? and prompted by asking " " What is the elephant doing ? This latter prompt was used only in the trial to , help the child understand the task. For the adult subjects, practice omitted for obvious reasons. They were simply informed the was Mac puppet that their task would be to guessthe meanings of nonsensewords. On each test trial , the subject first heard the stimulus sentenceand then saw the video-taped scene. The scenewas repeated for up to one minute as the subject watched. The experimenter repeatedthe stimulus sentenceat least

the same thematic-role assignmentto syntactic position : the subject is agent, the object is theme ' ' (or patient) . What we mean by attention to the syntax in solving the mapping problem for the ' ' ' membersof thesepairs is that the observers conjecture about whether the verb means chase or ' flee' is consequenton noticing which observedentity receivesthe subject agent slot. As we have emphasizedearlier, this is one reason why a priorly acquired nominal vocabulary is prerequisite to acquiring verb meanings.

C. Fisher et al. / Syntactic constraints

34S

once while the videotape played. The subject was asked " What does gorping rnean?" . Additional prompts were " What ' s happening?" or " What ' s going on?" . Subjects were encouraged to guess. Subjects who said nothing during the practice trial and during the first two experimental trials were dropped frorn the study. Sceneswere presented in two orders, chosen to allow all subjects to begin with less difficult items (as revealed during pilot testing) . The first order was feed/ eat, give/ receive, chaselfiee, cover/put , carry/ ride, push/fall . The secondorder was its reverse. The nonsenseverbs were presentedto the subjectsin one of three linguistic contexts: ' ' Neutral syntax ( No sentence) trials. A third of the subjects in each age group heard the nonsensewords in the syntactically uninformative context " " Look ! Ziking ! . These subjects were cued (by the -inK suffix, Brown 1957) that the novel word was a verb but receivedno infonnation about its specific syntactic behavior. Even if young children use syntactic evidencein real-life verb learning, this experimental condition withheld such evidence, allowing us to assessany blases subjects might show .in interpreting the scenes and providing a baselineagainst which to compare their perfonnance when given syntactic information . Sentencetrials. The remaining two -thirds of the subjects saw the scenes accompaniedby one of two introducing sentential contexts ( X or Y in table 1, for each scene). Each subject heard only one of the two sentencesfor a ' ' single scene. One group of subjectsheard the sentencesdesignated X in table ' ' 1, and the other group heard those designated V . We reiterate that the assignment of sentencesto the categories X and Y was entirely arbitrary . Thus, e.g., transitive feed was assignedto the Y category while transitive push was assignedto the X category. This arbitrary assignmentassuredthat each subject group heard some sentencesof both kinds. 4.1.4. Coding and scoring ' Subjects paraphrases were sorted into three categories :

( I ) ResponseX : Responsesthat fit sentence X (shown in table 1), both . For syntactically and in describing the scene, were coded as X responses ' " " I example, for scene , eat, or a phrasal equivalent ( He s drinking soup ), is congruent with the sceneand with the construal implied by the structure of sentenceX . Notice that a subject might be exposedto a Y sentence( The bunny is gorping the elephant) but give the X responseall the same. If so, the syntax of the input sentencehas failed to influence that response.

C. Fisher et alI

Syntactic constraints

(2) ResponseY: Responsesthat fit sentenceY (table 1), both syntactically and in describing the scene, were coded as Y responses . For example, for ' " scene1, feed, or a phrasal equivalent ( He s giving him medicine" ) would be congruent both with the scene and with the (causative) construal implied by sentenceY of this pair . 3 ( ) Other ( 0 ) : Failures to respond and responsesthat fit neither sentenceX nor Y were coded as Other. These were almost always relevant to the scene in some way (e.g., the response " They' re playing" for the give/ receivescene), but incongruent with the ditransitive syntax. Also included in this category were responsesthat fit both X and Y contexts, and therefore could not demonstrate sensitivity to the sentencestructure.

/ts 4.2. Resu To assessthe reliability of the coding, all of the responsesfrom the 3-yearold subjects were recoded by an independent coder who was blind to introducing context. The two coders agreedwith each other in 94% of coding decisions. Residual disagreements were resolved through discussion. The ' subjects responsesare summarized in table 2. This table shows the proportion of subjects in each age group who produced each type of responsefor each scene. We now discussthesefindings under severalrubrics: 4.2.1. Breadth of the hypothesisspace The scenarioswere designedto be quite simple, containing few distracting properties. Even so, as table 2 shows, child subjects gave a response that had to be relegated to the ' Other ' category in a substantial proportion of the trials (29% across all conditions for 3-year-olds, 23% for 4-year-olds) . This finding constitutes yet one more demonstration of the many- many relations between scene observation and verb interpretation . In several casescoded ' Other ' the , subjects explicitly mentioned both of the standard choices (e.g., ' ' and getting) . These responseswere quite common in the No sentence giving ' ' condition (36% of adults and 20% of children s total responses) but ' ' extremely rare in the sentencecontexts ( 3% of adults and 5% of children s responses). When a sentencecontext was provided, both children ( t ( 52) = 3.64, p < 0.001) and adults ( t ( 16) = 5.89, p < 0.001) were much less likely to propose both readings for the novel verb. This is a first indication that the syntactic contexts rein in the many interpretations made available by scene inspection, focusing subjects' attention on specific aspects of our scenanos.

C. Fisher et aI. I Syntactic constraints

347

Table 2 each foreach scene Proportion context response , byageandintroducing

-year-olds(n = 24) : Three eat/ feed 0.25 0.38 0.13 push/fall 0.38 ftee/chase 0.00 0.63 ride/carry 0.25 0.75 0.25 give/take 0.50 0.13 put/cover 0.25

0.38 0.50 0.38 0.00 0.25 0.63

o.so 0.88 0.38 0.63 0.50 0.38

0.13 0.13 0.13 0.25 0.00 0.25

0.38 0.00 0.50 0.13 0.50 0.38

0.2.5 0.2.5 0.00 0.2.5 0.2.5 0.38

0.7S 0.7S 0.88 O.SO 0.25 0.25

0.00 0.00 0.13 0.25 0.50 0.38

Mean

0.35

0.54

O.IS

0.31

0.23 0.56

0.21

0.60 1.00 0.40 0.40 0.50 0.40

0.20 0.00 0.10 0.20 0.00 0.20

0.20 0.00 0.50 0.40 0.50 0.40

0.10 0.50 0.00 0.20 0.30 0.20

0. 80 0.40 0.90 0.70 0.60 0.70

0.10 0.10 0.10 0.10 0.10 0.10

0.27

0.38

Four-year-o/ds (n= 30j : eat/ feed 0.10 0.70 0.00 push/fall 0.70 flee/chase 0.00 0.70 ride/carry 0.00 0.80 give/ take 0.40 put/ cover 0.20

0.10 0.10

0.20 0.30 0.30 0.20 0.50 0.60

Mean

0.13

0.41

0.35

0.55

0.11

0.33

0.1.2

0.68

0.10

Adults( n= J8) : eat/ feed 0.00 push/fall 0.50 flee/chase 0.00 ride/carry 0.17 give/take 0.50 put/cover 0.00

0.17 0.00 0.83 0.33 0.00 0.83

0.83 0.50 0.17 0.50 0.50 0.17

1.00 1.00 1.00 0.83 1.00 0.50

0.00 0.00 0.00 0.17 0.00 0.17

0.00 0.00 0.00 0.00 0.00 0.33

0.00 0.00 0.00 0.00 0.00 0.00

1.. 1.. 1.. 1.. 1.. 1..

0.00 0.00 0.00 0.00 0.00 0.00

Mean

0.36

0.44

0.89

0.06

0.06

0..

1..

0.00

0. 19

4.2.2. Narrowing the hypothesisspaceby attention to the syntax Each of the standard responseswas more frequent in the matching sentence context than in either of the other two introducing contexts. For example, ' ' subjectsgave more give -like responsesto give syntax than to receivesyntax or to the neutral condition. That is, X responseswere more likely in the context of sentenceX than sentenceY (children: t(34) = 5.04, adults: 1( 10) = 14.56, p ' context < 0.001) or the ' No sentence (children: 1(34) = 4.19, adults: 1( 10) = 5.81, p < 0.00I ). Similarly, Y responseswere more frequent in the context of sentenceY than sentenceX (children: t(34) = 8.83, adults: t( IO) = 16.14, p ' < 0.001) or ' No sentence (children: t(34) = 5.16, adults: 1( 10) = 32.73, p

348

C. Fisher et aJ. / Syntactic constraints

< 0,001) .9 The effect of syntax on the interpretation of thesenonsensewords emergesin the same way for each scenewith one exception (the 3 year old with consistent put subjectsparaphraseda nonsenseverb describing put/ cover syntax more often than cover syntax in all three introducing contexts)'. Overall, the result is that syntactic context had a powerful effect on subjects construal of the nonsenseverb. ' The children s behavior is epitomized most poignantly in several contrasting responsesto the scenesin the presenceof syntactic context. Consider the carry/ ride scene. Hearing the sentencethat treats the carrier (rabbit ) as " " ' subject, a child responds He s holding him on his back . But hearing the sentencethat treats the rider (monkey) as subject, another child responds " ' " He' s sitting on him s [sic] back . Clearly thesechildren watch the sceneand interpret what they seeto derive the verb meaning: They induce the meaning of its real-world accompaniments. Perhaps unlike the adults, from inspection ' they do not seemto be aware of a linguistic puzzle. All the same, the sentence heard exerts a strong - albeit implicit - influence on just what they think the ' ' scenedepicts: whether it is about the one who holds/ carries or the one who sits/ rides. This is the sensein which sentence-to -world pairing can sharply limit the search-spacefor verb identification. 4.2.3. Semanticblasesin the interpretation of verbs In addition to constituting a baseline for comparison to the sentence ' trials in our task , the ' No sentence condition provides a chance to look " for semantic blases in verb interpretation . After all , in observing th e chase/ " " flee scenewhile hearing Look ! Ziking ! , subjects are not really warranted in preferring one of these interpretations over the other . The scene fits them both . Our initial hypothesis was that subjects would therefore choose " chaseor flee ( " run away ) more or less at random in this condition , hence our original arbitrary division of the sentences(and their responses) into ' ' ' the ' X and V categories. But our subjects were anything but openminded in their guesses, as table 2 shows. All age groups had a preferred ' ' responsein the No sentence condition for 5 of the 6 scenes. They tended to 9 Preliminary analyses indicated that there was no effect of order (which of the stimuli the , so the two order groups subjects saw first ) on the probability of sentence-congruent responses were combined in this and all further analyses. The effect of syntax on verb paraphraseswas shown in a series of planned I -tailed I-testS. In each case the dependent variable was the proportion (arcsine transformed) of each response( X or V ), examined across sentencecontexts. Separateanalyses for the 3- and 4- year-olds yielded the same results in each case; to simplify presentation of the results, the two groups of children are pooled in the analysespresentedhere.

C. Fisheret al. / Syntactic constraints

349

describesceneI as one of feeding rather than eating, scene2 as pushing rather than fal /ing, scene 3 as chasing, not fleeing or running away, scene 4 as carrying rather than riding, and scene 5 as giving rather than taking or receiving. The adults showed these blases even more regularly than the children. In each of these cases, subjects evidently selected a ' more causal' or agentive participant in the scene, and took the verb to code the actions of that participant, who thus became sentential subject. Scene 6 (put/ cover) leavesthe agent choice unaffected and, interestingly enough, it is in this case only that children show no preferenceof choice in the ' No sentence' condition . Clearly, this unanticipated agency factor was strongly affecting our ' . subjects responses To study this semantic bias, we now recoded the X and Y sentences according to whether they matched or mismatchedthis agencybias. For feed/ eat and push/fal /, no difficulty in doing so arises: Only the feed and push sentencesmention the causal agent so theseinterpretations should be favored if there is a bias to conceivescenesas an agent acting on a thing affected. For chaselfteeand give/ receive, this agent- act- patient interpretation applies to both members of the pairs. For these, we relied on Dowty ' s ( 1991) ' proto ' agent classification schemein which several factors are postulated to lead to the choice of plausible agent: animacy (which does not distinguish for our stimuli), a-ctivity, and instigator of the action. Thus necessarilyit is chasingthat precedesand causesfleeing, and giving that precedesand causesreceiving. For the case of carry/ ride, no such principled distinction in the verb meaningsthemselvesexists for choosing the plausible agent. However, in our particular depiction of this act, the carrier (the rabbit ) was actively running across the screen with an inert monkey sitting on his back. Apparently this distinction of relative activity vs. passivenessled subjectsto choose the rabbit as instigator, thus plausible agent. (Had our sceneinstead shown, say, a child riding a mechanical bull , doubtless this choice would have been reversedfor this verb pair .) Table 3 reorganizes the findings according to this distinction between ' ' .lo Agentive (A ) versusNon Agentive and lessplausible agent (NA ) responses The table shows the proportion of subjects in each age group and intro 10 Because theagencyeffectwasunanticipated , theexperimentwasnot balancedsuchthat each . So that this differencewill not subjectwould hear an equal numberof A and NA sentences contaminatethestatisticalassessments here, all comparisons betweenthe A and NA contextsare within-subjects . All othercomparisons are betweensubjects , as before.

350

C. Fisheret alI Syntacticconstraints

ducing context who produced A (e.g., feed, carry) and NA (e.g., eat, ride) , as defined above. These values are shown only for the five relevant responses scenarios(that is, excluding the put/ cover scenewhich cast the same participant as subject in both context sentences ) . Across all presentation conditions NA responses(91) two to outnumber 187 A and all age groups, responses( ) one. There is an agency bias.

Means:

0.53

0.18

0.70

0.13

0.20

0.50

Four-year-olds(n = 30) : . eat feed fall push flee chase ride carry take give

0.10 0.10 0.10 0.80 0.40

0.10 0.00 0.00 0.00 0.10

0.80 1 0.90 0.70 0.50

0.10 0.00 0.00 0.20 0.00

0.20 0.50 0.10 0.20 0.30

0.60 0.40 0.40 0.40 0.60

Means:

0.66

0.04

0.78

0.06

0.26

0.48

Adults(n = 18) : eat feed fall push flee chase ride carry take give

0.17 0.50 0.83 0.33 0.50

0.00 0.00 0.00 0.17 0.00

1.00 1.00 1.00 1.00 1.00

0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.17 0.00

1.00 1.00 1.00 0.83 1.00

Means:

0.47

0.03

1.00

0.00

0.03

0.97

. Since ' Other ' responses are left out of this table , proportions not sum to I .

within each scene/ context cell may

Not surprisingly, this effectof the natural interpretationof the scenesis most powerfulwhereit is not contaminatedby mismatchingsyntacticinfor-

C. Fisher et at. / Syntactic constraints

mation. Thus the preferencefor agentive responsesappears most strongly in the ' No ' sentence' condition in table 3. This effect was assessedin post-hoc (Bonferroni adjusted) matched-pairs I-tests, which revealed that A choices were more frequent than NA ones in the ' No sentence' context for both children (t( 17) = 6.18, p < 0.001) and adults (1(5) = 5.93, p < 0.01) . 4.2.4. The interaction of syntax and semantics So far we have shown (table 2) that structure affects Ss' guesseswhen it is made available, in the X and Y (sentential) conditions of presentation; and that the agency bias (for which responseswere recoded as A and NA ) affects Ss' guessesheavily in ~he absenceof syntactic information (the ' No sentence' condition of table 3) . The question remains how these factors interact when (as in the real world of maternal speech) the new word is presentedin a full sentencecontext. To find out , we now look more closely at the finding (table 2) that , while the effects of syntactic context were very strong and reliable, they were nowhere near categorical for the child subj~ ts (though they were nearly so for the adults) . The reason is that while the syntactic introducing circumstances " supported the agencybias in some cases(as when the subject heard The " rabbit is biffing the elephant while watching a feed/ eat scene), the syntactic and semantic cues pulled in opposing directions in other cases(as when the " " subject heard The elephant is biffing while watching this same scene). To repeat, subjects tend to view this scene as a feeding scene (the semanticinterpretive influence), but to encode the intransitive sentencestructure as favoring eat (the syntactic influence). The joint action of thesetwo variables can be seenin table 4, which repeats the data of table 3, but summarizing across the five relevant scenarios. Thus table 4 shows the proportion of subjects in each group and introducing context who produced A and NA responses.Column 1 of this table shows the bias toward agentive responseswhich we have just documented. Column 2 shows a slight enhancementof this preference(slight, becausethe preference in this direction was already strong) when it is supported by the syntactic evidence; that is, in agentive syntactic contexts. Column 3 shows that the agencybias is heavily mitigated by mismatching syntactic evidence: When the syntax demands the NA interpretation, the A responseis actually dispreferred, no longer the modal responsefor any age group . That is, we seethe same effect of sentencecontext on verb interpretation when the normally disfavored non-agentive responsesare examined separately: These responsesare more frequent in the non-agentive than the agentive sentencecontext (chil -

C. Fisher et a/ . / Syntactic constraints

Table4 , by ageand introducingcontext Proportion. agentive(A) and non-agentive( NA) responses Age group

Introducing - context Non - agentive

No sentence

Agentive

A

NA

A

NA

A

Threes Fours Adults

0.53 0.66 0.47

0.18 0.04 0.03

0.70 0.78 1.00

0.13 0.06 0.00

0.20 0.26 0.03

0.50 0.48 0.97

Overall

0.57

0.08

0.81

0.07

0.18

0.61

. Since ' Other ' responses are left out of this table , proportions not sum to 1.

within each scene/ context cell may

4.2.5. The effect of age Inspection of table 4 also shows that there is an age effect on these response patterns , with adult responses almost categorical and child responses probabilistic with respect to the variables under investigation . The effect of age on the likelihood of producing frame-congruent responsesin the two sentence contexts was shown in an ANDV A with age group as a between-subjects effect, and sentence context (A versus NA ) as awithin subjects factor . The main effect of age was significant (F(2,45) = 17.36, P < 0.001) : adults were more likely than children to produce a response that fit the frame , but the two groups of children did not differ significantly from each other . There was also a significant main effect of sentence context (F( 1,45) = 5.33, p < 0.05) : Frame-congruent responseswere more likely in the A than the NA context for all age groups, the interaction of situation and syntax that we discussed earlier . However , the effects of age and sentencecontext did not interact (F(2,45) < 1) . Thus the effect of age is a simple one - adult performance was more stable with regard to both the syntactic and saliencevariables, but all groups took syntactic information into account when it conflicted with the agency bias as well as when it did not.

C. Fisheret alI Syntacticconstraints

4.2.6. Summary of thefindings The findings support the claim that a scene observed is insufficient for ' fixing the meaning of a new verb, for subjects construal was often one that mismatched the interpretations we had in mind. But in these stripped-down puppet scenarios, the kinds of interpretation we had expectedwere the most frequent ones for each age group and for each scenario (table 2). A bias in event representationinfluenced the subjects' choice of the plausible ' doer' vs. ' done-to ' in these scenes , independent of the syntactic introducing contexts (tables 3 and 4) . But the most powerful influence on the choice of responsein this regard was the structure of the sentenceheard, for the sentencesthat ' pitted syntax against semantics reversed what had been subjects modal preferencein the neutral contexts (tables 2 and 4) . As we hypothesized, the syntactic structure led to a principled choice of interpretation among those that were compatible with the scenein view. 4.2.7. An alternative interpretation of thefindings Before discussing a theory of verb mapping that comports with these findings, we digress to consider a persistent difficulty in interpreting experimental studies of word-learning. It is hard to rule out the hypothesis that subjects are relying solely on preexisting lexical knowledge in such tasks (Pinker this volume) . That is, children could take the nonsensewords to be synonyms for existing words, and respond with eat when appropriate in our task becausethey already know both the meaning and the syntax of that particular lexical item, rather than by interpreting the novel term in asyntax sensitive learning procedure. The standard remedy for this problem is to teach young children new words for novel objects or actions (e.g., Markman 1989, Gropen et al. 1991) . Many studies of syntactic bootstrapping use this strategy as well (Naigles 1990, Hirsh -Paseket al. 1988) . In the current experiment, even though it allowed for the possibility of glossing the nonsenseterm with a known verb, our child subjectsrejectedthis option in a substantial percentageof cases. Thirty -five percent of the 3-yearolds' and 48% of the 4-year-olds' responseswere phrasal descriptions rather " than single words (e.g., 441icking it off the spoon" instead of 44eating , 44giving " " " him medicine rather than 44feeding, 44trying to smell him rather than " ' 44chasing) . This is not to say that thesechildren didn t know the wordsfeed, eat, and chase. Rather, they seem to have taken the novel verbs to have specific meanings that were not equivalent to known verbs. This outcome is in line with much current work on the . uses of lexical contrast in the acquisition of word meaning. Young children seemdetermined to assign new

354

C. Fisher et al. / Syntactic constraints

meaningsto new words. They do not expect there to be exact synonyms. (For this effect with nouns, seee.g., Clark 1987, and Markman , this volume ; and for the analogous effects with verbs, Golinkoff et al., in prep. ; Kako , in prep.) We now ask whether the samepattern of syntactic effectson verb interpretation occurs for the phrasal responsestaken alone. If so, it is unlikely that ' the children were merely ' filling in the blank in the stimulus sentenceswith a known verb that fit , but were engaging in something more like the syntaxguided mapping procedure proposed here. Table 5 shows the obtained responsepatterns for the phrasal responses , 4old contributed one olds and . Three 3 year year omitting all single-verb glosses no data to this analysis. Given the reduction in the data entailed by ' recoding the responses in this way, the 3- and 4-year-olds data were combined. Inspection of table 5 shows that the same pattern of structural effects on verb interpretation remains. This pattern was reliable in independent . I -tests on the (arcsine transformed) proportion of X and Y responses X than sentenceY ResponseX was more likely in the context of sentence ' ' (1(30) = 2.94, p < 0.01) and the No sentence context (1(32) = 4.11, p < 0.001) . Similarly , responseY was more frequent given sentenceY than either of the other two introducing contexts (sentenceX : 1(30) = 6.59, p < 0.001; no sentence: 1(32) = 3.61, p < 0.001) . In sum, even when the children evidently did not interpret the nonsensewords as puppet synonyms for verbs in their known lexicon, there are strong and reliable effects of sentencestructure on construal.

5. DiseI RiOD We now outline a verb mapping procedure that comports with the findings just described, and with related effects in the language acquisition literature . 5.1. Extraction of linguistic formatives

A vexed problem in understanding language learning concerns how the child finds such units as word and phrase in the continuously varying sound " ' stream: how Henny-Penny s listeners interpreted her as saying The sky is " " " falling rather than This guy is falling . A vast literature now supports the

C. Fisheret al. / Syntacticconstraints

355

Table5 Proportion of X and Y responsesby introducing context ; phrasal responsesonly (3- and 4- yearaids together, n = 50) Scene

' No sentence' (n = 18)

SentenceX (n = 16)

eat/ feed push/ fall flee/ chase ride/carry give/ take put/ cover

0.00 0.33 0.00 0.11 0.00 0.00

0.17 0.06 0.17 0.28 0.11 0.06

0.13 0.75 0.38 0.38 0.13 0.00

0.00 0.00 0.06 0.06 0.00 0.13

0.00 0.31 0.00 0.25 0.00 0.13

0.13 0.50 0.56 0.50 0.19 0.25

Mean

0.07

0. 14

0.29

0.04

0. 11

0.35

SentenceY (n = 16)

view that thesesegmentationdecisionsare made by infants basedon prosodic and distributional cues in caretaker speech(for recent evidence and discussion , see Gleitman et al. 1988, Gerken et al. 1993, Fisher and Tokura , in 11 press; Kelly , this volume; Cutler , this volume; Brent, this volume) . In what follows, we presuppose these approaches to solving the segmentation problem , concentrating attention on acquisition of the phrase structure (which requires ~abeling as well as segmentationof phrases) and the word meanings. 5.2. Word- towor /d pairing and the acquisition of first nouns We take as given that the human learner expects sentencesto convey predicate/argument structure as organized by a phrase-structure grammar that conforms to X -bar principles. However, universal grammar leavesopen someparametersof the phrase structure of the exposUrelanguage; thesemust be set by experience. Before this learning occurs, the novice can recruit only the exigenciesof word use - the pairing of words to their extralinguistic contexts - to solve the mapping problem. 11 For evidenceon infant attentionto clauseboundingcuesseeHirsh-Paseket at. 1987; for -boundingcues,seeJusczyket al. 1992;and for word-boundingcuesseeGrosjeanandGee phrase 1987 , Kelly, this volume. Thereis alsoa literatureon adult speechproductionand perception , too vast for us to cite here investigatingthe physicalbasesof theseprosody-syntaxmappings -Cooper 1980 , Lehisteet at. 1976 , , Cooperand Paccia (but for seminalarticles, seeKlatt 1975 Cutler, this volume) and the availability of suchcuesin infant- directedspeech(Fernaldand Simon1984 , Fisherand Tokura, in press; Ledererand Kelly 1991 ).

C. Fisher et alI

Syntactic constraints

By default , the effect should be that youngest children can acquire only object tenDs (nouns, in the adult language) . This is because, as we described in introductory remarks, only the nouns occur in maternal speech in a tight time -lock with the situational contexts, and in ostensive contexts. And indeed one of the most striking findings in the languagelearning literature is that first words are nouns despite the fact that from the beginning the learner is exposed to words from every lexical category . We assume that these first words are assigned to the formal category noun on a semantic basis, as conjectured by Grimshaw ( 1981) and Pinker ( 1984) . 5.3. Setting the phrase structure , and first verbs

-to-world Our findings suggest that verb learning implicates a sentence as Pinker ( 1984) pairing procedure and cannot in general be accomplished, and others have advocated (seefootnote 3), by pairing the isolated verb to its observational contingencies. But then how much grammatical knowledge is required as input to verb learning? And how might children acquire this? We propose that the meanings of the first verbs, as well as relevant components of the phrase structure, are acquired by bootstrapping from a parti .al sentential representation (henceforth, PSR) that becomes available once some nouns have been learned: This consists of the known nouns and the unknown verb, as sequencedin the input sentence, e.g.

[ ... baby ... eat ... cookie ...] By hypothesis, it is this richer-than-the-word , poorer-than-the-phrase-structure , representationthat learners past the one-word (noun) stagefirst attempt to pair with the scenein view. There is some evidence that the PSR can aid in verb identification in two ways. First , the identity of the nouns can provide information about the selectional properties of the verb. Lederer et al. ( 199] ) showed that adults can identify about 28% of the verbs that mothers are uttering if , in addition to the sceneinformation , they are also told which nouns occurred with the verb in the maternal utterances. This level of performance is not great, but is a significant improvement over the 7% success rate that subjects achieve if shown only the video-taped scene. It is easy to seewhy having the nouns is so helpful : If you are told that baby and cookie

C. Fisher et alI

Syntactic constraints

357

occurred in construction with the mystery verb, eat becames a plausible ' ' conjecturejust becauseverbs that mean eat should selectfor edibles.12 The PSR yields a second advantage for verb mapping, provided that the learner also has implicit accessto the Projection Principle (roughly, that every argument position required by the verb will be reflected as a noun phrase in the surface sentence; Chomsky 1981) . So armed, the learner can make a secureconjecture as between an intended unary relation (such as fa //) vs. a binary relation (such as push), simply by counting the number of noun in the sentence.13 phrases Early use of this machinery is suggestedby findings from Naigles ( 1990), with babies 23- 25 months of age. They were shown a video-taped scenein which (a) a duck who by pushing on a rabbit ' s head forces the latter into a squatting position whilst (b) both the duck and the rabbit wheel their free arms in a circle.' Half the subjects were introduced to the scene with the sentence" The d uck is gorping the rabbit " and the other half heard " The duck and the rabbit are gorping " . Thereafter , two new videos were shown, one to the child ' s left and one to her right , along with the prompt " Find " gorping now ! . One of the new videos showed the duck forcing the rabbit to squat (but no arm -wheeling) and the other showed the two side by side wheeling their arms (but no forcing -to -squat) . The children who had been introduced to gorping within the transitive sentence now gazed longest at the causal scene while those who had heard the intransitive sentence looked longer at the noncausal scene. Here, as in the push/fall and feed/ eat scenesinvestigated in the present experiment, the one-argument structure did not substain a causal interpretation despite the agency bias in event representation. A related point is made by Fisher ( 1993), who showed unfamiliar agentpatient events to children aged three and five yeary but with the entities named only by pronouns. For instance, they saw one person causing another to rotate on a swiveling stool by alternately pulling on the ends of a scarf around the victim ' s waist. Half the subjects heard " She's blicking her around" and the other half heard " She' s blicking around" . The child ' s task was to point out , in a still photograph of the event, the one whose action was 12 We must acknowledge , however , that the Gleitman-Landau archiveof maternalspeech includesmanyexamples like " Don' t eat that paper!" , " We don' t eat the book, Bonnie" . 13 Of coursethis canonly work for simplesentences for, e.g., " Therabbit in thegrasshopped " containstwo NPswithin an to novicesarecharacteristically away argumentposition. But sentences short (approximately5 wordslong, on average ; Newport 1977 ) and so rarelyembodythis problem.

C. Fisher et a/. I Syntactic constraints

' labelled by the novel verb (" Point to the one who s blicking the other one " around" or " Point to the one who' s blicking around ) . The intent here was to put the children into the position of much younger learners who have access only to the PSR: They knew how many arguments were supplied to the new verb, but not which was which. Those who had heard the transitive frame confidently chose the causal agent as the blicker , while those shown the intransitive frame were willing to select the patient as blicker. Thus without being told which event participant has been cast as sentencesubject, preschooler interpreted a one-argument structure as incompatible with a causal interpretation . The PSR taken together with the scene observed will allow learners to acquire a crucial aspect of the phrase structure itself. Supposea child hears " " kick " for the first time in the frame " The bunny kicked the monkey . The agency bias,- as constrained by the minimal structure given by the PSR (namely, a 2-argument structure), will lead the learner to seek an agent patient interpretation of the scene. Provided that she knows the nouns bunny and monkey, she can annotate the phrase structure as shown in figure 1, marking bunny, the first noun in the structure, as the agent (Joshi and Rambow, in prep.) . This representation now matches two of the quasiuniversa properties of the category subject of transitive sentence there is in a transitive one noun which is both the agent and the leftmost noun 14 structure. There is a strong tendency for languagesto place subjectsbefore objects (Keenan 1976, Kayne 1992) . Further phrase-structure options can then be set based on this initial assignment.

" " " / / """ / MONKE BUNNY KICK

(agent)

Fig. 1

14 Notice that this claim differs from the subject -agentlink which has beeninvokedin the , Draine literatureto supportan initially asyntacticverb-learningprocedure(e.g., Grimshaw1981 . It is andHardy 1982 , andothers). In fact, subjectsareoftenpatientsor experiencers , Pinker1984 . To recognizethis whoseagentjust about universallysurfacesas subject the transitivesentence distinction requires , at minimum, PSR knowledgethat will revealthe numberof argument . positions

C. Fisher et aI. / Syntactic constraints

359

~ ~u~ ~0~=

f

Notice finally that if there are languagesin which objects precedesubjects, the assignmentof subject of the sentence(the NP immediately dominated by S) to the serially second NP is still possible, based on the PSR. In that case, the learner would have heard " Kicked the monkey the bunny" in the presence of a bunny-kicking -monkey scene. The observed scene identifies bunny as agent (hence subject of the transitive verb), consistent only with template (b) of figure 2.

-structure -structure . Fig. 2. (a) anSVOphrase ; (b) a VOSphrase template template

>A A =zz>~~ ~0ZZ ~

f

;

Severalother kinds of cue to the phrase-structure of the exposurelanguage have been suggested , and have plausibility . For instance, Mazuka ( 1993) that suggests branching direction in Japanesecan be deduced by observing intonational markings of moved relative clauses. Further , subject and object NP ' s vary strongly in maternal speech(the former is usually a monosyllabic pronoun and is often omitted even in non-prodrop languageslike English; Fisher and Tokura , in press; Gerken et ale 1993), and learners may be sensitiveto such probabilistic patterns (for discussion, seeKelly , this volume) . Hirsh -Pasekand Golinkot T( 1991) showed that 17-month -old children who utter only isolated nouns appreciate at least the rudiments of English word order and the way this maps onto thematic roles. If they heard a voice saying " " Big Bird tickles Cookie Monster they looked primarily at a video screen " " depicting this event, but if the voice said Cookie Monster tickles Big Bird , they looked at another video screen which depicted this latter event. These toddlers evidently were sensitiveto (something like) the idea that the doer of the action is the subject of the transitive sentence; and that in English the serially first noun is that subject. 5.4. Structural information narrows the searchspacefor verb mapping We have so far argued that PSR taken together with scenerepresentation allows the child a bootstrap into the phrase structure of the exposure

C. Fisheret oJ. I Syntacticconstraints

language. Once the full phrase structure is acquired, the learner can approach the perspective-changing verbs that we have studied. Disentanglement of the membersof thesepairs requires more than counting NP positions (which are the same for both interpretations) and fitting these to the logic of the observedsituation (which suits either choice) . Once the phrase structure has been bootstrapped from PSR, the learner can make this decision by inspecting the geometry of the tree to determine which noun is sentencesubject. If the plausible agent appears as subject with the give/ receive scene, then the situational and syntactic cues converge on give. But if the plausible agent appears in nonsubject position then it is not the agent, despite appearances. In the present experiment, we showed child responsivenessto these implications of structure for verb interpretation even in caseswhere they had to overcome a semantic bias in event interpretation to use it . 5.5. The informativeness of multiple frames

In principle, attention to the licensedrange of syntactic environments for a verb can provide converging evidence about its interpretation, just because these several environments are projections from the range of argument structures associatedwith that verb. This feature of syntactic bootstrapping is controversial (as opposed to the ' zoom lens' notion which appears to have gained wide currency) . So after describing the potential usefulness of framerange information for solving the mapping problem, we will discussavailable experimental evidence in its favor. Specifically, we will discuss the experimental manipulations deemedcritical by some skeptics (particularly , Pinker, this volume) for confirming the hypothesis. 5.5.1. The resolvingpower of frame rangesfor verb mapping In very many cases, a surface-structure/ situation pair is insufficient or even ' misleading about a verb s interpretation . One such case is the eat example that we mentioned earlier. The phrase structure is the same when the adult " " " " says Did you eat your cookie? as when he says Do you want the cookie? , and the two verbs are used by caretakers in situations where their interpretations can easily be mistaken. Subjects always come up with an action term that fits the observed sceneand the structure, and guesseat instead of want. In responseto the fact that the next sceneobservation does not support the eat conjecture (it may show, say, the mother offering a toy rather than a

C. Fisher et alI

Syntactic constraints

cookie to the child ), subjects now come up with yet another physical term (e.g., take) . They are mulishly resistant to conjecturing any mental term. While successiveobservations force them to change their minds about which physical-action term is the right one, they never seemto get the idea that this bias should be overridden altogether. This effect was shown by Lederer et al. ( 1991) with adults, and by Gillette ( 1992) with children. Examination of the further syntactic privileges of eat and want can resolve this problem . Eat occurs intransitively and in the progressive form while want does not . Want also occurs with tenseless sentence complements (Do you want to eat the apple?) . These distinctions are sufficient to disentangle the two verb construals , for only mental activity verbs license these constructions . (Note that force verbs, which also accept sentence complements, require an additional nominal position , e.g., Make him eat the apple! , but not . Make eat the apple.' ). As we will discuss presently, subjects seize upon this disambiguating structural information to find the right construal . A second example of residual problems unresolved by single frames, even ' though theseare paired with differing events, concernsa blind child s learning of the distinction between touch and see. Blind learners receiveobservational evidenceabout the meanings of both terms though it is perforce haptic and not visual. The blind child ' s first uses of see at age two were in the sense ' touch' e. . she commanded " Don ' t see that !" , g, , while pushing her brother from her record . The confusion arose , doubtless, becauseevery away player scenein which the blind child can see(' ascertain by perceptual inspection' ) is a scenein which she can touch. And both verbs occur most often in maternal ' " ' speechas simple transitives. Further syntactic experience( Let s seeif there s " cheesein the refrigerator ) can account for how the blind child could, as she did , come to distinguish between the two construals by age three (Landau and Gleitman 1985). As a more general example of the convergencethat frame ranges make available for verb mapping, consider the four verbs give, explain, go, and think. These verbs are cross-classified both conceptually and syntactically. Give and explain, different as they are in many regards, both describe the transfer of entities between two parties. Accordingly, they can appear in structures with three noun-phrase positions :

(7) Ed gavethe horseto Sally. (8) Ed explainedthe factsto Sally.

C. Fisheret alI Syntacticconstraints

In (7), a physical object (the horse) is transferred from Ed' s to Sally' s hand and in (8) abstract objects (the facts) are transferred from Ed' s to Sally' s mind. A noun phrase is required for each of the entities involved: the giver, the receiver, and that which is transferred betweenthem. It is this similarity in their meanings that accounts for the similarity in the structures that they accept. Verbs that describe no such transfer are odd in theseconstructions: (9) ( 10)

. Philip went the horse to Libby . . Philip thought the facts to Libby .

But there is another semantic dimension for these four verbs in which the facts line up differently . Explain and think concern mental events while give and go concern physical events. There is a typical surface reflex of this distinction also, namely, mental verbs accept sentencecomplements(expressa relation betweenan actor and a proposition) : ( 11) Jane thinks/explains that there is a mongoosein the parlor . ( 12) * Jane goes/gives that there is a mongoose in the parlor . The learner who appreciatesboth of thesemapping relations can deducefrom the range of syntactic environments that give expresses physical transfer while explain expresses mental transfer (that is to say, communication ; Zwicky 1971, Fisher et al. 1991) . Potentially there can be a rapid convergenceon the meaning of a verb from examination of the several structures in which it appearsin speech. Though there are hundreds of transfer verbs and scoresof cognition-perception verbs, there is a much smaller number of verbs whose meaningsare compatible with both these structures, and which therefore can expresscommunication (e.g., tell, shout, whisper) . Thus acrossuses, the syntax can significantly narrow the hypothesis spacefor the verb meaning. Perhaps the most important reason for postulating this cross-sentence ' ' procedure has to do with open roles . Though eat is logically a two -argument predicate, with an eater and an eatee, still one can say The baby is eating. It is often supposedthat learners would unerringly interpret a co-occurring scene as one of eating, and simply refuse this scene/sentenceas a learning opportunity becausethe scenedoesn' t line up with the required argument structure for eat (assuming, of course, that the child can' t ' hear traces') . But unfortunately , scenesare complex and thereforealmost always support falseconstruals if a single scene/sentencepair is to be decisive. For instance, the baby is sitting , smiling, and so forth , while she eats. Why not map one of these

C. Fisher et a/. I Syntactic constraints

inalienable acts onto the observed intransitive sentence? The advantage of cross-sentenceanalysis taken together with cross-sceneanalysis is that it can reveal the argument structure associatedwith the verb overall. Attention to severalstructural environments becomesan even more important capacity of the learning devicewhen we consider languagessuch as Chinese, which allows rampant omission of arguments in the surface structure.15 We have now hypothesized that the lexical entry for the verb is derived from observing its range of syntactic environments (which reveal the verb' s argument-taking properties) taken together with the observational environments ' ' (which reveal everything else.) This does not mean that each use of the verb instantiates each argument-taking component of the lexical entry . For instance, verbs like open and sink express a causal relation in some environments but not others (Carol opensthe door vs. The door opens). The gloss assignedto lopenl in the lexicon must be one that is compatible with both licensed semantic-syntactic environments. In contrast, die is noncausal (intransitive) only and kill is causal (transitive) only . Following Grimshaw ( 1992), we might render the entry for open as

CAUSE (y OPEN (x I I AGENT THEME '

' decomposing the meaning only to the level required to state the argument structure and thus to predict the surface structures. The semantic distinction between opening and closing is then derived from examining situational factors, though only God and little children know just how. There is no opening-vs.-closing syntactic reflex to aid them. This schemecertainly does not imply that open is interpreted as meaning ' an event which is causal and noncausal at the same time' . If that were true, then the larger a verb' s syntactic range the less it would mean. Rather, the lexical description predicts that open is causal when transitive, noncausal when intransitive. The interpretive choice among those made available by the lexical entry is (on any single use of the verb) derived computationally from the truth value of the sentencestructure. 15 To mention one more example pertinent to the verbs studied in our experiment, note that both eat and get can occur in transitive environments. But a clue to the transfer- of-possession senseof get is manifest in ditransitive Emily gets Jacquesan ice cream cone vs. . Emily eats Jacquesan ice cream cone. Similarly , there is a semantic correlate of object dropping (eat vs. want), see Resnik ( 1993).

C. Fisher et al. / Syntactic constraints

5.5.2. Documentationof the use of frame rangesin acquisition Theory all aside, the questions remain whether speech to novices, which tends to be quite simple, is sufficiently rich in structural information to support learning ; and whether observerswill use such information even if it is there in the input . The answer to both questions appears to be yes. The database provided by mothers to their young children is refined enough to support learning from frame ranges. Lederer et al. (in press) examined lengthy conversations of 8 mothers with their young children (MLU < 2.0) to find the verbs used most frequently within and across mothers. For the 24 most common verbs, a verb by syntactic-environment ' matrix was developed for each mother s speech. Within and across mothers, each verb was found to be unique in its syntactic range. Using a procedure devised by Fisher et al. ( 1991), it was determined that overlap in the syntactic environments predicted their semantic overlap to a striking degree. Much more important , there is also evidence that observers (at least adults in the laboratory ) will use this cross-sentence information for verb identification . This was shown by Lederer et al. ( 1991) with a version of the experiment that Pinker (this volume) has acknowledged would put framerange learning to the crucial test. Adult subjects were shown lists of actual syntactic structures used by mothers to their infants (MLU < 2.0) with all nouns and verbs converted " " to nonsense(e.g., " Rom GORPS that the rivenftak is grum , Can vany GORP the blicket ?" ) . Thus the input stimuli were designed to permit a strong test of the frame range hypothesis - no scenes, no content words , just syntactic frames. The subjects correctly identified 52% of the maternal verbs under this condition . This level of performance compares very favor ably to the 7% correct performance based on scene observation, the 13% level achieved by knowing only the nouns in the sentence, and the 28% correct performance achieved in the presence of nouns-plus-scenes. Not ' ' only does this show that some ideal frame range is informative for verb identification (as in Fisher et al. 1991) . It shows - by using as the stimulus set real maternal sentenceswhose content words have been converted to nonsense- that the frame range provided by mothers to infants under age two is sufficient to support a good measure of verb identification . In particular , subjects provided with frame-range information had no difficulty in conjecturing mental verbs. For these (e.g., want), the structural information is particularly useful while , as we discussedearlier, the sceneinformation is misleading owing to the bias to interpret scenesas depicting concrete actions.

C. Fisher et al. / Syntactic constraints

365

A difficulty with interpreting these results onto the child learning situation is that thesesubjects(when correct) were identifying old verbs that they knew, by definition : Perhaps they just looked up the frame ranges for these known verbs in their mental lexicons rather than using the frames to make semantic deductions. Becauseof this possibility , the pertinence of the findings is much more easily interpreted by inspecting the 48% of cases where the subjects failed to identify the maternal verb, guessingsomething else.16 The finding is that false guessesgiven in responseto frame-range information are semantically close to the actual verb the mother said (as assessedby the Fisher et al. semantic-similarity procedure) while false guessesin responseto sceneswere semantically unrelated to the verb the mother actually uttered. As syntactic ' bootstrapping predicts, the frame range put the subjects into the semantic ' even when neighborhood they did not allow convergenceto a unique verb construal. Note that 52% percent correct identification , while a significant improvement over 7% or 28% , is not good enough if we want to model the fact that verb learning by three-year-olds is a snap. They do not make 48% errors so far as we know , even errors close to the semantic mark. But as we have repeatedly stressed, syntactic bootstrapping is not a procedure in which the child is assumedto forget about the scene, or the co-occurring nominals, and attend to syntax alone (as Lederer et al. forced their subjects to do in this -to-world manipulation by withholding all other evidence). It is a sentence pairing procedure. Indeed, adding the real nouns to the frames without video in this experiment led to over 80% correct verb identification ; adding back the sceneyielded almost perfect performance. So if the child has available (as she does, in real life ) multiple paired scenesand sentences , we can at last understand why verb learning is easy. Collateral evidence from children for the use of multiple frames is at present thin , largely becauseit is difficult to get young children to cooperate while a lengthy set of structures/ scenesis introduced. (For this reason, the experiment presented in this article settled for studying the effect of single structures on the interpretation of single scenes , though we interpret the findings as a snapshot of an iterative process.) Supportive evidence comes from Naigles et al. ( 1993), who found that young children will alter their interpretation of known verbs in responseto hearing them in novel syntactic 16 This is analogousto the findingsof our experimentasorganizedin table5. Therewe looked only at instanceswherethe childrendid not comeup with the known verb, but indicatedsome newconstrualthrougha paraphrase . The findingswerethe sameas for the single-word glosses .

366

C. Fisher et a/. / Syntactic constraints

willnot(forareplication older children and adults environments , while usually oftheframe extension etale1992 and ,expansion , see ). Evidently Naigles in thelearning istaken asevidence foralteration oftheconstrual early range ' the word s isset but after extensive for that word , meaning experience process . theconstrual thesyntax loses itspotency tochange and 5.3. Whatsemanticcluesresidein thesyntax? We have suggestedthat the fonnal medium of phrase structure constrains the semantic content that the sentenceis expressing, thus providing crucial clues to the meaning of its verb. One such clue resides in the number of arguments: A noun phrase position is assignedto each verb argument; this will differentiate push from fall . Another concerns the positioning of the arguments: the subject of transitives is the agent, differentiating chasefrom flee . The case-marking and type of the argument also matters, e.g., verbs whose meaning allows expression of paths and locations typically accept prepositional phrases( The rabbit puts the blanket on the monkey; Jackendoff 1978, Landau and Jackendoff, in press), and verbs that expressmental acts and statesaccept sentential complements(John thinks that Bill is tall , Vendier 1972) . Of course one cannot converge on a unique construal from syntactic properties alone. Since the subcategorization properties of verbs are the ' syntactic expressionsof their arguments, it is only those aspectsof a verb s meaning that have consequencesfor its argument structure that could be - semantic distinctions are not represented in the syntax. Many - most fonnally expressedwith this machinery. An important example involves the manner in which an act is accomplished, e.g., the distinctions between slide, roll and bounce, which are not mapped onto differences in their syntactic behavior (Fillmore 1970) . All theseverbs require as one argument the moving entity and allow the causal agent and path of motion as other arguments; hence, The ball slid. rolled. bounced ( down the hill ) .. Kimberley slid. rolled. bouncedthe ball ( down the hill ) . The specificmanners of motion are expressed within the verb rather than surfacing as distinctions in their syntactic ranges. In sum, it is only th~ meaning of a verb as an argument-taking predicate that can be representedby the surface phrase structures (Rappaport et ale 1987, Fisher et ale 1991, Fisher, in press) . The structures can therefore reveal only certain global properties of the construal, such as whether the verb can express inalienable (intransitive) , transfer (ditransitive ), mental/perceptual (inflected sentencecomplement), and symmetrical (sensitivity of the frame to

C. Fisher et aI. I Syntactic constraints

the number of one of its arguments) contents, and whether it expresses an activity (progressive) or a state (simple present) . Overall, our view is not that there are ' verb classes', each of which has semantic components and (therefore) licensescertain structures. Rather we suggestthat verb frames have semantic implications (truth values), and verbs have meanings. Owing to the meaning of the verb, it will be uncomfortable and thus rarely or never uttered - in some frame, e.g., we don ' t say " Barbara looked the ball on the table" becauseno external agent can cause a ball to move just by looking at it (that would be psychokinesis). If the circumstances warrant , however, look can and will be used unexceptionally in this frame; for example, the rules of baseball make it possible to say (and sports announcers do say) " The shortstop looked the runner back to third base" . As for learners, we believe they note the frame environments in which verbs characteristically occur, and thus the argument structures with which their meanings ' ' typically comport . These ranges of typical structures are compatible with only small sets of verb meanings. Because the formal medium of phrase structure is revealing only of a restricted set of semantic properties, we cannot and have not argued that the verb mappings are learned ' from ' the syntax. Indeed we have just made clear that what most people think of as the ' meaning' (that open concerns being ajar while closeconcerns being shut) is nowhere to be found in the syntax of sentences . Rather, we have shown that the initial narrowing of the searchspacefor that meaning, by attention to the argument structure as revealedby the syntax, is the precondition for using the sceneinformation efficiently to derive the meaning. When babiesdo not appear to know the phrase structure, they learn few verbs; when adults and young children are required to identify verbs without phrase structure cues (as when told " Look ! Ziking !" or when presentedwith silent videos of mother- child conversation) again they do not convergeto a unique interpretation . We conclude that the phrase structure is the leamer' s version of a zoom lens for verb vocabulary acquisition.l ' 17 Pinker(1984 that, in therelativelyadvanced child, phrasestructuralinformation ) hypothesized couldbeusedfor anotherpurpose : to assign abstract words , thosethatareneitherthings noracts(e.g., situation , know suchasnounandverb: a newitemthatoccun ) to lexicalcategories in a verbpositionin thestructure is, in virtueof thatposition , a verb; andsoforth. Pinkertermed 'structure ' . This seems this procedure distributionallearning . But this dependent plausible will giveno clueto theverbmeaning otherthanthat thewordmeans'something procedure , ' distinctions , thepositionweadoptallowssemantic withintheverbclassto be verby. In contrast ' verbs because . Knowcanbe assigned extracted to theclassof 'mental it ~ pts tensed just sentence . Thisgrosssemantic classification complements , the burdenon observation accomplished is still to distinguish , realizeandsoforth. amongthink,know

368

C. Fisher et at. I Syntactic constraints

5.4. Quirks , provisos , and limitations The usefulness of form - to - meaning correspondences for verb learning is limited by several factors :

5.4.1. Language-specific linkages of syntax to semantics By no means all functions from syntax to semanticsare universal. There are within -languagequirks in thesemappings; for example, paths are usually encoded via prepositional phrases in English (e.g., come into the room) but occasionally not (e.g., enter the room) ; seeGruber 1965. Moreover, there are some systematic differencesin the mappings across languages(Talmy 1985) . It would be incoherent to suppose that language-specific correspondences (which themselvesmust be learned) could serve as input to verb learning at early stages. But linguistic- descriptivefindings due to Gruber ( 1965), Fillmore ( 1968), Levin ( 1985), Jackendoff ( 1972, 1978, 1990), Talmy ( 1985), Pinker ( 1989), and experimental evidencefrom Fisher et al. ( 1991) and Geyer et al. (forthcoming ) suggestthat the correspondencerules are broad in scope and sufficiently stable cross-linguistically to support a good measure of verb vocabulary learning. 5.4.2. Arguments, adjuncts, and phrase boundaries Another severe problem for our approach has to do with the syntactic analysis that children could perform on the sentencesheard. After all , I saw the book on the table must be analyzed differently from I put the book on the table. Otherwise see and put will be assumed to have the same number of argument positions, falsely suggestinga similarity in their conceptual structure . How is the correct parse of an utterance to be achieved? Practically speaking, the very short sentencesused to novices contain such structural ambiguities only very rarely. Moreover, there is evidence that height of attachment is robustly cued by prosody in speech to novices (Lederer and Kelly 1991) . All the same, the principled difficulty of deriving the intended parse from observation of a spoken sentence certainly complicates the proposed learning procedure (seeGrimshaw, this volume) . 5.4.3. Theproblem of polysemy , If a single phonological object has more than one, and unrelated, senses . Consider we have the for envisaged this createsanother problem procedure ' ' ' ' relate in the sensestell and conceptually connect . Such an item will have a variety of subcategorizationframes, but some of thesewill be consequencesof

C. Fisher et at. / Syntactic constraints

one of the meanings and others the consequenceof the other. Putting them together as a single frame-range should lead to chaos. The degree to which polysemy reducesthe plausibility of the use of multiple frames is unknown in detail (seeGrimshaw, this volume, for a pessimisticview) . However, there is some suggestiveevidence that the frame-ranges of verbs are well-correlated with their meaningsin the general case, despite this problem. The manipulations of interest in this regard were carried out in English by Fisher et al. ( 1991) and in Hebrew by Geyer et al. (forthcoming) . One group of subjects provided the frame ranges for a set of common verbs (they gave judgments of grammaticality of all the verbs in various syntactic environments ) . A second group of subjectsprovided semantic-relatednessjudgments for these verbs presented in isolation (with no syntactic context) . The question was whether the overlap in frame-ranges predicted the semantic relatednessamong the verbs. The answer is yes, massively- and in materially ' the same way for English and for Hebrew. The more any two verbs overlapped in their syntactic privileges, the closer they were judged to be in their meanings. Evidently , overlap in frame range provides a guide to semanticrelatednessthat (though probabilistic) is stable enough to contribute to the verb-learning feat. The three problems just described - variability of the mapping relations, alternate parsesfor input sentences , and polysemy - limit or at least complicate the. potential effectivenessof the procedure we have called syntactic bootstrapping. Thus any linguist or psychologist worth his or her salt can find counterexamples to the claim that frame-information , or frame-range information , always and perfectly predicts the relevant (argument-taking) properties of the verbs - just as counterexamples to the usefulness of situational information are easy to find. We must suppose, in consequence , that the learner draws on convergent cuesfrom prosody, syntax, and situation , as available, jiggling them all across instances to achieve the best fit to a lexical entry . That is, the internal structure of the child ' s learning procedure is likely to be quite mixed in the information recruited and probabilistic in how such information is exploited, sad as this seems. In the work presented, we could show only that syntactic evidence is on theoretical grounds crucial for working out certain mapping problems (those that involve perspective-taking verbs) and indeed is used by youngsters solving for these under some exquisitely constrained laboratory conditions. We take the outcomesto lend plausibility to the overall approach.

C. Fisher et al. / Syntactic constraints

6. Concl _ io-

We have proposed a learning procedure for verbs which requires that children be armed with at least some innate semantic/syntactic correspondence rules, and considerable abilities and dispositions to perform formal analyses on the speech they hear. As such, these ideas have often been rejected as too formidable to be used by babies sometimes by the same commentators who invoke highly abstract formal principles to account for ' the child ' s acquisition of syntax. In Pinker s ( 1989: 263- 264) words, if the ' mechanismis used at all it is used ... as a , sophisticated syntactic bootstrapping form of cognitive problem solving rather than a general interpretive ' ' ' linguistic scheme, a kind of riddle solving . In contrast, there is something so tangible and appealing to introspection about the idea of parsing of ongoing events that this is widely acceptedas a sufficient basis for lexical learning. But theseIntuition -derived theoretical blasescannot so lightly be accepted. They have fooled us before. For example, there is strong cross linguistic as at least are olds evidence that two year probably quicker - to quick their semanticaspects(Levy extract the formal aspectsof gender as to extract 1983) . Gordon ( 1985) has shown that young children are more attentive to the formal distinction between mass and count nouns in English than to the semanticcorrelates of this distinction . ' ' The present experiment documented only the focusing ( zoom lens ) aspect of the syntactic bootstrapping procedure. The successivenarrowing of the ' semanticconjecture that derives from observation of a verb s severallicensed structural environments was not tested, though prior experimentation we have cited demonstrates both the strength of these relations and use of this procedure by adults and children in verb identification . It is rather more surprising that there is little systematic evidence for the word -to-world pairing procedure either. An interesting exception is Gropen et al. ( 1991) . Children in this experiment heard new motion verbs while shown several example scenesalong with the syntactically uninformative sentence " " This is pilking . Most of them learned the verb meanings(though some did not) . Unfortunately , the teaching procedure included negative instances " " ( This is not pilking ) and specific correction wheneverthe children erred. So far as we know , such explicit negative evidence is not usually available to learners.is Still , no one can doubt that a salient motion (e.g., zigzagging in 18 For Gropenet aIls purposesthis unusualteachingenvironmentdid not matter, Their aim wasto showthat if the childrendid learnthe verb, theycouldprojectits argumentstructureand

C. Fisher et af. I Syntactic constraints

this experiment) can sometimesbe mappedagainst a spokenverb from the bare evidenceof observation. It is the overwhelming fallibility of such a word -to-sceneprocedure that we have emphasizedin this article. Therefore we have challenged the logic of observation alone as the input to verb learning. One such challenge is that too many verbs come in pairs that are just about always mapped onto the same situations, so cross-situational observation will never distinguish between them. Another is that some verbs encode concepts that are not observable at all , e.g., know or want. Another is that considerations of salience(the agencybias) are fatal to the possibility of verb learning in all the ' caseswhere the caretaker happensto utter some word which is not the most ' salient of all for the scenethen in view. To help redress these logical and practical problems we have shown that children can make significant use of structural information . At the sametime their construals of new verb meanings are affected by blases as to how to represent an event. But this influence from plausibility considerations acknowledged , the influence of structure is materially stronger and wins out in the majority of cases. Toddlers know that it is better to receive than to give when Santa is the indirect object of the sentence.

References . What someconceptsmight not be. ArmstrongS ., L.R. Gleitman and H. Gleitman, 1983 . 13 263 208 Cognition , esto grammar.In: E. Wanner, L.R. . Functionalistapproach Bates , 1982 , E. and B. MacWhinney Gleitman(eds.), Languageacquisition: The stateof the art, 173- 218. New York: Cambridge . UniversityPress . Paper . The acquisitionof non-basicsentences Beckwith , R., E. Tinker and L. Bloom, 1989 on LanguageDevelopment at the BostonUniversityConference , Boston. presented . Semanticfactorsin the acquisitionof rules for word useand sentence Bowerman , M., 1976 . In: D.M. Morehead construction , A.E. Morehead(eds.), Normal and deficientchild language . . Baltimore, MD : UniversityPark Press . In : es in lexical and syntacticdevelopment . Reorganizationalprocess Bowerman , M., 1982 E. Wanner, L.R. Gleitman(eds.), Languageacquisition: The stateof the art, 319- 346. New . York: CambridgeUniversityPress

. The answerwaspositive: Knowledgeof meaning henceits probablesurfacestructureprivileges ) just as the latter (again, predictssurfacestructure(though errorfully, see Bowerman1982 errorfully) predictsthe former.

C. Fisheret alI Syntacticconstraints Braine, M. and J. Hardy, 1982.On what casecategoriesthereare, why they are, and how they , speculation , and evidencefrom children. In : develop: An amalgamof a priori considerations E. Wanner, L.R. Gleitman(eds.), Languageacquisition: The stateof the art, 219- 239. New York: CambridgeUniversityPress . Brown, R., 1957 . Linguisticdeterminism . Journalof Abnormaland Social and the part of speech 55, 1- 5. Psychology Bruner, J.S., 1975 . From communication to language : A psychological . Cognition3, perspective 255- 287. . Child' s talk. New York: Norton. Bruner, J.S., 1983 N. 1957 . Syntacticstructures . New York: Mouton Publishers . , , Chomsky . Lectureson governmentand binding. Dordrecht: Foris. , N., 1981 Chomsky . The principle of contrast: A constraint on languageacquisition Clark, E.V., 1987 . In: B. MacWhinney(ed.), Mechanisms of languageacquisition , 264- 293. Hillsdale, NJ: Erlbaum. Clark, H.H. andJ.S. Begun, 1971 . Thesemantics of sentence . Languageand Speech , 14, subjects 34- 46. -Cooper, 1980 . Syntaxand speech . Cambridge , MA : HarvardUniversity Cooper, W. and J. Paccia . Press . Thematicproto-rolesand argumentselection . Language67(3), 547- 619. Dowty, D., 1991 Dromi, E., 1987 . Early lexicaldevelopment . New York: CambridgeUniversityPress . ' Fernald, A. andT. Simon, 1984 . Expandedintonationcontoursin mothers . speechto newborns Developmental 20( 1), 104- 113. Psychology . Lexicalentriesfor verbs. Foundationsof language4, 373- 393. Fillmore, C.J., 1968 . The grammarof hitting and breaking. In: R. Jacobs Fillmore, C.J., 1970 , P. Rosenbaum (eds.), Readingsin Englishtransformationalgrammar, 120- 133. Waltham, MA : Ginn. Fisher, C., in press. Structureand meaningin the verb lexicon: Input for a syntax-aidedverb . Languageand CognitiveProcess es. learningprocedure ' useof structuralcuesto verb . Preschoolers Fisher, C., 1993 . Paperpresented at the 60th meaning Annual meetingof the Societyfor Research in Child Development , NewOrleans , LA. Fisher, C. and H. Tokura, in press . Prosodyin speechto infants: Direct and indirectcuesto syntacticstructure.In: J. Morgan, C. Demuth(eds.), Signalto syntax. Hillsdale, NJ: Erlbaum. Fisher, C., H. Gleitmanand L. R. Gleitman, 1991 . On the semanticcontentof subcategorization frames . CognitivePsychology23, 331- 392. Fodor, J.A., 1981 . The presentstatus of the innatenesscontroversy . In: J.A. Fodor (Ed.), . Cambridge . , MA : MIT Press Representations Fritz, J.J. and G. Suci, 1981 . Facilitationof semanticcomprehension at the one-word stageof . Journalof Child Language9, 31 39. languagedevelopment Gentner, D., 1978 . On relationalmeaning : The acquisitionof verbmeaning . Child Development 49, 988-998. uentner, D., 1982 . Why nounsare learnedbeforeverbs: Linguistic relativity versusnatural , thoughtand , Volume2: Language partitioning. In: S.A. Kuczaj(ed.), Languagedevelopment culture, 301- 334. Hillsdale, NJ: Erlbaum. . Whenprosodyfailsto cuesyntacticstructure: Gerken, L., P.W. JusczykandD.R. Mandel, 1993 Nine-month-olds' sensitivityto phonologicalvs. syntacticphrases . Ms., State Universityof New York at Buffalo. asa predictorof verb Geyer, J., L.R. Gleitmanand H. Gleitman, forthcoming.Subcategorization : . . Evidence from modem Hebrew Ms. of , University Pennsylvania meaning

C. Fisheret al. / Syntacticconstraints

373

Gillette, J.A., 1992 . Discoveringthat /think/ means' think' : Theacquisitionof mentalverbs. Ms., . Universityof Pennsylvania . Effectsof situationalcueson the identificationof Gillette, J.A. and L. R. Gleitman, forthcoming nounsand verbs. . The pragmaticsof word order: Predictability Givon, T., 1986 , importance , and attention. : Benjamins . Amsterdam Gleitman, L.R., 1990 . The structuralsourcesof verb meanings . LanguageAcquisitionI , I- 55. . Wherelearningbegins: Initial Gleitman, L.R., H. Gleitman, B. Landauand E. Wanner, 1988 for languagelearning. In: E. Newmayer(ed.), The CambridgeLinguistic representations . , Vol. III : 150- 193. New York: CambridgeUniversityPress Survey . Languagein the two-year-old. Cognition Goldin-MeadowS., M. Seligmanand R. Gelman, 1976 4, 189-202. . C. Jacquetand K. Hirsh-Pasek . Lexicalprinciplesunderlythe Golinkoff, R.MR , in progress . learningof verbs. Ms., Universityof Delaware . Evaluatingthe semanticcategorieshypothesis : The caseof the count/mass Gordon, P., 1985 distinction. Cognition20, 209- 242. . Form, function and the languageacquisitiondevice. In : Ct . Baker, J.J. Grimshaw , J., 1981 , 165- 182. Cambridge , MA : MIT McCarthy(eds.), Thelogicalproblemof languageacquisition . Press . Argumentstructure. LinguisticInquiry Monograph18. Cambridge Grimshaw , J., 1990 , MA : . MIT Press and direct objects: Gropen, J., S. Pinker, M. Hollanderand R. Goldberg, 1991. Affectedness The role of lexical semanticsin the acquisitionof verb argumentstructure. Cognition 41, 153- 195. . Cognition25, . Prosodicstructureand spokenword recognition , F. and J.P. Gee, 1987 Grosjean 135- 155.. . Studiesin lexicalrelations. PhiD. dissertation Gruber, J.S., 1965 , MIT . . Basic-levelindividuals. Cognition48, 199-221. Hall, D.G., 1993 . Assumptionsabout word meaning: Individuationand basic Hall, D.G. and S. Waxman , 1993 . 64, 1550 - 1570 levelkinds. Child Development . Languagecomprehension : A newlook at someold Hirsh-Pasek , K. and RiM . Golinkoff, 1991 -Kennedy . In: N. Krasnegor themes , R. Schiefelbusch , D. Rumbaugh , M. Studderty (eds.), Erlbaum . 301 320 . Hillsdale NJ: and behavioral of , , Biological aspects languageacquisition . Syntactic Hirsh-Pasek , K., H. Gleitman, L. Gleitman, R. Golinkoff and L. Naigies, 1988 . Paper presentedat the Boston University bootstrap ping: Evidencefrom comprehension Conference on LanguageDevelopment , Boston. . Hirsh-Pasek , 1987 , K., D. Kemler-Nelson, P. Jusczyk , K. Cassidy , B. Drussand L. Kennedy Clausesare perceptualunits for younginfants. Cognition26, 269- 286. . Semanticinterpretationin generativegrammar. Cambridge Jackendoff , R., 1972 , MA : MIT . Press . Grammarasevidence for conceptualstructure.In: M. Halle, J. Bresnan Jackendoff , R., 1978 , G. Miller (eds.), Linguistictheory and psychologicalreality, 201- 228. Cambridge , MA : MIT . Press . . Semanticstructures . Cambridge Jackendoff , R., 1990 , MA : MIT Press . Ms., . Dependency JoshiA . and O. Rambow , in progress parsingfor phrasestructuregrammars . Universityof Pennsylvania

374

C. Fisher et alI

Syntactic constraints

, A. Woodward and , P.W., K. Hirsh-Pasek , DiG. Kemler-Nelson, L.J. Kennedy Jusczyk . Perceptionof acousticcorrelatesof major phrasalunits by young infants. J. Piwoz, 1992 24, 252- 293. CognitivePsychology -Nelson, in press. Syntactic units, prosody, and psychological P. and D. Kemler Jusczyk, reality during infancy. In : J. Morgan, C. Demuth (eds.), Signalto syntax. Hillsdale, NJ: Erlbaum. . Preschoolers Kako, E., in progress assignnovel verbs to novel actions. Ms., University of . Pennsylvania . A restrictivetheoryof word orderand phrasestructure.Paperpresented at the Kayne, R., 1992 Annual Meetingof GLOW, Lisbon, Portugal. ' . In: C.N. Li ed. Keenan . Toward a universaldefinitionof ' subject , E.L., 1976 ( ), Subjectand . topic, 305- 333. New York: AcademicPress . Vowel lengtheningis syntacticallydetenninedin a connecteddiscourse . Klatt, D.H., 1975 Journalof Phonetics3, 129- 140. : Evidencefrom the blind child. Landau, B. and L. R. Gleitman, 1985. Languageand experience . , MA : HarvardUniversityPress Cambridge . 'What' and ' where' in spatial languageand spatial Landau, B. and R. Jackendoff , in press . cognition. Brain and BehavioralSciences . Prosodiccorrelatesto the adjunct/complementdistinctionin Lederer , A., and M. Kelly, 1991 motherese . Papersand Reportsin Child LanguageDevelopment 30, Stanford, CA.. of cross-situationaland Lederer , A., H. Gleitmanand L.R. Gleitman, 1991.The infonnativeness cross-sententialevidencefor learningthe meaningof verbs. Paperpresentedat the Boston on LanguageDevelopment , Boston. UniversityConference . The syntacticcontextsof maternalverb Lederer , A., H. Gleitmanand L.R. Gleitman, in press ' use. In: M. Tomasello , W. Merriman (eds.), Beyondnamesfor things: Young childrens acquisitionof verbs. Hillsdale, NJ: Erlbaum. . The role of durationin disambiguating Lehiste , I., J.R. Oliveand L. Streeter , 1976 syntactically - 1202 . . Journalof the AcousticalSocietyof America60( 5), 1199 ambiguoussentences . Biologicalfoundationsof language . New York: Wiley. , E.H., 1967 Lenneberg . The perceptionof causalityin infants. PerceptionII , 173- 186. Leslie, A.M., 1982 Leslie, A.M. and S. Keeble, 1987. Do six-month-old infants perceivecausality? Cognition25, 265- 288. . Lexical semanticsin review: An introduction. In: B. Levin (ed.), Lexical Levin, B., 1985 semanticsin review. LexiconProjectWorking Papers , MA : The MIT , I : 1- 62. Cambridge . Centerfor CognitiveScience ' Levy, Y., 1983.It s frogsall the way down. Cognition15, 75-93. of the Mandler, J., 1991 . Prelinguisticprimitives. In: L.A. Sutton, C. Johnson(eds.), Proceedings -=-425. Berkeley 17th Annual Meetingof the BerkeleyLinguisticsSociety , 414 , CA: Berkeley . LinguisticsSociety . Categorization and namingin children: Problemsof induction. Cambridge Markman, E., 1989 , MA : MIT Press . . How can a grammaticalparameterbe set before the first word? Paper Mazuka, R. 1993 at the , RI. , BrownUniversity, Providence Signalto SyntaxConference presented . The perceptionof causality. London: Methuen. Michotte, A., 1963 . Journalof Child Language17, Naigies, L., 1990. Childrenusesyntaxto learn verb meanings 357- 374.

C. Fisher et aI. I Syntactic constraints

shifts in the constructionof verb . Developmental , L., A. Fowler and A. Helm, 1992 Naigles 7 . 403 427 . , meaningsCognitiveDevelopment . Childrenacquireword meaningcomponents , L., H. Gleitmanand L. R. Gleitman, 1993 Naigles . In: E. Dromi (ed.) Languageand development from syntacticevidence , 104- 140. Norwood, NJ: Ablex. . Structureand strategyin learningto talk. Monographsof the Societyfor Nelson, K., 1973 38. Research in Child Development . : Interrelationsin acquisitionand development . Concept Nelson, K., 1974 , word, and sentence Review81, 267- 285. Psychological . Motherese : The speechof mothersto youngchildren. In: N. Castellan , Newport, E.L., 1977 D. Pisoni, G. Potts(cds.), CognitiveTheory, Volume2. Hillsdale, NJ: Erlbaum. . Journalof Child Language . Ostensivedefinitionin vocabularyacquisition Ninio, A., 1980 , 7, 565- 573. . Cambridge . Languagelearnabilityand languagedevelopment Pinker, S., 1984 , MA : Harvard . UniversityPress . . Learnabilityand cognition. Cambridge Pinker, S., 1989 , MA : MIT Press . MA : MIT Press . W. V. . Word and 1960 , , , Quine object Cambridge . Lexicon . Levelsof lexical representation , 1987 , M., B. Levin and M. Laughren Rappaport . , MA : MIT Centerfor CognitiveScience ProjectWorkingPapers# 20. Cambridge . Paperpresentedat the CUNY . Selectionalpreferenceand implicit objects Resnick , P., 1993 Conference Sentence , Amherst, MA. Processing . In: Y. Levy, I. Schlesinger . The origin of relationalcategories , M. Braine , I., 1988 Schlesinger . NJ: Erlbaum . Hillsdale 121 178 in and es . eds , , language acquisition ( ), Categories process . On the natureof talk to children. In: E.H. Lenneberg Slobin, D.I., 1975 , E. Lenneberg(cds.), . New York: Holt, Rinehartand Winston. Foundationsof languagedevelopment : . L. 1985 Lexicalization , , patterns Semanticstructurein lexicalforms. In: T. Shopen(ed.), Talmy , Volume 3: Grammaticalcategoriesand the Languagetypology and syntacticdescription . lexicon. New York: CambridgeUniversityPress . Joint attentionon actions: Acquiringverbsin ostensive Tomasello , M. and A.C. Kruger, 1992 . Journalof Child Language19, 311- 333. contexts and non-ostensive . Ithaca, NY : CornellUniversity . Vendier,Z., 1972Rescogitans:An essayin rationalpsychology . Press . LinguisticInquiry II (2), 223-233. . In a mannerof speaking Zwicky, A., 1971

) 377- 410. North-Holland Lingua92 ( 1994

How could a child use verb syntax to learn verb semantics?* Steven Pinker Institute of Technology , Massachusetts , 10-016, Departmentof Brain and CognitiveSciences " MA02139 Cambridge , USA

I examine Gleitman' s ( 1990) arguments that children rely on a verb' s syntactic subcategorization frames to learn its meaning (e.g., they learn that seemeans ' perceivevisually' becauseit can appear with a direct object, a clausal complement, or a directional phrase). First , Gleitman argues that the verbs cannot be learned by observing the situations in which they are used, becausemany verbs refer to overlapping situations, and becauseparents do not invariably use a verb when its perceptual correlates are present. I suggestthat these arguments speak only against a narrow associationist view in which the child is sensitive to the temporal contiguity of sensory features and spoken verb. If the child can hypothesizestructured semantic representationscorresponding to what parents are likely to be referring to , and can refine such representationsacross multiple situations, the objections are blunted ; indeed, Gleitman ' s theory requires such a learning process despite her objections to it . Second, Gleitman suggeststhat there is enough information in a ' verb s subcategorization frames to predict its meaning ' quite closely' . Evaluating this argument ' requires distinguishing a verb s root plus its semantic content (what She boiled the water shares with The water boiled and does not share with She broke the glass), and a verb frame plus its semanticperspective(what Sheboiled the water shareswith Shebroke the glassand does not share with The water boiled) . I show that hearing a verb in a single frame only gives a learner coarse information about its semantic perspective in that frame (e.g., number of arguments, type of ' arguments); it tells the learner nothing about the verb root s content across frames (e.g., hot . Moreover a verb across all its frames also revealslittle about the verb , hearing bubbling liquid ) root ' s content. Finally , I show that Gleitman ' s empirical argumentsall involve experimentswhere children are exposed to a single verb frame, and therefore all involve learning the frame' s ' perspectivemeaning, not the root s content meaning, which in all the experiments was acquired ' by observing the accompanying scene. I conclude that attention to a verb s syntactic frame can ' help narrow down the child s interpretation of the perspectivemeaning of the verb in that frame, but disagreewith the claim that there is some in-principle limitation in learning a verb' s content

. Preparationof this paperwassupportedby NIH Grant HD 18381and NSF Grant BNS 9109766 . The ideasand organizationof this paper wereworked out in collaborationwith Jane Grimshaw on Language , and werepresented jointly at the 1990BostonUniversityConference . I thank Paul Bloom , JessGropen, Gary Marcus, an anonymousreviewer , and Development and comments . especiallyLila Gleitmanfor helpfuldiscussions - 3841/94/ $07.00 @ 1994- ElsevierScienceB.V. All rightsreserved 0024 - 3841( 93) EO044- 8 SSDJO024

378

S. Pinker / Verb syntax and verb semantics

' from its ~ituatinn ~ of use that could only be resolved by using the verb s set of subcategorization frames.

' 1. Introduction: The problem of learning words meanings When children learn what a word means, clearly they must take note of the circumstancesin which other speakersuse the word . That is, children must learn rabbit becausetheir parents use rabbit in circumstancesin which the child can infer that they are referring to rabbits. Equally obviously, learning word meaningsfrom circumstancesis not a simple problem. As Quine ( 1960), among others, has noted, there are an infinite set of meanings compatible with any situation , so the child has an infinite number of perceptually indistinguishable hypothesesabout meaning to choose among. For example, all situations in which a rabbit is present are also situations in which an animal is present, an object is present, a furry thing is present, a set of undetachedrabbit parts are 'present, a something-that -is-either-a-rabbit -ora Buick is present, and so on. So how does the child figure out that rabbit ' means ' rabbit ' , not ' undetachedrabbit part ? Word learning is a good example of an induction problem, where a finite set of data is consistent with an infinite number of hypotheses, only one of them correct, and a learner or perceiver must guess which it is. The usual explanation for how people do so well at the induction problems they face is that their hypotheses are inherently constrained: not all logically possible hypothesesare psychologically possible. For example, Chomsky ( 1965) noted that children must solve an induction problem in learning a language: there are an infinite number of grammars compatible with any finite set of parental . They succeed, he suggested sentences , because their language acquisition circuitry constrains them to hypothesize only certain kinds of grammatical rules and structures, those actually found in human languages, and because the kinds of sentenceschildren hear are sufficient to discriminate among this small set of possibilities. In the case of learning word meanings, too , .not all logically possible construals of a situation can be psychologically possible candidates for the ' meaning of a word . Instead, the hypotheses that a child s word learning mechanismsmake available are constrained in two ways. The first constraint comes from the representational machinery available to build the semantic ' structures that constitute mental representations of a word s meaning: a ' Universal Lexical Semantics, analogous to Chomsky s Universal Grammar

S. Pinker / Verb syntax and verb semantics

379

(see, e.g., Moravscik 1981, Markman 1989, 1990; Jackendoff 1990) . For ' ' example, this representational system would allow object with shapeX and ' ' ' object with function X as possibleword meanings, but not all the undetached ' ' parts of an object with shapeX , object with shapeX or a Buick' , and ' object ' and the surfacesit contacts . The second constraint comes from the way in which a child ' s entire lexicon may be built up ; on how one word ' s meaning ' may be related to another word s meaning (see Miller 1991, Miller and Fellbaum 1992). For example, the lexicons of the world ' s languages freely allow meronyms (words whose meanings stand in a part- whole relationship, like body- arm) and hyponyms (words that stand in a subset- superset relationship , like animal- mamma/), but do not easily admit true synonyms Bo ( linger 1977, Clark 1987, Miller and Fellbaum 1991) . A child would therefore not posit a particular meaning for a new word if it was identical to ' some existing word ' s meaning. Finally , the child would have to be equipped with a procedure for testing the possible hypotheses about word meaning against the situations in which adults use the words. For example, if a child ' ' thought that pet meant dog , he or she will be disabusedof the error the first time the word is used to refer to a fish. Although the problem of learning word meaningsis usually discussedwith regard to learning nouns, identical problems arise with verbs (Landau and Gleitman 1985, Pinker 1988, 1989; Gleitman 1990) . When a parent comments on a dog chasing a cat by using the word chase, how is the child to know that it means ' chase' as opposed to ' flee' , ' move' , ' go' , ' run ' , ' be a dog chasing' , ' chaseon a warm ' day , and so on? As in the case of learning noun meanings (indeed, learning in general), there must be constraints on the child ' s possible hypotheses. For example, manner-of -motion should be considered a possible component of a verb' s mental dictionary entry, but temperature-during -motion should not be. (See Talmy 1985, 1988; Pinker 1989, Jackendoff 1990, and Dowty 1991, for inventories of the semantic elements and their configurations that may constitute a verb' s semantic representation.) Moreover, there appear to be constraints on lexical organization (Miller 1991, Miller and Fellbaum 1991) . For example, verb lexicons often admit of co-troponyms (words that describe different manners of perfonning a similar act or motion , such as walk-skipjog) but , like noun lexicons, rarely admit of exact synonyms ( Bolinger 1977, Clark 1987, Pinker 1989, Miller and Fellbaum 1991) . Finally , the child must be equipped with a learning mechanism that constructs, tests, and modifies semantic representationsby comparing infonnation about the uses of verbs by other speakersacrossspeechevents (Pinker 1989) .

380

S. Pinker/ Verbsyntaxandverbsemantics

1.1. A novel solution to the word-learning problem In recent years Lila Gleitman and her collaborators have presenteda series of thorough and insightful discussionsof the inherent problems of learning verbs' meanings (Landau and Gleitman 1985, Hirsh -Pasek et ale 1988, Gleitman 1990, Naigles 1990, Lederer et ale 1989, Fisher et ale 1991and this volume) . Interestingly, Gleitman and her collaborators depart from the usual ' solution to induction problems, namely, seeking constraints on the learner s ' hypotheses and their relation to the learner s input data as the primary explanation. Rather, they argue that the learner succeedsat learning verb semanticsby using a channel of information that is not directly semantic at ' all. Specifically, they suggestthat the child infers a verb s meaning by using the kinds of syntactic arguments (direct object, clause, prepositional phrase) that appear with the verb when it is used in a sentence. Such syntactic ) are referred to in properties (e.g., whether a verb is transitive or intransitive ' ' ' ' ' various literatures as the verb s argument structure , argument frame , ' ' ' ' syntactic format , and subcategorization frame . 'Indeed, Gleitman and her collaborators argue that information about a verb s semantics, gleaned from observing the circumstances in which other speakers use the verb (e.g., ' ' learning that open means opening becauseparents use the verb to refer to opening things) is in principle inadequate to support the acquisition of the verb' s semantics; cues from the syntactic properties of the verb phrase are essential. This position has its roots in Brown ( 1957) and Katz et ale ( 1974), who showed empirically how children use grammatical information to help learn certain aspectsof word meanings. But it was given a stronger form in Landau ' and Gleitman s ( 1985) book Language and Experience: Evidencefrom the Blind Child. Landau and Gleitman point out that a blind child they studied acquired verbs, even perceptual verbs like look and see, rapidly and with few ' errors, despite the child s severeimpairment in being able to witnessdetails of the scenesin which the verbs are used. Moreover, they noted that a sighted ' child ' s task in learning verbs is different from the blind child s task only in degree, not in kind . Since the learning of verbs like see and know cannot critically rely on information from vision, Landau and Gleitman presented the following hypothesis: 'In essence crucialcues formatsfor a verbprovides ourpositionwill bethatthesetof syntactic . esof themeanings reflex surface theseformatsareabstract to theverbmeanings just because ' verb for some is attested that format , in information ... thereis verylittle anysinglesyntactic frames . However... thesetof subcategorization for that formatserves manydistinctuses

S. Pinker / Verb syntax and verb semantics

associated witha verbis highlyinformative aboutthemeaning it conveys . In fact, sincethe surface formsarethecarriersof criticalsemantic information of verbsis partly , theconstrual indeterminant withoutthe subcategorization . Hence information , in the end, a successful for verbmeaning mustrecruitinformation frominspection learningprocedure of themany - 139 formatsin whicheachverbparticipates .' ( 1985 : 138 grammatical ) For example, here's how a child hearing the verb glip in a variety of syntactic frames could infer various components of its meaning from the characteristic semantic correlates of those frames. Hearing I glipped the book (transitive frame, with a direct object), a child could guessthat glipping is something that can be done to a physical object. Hearing I glipped that the book is on the table (frame with a sentential complement), the child could infer that glipping involves some relation to a full proposition. Hearing I glipped the book from acrossthe room (frame with an object and a directional complement) tells him or her that glipping can involve a direction. Moreover, the absenceof Glip that the book is on the table! (imperative construction) suggeststhat glipping is involuntary, and the absenceof What John did wasglip the book (pseudocleft construction) suggeststhat it is not an action. With this information , the child could figure out that glip means ' see', becauseseeingis an involuntary nonaction that can be done to an object or a proposition from a direction. Note that the child could make this inference without seeinga thing, and without seeinganyone seeinganything. In her 1990paper laying out this hypothesisin detail and -discussingthe motivation for it , Gleitman calls this learning procedure ' ' syntactic bootstrapping , and offers it as a major mechanismresponsible ' for the child s successat learning verb meanings. The goal of the present paper is to examine the general question of how a child could use the syntactic properties of a verb to figure out its semantic properties. I will discuss several kinds of mechanisms that infer semantics from syntax, attempting to distinguish what kinds of inputs they take, how they work , what they can learn, and what kind of evidencewould tell us that children use them. I will focus on Gleitman' s ( 1990) thorough and forceful arguments for the importance of syntax-guided verb learning. After she puts theseargumentsin particularly strong form in order to make the best casefor them and to find the limits as to what they can accomplish, Gleitman settles on an eclectic view in which a set of learning mechanisms, some driven by syntax and some not , complement each other. I agree with this eclectic view and will try to layout the underlying division of labor among learning mechanismsmore precisely. In doing so, I will , however, be disagreeingwith some of the particular strong claims that Gleitman makes about syntaxguided learning of meaning in the main part of her paper.

S. Pinker I Verb syntax and verb semantics

2. What is learnedfrom what: Two preliminary clarifications Sentencescontain a great deal of information, and the child is learning many things at once from them. To understand how syntax can help in learning semantics , it is essentialto be clear on what kinds of information in a sentence ' ' are and are not syntactic, and what kinds of things that a child is learning are ' ' ' and are not semantic. Before examining Gleitman s arguments, then, I make someessentialdistinctions, without which the issuesare very difficult to study. 2.1. Linguistically-conveyedsemanticcontent is not the sameas syntacticform Gleitman' s hypothesesliterally refer to the acquisition of verb meaningsvia the use of syntactic information , specifically, the syntactic properties of the arguments that the verb appears with (e.g., whether it takes a grammatical object, a prepositional object, a sentential complement, or various combinations of thesearguments in different sentences ) . Note that this is not the same as claiming that the child uses semantic information that happens to be communicated by the linguistic channel. Sentences , obviously, are used to convey real world information , and children surely can infer much about what a verb means from the meanings ' of the other words in the sentenceand from however much of the sentences structure they are able to parse. For example, if someone were to hear I ' glipped the paper to shredsor I fliped the delicioussandwichand now I m' full ,' out that glip means something like tear presumably he or she could figure ' ' and flip means something like eat . But although these inferencesare highly ' specificand accurate, no thanks are due to the verbs syntactic frames (in this case, transitive) . Rather, we know what those verbs mean becauseof the semanticsof paper, shreds, sandwich, delicious, full , and the partial syntactic analysis that links them together (partial , because it can proceed in the absenceof knowledge of the specific subcategorization requirements of the verb, which is the data source appealed to by Gleitman) . In other words, ' ' inferring that tear means tear from hearing paper and shreds is a kind of contigencies, the same one cognitive inference using knowledge of real world ' tear' when means tear that infer to seeing paper being that could be used 's verb a of meaning from its torn to shreds. It is not an example learning . For this reason, with syntactic properties, the processGleitman is concerned ' a blind (or sighted) child can learn a great deal about a verb s meaning from the sentences the verb is used in , without learning anything about the ' . meaning from the verb s syntax in those sentences

S. Pinker / Verb syntax and verb semantics

383

Moreover, some of the information about how a verb is used in a sentence is based on universal features of semantics. For example, the sentenceI am ' ' ' glipping applescould inform a learner that glip can t mean like , becausethe progressiveaspect marked on the verb is semantically incompatible with the ' stativity of liking . Here, too , one can learn something about a verb s meaning from the sentencein which the verb is used, as opposed to the situation in which the verb is used, but the learning is driven by semantic information (in this example, that liking does not inherently involve changesover time), not syntactic information . Gleitman ( 1990) does not contest this distinction ; in footnote 8 on p . 27 and in footnote 26 (p. 379) of Fisher et al. ( 1991), she states that her arguments are not about the use of linguistically -conveyed information in general, but about the use of the syntactic properties of verbs per se. Nonetheless, the .distinction has implications that bear on her arguments in ways she does not make explicit . First , the distinction blunts the intuitive impact of two of Gleitman' s recurring arguments for the importance of syntactic infoimation : that blind children learn verbs' meanings without seeingtheir referent events, and that parents do not invariably use verbs in unique situations (e.g., they do not say opensimultaneously with opening something) . These phenomenasuggestthat children must attend to what parents say, not just what they do. The phenomena do not , however, lead by some process of elimination ; to the hypothesis that children are using the syntactic subcategorization properties of individual verbs. The children may just be figuring out the content of the ' sentences , and inferring a verb s semantics from its role in the events conveyed. Second, many of the supposedly syntactically-cued inferences that Gleit man appeals to may actually be ~ediantically cued in the same sensethat hearing a verb used with sandwich suggests that it involves eating. The ' ' subcategorization frames that Landau and Gleitman ( 1985), Gleitman ( 1990), and Fisher et al. ( 1991) appeal to are distinguished more by the semantic content of particular words in them than by their purely syntactic (i.e., categorical) properties. Indeed, most of the entries are not syntactically distinct subcategorization frames in the linguist ' s sense at all. Of the 33 entries listed in Appendix A of Fisher et al. ( 1991), two thirds are actually not syntactically distinct subcategorizationframes. Seventeenframes are syntactically identical V -PP frames differing only in the choice of preposition (e.g., in NP versus on NP) . (Fisher et al. did , to be sure, collapse these prepositions into a single frame type in the data analysis of their study.) Three are V -S'

384

S. Pinker I Verb syntax and verb semantics

frames differing only in the choice of complementizers(e.g., that S versus if S). There are V -NP -PP frames differing only in the choice of preposition (e.g., NP to NP versus NP from NP ; these were, however, collapsed in the analysis) . And three are not subcategorization frames at all but the morphosyntactic constructions imperative, progressive, and pseudo-cleft, which are syntactically well-formed with any verb (though some are awkward because of semantic clashes, such as involuntary verbs in the imperative) . The ' problem is that even if learners can use verbs patterning across these linguistic contexts, it is misleading to say that they would be relying on ' syntactic information . In most modem theories of verbs compatibility with prepositions and complementizers (see Jackendoff 1987, 1990; Pinker 1989, Grimshaw 1979, 1981, 1990), the selection is made on semantic grounds: for example, verbs involving motion in a direction can selectany preposition that involves a d~rection. There are verb- specific idiosyncrasies, to be sure (such as rely on and put up with), but even these may be treated as involving idiosyncratic semantic properties of the verb. Thus if a child notices that a verb takes across and over but not with or about, and infers that the verb involves motion , the child is not using syntactic information , but figuring out that an event involving the traversal of paths (inherent to the meaning of across and over) is likely to involve motion , just as an event that involves sandwiches and hunger is likely to involve eating.1 ' ' ' ' 2.2. The term syntactic bootstrapping and the opposition of syntactic and ' ' semantic bootstrap ping are

inR mislead

' ' It is unfortunate that Gleitrnan chose the term syntactic bootstrapping to ' refer to the processof inferring a verb s meaning from its set of subcategoriza 1 Note that some of the other linguistic contexts that Landau and Gleitman call ' subcategorization frames' are not subcategorizationframes either, but frozen expressionsand collocations that are probably idiosyncratic to English and henceno basis for learning. Theseinclude Look !, See?, Look! The doggie is running!, See? The doggie is running!, Come see lhe doggie, and look like in ' the senseof ' resemble. Since look and see are the only two verbs that Landau, Gleitman , and their collaborators discussin detail, if their learning scenariosfor thesetwo verbs adventitiously exploit particular properties of English, one has to be suspicious about the' feasibility of the scenario in the general case. More generally, Fisher, Gleitman, and Gleitman s claim that there are something like 100 distinct syntactic subcategorization frames, hence, in principle, 2100 syntactically distinguishable verbs, appears to be a severe overestimate. I think most linguists would estimate the number of syntactically distinct frames as an order of magnitude lower, which would make the estimated number of syntactically distinguishable verbs a tiny fraction of what Fisher et aI. estimate.

S. Pinker I Verb syntax and verb semantics

tion frames. She intended the term to suggestan opposition to my ' semantic ' bootstrapping (Pinker 1982, 1984, 1987, 1989), and one of the sectionsin her 1990paper is even entitled ' Deciding betweenthe bootstrapping hypotheses'. ' ' Though the opposition semantic versus syntactic bootstrapping is catchy, I suggestit be dropped. The opposition is a false one, becausethe theories are theories about different things. Moreover, there is no relationship between what Gleitman calls ' syntactic bootstrapping' and the metaphor of bootstraps , so the term makes little sense. Gleitman usesthe term ' semantic bootstrapping' to refer to the hypothesis that children learn verbs' meanings by observing the situations in which the verbs are used. But this is not accurate. ' Semantic bootstrapping' is not even a theory about how the child learns word meanings. It is a theory about how the child begins learning syntax. ' The bootstrapping problem' in grammar acquisition (see Pinker 1987) arises becausea grammar is a formal system consisting of a set of abstract elements, each of which is defined with respect to other elements. For example, the ' subject' of a sentenceis defined by a set of formal properties, such as its geometric position in the tree with respectto the Sand VP nodes, its ability to force agreement with the verb, its intersubstitutability with pronouns of nominative case, and so on. It cannot be identified with any semantic role, sound pattern, or serial position . The bootstrapping problem is: How do children break into the systemat the very outset, when they know nothing about the particular language? If you know that verbs agree with their subjects, you can learn where the subjects go by seeingwhat agreeswith the verb - but how could you have learned that verbs ' agree with their subjects to begin with , if you don t yet know where the ' ' subjectsgo? How can children lift themselvesup by their bootstraps at the very outset of languageacquisition, and make the first basic discoveriesabout the grammar of their languagethat are prerequisite to any further learning? Pinker ( 1982), following earlier suggestionsof Grimshaw ( 1981), suggested that certain contingenciesbetweenperceptualcategoriesand syntacticcategories , mediated by semantic categories, could help the child get syntax acquisition started. For example, if the child was built with the universal linking rule that , and they could infer from agentsof actions were subjectsof active sentences a sentence's perceptual context and the meaningsof some of its content words that a particular word referred to the agent of an action, the child could infer that that word was in subject position . Once the position of the subject is established as a rule or parameter of the child ' s nascent grammar, further kinds of learning can proceed. For example, the child could now infer that any new word in this newly-identified position must be a subject, regardless

S. Pinker I Verb syntax and verb semantics

of whether it is an agent; he or she could also infer that verbs must agree in person and n':1mberwith the element in that position . SeePinker ( 1984) and ( 1987) for a more precisepresentation of the hypothesis. The semantic bootstrapping hypothesis does require, as a background assumption, the idea that the semantics of at least some verbs have been acquired without relying on syntax. That is becausethe theory is about how ' ' syntax gets bootstrapped at the very beginning of learning; if all word meaningswere acquired via knowledge of syntax, and if syntax were acquired ' via knowledge of words meanings, we would be faced with a vicious circle. The semantic bootstrapping hypothesis is agnostic about how children have attained knowledge of these word meanings. Logically speaking, they could have used telepathy, surgery, phonetic symbolism, or innate knowledge of the English lexicon, but the most plausible suggestion is that the children had attended to the contexts in which the words are used. Gleitman takes this ' latter assumption (that the child s first word meanings are acquired by attending to their situational contexts), generalizesit to a claim that all verb meanings are acquired by attending to their situational contexts (i.e., even verbs acquired after syntax acquisition is underway), and refers to the ' ' generalizedclaim as semantic bootstrapping . But this is a large departure from its intended meaning. ' ' And what Gleitman calls, in contrast , syntactic bootstrapping , is not a different theory of how the child begins to learn syntax . Thus it is not an alternative to the semantic bootstrap ping hypothesis. (The only reason they could be construed as competitors is that semantic bootstrap ping assumesthat at least some verb meanings can be acquired before syntax, ' so a very extreme form of Gleitman s negative argument , that no verb meaning can be learned without syntax, is incompatible with it .) Moreover ' ' , since syntactic bootstrapping is a theory of how the child learns the meanings of specific verbs, and since it can only apply at the point at which the child has already acquired the syntax of verb phrases, it is not ' ' clear what it has to do with the bootstrap ping problem or the metaphor ' of lifting oneself up by one s bootstraps . For these reasons, I suggest that the term be avoided. Here is a somewhat cumbersome, but trans~arent and accurate set of ' . ' Semanticcueing of syntax refers to the semanticbootstrapping replacements ' ' hypothesis. Semantic cueing of word meaning refers to the commonplace assumption that meaningsare learned via their semantic contexts (perceptual ' ' or linguistic) . Syntactic cueing of word meaning is the hypothesis defended by Gleitman and her collaborators.

S. Pinker I Verb syntax and verb semantics

Now , in some contexts Gleitman does present a genuine alternative to the semantic bootstrapping hypothesis. She suggeststhat the child can use the prosody of a sentenceto parse it into a syntactic tree. Though she never specifiesexactly how this could be done, presumably the child would assume that pausesor falling intonation contours signal phrase boundaries. Having thus inferred a syntactic tree, the child could infer a verb' s meaning from the trees it appears in. Note , though, that the information that the child usesto get syntax acquisition started is not itself syntactic, but prosodic; the hypothesis can thus sensibly be called ' prosodic bootstrapping' . If both prosodic bootstrapping, and syntactic cueing of word meaning were possible, semantic bootstrapping would be otiose. But while it is plausible that the infant usesprosodic information to help in sentence analysis at the outset of language acquisition (e.g., to identify utterance boundaries), it is completely implausible that this information is sufficient to build a full syntactic tree for an input sentence(seePinker 1987) . The prosodic bootstrapping hypothesis, taken literally , is quite extraordinary . It is tantamount to the suggestion that there is a computational procedure that can parse sentencesfrom any of the world ' s 5,000 languageswhen the sentencesare spoken from behind a closeddoor (i.e., the sentencesare filtered so that only prosodic information remains) . Among the surprising corollaries to this claim is that it should be fairly easy for a person or machine to give a full parse to an English sentenceheard from behind a closed door , because the listener can use both the universal and the English-specific mappings between prosody and syntax, whereas the child supposedly is capable of doing it using only the universal mappings. If , on the contrary , we, knowing English, cannot parse a sentencefrom behind a closed door , it suggeststhat the young child , not knowing English, is unlikely to be able to do so either. Thus the claim that infants can bootstrap syntax from prosody must be viewed with considerableskepticism.2 ' Overviewof Gleitmans argumentsfor the syntactic cueingof verb semantics. With theseindependentissuesout of the way, we can now turn to Gleitman' s arguments for the importance of the syntactic cueing of verb meaning. These arguments fall into three categories. There are negative arguments: verb meanings cannot be learned from observation of situational contexts alone; therefore some other source of information is required. There is a positive ' ' 2 Moreover, many of the syntactic frames that Gleitman assumesthe child is discriminating in ' order to infer verbs meanings are prosodically identical, such as frames differing only in the specificprepositions or complementizersthey contain, like in versuson or that versus if (see, e.g., Gleitman 1990: table 2).

S. Pinker I Verb syntax and verb semantics

hypothetical argument: verb meanings could be learned from verb syntax; therefore verb syntax probably is that other source. And there are empirical arguments: Children in fact learn verb meaning from verb syntax. I will examine thesearguments separately. 3. The negative arguments : Verb meaDings can' t be learned from observation

Gleitman presentssix argumentswhy attending to the situations in which a verb is used (what she calls ' observation' ) is in principle inadequate to learn the verb' s meaning. I believe that none of the argumentsestablishes her main point , that there is an in -principle gap in observational learning that only syntactic subcategorization information can fill . There are two problems in the argument. ' 3.1. Arguments directed against 'observationlearning only refute learning by associative pairing

The first problem is that Gleitman' s arguments are not aimed at ' observation ' in general. They are aimed at a particular straw theory of observation. This foil , a version of one-trial associative pairing , has the child identify a verb~s meaning with the sensory features activated by the situation at the moment when a verb is uttered. But no one believesthis particular theory, so refuting it is ineffective in establishing in -principle limitations on observation; a few uncontroversial assumptionsmake Gleitman' s objections moot. Let me examine the arguments in order. 3.1.1. Multiply -interpretableevents Any single event is multiply -ambiguous as to which verb it exemplifies. Gleitman ( 1990) notes, for example, that most situations of pushing also entail moving. If a situation is describedas (say) The boy is pushing the truck, ' ' ' the child cannot know whether push means ' push or move . This point , however, only shows that children cannot learn the meaning of a verb from a single situation. But no one, not even the British associationists and their descendants , has ever suggestedthey do. Simply allow the child to observe how a verb is used across multiple situations (see, e.g., Pinker 1989: ch. 6), and the problem disappears. Sooner or later, push will be used for instancesof pushing without moving (e.g., pushing against a wall , or pushing

S. Pinker / Verb syntax and verb semantics

389

someone who holds his ground), and move will be used for instances of moving without pushing (e.g., sliding or walking) . To take another one of Gleitman' s examples ( 1990: 14), even though a single event may be describable as pushing, as rolling, and as speeding, most events are not. The child need merely wait for an instance of rolling without pushing or speeding, speedingwithout pushing or rolling , and pushing without rolling or speeding. SeeGropen et al. ( 1991a) for experimental demonstrations that children use this kind of information .

3.1.2. Pairedverbsthat describesingleevents Gleitman ( 1990: 16; seealso Fisher et al. 1991: 380) suggeststhat there are pairs of verbs that overlap 100% in the situations they refer to. For example, there can be no giving without receiving, no winning without beating, no buying without selling, and no chasingwithout fleeing. In fact, I doubt that pairs of verbs that refer to exactly the same set of situations exist (or if they do , they must be extremely rare.) Such pairs would be exact synonyms, and there is good reason to believe that there are few if any exact synonyms (Clark 1987, Bolinger 1977, Miller and Fellbaum 1991). To take just theseexamples, I can receivea packageeven if no one gave it to me; perhaps I wasn' t home. John, running unopposed, can win the election, ' though he didn t beat anyone, and the second-place Celtics beat the last-place Nets in the standings last year, though neither won anything. Several of my gullible college friends sold encyclopediasdoor to door for an entire summer, but in many cases, no one bought any; I just bought a Coke from the machine across the hall, but no one sold it to me. If John fled the city , no one had to be chasinghim ; Bill can chaseFred even if Fred isn' t fleeing but hiding in the garbagecan. I would certainly not claim that the learning of all thesedistinctions awaits the child ' s experienceof the crucially disambiguating situation. But a lot of it could, and more important , the in-principle arguments for an alternative that are based on putative total overlap among verb meanings are not valid if meaningsrarely overlap totally .

3.1.3. Thesubsetproblem In some cases, Gleitman suggests , verb learning is impossible even if verbs do not totally overlap in the situations to which they refer. If the situations referred to by Verb A are a supersetof the situations referred to by Verb B, a

390

S. Pinker I Verb syntax and verb semantics

child who mistakenly thought that Verb B had the same meaning as Verb A could never reject that hypothesis by observing how Verb B is used; all instances would fit the A meaning, too. The only disconfirming experience would be overt correction by parents, and there is good reason to believe that children cannot rely on such corrections. This argument is parallel to one commonly made in the acquisition of syntax (see, e.g., Pinker 1984, 1989; Wexier and Culicover 1980; Berwick 1985, Marcus 1993) . For example, move, walk, and saunter are in a superset relation ; any child that thought that saunter meant walk would do so forever, becauseall examplesof sauntering are also examplesof walking . But this is only a problem if the child is allowed to maintain synonyms in his or her vocabulary. If children do not like to keep synonyms around (see Carey 1982, Clark 1987, Markman 1989, for evidencethat they do not ), then if they have a verb A (e.g., walk), and also a verb B (saunter) that seemsto mean the same thing, they know something is wrong . They can look for additional meaning elementsfrom a circumscribed set to make the meaning of B more specific (like the manner of motion ) . Pinker ( 1989: ch. 6) outlines a mechanismfor how this procedure could work .

3.1.4. Thepoorfit of wordto world Gleitrnan suggeststhat even when a verb corresponds in principle to a unique set of situations, it is not , in practice, reliably ' used in that set of situations, so the child has no way of figuring out a verb s meaning based on the situations it actually is used in. For example, Landau and Gleitman showed that the blind child they ' ' studied learned haptic equivalents of the verbs look (roughly, palpate or ' ' ' ' explore haptically ) and see(roughly , sensehaptically ) . But , they found , her ' mother didn t use look and seemore often when object was near than when object was far. ' The point of this argument is unclear. Of course, the mother didn t ' ' necessarilyuse look when an object was near. Look doesnt mean an object is ' near~; it means ' look . The lack of correlation between some easily sensed property like nearness and use of a verb is only relevant if the child is confined to considering"lists of sensory properties as possible verb meanings. ' If children can entertain the concept of looking , in something like the adults ' sense(and Gleitman 1990: 4, assumesthey can), it doesnt matter how many sensoryproperties a verb fails to correlate with if those properties define only ' a crude approximation of the verb s actual meaning. ( This is a problem, for

S. Pinker I Verb syntax and verb semantics

example, with the conclusionsdrawn by Lederer et ale 1989.) All that matters is whether a child can recognize situations in which that correct concept applies. Gleitman ( 1990) then turns to a stronger argument. Even when one examinesgenuineinstancesof the concept corresponding to a verb' s meaning, one finds a poor correlation with instances of the parent uttering the verb. For example, in one study put was found to be used 10% of the time when there was no putting going on. Similarly , open was used when there was no opening 37% of the time. As Gleitman notes, this is not a surprise when one realistically considers how parents interact with their children. When a mother, arriving home from work , opens the door , she is likely to say, What ' did you do today?, not I m openingthe door. Similarly , she is likely to say Eat your peas when her child is, say, looking at the dog, and certainly not when the child is already eating peas. Indeed, Gleitman ( 1990: 15) claims that ' positive imperatives pose one of the most devastating challenges to any schemethat works by constructing word -to-world pairings' . The problem with this argument is that it , too , only refutes the nonviable theory of learning by associatepairing , in which verb meanings are acquired via temporal contiguity of sensory features and utterances of the verb. It doesn't refute any reasonableaccount, in which the child keeps an updated mental model of the current situation (created by multisensory object- and event-pe~ception faculties), including the likely communicative intentions of other humans. The child could use this knowledge, plus the lexical content of the sentence, to infer what the parent probably meant. That is, chldren need not assumethat the meaning of a verb consists of those sensoryfeatures that are activated simultaneously with a parental utterance of the verb ; they can assume that the meaning of a verb consists of what the parent probably meant when he or she uttered the word. Thus imperatives, where the child is ' not performing the act that the parent is naming, are not ' devastating . Certainly when a parent directs an imperative at a child and takes steps to enforce it , the child cannot be in much doubt that the content of the ' ' imperative pertains to the parents wishes, not the child s current activities. 3.1.5. Semantic properties closed to observation

' Gleitman considers this the ' most serious challenge to the idea that children learn verb meanings by attending to their nonsyntactic contexts. Mental verbs like think, know, guess, wonder, know, hope, suppose , and understandinvolve private events and states that have no external perceptual

S. Pinker / Verb syntax and verb semantics

correlates . Therefore children could not possibly infer their meanings obser vationally . One problem I see with this argument is that although children may not be able to observe other people thinking and the contents of others ' beliefs , they can observe themselves thinking and the contents of their own beliefs . Similarly , children may not know what their mothers are feeling , but they certainly know what they are feeling . And crucially , in many circumstances so do their mothers . When a parent comments on what a child is thinking or feeling , that constitutes information about the meanings of the mental state verbs they use. ' Moreover , there surely are ways to infer a person s mental state from his or her behavior . Indeed , the standard way that humans explain each other ' s behavior is to assume that it is caused by beliefs and intentions , which can only be inferred . This must be how adults , during ordinary speech production , know when to use mental verbs based on their own mental state or ' guessesabout others , even though there is no obvious referent event . There is no principled reason that children could not infer meanings of new mental verbs using exactly the same information that adults employ to use existing mental verbs accurately .

3.1.6.

Does a richer system of mental representation hurt or help the child ?

Gleitman suggeststhat if children are not temporal contiguity associatorsif they can entertain hypothesesabout causes, mental states, goals, speakers' intentions, and so on - their learning task is even harder. For the very richnessof such representational abilities yields a combinatorial explosion of logically possible hypothesesfor the child to test. This argument, however, seems to conftate two ideas: ' a rich set of ' ' ' hypotheses, and a set of rich hypotheses. Gleitman correctly points out that a rich (i .e., numerous) set of hypothesesis a bad thing if you' re a learner. But replacing her associative-pairing mechanismwith a cognitively more sophisticated one results in a set of rich (i.e., structured) hypotheses, not a rich set of hypotheses. And a set of rich hypothesesmay in fact be fewer in number than a set of impoverished ones (e.g., combinations of sensory features) in any given situation : creatures with complicated human brains see the world in only a few of the logically possible ways. Presumably there are many more hypothesesfor a learner who considersall subsetsof patchesof color and bits of fur and whisker than there are for a learner with a sophisticated objectrecognition system who obligato rily perceives these patches as a single

S. Pinker I Verb syntax and verb semantics

393

' rabbit ' . The whole

point of a rich computational apparatus is to reduce the interpretations of a sceneto the small number of correct ones. This is exactly what is neededto help solve the learning problem. 3.2. Problemsin understandingobservationallearning do not constituteevidence for syntactic cueing In much of her discussion, Gleitman attempts to place the burden of proof on anyone who believes that verb learning depends on observation, by identifying many areas of ignorance and difficult puzzles regarding how it could work . Indeed, anyone who thinks that a child can infer what a parent means from the situation and the nonverb content of the sentence must propose that a heterogeneouscollection of not -very-well specified routes to knowing - indeed, the entirety of cognition - is available for use in the ' learning of verb meanings. Moreover, any such proposal must deal with the fact that even the most perceptive child and predictable parent cannot be expectedto be in perfect synchrony all the time. Gleitman' s discussion contains penetrating and valuable analyses that clearly define central researchproblems in how children learn the meaningsof words. But to support the alternative claim that verb subcategorization information is crucial, it is necessaryto show that no theory of inferring communicative intent could ever be adequate, not that we currently don' t have one that is fully worked out . Moreover, Gleitman' s attempt to shift the burden of proof ultimately fails, becauseshe herself, at the end of the 1990 article and in Fisher et al. ( 1991 and this volume), concedes(in responseto some of the points I elaborate on in the next section) that someform of observational learning in indispensable. She notes that information about manner of motion , type of mental state, nature of physical change undergone, and so on, are simply not available in the syntax of subcategorization: ' the syntax is not going to give the learner information delicate and specific enough, for example, to distinguish among such semantically close items as break, tear, shatter, and crumble ... Luckily , these distinctions are almost surely of the kinds that can be culled from transactions with the world of objects and events' (Gleitman 1990: 35) . ' This concession, however, completely redirects the force of Gleitman s criticisms of observational learning. For the meaning components that Gleit man agrees are learned by observation are the very components that she, earlier in the article, claimed that observation cannot acquire! For example, the fact that open is often used when opening is not taking place (e.g.,

S. Pinker / Verb syntax and verb semantics

imperatives), and that open is not used when opening is taking place (e.g., when someoneenters the house), if it is relevant at all , pertains in full force to the ' delicate and specific' aspectsof the meaning of open (i.e., those aspects that differentiate it from syntactically identical close). Similarly, parents surely cannot be counted on to use break or tear when and only when breaking or tearing are taking place, respectively. Nonetheless, Gleitman concedesthat the meanings specific to open, break, and tear are somehow learned by observation. Thus it is not true, as she suggests( 1990: 48), that ' semantically relevant information in the syntactic structures can rescue ' observational learning from the sundry experiential pitfalls that threaten it . There are pitfalls , to be sure, but for most of the ones Gleitman originally discussed , syntax offers no rescue. What we need is abetter , non-associationist theory of observational learning. ' 3.3. Conclusionsabout Gleitmans argumentsagainst observationallearning Gleitman convincingly refutes a classical associationist theory of semantic learning, in which word meanings are acquired via temporal contiguity of sensoryfeaturesof the sceneand utterancesof the word. She also convincingly shows that to explain verb learning, we need a constrained representational ' systemfor verbs meanings, principles constraining how one verb is related to another in the lexicon, a learning mechanism that can construct and modify semantic representations over a set of uses of the verb, and a greater understanding of how children interpret events, actions, mental states, and other speakers' communicative intentions. But the arguments do not show that the full set of semanticcues to semanticsis so impoverished in principle that the child must use sets of syntactic subcategorization frames as cues instead, nor that syntactic cues provide just the information that semantic cues fail to provide. Rather, Gleitman herself assumesthat there exists some form of observationallearning powerful enough to acquire aspectsof meaning that her own arguments show to be hard to acquire.3 3 PaulBloomhaspointedout to methat argumentssimilarto Gleitman's wereoriginallymade ' . For example , Chomskyshowed ) in his reviewof Skinners VerbalBehavior by Chomsky( 1959 of their couldnot in generalbe learnedby hearingthe nounsin the presence that nounmeanings like Eisenhower . But Chomskyusedexamples referents , a propername, whosemeaningcouldnot possiblybe distinguishedusingsyntacticcuesfrom the thousandsof other proper namesthat that observationand syntacticcuesare not the only mustbe learned(e.g., Nixon). This suggests possiblemeansof learning. SeeBloom (this volume) for discussionof similar issuesin the . learningof noun meanings

S. Pinker/ Verbsyntaxandverbsemantics

395

4. ne positive hY)M)tbeticaI argmnent: Semantic infonnatio D in subcategorization frames

Gleitman and her collaboratorsgive a few specificexamplesof how a learner might use a verb's syntactic propertiesto predict aspectsof its . Unfortunately, they do not relate theseexamplesto a general meaning ' theory of the relation of syntaxto semanticsin verbs lexicalentriesand of how a learnercould exploit them. In this sectionI will attemptto fill this gap ' by layingout the logic of verbs syntaxand semanticsand the implicationsof that logic for learning. 4.1. Verb roots versus verb frames

The first question we must ask is, what do we mean by ' a verb' ? The term is ambiguous in a critical way, becausein most languagesa verb can appear in a family of forms, each with a distinct meaning component, plus a common meaning component that runs throughout the family . For example, many verbs can appear in transitive, intransitive , passive, double-object, prepositional object, and other phrases. In some cases the verb actually changesits morphological form acrossthesecontexts, though in English only the passiveis marked in this way. Following standard usagein morphology , we can say that all the forms of a given verb share the same verb root. We can then call the syntactically distinct forms of a given root its frames. For example, consider the matrix of verb forms in figure 1, where the existenceof a given root in a given frame is marked with an ' x' . The meanings of the x ' s differ along two dimensions. Let me use the term root meaningto refer to the aspectsof meaning that are preservedin a given root across all the frames it appears in ; that is, whatever aspectsof meaning The water boiled and I boiled the water have in common, and fail to share with The door openedand I opened the door. Let me use the term frame meaningto refer to the orthogonal dimension: the aspectsof meaning that are shared across all the roots that appear in a given frame; that is, whatever aspectsof meaning differentiate The water boiled from I boiled the water, and that The water boiled and The door openedhave in common. Note that root meanings are much closer to what people talk about when worrying about acquisition of word meaning. That is, the main problem in learning boil is learning that it is about hot liquid releasing bubbles of gas. This is the aspect of boil that is found in both its transitive and intransitive uses, that is, its root meaning. The root meaning corresponds to what we

396

S. Pinker I Verb syntax and ve.pbsemantics Frames

NP-

NP_NP

eat move boil

x

x

x x

x x

open kill die think tell know see look

x

x

NPJ

NP--PP

NP- NP-PP

NP- NP-S

Roots

x x x x x

x

x

x

x

x x

x

Fig. I think of as the content of a verb. The frame meaning - the fact that there must be an agent causing the physical change when the verb is used in the transitive frame, and that the main event being referred to is the causation, not the physical change- is just as important in understanding the sentence, but it is not inherently linked to the verb root boil. It is linked to the transitive syntactic construction, and would apply equally well to melt, freeze, open, and the thousands of other verb roots that could appear in that frame. This is a crucial distinction . 4.2. Learning about a verb in a single / rame

The first question that follows is, What can be learned from hearing a verb in one frame? Something, clearly, for frame semanticsand frame syntax are highly related. For example, it is a good bet that in A glips B to C, glip is a verb of transfer. The regularities that licensethis inference are what linguists call linking rules (Carter 1988, Jackendoff 1987, 1990; Pinker 1989, Gropen et al. 199Ia) . For example, if A is a causal agent, A is the subject of a transitive verb. Linking rules are an important inferential mechanism in semantic bootstrapping (semantic cueing of syntax at the outset of language acquisition ), in predicting how one can use a verb once one knows what it means, and in governing how verbs alternate between frames (see Gropen et al. 1991afor discussion) .

S. PinkerI Verbsyntaxandverbsemantics

397

One might now think : If syntax correlates with semantics, why not go both ' ways? If one can infer a verb s syntax from its semantics(e.g., in semantic ' bootstrapping), couldn t one just as easily infer its semanticsfrom its syntax? As Gleitman puts it ( 1990: 30) : ' The

turns semanticbootstrap syntacticbootstrap ping proposalin essence ping on its head. the mappingrulesfor semantics on to , thechild who understands Accordingto this hypothesis . The syntaxcan usethe observedsyntacticstructuresas evidencefor deducingthe meanings learnerobservesthe real-world situation but also observesthe structuresin which various wordsappearin the speechof the caretakers . Suchan approachcan succeed because , if the syntacticstructuresare truly correlatedwith the meanings , the rangeof structureswill be informativefor deducingwhichword goeswith whichconcept .'

I believethis argument is problematic. The problem is that a correlation is not the same thing as an implication. ' Correlation ' means ' many X ' s are Y ' s or ' ' ' ' ' ' many Y s are X s or both . Implication means if X , then Y , though not ' necessarilyvice versa . The asymmetry inherent in an implication is crucial to understanding how it can be used predictively. For example, if I feed two numbers (e.g., 3 and 5) into the sum-of function , the value must be 8. But if I guesswhich inputs led to a value of 8, I cannot know that they were 3 and S. Linking rules are implications . They cannot straightforwardly be used in the reversedirection. If a verb means ' X causesY to shatter' , then X is the subject of the verb. But if X is the subject of a verb, the verb does not ' ' necessarilymean X causesY to shatter . This asymmetry is inherent to the design of language. A grammar is a mechanism that maps a huge set of semantic distinctions onto a small set of syntactic distinctions (for example, thousands of kinds of physical objects are all assignedto the same syntactic ' ' category noun ). And becausethis function is many-to- one, it is not invertible. Now , if one castsaway most of the meaning of a verb (e.g., the part about shattering), there may remain some abstract feature of meaning that could map in one-to -one fashion to syntactic form . To the extent that that can be done, one could learn some things about a verb form ' s meaning from the frame that the verb appearsin. First , one can learn how many arguments the verb relates in that form , as in the difference between The water boiled (one argument) and She boiled the water (two arguments), or the difference betweendie (one argument) and kill (two arguments) . Second, one can infer ' ' something about the logical type of some of the arguments, like proposition ' ' (if the verb appearswith a clause) versus thing (if the verb appears with an NP) versus ' place/path' (if the verb appears with a PP) . That is, the syntax can help one distinguish betweenthe meaning offind in find the book andfind

398

S. PinkerI Verbsyntaxandverbsemantics

that the book is interesting; between shoot the man and shoot at the man; perhaps even between think, eat, and go. Third , the syntax of a sentencecan help identify which argument can be construed as the agent (viz., the subject) in caseswhere the inherent properties of the arguments (such as animacy) leave it ambiguous, for example, in kill versus is killed by, and chaseversus flee . Similarly , syntactic information can distinguish the experiencerfrom the ' stimulus in ' psych-verbs with ambiguous roles, such as Bill feared Mary and Mary frightened Bill . Fourth , syntactic information can help identify which ' ' argument is construed as affected (viz., the syntactic object) in events where several entities are being affected in different ways. For example, in load the hay and load the wagon, on cognitive grounds either the hay or the wagon ' ' could be interpreted as affected : the hay, becauseit changeslocation , or the wagon, becauseit changesstate from not full to full (similar considerations apply to the .pair of verbsfill and pour. The listener has to notice which of the two arguments (content or container) appearsas the direct object of the verb ' ' to know which one to construe as the affected argument for the purpose of understanding the verb in that frame. Gleitman and her colleaguesgive many ' ' examplesof these forms of learning, which I have called reverselinking (see Pinker 1989 and Gropen et al. 1991a, b for relevant discussion and experimental data) . ' Unfortunately , while one can learn somethingabout a verb form s meaning from the syntax of the frame it appears with , especially when there are a small number of alternatives to selectamong, one cannot learn much, relative to the full set of English verbs, becauseof the many-to -one mapping between the meanings of specific verbs and the frames they appear in. For example, one cannot learn the differences among slide, roll , bounce, skip, slip, skid, tumble, spin, wiggle, shake, and so on, or the differencesamong hope, think, pray , decide, say, and claim; among build, make, knit , bake, sew, and crochet; among shout, whisper, mumble, murmur, yell , whimper, whine, and bluster; among fill , cover, tile, block, stop up, chain, interleave, adorn, decorate and face, and so on. Indeed, Gleitman herself ( 1990: 35) concedesthis point in the quote reproduced above. In sum, learning from one frame could help a learner distinguish frame meanings, that is, what the water boiled has in common with the ball bounced and does not have in common with I boiled the water. But it does not distinguish root meanings, that is, the difference the water boiled and' the ball' bounced. And the root meaningsare the ones that correspond to the content ' ' ' of a verb, what we think of as the verb s meaning , especially when a given verb root appears in multiple frames.

S. Pinker I Verb syntax and verb semantics

399

The frame meanings (partly derivable from the frame) are closer to the ' perspective that one adopts relative to an event: whether to focus on one actor or another, one affected entity or another, the cause or the effect. Indeed in some restricted cases, differencesin perspectiveare most of what distinguishes pairs of verb roots, such as kill and die, pour and fill , or Gleitman' s example of chaseand flee . Gleitman ( 1990) and Fisher et al. (this volume) adopt a metaphor in which the syntax of a verb frame servesas a ' zoom lens' for the aspectsof the event referred to by the verb. This metaphor is useful, becauseit highlights both what verb syntax can do and cannot do. The operation of lens when aimed at a given scenegives the photographer three degreesof freedom, pan, tilt , and zoom, which have clear effects on the perspectivein the resulting picture. But no amount of lens fiddling can fix the vastly greater number of degreesof freedom defined by the potential contents of the picture - whether the lens is aimed at a still life , a nude, a ' 57 Chevy, or one' s family standing in front of the Grand Canyon. So I have no disagreement with Gleitman' s arguments that a syntactic frame can serve as a zoom lens, helping a learner decide which of several perspectiveson a given type of event (discernedby other means) a verb forces on a speaker. But becausethis mechanismcontributes no information about a verb' s content, it cannot offer significant help in explaining how children learn a verb' s content despite blindness, nor in explaining how children learn a verb' s content despite the complexity of the relationship between referent event and parental usage. '

4.3. Learning about a verbfrom its multiple frames Gleitman recognizes the limitations of learning about a verb's meaning

froma single frame :

' To be sure the number of such clause structures is , quite small compared to the number of possible verb meanings: It is reasonable to assume that only a limited number of highly general semantic categories and functions are exhibited in the organization that yields the subcategorization frame distinctions. But each verb is associated with several of these structures. Each such structure narrows down the choice of interpretations for the verb. Thus these limited parameters of structural variation , operating jointly , can predict possible ' meaning of an individual verb quite closely. (Gleitman 1990: 30- 32)

The claim that inspection of multiple frames can predict a verb' s meaning ' quite closely appearsto contradict the earlier quote in which Gleitman notes that syntactic information in general is not ' delicate and specific enough to '

400

S. Pinker I Verb syntax and verb semantics

" ' distinguish among ... semantically close items . To seeexactly how close the ' syntax can get the learner to a correct meaning, we must ask, What can be ' learned from hearing a verb in multiple frames? In particular , can a root ' meaning - the verb s content - be inferred from its set of frames, and if so, how? Unfortunately , though Gleitman and her collaborators give examples of how children might converge on a meaning from several frames, almost always using the problematic example of see(seefn. 1), they never outline the inferential procedure by which children do so in the general case. In Fisher et al. (this volume) they suggest that the procedure is simply the zoom lens ' ' (single-frame) procedure applied iteratively . They give the procedure as ' follows : In assigning a gloss to the verb, satisfy all semantic properties ' implied by the truth conditions of all its observed syntactic frames . But this cannot be right , for reasons they mention in the next paragraph. The truth conditions (what I have been calling ' frame meaning' ) that belong to a verb form in one frame do not belong to it in its other frames. So satisfying all of them will not give the root meaning or verb' s content. If we interpret ' ' satisfying all semantic properties as referring to the conjunctionof the frame meanings, we get the meaning of its most restrictive frame, which will be incompatible with its lessrestrictive frames. For example, the truth conditions for transitive boil include the presenceof a causal agent. But presenceof a causal agent cannot be among the semantic properties of boil across the board, for its intransitive version ( The water boiled) is perfectly compatible with spontaneous boiling in the absenceof any agent. But if we interpret ' ' satisfying all semanticproperties to be the disjunctionof frame meanings, the aggregation leads to virtually no inference at all. Consider again the frame involved in The water boiled. This intransitive frame tells the learner that the meaning of boil in the frame consists of a one-place predicate. Now consider a second frame, the one involved in I boiled the water. This transitive frame tells you (at most) that the meaning of boil in the frame consistsof causation ' of some one-place predicate. What do they have in common? One-place ' . Which is not very useful. It says nothing whatsoever about the predicate root meaning of boil, that is, that it pertains to liquid , bubbles, heat, and so on. This is a problem even for verbs that appear in many frames, for which the syntax would seem to provide a great deal of converging information (see Levin 1985, Pinker 1989) . For example sew implies an activity . Sew the shirt implies some activity performed on an object. Sew me a shirt implies an activity creating an object to be transferred to a beneficiary. Sew a shirt out

S. Pinker I Verb syntax and verb semantics

of the rags implies an activity transforming material into some object. What ' do theseframe meaningshave in common? Only ' activity ' . Not ' sewing . ' ' The conclusion is clear: you can t derive a verb s root meaning or content by iterating the zoom lens procedure over multiple frames and taking the resulting union or intersection of perspectives. 4.3.1. Can anything be learnedfrom multiple frames ? I do not wish to deny that there is some semantic information implicit in the set of frames a verb appearswith , nor that an astute learner could not , in principle, use this information . The example Gleitman uses most often, see, has clear intuitive appeal. But which general procedure is driving the inference about seeand other such cases? I can think of two. According to Gleitman, a set of argument frames implicitly poses the ' question, What notion is compatible with involving a physical object, ' involving a proposition, and involving a direction? The child deduces the ' .4 In other words this is a kind of ' , response seeing cognitive riddle-solving ' (Pinker 1989) ; it involves all of a leamer s knowledge, beliefs, and cognitive inferential power. I am not arguing either that children can or cannot solve such riddles. I am simply pointing out what would be going on if they could do so. In particular , note what they would not be doing . They would not be relying on any grammatical principle , and hence would not be enjoying the putative advantages of universal constrained linguistic principles to drive reliable inferences. That is, if guessing a verb' s meaning from its set of frames succeedsat all , it does so by virtue of the child ' s overall cognitive cleverness , and hence could suffer from the same unreliability of overall clevernessas ' inferring a speakers likely meaning from the knowledge of the situation. It is not a straightforward mechanicalprocedure that succeedsbecausethe frames ' ' are abstract surfacereflexes of the meanings (Landau and Gleitman 1985: 138) ' ' 4 Actually , the question and answer should be stated in tenDs of a family of notions , not ' notion ' becauseverbs like seethat can take either , objects or clausal complementsdo not exhibit ' ' ' ' a single content meaning across these frames: seeNP does not mean the same thing as see S . The latter is not even a perception verb: [ seethat the meal is ready does not entail vision. (Clearly ' not , becauseyou can t visually perceive a proposition .) Similarly , [ feel that the fabric is too ' smooth does not entail palpation ; it s not even compatible with it . And Listen! [ hear that the orchestrais playing is quite odd. ( Theseobservations are due to Jane Grimshaw.) Clearly there is ' ' a commonality running through each of thesesets, but it is a metaphorical one; knowing can be ' ' construed metaphorically as a kind of perceiving .

402

S. Pinker I Verb syntax and verb semantics

or because' rnuch of the [sernantic] information can be read off frorn the subcategorization frarnes thernselvesby a general scheme for interpreting thesesemantically' (Landau and Gleitman 1985: 142) . Moreover, the premises that would drive this riddle-solving are far rnore irnpoverished than the ' prernisesderived frorn inferring a speakers rneaning frorn the context. The latter can include any concept the child is capable of entertaining (sewing, boiling , and so on); the former are restricted to a srnaller set of abstract concepts like causability and taking a propositional argument. There is a second way that sets of syntactic frarnes could assist sernantic learning. That is via narrow argument-structure alternations. Often the verbs that can appear with a particular set of syntactic frarnes have surprisingly specific rneanings (see Levin 1985, in press; Pinker 1989, for reviews) . For exarnple, in English, the verbs that can appear in the double-object form but not the to-object form are verbs of irnplied deprivation like envy, bet, and . begrudge(e.g., I envied him his good looks/ envied his good looks of him) . Sirnilarly, verbs of rnanner of rnotion can alternate betweencausative-transitive and inchoative-intransitive forms (e.g., I slid the puck/ The puck slid) , but verbs of direction of rnotion cannot (e.g., I brought the book/ . The book brought) . An astute learner, in principle, could infer , frorn hearing I glipped him those things and frorn failing to hear I glipped those things of him, that glip involves sorne intention or wish to deprive sorneoneof sornething. But note ' that these regularities are highly specific to languagesand to dialects within a language. (For exarnple, . 1 suggestedher somethingis grarnrnatical in Dutch , and . I pushedhim the box is grammatical in sornedialects of English; seePinker 1989.) Exploiting thern requires first having acquired these subtle esand their syntacticbehavior in the dialect, presumablyby abstracting subclass subclass es frorn the sernanticsand syntax of individual verbs, acquired by the other rneans. This kind of inference dependson a good deal of prior learning of verbs' rneaningsin a particular language, and thus is rnost definitely not a ' caseof ' bootstrapping performed by a child to acquire the rneaningsof the verbs to begin with . ' In general, learning a verb s content or root rneaning frorn its set of ' ' syntactic frames ( syntactic bootstrapping ) is fundamentally different frorn ' ' learning its perspectiveor frame rneaning frorn a single frarne ( zoorn lens ) . ' Thus I disagreewith Gleitman s ( 1990) suggestionthat they are versions of a single procedure, or Fisher et alis suggestion(this volume) that one is sirnply the iteration of the other. There is a clear' reason why they are different. While there rnay be a universal rnapping betweenthe rneaning of a frarne and the syntax of that frame (allowing the lens to zoorn), there.is no universal

S. Pinker / Verb syntax and verb semantics

mapping betweenthe meaning of a root and the set of frames it occurs in (see Talmy 1985 and Pinker 1989 for reviews) . For examplet universal linking rules impl Yt roughlYt that an inchoative verb can appear in an intransitive framet and a causativeverb can appear in a transitive frame. And itts clearly possible for some roots to be able to have both causative and inchoative meanings (and hence to appear in both frames) . But itts an accident of English that slide appears in both framest but come and bring appear in one each. Thus the kinds of learning that are licensed by universalt reliablet grammatical linking regularities are restricted to differencesin perspective. A verbts content is not cued by anyone of its syntactic framest and at best might be related to its entire set of frames in a tenuoust language-specific way. S. Experimental evidence on children ' s learning of verb meanings from verb syntax

Gleitman ( 1990) reviews three ingenious experimental studies intended to ' support the claim that children can and do use a verb s syntax to learn its meanings. Though the experiments are important contributions to our understanding of the acquisition of verb meaning, they do not speak to Gleitman' s proposals about ' syntactic bootstrapping' , becausenone of the experiments involved the presentation of multiple frames to a child . Let me review them.

5.1. Hirsh-Paseket a/. ( 1988) One set of studies (Hirsh -Pasek et ale 1988) presented27-month -olds with two video screens. One showed a pair of puppets performing a simple action like flexing their own arms; the other showed one puppet causing the second to perform the action, by manipulating its arms. A recorded voice used a verb unknown to the child in one of two syntactic frames, either the intransitive ' ' ' Big Bird is flexing with Cookie Monster or the transitive Big Bird is flexing ' Cookie Monster . Children looked at the screenshowing the noncausal scene more often when the sentence was intransitive, and looked at the screen showing the causal scenemore often when the sentencewas transitive. Note that this study provided children with only a single frame, not multiple frames. And note as well that what the children were sensitiveto was a frame-specific perspective (causation by an agent), not a root -specific

404

S. Pinker I Verb syntax and verb semantics

content (flexing) . The content (flexing) was acquired through observation, not syntax; it was depicted on the video screen, and the child was watching it . Thus at best the children were demonstrating use of the zoom lens procedure; ' there was no opportunity for multiple frames to cue the verb s content. (At worst, the children were not acquiring any information about the verb at all , but were ignoring the verb and merely responding to the transitive and intransitive sentenceframes themselveswhen directing their attention.)

5.2. Naig/es( 1990 ) ' A secondstudy appearsto show children learning a verb s content from a single frame. In Naigles ( 1990), 24-month-olds first saw a video of a rabbit pushing a duck up and down, while both made large circleswith one arm. One ' ' group of children heard a voice saying The rabbit is gorping the duck ; another ' heard 'The rabbit and the duck are gorping . Then both groups saw a pair of screens , one showing the rabbit pushing the duck up and down, neither making arm circles, the other showing the two charactersmaking arm circles, neither ' ' pushing down the other. In responseto the command Wheres gorping now? ' Find gorping! , the children who heard the transitive sentencelooked at the screen showing the up- and-down action, and the children who heard the intransitive sentencelooked at the screenshowing the making-circlesaction.s What have the children learned, and from what source? Clearly, they ' learned most of the verb root s content - that gorp means pushing and/ or making circles, but not sliding or boiling or killing or dancing not by ' the video attending to the syntax, but by observing the scene; that s what depicted. Without the video, the children would have learned little if anything . What the children learned from the sentencesyntax was, once again, 5 Unfortunately , the Naigics experiment had a confound, commentedon by Gleitman, 1990, in footnote 13 on p . 43. The difference between the gorps could have been cued by the conjoined versussingular subjects, disregarding the verb syntax entirely . That is, the difference between The rabbit is gorping and The rabbit and the duck are gorping, with identical verb syntax, could have been sufficient for children to pick out a screento look at. In the first case, the children would look at what the rabbit alone was doing ; in the secondcase, they could look at what the rabbit and duck were doing simultaneously. This would be sufficient to direct their attention in testing to the push-down screenin the first case, the make-circles screenin the second. This confound, a good example of the difference between using linguistically - conveyed content as opposed to verbal syntax, could be eliminated by using the sentencesof Hirsh -Pasek et al. The voice-over ' ' ' would say either ' the rabbit is gorping the duck or the rabbit is gorping near the duck , Children in the first condition , if they could useverb syntax as a cue, would still find gorping in the up- anddown screen, children in the secondwould find it in the make-circles screen.

S. Pinker I Verb syntax and verb semantics

405

the meaning of the verb in its particular frame. The sentenceplayed to the first group of children told them that gorp is a two-place predicate, presumably a causative. It means either 'cause to pop up and down' or ' cause to make arm circles' . The sentenceplayed to the secondgroup told them that gorp is a one-place predicate, presumably an activity . It means either ' make arm circles' or ' pop up and down' . Once again, the syntactic frame cued only the coarse information of how many arguments were immediately related by the verb; the rest came from observation. One might object at this point that Naigles' s experiment has demonstrated what I have been arguing is impossible: children appear to have learned about a verb' s content (in the case, up-and-down versus make-circles) from the sentencein which it is used, and could not have learned that content from observation alone. But this is misleading. Successdependedcompletely on the fact that Naigles engineeredan imaginary world in which perspective and content were confounded, so that when children were using syntax to choose the right perspective, they got the right content, too , by happy accident. Note there is no grammatical constraint forcing or preventing either ' popping up and down' or ' making arm circles' from being exclusively transitive, exclusively intransitive, or alternating. Nor is there any real-world constraint that could cause creatures to make arm circles and to pop up and down in tandem. But Naigles' s teaching example exemplified both such constraints: only popping up-and-down was causable, and such causation took place in the presenceof arm circles. It was only these artificial contingencies that made the forms learnable by syntax rather than observation. Consider what would have happenedif the children had been shown a scenedepicting circles without pushing up and down or vice-versa. In that case, observation would have been sufficient to distinguish the two actions, with no syntax required. Now consider what would happen if the children had been shown an armcircling rabbit causing the duck simultaneously to pop up and down and to make arm circles. This is no more or less bizarre than the conjunction that Naigles did show children, where causation of popping up and down was simultaneous with uncaused arm-circling . In that case, neither the sentence The duck is gorping the rabbit nor the sentence The duck and rabbit are gorping would have distinguished the two kinds of motion . This shows that syntax is neither necessarynor sufficient- to distinguish the alternative content meanings of gorp, across all the different scenesin which it can be used; observation, in contrast, is sufficient. In sum, Naigles simply selected a contrived set of exposure conditions that penalized observation while letting syntax lead to the right answer by coincidence.

406

S. Pinker I Verb syntax and verb semantics

5.3. Fisheret a/. ( 1994) The third experiment is reported by Fisher et ale(this volume) . Here 4- yearolds viewed scenesthat were consistent with two verbal descriptions, like eat and feed, chaseand flee, ride and carry, give and take, or put and carry. A puppet described the action using a novel verb root like zorp. For each sentence, different children heard the verb used in one of two syntactic forms. Depending on the action, the forms differed in transitivity , in which animal was named as the subject or object, or in whether an animal or a thing was named as the direct object. The child was asked to describe the event for the puppet by providing it with the English translation of the puppet word . The children tended to provide the English word that corresponded to the syntactic frame that the puppet used. There is a problem, though, with interpreting this experiment as evidence ' for syntactically-cued verb learning : the children didn t learn any verbs. They simply had to select existing English verbs that were the translation of the puppet words (this is literally how the task was described to the children) . ' That is, the children weren t learing what eat or feed mean; they were simply demonstrating that by the time they had walked into the lab, they had previously learned (through whatever means) that eat can have a sole, actor argument whereasfeed can have an agent and a beneficiary/goal argument. The-task was: ' which English verb is consistent with the videos and has an ' ' does a verb mean if it agent subject and a beneficiary/goal object? not , what ' ' has an agent subject and a beneficiary/goal object? It s not surprising that children can be shown to know the answer to the first question, because that ' s exactly the knowledge they depend on every time they use the verb correctly in their speech. So the experiments show what children already know about English verbs, not how they learn English verbs. Gleitman ( 1990) and Fisher et ale (this volume) recognizethis problem but suggestthat it provides indirect evidencein favor of their position . I think it might show little or nothing about verb learning at all , becausethe knowledge being used could be specific to what the children know about English. that Suppose someone were to advance the (obviously wrong) hypothesis ' ' are with p people learn what a word means by' its sound words beginning ' causative, words beginning with f are inchoative. To test it , they show a ' scene that is ambiguous between falling and pushing, and say the puppet ' ' ' calls this f .... or the puppet calls this p .... . Subjectsguessfalling and pushing accordingly. But obviously they do it by tapping particular facts of English acquired beforehand, with no utility in learning any other language. That is,

~., S. Pinker I Verb syntax and verb i Pcmallti

407

there is no reason to think that the retrieval cue that the experimenter provides now, for existing knowledge, was ever used as a learning cue, in order to acquire that knowledge originally . Substitute ' transitive syntax' and ' intransitive ' ' ' ' ' syntax for p .... and f .... and one has the Fisher et al. experiment (this volume) - a test of whether children can use transitivity correctly as a retrieval cue for previously learned words when the content of the words is available observationally. 5.4. What experiment would show syntactic cueing of verb semantics? There is an extremely simple experiment that could test whether children can learn a verb root ' s semanticcontent from multiple frames. There could be no TV screen, or content words, just syntactic frames. For example, children would hear only . She;'pi /ked; She pi /ked me something; She pi /ked the thing from the other things; Shepi /ked the other things into the thing; Shepi /ked one ' thing to another, and so on. If children can acquire a verb s content from multiple frames, they should be able to infer that the verb basically means 'create ' by attaching (Levin 1985). (Of course, one would have to ensure that the child was learning a new meaning and not simply using the frames to retrieve an existing word , for reasonsmentioned in the preceding subsection.) Lest one think that this set of inputs is way too impoverished and boring for ' a child to attend to , let alone for the child to draw semanticconclusionsfrom , in the absenceof perceiving some accompanying real-world event, recall that this is exactly the situation that Landau and Gleitman assumethe blind child is in. It would be an interesting finding if children (or adults) could learn ' significant aspectsof a verb s content from syntactic cues, as this experiment would demonstrate. If Gleitman and her collaborators are correct, they should be able to do so.

6. Conclusions

I have gone over Gleitman' s arguments against the sufficiency of learning verb semantics by observation of semantic cues in the situations in which a verb is used, and her arguments for the utility and use of syntactic subcategorizatio information. I suggestthat a careful appraisalof thesearguments leads to the following conclusions. As Gleitman shows, temporal contiguity betweensensoryfeaturesand verb usagescannot explain the acquisition of verb meaning. What this suggestsis

408

S. PinkerI Verbsyntaxandverbsemantics

that the explanation of verb learning requires a constrained universal apparatus for representing verb meanings, principles governing the organization of the lexicon, a perceptual and conceptual system acute enough to infer which elementsof verb meaningsan adult in a situation is intending to refer to , and a learning procedure that can compare hypothesizedsemanticrepresentations acrosssituations. Gleitman has also convincingly demonstrated that single syntactic frames provide information about aspectsof the meaning of the verb in that frame ' ' (the zoom lens hypothesis) . This information is largely about the perspective that a verb forces a speaker to take with regard to an event. It includes the number of arguments, the type of argument, a focus on the cause or effect, and the choice of agent and affected entity when more than one is cognitively possible. As Gleitman points out , these are exactly the kinds of information that are difficult or impossible to infer from observing the situations in which a verb is used. I disagree, however, that multiple syntactic frames provide crucial information about the semantic content of a verb root across its different frames Gleitman calls ' syntactic bootstrapping' ) . There is no syntacticallywhat ( driven general inferential schemeby which such learning could work ; there is no empirical evidencethat children use it ; and it does not make up for any of the problems Gleitman notes in understanding how children learn about a verb' s meaning from observing the situations in which it is used. Indeed, the suggestion is incompatible with one of the basic design features of human language: a vast set of concepts is mapped onto a much smaller set of grammatical categories.

References . . Cambridge . The acquisitionof syntacticknowledge Berwick, R.C., 1985 , MA : MIT Press : The role of syntax-semanticsmappingsin the acquisitionof Bloom, P., 1994. Possiblenames . Lingua92, 297- 329(this volume). nominals . Meaningand form. London: Longman. BolingerD ., 1977 . Journalof Abnonnaland Social . Linguisticdeterminismand thepart of speech Brown, R., 1957 1 5. 55 , Psychology : The state of the art. In : E. Wanner, L.R. Gleitman . Semanticdevelopment Carey, S., 1982 (eds.), Languageacquisition:The stateof the art, 347- 389. New York: CambridgeUniversity . Press . On linking: Papersby RichardCarter. (LexiconProjectWorking Paper25). Carter. R.J., 1988 . B. Levin, C. Tenny(cds.). Cambridge , MA : MIT Centerfor CognitiveScience ' s 'Verbalbehavior'. 3 26 - 58. . A Review of B.F. Skinner N. 1959 , , , Language Chomsky

S. Pinker I Verb syntax and verb semantics

409

. . Aspectsof the theoryof syntax. Cambridge , MA : MIT Press , N., 1965 Chomsky . In : B. Mac. The principleof contrast: A constrainton languageacquisition Clark, E.V., 1987 . Hillsdale, NJ: Erlbaum. of languageacquisition Whinney(ed.), Mechanisms . Language67, 547- 619. . Thematicproto-rolesand argumentselection Dowty, D., 1991 . On the semanticcontentof subcategorization Fisher, C., H. Gleitmanand L.R. Gleitman, 1991 23, 331- 392. frames . CognitivePsychology . Whenit is betterto receivethan to Fisher, C., G. Hall, S. Rakowitzand L.R. Gleitman, 1994 give: Syntacticand conceptualconstraintson vocabularygrowth. Lingua 92, 333- 375 (this volume). . LanguageAcquisitionI , 3- 55. . The structuralsourcesof verb meaning Gleitman, L.R., 1990 . Complementselectionand the lexicon. Linguisticinquiry 10, 279- 326. Grimshaw , J., 1979 . In: C. L. Baker, . Fonn, function, and the languageacquisitiondevice Grimshaw , J., 1981 J.J. McCarthy(eds.), The logicalproblemof languageacquisition165- 182. Cambridge , MA : M IT Press . . . Argumentstructure.Cambridge Grimshaw , MA : MIT Press , J., 1990 . Affectedness and direct objects: Gropen, J., S. Pinker, M. Hollanderand R. Goldberg, 1991a in the acquisitionof verbargumentstructure.Cognition41, 153The role of lexicalsemantics 195. . Syntax and semanticsin the Gropen, J., S. Pinker, M. Hollander and R. Goldberg, 1991b acquisitionof locativeverbs. Journalof Child Language18, 115- 151. . Syntactic Hirsh-Pasek , K., H. Gleitman, L. R. Gleitman, R. Golinkoff and L. Naigles, 1988 . at the 13th Annual Boston : Evidence from bootstrap ping comprehensionPaperpresented on LanguageDevelopment . UniversityConference . Thestatusof thematicrelationsin linguistictheory. LinguisticInquiry 18, jackendoff, R.S., 1987 369- 411. . Semanticstructures . Cambridge . JackendoffiR.S., 1990 , MA : MIT Press . What's in a name? On the child' s acquisitionof Katz, B., G. Bakerand J. Macnamara , 1974 45, 269- 273. properand commonnouns. Child Development . Languageand experience . Cambridge Landau, B. and L. R. Gleitman, 1985 , MA : Harvard . UniversityPress . Input to a deductiveverb acquisition Lederer , A., H. Gleitman and L.R. Gleitman, 1989 . Paperpresentedto the 14thAnnual BostonUniversityConferenceon Language procedure . Development . Lexical semanticsin review: An introduction. In: B. Levin (ed.), Lexical Levin, B., 1985 in review. LexiconProjectWorkingPapers# I. Cambridge semantics , MA : MIT Centerfor . CognitiveScience Levin, B., in press . Englishverbclassesand alternations : A preliminaryinvestigation . Chicago, IL : Universityof ChicagoPress . . Negativeevidencein languageacquisition . Cognition46, 53- 85. Marcus, G.F., 1993 Markman, E.M., 1989 . Categorization in children: Problemsof induction. Cambridge , MA : MIT Press / BradfordBooks. . Constraintschildrenplaceon word meanings . CognitiveScience14, 57Markman, E.M., 1990 77. Miller, G.A., 1991 . The scienceof words. NewYork: ScientificAmericanLibrary. Miller, G.A. and C. Fellbaum, 1991 . Semanticnetworksof English. Cognition41, 197-229. . How do wordsgettheirmeanings ? TheJournalof Philosophy Moravcsik 78, 5-24. , J.M.E., 1981

410

S. Pinker / Verb syntax and verb semantics

. Journalof Child Language17. . Childrenusesyntaxto learn verb meanings Naigles. L.. 1990 357- 374. . In: J. Bresnan(ed.). . A theoryof theacquisitionof lexicalinterpretivegrammars Pinker. S.. 1982 . of grammaticalrelations.Cambridge The mentalrepresentation . MA : MIT Press . Cambridge . Languagelearnabilityand languagedevelopment Pinker. S.. 1984 . MA : Harvard . UniversityPress . In: B. MacWhinney(ed.). . The bootstrap Pinker. S.. 1987 ping problemin languageacquisition . Hinsdale of languageacquisition Mechanisms . NJ: Erlbaum. . Resolvinga learnabilityparadoxin the acquisitionof the verb lexicon. In : Pinker. S.. 1988 . Baltimore. MD : Brookes . M. Rice. R. Schiefelbusch (eds.). The teachabilityof language . Learnabilityand cognition. The acquisitionof argumentstructure.Cambridge Pinker. S.. 1989 . MA : MIT Press / BradfordBooks. . . Word and object. Cambridge . MA : MIT Press Quine. W.V.O.. 1960 . Lexicalizationpatterns: Semanticstructurein lexicalforms. In: T. Shopen(ed.). Talmy. L.. 1985 . Vol. III : Grammaticalcategoriesand the Languagetypology and syntacticdescription . lexicon. NewYork: CambridgeUniversityPress . Forcedynamicsin languageand cognition. CognitiveScience12. 49- 100. Talmy. L.. 1988 . Cambridge . Formal principlesof languageacquisition Wexier. K. and P. Culicover. 1980 . MA : MIT Press .

) 411- 430. North-Holland Lingua92 ( 1994

* Lexicalreconciliation Jane Grimshaw . RutgersUniversity . 18 Seminary Departmentqf Linguisticsand Centerfor CognitiveScience Place. NeM' BrunsM'ick. NJO8903 . USA In the context of current research in lexical representation there are two fundamental ideas about lexical learning. One holds that the semantics of a word is critically involved in the acquisition of its syntax, another holds that the syntax of the word is critically involved in the acquisition of its semantics. This paper examines the positive results and limitations of the two views, and proposes a reconciliation of the two in which a hypothesized meaning based on observation is the input to the linguistic mapping principles. Thesederive a predicted s-structure, checkedagainst observedsyntax. Learning occurs when the two match.

1. Introduction The now standard lines of reasoning concerning universal properties of languageand languagelearnability lead inexorably to the conclusion that the lexicon that is eventually masteredby a languagelearner is every bit as much a function of the systemmastering it as it is of the linguistic data the learner is exposed to. This represents something of a change - in earlier days of generativegrammar, the lexicon was often defined as a collection of idiosyncratic information . The assumption was that lexical information was simply what was unpredictable and unprincipled, and what therefore must be learned about the language. Since the lexicon was in no way determined by VG , its properties dependedonly on what was observed. In the past ten or fifteen years, we have achievedgreater understanding of the structure of lexical systems, as a result of work on the representation of .

This paper grew out of joint work with Steve Pinker, which was presented at the Boston University Language Development Conference in 1990, and owes much to his input and influence. Many discussionswith Lila Gleitman helped to clarify the arguments immensely, as did conversations with Barbara Landau. Thanks are also due to Alan Prince, the Rutgers-Penn lexicon group. participants at the University of Pennsylvania Lexical Learning Conference in 1992. and the Workshop on Thematic Roles in Linguistic Theory and LanguageAcquisition held at the University of Kansas in 1993, and an anonymous reviewer. . Science8 .v . All - 3841/94/$07.00 @ 1994- Elsevier 0024 rights reserved SSDI 0024 3841( 93) EO045 9

412

J. GrimshawI Lexical reconciliation

lexical items, and the grammatical processes which involve them (Grimshaw 1990, Jackendoff 1990, Levin and Rappaport 1992, Pinker 1989, Williams , in press, Zu bizarreta 1987). With this has come the recognition that lexical systems,just like other linguistic domains, are subject to universal constraints and a high degree of language-internal consistencey. ( For a particularly interesting perspectiveon the relationship between the theory of words and the notion of lexical listing , seeWilliams 1992.) A change in perspectivehas resulted. If the endpoint of the acquisition processis highly principled, then the acquisition processitself must be interesting. The acquisition issuesthat have been addressedin recent researchseemto fall into two subgroups. First , especiallysince Baker ( 1979) there has been an important seriesof studies which attempt to uncover the principles behind the acquisition of patterns of alternation shown by verbs: including Bowerman ( 1974, 1990), Choi and Bowennan ( 1991), Grimshaw ( 1989), Gropen et al. ( 1991), Pinker ( 1989), Randall ( 1990). These questions have been the concern in particular of theories which attempt to make use of mapping between semanticsand syntax. Second, some recent work has investigatedthe question of how the relatively gross meaning of a word can be acquired (Landau and Gleitman 1985, Gleitman 1990, Naigles 1990, Fisher et al. 1991), looking into theories which make use of mapping of some kind between syntax and meaning. Research of the first kind asks how learners determine that give ' ' , We gave participates in the dative alternation ( We gave a book to someone , someonea book) whereas donate does not ( We donated a book to someone * We donatedsomonea book) . Researchof the second kind asks how learners ' ' ' ' detennine that a particular morpheme means give , rather than hold . Whether these are ultimately different questions, with different answers, remains to be seen.

The linguisticsof the lexicon The claim that observation of the world is inadequate for the acquisition of lexical syntax is no surprise - the argument here is just the same as the argument for any other kind of syntax. The theory of lexical syntax involves well-formedness principles for argument structure and its interface with lexical semantics and d-structure: the principles are universal, and not plausibly learned. With respect to lexical meaning, though, this position is perhaps more surprising, although arguments from lexical semantic universals , patterns of errors in learning and so forth , show the effects of va in

J. Grimsha~' I Lexical reconciliation

413

lexical semanticsas elsewhere. The theory of lexical semanticsplays the same role in constraining possible word meanings and the range of solutions that the child will entertain, as the theory of Universal Grammar plays in syntax. For example, it seemsto be the case that while a causative meaning can be expressedeither by a single morpheme, as in ( I a), or by a phrase, as in ( I b), a causative-of -a-causative meaning can never be expressedby a single morpheme , even though it can be expressedby a phrase, as in (c) . ( Ia ) The chemical killed the grass. cause-to -die ( I b) The chemical causedthe grass to die. cause-to-die lc ( ) Overapplication causedthe chemical to kill the grass. cause-to-cause-to -die This restriction is a consequenceof the principles which organize lexical semantic structure and argument structure, some of which are discussedin the work cited above. Thus linguistic theory provides a constrained representational system which limits the solutions available to the learner of the lexicon, along lines familar from other domains. Starting from the simplest point of view, for each morpheme in the lexicon, there are- two sorts of information to be detennined by a learner. These are the syntax of the lexical item, which includes its syntactic category and its subcategorization properties, however they are expressed, and its semantics. Minimally , for example, a learner must detennine that hope is a verb which occurs with a sentential complement, and that it expresses a relation of a certain kind betweenan individual , the hoper, and a proposition or situation ' ( We hope that it won t rain tomorrow, We hope/ or rain tomorrow) . Much current work on lexical theory attempts to uncover the principles of UG which constrain the relationships among various aspects of the lexical representation of predicates and between the lexical representation and the syntactic realization. (On the latter seeespeciallyBaker 1988, Pesetsky1992.) The picture that emergesfrom this researchprogram is one in which the range of syntactic configurations associatedwith a verb is highly predictable from its semantics, once parametric syntactic variation is taken into account. The lexical semantic representation of a predicate determines its argument structures, its argument structures plus parametric properties of phrase structure entail the d-structure configurations of the predicate, and the

414

J. GrimshawI Lexical reconciliation

d-structure plus further parametric effectsentail the s-structure configurations the predicate appearsin. The existenceof principles of thesekinds guarantees that certain relationships will obtain betweenthe s-structure of a verb, noun, or adjective, and its meaning. This basic point is fundamental to almost all current ideas about lexical learning. However, the principled character of the relationship between meaning and syntax is sometimes questioned (for a recent example see Bowennan 1990), so it is important to clarify the nature of the claim that the relationship is indeed principled . The key point is that the notion of mapping is a highly abstract one, in two important respects. First , mapping from lexical semantics is onto d-structure, not s-structure, although all that the language learner and the linguist can observe is the sstructureS -structure of course is affected by other considerations also, some cross-linguistically variable, and thus reflects linking only quite indirectly . In fact, it is not possible to draw conclusions about linking directly from dstructure either, since d-structure itself is subject to parametric variation . All predictions of a mapping theory are made modulo parametric properties, both lexical and grammatical, of the language. Second, mapping does not take as its input an event, but a semantic representation. That is to say, the theory of UG does not say anything at all about how events are described syntactically. What it does say something about is how particular semantic representationsare expressedsyntactically (modulo the parametric properties of the language, asjust discussed). To say, ' then, that a verb which ' means x will map into the syntax in a particular way, is to say that a verb with a particular lexico-semantic representationwill have a particular syntax. Another verb which superficially appears to have the same meaning, in that it describes (at least) approximately the same events, may have a very different syntax. A particularly neat example of this type is the contrast discussedby Levin and Rappaport Hovav ( 1992) between the English blush and Italian arrossire. Both are used to describe the same event, and in this sense they mean the same, yet the Italian verb has unaccusativesyntax while the English verb has unergative syntax. Levin and Rappaport Hovav show that the semantic representations of the verbs are different: arrossire is a change of state predicate, while blush is not. Thus despite their similarity in tenDS of the events they describe, the verbs have different lexical semantic representations, and this is what determines their syntax. The principles which relate one level of representation to another relate the lexical semanticsof a predicate to its (lexical) syntax : they do not relate events in the world to (lexical) syntax.

J. GrimshawI Lexical reconciliation

415

3. Situations and sentences

thelearning asin (2): consider , wecanpose ably problem Idealizing (2) SITUATION\ OBSERVER > WORD REPRESENTATION SENTENCE / The input to the learning systemis a pair of an s-structure, and a representation of a world situation. This representation will be an interpretation of a perceived event, for example, based on observation, previous observations, surrounding discourse, etc. How then does a learner arrive at the lexical representation for the verb in the sentence, given this kind of learning situation? Two ideas form the focus of a large amount of current researchon the question. The first idea is that analysis of the situation can make it possible to determine the meaning of a word , and that the meaning of a word in turn makes it possible to determine its lexicals Yl:ltax. This type of proposal has been very fully developed by Pinker ( 1989), in an important extension of earlier work , which concentrated on the question of how meaning might play a role in allowing the child to perform an initial syntactic analysis upon which the syntactic system of the languagecould be erected, and on ways in which the meaning of a verb could make aspects of its syntax predictable to a learner; Grimshaw ( 1981), Pinker ( 1984, 1989) . The second idea which has emerged on how linguistic relations play their part in acuisition, is that analysis of the sentencemakes it possible ta determine (parts of) its semantics (seeLandau and Gleitman 1985, Fisher et al. 1991, Fisher et al., this volume, Gleitman 1990) . These two ideas about mapping are sometimes contrasted ' under the rubrics ' semantic bootstrapping' and ' syntactic bootstrapping (but seePinker, this volume, for discussion) . ' ' Semantic bootstrapping has one extremely important property : it makes direct use of the principles mapping from lexical meaning to lexical syntax discussedearlier. I will not illustrate this in detail - the works referred to above contain many examples. Nonetheless, it would probably be incorrect to maintain that an analysis of the situation a/one is the input to the learner. Becauseof the indirect relationship between events and semantic representations of verbs, discussedin the previous section, it is not easy for an observer ' to determine a verb s meaning from an event. Gleitman and colleagues(see Gleitman 1990and Fisher et al., this volume, for examples) have looked into this point , showing that eventstypically have multiple construals, hencemany

416

J. GrimshawI Lexical reconciliation

verb meaningswill be compatible with most events. As an example, consider a verb which has a causedchange of state interpretation, with a semantic ' ' representationof the form : x causesy to become- . Such verbs often have a changeof state counterpart, which can utilize the same morpheme, as with melt, or a different morpheme, as in kill / die. The change-of -state version has a semantic representation of the form : ' y becomes- ' .

(3) Wemelted the ice/ the ice melted We killed the dragon/ the dragon died The problem for word learning from world situations is that circumstances that can be described by one member of the pair of verbs, can often be describedequally well by the other: seethe discussion of give and receivein Fisher et al. ,(this volume) for example. The causative entails the change of state, hencewheneverthe causativedescription is true of some state of affairs in the world , the change-of -state description is true also (although not vice versa) . So if a learner guessesthe causative (kill ) when in fact the verb has a change-of -state meaning (die), and if word learning is based on world observation alone, recovery is only possible if some situation tells the learner that even when there is no possibility of construing the event as involving an ' ' agent, die is still used. If a learner choosesthe change-of -state meaning ( die ) ' ' for a: morpheme which is in fact the causative ( kill ), there is no way to correct the mistake, becausethere is no world situation where the inchoative is inappropriate and the causative appropriate. The correct meanings must therefore be assignedto the membersof the pairs by some other means, since learning by world observation seemsat worst to be impossible, and at best to require that juvenile speakershave accessto a disambiguating situation for every pair of meanings like this. But if observation of the world is not enough, other means must be available for lexical learning. This is where key information contained in the sentencein (2) offers the most promising avenue for successfulword learning, very roughly along the lines sketched by Gleitman ( 1990). The fundamental idea is that the linguistics of words itself makes learning words possible. It is the language, and not the world , that supports the processof word learning. Returning to the kill / die problem, although the situations in which the two morphemesare used are not good sourcesof information about the meanings of the verbs, the two are linguistically different in a crucial respect, and this linguistic difference makes word learning possible. What I will for now term, loosely, the ' transitivity ' of kill and causative melt contrasts with the

J. GrimshawI Lexical reconciliation

417

' " intransitivity of die and inchoative mell, and this property is correlated with the meaningsin a way that can be exploited by a languagelearner. What aspects of verbal meaning can in principle be deduced from the . syntactic context a verb appears in ? The proposal advanced by Landau and Gleitman ( 1985) is that learners use information about the surface syntax of a clause to determine (aspects of) the meaning of the verb in the clause. Specifically, they suggestthat the subcategorization frame of a verb contains the critical information . It is easyto seethat this idea will be useful in solving the kill / die problem : since kill is subcategorizedfor an NP complement and die is not , a learner who knew the subcategorizationsin advance could use them to choose the right morpheme for the right meaning. (Presumably variability resulting from parametric variation can be factored out of the situation, and a sufficiently abstract view of the subcategorization sets will make it possible to treat the subcategorization of a verb in one language as being the same as the subcategorization of a verb with the same meaning in another language, despite superficial differencesin the syntactic systemsin the two cases.) When the issueis consideredin a more precisefashion, however, a number of considerations suggestthat subcategorization frames are not the optimal source of information . (See Pinker, this volume, and Fisher et al., this volume, for additional remarks on the limitations of frames for learning semantics.) First , -observable context alone cannot determine what the subcategorization frames of a verb are. Arguments figure in subcategorization frames, adjuncts do not. However both occur on the same side of the verb in a languagelike English, hencethere is no positional evidenceto distinguish one from the other. This problem arises wherever adjuncts and arguments have the same form , e.g. with PPs. Consider the examplesin (4) : while write takes an optional PP adjunct, put has an obligatory PP argument. Similarly , last has a temporal argument, while wriggle occurs with a temporal adjunct. (4a) (4b) (4c) (4d)

He wrote a book in his room. He put a book in his room. The performance lasted for an hour. The performer wriggled for an hour.

Without knowing which expressionsare arguments it is not possible to know what frames the verb appearsin , but in order to know which expressionsare arguments it is necessaryto know what the verb means. Thus it is not clear what role the subcategorizationframes could have in the acquisition of verbal

418

J. Grimshaw/ Lexical reconciliation

meanings in cases where there is no clear independent indication of the argument or adjunct status of a phrase associatedwith the verb. A secondkind of limitation arisesbecauseof the existenceof large numbers of many-to -one semantics-to -syntax mappings. For example, the set of verbs which subcategorizesfor NP is both enormous and extremely disparate semantically:

(5a) He weighedthe tomatoes. (5b) He weighed300pounds. (6a) He becamea doctor. (6b) He shot a doctor. (7a) He asked someonethe time. (7b) He asked someonea question. So the fact that a verb takes an NP complement is not very informative as far as its meaning is concerned. (Of course it is not completely uninformative , since it does make it possible to eliminate a number of candidates: ' put ' and ' die' for , example, are not possible meanings for any of the verbs in (5H7 ) .) The reason that the syntax here is comparatively uninformative is that many different meaningsare mapped onto a syntactic expressionlike NP. Syntactic mapping based on a single subcategorization frame will be maximally effective when the mapping is one-to-one or just few-to -one. This is probably the casewith , for example, sentential complements. A verb with a sentential complement must draw its meaning from a relatively small set of possibilities, which include verbs of communication (say, announce, state), verbs of logical relation (entail, presuppose ), and verbs of propositional . attitude (hope, believe) Being aware of the limitations and problems of mapping from subcategorization frames to meaning in cases such as the ones just discussed, the researchersworking on syntactic mapping have investigated the idea that sets of subcategorizationframes, and not just single frames, playa role in the acquisition of verbal meaning (see especially Landau and Gleitman 1985, Fisher et al. 1991, and the discussion of ' frame range' in Fisher et al., this volume) . The learner will examine the set of subcategorizationsthat a verb appearsin , and discover properties of its meaning in this way. Note that one strong disadvantageof this position is that it is not possible to learn a meaning for a morpheme on exposure, even repeatedexposure, to a single sentencetype. Analysis acrosssentencetypes will be required, since the

J. Grimshaw/ Lexical reconciliation

419

entire set of frames that the verb appearsin , or some approximation thereof, must be observedbefore its meaning can be detennined. There is an important precondition for the successof the ' sets of frames' idea. In order for learners to be able to use a verb' s subcategorizationset to determine its meaning, the learner must know in advancewhich set goeswith which meaning. This is possible only to the extent that the subcategorization set/ meaning mapping is cross-linguistically stable, or shows only parametric variation. So the question is whether UG detenninesthe subcategorizationset associatedwith a verb, or not. This issueturns out to be highly problematic the reason is that the total subcategorizationset of a verb is a function of the set of subcategorizationsin which each senseof the verb participates. And the way sensesare distributed acrossmorphemesis not unifonn acrosslanguages. (This point is also made in Pinker, this volume.) Let us consider why this should be so. The relevant principles of UG govern the mapping between the meaning of a verb and its syntax in that meaning. In addition , UG regulates the set of possible semantic alternations that a morpheme may participate in , such as the causative/change-of -state relation. If each morpheme had exactly one sense, or if the alternative senses of a morpheme were always regulated by UG , this would exhaust the situation. But in fact, a morpheme can have severalsenses , each with its own UG detennined syntax. and each participating in alternations in the ways prescribedby UG . UG says little or nothing about the complete set of senses the verb has, and therefore little or nothing about the total set of subcategorilations of the morpheme. UG only detennines the properties of the individual sensesand those that are related grammatically. As an example, consider the verb shoot. It has at least two senses: one exemplified in She shot the burg/ar , and one in The burglar shot out of the room. In the first sense, shoot takes an NP complement, in the second a PP. In both casesthe subcategorizations are highly predictable. shoot- l is like, say, stab, and takes an individual / NP as its complement, while shoot-2 is like , say raced, and takes a directional PP as its complement. The subcategorization of each senseis completely in accordancewith the theory of grammar, but nothing about the theory of grammar determines that shoot will have these two senses . Hence the theory of grammar cannot possibly predict that shoot will have these two subcategorization frames, and a learner could not know this in advance. In caseslike these, the sensesof the verb do not seemto be related by UG at all , even though they are all realized by a single morpheme. Presumably they are related by association, which depends on semantic field and other

420

J. GrimshawI Lexicalreconciliation

cognitively real but grammatically irrelevant factors. Thus it is probably not an accident that shoot has a verb-of-motion meaning, and stab does not , since shooting involves a rapidly moving bullet. The point is that the relationship is associativein character, and not UG determined. Similarly, consider the case of know as in I know her and in I know that she is here. The two usesof the verb are cognitively close - both are state predicatesand both are psychological . Hence it is presumably much more likely that these two senseswill be clustered together into a single morpheme than that say, eat and one of the two sensesof know will be clustered together. The probabilities involved are not a matter of UG , however, and of course the clustering of sensestogether under a single morpheme is notoriously variable cross-linguistically . The two sensesof know correspond to different morphemes in many other languages: French for one. What is involved in these effects is the spreading of morphemes across meanings. The results range in character from pure ' w here no homonyms significant semantic relation can be seenat all , through relations that seem cognitively natural , and that probably reflect various clustering effects, to relations that are grammatical in character. The point is, though, that the entire subcategorizationset associatedwith a morpheme is simply the sum of the subcategorization sets for each meaning of the morpheme. While it is true that the subcategorization set for each meaning is highly principled and strongly related to the meaning, the same is not true for the entire set. Each individual subcategorization frame reflects the systematicproperties of the form -meaning correspondences , but the entire set of subcategorizationsalso reflects accidental combinations of subcategorizations ' , which result from a single morpheme s appearing in severaldifferent . It follows that there is no stable mapping between the full set of senses subcategorizations that a morpheme appears in and the meaning of the morpheme. The full set of subcategorizationswill depend, not just on UG , but also on the range of meanings that the morpheme assumesin a given language. The conclusion is, then, that a predictable relationship betweensubcategorizations and morphemes does not hold : the predictable relationship is between subcategorizations and a particular sense of the morpheme. It follows that it is not possible to use the entire subcategorization to learn meaning, becausethe entire subcategorizationset is not associatedwith a single meaningin thefirst place. As a result, a learning mechanismbasedon subcategorization setswill give , typically the most common poor results on morphemes with many senses ones. It will predict many errors, or failure of learning. Suppose that the

J. Grimshaw/ Lexical reconciliation

learning device tracks the syntactic context of the morpheme and attempts to determine its meaning based on t.his information . It will collect up all observed subcategorizationsfor a morpheme, disregarding the fact, which it ' cannot by definition be sensitive to , that verb s meaning differs across these subcategorizations. It will then assign to the verb the meaning that would be assigned to a verb which, under a single meaning, occurred in this set of frames. But this will be incorrect, except in the caseof a verb with exactly one sense. For verbs with multiple senses , if the procedure succeedsin assigninga will at all it one , meaning give meaning to the verb instead of several, and the meaning it gives might well be completely unrelated to any actual meaning of the verb being learned. Without knowing about meaning it is not possible to know which subcategorizationsshould be grouped ,ogether, and which shouldbe kept separate. What of lexical entries in which the relationship betweenthe alternatives is regulated by UG , such as the causative/change of state examples discussed above (e.g. melt/ melt) ? Supposethat there are two subcategorization frames associatedwith each of these verbs, one transitive and one intransitive. Can this set of frames be exploited to learn the meaning of the verb? (Note that the existenceof the alternation cannot be crucial for learning the meaningsof the members, becausemany verbs, such as kill and die do not alternate, but are still learned.) Here again the answer is negative, becausethe two occurrencesof the morpheme have different meanings, under standard assumptions about lexical representation. 1 So there cannot be a single sensespecified by UG as associatedwith this subcategorization set. Note that it also cannot be the casethat the shared meaning, i.e. what the two casesof the verb have in common, is learned from the subcategorization set either, becauseall causative /change of state pairs have the same subcategorization set, even though each pair has its own meaning. Clearly what we must aim for is a learning procedure that uses the alternation as a clue to the semantic analysis of the verbs, that might tell the learner that these verbs have changeand causedchangemeanings. However,

1 However, Grimshaw in work in progress ( 1993) denies these assumptions for all cases of alternations involving no overt morphology , proposing instead that there is only one meaning for ' e.g. melt in its causative and inchoative uses. The ' alternation is just the result of the meaning interacting with clausal structure. Further work is required to seehow this ultimately bearson the ' ' learning questions addressedhere, but it seemsthat in this senseof meaning , observation of the two clause structures associatedwith melt is essential for arriving at the correct analysis. This conclusion will hold for the UG -governed alternations only .

422

J. GrimshawI Lexicalreconciliation

just observing subcategorization alternations will not achieve this result, becausethere are other transitive/ intransitive alternations like eat and leave.

(8a) We. meltedthe ice. (8b) The ice melted. (9a) (9b) (9c) ( l Oa) ( lOb) ( I Oc)

We left the room. We left. . The room left. We ate the ice. We ate. * The ice ate.

What will guarantee successis to take into account the properties of the arguments, and not subcategorization alone; it is the fact that the subject of one caseof the verb correspondsto the object of the other casesthat reliably distinguishes change/ causedchange pairs from eat and leave. One final point concerning the causative/change-of -state pairs. In fact both are ' transitive' in d-structure, according to the unaccusativehypothesis (Perlmutter 1978, Burzio 1986), which means that technically they both have transitive subcategorizations. Again , this suggests that subcategorization frames are not exactly the right place for the child to look for help in figuring out verb meaning. We can summarizethe conclusionsso far as follows. Mapping from meaning onto syntax can success fully exploit a set of principles of Universal Grammar which relate lexical meaning ultimately to surface syntax. However, determining the meaningjust from observation of the world seemsto require multiple exposuresacrosssituations in many cases, and may be impossibleunder certain circumstances , when the world is particularly uninformative. On the other hand, fully exploit linguistically mapping from syntax onto meaning promises to success encodedinformation about a verb' s meaning. However as formulated so far, it seemsto require multiple exposuresacross sentences , and also may be in is uninformative. cases when , syntax impossible many

4. Reconciliation Clearly then we seeka model which preservesthe advantagesof both kinds of ideas: which makes it possible to use va mapping principles to regulate

J. GrimshawI Lexical reconciliation

423

syntax, and to use surface syntax to regulate the semantic analysis. One way in which it is possible to combine the essentialgood effects of both types of mapping gives them different roles in the learning process: the semantics-tosyntax mapping principles provide a predictive mechanism, and the observed s-structure provides a checking mechanism. This is the basisof Reconciliation. (Wilkins 1993 explores a rather similar model for lexical learning, with particular emphasison the acquisition of morphology .) Reconciliation

(1) (2)

(6) ( 7) (8)

The learner interprets a scene or situation , hears a sentence and detects the verb. The learner finds a relationship R among participants in the situation ( entities , propositions etc.) that is sensible given the interpretation of the observed situation . The learner checks that R involves participants consistent with the content of the ( candidate argument ) expressions in the sentence, and rejects an R that does not meet this requirement . The learner constructs a lexical conceptual structure which is consistent with R , and assigns candidate argument expressions in the sentence to argument positions in the lexical conceptual structure . This lexical conceptual structure is fed through the semantics- to -syntax mapping principles of UG in their language particular instantiation . The s-structure predicted by step 5 is compared to the observed s-structure . If they do not match then no learning takes place . If they do match then the morpheme is entered into the lexicon with the hypothesized lexical conceptual structure .

A few comments are required about the steps involved in reconciliation. Step 2 excludes situations where the interpretation of the event is one of throwing , say, and R is a relationship betweenpropositions. Step 3 constrains the device to considering Rs that expressrelationships betweenthe right kind of entitities: if the sentencecontains two NPs, and one is the ball, then a verb ' ' meaning say is not a candidate, since it is not a possible relationship betweena ball and some other entity . Similarly , with the NPs John and the hal/, R cannot be a verb meaning 'dress', although it could mean ' throw ' . Whether this should depend on probabilities given real world knowledge, or only on strictly linguistic selectional restrictions, I leave open. Note that step 3 is one way in which the sentenceconstrains the processof word learning,

424

J. GrimshawI Lexical reconciliation

not through the syntactic fonn of the arguments of the predicate but through gross consistency between the (candidate) arguments and the relationship expressedby a candidate R. This step is prelinguistic , in the sensethat it does not rely on the linguistic representation of the arguments, just on their gross meaning. It is quite different, then, from the effects of the linguistic mapping and checking involved in later stages. An interesting question is whether some ' ' of the effectsdescribedin the literature on syntactic bootstrapping are really due to this (non-linguistic) process, see Pinker (this volume) . At Step 4, however, the procedure is now linguistic in character. If R is a causal relationship between two entities, a causative lcs will be constructed. The system is now working with a linguistic representation, rather than just with construals of events, and conceptual properties of relationships and entities. With respect to the final steps, note that an incorrect representation will be entered into the lexicon, if the match discoveredin R6 is accidental. We must assumethat general principles of decay will eventually result in eradication of mistaken entries (cf. Draine 1971), since they will result in matches less often ' than correct entries. Moreover, the notion of a ' match requires further explication. The definition could require identity of fonn between the observed sentence and the predicted sentence, or it could allow for limited inconsistencies , in particular with respect to various ways of reducing the number of arguments that are actually expressedin a clause. Imperatives lack an overt subject, for example, there are elliptical fonns , and there are verbs like wash and eat, which can be syntactically intransitive while apparently ' ' maintaining a two -place semantic structure. Cognate objects (e.g. to die a ' ' peaceful death) and fake reflexives (e.g. to behaveoneself) pose the opposite problem, and a more detailed treatment is required than I will give here. The reconciliation model incorporates aspectsof both semanticand syntactic ' ' bootstrapping . It crucially involves mapping from a posited meaning to a syntactic fonn . It also exploits the surface syntax to constrain solutions. A simple result of the model is that the number, the position , and the fonn of the syntactic arguments of a predicate will constrain the semantic representation it is given. This is becauseof the grammatical principles regulating the lcs- syntax relation. Suppose for example, that a learner hears a sentencecontaining the verb give: Mary is giving the package to the boy, and observesan event in which one individual hands a package to another individual . Suppose the learner interprets the event as involving a three-place logical relationship of transfer . This ' R ' will be consistent with the content of the candidate of possession argument expressionsin the sentence: Mary , the package, and the boy, since

J. GrimshawI Lexical reconciliation

425

theseare the right kind of entities to participate in such a relationship. Now the learner constructs an Ics for R (step 4), say x acts to transfery to z, and assignsthe candidate argument expressionsin the sentenceto positions in the lcs of the predicate. Supposethe learner assignsMary to x, the packageto y, and the boy to z. Then given the semantics-to-syntax mapping principles as they work out in English, the predicted s-structure is Mary verbsthe package to the boy. Since this is the observed s-structure, this experienceyields an Ics for give. (Note that the wrong assignmentof candidate argument expressions to lcs positions would not have yielded a match, hence no learning would have occurred.) Suppose, on the other hand, that the learner construes the event as one of holding or getting, interpretations that are equally consistent with gross observation. Now the learner will posit a two-place ' R ' . At step 3 the procedure might already break down, if the learner can decide that all of the candidate arguments in the clausemust be actual arguments, since there are 3 candidate arguments but only two actual arguments. Perhaps this is not so easy, however, since one or other of the phrases might be involved in some adjunct role. Assuming then, that the procedure will not halt at step 3, what ' happens? The learner now constructs one of the two Ics s, for hold or get, and examines the predicted s-structures that correspond to them. Let us assume that the Ics' s are something like : x havey , x come to havey . There is no way for these Ics representationsto yield the observed s-structure. They have the wrong number of arguments and since there is no way to treat the PP to the boy as an adjunct there is no way to reconcile the observed s-structure with the predicted s-structure. In addition , the arguments will be in the wrong ' ' syntactic position . Semantics-to-syntax mapping will place the getter or ' ' holder the ( boy) in subject position , but this will contradict the observed s-structure which has the giver (Mary ) in subject position . In general, the number of arguments in the observedsentencewill have to match the number of arguments of R : if a predicate expresses an n-place relationship it will have n syntactic arguments, and if it has n syntactic argumentsit will be a logical n place predicate. (I set aside here the disruptions to this generalizationmentioned above.) This is, more or less, what the Theta Criterion and the Projection Principle guarantee(Chomsky 1981) : the number of phrasal argumentsin the syntax is the sameas the number of (open) logical positions in the lexical representation of the predicate, and the syntactic derivation cannot changethe number of arguments. By the samereasoningReconciliationresolvesthe kill / die problem discussed above. A learner can conclude that die can be a change-of -state predicate but

426

J. GrimshawI Lexical reconciliation

not a causative, that kill can be causativebut not change-of -state, and that melt can be both. This can be determinedjust on the basis of a single observation each for kill and die, and two observationsof melt, one in each complement structure. A verb which means to causesomething to change state must have two arguments, a verb which means to change state must have one. Hence, since kill has two syntactic arguments, it must have two semantic arguments and cannot have a change-of-state meaning. Since die has one syntactic argument it must have one semantic argument and cannot be causative. The syntactic form provides the information that there are two semanticarguments, which provides the necessaryinformation about possibleverb meanings. As we saw in section 3, in order to properly identify instances of the alternation betweencausativeand change-of -state meanings, it is necessaryto take into account the fact that the subject of one caseof the verb corresponds to the object of the other : We melted the ice, The ice melted. The properties of the arguments are essentialto distinguishing theseverbs from verbs like leave and eat. If intransitive eat is mistakenly analyzed as having a change of state meaning, its predicted s-structure will have what is eaten in subject position The ice ate. The observed s-structure will have the eater in subject position We ate. Hence the wrong analysis will be rejected. Similarly , if a learner fails to assigna change-of -state analysis to melt, treating it instead as having an lcs like intransitive leave or eat, the predicted s-structure will have the agent in the subject position ( We melted), while the observed s structure will have melted ice The in ). subject position ( entity undergoing the change of state Once again, the error will be avoided. The syntactic form of an argument will similarly constrain meaning. The the syntactic form dependson the semantic properties of the argument, hence and its semantics about information , can provide syntax of an argument hence about the semantics of a predicate. Sentential complements, for instance , will occur with verbs of propositional attitude (e.g. believe), verbs of ) . Reconciliation logical relation (e.g. entail) and verbs of saying (e.g. announce of errors by certain will fact this that types effect desired the has prevent with a is child a in which one is roughly event playing learners. Supposethe the to mean are that dog or being you dog and an adult says either I think ' ' it thinks learner the and That dog will kill you. If the verb means think ' ' while means kill , the predicted s structure will contain an NP complement . Similarly if the the observed s-structure will contain a clausal complement ' and the learner thinks it means ' think ' , the predicted s' kill verb means structure will contain a sentential complement while the observed s structure will contain an NP.

J. Grimshaw/ Lexical reconciliation

427

In sum, under Reconciliation, the number, position, and form of the arguments of a predicate will all constrain the interpretations that can be assignedto that predicate. Of course, the samegeneral point can be made for many other kinds of linguistic infonnation : hearing a verb in the passive form , in the progressive, with a particular aspectual modifier and so forth , will similarly provide constraining information about the verb' s semantics, eliminating many posited, but incorrect, lexical semanticrepresentations. This captures the essenceof the issue addressed by syntactic bootstrapping : it provides a more precisecharacterization of the idea that the languagecan be used to map from observation to verb meanings. In so doing, however, it makes crucial use of the notions behind semantic bootstrapping, concerning the mapping betweensemanticsand syntax.

S. Conclusion

The important properties of Reconciliation are these: - It does not depend on exposuresto multiple sentence types, in the sense that neither cross-situational analysis, nor cross-sentential analysis is involved in setting up the lcs representations. Therefore the problems posed the variable sensesof a morpheme discussedin section 3 do not arise. by - It usessemantics to predict syntax where Universal Grammar makes this possible. - It usessyntax to eliminate wrong semantic candidateswhere possible. I emphasize again that I have addressed here the question of learning basic lexical semantic representations, and not the learnability problem discussedin much of the literature , which concerns the problem of determining which verbs participate in which ' alternations ' , see especially Pinker ( 1989) and referencestherein. Within the terms of the present discussion we could see this as the question of how the system should proceed when a single morpheme would be assignedmultiple representations, whether multiple Ics' s, or multiple syntactic configurations , but I will not explore the issue here. Also unexplored is the issue of how morphologically complex items are analyzed. Even with a procedure like Reconciliation, which exploits a full set of grammatical principles, there is no way to savea learner from having to learn some word meanings simply from observation. There are many are sets of

J. Grimshaw/ Lexical reconciliation

words which have absolutely identical linguistics. Such setsinclude the setsof causative verbs (kill , melt, burn) and change-of -state verbs (die, melt, burn) . Semantics-to -syntax mapping guaranteesthat each member of the set will be syntactically indistinguishable (just as the names of animals, cat versus dog for example, are not syntactically distinct ) . So examination of the surface ' syntax will not inform a learner as to whether a verb means to become ' ' ' liquid or to becomesolid . The general situation is that it is possible to use the surface syntax to constrain analysesof the semanticstructure of a verb, but not its semanticcontent: the fact that a verb is a change-of -state verb, but not the fact that it expresses a particular change of state. This is a matter of semanticcontent only and is not reflected in the syntax of a verb at all. Presumably, then, the semantic differences among members of these sets must be learned from observation about the world in some sense. However, this does not necessarilymean that the differencesare perceptual in character. A vast quantity of information about the world can be encodedlinguistically but is not linguistic itself. Thus a child can observe that melt is used of , for example, ice, while burn is used of , for example, paper. This is sufficient for the child to conclude that the meaningsare as they are rather than the other way around. In this way, it is possiblefor a child to know what a word means without ever having observed an event which would count as an occurrence of what the verb describes. Indeed if this were not the case it would be impo'Ssibleto understand how meaning differencesamong unobservableverbs are acquired: think, hope, imagine. For this reason, language is a source of essential information for lexical learning in two respects. As just discussed, languagecan convey information about word meaning which is orders of magnitude more informative than observation of the world can be. Second, by virtue of the grammatical principles that govern it , language constrains the possible representationsof words in ways that learnerscan exploit in word learning. Reconciliationis one way in which this might happen.

References . Syntactictheoryand the projectionproblem, LinguisticInquiry 10, 533- 581. Baker, C.L., 1979 . Chicago, IL : . Incorporation: A theory of grammaticalfunction changing Baker, M., 1988 . Universityof ChicagoPress . Learningthe structureof causativeverbs: A study in the relationshipof Bowerman , M., 1974 . Papersand Reports on Child Language and syntacticdevelopment semantic cognitive, . 8. StanfordUniversityDepartmentof Linguistics Development

J. Grimshaw/ Lexicalreconciliation

429

. Mappingthematicrolesonto syntacticfunctions: Are childrenhelpedby Bowerman , M., 1990 - 1289 . innatelinking rules? Linguistics28, 1253 . On two typesof modelsof the internalizationof grammars . In: D. Siobin Braine, M. D.S., 1971 of grammar: A theoreticalsymposium , 153- 186. New York: Academic (ed.), The ontogenesis Press . -bindingapproach . Italian syntax: A government . Dordrecht: Reidel. Burzio, L., 1986 . Learningto expressmotion eventsin Englishand Korean. Choi, S. and M. Bowerdian , 1991 Cognition41, 83- 121. . Lectureson governmentand binding. Dordrecht: Foris. , N., 1981 Chomsky . Language . The notion of sourcein languageacquisition Clark, E.V. and K.L. Carpenter , 1989 65, 1- 30. . On the semanticcontentof subcategorization Fisher, C., H. Gleitmanand L. R. Gleitman, 1991 . CognitivePsychology 23, 331- 392. frames Fisher, C., D.G. Hall, S. Rakowitzand L. Gleitman, this volume. Whenit is betterto receive than to give: Syntacticand conceptualconstraintson vocabularygrowth.. Lingua92, 333- 375 (this volume). . LanguageAcquisition1, 3- 55. . The structuralsourcesof verbmeanings Gleitman, L., 1990 . ForD1 Grimshaw , J:, 1981 , function, and the languageacquisitiondevice. In: C.L. Baker, J.J. , MA : MIT , 165- 182. Cambridge McCarthy(cds.), Thelogicalproblemof languageacquisition Press . Grimshaw, J., 1989. Getting the dative alternation. In : I. Laka and A. Mahajan (eds.), Functional heads and clause structure, 113- 122. MIT Working Papersin Linguistics Volume 10. . Argumentstructure. LinguisticInquiry Monograph18. Cambridge Grimshaw , J., 1990 , MA : . MIT Press . The leastlexicon. Colloquiumof the Institute for Researchin Cognitive Grimshaw ; J., 1993 . Science , Universityof Pennsylvania . Affectedness anddirectobjects: The Gropen, J., S. Pinker, M. Hollanderand R. Goldberg. 1991 in the acquisitionof verbargumentstructure.Cognition41, 153- 195. roleof lexicalsemantics . . Cambridge . Semanticstructures Jackendoff , R., 1990 , MA : MIT Press . Cambridge . Languageand experience Landau, B. and L. R. Gleitman, 1985 , MA : Harvard . UniversityPress . Ms. : At thesyntax-semantics interface . Unaccusativity Levin, B. and M. RappaportHovav, 1992 . Journalof Child Language17, . Childrenusesyntaxto learn verb meanings , L., 1990 Naigles 357- 374. . BerkeleyLinguistic . Impersonalpassives and the unaccusative Perlmutter , D., 1978 hypothesis Society4, 157- 189. . Ms., MIT . and cascades . Zero syntax, Vol. 1: Experiencers , D., 1992 Pesetsky . Cambridge . Languagelearnabilityand languagedevelopment Pinker, S., 1984 , MA : Harvard . UniversityPress . Learnabilityand cognition: The acquisitionof argumentstructure.Cambridge Pinker, S., 1989 , . MA : MIT Press ? Lingua92, 377- 410 . How coulda child useverbsyntaxto learnverbsemantics Pinker, S., 1994 (this volume). . Linguistics of languageacquisition . Catapultsand pendulums : The mechanics Randall, J., 1990 - 1406 . 28, 1381

430

J. Grimshaw/ Lexica~reconciliation

Wilkins, W., 1993 . Lexicalleamingby error detection . Ms., ArizonaStateUniversity. . Ms. Williams, E., 1992 . Remarkson lexicalknowledge Williams, E., in press . Thematicstructurein syntax. Cambridge . , MA : MIT Press Zubizarreta, M.-L., 1987 . Levelsof representation in the lexiconand syntax. Dordrecht: Foris.

Section 6

Procedures for verb learning

) 433- 470. North-Holland Lingua92 ( 1994

433

Surface cues and ro bust inference as a basis for the early acquisition of subcategorization frames * M.R. Brent Departmentoj Cognitive Science. Johns Hopkins University. Baltimore. MD 21218. USA

How could children possiblyacquiretheir first subcategorization frames? The hypothesis that they directly observethe syntactic structuresof the utterancesthey hear raisestwo . First, how can childrenparseinput utterancesreliably without alreadyknowingthe questions , how do childrensurvivethe ungrammati syntacticpropertiesof all the wordsin them? Second calor misconstruedutterancesthey inevitably encounter ? This paper suggestsa specific inferencealgorithm that substantiallyreducesthe effectsof ungrammaticalor misconstrued , theycan useapproximatecues input. Sincechildrenmusthavesomesuchinferenceprocedure to determinesyntacticstructure. In particular, theycanusestring-local surfacecuesratherthan . Suchcuesmake it possibleto discoverrelevantsyntacticstructurein an global constraints utterancewithout alreadyknowingall the wordsin it. This papersuggests a possiblesetof cues for English subcategorizationframes that assumesonly the ability to detect the ends of utterancesand knowledgeof a few function morphemesand proper names . Simulation on naturallyoccurring, child-directedEnglishshowthat thesecues, combinedwith experiments the proposedinferencemechanism , do surprisingly well at discoveringsubcategorization frames.

1. Introduction

Individual words differ in the syntactic types of the phrases that can representtheir semantic arguments. For example, watch takes a noun-phrase argument while look does not.

( I a) JanewatchedBob ( I b) *Janelooked Bob .

This work benefited enormously from conversations with Bob Berwick, Lou Ann Gerken, Lila Gleitman, Jane Grimshaw, and Elissa Newport . Any remaining faults ... they probably tried to talk me out of. . ScienceB.V. All - 3841/94/ $07.00 ce 0024 > 1994- Elsevier rights reserved SSDJ0024- 3841(93) EO046- A

M . R. Brent I Surface cuesand robust inference

434

Similarly , pretend takes a tensed clause or an infinitive argument while play does not .

(2a) (2b) (2c) (2d)

Jane is pretending to be grown up * Jane is playing to be grown up Jane is pretending she is grown up * Jane is playing she is grown up

Chomsky ( 1965) referred to these properties of words as their subcategorization frames. In general, a word may have several subcategorization frames, just as it may have several syntactic categories. Within a given language, the subcategorizationframes and the meaningsof words are strongly correlated ( Fisher et al. 1991, Levin 1993, Zwicky 1970) . For example, all English words that take three semantic arguments of which " one is realized as a direct object and one as a tensedclause, as in I told him " ' I m happy , are verbs of communication (Zwicky 1970) . Examples include tel/, write, fax , inform, warn, advise, and so on. It has been hypothesizedthat the correlation between subcategorization frames and meanings plays an important role in lexical acquisition. Pinker and colleagues have proposed that children typically learn the meaning of a word first , then exploit the regular correspondencebetween meaning and subcategorization to infer its subcategorization frames (Pinker 1984, 1989) . Conversely, Gleitman and colleagues have proposed that children often learn the subcategorization frame first , then exploit the correspondenceto restrict their hypothesesabout its possible meanings (Landau and Gleitman 1985, Gleitman 1990) . These two proposals are known as the semantic bootstrapping and syntactic boot 1 strapping hypotheses, respectively.

1.1. Theproblem ' Both semanticand syntactic bootstrapping depend on two-year-olds ability to learn the subcategorization frames of some words without relying on their meanings. This is obvious in the caseof syntactic bootstrapping if children are to rely on the subcategorizations of some particular word to infer its meaning they cannot also rely on its meaning to infer its subcategorizations. In the case of semantic bootstrapping, the problem is that languages vary 1 Semanticbootstrapping also refersto the more generalhypothesisthat lexical semantics ' , Pinker 1984 ). providesthe basisfor childrens acquisitionof lexicalsyntax(Grimshaw1981

M. R. Brent/ Surfacecuesandrobustinference

435

substantially in which particular subcategorization frames correspond to which particular meanings. This variation implies that children must learn the language-particular correspondencebefore using it to infer the subcategorizations from meanings. To learn the correspondence they must learn the meaningsand subcategorizationsof some words independently (Pinker 1989, Gropen et al. 1991). Although the semantic and syntactic bootstrap ping hypotheses have generated a substantial literature , no non-semantic procedure by which children might learn subcategorizations has been developed. There have been insightful proposals about the general type of cues that might be exploited in such a procedure, notably prosodic cues to syntactic structure such as voice pitch, pausing, and vowel duration ( Fisher and Tokura 1993, Lederer and Kelly 1992, Morgan 1986, Fernald and Kuh11987, Hirsh -Pasek et al. 1987, Jusczyk et al. 1993, Kemler -Nelson et al. 1989, Mehler et al. 1988) . This proposal, known as the prosodic bootstrap ping hypothesis, holds that children can recover partial syntactic bracketing (but not category labels) from prosodic cues. However, no explicit procedure has been proposed for recovering bracketings nor for extracting linguistic regularities from them on the basis of prosodic cues. It seemsthat young children must infer some of the lexical syntax of their languages- the syntactic facts about individual words - from larger syntactic structures. But it is difficult to see how children could identify syntactic structure in an utterance without already knowing the syntactic functions of some of the words in the utterance. This posesan apparent paradox : to learn lexical syntax, children must recover the syntactic structure of the input ; to recover syntactic structure, they must already know lexical syntax - the bootstraps need bootstraps.

1.2. Hypotheses This paper investigatesthe possibility that children first learn the syntactic functions of a few words that are extremely common and highly informative about syntactic structure, then exploit these words as probabilistic cues to key syntactic structures in the input utterances. The function morphemes - prepositions, determiners, inflection , pronouns , auxiliary verbs, and complementizers - are typically the shortest, most common , and most syntactically informative words in a .language - ideal starting points for learning syntax (Morgan et al. 1987, Valian and Coulson 1988) . The fact that children do not use function morphemes consistently until quite late

436

M . R. Brent ! Surface cuesand robust inference

has suggestedto some that young children do not know them. Logically , however, children must know some function morphemes by the time they are learning subcategorization frames - they cannot observe that look subcategorizes for a prepositional phrase without the ability to recognize even one preposition . Further , the next section reviews a substantial body of empirical evidence that English-speaking children know the syntactic privileges of a number of function morphemes before they are two -and-a-half years old . Although function morphemes provide a great deal of infonnation about syntactic structure, they do not provide enough for complete, unambiguous syntactic parsing. This paper investigatesthe hypothesis that young children can learn lexical syntax on the basis of partial and uncertain syntactic analyses of input utterances, and that they can cope with the resulting misconstruals using statistical inference. 1.3. Context and methodology

From one perspective, the current approach resemblesparameter-setting models (Chomsky 1981, 1986; Lightfoot 1991) : it assumesa fixed, finite menu of subcategorization frames from which a lexical entry is selectedfor each verb. For each verb V and each subcategorizationS , the presenceor ' absence of S in V s lexical entry can be seen as a binary -valued lexical parameter. The function -morpheme cues proposed here can be seen as triggers for the subcategorization parameters. For present purposes it does not matter whether the menu of possible subcategorizationframes is innate or acquired- only that knowledgeof the menu is independentof the mechanisms children use to selectfrom it. The statisticalinferencecomponent of the current proposal can be seenas redeeminga promissory note left (often unsigned) by work on parameter-setting models. As Lightfoot ( 1991: 19) puts it : 'We shall need to characterizethe robustnessof the data which may act as a trigger. . One can be sure that Robustnessis presumablya function of saliencyand frequency " " arenot alwayssetby singleevents;that wouldmakethechild too triggerhappy parameters , and inclinedto draw long-tenDconclusions(a metaphor ) from insufficientdata. However than others.' someparameters may requiremoretriggeringexperience A precise algorithm for determining the amount of data required to set ' ' each parameter is represented in section 3 (and in the Appendix ) . This appears to be the first explicit proposal in this domain . This work suggests that weighing linguistic evidence is by no means a trivial problem ; indeed , as

M.R. Brent/ Surface cuesandrobustinference

437

a major component of any realistic mechanism for language acquisition, robust inference deservesmuch more attention t~an it has received to date. Further , robust inference procedures make available new approaches to languagelearning problems and new methods for evaluating them. Ultimately, this gives the work presented here a character quite different from that of both parameter-setting models and learnability models (e.g. Wexier and Culicover 1980, Morgan 1986). The ability to cope with occasional input errors implies the ability to use probabilistic or approximate cues. In English, for example, the word to followed by a word that can be interpreted as an uninflected verb usually marks an infinitive . However, the sentenceJohn drove to fish markets allover town is an exception to this generalization. If such exceptionsare rare enough (and distributed appropriately) the generalization can still be used as a cue; an inference procedure can discount evidence generated exclusively by the exceptions. The infinitive cue is local, in the sensethat it can be used without regard to the rest of the sentence, if there is a mechanism for handling the resulting errors. It can also be thought of as a surface cue (Kimball 1973). By proposing that children use surface cues, the present model comes closer to contacting the actual input than parameter-setting and learnability models, where it is typically assumed that the input to syntax learning includes bracketeddeep-structure or surface-structure representations . Subcategorization frames can be read directly off the deep structure, so assuming such inputs is tantamount to assumingaway the problem under investigation here. Closer contact with the input makes it natural to test models of this kind by computer simulation on naturally occurring input ( Bever 1991). This paper reports on a simulation using transcripts of the child-directed English from the CHILDES database(~ acWhinney 1991). The resultsof the simulation support the notion that it is possible to learn subcategorization frames without parsing, without knowing much lexical syntax aside from a few function words, and without relying on semantics. As a method of investigation , simulation is quite different from the theoretical arguments presented by Wexier and Culicover , Lightfoot , Morgan , and others. One important difference is that it takes account of the quantitative structure of naturally occurring , child -directed language. It has been suggestedthat the presence of a few, high -frequency syntactic markers like the common function words may be a critical property that makes languageslearnable (Morgan et al. 1987, Valian and Coulson 1988) . For example, if English had fifty equally frequent determiners, fifty verbal inflections , fifty complementizers, and so on , it might not be learnable.

438

M . R. Brent / Surface cuesand robust inference

Simulation experiments are responsive to such quantitative properties of language, while learnability theory is not. Similarly , child -directed speech tends to contain shorter sentencesand more questionsthan adult-directed speech, and to contain repeatsof the same phrase in different contexts (Morgan 1986, Newport et al. 1977) . It has been suggestedthat this may either aid or disrupt language acquisition. In either case, simulation experiments using transcripts of child -directed speechreflect the distributional properties of child -directed sentences , whereas theoretical do not. arguments Although simulation has a number of advantages, no one mode of investigation resolves all questions. While some of the strategies embodied in the cues for English will carryover to some other languages, the cues themselves are necessarilyspecific to English, and this limits the conclusions one can draw from simulation experiments that use them. Further , the present simulation takes as its input orthographic transcripts of child -directed speech. Such transcripts are closer to the input children actually receive than deepstructure trees, but they are much more abstract than an acoustic signal. 1.4. Organization

The remainder of this paper is organized as follows. Section 2 reviews the evidence that children younger than two -and-a-half years old are able to recognizespecific function morphemesand to exploit them in understanding, despite their failure to use them reliably in speech. Section 3 develops one possible implementation of the current proposals. Section 4 presents experiments using a computer model and transcripts of child -directed speechfrom the CHILDES corpus. Finally , section 5 summarizes the results of this investigation and draws conclusions. 2. Children ' s resources

One of the major hypotheses advanced here is that children rely on the distribution of function morphemesin the early acquisition of subcategorization frames. This hypothesis is only plausible if young children recognize the syntactic privileges of function morphemes. In addition , the model presented in section 3 assumesthat young children know some proper names by the time they need to learn syntactic frames. Finally , the model assumesthat children can detect the ends of utterances using prosodic cues such as pause

M. R. Brent! Surface cuesandrobustinference

439

length, pitch , vowel duration , and volume. This section reviews the empirical basesfor theseassumptions. 2.1. Function morphemesand names The fact that young children do not use function words reliably has sometimesbeenexplainedby claiming their grammarslack function morphemes ' ( Radford 1988, 1991). It has been further proposed that children s putative ignorance of function morphemesresults from a perceptual failure - they do not hear or cannot distinguish function morphemes in most languages becausethey are unstressed(Gleitman et al. 1988) . However, a substantial . body of convergentevidencehas accumulatedagainst eachof thesehypotheses Briefly, it has been shown that : ( I ) Even infants can distinguish unstressed syllables as well as stressed syllables (Jusczyk and Thompson 1978, Williams and Bush 1978) . This refutes the perceptual 4explanation' of children' s putative ignorance of function morphemes. ' (2) Constraints on the metrical and prosodic phonology of children s speech provide better explanationsof their patterns of omission than simple ignorance (Demuth 1993, Gerken and McIntosh 1993, Gerken 1992, Wilnen et al. 1992). This removesthe motivation for the ignorancehypothesis. (3) Children are sensitive to the presence and correct usage of function morphemes in adult speech, as assessedby a variety of techniques (see below). This directly refutes the ignorance hypothesis. The direct evidence that children are sensitive to the presenceand correct usageof function morphemesis summarized below: ( I ) On imitation tasks, young children (age 1; 11- 2;6, MLU 1.30- 2. 00, M = 1.73) omit English function morphemes more often than unstressed nonsense syllables (Gerken et al. 1990) . This effect persists when the sentencesare synthesizedautomatically to control for possible prosodic differencesbetweengrammatical and ungrammatical sentencesspoken by humans (age 2;0- 2;6, MLU 1.57- 2. 60, M = 2. 07), and when the nonsense words are composed of phonemic segmentssimilar to those of English function morphemes (age 2;0- 2;6, M MLU = 2.20) . This indicates that children younger than two-and-a-half distinguish between English function morphemes and similar nonsense syllables - i.e. they know the

440

M. R. BrentI Surfacecuesandrobustinference

specific segmental content English function morphemes, if not their syntactic and semantic functions. (2) Young children tend to interpret a novel noun as referring to a class of " " objectswhen it is taught with an article ( This is a dax ). When it is taught " " without an article ( This is dax ) they tend to interpret it as referring to a unique individual (Katz et al. 1974, Gelman and Taylor 1984). (GT subjects age2;2- 3;0, mean 2;6. KBM : girls, mean age 1;5.) This indicatesthat young children recognize some English detenniners and that they distinguish betweencommon and proper nouns both syntacticallyand semantically. 3 ( ) Young children understand grammatical English sentencessignificantly more often than sentencesin which the function morphemes are deleted (Shipley et al. 1969, Petretic and Tweeney 1976), or in which function words are replaced by nonsensesyllables (Petretic and Tweeney 1976) (SSG: age 1;7- 2;5 mean 2; 1, median-LV I 1.85. PT: mean age 2;3, MLU 1.07- 1.66) . The effect persists when function morphemes are replaced by others of the wrong category, as in * Find was bird for me, where was is substituted for a determiner (Gerken and McIntosh 1993) . ( Least mature group, MLU < 1.50, M = 1.15, produced no articles during session.) This indicatesthat two-year-olds who almost never usefunction morphemes can neverthelessdistinguish their syntactic privileges. Taken together, these experiments demonstrate that ~hildren younger than two-and-a-half years old, who almost never produce function words, know the segmental, syntactic, and semanticproperties of determinersand auxiliaries. In light of this evidence, there is no reason to doubt that they have similar facility with pronouns and inflectional suffixes.

2.2. Prosody The prosodic properties of a sentenceare linguistically significant properties of its sound including intonation and rhythmic properties like vowel duration , pausing, and volume. These properties are believed to be governed by rules that are distinct from but sensitiveto syntactic structure (Nespor and Vogel 1982, Selkirk 1981) . It has been hypothesized(under the rubric of prosodic bootstrapping) that young children recover a sybstantial amount of syntactic structure, perhaps even an unlabeled bracketing of all major syntactic constituents, on the basis of prosodic information (Morgan 1986). Substantiatingthis hypothesisrequires, at least, establishingthat young children attend to prosody and that it is possible

M . R. Brent ! Surface cuesand robust inference

to recover syntactic structure from the prosodic properties of sentences without already knowing the language. There is ample evidence for the former ( Fernald and Kuh11987, Hirsh -Paseket al. 1987, Jusczyk et al. 1993, Kemler-Nelson et al. 1989, Mehler et al. 1988), but little for the latter. A number of studies have been done, but they used carefully constructed pairs of ambiguous sentences , read aloud ( Lederer and Kelly 1992, Morgan 1986). As Fisher (in press) points out , the prosodic properties of reading aloud may well be different from those of fluent speech, and the syntactic properties of the constructed sentencesare certainly different from those of the very short sentencesthat typify speechdirected at young children. Further , the readers in some of these experiments may have been aware that the sentenceswere ambiguous, and hencetried to disambiguate them prosodically. However, Fisher and Tokura ( 1993) have shown that the boundaries of utterances - sentencesor sentencefragments - can be predicted from the prosodic properties of natural , infant -directed speech in both English and . For each syllable in samplesof natural speechby mothers to their Japanese 8- and l4 -month-old children, they measuredpauseduration , vowel duration , fundamental frequency excursion within each vowel, and average amplitude within each vowel. Each syllable was labeled as utterance-final , phrase-final , word -final , or non-final and a discriminant analysis was performed on onehalf of the sample in each language. When the resulting classification was applied to the other half of the sample, it correctly predicted 61% of all utterance boundaries (false negative error 39% ) . More importantly , 93% of all the utteranceboundariesit predicted were correct (false positive error 7 % ), even though only 28% of all syllableswere utterance-final. Similar resultswere obtained for predicting utterance-final syllablesin Japane~ , using a classification function basedon the Japanesesample. Remarkably, the classifier based on the English sample performed almost as well at predicting utterance-final syllables in Japaneseas did the one based on the Japanesesample. It is possible, then, that a single classification procedure for finding utterance boundariesis innate and universal. When the samemethod was applied to find a classificationfunction for predicting syllablesthat end either an NP or a VP, but not a whole utterance, the results were roughly at chance.

3. An implementation This section describesone possible implementation of a learning strategy based on function morpheme cues and statistical error reduction. This

442

M . R. Brent / Surface cuesand robust inference

implementation servesas a carefully worked out example that clarifies some of the implications of such a strategy. It also servesas the instrument for the simulation experiment presentedin the next section. 3.1. Overview This subsection explains how the model works in general terms ; the following subsections fill in the details . First , note that subcategorization frames , as described by Chomsky ( 1965), do not distinguish between tensed clauses and infinitives - both are described as sentences. Sentential complements are realized as tensed clauses or infinitives as a result of lexical features and transformations . The cues described below , however , recover only the syntactic forms of arguments as they occur in utterances . (Boguraev and Briscoe , 1987, provide evidence that it is usually ' possible to recover the lexical features required in Chomsky s analysis from surface argument types mechanically , but that question is not addressed here.) In addition , this implementation recovers some information that is usually thought of as function word selection , rather than subcategorization . the To avoid confusion , representations recovered in this implementation are referred to as syntactic frames , or simply frames . The syntactic frames inves.tigated here , excluding selection for prepositions and complementizers , are shown in table I . Table I The syntacticframesstudiesin the simulation

SF description

Good example

Badexample

NP only tensedclause infinitive PP only NP & clause NP & infinitive NP & NP NP & PP

greet them ' hope he ll attend to hope attend listen to me ' tell him he s a fool want him to attend tell him the story put it on the table

* arrive them * tell he' ll attend * greet to attend * put on it ' * yell him he s a fool * to attend him hope * want him the story * Iisten him to me

The task of learning follows :

syntacticframes for verbs can be broken down as

M.R. Brent! Surface cuesandrobustinference

443

( 1) Collect observations about the syntactic frames manifested in each input utterance: (a) Identify a verb, V , in the utterance (b) Identify phrasesin positions where complementsof V can occur (c) For each potential complement, C, determine whether C is in fact a subcategorizedcomplement of V 2 Make inferencesabout the lexicon by analyzing the observations in such ( ) a way that the effects of ungrammatical and misconstrued input are minimized Step I and Step 2 could be interleaved, so the lexicon is potentially updated after each input utterance, but the implementation presented here draws conclusions only after a body of observations has been collected, since this approach is con,ceptually simpler. The implementation presented here includes cues for collecting observations in English and a novel inference method that can be applied to other cues and to other languages. The cues assumeno knowledge of lexical items except for proper namesand a few function morphemes. Although they refer to categorieslike determiner and pronoun , the initial lexicon of items in each category need not be complete. The cues used for detecting verbs in this implementation have two logically separable components: one that identifies words that can be verbs, and another that excludes occurrencesof such words in contexts where they are unlikely to befunctioning as verbs. As each utterance is processed, any word that has already occurred both with and without the suffixing is assumedto have a verbal sense. All occurrencesof such a word are taken to be verbal occurrencesexcept when preceded by a known determiner or . preposition verbs are very rare in those positions. The cuesusedfor identifying phrasesin this implementationare as follows (see table 3, page450). Proper nouns and pronouns are taken as noun phrases( NPs). (This classof words is called lexical NP, or LNP , hereafter.) Detenniners are also taken as signalingNPs. ( Theclassof detenninersand LNPs is referredto as NPLE , for NP left- edge, in table 3.) Prepositionsfollowed by NPs are taken as PPs. To followed by a word that has alreadyoccurredboth with and without the suffixing is taken as marking an infinitive. A complementizer(such as that) followed by a determineror NP is taken as the beginningof a tensedclause. So, too, is an NP followed by a previouslyidentified verb. Phrasesare only relevant when they are in a position to serve as part of a syntactic frame that the learner is looking for . For example, there is no

444

M.R. Brent! Surfacecuesandrobustinference

frame that contains a tensedclausefollowed by an NP , so the implementation does not look for that sequenceof phrases. The problem of whether a phrase is in fact a subcategorizedcomplement of a given verb is resolved in an approximate manner, leaving the errors to be cleanedup by the inferencecomponent. In particular, a phrase is taken to be a complement of a verb if it follows that verb immediately, or if it follows a proper name or pronoun that follows the verb immediately. Further , NPs are taken as complementsonly if they are unlikely to be subjects- specifically, if they are followed by an utterance boundary or a cue for another complement phrase. Although the details of this strategy are specific to English, there appears to be a cross-linguistic tendency for subcategorizedcomplements to occur near their verbs. The statistical inference component takes as input the number of times each verb has occurred with cues for each frame, along with the total number of times each verb has occurred in any context. Thus, the implementation keepstrack of three things as it collects observations: ( I ) Which words have occurred both with and without -inK, not counting occurrencesprecededby a preposition or a detenniner. These words are . treated as having verbal senses (2) How many times each word with a verbal sense has occurred, not c' Ountingoccurrencesprecededby a preposition or a detenniner. 3 ( ) How many of those occurrenceswere followed by cues for each of the syntactic frames. This infonnation is kept in a data structure called an observationstable. An alphabetically contiguous portion of the observations table from an experimental run is shown in table 2. Each row representsdata collected from one of word fonns , an -ing fonn and its stem form . The first column, titled pair V , contains the total number of times the word occurs in positions where it could be functioning as a verb. Each subsequentcolumn representsa single frame. ' CI' stands for a tensed clause argument with that or null Cornple' ' ' mentizer, ' in ' , ' on' , and ' to for PPs headed by those prepositions, inf for ' ' infinitive V Ps, and wh for clauses headed by what or where. The number appearing in each row and column representsthe number of times that the ' row' s verb co-occurred with cues for the column s frame. Zeros are omitted. For example, pretend and pretending occurred a combined total of 14 times, excluding those occurrencesthat followed detenninersor prepositions. Two of those occurrenceswere followed by a cue for a tensedclauseand one by a cue

M . R. Brent ! Surface cuesand robust iliference

445

Table2 Rawobservations , for the first 7Sverbsin alphabeticalorder , beforestatisticalprocessing

V NP NPNPNPclNPinNPinfNPonNPtoNPwhcl in inf on to wh

bring brush button buy call carry catch cheat chew clean close color comb come cook cry cut dance do draw drink drive eat even fall feed feel find fish fix fly forget get give go grow have

-

21 118 17 52 54 25 12 88 19 6 3 13 5 2 1 15 1 10 97 43 99 1 25 6 298 6 19 1 42 5 16 5 21 -

add ask be blow

142418410 23 I 33 4 19 4 254 64 13 46 25 4 55 15 103 34 17 32 15 6 6 I 398 84 3 91 4 8 982 5 3 506 59

2 1 1

2

162 1

1 7

1 3

2

I I 24 113 3 26

3 9

3

3 S6 3

M.R. BrentI Surfacecuesandrobustinference

446

Table2 (Cont.) V

jump keep kick knock

in inf on to wh

32 48 23 27 5 3 24 27 3 34 5 2 17 I 13 7 12 ' 15 6

laugh lay leave lick lie lift listen live look make meet miss move mow

824 4 18342 2 I 31 II 30 5 2

nap open pat pick pitch play pour pretend pull push put rain

123SS 41 28 66 3 2 228 I 12 14 S7 9 I 44 S I 99S46 I 7

18 26

I 12 21

9 I -

hide hold hurt invent

NP NPNP NPcl NPin NPinf NPon NPto NPwh cl

2S

I I 5

for an infinitive . The observations table is the input to the statistical inference procedure. The statistical inference component of this implementation has two parts : one estimatesthe error rate for each frame, and the other usesthis estimate to weigh the evidenceprovided about that frame by multiple occurrencesof each verb. In particular , the latter component considers the number of times the

M . R. Brent I Surface cuesand robust inference

447

verb occurred with cues for the frame and the number of times it occurred without , much as one would consider the number of headsand the number of tails turned up by a coin to determine whether or not it is fair . The method of estimating the error rate was designed specifically for this implementation, while the method for weighing the evidence in light of the error rate is a standard one. The error estimation procedure is based on some simplifying approximations . First , for each frame , a single source of error dominates all others in terms of frequency. Second, errors from this source are distributed evenly across verbs. Third , verbs that in fact have a given frame occur with cues for that frame with significantly higher probability than verbs that do not have it . Put another way, false alarms are less frequent than true alarms. To the extent that these approximations hold , the verbs with cues for a given frame consist entirely of false whose co-occurrence ' alarms should have a relatively low rate of co-occurrence with those cpes, and their rates of co-occurrence should be distributed binomially . These facts can be used to identify a sample of verbs whose co-occurrence with cues for a frame consist entirely of false alarms, or miscues. This sample can then be used to estimate the error rate. The remainder of this section provides a more detailed discussion of the implementation and the rationale behind its design. Readers interested primarily in the experimental results can proceed to section 4 without loss of continuity . 3.2. Collecting observations

3.2.J. Finding verbs Whyfind verbs? Although verbs are not the only words with subcategorization frames, their syntactic properties are sufficiently different from those of other words that they must be analyzed separately. One important difference betweenverbs and nouns is illustrated in (3) :

(3a) (3b) (3c)

John liked to pretend[cp he wasat the theater] *John liked to play [cp he wasat the theater] John liked the play [cp he sawat the theater]

(3a) is grammatical because the verb pretend takes a tensed clause (CP) argument; (3b) is ungrammatical because the verb play does not ; (3c) is grammatical becauseplay is functioning as a noun and the following clauseis

448

M . R. Brent / Surface cuesand robust inference

a relative clause rather than an argument of play . Any common noun can have a relative clause, so noticing a tensedclauseis not informative about the syntactic frames of a preceding noun, but it is informative about those of a preceding verb. A similar difficulty arises from the fact that , in English, verbs can take NP complementsbut nouns cannot:

' ] [NPthe world s largestbank] (4a) We plannedto [" purchase ' ] [NPthe world s largestbank] wasprofitable (4b) . Our [N purchase Further, the NP complementof a verb tendsto becomea PP headedby of in the verb's nominalization: ' ' (5) Our [N purchase] of [NP the world s largest bank] was profitable Thus if observations of purchase the noun and purchase the verb were not distinguished, they would provide contradictory evidence about whether or not purchasetakes an NP complement. This has consequencesfor nouns that are not ambiguous as well - cues indicating that a word has an NP complement should be disregarded in the presenceof evidencethat the word is a noun. The proposed statistical inference method adds to the importance of distinguishing between nouns and verbs. It is based on the simplifying assumption that miscuesfor a given frame are distributed fairly evenly across words. This assumption is violated if data from nouns and verbs are pooled. To see this, suppose that no discrimination were made between nouns and verbs. Further , suppose that a clause following a word W were taken as evidencethat W takes a tensedclause argument. Example (3) shows that the likelihood of that evidencebeing incorrect is much higher if W is a noun than if W is a verb. Thus, distinguishing betweennouns and verbs is important for obtaining a relatively uniform distribution of errors. This would be true even if there were no lexical ambiguity . Cuesfor finding verbs. In English, there are several function -morpheme cues that provide evidenceabout major category, but none of them is ideal. In general, the simplest cues tend to be either less reliable or less common than more complex ones. For example, a simple, reliable cue is that words following a form of be and ending in - ing (progressives) are very likely to be verbs. Unfortunately , a number of words occur in the progressiveconstruc-

.MR . Brent / Surface cuesand robust inference

449

tion very rarely, if ever. These include some of the most common and most syntactically interesting words in child -directed speech, such as know, want, like and see. The following two-part procedure for identifying verbs is a reasonable compromise among simplicity , frequency, and reliability : first , identify words that occur as verbs ; second, when such a word occurs, assume it is functioning as a verb , except in the presence of evidence to the contrary . All words that can occur as verbs can occur both with and without the final syllableing , and all but a few words with that property can occur as verbs. Even mental state verbs like know, which rarely occur in the progressive, have nominalizations ending in -ing (Knowing how to read a map is useful) . A learner who has identified potential verbs can now proceed by trying to weed out their non -verbal occurrences. Occurrences , prepositions , or non -auxiliary verbs, for example, following determiners ' are unlikely to be verbal . The simulations are based on the -ing cue for potential verbs and the determiner and preposition cues for non -yerbal occurrences.2 However , these cues are not without difficulties . Nominalizations of mental states like knowing, linking , and wanting appear to be rare in child -directed speech, perhaps becausethey are semantically abstract. A second problem is that some ambiguous words occur as nouns more often than they occur as verbs, even when occurrences following determiners and prepositions are excluded. 3.2.2. Identifying potential complements This implementation focuses on four complement types: Noun phrase NP ( ), prepositional phrase (PP), infinitive verb phrase (VP), and tensedclause ' or ' compleinentizer phrase (CP) . Within thesemajor categories, PPs and CPs are distinguished by their head prepositions and complementizers, respectively . For example, a CP headedby that is treated as a separatecomplement type from one headed by what or where. The goal is to find representativesof each complement type that occur frequently in the input , and that can be identified without the need to know a lot of other words. The words that must be known to identify these ' ' 2 In speechthere is only one allomorph of the morpheme ling / . In written English the ing suffix can induce spelling changes at the end of the stem, and the simulation includes some techniques for negotiating those spelling changes. Description of these techniques has been omitted since the orthographic problem they solve does not occur in auralleaming .

450

M.R. Brent/ Surfacecues and

robust inference

representatives should be very common and regular, preferably without morphological variation . In terms of these criteria , pronouns and proper names (i.e., lexical NPs) are ideal representatives of the noun phrases. Determiners are also in identifying NPs. Given these NPs, a simple cue for prepositional phrasesis a preposition followed by an NP. The best cues for infinitive V Ps and tensedclauses(CPs) take advantageof the verb list that is being learned from the -ing cue. Specifically, any verb that follows to is interpreted as an infinitive . Similarly , any lexical NP followed by a verb is taken to be the subject of a CP. In addition , a complementizer followed by a lexical NP - as in that he or where it - is taken to head a CP. The complementizer that is frequently omitted , so the first CP cue, which does not rely on explicit complementizers, is important . The English cues outlined above are summarized in table 3.3 NP left-edge (NPLE ) refers to the class of lexical NPs and determiners. Table3 The cues used by the simulation for identifying complements . LNP (Lexical NP) includes . NPLE (NP left edge) includesLNP and the determiners pronounsand personalnames

Phrasalcategory

Cue

NP

NPLE

Example Don' t eat that.

pp VP(inf) CP

P NPLE 10V C NPLE (C) LNPV

Put your toy on die ftoor. Do you like to dance? I hope that you like it . I know you drink vodka.

I sawa eat.

3.2.3. Determining subcategorization Once the learner has found a verb and a potential complement phrase, it must detennine whether the phrase is in fact a subcategorizedcomplement of the verb. This implementation takes a phrase P as a complement of a verb V if : ( 1) P is ' near' enough to V , and (2) P is unlikely to be the subject of 3 Thesecuesdo not distinguishbetweenthe headsof prepositionalphrasesand post-verbal " " , like the up in I looked up an obscureword yesterday. Particlescan appeareither particles " betweenthe verband the NP or after the NP, as in I look themup everyday" . In the fonDer casethe cueswould indicatea PP headedby up. It is not difficult to imagineproceduresfor analyzingthe output. of the current systemto discoverthe particle alternation, but such exceedthe scopeof the currentpaper. procedures

M .R. Brent ! Surface cuesand robust inference

anotherclause. P is near enoughto V if P follows V immediatelyor if one lexical NP (LNP ) intervenes. It follows that a verb is recognized as taking two complementsin a singlesentenceonly if the first is an NP. A phraseP is deemed unlikely to be a subject if either, (i) it is an NP followed by an utterance boundary or a cue for another argument phrasetype, or (ii ) if it is not an NP. The sametwo- part strategymay be applicableto many languages , although the specificcriteria for nearnessand non-subjecthoodmay be different. Like all syntactic cues, the nearness criterion is imperfect. Indeed it is incorrect for several types of grammatical sentences , such as: (6) John put [NP the toy Mary wanted] on the floor In (6) the nearnesscriterion classifies ' on the floor ' as a complement of 'anted, when in fact it is a complement of put . The argument-adjunct distinction poses another problem for the nearness criterion. (7a) John wanted ( * in order) to pursuea career in finance [Argument] (7b) John resigned(in order) to pursuea career in finance [Adjunct ] In (7a) the infinitive VP ' to pursue ...' is a subcategorizedcomplement of " 'ant in the sensethat : (i ) removing it renders the sentenceungrammatical, and (ii ) its semantic function as an aspiration is a special lexical property of " 'ant and a few dozen other verbs. Thus, the nearnesscriterion gives the correct result. The infinitive VP in (7b), by contrast, can be removed without affecting the grammaticality of the sentence, and its semantic function as the purpose for which an action was carried out is not special to a small class of verbs. The infinitive in (7a) is called an argument while the one in (7b) is called an adjunct. Purpose adjuncts can be diagnosed by inserting ' in order' before the infinitive - if the infinitive is an argument the result will be very bad. but if it is an adjunct the result will be a perfect paraphrase. The nearnesscriterion fails to distinguish between subcategorizedarguments and non-subcategorizedadjuncts. However, it should be noted that no syntactic analysiscan distinguish arguments from adjuncts, unlessit already knows the syntactic frames of the verbs in question. Although purpose adjuncts can be easily diagnosed in most cases, it is worth noting that there are no satisfactory, necessaryand sufficient conditions for distinguishing between arguments and adjuncts in general (Adams and Macfarland 1991).

M. R. Brent/ Surfacecuesandrobustinference

3.3. Robustinference Children hear ungrammatical utterances, and probably misconstrue some grammatical utterances, but these aberrations appear to have no impact on their ultimate grammars. This would not be possible if they drew irrevocable conclusions on the basis of single observations. This section suggests a method that appears suitable for weighing evidence from multiple observations . It exploits the standard method of hypothesis testing, along with a novel method for estimatizing the necessaryparameters. In the following discussion, verbs whose lexical entries permit some particular syntactic frame S are described as +S verbs; the remainder are describedas - S verbsS verbs may occur with cues for S either becauseof ungrammatical input or becausethe cues are imperfect diagnostics. Both of thesesituations are referred to as miscues. The occurrence of miscuescan be thought of as a random process. Each occurrenceof a verb V has some non-zero probability of being followed by a cue for a frame S, even if V cannot grammatically occur in S. If this model is accurate, the more times V occurs, the more likely it is to occur at least once with a cue for S. The intransitive verb arrive, for example, will eventually occur with a cue for an NP argument, if enough text is considered. A learner that treats a single occurrence of a verb followed by a cue as conclusive evidencewill eventually come to the false conclusion that arrive is transitive. In other words, the information provided by the cues will eventually be washed out by the noise. This problem is inherent in learning from naturally occurring language, since infallible parsing is not possible. The only way to prevent it is to consider the frequency with which each verb occurs with cues for each frame. In other words, to consider each occurrence of V without a cue for S as a small bit of evidenceagainst V being able to occur in frame S. Given a syntactic frame S, the statistical model treats each verb V as analogous to a biased coin, and each occurrence of V as a flip of that coin. An occurrencethat is followed by a cue for S correspondsto one outcome of the coin flip , say heads; an occurrence without a cue for S corresponds to tails.4 The problem is to determine when a verb occurs with cues for Soften enough that all those occurrencesare unlikely to be miscues.

4 Given a verb V , the outcomes of the coins for different S' s are treated as approximately independent, even though they cannot be perfectly independent. Their dependencecould be modeled using a multinomial rather than a binomial model.

M . R. Brent I Surface cuesand robust inference

453

3.3.1. Binomialfrequency data The statistical model is basedon the following approximation : for fixed S, all - S verbs have the sameprobability of being followed by a cue for S. Let 1l'- s stand for that probability . 1l'- s may vary from frame to frame, but not from verb to verb. Thus, miscuesmight be more common for tensedclauses than for NPs, but the working hypothesis is that all in transitives, such as .\"aunter and arrive, are about equally likely to be followed by a cue for an NP argument. If the miscue rate were known then we could use the standard hypothesis testing method for inference from binomial frequency data. For example, suppose1l'- s = 0.05 - on average, one in twenty occurrencesof aS verb should be followed by a cue for S. If some verb V occurs 300 times in the corpus, and 30 of those occurrencesare followed by cues for S, then V is unlikely to have probability 0.05 of being followed by a cue for S, and henceV is unlikely to be - SoSpecifically, the chance of flipping 30 or more heads out of 300 tossesof a coin with a five percent chance of coming up headseach time is less than three in 10,000. On the other hand, it is not all that unusual to flip 3 or more heads out of 30 on such a coin - it happens almost one time in five. If a verb occurs 30 times in the corpus and 3 of those occurrencesare followed by cues for S, it is quite possible that V is - S and that the occurrenceswith cues for S are explained by the five percent miscue rate of - S verbs. In order to use the hypothesistesting method we need to estimateils , the probability that an occurrenceof a - S verb V will be followed by a cue for S. For now, assumethat ils is known - a method for estimating 1l'_" is described below. Also assumethat, for each +S verb, V , the probability that V will be followed by a cue for S is greaterthan il _" . Other than that, no assumptionsare made about the probability that a +S verb will be followed by a cue for S. For , such as cut and walk, may have quite example, two verbs with transitive senses different frequenciesof co-occurrencewith cuesfor NP. It does not matter what thesefrequenciesare as long as they are greaterthan il - NP' If a coin with probability p of flipping heads is flipped n times, the probability of its coming up heads exactly m times is given by the binomial distribution : P (m,nip) = The probability sum :

n '. pm(lp m! (n - m) !

) n- m

of coming up heads m or more times is given by the obvious

454

M .R. Brent / Surface cuesand robust inference

n P(m+ ,nip) = LP (i ,nip) ism Analogously, P(m + ,nix _s) gives the probability that m or more occurrences of a - S verb V will be followed by a cue for S out of n occurrencestotal . If m out of n occurrences of V are followed by cues for S, and if P(m+ ,nix _s) is quite small, then it is unlikely that V is - So That is, the observeddata would be quite unlikely if V were - S and hencehad probability X- s of being followed by a cue for S. Traditionally , a threshold less than or equal to 0.05 is set, such that a hypothesis is rejected if , assuming the hypothesis were true, the probability of outcomes as extreme as the observed outcome would be below the threshold. The confidence attached to this . conclusion increasesas the threshold decreases 3.4. Estimating the miscue rate

As before, assumethat an occurrenceof a - S verb is followed by a cue for S with probability 1t'- S' Also as before, assumethat for each +S verb V , the probability that an occurrenceof V is followed by a cue for S is greater than 1t'- s' It .is useful to think of the verbs in the corpus as analogous to a large bag of coins with various blases, or probabilities of coming up heads. The only assumption about the distribution of blasesis that there is some definite but unknown minimum bias 1t'- s.s Determining whether a verb appears in frame S is analogous to determining, for some randomly selectedcoin, whether its bias is greater than 1t'- s' The only available evidencecomes from selecting a number of coins at random and flipping them. The previous section showed how this determination can be made given an estimate of 1t'- s' Suppose a series of coins is drawn at random and flipped N times. Each coin is assignedto a histogram bin representingthe number of times it comes up heads. At the end of this sampling procedure bin i contains the number of coins that came up headsexactly i times out of N. Such a histogram is shown in figure I , where N = 40. If N is large enough and enough coins are flipped N times, one would expect the following : ( I ) The coins whose probability of turning up heads is 1l'- s ( the minimum ) should cluster at the low - heads end of the histogram . That is , there 5 If the numberof coinsis takento be infinite thenthe blasesmustbe boundedaboveil -s'

Numbe of Coins 08Numb 16 24 32 40 of Head Flipp M . R. Brent / Surface cuesand robust inference

455

Fig. I. A histogramillustratinga binomiallyshapeddistributionin the first 8 bins.

should be some 0 :$ jo :$ N such that most of the coins that turn up jo heads or fewer have probability ils , and, conversely, most coins with probability ils turn upjo heads or fewer. (2) Supposejo were ' known. Then the portion of the histogram belowjo should havea roughly binomial shape. In figure I , for example, the first 8 bins have roughly the shapeone would expectif jo were 8. By contrast, the first 16 bins do not have the shapeone would expectif jo were 16- their height drops to zero for two stretches before rising signifi"cantlyabove zero again. Specifically , the shapeshould be roughly P( i,Nips ), wherei is the histogrambin, N is the fixed samplesize, and Psis an estimateof ils . (3) Supposeagain that jo were known. Then the averagerate at which the coins in binsio or lower flip headsis a good estimateof 7Lc. The estimation procedure tries out each bin as a possibleestimate of jo , the point of separation between the - S verbs at the low frequency end and the +S verbs at the high frequency end (Item 1) . Each estimate of jo leads to an estimate of ils (Item 3), and hence to an expected shape for the first jo histogram bins ( Item 2). Each estimatej of jo is evaluated by comparing the predicted distribution in the first j bins to the observed distribution - the better the fit , the better the estimate. The actual situation with verbs is slightly more complex than the one outlined above. The total number of occurrencesof each verb in any given input varies widely - look may occur thousands of times in a corpus where jostle occurs only once. Thus, the rates at which - S verbs co-occur with cues for S are not distributed according to a single binomial curve with sample-size N and mean Nx ils .6 Rather, they are distributed accordingto a superposition 6 In the context of machine learning, Brent ( 1993), presents an inference procedure that equalizesthe sample size by going through the input corpus twice, once for estimating it - . , and

456

M . R. Brent ! Surface cuesand robust inference

of binomial curves with different sample sizes N ( V) and means N( V) x X- S' This affects the details of the method but the underlying idea is unchanged: evaluate hypotheses about the :i :S boundary by comparing the observed distribution of - S verbs to the expected distribution for each hypothesized boundary. A formal specification of the estimation procedure is given in the Appendix . Verbs with various numbers of occurrencesin the input can be used to estimate the miscue rates, but there must be some minimum number of occurrencesbelow which verbs are not included in the estimation procedure. If verbs that occur very few times were included then there would be large sampling error and hence the +S verbs and the S verbs might overlap in However their rate of co occurrencewith cuesfor S. , there is no way to know the minimum sample size needed to keep the overlap betweenS and - S verbs acceptably low. On the other hand, the number of observations of each verb increases with time, so the learner can afford to keep raising its minimum . As this minimum is raised the degreeof overlap betweenS andS ' verbs will continue to decreaseand the leamer s conclusions will become increasingly stable and reliable. The learner will not know how stable or reliable its conclusions are at anyone time, but there is no reason to expect that child learners know or care. ' ' Note that the child s inferenceproblem is quite different from the scientists. Scientists need to know whether or not the sample is big enough that the experiment can be stopped and the results published. Children learning language do not face any analogous choice they never stop collecting data and they may draw provisional conclusions as necesary, even when the likelihood of error is not known. This continual raising of the minimum ' samplesize resemblesGold s identification in the limit (Gold 1967), where the learner always convergeson the right grammar but has no way to know when it has done so.

4. Experiment This experiment investigateswhether the implementation described in the previous section, and by extension the hypothesis of surface functional cues once for drawing lexical conclusions. On the first pass, a verb is ignored after some preset sample size has been reached. Multiple passesare not possible for a child leamer, so a new variant of the procedure was developed for the current problem.

M.R. Brent/ Surfacecuesandrobustinference

457

plus statistical inference, constitutes an effective strategy by which two -yearolds could learn subcategorization frames. The method is to simulate the proposed implementation on a computer. The input to the simulation is transcripts of English speechby mothers to their young children. 4.1. Methods

The input corpus for the experiment consists of 31,782 utterancesby adult caretakers to children between 1;0 and 2;6. The utterances were taken from the CHILDES database(MacWhinney 1991). The particular trtanscripts were as follows : all of Bates's ' free play' and ' snack' transcripts, but not the reading aloud transcripts (Bates et al. 1988); all of the Bernstein- Ratner transcripts ( Bernstein- Ratner 1987) ; all of the Higginson transcripts for children betweenages 1;0 and 2;6 (Higginson 1985) ; and the Warren-Leubecker (Warren- Leubecker and Bohannon 1984) transcripts for children 2;6 and younger. In all cases, exactly one adult and one child were present throughout . All lines except the adult ' s speechwere removed. The pause markers ( # ) were left in and taken as utterance boundaries, along with the ordinary punctuation marks. All diacritics between square brackets were removed. Angle brackets (used as scope markers for diacritics) were removed but the words between them were not. A few changeswere made to enforce consistent ' use of the compound marker ' + ' , including the placement of a ' + ' rather than a space between two-part names like Santa Claus and Mickey Mouse. Otherwise, transcription errors were not corrected.

. The observations were collectedexactly as described Collectingobservations in section3. Statistical inference. The algorithm given in the Appendix was used with I = 40 and minsample = 40. Given the estimated error rate for each syntactic frame, verbs are reported as + S if their rate of co-occurrencewith cuesfor S would have probability below 0.02 under the null ( - S) hypothesis. Vocabu/ary . I n order to make use of the cues, the learner needs to know some function morphemes. In this experiment the program made use of an initial vocabulary chosen from the most common words in the corpus '' ' ' ' ' '' (table 4) . Besidesthe four utterance-boundary markers, . , 1 , # , and ! , these include the pronouns you, that, it , this, I , we, he. and they; the determiners the, a, and your ; the complementizerswhat, that, and where; and

458

M . R. Brent / Surface cuesand robust inference

Table 4

The SOmostcommonorthographicforms in the corpus

the prepositions to, in and on. Auxiliaries and some modals are also common, and these can be exploited by cues. However, one of the goals of this experiment was to determine how little lexical knowledge the child could make do with , so these function morphemeswere not exploited. The demonstrative ' ' pro -PPs here and there were not exploited either, though they would probably be very useful for learning about locative PPs. Interjections, negations , and conjunctions are of little use. The fifty most frequent words also include the open classverbs put , look, want, and like, but the simulation does not know any open class words at the outset. In addition , the simulation started with a lexicon of all 158 proper names that occurred at least twice in the corpus.

4.2. Results This simulation identified a total 126 verbs, of which 76 were assignedat least one syntactic frame. The output of the simulation for those 76 verbs is shown in table 5, while the 50 verbs that were not assignedany frame are listed in table 6. Each row of table 5 representsa single verb. The symbols ' in appearing in each row representthe frames assignedto that row s verb the to stand and in on simulation . The symbol N P stands for an NP complement, , for PPsheadedby those prepositions, wh stands for a tensedwh clause, c/ for a tensed clause with that or a null complementizer, and if for a tensed if clause. When one of these symbols follows NP , it signifies a two -argument frame in which the first argument is realized as an NP. For easy referenceby frame, all the symbols for a given frame are aligned in one column.

~I0

. ~c~ ~] ~t~~

Table5

NPin

NPon NPto

NPwh cl

in

inf

NPwh

ZQZQ

ZZZZ Q ZZZZZZZZZZZZZZ Q ZZZZ Q

NPNP

"

z ~0c

ZZQ

ZZZ Q

NPNP

.c

z&0 z!0

NPNP

zzzz ~

:(~ ~

NP

z~0

Z Q Z Q ~~~~-~~ l

The 76 verbs that were assignedat least one syntactic frame

on

to

wh

M . R. Brent I Surface cuesand robust inference

460

Table5 (Coot .) NP pull

NP

push

NP

put

NP

read

NP

NPNP

NPin

NPon NPto

NPin

NPon

NPwh cl

in

inf

on

to

NPNP

record

wh in

ride

NP

rock

NP

roll

NP

say

NP

set

NP

sew

NP

show

NP

NPNP

sing

NP

NPNP

NPto

NPto

in

sit

on on

sleep smell

NP

spill

NP

splash

NP

in inf

start take

NP on

talk

.

cl

throw

NP

touch

NP

NPto

inf

trot try

NP

turn

NP

understand

NP

wait

NP

inf

wh

in

walk wash

NP

watch

NP

on

wh on

water wear

to

NPwh

NP

tell think

wh

NP

Table6 The 50 words that were identified as verbs but were not assignedany syntactic frames add button clean color come cook dancefraw evenfall fish fly forget grow hide inventjump keep kick knock laugh lay lie move mow nap pick pitch pour rain ring rub run scratch share shavesnow stack stand star step stick swim swing umm wake wavework write zip

M . R. Brent / Surface cuesand robust inference

These results should be interpreted in light of the fact that the statistical ' or inference technique does not answer 4yes 4no' as to whether a given verb ' or 4noreliable indication of has a given frame. Rather, it answers4yes yes so ' far . It should also be noted that a number of the verbs discovered have auxiliary or modal functions, including be, do, go and have. These are generally omitted from the following discussion, since they have special properties that must in any case be learned by some process that does not directly involve subcategorization frames. Now consider the results for each syntactic frame. Direct object ( N P) . 59 verbs of the 126 verbs were deemed acceptable with a direct object and no other complement (NP ). Most of these are correct. Most common transitive verbs were assignedNP , while most intran sitives, including come, fall , laugh, listen, lie, live, look, nap, pretend, rain, run, .vit, sleep, snow, step, talk and trot , were not. Many other words that are intransitive except in specializedsenses(swim the channel, work the metal) are not assignedNP either. It may be that these sensesare rare in child -directed speech. The only error among the verbs assigned NP is put , which requires a location phrase (typically a PP) . This is caused by extraction in where questions. Cry is assignedNP for the wrong reasons: the cues do not detect " " . inversion ( Oh no, cried Pinocchio) ; and the transcripts are inconsistent ' about pause marks before names used in address( The bug // cry June) . Double-object ( NPNP ) and the dative alternation. Ask, buy, do, give, read, show and sing are deemed acceptable with two NP complements (NPNP ). These are all correct. The other half of the dative alternation is the NPto frame, which is correctly assigned to give, show, bring, roll , and throw. Thus, the dative alternation is observedfor give and show. Bring, feed, roll , throw and read can alternate, but they are assignedonly one of the two frames. Infinitive VP (in/) . Feel, get, go, have, pretend, start , trot , and try are all assignedthe inf frame. Of these, feel is clearly incorrect, resulting from six ' repetitions of the sentenceYou don t know how good it feels to washmy ears and scrub my heels. In fact, feel requires an adjective as well as an infinitive , but the adjective has been fronted with how in this sentence. The inclusion of trot also results from repetition of an unusual nursery-rhyme construction as off we trot to play . The correct analysis of this utterance is not clear. The remaining verbs appear to license an infinitive complement, although in

462

M . R. Brent I Surface cuesand robust inference

semanticallydiversecapacities. In particular, go, have, and possibly get license an infinitive in their capacity as auxiliaries or modals. Tensedclauses( NPwh, wh, cl) . TeJ/ and ask are both deemedacceptable with both a direct object and a clauseheadedby a wh word , as in Ask Daddy what he wantsfor lunch. As expected, theseare both communication verbs. Find, look, record, understand and watch are deemed acceptable with a clauseheadedby a wh word as their sole argument, as in Look what I 've got ! Look is interesting becauseit is in fact acceptablewith a wh clause but not with a direct object nor with a that clause. Do, make, pretendand think are judged to take a clauseheadedby that or a clausewith no overt complementizer(cl). Pretendand think are canonicalmental state verbs. The relative poverty of such verbs in the output is discussedbelow. Make is mistaken as taking a tensedclausebecauseit takes an NP and a bare infinitive, as in I can't make John eat his peas. Becausethe cues are check , John eat ... is mistaken for a tensed clause. In many casesthis agreement construction could be distinguishedfrom a tensedclause, using either pronoun caseor subject-verb agreement . However, such knowledgewas not provided in the cuesfor this simulation. Finally, do is judged to take a tensedclausebecause inverted questions like Do you know how to swim? are mistaken for tensed clauses . NPin , NPon. The NPto frame was discussedunder the dative alternation. Put .was assignedNPon and NPin , while leavewas assignedNPon. These are correct, though leaveis acceptablewith NPin as well. in , on, to. Go, listen and talk were correctly assignedthe to frame. For on and in it is difficult to determine which cases represent subcategorized argumentsand which representadjuncts. Some usesof on and in clearly mark participants in the action, as in drive in the car and talk on the phone. Others clearly representlocational adjuncts, as in sleepin the baby bed and splash in the pool. A number of exampleswere difficult to classify. 4.3. Discussion

Overall, this experiment suggeststhat syntactic frames can be identified in child-directedEnglish using relatively simplecuesbasedon function morphemes, proper nouns, and utterance boundaries, in combination with statistical inference. However, the cues suggestedin section 3 are slightly too simple to achievehigh accuracy. Specifically, they do not distinguishbetweeninterrogative . This led to the erroneous assignmentof NP to put, and declarative sentences inf to / eel, and cl to do (although do is independentlymarked as idiosyncratic).

M . R. Brent / Surface cuesand robust inference

Since the declarative/interrogative distinction is clearly marked prosodically, children probably have accessto it . One potential problem is suggestedby the fact that locational adjuncts were sometimesmistaken for arguments in the simulation. However, it is not clear whether anyone can distinguish the two cases reliably in naturally . Until the nature of the argument/adjunct distinction is occurring sentences better understood, it is difficult to seehow the extent of the problem can be assessedand cues for resolving it investigated. In contrast to the generally positive results for identifying complements, this simulation failed to demonstrate that simple cues based on function morphemes can identify all verbs that occur frequently in child-directed English. A number of high-frequency mental state verbs, notably know, want, see, hear, like and love, are missing from tables 5 and 6. Thesewords were not detected becausethey do not occur in the input corpus with the suffixing . Mental state verbsdo not generallyoccur in the progressiveaspectand their nominalizations (e.g. knowing, liking) appear to be very in child-directed speech. The transcriptsusedin this experimentcompriseabout 56 hours of interaction, so it is possible that more text would reveal that nominalizations of mental state verbs are not intolerably rare. Further , for most of the transcripts the child was under 2;0, and it seemslikely that the use of semantically abstract words increaseswith the child ' s age. Nonetheless, the -ing suffix seems a ' precarious basis on which to identify mental state verbs - children s true strategiesfor identifying verbs are doubtless more complex. One possibility is that they use more complex function -morpheme cues that exploit other inflections, auxiliaries, modals, and pronoun case ( Brent 1991) . Since these cues are not without exceptions, it would make sense to posit statistical inferencemechanismfor lexical category cuesas well as for subcategorization cues. In fact, such an inference system would be useful even in the current simulation, where water and even are both erroneously recorded as having verbal occurrences. ( Both have verbal senses , but they do not occur as verbs in the input corpus.) Another possibility is that children understandutterances well enough to recognizethat certain words stand for actions or states, and that they have an innate predispositionto classifytheseas verbs (Grimshaw 1981). The importance of the statistical inference component can be seenclearly by comparing .its input , the raw observation shown in table 2, with its output , the lexicon shown in table 5. The segmentof the observation table shown in table 2 contains a number of caseswhere a verb co-occurs with cues for a frame that the verb does not in fact have in its lexical entry . These include comeand look with a cue for NP , color, come and get with cues for a tensed

464

M . R. Brent I Surface cuesand robust inference

clause, and come with a cue for an infinitive . As a result of the inference procedure, none of these miscues lead to errors in the lexicon shown in table 5.

5. Summary and conclusions

This investigation began with the question, How could children possibly acquire their first subcategorization frames? Knowing verb meanings does not help until the language-particular correspondencebetween meaning and subcategorizationalternations can be learned. Direct observation of syntactic structures might work , except that it is difficult to imagine how children could parse an utterance reliably without already knowing the syntactic properties Qf the words in it . However, function morphemes are among the most frequent morphemes, and they would seem to provide a good deal of probabilistic information about syntactic structure. Further , English speaking of the to have children appear syntactic privileges of some passiveknowledge -half. The question is how and a individual function morphemes by age two useful the information carried by function morphemes is and how it can be exploited to learn subcategorization frames reliably . To answer that question, a collection of simple cues was devised for English, using a vocabulary of twenty frequent function morphemes, under the assumption that children recognize the common proper names in their environment as NPs and can detect utterance boundaries prosodically . Becausethe information provided by these cues is probabilistic , inferences must be drawn from multiple occurrencesof each verb. Thus, a statistical inference procedure was developed. The cues and the inference procedure were simulated on transcripts of child -directed speech. This experiment showed that probabilistic surface cuescombined with statistical inference can determine syntactic frames with surprisingly high accuracy. Thus, young children could learn subcategorizationframes from sentenceswhosemeanings they do not fully understand. There is no doubt that two- year olds comprehend in some cases. However, many utterances, and this probably helps them parse the current experimentssuggestthat they can also learn from utterancesthey do not understand and cannot parse completely. However, some important questions remain open. Specifically, the hypothesis that young children can use their knowledge of function morphemesas cues for specific phrasal categories remains unproven. Even if they can, one would still want an explanation of how English-speaking children come to

M.R. BrentI Surface cuesandrobustinference

465

know that , for example, to followed by an uninflected verb is usually an infinitive , while I followed by an uninflected verb is not. In the simulation it was assumedthat the learner knows a priori , perhaps even innately, which sequencesof phrasesconstitute possible syntactic frames. To make use of this knowledge, it is not enough to know that , for instance, to followed by an uninflected verb is an instance of some linguistically significant category; rather, it is essentialto know that it is an instanceof the category represented ' . by infinitive in the a priori list of frames. An alternative possibility is that , at least at this early stage, children are simply forming distributional classesof verbs that reflect their ability to co-occur with various sequencesof function morphemes. Thus, English-speakingchildren might not know that to followed by an uninflected verb is an infinitive , per se. Rather, they might know only that co-occurrence with this sequenceis a significant distributional property of verbs, and that it is a different one from co-occurrencewith I followed by an uninflected verb. To put this question in context, however, note that English-speaking children do eventually come to know the relationship between sequencesof function morphemes and syntactic structure. This ' phenomenon needs explanation regardless of one s theory about the early acquisition of subcategorization frames, so the current proposals incur no extra explanatory burden. Finally , note that statistical inferencemechanismsof the sort proposed here have an important role to play in language acquisition research in general. Developing such mechanismsis challenging becausethere is often no .a priori basis for predicting frequency distributions in linguistic domains. This leaves simulation as one of the only tools for determining when an inference mechanism is adequate to a particular language learning problem. More experimentswith inferenceproceduresare neededto help explain the robustness of language acquisition, and to establish that individual learning strategies can be implemented robustly .

Appendix : Error rate estimation

This appendixlays out the procedureusedfor estimatingthe error ratesXI in the experiment. This presentationis designedfor clarity rather than efficiency. Let N( V) be the total number of occurrencesof verb V, and letf ( ViS) be the number of times V co-occurs with cues for S. The relative frequency of V with cuesfor S is the ratiof ( Vis) fN ( V) , a number betweenzero and one. The observed distribution of relative frequenciescan be plotted on a histogram

466

cuesandrobustinference MR . Brent! Surface

whose bins correspond to equal-sized subintervals of the unit interval (e.g. figure I , page 455) . By assumption, there is some bin jo such that verbs in bins jo and lower are - S while those in bins abovejo are +S. The estimated miscue rate is the weighted averageof the relative frequenciesof - S verbs by assumption, the verbs with relative frequency below jo . The procedure ESTIMATE- MISCUE- RATE estimatesthe miscue rate for a frame S by first estimating the :I: S separator, jo , for S. The inputs to the - the list of observed verbs, N - a procedure are S - the frame, verbs function returning the total frequency for each verb, f - a function returning the total frequency of each verb with cues for each frame, bin - count - the number of histogram bins to use in plotting the relative frequencies, and min - sample - the minimum number of times a verb must occur to be used in this estimation. Each bin is evaluated as a possible value for jo , and the one with tl1e highest evaluation is selected. The variable cutoff - bin stores hypothesesabout jo while they are being evaluated. The estimation procedure involves only the - S verbs - by hypothesis, those with relative frequency below cutoff - bin . Thus, the variables total - verb - ocC':urrences and refer to the total counts for verbs with relative total - cue - occurrences bin . Each hypothesis about jo (.stored in frequency below cutoff bin ) generatesa hypothesis about the miscue rate ils , which is cutoff stored in the variable rate . In particular , the miscue rate is hypothesizedto to total - verb - occurren be the ratio of total - cue - occurrences ces , where, again, theserefer only to the - S verbs. This hypothesis predicts the shape of the distribution of relative frequenciesfor - S verbs. In particular , it predicts that the distribution should be a mixture of binomial distributions with a common mean, the miscue rate, but different different sample sizes, corresponding to the number of times each verb occurs in the input . The procedure evaluates each hypothesis about jo (and the consequent hypothesisabout the miscue rate) by comparing the predicted distribution forS verbs to the observeddistribution . The comparison is done by summing the squares of the differences between the observed and predicted distributions of - S verbs at each bin. At bins above cutoff - bin the observed distribution of - S verbs is taken to be uniformly zero. ESTl MATE- MISCUE- RATE( S, verbs , N, f , bin - count , min - sample ) total - verb - occurrences = O, total - cue - occurrences = O best - sum- of - squares = infinity For cutoff - bin = 1 to bin - count observed - dist [ cutoff - bin ] = 0

M . R. Brent I Surface cuesand robust inference Set up the Oth bin of the histogram

of observed co - occurrence

467 rates

For Veverbs

f (ViS ( bin - count * N(V) If ( verb - bin = 0 ) And ( N ( V ) ~ min - sample ) observed - dist [ O] = observed - dist [ O] + 1 total - cue - occurrences = total - cue - occurrences + f ( ViS ) total - verb - occurrences = total - verb - occurrences + N ( V )

verb - bin = roundoff

Try each bin as a hypothesis about the ~ S cutoff For cutoff

- bin

= 1 to bin - count

Add in verbs that were + 8 under the previous hypothesis about the ::i:Scutoff, but are - 8 under the new hypothesis For Veverbs f (ViS N( V ) - bin ) And ( N ( V ) ~ min - sample ) If ( verb - bin = cutoff observed - dist [ verb - bin ] = observed - dist [ verb - bin ] + 1 = total - cue - occurrences total - cue - occurrences + f ( V, S ) = total - verb - occurrences total verb - occurrences + N(V )

verb - bin

= roundoff

( bin - count

Estimate error rate assuming cutoff

- bin

*

is :i:S separator

raet = total - cue -occurrences total verb occurrences - s = rate Compute expected distribution over - S verbs assuming 1C For Veverbs

f (VtS ( bin - count * N(V) ( verb - bin ~ cutoff - bin ) And ( N ( V ) ~ min - sample )

verb - bin = roundoff If

The expected distribution is a mixture of binomials with a common mean, rate , but different sample sizes, N ( V) . Compute probability

M . R. Brent I Surface cuesand robust inference

distribution for verb V, scale it to bin - count bins, and add it to the mixture. For orig - bin = O toN ( V ) Scale orig

- bin

to the mixture histogram .

- bin t * Orig ) N(V) Add the binomial distribution for rate , and N ( V ) to the mixture expected - dist [ bin ] = expected - dist [ bin ] + - bin N(V) * rateorig l - rate ) ( N ( V ) orig bin ) ( ( ) orig - bin Compareobservedand expecteddistributions, assumingcutoff - bin is the :i:Sseparator sum- or - squares = 0 For bin :: 0 to bin - count sum- of - squares = sum- of - squares + ( expected - dist [ bin ] observed - dist [ bin ] ) 2 If sum- or - squares < best - sum- of - squares best - rate = rate , best - sum- of - squares = sum- or - squares Return best - rate . b in = roun d 0 ff

. ( b incoun

References of the 2nd Annual . In: Proceedings . Testingfor adjuncts Adams, L. and T. Macfarland, 1991 . Formal LinguisticsSocietyof Meeting of the Formal LinguisticsSocietyof Midamerica . Midamerica . From first wordsto grammar: Individualdifferences Bates , 1988 , E., I. Brethertonand L. Snyder . . Cambridge mechanisms and dissociable , MA : CambridgeUniversityPress -Ratner, N., 1987 . In: K. Nelson, A. vanKleeck . The phonologyof parentchild speech Bernstein ' , Vol. 6. Hillsdale, NJ: Erlbaum. (eds.), Childrens language . In: N. . Thedemonsand the beast- Modularand nodularkindsof knowledge T.G. 1991 Bever, , 212 252 . Hillsdale to es . , N. eds Connectionist , processing , language ), approach ( Sharkey Reilly NJ: Erlbaum. : Utilising the . Largelexiconsfor naturallanguageprocessing , 1987 , B. and T. Briscoe Boguraev grammarcodingsystemof Idoce. ComputationalLinguistics13(3), 203- 218. framesfrom unrestrictedEnglish. . Automaticacquisitionofsubcategorization Brent, M.R., 1991 . of Institute Massachusetts Ph.D. thesis , Technology . Fromgrammarto lexicon: Unsupervised Brent, M. R., 1993 learningof lexicalsyntax. Computational Linguistics19, 243 262. . . Aspectsof the theoryof syntax. Cambridge , MA : MIT Press , N., 1965 Chomsky

M . R. Brent ! Surface cuesand robust inference

469

. Lectureson government . N.. 1981 and binding. the Pisalectures Chomsky , Volume9 of Studies in generative grammar. Dordrecht: Foris. . Knowledgeof language . N., 1986 . Convergence . Westport, cT : Praeger . Chomsky Demuth. K.. 1993 . Issuesin the the acquisitionof the sesothotonal system . Journalof Child Language20. 275- 301. Fernald. A. and P. Kuhl, 1987 . Accousticdeterminants of infant preference of motheresespeech . Infant Behaviorand Development10, 279- 293. Fisher. C.. H. Gleitmanand L.R. Gleitman, 1991 . On the semanticcontentofsubcategorization frames . Journalof CognitivePsychology 23(3), 331- 392. Fisher. C. and H. Tokura, 1993 . Acousticcuesto clauseboundariesin speechto infants: Cross. Manuscript , Departmentof Psychology linguisticevidence , Universityof Illinois. Gelman. S.A. and M. Taylor, 1984 . How two-year-old childreninterpretproper and common namesfor unfamiliarobjects . Child Development 55, 1535 - 1540 . Gerken. L.. 1992 . Young children's representationof prosodic phonology: Evidencefrom ' s weak -speaker . Departmentof Psychology English syllableproductions , SUNY Buffalo. Gerken. L.. B. Landauand R.E. Remez . Functionmorphemes in youngchildren's speech , 1990 perceptionand production. Developmental 26( 2), 204- 216. Psychology Gerken. L. and B.J. McIntosh, 1993 . The interplayof functionmorphemes and prosodyin early . Developmental 29, 448- 457. language Psychology Gleitman. L.. 1990. The structural sourcesof verb meanings . LanguageAcquisition 1( 1), 356. Gleitman, L.R., H. Gleitmanand E. Wanner, 1988 . Wherelearningbegins: Initial representations for languagelearning, 3- 48. New York: CambridgeUniversityPress . Gold. E.M.. 1967 . Languageidentificationin the limit. Informationand Control 10, 447- 474. Grimshaw . Form, function, and the languageacquisitiondevice, 165-210. Cambridge . J., 1981 , MA : MIT Press . . Affectedness anddirectobjects: The Gropen. J.. S. Pinker. M. Hollanderand R. Goldberg, 1991 role of lexicalsemanticsin the acquisitionof verb argumentstructure . Cognition41, 153195.Higginson . Fixing-assimilationin languageacquisition . RiP.. 1985 . PhiD. thesis , Washington StateUniversity. Hirsh-Pasek . K., D.G. Kemler-Nelson, P.W. Jusczyk . Clausesare , K. Write and B. Druss, 1987 perceptualunits for younginfants. Cognition26, 269- 86. . P.W.. K. Hirsh-Pasek , DiG. Kemler-Nelson, L. Kennedy Jusczyk , A. Woodwardand J. Piwez, 1993 . Perceptionof accousticcorrelatesof major phrasalunits by young infants. Cognitive (in press ). Psychology . Perceptionof a phoneticcontrast in multisyllabic . P.W. and E. Thompson , 1978 Jusczyk utterances 23, 105- 109. by 2-month-old infants. Perceptionand Psychophysics Katz. N.. E. Bakerand J. MacNamara . What' s in a name? A studyof how childrenlearn , 1974 commonand propernames . Child Development 45, 469- 73. Kemler-NelsonD.G., K. Hirsh-Pasek . How theprosodic , P.W. Jusczykand K.W. Cassidy , 1989 cuesin motheresemight assistlanguagelearning. Journalof Child Language16, 55- 68. Kimball. J., 1993 . Sevenprinciplesof surfaceparsingin naturallanguage . Cognition2( 1), 15- 47. Landau, B. and L. Gleitman, 1985 . Languageand experience . Cambridge , MA : Harvard . UniversityPress Lederer . Prosodicinformation for syntacticstructurein parental . A. and M.H. Kelly, 1992 . Manuscript . , Departmentof Psychology speech , Universityof Pennsylvania

M . R. Brent I Surface cuesand robust inference

. Chicago : A preliminaryinvestigation and alternations , IL : . Englishverbclasses Levin, B., 1993 . Press of University Chicago . . Cambridge . How to setparameters , MA : MIT Press LightfootD .W., 1991 . TheCHILDES project: Toolsfor analyzingtalk. Hillsdale, NJ: Erlbaum. , B., 1991 MacWhinney .A Mehler, J., P.W. Jusczyk , G. Labertz, H. Halsted, J. Bertonciniand C. Amiel-Tison, 1988 precursorof languageacquisitionin younginfants. Cognition29, 143 178. , and Morgan, J., 1986. Drom simple input to complex grammar. Learning, development . : MIT Press MA . , conceptualchangeCambridge . Structuralpackagingin the input to language Morgan, J., R.P. Meierand EL . Newport, 1987 learning: Contributionsof prosodicand morphologicalmarkingof phrasesto the acquisition . CognitivePsychology19, 498- 550. of language . Prosodicdomainsin externalsandirules. In: H. vander Hulst, N. Nespor, M. and I. Vogel, 1982 , VolumeI. Dordrecht: Foris. Smith(eds.), The structureof phonologicalrepresentations I'd ratherdo it myself: Someeffects . Mother 1977 H. Gleitman and E. L. Gleitman , Newport, , and non-effectsof maternalspeechstyle. In: C.E. Snow, C.A. Ferguson(eds.), Talking to . children: Languageinput and acuisition, 109- 149. NewYork: CambridgeUniversityPress of ? Journal . Does 1976 and RiD. P.A. Petretic production , precede comprehension , Tweeney Child Language4, 201- 209. . Cambridge , MA : Harvard . Languagelearnabilityand languagedevelopment Pinker, S., 1984 . Press University , . Learnabilityand cognition: The acquisitionof argumentstructure. Cambridge Pinker, S., 1989 . MA : MIT Press of the PhilologicalSociety86, 1. Transactions . Smallchildren's smallclauses Radford, A., 1988 43. . The syntaxof nominalargumentsin earlychild English. LanguageAcquisition Radford, A., 1991 I , 195-223. . In: J. Laver, J. Anderson(eds.), . On the natureof phonologicalrepresentation Selkirk, E., 1981 : North-Holland. of speech Thecognitiverepresentation , 379- 388. Amsterdam : Free the . A in 1969 L.R. Gleitman Smith and E.F. C.S. , acquisitionof language study , , Shipley . Language45, 322- 342. to commands responses . Anchor points in languagelearning: The role of marker Valian, V. and S. Coulson, 1988 and of . Journal Language27, 71- 86. Memory frequency : . Intonation patternsin child-directedspeech Warren-Leubecker , 1984 , A. and J.N. Bohannon . - 1385 . Child Development55, 1379 Mother-fatherspeech . Cambridge , MA : . Formalprinciplesof languageacquisition Wexier, K. and P. Culicover, 1980 . MIT Press elementsin . The (non)realizationof unstressed Wilnen, F., E. Krikhaar and E. den Os, 1993 : A rhythmicconstraint.Journalof Child Language(in press ). children's utterances . The discriminationby younginfantsof voicedstopconsonants Williams, L. and M. Bush, 1978 . - 1225 with and without releasebursts. Journalof the AcousticalSocietyof America63, 1223 . 223 233 2 . of manner . In a , A. 1970 speakingLinguisticInquiry Zwicky, ,

-Holland 92(1994 - 480 . North ) 471 Lingua

Acquisition Mark

471

of verb categories *

Steedman

, Universityof Pennsylvania Departmentof Computerand Information Science , 200 South 33rdStreet, Philadelphia PA 19104 6389, USA , The paper was deliveredas a commentaryupon Michael Brent's presentation'Acquisition of subcategorizationframes using aggregatedevidencefrom local syntactic cues' to the PennLanguageAcquisitionConference . It arguesin , IRCS, Universityof Pennsylvania , 1992 ' of errorsand supportof usingstatisticaltechniqueslike Brents to minimisethe consequences that the casefor believingthat childrenacquiresubcategorisation , but concludes and misanalyses otheraspectsof syntaxon the basisof semanticand contextualcuesremainsstrong.

1. Introduction

The question of how children acquire lexical entries for verbs, and in particula~ their subcategorisation frames is one of the central questions ' concerning the child s acquisition of syntax. Its importance is enhanced by the recent tendency in theories of grammar to gravitate to a lexicalist position, and the role of verbs as the head of their clause. How do children do it , given the non-determinacy and automata-theoretic complexity of the syntax itself, and the unsystematic presentation and error-proneness of the ' linguistic data that they apparently have to make do with ? Michael Brent s paper in this volume shows how the statistical technique of binomial error estimation can be used to minimise the effect of contamination in the data available to the child languagelearner, arising either from errors in the input itself or errors in the child ' s analysesof the input sentences . The technique is demonstratedby applying it to the sentencesof a corpus of actual adult- child conversations, to derive subcategorisation frames for verbs from analyses basedon imperfectly reliable local syntactic cues defined in terms of sequences of inflectional morphemes, function words and lexical NPs. As Brent . Thanksto Lila GleitmanandJeffSiskindfor readingthedraft. Theresearch wassupportedin , and CISEliP , CDA88-22719 , DARPA grant part by NSF grantnos. IRI90- 18SI3, IRI90- 16S92 -90-J- 1863 no. NOOOI4 . , and ARO grant no. DAALO3-89-COO31 - 3841/94/ $07.00 ~ 1994- ElsevierScienceB.V. All 0024 SSDI 0024- 3841( 93) EO047- B

right~reserved

472

M . SteedmanI Acquisition of verb categories

points out , these two aspects of the work are quite independent : binomial error estimation could be used to minimise the influence of errors arising from imperfect analysis procedures of any kind at all , including those based on semantic and prosodic information , as well as syntactic . The present paper considers the part that all of these sources of information may play .

2. Syntax The specific application of this technique to low -level syntactic cues, rather than these richer sources of information , in this and related work in Brent ' s ( 1991) thesis can be argued to deliver two further important results. First , it demonstrates a practical technique that actually can be used to automatically build lexicons on the basis of large volumes of text . ' Although this point is not discussed in Brent s present paper, it is worth emphasising. Hand -built dictionaries are inevitably very incomplete with respect to the exhaustive listing of subcategorisation properties that are needed for many computational applications . Techniques base~ on sim' ' plified syntactic properties which probabilistically compile out syntactic' ' and semantic properties of what a linguist would regard as the grammar , and working on the basis of statistical properties of their distribution over a large corpus , may well represent the only practicable possibility for automatically extending such dictionaries . Full -blown deterministic parsing of the corpora of the requisite size using linguistically respectable grammars and/ or semantic interpretations and deterministic parsing , is impracticably expensive computationally , using existing techniques, to the extent that it is possible at all . Second, the present study demonstrates the important fact that the information needed to determine verb subcategorisationsactually is there in the distribution of these very low-level properties in input of the kind that children are actually exposedto. For example, one of the apparent problems for acquisition of subcategorisationframes on the basis of syntactic information alone is the systematicambiguity in all languagesbetweensubcategorised arguments and nonsubcategorised adjuncts, illustrated for English by the following pair of sentences:

( la) We put Harry on the bus ( 1b) We met Harry on the bus

M . Steedman/ Acquisition of verb categories

473

How can the child avoid erroneously subcategorising meet like put ? Brent ' points out that it doesn t matter if they do, because it is (presumably universally) the case that the relative frequency with which the sequence V N P PPOll occurs will be significantly higher for verbs like put that subcategorisefor NPs and on PPs than for those like meet which subcategorise for NP and only allow PP as an adjunct. Binomial error estimation is able to distinguish the two distributions , and reject the child ' s spurious evidence from analyses suggesting that meet subcategorises for the PP. Similar results seemto follow for spurious occurrencesof subcategorisations arising from extraction, as in who did you put on the bus, which might appear otherwise to suggestthat put might subcategorisefor PP alone. It is therefore reasonableto ask whether the child languagelearner actually makes use of such purely syntactic cues to learn the lexical categories of verbs. Here Brent is extremely cautious, and goesout of his way to acknowledge the possible involvement of prosodic and semanticcuesas well. He notes in passing that there are a number of open questions that need answering before we can be quite comfortable with the assumption that the child is using the closed-class cues. The most important is that both the function words themselvesand the cue sequencesbasedon them are language-specific. The question arisesof how the child can possibly catch on to the fact that it , that and the are cue words, much lessthat the sequenceV it the suggeststhat V is probably a ditransitive verb, while the sequenceV that the suggeststhat V is probably a complement verb. It is hard to see that there is any alternative to knowing , besidesthe set of possible subcategorisation frames, (a) the precisesyntactic significanceof each closed class word as NP , Specof CP, etc., and (b) some statistics about possible corpora, including facts such as that complement-taking verbs are more common than ditransitives. 1 I shall remain equally cautious in the face of such open questions, and certainly would not wish to claim that the child cannot be using such cues. However, as long as these questions remain open, it also remains unclear whether we have escapedwhat Brent identifies as the ' chicken-or -egg problem' of apparently needing to know some syntax to apply this procedure. This suggeststhat there may be some point to asking ourselves what other resources the child could call upon, and in particular whether the two alternatives that Brent mentions, prosody and semantics, can help a child 1 It is not enough to assumethat the child simply looks out for verbs followed by all possible ' ' ' ' of sequences cue words, classifying verbs as ;t + the verbs , that + the verbs , etc. Such a classification does not determine a subcategorisation.

474

M . SteedmanI Acquisition of verb categories

learn the first elements of syntax, including their first subcategorisations, in the face of the kind of uncertainties in the input which he identifies.

3. Prosody Although Brent shows how misanalysesarising from the argument/adjunct ambiguity can be overcome, the consequencesof some other quite similar sorts of ambiguity , such as that betweenprepositions and particles illustrated below, are not so easily eliminable by distribution -basedmethods, since verbs 2 subcategorisefor ambiguous items like up in both its guises:

(2a) Werang up the hotel (2b) Weran up the hill In the case of particles and prepositions it seemsintuitively highly likely that prosody disambiguates the two. Lederer and Kelly ( 1991) have shown that adults can reliably identify which of the two a speakerhas uttered. They have shown similar effects for the argument/adjunct alternation. Kelly ( 1992 and this volume) presents results which suggest that a number of further apparent ambiguities are also correlated with prosodic distinctions. It is true that none of theseprosodic discriminators are invariably present. Nor does it seemat all likely that all the relevant ambiguities are marked in this way. For example, I know of no evidence that the V PP sequence arising from extraction in (a), below, differs in any prosodic repect from that in (b) : (3a) Who did you put on the bus (3b) Who did you run up the hill with However, where the information is marked, it may well be reliable enough to be used as evidenceunder appropriate distribution -based techniques such as ' Brent' s own, especially when we recall that adult s speech to children is characterisedby exaggeration of normal intonation contours. Z Brent suggeststhat spurious analyses of verbs like ring as subcategorising for PP can be eliminated by observing sets of subcategorisations, presumably meaning that we can reclassify verbs that have been assigned subcategorisationsof both PP and NP + P. However, this is a distinct (and language-specific) complication to the proposal, and appears likely to conflict with the other usesthat have been proposed for such sets. The prosodic cues discussedbelow would allow this particular complication to be eliminated from his account.

M . Steedman/ Acquisition o/ verb categories

475

However, a word of caution is in order here. Adult speakers do not actually use intonation to indicate syntactic structure, but to convey the distinctions of discourse meaning that are variously described in tenDS of ' focus' or of ' ' ' ' , oppositions such as topic/ comment , given/ new and the like. While the elements that are marked in this way correlate with syntactic structure, this is for semantic reasons, rather than for ease of processing. When adults exaggerate intonation contours in speaking to children, it is extremely unlikely that they are using the intonational markers in any very different way. It is therefore quite possible that children use this infonnation as a semantic, rather than syntactic, cue, as part of the third strategy under consideration here.

4. Semantics

As soon as it was appreciated that even quite trivial classesof grammar cannot be learned by mere exposureto their stringsets, and that there appears to be little evidencethat any more explicit guidance is provided by adults, it was obvious that some other source of infonnation , ' innate' in the sensethat it is available to the child prelinguistically, must guide them in acquiring their grammar. As has often beenpointed out , the only likely candidate is semantic 3 interpretation or the related conceptual representation. However inadequate ' o~r formal (and even infonnal ) grasp on the child s prelinguistic conceptualisation of the conversational situation , there can be no doubt that it has one, for even non-linguistic animals have that much. There can therefore be no doubt that this cognitive apparatus, for reasons which have nothing to do with languageas such, partitions the world into functionally relevant ' natural kinds' of the kind investigated by Landau in this volume, individual entities, including events, propositions, and such grammatically relevant notions as actual and potential participants and properties of those events, as well as the attitudes and attentional focus of other conversational participants. Since the main thing that syntax is for is passing concepts around, the belief that syntactic structure keeps as close as possible to semantics, and that in both J In thecontextof modernlinguistics , the suggestion goesbackat leastto Chomsky( 1965: 5659) and Miller ( 1967 ). But of courseit is a mucholder idea. SeePinker( 1979 ) for a reviewof someproposedmechanisms , includingthe importantcomputationalwork of Anderson( 1977 ), and seeGleitman( 1990 ) for somecogentwarningsagainstthe assumptionthat suchsemantic havetheir origin solelyin presentperceptionand thematerialworld in anysimple representations senseof that term.

476

M . SteedmanI Acquisition l?f verb categories

evolutionary and child language acquisition terms, the early development of syntax amounts to little more than hanging words onto the preexisting armatures of conceptual structure is so simple and probable as to amount to the null hypothesis. 4 Of course, as Chomsky has repeatedly pointed out , this realisation gets us practically nowhere. We have such a poor grasp of the nature of the putative underlying conceptualstructures that it is difficult to even designexperimental tests of the claim (quite apart from the other difficulties that arise in doing experimentswith prelinguistic childrenS Gleitman and others in the present volume have made considerable headway in the face of thesedifficulties, but there is a long way to go. For similar reasons to do with limitations on current knowledge, it does not seemto constrain syntactic theory in any very ' useful way. Right now (and this is Chomsky s substantive point ), the most reliable entry to the human system of language and symbolic cognition that ' we have comes from the linguists phenomenological grasp of the syntactic epiphenomenon, which has only just begun to look as though it is yielding some insight into the underlying conceptual structure. Nevertheless, the claim that semantics is the precursor of syntax is not without content, and has consequencesfor the question at hand. In particular , it immediately entails that if we are asking ourselveswhy children do not classify meet as subcategorisingfor NP PP on the basis of sentenceslike ( lb ), we met Harry on the bus, then we are simply asking the wrong question. A child who learns this instance of this verb from this sentencemust start from the knowledge that the denoted event is a meeting, and that this involves a ' transitive event concept. It usually never crossesthe child s mind that meet might subcategoriselike put , becausethe conceptual representation usually ' doesn t suggestthat. Once again, taking this position raisesmore questions than it answers. We are only just beginning to make sense of the complex mapping between surfacegrammatical roles like subject and object, and the underlying thematic roles that seemto be characteristic of the conceptual level. (I am particularly thinking of recent work by Grimshaw 1990.) It also raises the question of 4 The use of the words ' little more' rather than ' nothing more' is important . It would not be surprising to find that some part of syntax perhaps the observed constraints upon consistent had its origin elsewherethan in semantics. orders across headsand complements 5 I am not saying that logicians and my fellow computer scientists do not have interesting fonnalisms for representingconceptual structures. In fact these systemsare the main source of fonnal theoretical devices that linguists have to draw on. But as knowledge representation systems, none of them as yet seemparticularly close to the human one.

M. Steedman I Acquisitionof verbcategories

477

' whether the child s conceptual representation really can be used reliably in this fashion , which Pinker ( 1989) has called ' semantic bootstrapping ' , and , if not . how the child can cope with its unreliability .

S. Syntacticandsemantic Gleitman ( 1990) argues very persuasively that the child must often find itself in a situation which is ambiguous with respect to the conceptual representation. To take one of her examples, a child who is being read a story from their picture-book about a fox and a rabbit may have insufficient information from their understanding of the story and from a picture of the fox running after the rabbit to tell whether an unknown verb in an adult sentenceof the form ' The fox is V ER Sing the rabbit ' should be associated with the concept of chasing, or the concept of fleeing. In a number of elegant experiments, she and Landau and their colleagueshave shown that children who are artificially placed in this situation identify whichever conceptual representationis consistent with the syntactic form of the sentence. Since this is the only information that appearsto be available to the children as a basis for the decision, they argue that children are capable of using a process of ' ' syntactic bootstrapping to aid them in learning the subcategorisationsof verbs. Such a processrequires the child to generalisefrom its existing partial ' knowledge of syntax and verbal subcategorisations, perhaps via the linking ' ru1es discussedby Pinker and his colleagues(cf. Gropen et al. 1991). Gleitman' s proposal is in principle entirely consistent with semantic boot' strapping in the senseoutlined here. It is actually quite likely that the child s ' conceptual representation isn t much more underspecifiedthan being ambiguous as to whether this situation is an instance of transitive chasing or transitive fleeing, and a few other equally relevant propositions. That is, we can probably assumethat the child knows what it is to be ' read a story' , that the story is ' about the fox trying to eat the rabbit ' , and a lot of relevant facts such as that ' to eat something you have to catch it ' , ' rabbits don' t like being ' ' eaten , and so on. (If they don t know stuff like this, then they may be in a position to learn the nouns, but probably not to learn the verbs.) In this case, there aren' t likely to be that many other possibilities, and syntactic bootstrapping may well reduce the set to one possible meaning. If so, it may well do so in the face of syntactic complications that are irrelevant to subcategorisations, such as the presenceof modifiers or adjuncts. Their presencewill not prevent the child from acquiring the meaning of the verb, which it must have already

478

M . Steedman/ Acquisition of verb categories

as one of a number of alternatives that syntactic bootstrapping correctly disambiguates, if learning is to take place at all. One piece of circumstancial evidence in support of this conjecture is that adults work this way too. There is increasing experimental evidencethat the adult sentenceprocessing mechanism deals with the huge degree of nondeterminism that arises from natural grammars by appealing to meaning, filtering out the myriad spurious paths that the grammar permits on the basis of whether they make sense, both on the basis of sentence-internal semantics, and of reference and extension in the context. This semantic filtering of spurious paths which would otherwise overwhelm the computational resources of the processor has been claimed to go on continually at every point in the parsing process, with very fine ' grain' , probably more or less word by word . (See Steedman and Altmann 1989 and Clifton and Ferreira 1989 for referencesand arguments pro and contra this proposal, which ultimately comes from computer science, particularly in work by Winograd 1972.) Nevertheless, it may well be the case, as Gleitman suggests , that children are frequently much more at sea than this, and may even have much larger sets of propositions in mind , most or even all of which are irrelevant to the adult meaning. However, recent computational work by Siskind ( 1992) shows that a processof intersecting such setson successiveencounterswith the verb can be used to eliminate the spurious meanings. 6 Of course, children are not adults, and neither are they mind readers, and a ' ' meaning that seems appropriate to them over a number of iterations of this ' ' ' ' processmay not be the same as the adult s. The child s concept of chasing ' (we may imagine as an extension of Gleitman s example) may be overspecifically restricted to an activity of attempting to catch by running . In this case, ' ' their own future use may be characterised by undergeneralisation - for example, they may be unwilling to agreethat a similar scenario involving cars is chasing. There is of course a huge literature that has revealedthe fine detail of this process. 7 There are also instancesof overgeneralisation, and possibly 6 Two problems which Siskind leaves open are the problem of polysemous verbs, and the problem that ariseswhen the set of putative meaningsderived from an occurrenceof the verb are all spurious. Both of these eventualities will lead to empty intersections. One simple tactic that might serveto distinguish them and thereby be used to maintain a coherent lexicon would be to respond to an empty intersection by keeping both entries, relying on a tactic like binomial error estimation to distinguish between true polysemous lexical entries and spurious ones on distributional grounds. 1 For example, Brown ( 1973), Bowerman ( 1973) and Clark ( 1973), and Carey ( 1982), the last including an extensivereview.

M. Steedman I Acquisitionof verbcategories

479

' even more bizarre ' complexes, revealed in non-standard lexical meanings. There is also evidencethat children predict new lexical entries that they have not actually encountered, via lexical rules such as the rule that generates causative verbs from certain adjectives, such as cool. This process may on occasiongive rise to non-standard lexical causatives, as in # It colds my hand, either becauseof slightly non-standard lexical rules, or becausestandard rules are applied to slightly non-standard lexical entries. (SeeBowerman 1982and referencestherein.) The way in which children successivelymodify non-standard lexical items to approximate the adult lexicon is the most challenging and least wellunderstood part of the process. But the undoubted fact that the processes of syntactic and semantic bootstrapping appear to iterate in this way suggests that together they may constitute the processby which children gain accessto concepts which are not immediately available to prelinguistic sensory-motor cognition , and may thereby provide the force behind the explosive change in cognitive abilities that coincides, both in evolutionary and in child -developmental terms, with the appearanceof language. 8 Computational models of the kind proposed by Brent and Siskind will continue to provide the only way in which theories of this process, such as syntactic, prosodic and semantic ' ' bootstrapping , can be developed and evaluated.

References Anderson . Inductionof augmentedtransitionnetworks . CognitiveScienceI , 125, John, 1977 157. . Earlysyntacticdevelopment . Cambridge : CambridgeUniversityPress . Bowerman , Melissa , 1973 . Reorganisational . In : Bowerman esin lexicaland syntacticdevelopment , Melissa , 1982 process Eric Wanner, Lila Gleitman (eds.), Languageacquisition: The state of the art, 319- 346. : CambridgeUniversityPress . Cambridge Brent, Michael, 1991 . Automatic acquisitionof subcategorisation frames from unrestricted , MIT , CambridgeMA. English. UnpublishedPhD . dissertation Brown, Roger, 1973 . A first language : The early stages . Cambridge , MA : Harvard University Press . . Semanticdevelopment . In: Eric Wanner, Lila Gleitman(eds.), Language , 1982 Carey, Sasan . : CambridgeUniversityPress acquisition:The stateof the art, 347- 389. Cambridge . What's in a word? In: T. Moore (ed.), Cognitivedevelopmentand the Clark, Eve, 1973 . New York: AcademicPress . acquisitionof language 8 SeeVygotsky( 1962 on the natureof this processand its relation ) for someearlyspeculations -motor development to Piagetiansensory . and seeOleron ( 1953 ) for some ) and Furth ( 1961 earlystudiesof the effectsof deprivation. suggestive

480

M . Steedman/ Acquisition of verb categories

. The young word maker. In: Eric Wanner, Lila Gleitman(eds.), Language Clark, Eve, 1982 : CambridgeUniversityPress . acquisition:The stateof the art, 39()..428. Cambridge . Aspectsof the theoryof syntax. Cambridge . , Noam, 1965 , MA : MIT Press Chomsky Clifton, Charlesand FernandaFerreira, 1989 . Ambiguity in context. Languageand Cognitive Process es4, 77- 104. Furth, Hans, 1961 . The influenceof languageon the development of conceptformationin deaf children. Journalof Abnormaland SocialPsychology 63, 186- 389. Gleitman, Lila, 1990 . The structuralsourceof verbmeanings . LanguageAcquisitionI , 3- 55. Grimshaw . Argumentstructure. Cambridge . , Jane, 1990 , MA : MIT Press . Affectedness and , StevenPinker, MiceileHollanderand RichardGoldberg, 1991 Gropen, Jess direct objects: The role of lexicalsemanticsin the acquisitionof verb argumentstructure. Cognition41, 153- 196. . Usingsoundto solvesyntacticproblems . Psychological Review99, 349Kelly, MichaelH., 1992 364. . Domain-generalabilitiesappliedto domain-specifictasks: Kelly, MichaelH. andS. Martin, 1994 to in . Lingua92, 105- 140. (this , cognition, and language Sensitivity probabilities perception volume). Lederer . Prosodicinformationfor syntacticstructurein parental , Anneand MichaelKelly, 1992 . Paperpresented to 32ndMeetingof the Psychonomic . , SanFrancisco , 1991 speech Society Miller, George . The psychologyof communication . Harmondsworth : Penguin . , 1967 . Conceptualthinkingof the deaf. AmericanAnnalsof th~ Deaf98, 304- 310. Oleron, Paul, 1953 Pinker, Steven . Formalmodelsof languagelearning. Cognition7, 217- 283. , 1979 Pinker, Steven . Learnabilityand cognition. Cambridge . , 1989 , MA : MIT Press Siskind, Jeffrey, 1992 . Naivephysics , eventperception , lexicalsemantics , and languageacquisition . UnpublishedPhD . dissertation , MIT , Cambridge , MA. Steedman . Ambiguity in context: A reply.. Languageand , Mark and Gerry Altmann, 1989 es4, 105- 122. CognitiveProcess . Thoughtand language Lev 1962 . Cambridge . , , , MA : MIT Press Vygotsky . Understanding naturallanguage . New York: AcademicPress . Winogard, Terry, 1972

Appendix -fromthe Symbols Phonetic International with , Alphabet examples

a a .. A aJ au b

I j

st.QP , C2P Im ~ , ~ ther f.@ :t, pl.@ :cid byt, ft.Q.Qd ~ , kite ~ I, h~ , ho,ho ,hoss ~ balo, pi~ go, hag w~ e, rase b.@ :!t, d~ gmong, readily w~ ther, ! DY felt, left So, les hat, how ' Ohio, ahead ~ I, ~ if, remit ~et, can~on

k I

~ , bac -k lip, bill

tl d d3 el ~ f g h h I

m n IJ ou .') 0 .')1 P r s I t 9 6 u u 3r v w x

z 3

man, him ,n.ose, aImoy wi!!,l, sink gQ, SQY1 w.@ .lk , I.@ .w w.@ .Y.t .nt, astron.@ v bQil, .Qice eie, nae rim, parrot .'-imon, bog mip, f~ al lawn, walkeg ! !hiD, mytJ !l!en, wea!lier r.QQf , ft~ l.QQk , fyll bird, wQ!St yoice, giye m , qyail (Spanish ) proteFr , (Dutch) ka~ el ~ I, ~ lealure, garale

Bibliography

Abbs, J. H., &; Cole, K. J. ( 1982 ). Considerationsof bulbar and suprabulbar . In afferentinfluencesupon speechmotor coordinationand programming . motor control . Eds A. Person & ; Lubker J. B. Lindblom Grillner S. ), Speech , ( , , . Oxford: PergamonPress . EdinburghUniversity Abercrombie , D. ( 1967 ). Elementsof generalphonetics . Press , andthought.Hillsdale, NJ: Lawrence Anderson , memory , J. R. ( 1976 ). Language Erlbaum. . Cambridge , MA: Harvard Anderson , J. R. ( 1983 ). Thearchitectureof cognition Press . University . . CambridgeUniversityPress ). PlanningEnglishsentences Appelt, E. A. ( 1985 Arbib, M. A. ( 1981 ). Perceptualstructuresand distributedmotor control. In : Vol. 2. Motor control . Thenervoussystem V. Brooks(Ed.), Handbookof physiology . Bethesda Society. , MD: AmericanPhysiological Atkinson, J. M., &; Drew, P. ( 1979 ). Orderin court. London: MacMillan. . Austin, J. L. ( 1962 ). Howto dothingswithwords.Oxford: ClarendonPress ). Output editing for lexical Baars, B. J., Motley, M. T., &; MacKay, D. ( 1975 . Journal of VerbalLearningand statusfrom artificiallyelicitedslipsof the tongue . 391 14 382 VerbalBehavior , , . . Oxford: ClarendonPress A. 1986 , ( ). Workingmemory Baddeley roles. Berlin ,functionsandsemantic Barth, E. M., &; Wiche, R. T. P. ( 1986 ). Problems : DeGruyter. , MA: MIT Barwise ). Situationsand attitudes. Cambridge , J., &; Perry, J. ( 1983 . Press production. Bates,E., &; Devescovi , A. (forthcoming). Competitionand sentence sentence studies . Cross processing of In B. MacWhinney&; E. R. Bates(Eds), linguistic . . CambridgeUniversityPress

Bibliography

502

Bates, E., & MacWhinney, B. ( 1982) . Functionalist approaches to grammar. In E. Wanner & L . Gleitman ( Eds.), Language acquisition: The State of the art . Cambridge University Press. Bates, E., & MacWhinney, B. ( 1987) . Competition , variation , and language learning. In B. MacWhinney ( Ed.), Mechanismsof language acquisition. Hills dale, NJ: LawrenceErlbaum. Bates, E., Masling, M ., & Kintsch , W . ( 1978) . Recognition memory for aspectsof dialogue. Journal of Experimental Psychology: Human Learning and Memory, 4, 187- 197. Bauer, L . ( 1983) . English word-formation. Cambridge University Press. Beattie, G. ( 1978), Floor apportionment and gaze in conversational dyads. British Journal of Social and Clinical Psychology, 17, 7- 16. Beattie, G. ( 1980). Encoding units in spontaneousspeech: Some implications for the dynamics of conversation. In H . W. Dechert & M . Raupach ( Eds.), Temporal variablesin speech.The Hague: Mouton . Beattie, G. ( 1983) . Talk: An analysis of speechand non- verbal behavior in conver-

Press . sation . MiltonKeyes : OpenUniversity " ?" - A replyto Power Beattie , , G. ( 1984 ). Are therecognitive rhythmsin speech - 195 . . Language andSpeech 1983 , 27, 193 Beattie , G., & Butterworth, B. ( 1979 ). Contextualprobabilityand word frequen . Language andSpeech in spontaneous asdetenninants of pauses , 22, speech 201- 211.

es. Norwood, NJ: Ablex. , andprocess , R. de. ( 1980 ). Text, discourse Beaugrande Bell-Berti, F., & Harris, K. ( 1979 ). Anticipatory coarticulation: Someimplicatio from a study of lip rounding. Journal of the AcousticalSocietyof . America,65, 1268- 1270 : Evidencefrom Germanspeech ). Thecaseagainstaccommodation Berg, T. ( 1987 errordata. Journalof MemoryandLanguage , 26, 277- 299. Bierwisch , M. ( 1965 ). Regelnfiir die IntonationdeutscherSitze. In M. Bierwisch jiber Akzent und Intonationim a VI/. Untersuchungen (Ed.), StudiaGrammatic -Verlag. . Berlin: Akademie Deutschen Bierwisch , M. ( 1982 ). Linguisticsand languageerror. In A. Cutler (Ed.), Slipsof . Berlin: Mouton. thetongueandlanguage production . In Bierwisch , M. ( 1986 ). On the natureof semanticform in natural language : F. Klix & H. Hagendorf (Eds.), Human memoryand cognitivecapabilities . Amsterdam : North-Holland. andperformances Mechanisms on syntacticstructure Bock, J. K. ( 1977 ). The effectof pragmaticpresupposition . Journal of VerbalLearningand VerbalBehavior in questionanswering , 16, 723- 734.

Bibliography

503

Bock, J. K. ( 1982 ). Towards a cognitive psychologyof syntax: Infonnation , 89, to sentencefonnulation. PsychologicalReview contributions processing 1- 47. Bock, J. K. ( 1986a ). Meaning, sound, and syntax: Lexicalpriming in sentence : Learning, Memory, and Cognition . Journal of ExperimentalPsychology production . 586 /2 575 , , in languageproduction. CognitivePsychology Bock, J. K. ( 1986b ). Syntacticpersistence , /8, 355- 387. Bock, J. K. ( 1987a ). Coordinatingwordsand syntaxin speechplans. In A. Ellis (Ed.), Progressin the psychologyof language(Vol. 3). London: Lawrence Erlbaum. Bock, J. K. ( 1987b ). An effectof the accessibilityof word fonDS on sentence . Journalof MemoryandLanguage , 26(2), 119- 137. structures Bock, J. K., & Irwin, D. E. ( 1980 ). Syntacticeffectsofinfonnation availabilityin , / 9, sentenceproduction. Journal of Verbal Learning and Verbal Behavior 467- 484. Bock, J. K., & Warren, R. K. ( 1985 ). Conceptualaccessibilityand syntactic . fonnulation in sentence , 2/ , 47- 67. structure Cognition ' . Unpublished pronouns Boehme of Germanpossessive , K. ( 1983 ). Childrens understanding doctoraldissertation , NijmegenUniversity. . . London: Penguin BolingerD . (Ed.). ( 1972 readings ). Intonation:Selected , 2, BolingerD . ( 1983 ). Wheredoes intonation belong? Journal of Semantics 101- 120. BolingerD . ( 1986). Intonationanditsparts. London: EdwardArnold. : Bond, Z. S., & Small, L. H. ( 1984 ). Detectingandcorrectingmispronounciations 283 . 279 / 2 . Journalof Phonetics , , A noteon methodology . Languageand and Boomer, D. S. ( 1965 grammaticalencoding ). Hesitation . 158 148 8 , , Speech Boomer,D. S., & Laver, J. D. M. ( 1968 ). Slipsof the tongue. Journalof Disorders in Fromkin, V. A. (Ed.). ( 1973 12 . 2 3 ). Speech Communication , , (Reprinted of . TheHague: Mouton.) errorsaslinguisticevidence intonationandlanguage BrazilD ., Coulthard, M., & Johns, C. ( 1980 ). Discourse . Burnt Mill , Harlow: Longman. teaching of someproblemsof Bransford,J. D., & Johnson , M. K. ( 1973 ). Considerations . New York: Visual Ed. W. G. Chase . In informationprocessing ( ), comprehension . AcademicPress Bresnan , J. ( 1978 ). A realistic transfonnational grammar. In M. Halle, J. Bresnan& G. A. Miller (Eds.), Linguistictheoryand psychologicalreality. . , MA: MIT Press Cambridge

504 Bresnan , J. (Ed.). ( 1982 ). The mentalrepresentation of grammaticalrelations. . , MA: MIT Press Cambridge Broadbent , D. E. ( 1975 ). The magic number sevenafter fifteen years. In A. & A. Wilkes . London: John (Eds.), Studiesin long term memory Kennedy Wiley. Brodda, B. ( 1979 ). Nagot om de Svenskaordensfonotax och morfotax. (Paper No. 38). Universityof Stockholm,Instituteof Linguistics. Browman, C. P. ( 1978 ). Tip of the tongueand slip of the ear. Implicationsfor . UCLA WorkingPapersin Phonetics , 42, 1- 149. language processing Browman,C. P., & Goldstein, L. M. ( 1986 ). Towardsan articulatoryphonology. . Yearbook 3 219 252 , , Phonology . Brown, G., Currie, K. L., & Kenworthy, J. ( 1980 ). Questionsof intonation London: CroomHelm.

. CambridgeUniversityPress . Brown, G., & Yule, G. ( 1983 ). Discourse analysis : Theexplicit Brown, P., & Dell, G. S. ( 1987 ). Adaptingproductionto comprehension . CognitivePsychology mentionof instruments , 19, 441- 472. Brown, R. ( 1973 , languageand reality. AmericanPsychologist , ). Schizophrenia 28, 395- 403. " " . Journal Brown, R., & McNeill, D. ( 1966 ). The tip of the tongue phenomenon , 5, 325 337. of VerbalLearningandVerbalBehavior ' . NewYork: Norton. Bruner, J. S. ( 1983 ). Childs talk: Learningto uselanguage . Stuttgart: Fisher. Buhler, K. ( 1934 ). Sprachtheorie Burkle, B., Ninnaier, H., & Hernnann, Th. ( 1986 ). Vondir aus . . . Zur horer. (Bericht 10). Mannheim : Universityof Mannheim lokalenReferenz , bezogenen im SozialenKontext. undSprachverstehen Forschergruppe Sprechen Butterworth, B. ( 1972 ). Semanticanalyses of thephasingoffluencyin spontaneous . Unpublisheddoctoraldissertation , Universityof London. speech

Butterworth, B. ( 1975 . Journalof ). Hesitationand semanticplanningin speech Research , 4, 75- 87. Psycholinguistic Butterworth, B. ( 1980a ). Someconstraintson modelsof languageproduction. In B. Butterworth(Ed.), Language : Vol. 1. Speech andtalk. London: Academ production Press . Butterworth, B. ( 1980b . In B. Butterworth ). Evidencefrom pausesin speech : Vol. 1. Speech andtalk. London: AcademicPress . (Ed.), Language production Butterworth, B. (Ed.). ( I 98Oc : Vol. 1. Speechand talk. ). Languageproduction London: AcademicPress . Butterworth, B. ( 1982 ). Speecherrors: Old data in searchof new theories. In A. Cutler(Ed.), Slipsof thetongueandlanguage . Berlin: Mouton. production

505

Bibliography

Butter worth, B. ( 1983a). Lexical representation. In B. Butter worth ( Ed.) . Language production: Vol. 2. Development. writing and other language processes. London: Academic Press. Butter worth, B. ( Ed.) . ( 1983b) . Languageproduction: Vol. 2. Development. writing and other languageprocesses. London : Academic Press. Butter worth, B. (forthcoming) . Lexical access in speech production . In W. Marslen-Wilson ( Ed.). Lexical representationand process. Cambridge, MA : MIT

. Press Calvert, D. R. ( 1980 ). Descriptive

phonetics . New York : Decker .

Carr , T . H . ( 1979). Consciousnessin models of human infonnation processing: Primary memory, executive control , and input regulation. In G. Underwood & : Vol. 1. Psychologicalissues. London : R. Stevens(Eds.), Aspectsof consciousness . Academic Press Carroll , J. M . ( 1958) . Process and content in psycholinguistics. In R. Patton (Ed.), Current trends in the description and analysis of behavior. University of Pittsburgh Press. Carroll , J. M . ( 1983a). Nameheads. CognitiveScience, 7, 121- 153. Carroll , J. M . ( 1983b) . Toward a functional theory of names and naming. Linguistics , 21, 341- 371. Chafe, W. L . ( 1970). Meaning and the structure of language. University of Chicago Press. Chafe, W . L . ( 1976) . Giveness, contrastiveness, definiteness, subjects, topics, and point of view. In C. N . Li (Ed.), Subjectand topic. New York : Academic Press. Chafe, W . L. ( 1977) . The recall and verbalization of past experience. In R. W . Cole (Ed.), Current issuesin linguistic theory. Bloomington : Indiana University Press. Chafe, W. L . ( 1979) .The flow of thought and the flow of language. In T . Givon ( Ed.), Syntax and semantics: 12. Discourse and syntax. New York : Academic Press. Chafe, W . L . ( 1980). The deployment of consciousnessin the production of a narrative . In W . L . Chafe (Ed.), The pear stories. Cognitive. cultural . and linguistic aspectsof narrative production. Norwood , NJ: Ablex. Chomsky, N ., & Halle , M . ( 1968) . The sound pattern of English. New York : Harper and Row. Clark , E. V. ( 1970) . How young children describeevents in time. In GiB . Flores d' Arcais & W . J. M . Levelt ( Eds.), Advancesin psycholinguistics. Amsterdam: North -Holland . Clark , E. V . ( 1974) . Nonnal states and evaluative viewpoints. Language, 50, 316- 332.

Bib" ograph ~'

506

Clark , E. V. ( 1977) . From gesture to word: On the natural history of deixis in languageacquisition. In J. S. Bruner & A . Garton ( Eds.), Human growth and develop . Oxford : Wolfson College. Clark , E. V. ( 1987) . The principle of contrast: A constraint on languageacquisition . In B. MacWhinney (Ed.), Mechanismsof language acquisition. Hillsdale, NJ: LawrenceErlbaum. Clark , E. V. ( 1988) . On the logic of contrast. Journal of Child Language, 15, 317- 335.

and the acquisitionof language . New York: Academic Cognitivedevelopment Press . Clark, H. H. ( 1979 ). Respondingto indirectspeechacts. CognitivePsychology , 4, 430- 477.

Clark, H. H. ( 1985 . In G. Lindsey& E. Aron). Languageuseand languageusers son(Eds.), Thehandbook social . New York: of psychology HarperandRow. -Papi& Clark, H. H., ( 1987 ). Four dimensionsof languageuse. In M. Bertuccelli J. Verschueren . Amsterdam : JohnBenjamins . (Eds.), Thepragmaticperspective Clark, H. H., & Clark, E. ( 1977 andlanguage . An introductionto psycholi ). Psychology . NewYork: HarcourtBrace. Clark. H. H., & Gerrig, R. J. ( 1984 ). On the pretensetheoryof irony. Journalof : General , 113, 121- 126. Experimental Psychology Clark, H. H., & Haviland, S. E. ( 1977 and the given-newcontract ). Comprehension . In R. O. Freedle(Ed.), Discourseproductionand comprehension . New York: AcademicPress .

Clark, H. H., Schreuder , R., & Buttrick, S. ( 1983 ). Commongroundand the underst of demonstratives . Journalof VerbalLearningand VerbalBehavior , 22, 245- 258. Clark, H. H., & Schunk, D.H. ( 1980 to polite requests . Cognition ). Politeresponses , 8, 111- 143. Clark, H. H., & Wilkes-Gibbs, D. L. ( 1986 . ). Referringasa collaborativeprocess , 22, 1- 39. Cognition

Clements . PhonologyYearbook , G. N. ( 1985 ). The geometryof phonologicalfeaturies , 2, 225- 252. Clements . A generative , G. N., & Keyser, S. J. ( 1983 ). CV Phonology theoryof the . , MA : MIT Press syllable.Cambridge

Bibliography

507

Cohen, A . ( 1965) . Versprekingen als verklappers: bij bet proces van spreken en verstaan. Forum der Letteren, 6, 175- 186. Cohen, A . ( 1968). Errors of speechand their implication for understanding the strategy of the languageusers. Zeitschrift fUr Phonetik, 21, 177 181. ' Cohen, A ., Collier , R., & Hart , J. t ( 1982) . Declination: Construct of intrinsic feature of speechpitch? Phonetica, 39, 254- 273. ' Cohen, A ., & Hart , J. t ( 1965) . Perceptual analysis of intonation patterns. Proceedings of the Vth ICA , A16. Liege, Belgium. Collier , R., & Gelfer, C. E. ( 1983) . Physiological explanations of FO Declination. In M . P. R. Van den Broecke & A . Cohen (Eds.), Proceedingsof the Tenth International . Dordrecht: Foris. Congressof PhoneticSciences Cooper, W . E. ( 1976) . Syntactic control of timing in speechproduction. Unpublished doctoral dissertation, MassachusettsInstitute of Technology. Cooper, W. E., & Danly M . ( 1981) . Segmentaland temporal aspectsof utterance . 106 115 final lengthening. Phonetica, 38, Cooper, W . E., & Eady, S. J. ( 1986). Metrical phonology in speechproduction . Journal of Memory and Language, 25, 369- 384. Cooper, W . E., & Paccia-Cooper, J. M ., ( 1980) . Syntax and speech. Cambridge, MA : Harvard University Press. . Grossmann, L . J. Cooper, W. E., & Ross, J. R. ( 1975) . World order. In RE San, & T . J. Vance ( Eds.), Papersfrom the parasessiononfunctionalism. Chicago Linguistic Society. . Cooper, W. E., & Sorensen, J. M . ( 1981). Fundamentalfrequency in sentenceproduction . New York : Springer. Cooper, W. E., Tye-Murray , N ., & Nelson, L . J. ( 1987) . Detection of missing words in spoken text. Journal of PsycholinguisticResearch, 16, 233- 240. Crompton , A . ( 1982) . Syllables and segmentsin speechproduction . In A . Cutler (Ed.), Slips of the tongueand languageproduction. Berlin: Mouton . Cruse, D . A . ( 1986). Lexical semantics. Cambridge University Press. Cruttenden, A . ( 1986) . Intonation. Cambridge University Press. Cutler , A . ( 1980a). Errors of stress and intonation . In V. A . Fromkin (Ed.), Errors in linguistic performance. New York : Academic Press. Cutler , A . ( 1980b) . Syllable omission errors and isochrony. In H . W. Dechert & M . Raupach ( Eds.), Temporal variables in speech. Studies in honour of Frieda Goldman- Eisler. The Hague: Mouton . Cutler , A . ( Ed.) . ( 1982a). Slips of the tongue and languageproduction. Berlin: Mouton . Cutler , A . ( 1982b) . Speecherrors; A classified bibliography. Bloomington : Indiana Linguistics Club .

Bibliography

508

Cutler , A . ( 1983a). Lexical complexity and sentenceprocessing. In G. B. Flores d' Arcais & R. J. Jarvella ( Eds.), The processof languageunderstanding. Chichester: John Wiley . Cutler , A . ( 1983b). Speakers' conceptions of the function of prosody. In ACutle . R. Ladd ( Eds.), Prosody: Models and measurements . Heidelberg: Springer. Cutler , A . ( 1987) . Speaking for listening. In A . Allport , D . G. MacKay , W. Prinz, & E. Scheerer( Eds.), Languageperception and production: Relationships betweenlistening, speaking, readingand writing . London : Academic Press.

Cutler , A ., & Clifton , C., Jr. ( 1984) . The use of prosodic information in word recogn . In H . Bouma & D . G. Bouwhuis ( Eds.), Attention and performance: X. Control of languageprocesses. Hillsdale, NJ: Lawrence Erlbaum. Cutler , A ., & Isard, S. ( 1980). The production of prosody. In B. Butter worth ( Ed.), LanguageProduction: Vol. J. Speechand talk. London : Academic Press. Cutler , A ., & Ladd, D . R. ( Eds.) . ( 1983). Prosody: Models and measurements . Heidelberg: Springer. Cutler , A ., & Pearon, M . ( 1986) . On the analysis of prosodic turn - taking cues. In C. Johns- Lewis (Ed.), Intonation in discourse. London: Croom Helm. Daniloff , R. G., & Hammarberg, R. E. ( 1973) . On defining coarticulation . Journal of Phonetics, J, 239- 248. Danks, J. H . ( 1977). Producing ideas and sentences . In S. Rosenberg (Ed.), Sentenceproduction: Developmentsin research and theory. New York : John Wiley . Dauer, R. M . ( 1983) . Stress-timing and syllable-timing reanalyzed. Journal of Phonetics, JJ, 51- 62. Davidson-Nielson, N . ( 1975). A phonological analysis of English sp, st, sk, with specialreferenceto speecherror evidence. Journal of the IPA , 5, 3- 25. Deese, J. ( 1980) . Pauses, prosody, and the demandsof production in language. In H . W . Dechert & M . Raupach (Eds.), Temporal variables in speech. Studies in honourof Frieda Goldman-Eisler. The Hague: Mouton . Deese, J. ( 1984) . Thought into speech: The psychologyof a language. Englewood Cliffs , NJ: Prentice- Hall .

Dell , G. S. ( 1980). Phonological and lexical encoding in speechproduction: An analysis of naturally occurring and experimentally elicited speech errors. Unpublis doctoral dissertation, University of Toronto . Dell , G. S. ( 1984) . Representation of serial order in speech: Evidence from the repeated phoneme effect in speecherrors. Journal of Experimental Psychology: Learning, Memory, and Cognition, 10, 222- 233.

Dell , G. S. ( 1985). Positive feedback in hierarchical connectionist models: Applicatio to languageproduction . CognitiveScience, 9, 3- 23.

Bibliography

509

Dell, G. S. ( 1986 production ). A spreadingactivationtheoryof retrievalin sentence . Psychological Review , 93, 283- 321. Dell, G. S. ( 1988 ). The retrievalof phonologicalformsin production: Testsof , 27, predictionsfrom a connectionistmodel. Journalof Memoryand Language 124- 142. Dell, G. S., & Reich, P. A. ( 1977 ). A model of slips of the tongue. The Third L A C US Forum, 3, 448- 455.

Dell , G . S., & Reich, P. A . ( 1980) . Toward a unified theory of slips of the tongue. In V. A . Fromkin (Ed.), Errors in linguistic performance: Slips of the tongue, ear, pen, and hand. New York : Academic Press. Dell , G. S., & Reich, P. A . ( 1981) . Stagesin sentenceproduction : An analysis of speecherror data. Journal of Verbal Learning and VerbalBehavior, 20, 611 629. Denes, P. B., & Pinson, E. N . ( 1963). The speechchain. Baltimore: Bell Telephone Laboratories. De Pijper, J. R. ( 1983) . Modelling British English intonation: An analysis by resynthesis of British English intonation. Dordrecht : Foris. De Rooy , J. J. ( 1979) . Speechpunctuation: An acoustic and perceptual study of some aspects of speechprosody in Dutch. Unpublished doctoral dissertation, Utrecht University . De Smoot, K ., & Kempen, G. ( 1987) . Incremental sentenceproduction , self. correction, and coordination . In G. Kempen (Ed.), Natural languagegeneration: Recent advancesin artificial intelligence, psychology and linguistics. Dordrecht: Kluwer . Deutsch, W . ( 1976) . SprachlicheRedundanzund Objektidentifikation. Unpublished doctoral dissertation, University of Marburg . Deutsch, W., & Jarvella, R. J. ( 1983) . Asymmetrien zwischen Sprachproduktion . In C. F. Graumann & T . Herrmann (Eds.), Karl Biihlers und Sorachverstehen ~.-

. Axiomatik. Frankfurt: Klostermann

Deutsch , T. ( 1978 , W., & Pechmann ). Ihr , dir, or mir? On the acquisitionof pronouns in Germanchildren. Cognition , 6. 155- 168. of Deutsch , T. ( 1982 , W., & Pechmann ). Socialinteractionand the development . Cognition definitedescriptions , 11, 159- 184. :A Diehl, R. L. ( 1986). Coproductionanddirectperceptionof phoneticsegments critique . Journal of Phonetics, 14, 61- 66.

DikS . C. ( 1978) . Functionalgrammar. Amsterdam: North -Holland . DikS . C. ( 1987) . Linguistically motivated knowledge representation. In M . Nagao (Ed.), Language and artificial intelligence. Amsterdam: Elsevier Science Press. . Cambridge Dixon , R. M . W . ( 1972). The Dyirbal languageof North Queensland University Press.

Bibliography

510

Dogil , G. ( 1984). Grammatical prerequisitesto the analysis of speechstyle: fast/ casual speech. In D . Gibbon & H . Richter ( Eds.), Intonation, accentand rhythm. Berlin: De Gruyter . Doherty , M . ( 1987) . Epistemicmeaning. Heidelberg: Springer. Donders, F. C. ( 1868). Die Schnelligkeit psychischerProzesse . Archiv Anatomie und Physiologie, 657- 681. Du Bois, J. W. ( 1974) . Syntax in mid -sentence. Berkeley Studies in Syntax and Semantics(Vol . I : 111 . 1- 111.25) . Berkeley: University of California , Institute of Human Learning and Department of Linguistics. DuncanS ., & Fiske, D . W. ( 1977) . Face- to-face interaction: Research, methods and theory. Hillsdale, NJ: LawrenceErlbaum. Easton, T . A . ( 1972) . On the normal uses of reflexes. American Scientist, 60, 591- 599. Ehrich, V. ( 1982) . The structure of living spacedescriptions. In R. J. Jarvella & W. Klein ( Eds.), Speech,place and action. Studiesin deixis and related topics. Chi chester: John Wiley . Ehrich, V. ( 1985). Zur Linguist ik und Psycholinguistik der sekundaren Raumdeixis. In H . Schweizer (Ed.), Sprache und Raum. Stuttgart : Metzlersche Verlagsbuchhandlung. Ehrich V. ( 1987) . The generation of tense. In G. Kempen ( Ed.), Natural language generation. Dordrecht: Martinus Nijhoff . Ehrich, V., & Koster, C. ( 1983). Discourse organization and sentenceform : The structure of room descriptions in Dutch . DiscourseProcesses, 6, 169- 195.

Eibl - Eibesfeldt, I . ( 1974) . Similarities and differences between cultures in expressi movements. In S. Weitz (Ed.), Non- verbal communication. Oxford University Press. Eibl -Eibesfeldt, I . ( 1984). Die Biologie desmenschlichenVerhaltens: Grundrissder Humanethologie: Vol. J. Munich : Piper. Eriksen, C. W., Pollock , M . D ., & Montague, W. E. ( 1970). Implicit speech: Mechanisms in perceptual encoding? Journal of Experimental Psychology, 84, 502- 507. Fauconnier, G. ( 1985). Mental Spaces: Aspectsof meaningconstructionin natural language. Cambridge, MA : MIT Press. Fay, D . ( 1980a). Performing transformations. In R. Cole ( Ed.), Perception and production offluent speech..Hillsdale, NJ: Lawrence Erlbaum. Fay D . ( 1980b). Transformational errors. In V. A . Fromkin ( Ed.), Errors in lin guisticperformance: Slips of the tongue, ear, pen, and hand. New York : Academic Press. Fay, D . ( 1982) . Substitutions and splices: a study of sentenceblends. In A . Cutler ( Ed.), Slips of the tongueand languageproduction. Amsterdam: de Gruyter .

Bibliography

511

Fay, D ., & Cutler , A . ( 1977) . Malapropisms and the structure of the mental lexicon. Linguistic Inquiry , 8, 505- 520. Feigenbaum, E. ( 1963) . The simulation of verbal learning behavior. In E. Feigenbaum & J. Feldman ( Eds.), Computersand thought. New York : McGraw -Hill . Fel' dman, A . G. ( 1966a). Functional tuning of the nervous systemwith control of movement or maintenanceof a steady posture. II . Controllableparameters of the muscles. Biophysics, JJ, 565- 578. Fel' dman, A . G. ( 1966b) . Functional tuning of the nervious system with control of movement or maintenanceof a steady posture. III . Mechanicographic analysis of the execution by man of the simplestmotor tasks. Biophysics, JJ, 766- 775. Fillmore , C. J. ( 1968) . The casefor case. In E. Bach & R. T . Harms (Eds.), Universals of linguistic theory. New York : Holt , Rinehart and Winston. Fillmore , C. J. ( 1973) . May we come in? Semiotica, 9, 98- 115. Fillmore , C. J. ( 1977) . The case for case reopened. In P. Cole & J. M . Sadock ( Eds.), Syntax and semantics: Vol. 8. Grammatical relations. New York : Academic Press. Fillmore , C. ( 1982). Towards a descriptive framework for spatial deixis. In R. J. Jarvella & W. Klein ( Eds.), Speech,place, and action: Studiesin deixis and related topics. Chichester: John Wiley . Flores d' Arcais, G. B. ( 1975) . Some perceptual determinants of sentenceconstruction . In G . B. Flores d ' Arcais ( Ed.), Studiesin perception. Milan : Martel lo . Flores d ' Arcais, G . B. ( 1987a). Automatic processes in languagecomprehension. In G. Denes, C. Semenza, P. Bisiacchi, & E. Andreewsky ( Eds.), Perspectivesin . Hillsdale, NJ: Lawrence Erlbaum. CognitiveNeuropsychology ' Flores d Arcais, G . B. ( 1987b). Perceptual factors and word order in event descriptions . In G. Kempen ( Ed.), Natural languagegeneration: New results in artificial intelligence, psychologyand linguistics. Dordrecht: Martinus Nijhoff . Flores d' Arcais, G. B., & Schreuder, R. ( 1987). Semanticactivation during object , 49, 153- 159. naming. PsychologicalResearch Fodor , J. A . ( 1975) . Thelanguageof thought. Hassocks: Harvester Press. Fodor , J. A . ( 1983) . Themodularity of mind. Cambridge, MA : MIT Press. Fodor , J. A . ( 1985) . Precis of the modularity of mind. The Brain and Behavioral

Sciences. 8 , 1- 42 .

Fodor , J. A . ( 1987). Modules, frames, fridgeons, sleepingdogs, and the music of the spheres. In J. L. Garfield (Ed.), Modularity in knowledgerepresentationand natural-languageunderstanding.Cambridge, MA : MIT Press. Fodor , J. A ., Bever, T . G ., & Garrett , M . F. ( 1974) . The psychologyof language: An introduction to psycholinguistics and generative grammer. New York : McGraw - Hill .

Bibliography

512

Fodor , J. A ., Garrett , M . F., Walker , E. C. T ., & Parkes, C. H . ( 1980) . Against definitions. Cognition, 8, 263- 367.

Folkins , J. W ., & Abbs, J. H . ( 1975) . Lip and motor control during speech: Respons to resistive loading of the jaw . Journal of Speechand Hearing Research, 18, 207- 220. ' Ford , M . ( 1982) . Sentenceplanning units: Implications for the speakers representat . In J. Bresnan (Ed.), The of meaningful relations underlying sentences mental representationof grammatical relations. Cambridge, MA : MIT Press. Ford, M ., & Holmes, V. M . ( 1978). Planning units in sentenceproduction. Cognition , 6, 35- 53. Foss, D . J., & Hakes, D . T . ( 1978) . Psycholinguistics. An introduction to the psychologyoflanguage. Englewood Cliffs , NJ: Prentice- Hall . Fowler, C. A . ( 1980) . Coarticulation and theories of extrinsic timing . Journal of Phonetics, 8, 113- 133. Fowler, C. A . ( 1983) . Realism and unrealism: a reply. Journal of Phonetics, 11, 303- 322. Fowler, C. A . ( 1986a). An event approach to the study of speechperception from a direct-realist perspective. Journal of Phonetics, 14, 3- 28. Fowler, C. A . ( 1986b) . Reply to commentators. Journal of Phonetics 14, 149- 170. " " " ' " Fowler, C. A ., & Housum, J. ( 1987) . Talkers signalling of new and old ' words in speechand listeners perception and use of the distinction . Journal of Memory and Language, 26, 489- 504.

Fowler, C. A ., Rubin , P., Remez, R. E., & Turvey , M . T . ( 1980) . Implications for speechproduction of a general theory of action. In B. Butter worth ( Ed.), Language production: Vol. 1. Speechand talk . London: Academic Press. Fowler, C. A ., & Turvey, M . T . ( 1980) . Immediate compensation in bite- block speech. Phonetica, 73, 306- 326.

Francik , E. P., & Clark , H . H . ( 1985) . How to make requeststhat overcome obstacles to compliance. Journal of Memory and Language. 24, 560- 568. Franck, D . ( 1980) . Seven sins of pragmatics: Theses about speech act theory, conversational analysis, linguistics and rhetorics. In A . Foolen, J. Hardeveld, & . Nijmegen: Xeno. D . Springorum (Eds.), Conversatieanalyse . Ober Vergessen Freud, S. ( 1904). Zur Psychopathologiedes A//tagslebens , Versprechen, Vergreifen, Aberglaubeund [rrtum. Frankfurt am Main : Fischer, 1954. (An English translation of Chapter V : Slips of the tongue, appeared in V . A . Fromkin ( Ed.) . ( 1973) . Speecherrors as linguistic evidence.The Hague: Mouton ) .

Friederici. A ., & Levelt, W . J. M . ( 1986) . Cognitive processes of spatial coordina , 73, 455- 458. assignment. Naturwissenschaften

513 ). Simulation of human long-term memory. Psychological Frijda, N. ( 1972 Bulletin, 77, 1 31. . Fromkin, V. A. ( 1971 ). The non-anomalousnature of anomalousutterances . errors . 1973 Ed. V. A. Fromkin in: . 27 52 47 ) Speech ( ) ( , , (Reprinted Language . TheHague: Mouton). aslinguisticevidence . The Hague: Fromkin, V. A. (Ed.). ( 1973 ). Speecherrorsas linguisticevidence Mouton. . Slipsof thetongue , Fromkin, V. A. (Ed.). ( 1980 ). Errorsin linguisticperformance . ear, pen, andhand. NewYork: AcademicPress ). The linguisticevidenceof speecherrors. BRNO Studiesof English Fry, D. ( 1969 74. 8 69 , ,

subsequent recall . Memory and Cognition , 1, 213 216.

. R. Ladd ). A generativemodelof intonation. In ACutlerD Garding, E. ( 1983 . : . measurements Models and : Eds . Heidelberg Springer ), Prosody ( andnaturalGarfield, J. L. (Ed.). ( 1987 representation ). Modularityin knowledge . . Cambridge , MA: MIT Press understanding language Garnham, A., Shillcock, R. S., Brown, G. D. A., Mill , A. I. D., & Cutler, A. ). Slips of the tonguein the London-Lund corpus of spontaneousconversations ( 1982 . Berlin: . In A. Cutler(Ed.), Slipsof thetongueandlanguage production Mouton. , S. M., & Dell, G. S. ( 1984). Someneurolinguisticimplicationsof preGarnsey , 23, 64- 73. articulatoryeditingin production. BrainandLanguage Garrett, M. F. ( 1975 ). The analysisof sentenceproduction. In G. Bower : Vol. 9. New York: Academic Ed. ( ), Psychologyof learningand motivation . Press esin sentence Garrett, M. F. ( 1976 production. In R. J. Wales ). Syntacticprocess : . Amsterdam es to languagemechanisms & E. Walker (Eds.), New approach North Holland.

514

Garrett , M . F. ( 1982) . Production of speech: Observations from normal and pathologicallanguage use. In A . W . Ellis ( Ed.), Normality and pathology in cognitive functions. London : Academic Press. Garrod , S., & Anderson, A . ( 1987) . Saying what you mean in dialogue: A study in conceptualand semanticcoordination . Cognition, 27, 181- 218. Garrod , S., & Sanford, A . ( 1983) . Topic dependent effects in language processing . In G. B. Flores d' Arcais & R. J. Jarvella ( Eds.), The processof language . Chichester: John Wiley . understanding Gay, T . ( 1981) . Mechanisms in the control of speechrate. Phonetica, 38, 148 158. Gay, T ., & Turvey , M . T . ( 1979) . Effects of efferent and afferent interference on speechproduction : Implications for a generative theory of speech-motor control . Copenhagen: Institute of . 9th International Congressof Phonetic Sciences . Phonetics Gazdar, G. ( 1979). Pragmatics: Implicature, presupposition, and logical form . New York : Academic Press. Gazdar, G., Klein , E., Pullum, G., & Sag, I . ( 1985) . Generalizedphrasestructure grammar. Oxford : Blackwell. Gee, J. P., & Grosjean, F. ( 1983). Performancestructures: A psycholinguistic and linguistic appraisal. CognitivePsychology, 15, 411- 458.

Glaser, W. R., & Dungelhoff, F. J. ( 1984) . The time course of picture-word interference . Journal of Experimental Psychology: Human Perception and Performance , 10, 640- 654. Goffman , E. ( 1981) . Formsof talk. Oxford : Blackwell. Goguen, J. A ., & Linde , C. ( 1983) . Linguistic methodologyfor the analysisof aviation accidents. ( NASA Contractor Report 3741) .

Goldman , N . ( 1975) . Conceptual generation. In R. Schank ( Ed.), Conceptualinformatio processing.Amsterdam: Nnrth -Hnlland . of words in and the predictability Goldman - Eisler , F . ( 1958 ) . Speech production context. Quarterly Journal of ExperimentalPsychology, 10, 96- 106. Goldman-Eisler, F. ( 1967) . Sequential temporal patterns and cognitive processes in speech. Languageand Speech, 10, 122- 132.

515 Goldman-Eisler, F. ( 1968) . Psycholinguistics: Experimentsin spontaneousspeech. New York : Academic Press. Good , D . A ., & Butter worth, B. ( 1980) . Hesitancy as a conversational resource: Somemethodological implications . In H .W. Dechert & M . Raupach ( Eds.), Temporal variablesin speech.The Hague: Mouton . Goodglass, H ., Theurkauf , J. C., & Wingfield , A . ( 1984) . Naming latencies as evidence for two modes of lexical retrieval. Applied Psycholinguistics, 5, 135146. Goodwin , C. ( 1981) . Conversationalorganization: Interaction betweenspeakers and hearers. New York : Academic Press. Gordon , H ., & Meyer, D . E. ( 1987). Hierarchical representation of spoken syllable order. In A . Allport , D . G. MacKay , W . Prinz, & E. Scheerer(Eds.), Language perception: Relations between listening, speaking, reading and writing . London: Academic Press. Grice, H . P. ( 1957) . Meaning. PhilosophicalReview, 66, 377- 388. ' Grice, H . P. ( 1968) . Utterer s meaning, sentencemeaning and word meaning. 4, 225- 242. Vol. Foundationsof Language, Grice, H . P. ( 1975). Logic and conversation. In P. Cole & J. L . Morgan ( Eds.), Syntax and semantics: 3. Speechacts. New York : Academic Press. Grice, H . P. ( 1978) . Some further notes on logic and conversation. In P. Cole ( Ed.), Syntax and Semantics: Vol. 9. Pragmatics. New York : Academic Press. Grimes, J. E. ( 1975) . The thread of discourse.The Hague: Mouton . ? Prediction and prosody in the online Grosjean, F. ( 1983) How long is a sentence processingof language. Linguistics, 21, 501 529. Grosjean, F., Grosjean, L ., & Lane, H . ( 1979). The patterns of silence: Performance structures in sentenceproduction . Cognitive Psychology, 11, 58- 81. Grosz, B. J. ( 1981) . Focussing and description in natural languagedialogues. In A . K . Joshi, B. L . Webber, & I . A . Sag (Eds.), Elementsof discourseunderstanding . Cambridge University Press. Grosz, B. J., Joshi, A ., & Weinstein, S. ( 1983). Providing a unified account ofdefi nite noun phrasesin discourse. Proceedingsof the 21st Annual Meeting of the Associationfor ComputationalLinguistics. Association for Computational Linguistics. Grosz, B. J., & Sidner, C. L. ( 1985) . Discourse structure and the proper treatment of interruptions . Proceedingsof the International Joint Conferenceon Arti ficiallntelligence . Los Altos , CA : Morgan Kaufman . Gruber , J. S. ( 1965) . Studies in lexical relations. Bloomington : Indiana Linguistics Club. Gumperz, J. J., & Hymes, D . ( Eds.) . ( 1972) . Directions in sociolinguistics. New York : Holt , Rinehart and Winston.

Bibliography

516

Gussenhoven, C. ( 1984) . On the grammar and semantics of sentenceaccents. Dordrecht: Foris. Hale, K . ( 1981). On the position of Warlpiri in a typology of the base. Bloomington: Indiana Linguistics Club. Halle, M ., & Mohanan, K . P. ( 1985) . Segmentalphonology of modem English. Linguistic Inquiry , 16, 57- 116. Halle, M ., & Stevens, K . N . ( 1964) . Speechrecognition: A model and a program for research. In J. A . Fodor & J. J. Katz (Eds.), The structure of language: Readingsin thephilosophyof language. Englewood Cliffs , NJ: Prentice- Hall . Halliday , M . A . K . ( 1967a). Notes on transitivity and theme in English: II . Journal of Linguistics, 3, 199- 244. Halliday , M . A . K . ( I 967b) . Intonation and grammar in British English. The Hague: Mouton . Halliday , M . A . K . ( 1970) . A coursein spoken English: Intonation. Oxford University Press. Halliday, M . A . K ., & Hassan, R. ( 1976). Cohesionin English. London: Longman. Hammarberg, R. ( 1976) . The metaphysicsof coarticulation . Journal of Phonetics, 4, 353- 363. Hammarberg, R. ( 1982) . On redefining coarticulation . Journal of Phonetics, 10, 123- 137. Hankamer, J. (forthcoming) . Morphological parsing and the lexicon. In W . D . Marslen-Wilson (Ed.), Lexical representationand process. Cambridge, MA : MIT Press.

Harley, T. A. ( 1984 ). A critique of top- down independentlevelsof speechproduction : Evidencefrom non-plan-internal speecherrors. CognitiveScience , 8, 191- 219. Hart, J. 't , & Collier, R. ( 1975 . ). Integratingdifferentlevelsof intonationanalysis Journalof Phonetics , 3, 235- 255. Hawkins, J. ( 1983 . NewYork: AcademicPress . ). Wordorderuniversals Hawkins, J. ( 1984 . ) Modifier head or functionargumentrelations in phrase structure?Lingua, 63, 107- 138. Hawkins, P. R. ( 1971 . Language ). The syntacticlocationof hesitationpauses and , 14, 277- 288. Speech Hayes, B. ( 1986). Inalterability in CV Phonology. Language, 62, 321- 351. Heeschen, C., Ryalis, J., & Hagoort, P. (forthcoming) . Psychological stress in Broca' s versusWernicke' s aphasia. Journal of Clinical Linguistics and Phonetics. Henderson, A ., Goldman -Eisler, F., & Skarbek, A . ( 1966). Sequential temporal patterns in spontaneousspeech. Languageand Speech, 9, 207- 216.

517 Hernnann , T . ( 1983). Speechand situation: A psychologicalconceptionof situated speaking. Heidelberg: Springer. Hernnann , T ., Burkle , B., & Nirmaier , H . ( 1987). Zur horerbezogenenRaumref. ( Bericht 12) . University of Mann erenz: Horerposition und Lokalisationsaufwand Helm, ForschergruppeSprechenund Sprachverstehenim sozialenKontext . . Bern: Herrmann, T ., & W. Deutsch ( 1976) . Psychologieder Objektbenennung Huber. Hill , C. ( 1982). Up/ down , front / back, left/right : A contrastive study of Hausa and English. In J. Weissenborn & W. Klein (Eds.), Here and there: Crosslinguistic studieson deixis and demonstratum. Amsterdam: John Benjamns. Hockett , C. F. ( 1967) . Where the tongue slips there slip I . To honor Roman Jakobson: Vol. 2. The Hague: Mouton . Hoenkamp, E. ( 1983). Een computermodel van de spreker: Psychologischeen lin guistischeaspecten.Unpublished doctoral dissertation, Nijmegen University . Holmes, V. M . ( 1984) . Sentenceplanning in a story continuation task. Language and Speech,27(2), 115- 134. Holmes, V. M . (forthcoming). Hesitations and sentenceplanning. Languageand Cognitive Processes. Hopper, P. I. ( 1979). Aspectand foregrounding in discourse. In T . Givon ( Ed.), Syntax and Semantics: Vol. / 2. Discourseandsyntax. New York : AcademicPress. Hotopf , W. H . N . ( 1980) . Semantic similarity as a factor in whole-word slips of the tongue. In V. A . Fromkin ( Ed.), Errors in linguistic performance: Slips of the tongue, ear, pen, and hand. New York : Academic Press. Humphreys, G. W ., Riddoch , M . J., & Quinlan, P. T . ( 1988) . Cascadeprocesses in picture identification . Cognitive Neuropsychology , 5, 67- 104. Huttenlocher, J., & Kubicek , L . F. ( 1983) . The source of relatednesseffects on naming latency. Journal of Experimental Psychology: Learning, Memory and Cognition, 9, 486- 496. Isaacs, E. A . & Clark , H . H . ( 1987) . Referencesin conversation betweenexperts and novices. Journal of ExperimentalPsychology: General, / / 6, 26- 37. Jackendoff, R. ( 1972) . Semantic interpretation in generative grammar. Cambridge , Mass: MIT Press. Jackendoff, R. ( 1977) . X Syntax: A study of phrase structure. Cambridge, MA : MIT Press. Jackendoff, R. ( 1983). Semanticsand cognition. Cambridge, MA : MIT Press. Jackendoff, R. ( 1987a). On Beyond Zebra: The relation of linguistic and visual information . Cognition, 26, 89- 114. Jackendoff, R. ( 1987b) . Consciousnessand the computational mind. Cambridge, MA .: MIT Press.

Bibliography

518

Jaffe, J., Breskin, S., & Gerstman, L. J. ( 1972). Random generation of apparent speechrhythms. Languageand Speech, 15, 68- 71. James, D . ( 1972) . Some aspects of the syntax and semantics of interjections. Papersfrom the Eighth RegionalMeeting. Chicago Linguistic Society.

James, D . ( 1973) . Another look at, say, somegrammatical constraints on, ok , interjectio and hesitations. Papersfrom the Ninth Regional Meeting. Chicago Linguistic Society. James, W. ( 1892) . Textbook of psychology. London: Macmillan . James, W. ( 1893) . Theprinciples of psychology: Vol. 1. New York : Holt . Jarvella, R. J, & HermanS . J. ( 1972). Clause structure of sentencesand speech , 11, 381- 384. processing. Perceptionand Psychophysics Jarvella, R. J., & Klein , W. (Eds.) . ( 1982). Speech, place and action. Studies in deixis and related topics. Chichester: John Wiley,

Jassem, W ., Hill , D . R., & Witten , I . H. ( 1984) . Isochrony in English speech: Its statistical validity and linguistic relevance. In D . Gibbon & H . Richter (Eds.), Intonation , Accent, and Rhythm. Berlin: De Gruyter . Jefferson, G. ( 1978) . Sequential aspects of story-telling in conversation. In J. Schenkein ( Ed.), Studies in the organization of conversationalinteraction. New York : Academic Press. Jefferson, G. ( 1985) . An exercisein the transcription and analysis of laughter. In T . A . van Dijk ( Ed.), Handbook of discourseanalysis: Vol. 3. London: Academic Press. Johnson-Laird , P. N . ( 1983) . Mental models. Cambridge University Press. Johnson-Laird , P. N ., & Garnham, A . ( 1980) . Descriptions and discourse models. Linguistics and Philosophy, 3, 371- 393. Jones, H . G. V ., & Langford, S. ( 1987) . Phonological blocking in the tip of the tongue state. Cognition, 26, 115- 122. Jorgensen, J., Miller , G. A ., & Sperber, D . ( 1984) . Test of the mention theory of irony . Journal of Experimental Psychology: General, 113, 112- 120. Kaisse, E. M . ( 1985). Connectedspeech: The interaction of syntax and phonology. New York : Academic Press. Kamp , H . ( 1984) . A theory of truth and semantic representation. In J. Groenendijk , Th . Jansen, & M . Stokhof (Eds.), Truth , interpretation and information. Dordrecht: Foris. Kao , H . S. R., Galen, G. P. van, & Hoosain, R. ( Eds.) . ( 1986) . Graphonomics : Contemporaryresearchin handwriting. Amsterdam: North - Holland . Keele, S. W. ( 1981) . Behavioral analysisof movement. In V . Brooks ( Ed.), Handbook of physiology. The nervoussystem: Vol. 2. Motor control. Bethesda, MD : American Physiological Society.

Bibliography

519

" " Keenan, E. L . ( 1976). Toward a universal definition of subject . In C. N . Li . Press ( Ed.), Subjectand topic. New York : Academic Keenan, J., MacWhinney, B., & Mayhew, D . ( 1977) . Pragmatics in memory: A study of natural conversation. Journal of Verbal Learning and Verbal Behavior, 16, 549- 560. Kelly , M . H ., & Bock, J. K . ( 1988). Stressin time. Journal of Experimental Psychology : Human Perceptionand Performance, 14, 389- 403. Kelso, J. A . S., Holt , K . G ., Kugler , P. N ., & Turvey , M . T . ( 1980) . On the concept of corrdinative structures as dissipative structures: II . Empirical lines of convergence . In G. E. Stelmach & J. Requin (Eds.), Tutorials in motor behavior. Amsterdam: North - Holland . Kelso, J. A . S, Saltzman, E. L., & Tuller , B. ( 1986) . The dynamical perspectiveon speechproduction data and theory . Journal of Phonetics, 14, 29 59. " " Kelso, J. A . S., Tuller , B., & Harris , K . S. ( 1983) . A dynamic pattern perspective on the control and coordination of movement. In P. F. MacNeilage ( Ed.), The production of speech.New York : Springer. Kelso, J. A . S., Tuller , B., & Harris , K . S. ( 1986) . A theoretical note on speech timing . In J. S. Perkell & D . Klatt ( Eds.), In varianceand variability in speechpro . Hillsdale, NJ: LawrenceErlbaum. cesses Kelso, J. A . S., Tuller , B., Vatikiotis , E., & Fowler , C. A . ( 1984). Functionally specificarticulatory cooperation following jaw perturbations during speech: Evidence for coordinative structures. Journal of Experimental Psychology: Human Perceptionand Performance, 10, 812- 832. Kempen, G. ( 1987). A framework for incremental syntactic tree formation . Proceedings of the Tenth International Joint Conferenceon Artificial Intelligence. Los Altos , CA : Morgan Kaufmann . Kempen, G. ( 1988) . Language generation systems. In I . Batori , W . Lenders, & W . Putschke(Eds.), Computationallinguistics: An international handbookon computer orientedlanguageresearchand applications. Berlin: de Gruyter . Kempen, G., & Hoenkamp, E. ( 1982) . Incremental sentencegeneration: Implications for the structure of a syntactic processor. In J. Horecky ( Ed.), Proceedings of the Ninth International Conference on Computational Linguistics. Amsterdam: North -Holland . Kempen, G., & Hoenkamp, E. ( 1987) . An incremental procedural grammar for sentenceformulation . CognitiveScience, 11, 201- 258. Kempen, G., & Huijbers , P. ( 1983) . The lexicalization process in sentenceproduction and naming: Indirect election of words. Cognition, 14, 185- 209. Kenstowicz, M ., & Kisseberth, C. ( 1979) . Generativephonology. New York : Academic Press. Kent , R. D . ( 1986) . Is a paradigm change needed? Journal of Phonetics, 14, 111- 115.

Bibliography

520

Kintsch , W . ( 1974) . The representation of meaning in memory. Hillsdale, NJ: LawrenceErlbaum. Kintsch , W., & Bates, E. ( 1977) . Recognition memory for statements from a classroom lecture. Journal of Experimental Psychology: Human Learning and Memory, 3, 150- 168. Kiparsky , P. ( 1982). From cyclic phonology to lexical phonology . In H . van der Hulst ( Ed.), The structure of phonological representations : Part I . Dordrecht: Foris. Klapp , S. T . ( 1974). Syllable-dependent pronounciation latencies in number naming, a replication. Journal of Experimental Psychology, 102, 1138- 1140. Klapp , S. T ., Anderson, W. G., & Berrian, R. W. ( 1973) . Implicit speechin reading , reconsidered. Journal of Experimental Psychology, 100, 368- 374. Klapp , S. T ., & Erwin , C. I . ( 1976) . Relation between programming time and duration of the responsebeing programmed. Journal of Experimental Psychology : Human Perceptionand Performance, 2, 591- 598.

Klapp , S. T ., & Wyatt, E. P. ( 1976). Motor programming within sequencesof response . Journal of Motor Behavior, 8, 19- 26. Klatt , H . ( 1975) . Vowel lengthening is syntactically determined in a connected discourse. Journal of Phonetics, 3, 129- 140. Klein , W. ( 1978). Wo ist hier: Praliminarien zu einer Untersuchung der lokalen Deixis. LinguistischeBerichte, 58, 18- 40. Klein , W . ( 1979) . Wegauskunfte. Zeitschrift fUr Literaturwissenschaftund Lin guistik, 9, 9- 57. Klein , W. ( 1982). Local deixis in route directions. In R. Jarvella & W. Klein ( Eds.), Speech, place, and action: Studies in deixis and related topics. Chichester: John Wiley . Klein , W. ( 1984) . Buhler' s Ellipse. In C. F. Grauman & Th . Herrmann ( Eds.), Karl Biihlers Axiomatik . Funfzig Jahre Axiomatik der Sprachwissenschaften . Frankfurt : Klosterman. Klein , W., & Stutterheim, C. von ( 1987) . Quaestio und referentielle Bewegungin Erzihlungen . LinguistischeBerichte, / 09, 163- 183. Kloker , D . ( 1975). Vowel and sonorant lengthening as cues to phonological phrase boundaries. Journal of the Acoustical Society of America, 57 (Suppl. I ), 33- 34. Kohler , K . J. ( 1986) . Invariance and variability in speechtiming : From utterance to segment in German. In J. S. Perkell & D . Klatt ( Eds.), Invariance and variability in speechprocesses. Hillsdale, NJ: LawrenceErlbaum. Kohn , S. E., Wingfield, A ., Menn, L ., Goodglass, H ., Berko Gleason, J., & Hyde, M . ( 1987). Lexical retrieval: The tip of the tongue phenomenon. Applied Psycholinguistics,8, 245- 266.

Bibliography

521

" " Koriat , A ., & Lieblich , I . ( 1974) . What does a person in a TOT state know that ' " ' " a person in a don t know state doesn t know. Memory and Cognition, 2, 647 655. Koriat , A ., & Lieblich , I . ( 1977) . A study of memory pointers. Acta Psychologica, 41, 151- 164. Kosslyn, S. M . ( 1980) . Image and mind, Cambridge, MA : Harvard University Press. ' Kowal , S., Bassett, M . R., & O Connell, D . C. ( 1985) . The spontaneity of media Journal , 14, 1- 18. interviews. of PsycholinguisticResearch Kozhevnikov , V . A ., & Chistovich, L . A . ( 1965) . Speech: Articulation and Perception . Cl earing housefor Federal Scientific and Technical Information . Joint Publications Research Service. Washington, DC : U .S. Department of Commerce . Krauss, R. M ., & Weinheimer, S. ( 1964) . Changesin referencephrasesas a function of frequency of usage in social interaction: A preliminary study. Psychonomic Science, 1, 113- 114. Kroch , A ., & Hindle , D . ( 1982) . On the linguistic character of non-standard input . Proceedingsof the 20th Annual Meeting of the Association for Com . for Association . Computational Linguistics putational Linguistics Kucera, H ., & Francis, W . N . ( 1967) . Computational analysis of present-day AmericanEnglish. Providence: Brown University Press. LaBerge, D ., & Samuels, S. J. ( 1974) . Toward a theory of automatic information processingin reading. Cognitive Psychology, 6, 293 323. Labov, W. ( 1972) . Languagein the inner city . Philadelphia: University of Pennsylvania Press. Labov, W ., & Fanshel, D . ( 1977) . Therapeuticdiscourse, New York : Academic Press. Lackner, J. R., & Tuller , B. H . ( 1979) . Role ofefference monitoring in the detection . of self-produced speecherrors. In WE Cooper & E. C. T . Walker ( Eds.), Sentenceprocessing: Psycholinguisticstudiespresented to Merrill Garrett. Hills dale, NJ : Lawrence Erlbaum. Ladd , D . R. ( 1980) . The structure of intonational meaning. Bloomington : Indiana University Press. Ladd, D . R. ( 1984) . Declination: A review and some hypotheses. Phonology Yearbook, J, 53- 74. Ladd , D . R. ( 1986) . Intonational phrasing: The casefor recursiveprosodic structure . Phonology Yearbook, 3, 311- 340. Ladd, D . R., Scherer, K ., & Silverman, K . ( 1986) . An integrated approach to studying intonation and attitude. In C. Johns Lewis ( Ed.), Intonation in discourse . London : Croom Helm.

Bibliography

522

Lakoff, R. ( 1970 . Language ). Tenseandits relationsto participants , 46, 838- 849. Laver, J. D. M. ( 1973 ). The detectionand correctionof slipsof the tongue. In V. A. Fromkin(Ed.), Speech errorsaslinguisticevidence . TheHague: Mouton. Laver, J. D. M. ( 1980 . in the ) Monitoring systems neurolinguisticcontrol of speechproduction. In V. A. Fromkin (Ed.), Errors in linguisticperformance : , ear, pen, andhand. NewYork: AcademicPress Slipsof thetongue . Lee, B. S. ( 1950 . Journalof the Acoustical ). Effectsof delayedspeechfeedback Societyof America,22, 824- 826. Lehiste,I. ( 1970 als. Cambridge ). Suprasegment . , MA: MIT Press Lehiste, I. ( 1975 . In A. Cohen& S. G. ). The phoneticstructureof paragraphs Nooteboom (Eds.), Structure and processin speechperception . Heidelberg : Springer. Lehiste,I. ( 1977 . Journalof Phonetics ). Isochronyreconsidered , 5, 253- 263. , E. H. ( 1967 Lenneberg ). Biologicalfoundations . New York: John of language Wiley. Levelt, W. J. M . ( 1974) . Formal grammars in linguistics and psycho/inguistics. (3 Vols.) . The Hague: Mouton .

Bibliography

523

Levelt, W. J. M., & Kempen, G. ( 1975 ). Semanticand syntacticaspectsof remembering . In A. Ken: A reviewof somerecentcontinentalresearch sentences . : John . London term in Wiley . memory nedy& A. Wilkes(Eds), Studies long Levelt, W. J. M., & Maassen , B. ( 1981 ). Lexicalsearchand order of mentionin W. J. M. Levelt(Eds.), Crossingtheboundaries & W. Klein . In sentence production . Dordrecht: Reidel. to ManfredBierwisch . Studiespresented in linguistics ). Pointingand voicingin Levelt, W. J. M., Richardson, G., & La Hey, W. ( 1985 24 and . Journalof Memory Language , , 133- 164. deicticexpressions , E. ( 1978 ). Structureand useof Levelt, W. J. M., Schreuder , R., & Hoenkamp in the . Recentadvances Eds T. Smith P. & R. N. In ), motion. ( of verbs Campbell . Press Plenum York: . New of language psychology . In G. Kempen Levelt, W. J. M., & Schriefers , H. ( 1987 ). Stagesof lexicalaccess , psychology . Newresultsin artificial intelligence (Ed.), Naturallanguagegeneration . Dordrecht: MartinusNijhoff. andlinguistics ). Ambiguoussurface , G. R. E. ( 1970 Levelt, W. J. M., Zwanenburg , W., & Ouweneel , 6, 260- 273. of Language structureandphoneticform in French. Foundations . . CambridgeUniversityPress Levinson,S. C. ( 1983 ). Pragmatics , similarity, Levitt, A. G., & Healy, A. F. ( 1985 ). The rolesof phonemefrequency and availability in the experimentalelicitation of speecherrors. Journal of , 24, 717- 733. MemoryandLanguage : A philosophicalstudy. Cambridge , MA: Harvard Lewis, D. K. ( 1969 ). Convention . Press University , D. S., & Studdert-Kennedy, M. Liberman, A. M., Cooper, F. S., Shankweiler Review code. Psychological , 74, 431- 461. ). Perceptionof thespeech ( 1967 Liberman, M., & Pierrehumbert , J. ( 1984 ). Intonational invariance under & R. T. Oehrle(Eds.), Language M. Aronoff changesin pitch rangeand length. In to Morris Halle by his : Studiesin phonologypresented andsoundstructure . . Cambridge teacherandstudents , MA: MIT Press Lieberman , A ( 1985 ). Measuresof intonationof read , P., Katz W., & Jongman and spontaneous speechin AmericanEnglish. Journalof the AcousticSocietyof America,77, 649- 657. . (SpeechTransmissionLaboratoryReport Lindblom, B. ( 1963 ). On vowelreduction . 29). Stockholm:RoyalInstituteof Technology . (Quarterly Lindblom, B. ( 1968 ). Temporalorganizationof syllableproduction . of Institute : . 2 Stockholm Technology 3 Status and Royal ) Report Progress . In P. F. MacNeilage(Ed.), Lindblom, B. ( 1983 ). Economyof speechgestures . NewYork: Springer. Theproductionof speech of somefixedLindblom, B., Lubker, J., & Gay, T. ( 1979 ). Formantfrequencies simulation motor of model and a by predictive mandiblevowels programming speech . Journalof Phonetics , 7, 147- 161.

Bibliography

524

Lindblom B., & MacNeilage, P. ( 1986) . Action theory: Problems and alternative approaches. Journal of Phonetics, 14, 117- 132. Linde, C., & Goguen, J. ( 1978) . The structure of planning discourse. Journal of Social and Biological Structures, 1, 219- 251.

Linde, C., & Labov , W. ( 1975) . Spatial networks as a site for the study of language and thought . Language, 51, 924- 939. Lindsley, J. R. ( 1975). Producing simple utterances: How far do we plan? Cognitive Psychology, 7, 1- 19. Lindsley, J. R. ( 1976). Producing simple utterances: Details of the planning process . Journal of PsycholinguisticResearch,5, 331- 351. Linell , P. ( 1979). Psychological reality in phonology. Cambridge University Press. Longuet-Higgins, H . C. ( 1987). Mental Processes. Cambridge, MA : MIT Press.

Lounsbury, F. G. ( 1954). Transitional probability , linguistic structure, and systems of habit-family hierarchies. In CE . Osgood & T . Sebeok ( Eds.), Psycholing : A survey of theory and researchproblems. Bloomington : Indiana University Press. Lyons, J. ( 1977) . Semantics(2 Vols.) . Cambridge University Press. Lyons, J. ( 1981). Language, meaningand context. Bungay: Fontana Press. MacKay, D . ( 1970) . Spoonerisms: The structure of errors in the serial order of speech. Neuropsychologia , 8, 323- 350 (Reprinted in V. A . Fromkin ( Ed.) . ( 1973) . errors as . The Hague: Mouton .) Speech linguistic evidence MacKay , D. ( 1972). The structure of words and syllables: Evidence from errors in speech. CognitivePsychology, 3, 210- 227. MacKay , D . ( 1982) . The problem of flexibility , fluency, and speed-accuracy trade-off in skilled behavior. PsychologicalReview, 89, 483- 506. MacKay , D . ( 1987). The organization of perception and action: A theoryfor language and other cognitiveskills. New York : Springer. Maclay, H ., & Osgood, C. E. ( 1959) . Hesitation phenomena in spontaneous English speech. Word, 15, 19- 44.

, P. F. Motor control and serialorderingof speech MacNeilage . Psychological Review , 1970 , 77, 182- 196. , P. F. ( 1980 MacNeilage ). Emergingconceptsin speechcontrol. In G. S. Stelmach& J. Requin(Eds.), Tutorialsin motor behavior . Amsterdam : North-

Holland .

MacNeilage, P., &. Ladefoged, P. ( 1976) . The production of speechand language. In: E. c . Carterette &. M . P. Friedman ( Eds.), Handbook of perception: Vol. 7. New York : Academic Press.

Bibliography

525

, 53, 152- 168. , B. ( 1977 ). Startingpoints. Language MacWhinney ). Sententialdevicesfor conveyinggivenness , B., & Bates, E. ( 1978 MacWhinney cross cultural : A study. Journalof VerbalLearning and newness developmental andVerbalBehavior , / 7, 539- 558. ). The impact of cue validity on , B., Bates, E., & Kiegl, R. ( 1984 MacWhinney sentence interpretationin English, German, and Italian. Journalof VerbalLearning andVerbalBehavior , 23, 127- 150. Makkai, A. ( 1972 ). Idiomstructurein English.TheHague: Mouton. Malinowski, B. ( 1920 ). Classificatoryparticlesin the languageof Kiriwina. , / , 33- 78. (LondonInstitution). Bulletinof theSchoolof OrientalStudies ObFaktorenbeim Verstehen iiberspezifizierter ). Sensorische Mangoid, R. ( 1986 . Frankfurt: PeterLang. jektbennennungen , / 7, Marshall, J. C. ( 1984). Multiple perspectiveson modularity. Cognition 209 242. ). The temporalstructure Marslen-Wilson, W., & KomisarjevskyTyler, L. ( 1980 . Cognition , 8, 1- 71. of spokenlanguage understanding ). Producinginterpretable Marslen-Wilson, W., Levy, E., & KomisarjevskyTyler, L. ( 1982 . In R. J. of reference and maintenance : The establishment discourse and related in deixis : Studies and action , place, Jarvella& W. Klein (Eds.), Speech . : John . Chi chester Wiley topics esis speechunderstanding Marslen-Wilson, W., & Tyler, L. ( 1981 ). Centralprocess Transactions . Philosophical of theRoyalSocietyLondon,B259, 317- 332. Martin, J. G., & Strange , W. ( 1968). The perceptionof hesitationin spontaneous andPsychophysics . Perception , 3, 427- 432. speech . Hillsdale, NJ: Lawrence McNeill, D. ( 1979 ). The conceptualbasisof language Erlbaum. . Cambridge : A newapproach , MA: Harper McNeill, D. ( 1987 ). Psycholinguistics andRow. . Berlin: Behr. ). AusdemLebenderSprache Meringer, R. ( 1908 . Stuttgart: Gound Verlesen Meringer, R., & Mayer, K. ( 1895). Versprechen and D. A. A. Cutler schensche , with introductoryessayby Verlag. (Re issued . : JohnBenjamins) ). Amsterdam Fay( 1978 : A primingstudy. production encodingin language ). Phonological Meyer, A. ( 1988 . dissertation doctoral , University Nijmegen Unpublished 1985 P. C. Gordon E. & D. ). Speechproduction: Motor programming , ( , Meyer, . Journalof MemoryandLanguage , 24, 3- 26. of phoneticfeatures , plus or minustwo: Somelimits Miller, G. A. ( 1956 ). The magicalnumberseven Review information. Psychological , 63, 81- 97. on our capacityfor processing

Bibliography

526

Miller, G. A. ( 1969 . ). A psychologicalmethod to investigateverbal concepts Journalof MathematicalPsychology , 6, 169- 191. Miller, G. A. ( 1978 ). Semanticrelationsamongwords. In M. Halle, J. Bresnan , & G. A. Miller (Eds.), Linguistictheoryand psychological , reality. Cambridge MA: MIT Press .

Miller, G. A. ( 1982 . In ). Someproblemsin the theoryof demonstrative reference R. J. Jarvella& W. Klein (Eds.), Speech . place. andaction: Studiesin deixisand relatedtopics.Chichester : JohnWiley. -Laird, P. N. ( 1976 Miller, G. A., & Johnson . Cambridg ). Languageandperception . , MA: HarvardUniversityPress Mittelstaedt, H. ( 1983 ). A new solution to the problemof subjectivevertical. , 70, 272- 281. Naturwissenschaften Mohanan, K . P. ( 1982). Grammatical relations and clause structure in Ma layalam. In J. Bresnan (Ed.), The mental representationof grammatical relations. Cambridge, MA : MIT Press.

Moll , K . L ., & Daniloff , R. G. ( 1971) . Investigation of the timing of velar movements during speech. Journal of the AcousticalSocietyof America, 750, 678- 684.

Montague, R. ( 1974). Formal philosophy: Selectedpapers. New Haven: Yale University Press.

in linguisticperformance : Slipsof thetongue , ear, pen, andhand. New York: Academi Press . Motley, M. T., Baars, B. J., & Camden , C. T. ( 1981 ). Syntacticcriteriain prearticulatoryediting: Evidencefrom laboratory-inducedslipsof the tongue. Journal Research of Psycholinguistic , 10, 503- 522. Motley, M. T., Camden , C. T., & Baars, B. J. ( 1982 ). Covert formulationand editingof anomaliesin speechproduction: Evidencefrom experimentally elicited slipsof thetongue.Journalof VerbalLearningandVerbalBehavior , 21, 578- 594. Munro, A. ( 1977 . Unpublisheddoctoraldissertation ). Speechact understanding , Universityof California, SanDiego. Nakatani, L. H., O' Connor, J. D., & Aston, C.H. ( 1981 ). Prosodicaspectsof AmericanEnglishspeech , 38, 84- 106. rhythm. Phonetica

Neilson, M. D., & Neilson, P. D. ( 1987 ). Speechmotor control and stuttering: A -motor processing . SpeechCommunication computationalmodelof adaptivesensory , 6, 325- 333.

527 Nespor, M ., & Vogel, I. ( 1986) . Prosodicphonology. Dordrecht: Foris. Newell, A ., & Simon; H . A . ( 1972) . Human problem solving. Englewood Cliffs , NJ: Prentice-Hall . Nolan , F. J. ( 1982) . The role of action theory in the description of speechproduction . Linguistics, 20, 287- 308. Noordman , L. G. M . ( 1979) . lnferringfrom language. Heidelberg: Springer. Noordman -Vonk , W. ( 1979) . Retrieval from semantic memory. Heidelberg: Springer. Nooteboom , S. ( 1967). Some regularities in phonemic speecherrors. Annual Progress Report. Institutefor PerceptionResearchIPO , 2, 65 70. Nooteboom , S. ( 1972) . Production and perception of vowelduration. Unpublished doctoral dissertation, Utrecht University . Nooteboom , S. ( 1973) . The tongue slips into patterns. In V. Fromkin (Ed.), Speecherrors as linguistic evidence.The Hague: Mouton . Nooteboom, S. ( 1980) . Speaking and unspeaking: detection and correction of phonological and lexical errors in spontaneousspeech. In V. A . Fromkin ( Ed.), Errors in linguistic performance. New York : Academic Press. Nooteboom, S., & Terken, J. M . B. ( 1982) . What makes speakersomit pitch accents ? An experiment. Phonetica, 39, 317- 336. NormanD . A ., & Rumelhart, D . E. ( 1975) . Explorations in cognition. San Francisco: Freeman. Nunberg, G. ( 1979) . The non-uniquenessof semantic solutions: Polysemy. Linguistics and Philosophy, 3, 143- 184. <>Chs-Keenan, E. ( 1976). The universality of conversational implicature . Language in Society, 5, 67- 80. O' Connell, D ., & Kowal , S. ( 1983). Pausology. In W. A . Sedelow& S. Y . Sedelow (Eds.), Computersin LanguageResearch: Part I . Formalization in literary and discourse analysis. Berlin: Mouton . ' O Connor , J. D ., & Arnold , G. F. ( 1973) . Intonation of colloquial English. (2nd ed.) . London: Longman. Ohala, J. J. ( 1978) . Production of tone. In V. Fromkin ( Ed.), Tone: A linguistic survey. New York : Academic Press. Ohman, S. E. G. ( 1966) . Coarticulation in VCV utterances: Spectrographic measurements . lournalofthe Acoustical Societyof America, 39, 151- 168. Ojemann, G. A . ( 1982) . Interrelationships in the localization of language, memory, and motor mechanisms in human cortex and thalamus. In R. A . Thompson & J. R. Green ( Eds.), New perspectivesin cerebral localization. New York : Raven Press.

Bibliography

528

Oldfield , R. C. ( 1963) . Individual vocabulary and semantic currency: A prelimina study. British Journal of Social and Clinical Psychology, 2, 122- 130. Oldfield , R. C., & Wingfield , A . ( 1965) . Responselatencies in naming objects. Quarterly Journal of Experimental Psychology, 17, 273- 281. Oiler , D . K ., & MacNeilage, P. F. ( 1983) . Development of speechproduction . In P. F. MacNeilage ( Ed.), Theproduction of speech.New York : Springer. Olson, D . R. ( 1970) . Language and thought : Aspects of a cognitive theory of semantics. PsychologicalReview, 77, 257- 273. Osgood, C. E. ( 1971) . Where do sentencescome from? In D . Steinberg & L . Jakobovits ( Eds.), Semantics: An interdisciplinary reader in philosophy, linguistics andpsychology. Cambridge University Press. Osgood, C. E. ( 1980) . Lectureson languageperformance. New York : Springer. Osgood, C. E., & Bock, J. K . ( 1977). Salienceand sentencing: Some production principles. In S. Rosenberg( Ed.), Sentenceproduction: Developmentsin research and theory. Hillsdale, NJ: LawrenceErlbaum. Paivio, A ., Yuille , J. C., & Madigan, S. ( 1968). Concreteness , imagery, and meaningfulnessvaluesfor 925 nouns. Journal of ExperimentalPsychology, 76, ( I , Pt. 2). Palermo, D . S., & Jenkins, J. J. ( 1964) . Word association norms. Minneapolis: University of Minnesota Press. Parker, F., & Walsh, T . ( 1985) . Mentalism vs physicalism: A comment on Hammarberg and Fowler. Journal of Phonetics, 13, 147- 153. Partee, B., & Rooth , M . ( 1983) . Generalizedconjunction and type ambiguity . In R. Bauerle, C. Schwarze, & A . von Stechow ( Eds.), Meaning, useand interpretation of language. Berlin: De Gruyter . Pechmann, T . ( 1984) . Oberspezifizierungund Betonungin referentieller Kommunikation. Unpublished doctoral dissertation, Mannheim University . Perdue, C. ( Ed.) . ( 1984) . Secondlanguageacquisition by adult immigrants: Afield manual. Rowley, MA : Newbury House. Perfetti, C. A ., & Goldman , S. R. ( 1975) . Discourse functions of thematization and topicalization . Journal of PsycholinguisticResearch,4, 257- 271. Perkell, J. S. ( 1969) . Physiologyof speechproduction: Resultsand implications of a quantitativecineradiographicstudy. Cambridge, MA : MIT Press. Perkell, J. S. ( 1980) . Phonetic features and the physiology of speechproduction . In B. Butter worth (Ed.), LanguageProduction: Vol. 1. Speechand talk. London: Academic Press.

Petrie, H . ( 1988) . Semantic and syntactic processes in speechproduction. Unpublish doctoral dissertation, University of London. Pierrehumbert, J. ( 1981). Synthesizing intonation . Journal of the Acoustical Societyof America, 70, 985- 995.

529

. Ann Arbor: Universityof Pike. K . ( 1945) . The intonation of American English Michigan Press. . Cambridge , Pinker. S. ( 1985) . Language - - learnability and languagedevelopment MA : Harvard University Press. Posner, M . I ., & Snyder, C. R. R. ( 1975) . Attention and cognitive control . In R. Solso ( Ed.), Information -processing and cognition: The Loyola symposium. Hillsdale, NJ: Lawrence Erlbaum. Power, M . J. ( 1983) . Are there cognitive rhythms in speech? Language and Speech,26, 253- 261. Power, M . J. ( 1984) . Are there cognitive rhythms in speech?- A reply to Beattie ( 1984) . Languageand Speech,27, 197- 201. Premack, D . ( 1976) . Intelligence in ape and man. Hillsdale, NJ: Lawrence F.rlhaum. Prentice, J. L. ( 1966) . Response~trength of single words as an influence in sentence behavior. 10urnalof Verbal Learning and Verbal Behavior, 5, 429- 433. Prentice, J. L . ( 1967). Effects of cuing actor vs. cuing object on word order in sentenceproduction . PsychonomicScience,8, 163- 164. " Priller , J., & Mittenecker, E. ( 1988) . Experimente zum Unterschied von Wort " " " auf der Zunge und Gefiihl des Wissens . Zeitschrift /iir experimentelle und angewandtePsychologie, 35, 129 146. Prince, A . ( 1983) . Relating to the grid . Linguistic Inquiry , 14, 19- 100. Prince, E. ( 1981) . Toward a taxonomy of given-new information . In P. Cole (Ed.), Radicalpragmatics. New York : Academic Press. Pullum , G. K ., & Zwicky , A . M . ( 1988). The syntax-phonology interface. In F. J. Newmeyer( Ed.), Linguistics: TheCambridgeSurvey. Cambridge University Press. Pylyshyn, Z . W. ( 1984) . Computation and cognition: Toward a foundation for cognitive science. Cambridge, MA : MIT Press. Reason, J., & Lucas, D . ( 1984) . Using cognitive diaries to investigate naturally occurring memory blocks. In JE . Harris & P. E. Morris ( Eds.), Everyday . London : Academic Press. memory, actionsand absentmindedness Recanati, F. ( 1986). On defining communicative intentions. Mind and Language, 1, 213- 242. Redeker, G. ( 1986) . Languageusein informal narratives: Effects of social distance and listener involvement. Unpublished doctoral dissertation, University of California , Berkeley. Reichenbach, H . ( 1947) . Elementsof symbolic logic. New York : Free Press. Reinhart, T . ( 1982) . Pragmatics and linguistics: An analysis of sentencetopics. Philosophica, 27, 53- 94.

Bibliography Reinhart, T . ( 1984) . Principles of Gestalt perception in the temporal organization of narrative texts. Linguistics, 22, 779- 809.

RochesterS . R., & Gill , J. ( 1973) . Production of complex sentencesin monologue and dialogues. Journal of Verbal Learning and Verbal Behavior, 12, 203- 210. RochesterS ., & Martin , J. R. ( 1979) . Crazy talk. A study of the discourseof schizophrenicspeakers. New York : Plenum Press. Rosch~E. ( 1975). Cognitive referencepoints . Cognitive Psychology, 7, 532- 547. Rosch, E., Mervis , C. B., Gray , W., Johnson, D ., & Boyes- Braem, P. ( 1976) . Basicobjects in natural categories. Cognitive Psychology, 8, 382- 439. Ross, J. R. ( 1970). On declarative sentences . In R. A . Jacobs& P. S. Rosenbaum ( Eds.), ~ eadingsin transformationalgrammar. Waltham , MA : Ginn . Roudet,L . ( 1910) . Elementsde phonetiquegenerale. Paris: Librairie Universitaire H . Welter.

Rubin , D . C. ( 1975) . Within word structure in the tip -of -the- tongue phenome . Journal of VerbalLearning and Verbal Behavior, J4, 392- 397. Rumelhart, D . E., McClelland , J. L ., & the POP ResearchGroup . ( 1986) . Parallel distributedprocessing. (2 Vols.) . Cambridge, MA : MIT Press.

Ryave, A . L . ( 1978) . On the achievementof a seriesof stories. In J. Schenkein (Ed.), Studies in the organization of conversationalinteraction. New York : Academic Press.

Sacks, H . ( 1978) . Some technical considerations of a dirty joke . In J. Schenkein (Ed.), Studies in the organization of conversationalinteraction. New York : Academic Press. Sacks, H ., Schegloff, E. A ., & Jefferson, G. ( 1974) . A simplest systematicsfor the organization of turn -taking in conversation. Language, 50, 696- 735.

Sadock, J., & Zwicky , A . ( 1985) . Speechact distinctions in syntax. In T . Shopen ( Ed.), Languagetypology and syntactic description: Vol. J. Clausestructure. Cambridge University Press.

Saltzman, E., & Kelso, J. A . S. ( 1987) . Skilled actions: A task- dynamic approach . PsychologicalReview, 94, 84- 106. Sanford, A . ( 1985) . Cognition and cognitivepsychology. London: Weidenfeld and Nicholson. (Also: Hillsdale, NJ: LawrenceErlbaum) . Sanford, A . and Garrod , S. ( 1981) Understandingwritten language: Explorations . Chichester: John Wiley . of comprehension beyondthe sentence Schaffer, D . ( 1983) . The role of intonation as a cue to turn taking in conversation . Journal of Phonetics, 11, 243- 259. Schank, R. C. ( 1972) . Conceptual dependency: A theory of natural language understanding. CognitivePsychology, 3(4), 552- 631.

531

Bibliography

Schank, R. C. ( 1975) . Conceptual information processing. Amsterdam: North Holland . Schegloff, E. A . ( 1972). Sequencingin conversational openings. In J. J. Gumperz & D . H . Hymes ( Eds.), Directions in psycholinguistics. New York : Holt , Rinehart and Winston.

Scheg ;loffE . A. ( 1979 ). The

relevance of repair to syntax -for -conversation . In

T . Givon (Ed.), Syntax and semantics. ( Vol. 12). New York : Academic Press.

macro link . Berkeley: University of California Press. Schegloff, E. A ., Jefferson, G., & Sacks, H . ( 1977). The preference for selfcorrection in the organization of repair in conversation. Language, 2, 361- 382. Schegloff, E. A ., & Sacks, H . ( 1973) . Opening up closings. Sem;ot;ca, 7, 289 327. Schenkein. J. ( 1980) . A taxonomy for repeating action sequencesin natural con,), Language,nrndurtinn,. Vol. J. Speechand talk . versation. In B. Butter worth . Press Academic London: Scherer, K . ( 1986) . Vocal affect expression: A review and a model for future research . PsychologicalBu//etin, 99, 143- 165. Schiffrin, D . ( 1985). Everyday argument: The organization of diversity in talk . In T . A . van Dijk (Ed.), Handbook of discourseanalysis. ( Vol. 3) . London : Academic Press. Schlesinger, I . M . ( 1977) . Production and comprehensionof utterances. Hinsdale, NJ: LawrenceErlbaum. Schmerling, S. ( 1976). Aspects of English sentencestress. Austin : University of TexasPress. Schneider, W ., & Shiffrin , R. M . ( 1977) . Control led and automatic human information processing: I . Detection, search and attention. PsychologicalReview, 84, 1- 66. Schouten, M . E. H ., & Pols, L . C. W. ( 1979) . Vowel segmentsin consonantal contexts: A spectral study of coarticulation . Journal of Phonetics, 7, (Part I ) -

1- 23, (PartII ) 205- 224.

Schreuder, R. ( 1978) . Studiesin psycholexicology: With special referenceto verbs o/ motion. Unpublished doctoral dissertation, Nijmegen University . Schriefers, H . ( 1985) . On semanticmarkednessin languageproduction and verification . Unpublished doctoral dissertation, Nijmegen University . Scollon, R., & Scollon, S. B. K . ( 1981). Narrative. literacy andface in interethnic

. Norwood, NJ: Ablex. communication Scott, D. ( 1982 ). Duration as a cue to the perceptionof a phraseboundary. Journalof theAcousticSocietyof America,71, 996- 1007.

Bibliography

532

Searle, J. R. ( 1979). Expressionand meaning: Studies in the theory of speechacts. Cambridge University Press. Selkirk, E. ( 1972). Thephrasephonology of English and French. Unpublished doctoral dissertation, MassachusettsInstitute of Technology.

Selkirk, E. ( 1982). The syntax of words (Linguistic Inquiry Monograph 7) . Cambridge , MA : MIT Press. Selkirk, E. ( 1984a). Phonologyand syntax: The relation betweensoundand structure . Cambridge, MA : MIT Press. Selkirk, E. ( 1984b) . On the major class features and syllable theory. In M . Aronoff & R. T . Oehrle (Eds.), Languageand soundstructure: Studies in phonology presentedto Morris Halle by his teacherand students. Cambridge, MA : MIT Press. Senft, G. ( 1985) . Klassifikationspartikel im Kilivila - Glossen zu ihrer morpho logischen Rolle, ihrem Inventar und ihrer Funktion in Satz und Diskurs. Lin guistischeBerichte, 99, 373- 393. Seuren, P. A . M . ( 1978) . Grammar as an underground process. In A . Sinclair, ' R. J. Jarvella, & W . J. M . Levelt ( Eds.), The child s conception of language. Heidelberg: Springer. Seuren, P. A . M . ( 1985) . Discoursesemantics. Oxford : Blackwell. Seuren, P. A . M . ( 1988) . Presupposition. In A . von Stechow & D . Wunderlich ( Eds.), Handuchder SemantikfHandbookof semantics. Frankfurt : Athenium . ' Seymour, P. H . K . ( 1979) . Human visualco-,nition. New York : St. Martin s Press. Shattuck- Hufnagel, S. ( 1979) . Speecherrors as evidencefor a serial order mechanism in sentenceproduction . In WE . Cooper & E. C. T . Walker (Eds.), Sentence processing: Psycholinguistic studies presented to Merrill Garrett. Hillsdale, NJ: LawrenceErlbaum. Shattuck-Hufnagel, S. ( 1982). Three kinds of speecherror evidencefor the role of grammatical elements in processing. In L . K . Obler & L . Menn (Eds.), Exceptional languageand linguistics. New York : Academic Press. Shattuck- Hufnagel, S. ( 1983) . Sublexical units and suprasegmentalstructure in speechproduction planning. In P. F. MacNeilage ( Ed.), Theproduction of speech. New York : Springer. Shattuck- Hufnagel, S. ( 1985) . Context similarity constraints on segmentalspeech errors: An experimental investigation of the role of word position and lexical stress. In J. Lauter ( Ed.), On the planning and production of speechin normal and hearing-impaired individuals: A seminarin honour of S. Richard Silverman(ASHA Reports, 15) . Shattuck- Hufnagel, S. ( 1986) . The representation of phonological infonnation during speechproduction planning: Evidence from vowel errors in spontaneous speech. Phonology Yearbook, 3, 117- 149.

Bibliography

533

Shattuck- Hufnagel, S. ( 1987) . The role of word onset consonants in speech production planning: New evidence from speech error patterns. In E. Kelier & M . Gopnik ( Eds.), Sensory processes in language. Hillsdale, NJ: Lawrence Erlbaum. Shattuck- Hufnagel, S., & Klatt , D . ( 1979) . The limited use of distinctive features and markednessin speechproduction : Evidence from speecherror data. Journal of VerbalLearning and Verbal Behavior, 18, 41 55. Shepard, R. N ., & Metzler, J. ( 1971). Mental rotation of three dimensional objects . Science, 171, 701- 703. Shepard, R. N ., & Hurwitz , S. ( 1984) . Upward direction , mental rotation , and discrimination of left and right turns in maps. Cognition, 18, 161- 193. Siegel, D . ( 1974). Topicsin Englishmorphology. New York : Garland. Siegel, G. M ., & Pick, H . L. ( 1976) . Auditory feedback in the regulation of voice. Journal of the AcousticalSocietyof America, 56, 1618- 1624. Slobin, D . ( 1982). Universal and particular in the acquisition of language. In E. Wanner & L . R. Gleitman (Eds.), Languageacquisition: The state of the art . Cambridge University Press. Smith, E., & Medin , D . L . ( 1981) . Categories and concepts. Cambridge, MA : Harvard University Press. Sonnenschein , S ( 1982). The effects of redundant communications on listeners: When more is less. Child Development, 53, 717- 729. Sonnenschein , S. ( 1984) . The effect of redundant communication on listeners: different types may have different effects. Journal of Psycholinguistic Research Why , 13, 147- 166. Sorokin, V. N . Gay, T ., & Ewan, W. G. ( 1980) . Some biomechanicalcorrelatesof jaw movement. Journal of the Acoustical Societyof America, 68(SI ), S32(A ) . Sowa, J. F. ( 1984). Conceptualstructures. Reading, MA : Addison-Wesley. : Communicationand cognition. Oxford Sperber, D ., & WilsonD . ( 1986). Relevance Blackwell. : Sridhar, S. N . ( 1988) . Cognition and sentenceproduction: A cross-linguistic study. New York : Springer. Steedman, M . J. ( 1982) . Reference to past time. In R. J. Jarvella & W. Klein (Eds.), Speech, place and action: Studies in deixis and related topics. Chichester: John Wiley . , Steedman, M . J., & Johnson- Laird , P. N . ( 1980) . The production of sentences utterancesand speechacts: Have computers anything to say? In B. Butter worth (Ed.), Languageproduction: Vol. 1. Speechand talk . London : Academic Press. Sternberger, J. P. ( 1982) . The nature of segmentsin the lexicon: Evidence from speecherrors. Lingua, 56, 43 65.

Bibliography

534

Stemberger, J. P. ( 1983a). Speecherrors and theoretical phonology: A review. Bloomington : Indiana Linguistics Club. Stemberger, J. P. ( 1983b) . The nature of Irl and III in English: Evidence from speecherrors. Journal of Phonetics, 11, 139- 147. Stemberger, J. P. ( 1984) . Lexical bias in errors in languageproduction: Interactive . editors. and perceptual blases. Unpublished manuscript, Carnegiecomponents Mellon University , Pittsburgh. Stemberger, J. P. ( 1985a). An interactive activation model of languageproduction . In A . W. Ellis (Ed.), Progressin the psychologyof language: Vol. 1. Hillsdale, NJ: LawrenceErlbaum. Stemberger, J. P. ( 1985b). The lexicon in a model of languageproduction. New York : Garland Publishing. Stemberger, J. P., & MacWhinney, B. ( 1986) . Form -oriented inflectional errors in languageprocessing. Cognitive Psychology, 18, 329- 354. Stemberger, J. P., & Treiman, R. ( 1986) . The internal structure of word -initial consonantclusters. Journal of Memory and Language, 25, 163- 180. Sternberg, S., Monsell, S., Knoll , R. L ., & Wright , C. E. ( 1978). The latency and : Comparisons of speechand typewriting . duration of rapid movement sequences In G. E. Stelmach (Ed.), Information processingin motor control and learning. New York : Academic Press. Sternberg, S., Wright , C. E., Knoll , R. L., & Monsell, S. ( 1980) . Motor programs in rapid speech: Additional evidence. In R. A . Cole ( Ed.), Perceptionand production offluent speech.Hillsdale, NJ: Lawrence Erlbaum. Stevens, K . N . ( 1983). Design features of speechsound systems. In P. F. MacNeilage ( Ed.), Theproduction of speech.New York : Springer.

Stevens, K ., & Perkell, J. S. ( 1977). Speechphysiology and phonetic features. In M . Sawashima& F. S. Cooper (Eds.), Dynamic aspectsof speechproduction. University of Tokyo Press. Svartvik, J., & Quirk , R. (Eds.) . ( 1980). A corpus of English conversation. Lund: CWK Gleerup. Tannenbaum, P. H ., Williams , F., & Hillier , C. S. ( 1965). Word predictability in the environments of hesitations. Journal of Verbal Learning and Verbal Behavior, 4, 134- 140.

Tannenbaum, P. H ., & Williams , F. ( 1968). Generation of active and passivesentence as a function of subject and object focus. Journal of Verbal Learning and Verbal Behavior, 7, 246- 250. Tanz, C. ( 1980). Studies in the acquisition of deictic terms. Cambridge University Press. Taylor , I . ( 1969). Content and structure in sentence production . Journal of VerbalLearningand VerbalBehavior, 8, 170- 175.

Bibliography

535

Tent, J., & Clark , J. E. ( 1980) . An experimental investigation into the perception of slips of the tongue. Journa/ of Phonetics, 8, 317- 325. Terken, J. M . B. ( 1984). The distribution of pitch accents in instructions as a function of discoursestructure. Languageand Speech,27, 269- 289. Thomassen, A . J. W. M ., & Teulings, H . L . ( 1985) . Time, sizeand shapein handwriting : Exploring spatio- temporal relationships at different levels. In J. A . Michon & J. Jackson( Eds.), Time. mind and behavior. Heidelberg: Springer. Thorson, N . ( 1983) . Two issuesin the prosody of standard Danish. In A . Cutler & R. D. Ladd(Eds.), Prosody: Models,and measurements. Heidelberg : Springer . Treiman, R. ( 1983) . The structure of spoken syllables: Evidence from novel word games. Cognition, 15, 49- 74. Treiman, R. ( 1984). On the status of final consonant clusters in English syllables. Journal of Verbal Learning and Verbal Behavior, 23, 343- 356. Turner , E., & Rommetveit, R. ( 1968) . Focus of attention in recall of active and . Journal of Verbal Learning and VerbalBehavior, 7, 543- 548. passivesentences Turvey, M . T . ( 1977) . Preliminaries to a theory of action with referenceto vision. In R. Shaw & J. Brandsford ( Eds.), Perceiving, acting, and knowing: Toward an ecologicalpsychology. Hillsdale, NJ: Lawrence Erlbaum. Tyler, L ., & Wessels, J. ( 1983). Quantifying contextual contributions to word , 34, 409- 420. recognition processes. Perceptionand Psychophysics Umeda, N . ( 1975) . Vowel duration in American English. Journal of the Acoustical Societyof America, 58, 434- 445. Vaissiere, J. ( 1983) . Language-independent prosodic features. In ACutlerD . Heidelberg: Springer. . R. Ladd (Eds.), Prosody: Models and measurements Valian , V. V. ( 1971) . Talking, listening and linguistic structure. Unpublished doctoral dissertation, Northeastern University , Boston. Van Bezooijen, R. ( 1984) . Characteristicsand recognizability of vocal expression of emotion. Dordrecht: Foris. Van den Broecke, M . P. R., & Goldstein, L. ( 1980). Consonant features in speech errors. In V . A . Fromkin ( Ed.), Errors in linguistic performance: Slips of the tongue, ear, pen, and hand. New York : Academic Press. Van der Sandt, R. A . ( 1982) . Kontekst en presuppositie. Nijmegen Institute of Semantics. Van Dijk , T . A . ( 1985). Handbook of discourseanalysis. (4 Vols.). London : Academic Press. Van Galen, G. P., & Teulings, H . L . ( 1983) . The independentmonitoring of form and scalefactors in handwriting . Acta Psychologica, 54, 9- 22. Van Wijk , C. ( 1987). The PSY behind PHI : A psycholinguistic model for performance structures. Journal of PsycholinguisticResearch , 16, 185- 199.

Bibliography

536

Van Wijk , C., & Kempen, G. ( 1987). A dual system for producing self-repairs in spontaneousspeech: Evidencefrom experimentally elicited corrections. Cognitive Psychology, / 9, 403- 440. Vorberg, D . ( 1985). Unerwartete Folgen von zufalliger Variabilitat : Wettlauf Modeile fur den Stroop- Versuch. Zeitschrift fiir experimentelle und angewandte Psychologie,32, 494- 521. Wales, R., & Toner, H . ( 1980). Intonation and ambiguity . In W. Cooper & E. Walker ( Eds.), Sentenceprocessing: Psycholinguisticstudiespresentedto Merrill Garrett. Hillsdale, NJ: LawrenceErlbaum. Walton , D . N . ( 1982). Topical relevance in argumentation. Amsterdam: John Benjamins. Webber, B. L . ( 1981) . Discourse model synthesis: Preliminaries to reference. In A . K . Joshi, B. L . Webber, & I . A . Sag (Eds.), Elementsof discourseunderstanding . Cambridge University Press. Webelhuth, G. ( 1985). German is configurational . The Linguistic Review, 4, 203- 246. Weissenborn, J., & Klein , W . (Eds.) . ( 1982) . Here and there: Cross-linguistic studies on deixis and demonstration. Amsterdam: John Benjamins.

Wells, G., MacLure , M ., & Montgomery , M . ( 1981) . Some strategies for sustainin conversation. In P. Werth (Ed.), Conversationand discourse: Structure and interpretation. London: Croom Helm. Wells, R. ( 1951). Predicting slips of the tongue. The Yale Scientific Magazine, 26, 9- 30. (Also in V. A . Fromkin ( Ed.), ( 1973) . Speecherrors as linguistic evidence. The Hague: Mouton .) Whorf , B. L . ( 1956). Language. thought. and reality . Cambridge, MA : MIT Press. Wickelgren, W. A . ( 1969) . Context-sensitive coding, associative memory, and serial order in (speech) behavior. PsychologicalReview, 76, 1- 15. Wickelgren, W . A . ( 1976) . Phonetic code and serial order. In E. C. Carterette & M . P. Friedman ( Eds.), Handbook of Perception: Vol. 7. New York : Academic Press.

Willems, N . ( 1983) . English intonation from a Dutch point of view: An experiment investigationof English intonation producedby Dutch native speakers. Dor drecht: Foris. Williams , E. ( 1983) . Semantic and syntactic categories. Linguistics and Philosophy, 6, 423- 446. Wilshire, C. ( 1985) . Speecherror distributions in two kinds of tongue twisters. Unpublished bachelor thesis, Monash University, Clayton , Victoria . ' WilsonD ., & Sperber, D . ( 1981) . On Grice s theory of conversation. In P. Werth (Ed.), Conversationand discourse: Structure and interpretation. London: Croom Helm.

Bibliography

537

WilsonD ., & Sperber, D . ( 1988) . Mood and the analysis of non-declarative sentences . In J. Dancy. J. Moravsik . & C. Taylor (Eds.), Hunwmagency: Language.

. dutyandvalue.StanfordUniversityPress ). Effectsof frequencyon identificationand namingof objects Wingfield, A. ( 1968 . AmericanJournalof Psychology , 8/ , 226- 234. Winterhoff-Spurk, P., Herrmann, T., & Weindrich, D. ( 1986 ). Requestingrewards : A study of distributivejustice. Journalof Languageand SocialPsychology , 5, 13- 31. Wunderlich, D. ( 19] 6). Towardsan integratedtheoryof grammaticaland pragmatical , methods . In A. Kasher(Ed.), Languagein focus: Foundations meaning Bar-Hillel. Dordrecht: Reidel. . Essaysin memoryof Yehoshua andsystems . . CambridgeUniversityPress Wunderlich , D. ( 1979 of linguistics ). Foundations . In F. Coulmas(Ed.), A Festschrift Wunderlich, D. ( 1981 ). Linguisticstrategies . TheHague: Mouton. for nativespeaker Wunderlich,D., &.. Reinelt, R. ( 1982 ). How to gettherefrom here. In R. Jarvella & W. Klein (Eds.), Speech , place, andaction: Studiesin deixisandrelatedtopics. : JohnWiley. Chichester . Leipzig: Kroner. Wundt, W. ( 1896 ). GrundrissderPsychologie Wundt, W. ( 1900 (2 Vols.). Leipzig: Kroner. ). Die Sprache ' ). I recognizeyour facebut I cant rememberyour name: Yarmey, A. D. ( 1973 . Furtherevidenceon the tip- of the tonguephenomenonMemoryand Cognition , 1, 287- 290. . In M. A. Campbell(Ed.), ). On gettinga word in edgewise Yngve, V. H. ( 1970 the Sixth , Society. University Linguistic Chicago Meeting Regional Papersfrom . of ChicagoPress Zubin, D. A., & Choi, S. ( 1984). Orientationand gestalt: Conceptualorganizing . In D. Testen, V. Mishra & J. Drogo principlesin the lexicalizationof space . ChicagoLinguistic on Lexical Semantics the Parasession . Eds ), Papersfrom ( Society. ). La frontieredu , G. R. E., & Levelt, W. J. M. ( 1977 , W., Ouweneel Zwanenburg mot enFran~ is. Studiesin Language , 1, 209- 221. . In P. M. Peranteau , J. N. Levi, & G. C. Zwicky, A. M. ( 1972). On casualspeech the the . Phares(Eds), Papersfrom ChicagoLinguistic EighthRegionalMeetingof , 607- 615. Society ). Cliticizationversusinflection: English Zwicky, A. M., & Pullum, G. K . ( 1983 n' t. Language , 59, 502- 513.

Author

Index

Abbs, J. H., 439, 446, 451 Abercrombie , D., 392 Anderson , J. R., 17, 19- 2I , 73- 75, 77, 93, 154 Anderson , W. G., 414 Appelt, D. A., 124 Arbib, M. A., 446 Arnold, G. F., 307, 310 Aston, C. H., 388 Atkinson, J. M., III Austin, J. L., 58

Boomer , D. S., 23, 247 -Braem , P., 223 Boyes Bransford , J. D., 113 BrazilD., 315 Breskin , S., 127

212- 213 Bauer , L., 185

Bresnan , J., 74, 161- 162. 170, 189, 193, 197, 235, 338 Broadbent , D. E., 21, 119 Brodda, B., 449, 453 Browman , C. P., 320, 327, 445 Brown, G., 23, 36, 112- 113, 147- 148, 307, 400, 403- 405 Brown, G. D. A., 199 Brown, P., 137 Brown, R., 113, 320 Bruner, J. S., 29 Buhler, K., 45- 46, 53- 54, 89 Burkle, B., 51- 52, 156 Butterworth, B., 23, 36, 109, 126, 128, 183, 186, 203, 217- 218, 222, 231 Buttrick, S., 48

, R. de, III Beaugrande Bell-Berti, F., 443 Berg, T., 249 Berko-Gleason , J., 320 Berian, R. W., 414 Bever , T. G., 23 Bierwisch , M., 93, 179, 216, 218, 248 Bleasdale , F. A., 320 Bloem, I., 400 Bock, J. K., 14, 16, 20, 23, 192, 236, 261, 263- 265, 267, 273- 275, 278- 281, 303, 394- 396 Boehme , K., 46 BolingerD ., 165, 307 Bond, Z. S., 471

Calvert , D. R., 425,428,434 Camden , C. T., 461,465,468 Carr, T. H., 20 Carroll,J. M., 196 , 262 Chafe . W. L.. 23, 36, 119 , 138 , 148 , 267 Chistovich , L. A., 23, 35I Choi, S., 52 , N., 305 Chomsky Clark, E. V., 23, 46, 54, 85, 128 , 139 , 200, 213,229,267 Clark, H. H., 23, 43, 48- 49, 59, 65- 67, 85, 94, 120 , 229, , 136 , 148 , 185 , 122 , 128 , 133 267 Clark, J. E., 470- 471 Clements , G. N., 290,292,297

Baars,B. J., 356- 358, 461- 462, 464- 465, 468, 477 , A., 10 Baddeley Barth, M., 74 Barwise , J., 74 Bassett , M. R., 391 Bates , E., 121, 147, 262, 265. 268, 270,

- 128 Beattie . G.. 23. 33.35.68. 126 . 203

Author Index Clifton, C., Jr., 464 Cohen,A., 214, 399 Cole, K. J., 451 Collier, R., 307, 310, 400, 404- 405 Cooper, F. S., 441 Cooper, W. E., 14, 264, 388- 389, 394- 396, 400, 471 Coulthard, M., 315 Craik, F. I., 320 Crompton, A., 327, 329, 336, 340, 346, 348 Cruse, D. A., 187 Cruttenden , A., 307, 310, 314 Currie, K. L., 113, 307, 400, 403- 405 Cutler, A., 35, 165, 171- 172, 184, 186, 199, 214, 247, 255, 307, 326, 344, 348, 355, 372, 392, 396- 397, 462- 464, 477, 495 Daniloff, R. G., 444 Danks, J. H., 24, 27 Danly, M., 388- 389 Dauer, R. M., 393- 394 -Nielson, N., 336 Davidson Deese , J., 65, 306, 316, 386, 390, 393 Dell, G. S., 13- 14, 16, 19- 20, 137, 211, 218, 319, 330- 331, 334, 342, 351- 354, 356- 360, 362- 363, 366, 469, 474- 476 Denes , P. B., 426, 431 De Pijper, J. R., 400, 403 De Rooij, J. J., 381 De Smedt,K., 235, 253, 468, 486 Deutsch , W., 46, 121, 129, 131, 133 Devescovi , A., 262, 268, 270 Diehl, R. L., 444 Dik, S. C., 74 Dixon, R. M. W., 55 Dogil, G., 370, 408 Doherty, M., 102 Donders,F. C., 200 Drew, P., III DuBois, J. W., 485 Dfingelhoff,F. J., 226- 227 DuncanS., 34 Eady, S. J., 394- 396 Easton,T. A., 448 Ehrich, V., 51- 52, 58, 112, 142, 155- 157 Eibl-Eibesfeldt , I., 30, 37 Eriksen,C. W., 414 Erwin, C. 1., 416 Ewan, W. G., 449

540 Fanshel , D., 112 Fauconnier , G., 94 Fay, D., 14, 184, 218, 248, 254- 256, 355,

462 , E., 205 Feigenbaum Fel'dman , A. G., 438 Fillmore , C. J., 46, 48, 53- 55, 90, 97- 98, 103 , 266,272 FiskeD. W., 35 Floresd'Arcais , G. B., 20, 223-224,226, 262,269 Fodor,J. A., 15,20, 22- 23, 74, 94, 201 Folkins , J. W., 439,446 Ford, M., 23, 256-259,282 Foss , D. J., 23 Fowler , C. A., 147 , 438- 440,444- 447, 449- 450,452- 453 Francik , E. P., 136 Francis , W. N., 219 Franck , D., 39 Freud , S., 214,217,234 Friederici , A., 49 Frijda,N., 74 Fromkin , V. A., 184 , 200,214,216,218221,247- 248,325,328,330- 332,335336,338,340,347-350 Fry, D., 24 , 0 ., 332,388- 389 Fujimura , H., 444 Fujisaki Gallistel , C. R., 435 Gardiner , J. M., 320 , E., 307 Garding Garfield , J. L., 22 Garnham , A., 114 , 199 , S. M., 469 Garnsey Garrett , M. F., 14,23- 24, 201,204,216, 218,220- 222,247-249,255,282,303, 324,336,371- 372,378,497 Garrod , S., 119 , 154 Gay, T., 388,439,445,449 Gazdar , G., 41, 63- 64, 170 Gee , J. P., 259,386,390- 391 Gelfer , C. E., 400 , R. J., 43 Gerrig Gerstman , L. J., 127 Gill, J., 259 Glaser , W. R., 226- 227 - 112 Gotrman , E., 111 , 495 ~J. A., 112 Goguen Goldman , N., 205- 206

541

Author Index Goldman, S. R., 263 Goldman-Eisler, F., 23, 126, 128, 203- 204, 259 Goldstein,L., 327, 350, 445 Good, D. A., 128 , H., 320 Goodglass Goodwin, C., 33, 36- 37 Gordon, P. C., 351, 419 Gray, W., 223 Grice, H. P., 30, 39- 41, 43- 44, 47, 53, 59, 66- 67, 69, 123, 129, 133, 148, 158, 213 Grimes,J. E., 23 , F., 36, 259, 386, 390- 391 Grosjean Grosjean - , L., ~86 Grosz, B. J., 112- 113, 119 Gruber, J. S., 90 , J. J., 42 Gumperz Gussenhoven , C., 307

, P., 400 Hagoort Hakes , D. T., 23 Hale,K., 169 Halle,M., 289,291,305,441 , M. A. K., 23, 35, 165 , 271,307, Halliday

- 393 392

Hankamer . G.. 185

, T. A., 122 , 129 , 218,221 Harley Harris,K. S., 439,443 Hart, J. 't, 307,310,399,404- 405 Hassan , R., 271 Haviland, S. E., 120 Hawkins,J., 82, 260 Hawkins. P. R.. 23

. B.. 292 Hayes . A. F.. 330 . 339 . 350 Healy Pc D. c.. 400 J-Jeesch

Henderson , A., 126- 127 Hennan. S. J.. 121 Hemnann, T., 23, S2, 66, 119, 124, 131, I ~~- I~S

Hill. C.. SO . 52 Hill, D. R., 393 Hillier, C. S., 203 Hindle , D., 497 Hockett , C. F., 482

, E., 14, 24- 25, 28, 162, 206, Hoenkamp 235- 237, 240, 244, 249, 252- 253, 275, 282 Holmes,V. M., 23, 256- 260, 282 Holt, K. G., 447 Hoosain, R., 435

Hopper, P. I., 137 Hotopf, W. H. N., 199, 216, 219- 220 Housum,J., 147 Huijbers, P., 187, 231, 258 , G. W., 230, 232 Humphreys Hurwitz, S., 154 Huttenlocher , J., 225- 226, 230 M. 320 Hyde, , Hymes,D., 42 Irwin, D. E., 263, 265 Isaacs , E. A., 133, 148 lsard, S., 165, 171- 172, 247, 392

Jackendoff , , R., 74, 79, 84, 87, 90, 94, 168 170 Jaffe , J., 127 James , D., 482 James , W., 119 , 320 Jarvella , R. J., 46, 121 Jassem , W., 393- 394 Jefferson , 458 , G., 23, 31, 33, 111 Jenkins , J. J., 216 Johns , C., 315 Johnson , D., 223 Johnson , M. K., 113 -Laird, P. N., 46, 58, 72, 74, 84, Johnson - 114 113 , 195 , 200,207- 209 Jones , H. G. V., 320- 321 , A., 399 Jongman , J., 43 Jorgenson Joshi, A., 119

Kaisse , 302,367- 369,377,408 , E. M., 166 Kamp,H., 114 Kao, H. S. R., 435 Katz, W., 399 Keele , S. W., 416 Keenan , E. L., 194 Keenan , J., 121 Kelly, M. H., 394-396 Kelso , J. A. S., 439- 440,446- 447,449450,452 KelterS., 122 , 221,273,275 , 187 , , 162 , G., 14,24- 25, 28, 121 Kempen 231,235- 237,240,244,249,253,258, 275,282,468,486- 489,491,497 Kenstowicz , M., 166 Kent, R. D., 440 Kenworthy,J., 113 , 307,400,403-405 292 S. J. , , Keyser Kiegl, R., 270

Author Index

Kintsch , W., 74, 121 , P., 286 Kiparsky Kisseberth , C., 166 Klapp,S. T., 410,414- 419,454,473 Klatt, D. H., 330,339-340,350,389 - 137 Klein, W., 46, SO , 52- 53, 89, 112 , 136 , 139 Kloker,D., 389 Knoll, R. L., 417,419-421 Kohler,K. J., 436 Kohn, S. E., 320 Koriat, A., 320 , S. M., 72 Kosslyn Koster , C., 112 , 142 Kowal,S., 391 Kozhevnikow , V. A., 23, 351 Krauss , R. M., 148 Krach,A., 497 Kubicek , L. F., 225,230 Kucera , H., 219 , P. N., 447 Kugler la BergeD., 20 - 112 Laboy , W., 111 , 139 , 143 Lackner , J. R., 471- 472,474 Ladd.D. R.. 165 . 176 . 307.309.385.399 , P., 438 Ladefoged LaHeij, W., 438 Lakoff,R., 57 Lane , H., 386 , S., 320,321 Langford Laver , J. D. M., 14,247,468

Linde,C., 112 , 139 , 143 , J. R., 258 Lindsley Linell, P., 285 -Higgins , H. C., 297 Longuet , F. G., 203 Lounsbury Lovins , J. B., 332 Lubker , J., 439,445 Lucas , D., 320 , J., 44, 46, 55, 57- 58, 62, 197 , 207 Lyons Maassen , B., 16, 162 , 231,258,276-278, 280- 281 , D., 19,331- 332,350-351,353, MacKay 356- 357,462,465,470,474- 477,498 , H., 199 , 203 Maclay MacLure , M., 36 , P. F., 438,442,445- 446,451 MacNeilage , B., 97, 121 , 147 , 265,267, MacWbinney 270,272- 273,344,348 , S., 264 Madigan Makkai,A., 187 Malinowski , B., 104 , R., 131 Mangoid Marshall , J. C., 22 - 148 -Wilson Marslen , W., 35, 147 , 473, 492 Martin, J. G., 203 Martin, J. R., 113 , M., 121 Masling , K., 214,216- 217,248 Mayer , D., 121 Mayhew McNeill,D., 12,23- 24, 74, 94, 285,320 Medin,D. L., 183 Menn,L., 320 , R., 214,216- 217,248,463-464, Meringer 470,478,480 Mervis , C. B., 223 Metzler , J., 72 , A., 353-354,360,409- 410 Meyer , D. E., 351,419 Meyer Mill, A. I. D., 199 Miller, G. A., 21, 43, 46, 48, 58, 74, 119 , 183 , 185 , 195 , 200,207- 209,212 Mittelstaedt , H., 49 Mittenecker , E., 320 Mohanan , K. P., 163 , 289,291 Moll, K. L., 444 Monsell , S., 417,419-421 , R., 74, 77, 79, 88 Montague , W. E., 414 Montague , M., 36 Montgomery Morton,J., 201,203

543

Author Index Motley, M. T., 356, 461- 462, 464- 466, 468- 469, 474, 477 Munro, A., 139 Nakatani, L. H., 388- 390, 392- 393 Neilson, M. D., 451 Neilson, P. D., 451 Nelson, L. J., 471 Nespor, M., 302- 303, 370, 373, 378, 381, 392, 408 Newell, A., 17, 143, 321 Ninnaier, H., 51 Nolan, F. J., 440 Noordman, L. G. M., 212 Noordman-Vonk, W., 183 Nooteboom , S., 149, 273, 350, 357, 387390, 463- 464, 470, 478, 480, 485 Nonnan, D. A., 74 Nunberg,G., 47- 48, 53, 134

-Keenan Ochs . E.. 41- 42 O'Connell . D. C.. 391 O'Connor . J. D.. 307.310 O'Connor . K. D.. 388 Ohala . J. J.. 400 Ohman . S. E. G.. 453 . G. A.. 441 Ojemann Oldfield . 230 . R. C.. 199 Oiler. D. K.. 446 Olson . D. R.. 129 , 199 , 203,267, C. E., 23, 44, 148 Osgood 269,272 Ouweneel , 367 , G. R. E., 166 -Cooper,J. M., 14, 389 Paccia Paivio, A., 264 Pa1enno , D. S., 216 Parker, F., 444 Parkes , C. H., 201 Partee , B., 89 Pearson , M., 35 Pechmann , T., 46, 129- 133, 150, 273 C. Perdue , , 42 Perfetti, C. A., 263 Perkell,J. S., 442- 444, 446, 453, 456 Perry, J., 74 Petrie, H., 126 Pick, H. L., 463 Pierrehumbert , J., 307 Pike, K., 332 Pinker, S., 192, 194, 197- 198 Pinson,E. N., 426, 431

Pollock , M. D., 414 Pols,L. C. W., 432 Posner , M. 1., 20, 73 Power , M. J., 127 Premack , D., 73 Prentice , J. L., 262- 263 Priller,J., 320 Prince , A., 298 Prince , 146 , 148 , E., 114 Pullum , G. K., 375- 376,378,408 , Z. W., 22 Pylyshyn , P. T., 230 Quinlan Quirk, R., 2, 43 Reason , J., 320 Recanati , F., 59 Redeker , G., 148 Reich, P. A., 218, 351, 356 Reichenbach , H., 56 Reinelt, R., 112, 139 Reinhart,T., 98- 99, 137, 265 Remez , R. E., 438, 450, 453 Richardson, G., 438 Riddoch, M. J., 230 RochesterS. R., 113, 259 Rommetveit , R., 262- 263 Rooth, M., 89 Rosch, E., 155, 223 Ross,J. R., 65, 264 Roudet, L., 387 Rubin, D. C., 320 Rubin, P., 438, 450, 453 Rumelhart , D. E., 19, 74 Ryalis, J., 400 Ryave,A. L., III Sacks , H., 23, 31, 33, 35- 36, 39, 111,459 Sadock , J., 63 Saltzman , E. L., 439- 440, 446, 450

Samuels , S. J., 20 Sanford , A., 113 , 119 Schaffer , D., 35 Schank , R. C., 74, 90, 139 , 458, , E. A., 23, 31, 36, 112 Schegioff 460-461,463,497 Schenkein , J., 122 , 273 Scherer , 307 , K., 102 Schiffrin , D., 38 , I. M., 23, 104 Schlesinger , S., 165 Schmerling Schneider , W., 20

S44

Author Index

Schouten , M. E. H., 432 Schreuder , R., 48, 206,212,223-224,226 Schriefers , 227- 230 , H., 188 Schunk , D. H., 65- 66 Scollon , R., III Scollon , S. B. K., III Scott , D., 391 Searle , J. R., 60- 61, 64- 65, 69 - 166 Selkirk , 174 , 176 , 287,289, , E., 165 291-292,298- 299,303- 306,361,373374,379- 380,382- 383,387,391,394, 420 Senft , G., 104 Seuren , 197 , 114 , 118 , P. A. M., 22, 74, 101 , P. H. K., 223 Seymour Sbankweiler , D. S., 441 -Hufnagel Shattuck , S., 319,321,330- 332, 334- 335,338- 343,346,350,357- 358, 360,362 , R. N., 72, 154 Shepard Shitrrin , R. M., 20 Sbillcock , R. S., 199 - 113 Sidner , 119 , C. L., 112 , D., 286 Siegel , G. M., 463 Siegel Silvennan , K., 307 Simon , H. A., 17, 143 Skarbek , A., 126 Siobin , D., 104 Small , L. H., 471 Smith , E., 183 , C. R. R., 20, 73 Snyder Sonnenschein , S., 131 Sorensen , J. M., 400 Sorokin , V. N., 449 Sowa , J. F., 74 , D., 40- 41, 43- 44, 59, 63- 64, 73, Sperber 101 Sridhar , S. N., 148 , 265,267- 270,272-273 Steedman , M. J., 57- 58, 207 , J. P., 218,247,249,256,326, Sternberger 330,332,334,336- 338,344,348,351, 356- 357,371- 372,474 , S., 414,416- 421,454 Sternberg Stevens , K., 432-433,440- 442,446 , W., 203 Strange -Kennedy Studdert , M., 441 - 137 Stutterheim , C. von, 136 Svartvik , J., 2, 43 Tannenbaum , P. H., 203, 262 Tanz, C., 46

Taylor, I., 259 Tent, J., 470- 471 Terken, J. M. B., 132, 147, 149, 273 , H. L., 435 Teulings Thomassen , A. J. W. M., 435 Thorson, N., 307 Toner, H., 166 Treiman, R., 330, 333, 336, 338 Tuller, B., 439, 446, 452, 471- 472, 474 Turner, E., 262- 263 Turvey, M. T., 438- 439, 446- 448, 450 Tye-Murray, N., 471 Tyler, L., 35, 147- 148, 473, 492 Umeda,N., 389 Vaissiere , J., 307, 389 Valian, V. V., 257 Van Bezooijen , R., 102, 307 VandenBroecke , M. P. R., 350 Vander Sandt, R. A., 118 Van Dijk , T. A., 112 VanGalen, G. P., 435 VanWijk, C., 259, 303, 391, 468, 487- 489, 491, 497 Vatikiotis-Bateson , E., 452 Vogel, I., 302- 303, 370, 373, 378, 381, 392, 408 Vorberg, D., 227

Wales , R., 166 Walker , E. C. T., 201 Walsh , T., 444 Walton . D. N.. 112

Warren, R. K., 192, 261, 263- 265 Webber , B. L., 114 Webelhuth , G., 163 Weindrich,D., 135 Weinheimer , S., 148 Weinstein , S., 119 Weissenborn , J., 46, 53 Wells, G., 36 Wells, R., 332 Wessels , J., 35 Whorf, B. L., 103 Wiche, R. T. P., 74 , W. A., 357 Wickelgren Wilkes-Gibbs, D. L., 122, 133, 148 Willems, N., 399 Williams, E., 89 Williams, F., 203, 262 Wilshire, C., 343

Author Index

WilsonD., 40- 41, 43-44, 59, 63- 64, 73, 101 , A., 230,320 Wingfield -Spurk Winterhoff , P., 135 Witten,I. H., 393 Wright,C. E., 417,419-421 Wunderlich , 139 , D., 40, 52, 64, 112 Wundt,W., 23, 26, 28, 96, 119 , 162 , 235, 239,261,275,417 . Wyatt,E. P., 417 , A. D., 320 Yanney , V. H., 36 Yogve YuiUe , J. C., 264 - 113 - 148 Yule,G., 23, 36, 112 , 147

Zubin , D. A., 52 , 367 Zwanenburg , W., 166 - 376 Zwicky . A.. 63. 369 . 375 . 378 . 408

S4S

Subject

Index

Abdominal muscles, 422 Absolutive. SeeCase

Active voice. SeePassive Activation

Accent, 297, 328. SeePitchaccent , Word accent contrastive , 132- 133, 177 default, 177- 178 intonational, 309 andisochrony , 392, 395, 397 nuclear,309 , 309- 310 primary, secondary 147 prosodic, 130, in shifts, 372 , 247, 282, 372 stranding Accentuation , 25, 132, 179, 283, 374. See alsoDe-accentuation of discriminatinginformation, 132 of referentialexpressions , 150 . SeeLexicalaccess Access Accessibility , 236, 260- 261, 264- 266, 275, conceptual 278 lexical, 16, 260, 276- 277 , 281 phonological of phoneticplan, 285 status(index) of referents , 108, 144- 146, 149, 151, 159, 163, 237- 238, 246, 270271, 401 of word fonns, 16, 276 errors), 249 Aa: ommodation(in speech ofcase, 249 in connected , 364, 408, 411 speech metrical, 299 , 364 morphological andallophonic,340, 364 segmental andstranding , 246 . SeeCase Accusative function. SeeDeixis Acknowledging

, 198, 229, 275, 278 conceptual lemma, 181, 195, 197, 200- 201, 219- 221, 232, 244, 280 , 202, 204 logogen , 93, 96, 197, 231- 232 meaning , 7, 231- 232, 280, 320, 397 phonological Activationspreading , 18- 20, 75, 211- 212, 225, 233, 319- 320, 366- 367 theoryof self-monitoring, 465, 467, 474477 erron, 234, 346- 347, theoryof speech 351- 363 Acousticgoals, 441- 442 Actor. SeeThematicroles Adjunct, 169 modeltheory, 45I Adaptive ' Address in discourse model, 99, 118, 146, 151, 152, 159, 260, 271 in form lexicon, addressframes , 165, 187, 191, 321- 329, 335- 337, 344, 346, 348349, 361- 362, 367, 371, 373, 406, 408410, 419 in linearization , 144 Addressingfailures oflemmas, 214- 222 of lexicalform, 162, 331, 346, 349, 351, 358 Adjacencypain, 36 Adjective, 148, 167- 169, 194- 196, 236, 253, 476 affixes , 104 in compounds , 288 eliding, 271 , 178 focusing

Subject Index Adjective(cont.) marked/unmarked , 229- 230 nodes , 476 , 180, 251, 255 phrase andprelexicalhesitation , 203 root, 286- 287 andword accent , 306, 374, 393 Adverbial deictic, 48, 69 modal, 62, 403 temporal,55, 58, 69, 178 Affix, 182- 183, 286, 288- 289, 299 in activationspreading , 352- 353, 355 inflectional , 290 in spelloutprocedures , 324- 331, 344, 349, 352, 362 root (classI , primary), 185, 286- 288, 294, 299 shift, 372 word(classII , secondary ), 286- 289, 294 Affricatives , 434 Agent. SeeThematicrole Agency,270 andsaliency , 268- 269, 276 , 185, 289 Agglutinativelanguage , 438 Agonist/antagonist , 22, 469 Agrammatism Alpha/gammaloop, 436, 437 Allomorph, 301- 302, 317, 325, 331, 364, 367- 370, 375- 377, 406 Allophone, 296- 298, 327- 328, 340- 341, 362, 434 Alternativeplans, 217 , 299, 306, 374, 394- 396 Alternatingstress Alveolarridge, 296, 430, 433, 442 Alveolars,433 Ambiguity andconversational maxims,41, 129 of intentions , 67 of networkrepresentations , 77 in referentialexpressions , 129 repairof, 461, 491 , 295 Ambisyllabicity Amnesticdisorders , 184 Anacrusis , 308, 373, 393- 394, 396 Anaphora, ISO, 176, 271, 283 Anteriorspeech sounds , 296, 340, 433, 443 errors, 347- 348, Anticipationin speech 358- 359, 372, 468 Anticipatoryretracingin self-repairs, 490- 491

548 errorsand repairs, Antonymsin speech 185, 216, 219, 496 , 139, 143 Apartmentdescriptions , 26, 119 Apperception , 119 Apprehension Appropriateness repair. SeeRepair Areafunction, 445 , 337 Archiphoneme Argument , 169- 170, 174- 178, 189, 191-conceptual 195, 245 external, 169, 182 Arousal(emotional ), 469 , 138- 144. See Arrangingfor expression a/so Linearization Article, 167- 170, 203, 238 definite/indefinite,271- 273 in phonologicalwords, 377 in shifts, 371 Articulation, Articulator, 7, 9, 22- 28, 102, 180, 229, 259, 284- 285, 291, 296, 304, 318, 328, 345, 361, 364, 410- 411, 413457, 473, 475, 497 in delayedauditoryfeedback , 477 effort, 408 interruptionof, 13 latencyof, 230, 259, 473 monitoringof, 462, 472, 476 precisionof, 396, 410 Articulatornetwork, 450 Articulatory buffer, 12, 28, 386, 473 , 295, 297, 318, 326, 346, 362, 409 gesture plan, 12- 13, 24, 28, 297, 318, 329, 386 placeof, 296, 472 units, 332 Artificial intelligence , 124, 205, 207 , 424 Arytenoidcartilage Aspect, 55, 57, 100, 102- 103, 106, 165, 183, 191, 233 sounds , 296- 297 Aspirationof speech Assimilation , 369- 371, 406, 408, 411 Assertion , 3, 10, 17, 300 Assertivespeech act. SeeSpeech act . SeeWord asSociation Association Associative intrusions,214- 215, 219, 221, 234 relationsin lexicon, 183- 184, 234 Attention of addressee , 5, II , 35, 59, 68, 108, 110, 136, 144, 147, 150- 152, 270, 311, 316

Subject Index in lexicalretrieval, 21 in monitoring, 9, 28, 463- 467, 471- 473, 487 andpausing , 127 in planning, 20, 28, 82, 89, 119, 126- 129, 158, 257 selective , 463, 465, 467, 498 span, 119 Attentional effort, 157 , 119, 141, 172, 176, 178, 316 focusing lapse129, 145, 487 pointer, 172 resources , 21, 27, 124, 126, 249, 467 's 44 57 61- 63, 90, Attitude(speaker ), , , 101 102, 171, 307, 311- 364, 366, 402, 478. SeealsoCommitment Audition, 9, 13, 475 Auditory feedback , 443- 445, 474, 477. Seealso Delayedauditoryfeedback goal, 442, 444- 445, 456 , 440, 442, 445, 456 targets , 2, 20- 22, 27Automaticityof processing 28, 249- 250, 259, 270, 282, 447- 448, 477 Auxiliaries,7, 57, 193- 194, 197- 198, 253, 299 modal, 197 reductionof, 302, 364, 375, 377- 378, 406 Awareness , 20- 22 BABEL, 205 Backchannelbehavior,36, 430 , 71, 97, 137, 267, 315, 400 Backgrounding , 235, 253 Backtracking Backwardsuppletionof presupposition , 118, 146 Basicclausepartitioning, 256- 257 Basiclevelterms, 223 Basicmetricalpatternof word, 284, 298299, 323, 373- 374, 382, 384 Basicobjectlevel, 223, 234 Beat, 297- 299, 305, 383 demi, 304- 305 movement , 306, 373- 374, 377, 382, 384, 394 silent, 380- 382, 385, 387, 391, 408, 420 Bernoullieffect, 426- 427 Bilabials,370, 432 Biteblockexperiments , 439, 445- 446, 451- 452

549 Blend. SeeErrors , 110- 111, 114, 140, 143 Bookkeeping tone Boundary , 171, 180, 310, 381, 390, 401- 403, 421 Brain, 18, 445- 446 Breakdownpattern, 22 Breakoption for phonologicalphrase , 304- 305, 308, 373, 379- 381, 385- 387, 390, 400, 402 , 413, , speechversusnon-speech Breathing 423, 425, 429, 447- 449, 455 Call hierarchyin lP G, 244, 252, 256- 259 andplanningunits, 257, 259 Canonicalsettingfor speech , 29- 30, 45, 49, 68 55, Case absolutive , 169 accusative , 89, 163, 243, 251, 253- 254, 263 dative, 163, 181, 193, 243, 248- 249, 254 , 168- 169 ergative marking, 161, 163, 169, 179, 235, 248, 254 nominative , 6- 7, 163, 168, 240, 243, 248, 249, 25I , 253 Casualspeech , 368- 369, 406 Categorialprocedurein lP G, 238- 242, 244- 245, 250- 251, 254- 256, 261, 275, 282 of experience , 78, 94- 95 Categories in selfrepairing convention Categoryidentity , 493- 494 Causativeverbs, 54 C-command , 377 , 61 Challenges Changeof state, 62, 268 errors, 346- 347 Checkofffailurein speech , 138Chronologicalorderin linearization 139, 159 Citation formsof words, 364, 373- 374, 382, 405, 407, 411 Classificatory particles, 104 Clause basic, 256- 258, 283, 303 finite, 256- 258, 308 main, 167, 250, 252, 255, 308 subordinate , 254- 256, 258 surface , 23, 256- 257 Clitic, 307, 375, 317- 378 Cliticization, 302, 307, 364, 370- 371, 373, 375- 378, 380, 406- 408, 411

Subject Index

550

Cluster(of phonologicalsegments ), 293Competingplans, 217 294, 319- 320, 325, 332- 338, 340, 342Competitionmodel. 265. 270 344, 348- 349, 360- 361, 363, 369, 409, , 7, 169- 170, 175- 176, 180, Complement 419, 453 238- 239, 242, 250- 251, 255 nounphrase , 327, 335, 337- 338, 344, , 178, 196 composition 348- 349, 361- 363, 409 sentential , 190. SeealsoS-COMP verbal, 169, 190, 197. SeeqlsoV-COMP spellout, 325, 337, 343- 344, 349 . Coarticulation , 7, 370, 490, 438, 443- 444, , 200- 201 Componentiality 454, 457 , 185- 186, 286, 288- 290, 382 Compound Cocktailpartyeffect,462 , 205, 235 Computermodels Coda Conceptual . SeeAccessibility , 379procedurefor phonologicalphrase accessibility 380 , 169- 170, 172, 175- 176, 178, argument of syllable,293, 294, 324- 325, 327- 330, 189- 195, 245 332- 335, 338, 344, 348- 349, 353, 357, core, 212 360- 362, 407- 410 feature,71, 104, 157, 204 function,.79- 82, 189, 241 Cognitive effort, 128, 157, 344 intrusionandspeech errors, 214- 215, 217, 219- 220, 234 , 22 impenetrability load, 128 modification , 82- 83, 196 , 154, 159 stylein linearization , 147, 149- 151 prominence relations, IS, 27, 184, 216 , 73, 202- 204 system Coherence in lemma, 188- 189, 195- 197, specification 200, 217, 233, 237, 241, 245 , 118, 128 conceptual of discourse , 283 , 6, 26, lOS, 181, 186, Conceptualization 205, 209. SeealsoMessage , 275 syntactic generation Cohesion , 236, 271 , 9- 10, 14, 16, 21, 25, 27, 70, Conceptualizer Cohesive 95, 105, 202, 237 Condition/actionpair (in productions , 236, 271, 461 grammaticalencoding ), 10, reference 17, 72, 124- 125, 207, 240 , 271- 273 syntax, 273 , 163, 166- 169, Configurationallanguage Co- hyponym, 183- 184, 219 179- 180, 270 Collaborationin conversation effect , 65, 133 Congruency Color naming, 131, 183, 207 in usingcomparatives , 228- 229 errorsand repairs,463, 466, 496 in topica1ization , 262, 268 -and-executionstagein articulation in questionanswering Command , 273 , 419 , 266 Congruentgrammaticalencoding Commissives . SeeSpeech act Conjunct, 264 Commitment , S, 61- 64, 101- 102, 123, 134, , 82, 167, 194, 276- 281, 367, Conjunction 314, 402 377, 499 Connected , 57, 62- 63, 68, 101 episternic , 284, 302- 319, 323, 325, speech deontic, 62, 64, 68, 101 329, 343- 361, 364- 412 Commonground, 112, 115, 118, 120, 124, Connectionism , 18- 20, 211, 474- 478. See 132, 135, 146 alsoActivationspreading Communicative Connective , 44, 483 goal, S, 11, 107, 115, 128, 136, 282 , 139- 143 Connectivityin linearization intention, 4- 5, 21, 30, 58- 60, 62, 68- 69, principleof, 140 107- 110, 114, 119, 123- 125, 137, 144, Connotation , 183 157- 159. SeealsoIllocutionaryforce Consonant , 183, 187, 234 Comparatives , 295 ambisyllabic cluster. SeeCluster , Compensatory articulatorymovements 438, 439, 443, 445, 451- 452 co-productionwith vowel, 453

Subject Index

551 Convention , 38, 42, 44, 63, 65, 67- 68 in repairing,492- 494 Conventional forms, 6, 65, 315 rules, 38, 41- 43 , degreeof, 66- 67 Conventionality in lexicalaccess , 199- 20I , Convergence 204, 207, 209- 210, 212, 214, 233 Conversation , 1- 8, 27- 46, 53, 56, 58, 6069, 111- 117, 119- 123, 133, 145, 172, 179, 306, 390, 404, 458, 460, 466, 471 from, 31, 33, 36- 37 closing,disengaging , 42, 58, 63, 111- 112 everyday opening , engagingin, I , 31, 36 rulesof, 460 , 36 sustaining skill, 21 , 30- 31, 38Cooperationin conversation 39, 44, 67, 133- 134 Cooperativeprinciple, 39- 40, 53 , exploitationof, 43- 44, Cooperativeness 133 Coordinationandself-repair, 486- 489 Coordinativestructures , 440, 447- 450, 452, 456 457 Coproductionof articulatorygestures , 444- 445 Coremeaning , 212, 214, 216, 219 Coreprinciple, 213 sounds Coronalspeech , 296, 301, 433, 442, 453 , 424 Cricoarytenoidmuscles , 424 Cricothyroidmuscles Cross-linguisticcomparisons , 265, 268, 272, 283 Culture, 37, 42, 44, 62- 65, 104, 139, 497 Currentnodein activationspreading , 353355, 358- 360, 475, 477 -next, 31, 34, 36, 38 -selects Current-speaker , 126- 128 Cyclesin monologues Dative. SeeCase Deaccentuation , 147- 150, 283 Deactivation , decayof activation,234, 3 359 Deafness , 444, 446 Debate, 112, 126 -makingstage Decision , 228 Decisiontable, 207- 210, 213, 233 . SeeKnowledge Declarativeknowledge Declarativemood, sentence type, 101

Index Subject Declaration . SeeSpeech act Declination , 398- 400, 421- 422, 427 , 256 Deepstructure , 23, 259. SeealsoBasicclause Deepclause Defaultaccent , 177- 178 Definite/indefinite , 114, 120, 152, 197 expressions referent , 99, 146, 170, 175, 267, 271- 273 Deictic adverbial , 48, 69 , 29, 44- 46, 102 anchoring characterof speech , 30 - 53, 69, 153, 154 coordinatesystem , 48, SO origin (origo), 46, 48, 50, 52- 55, 58, 69 model, 53 , 137, 153- 155 perspective pronoun, 5, 45 proximity, 103 region, 47, 52- 54 shift, 45 shift of origin, 54, 58, 69 verb, 54 Deicticreference , 51, 53 primary, 48- SO , 49- 51, 69 secondary Deixis functionof, 54- 5~ acknowledging 53 54, 69 by analogy, , 100- 103 anchoringof utterance coordinationproblem, SO delimitationproblem, 52- 53 discourse , 46 identifyingfunctionof, 47- 48 informingfunctionof, 48- 54 , 5, 45 person place, spatial, 45- 55, 58, 69, 103 social, 45- 46 time, 45- 46, 55- 58, 106, 137 typesof, 44- 46 , 477 Delayedauditoryfeedback Demonstrative , 104 determiner , 47 reference , 48, 53 Dentalspeech sounds , 433 Deonticmodality. SeeCommitment Derivation(morphological ), 183- 184, 286- 287, 289, 316, 326, 344 Destination , in lP G, 239- 253, 256, 258, 282, 303 error, 246, 248- 249, 282 functional, 239, 241, 243, 248- 249 , 252- 253, 256, 258, 304 hierarchy

552 Detectionof errors, 463, 471- 472 Determiner , 169, 170, 194, 196- 198, 246247, 271, 299 Diacriticfeatures , 165, 167, 170, 174, 183, 190- 193, 196- 198, 233, 238- 244, 247251, 275, 288- 289, 321- 322, 324, 329331, 343- 344, 352, 362, 365- 366, 373, 377, 382 Diphthong, 338, 409, 432, 434, 440 Direct/indirectspeech act. SeeSpeech act Directionascategoryof experience , 74, 78, 91, 94 Directionalterms, 53, 91, 153- 154, 195 andgazemovement , 153- 154 Directive. SeeSpeech act Discourse coherence of, 283 context, 53, 110, 115, 158, 183, 461, 463 model, 9, 113- 120, 122- 123, 126, 132, 134, 144- 152, 159, 163, 171, 175, 271, 273 record, 8, 10, 111- 123, 146, 172,460 situation, 10, 107, 110 topic, 99, 112- 113, 115, 12I, 152, 158159, 460 type, 111- 112, 126 Discriminatinginformationin object naming, 131- 132, 134 Discriminationnetwork, 205- 207, 210, 233 Distraction, 216- 217 Distinctivefeature,296- 297, 330, 340- 341, 350, 354, 358, 362, 440- 442, 477, 456 Distributed control, 204 , 20 processing , 211 representation Drawl, 34- 35, 420- 421, 455, 477 , 18, 24, 281, 477, 498 Dysftuency Editing, 458- 499 distributed,467- 469 , 459, 478, 482- 484, 491, 494, expressions 496, 498- 499 lexicalbiasin, 465, 477 , 466, 468- 469 prearticulatory semanticbiasin, 465 syntacticbiasin, 465 theory, 465, 467- 468, 470, 476, 498 of taboowords, 461, 468- 469 Editor, 459, 466, 469- 470, 474 of vocaltract, 449, 453 Eigenfrequency Ellipsis, 153, 253- 254, 273

553

Subject Index , 89- 90, 105 Ellipticalmessages Emotion, 38, 70, 102, 171, 307, 311, 316, 364, 366, 403, 410, 469 Empathy,312 , 300, 404 Emphaticstress , 167, 170, 253, 375 Emptyelements Encliticization , 375, 380, 407- 408, 411 . SeeKnowledge knowledge Encyclopedic , 132- 133, 136, Endophoricredundancy ISO . SeeConversation Engaging Epiglottis,424, 428 . SeeCommitment Epistemiccommitment Ergative.SeeCase Errors

in other-producedspeech , 470- 471 402 398 397 394 , , Euphony, Euphonyrules, 384, 394, 396, 411 Eventasconceptualcategory , Event function, 26, 74, 78- I06pass;m, 157, 159, 163, 188- 189, 241 Eventdescription , 261- 262, 267- 270, 272, 277- 278 Eventtimein deixis, 55- 58 Examinations , 111- 112 Excitement , 102 Executivecontrol. SeeControl Exhalation,295, 422, 447 , 295 Exhalingspeechsounds , 132- 134, 136 Exophoricredundancy , 497 Expansions , 422, 472 Expiratorymuscles act . SeeSpeech Expressives functionof prosody, 306- 307, Expressive 317 Exploitationof maxims,43- 44 Externalargumentof verb, 169, 182 Externalloop in monitoring, 472- 474 Extrinsictiming, 436, 444 Eyebrowflash, 37 Eyecontact, 35. SeealsoGaze Facial expression, 68, 430 Falsetto, 424 False vocal chords. 424

Fastspeech , 302,368- 371,385,409,487 , , 1- 2, 8, 12, 18, 24, 28, 30, 34, 128 Fluency - 204,222,239,245,282, - 128 127 , 199

SS4

Subject Index

, 430 Genioglossus Gesamtvonte Uung,23. 26 Gestaltof object/scene . 131- 132. 156. 159 Gesture . 8. 23. 35. 45. 47. 53. 59. 68 articulatory. 291. 295. 297. 318. 326- 327. 346. 362. 364. 388, 409. 414. 434- 435. 438. 440. 444, 446 Giveninformation. 6. 98. 114. 311. 315 Givenness . 99. 100 Glottal speechsounds . 381. 433 Glottal stop. 381 Glottis. 309. 424. 427- 429. 433. 445 Goal communicative . 5. II . 107. 117. 128. 136. 282 elaboration . 127 Grammar. 192. 236 , 70 message IncrementalProduction . 235. 249. 282 LexicalFunctional. 162. 235 of repair. 489 Grammatical . 9. II . 105, 179. 181- 283passim . encoding 375, 422 relations. 15. 27. 161. 179. 285 Grammaticalfunction. 97. 152. 161- 163. 167- 170. 179- 180. 189- 190. 192- 195. 198. 222. 233, 235- 236. 238- 239. 244245, 250. 256. 260- 261. 264- 270. 275. 279. 281. 283 hierarchyof, 192. 194. 233, 260. 266. 269. 275 Gravityanddeixis. 49 Grid. metrical. 298- 299. 304- 305. 317. 323. 366. 374. 378. 380- 384. 411

Galvanicskin response , 469 Gap (pause ), 33, 78 Gaze in convenation, 8, 35, 36, 59, 68 movementanddi~ tional tenDs, 153- 154 tour, in scenedescriptions , 153

Gender , 168 , 231,238 Generic verb,206

Haplology. SeeErron Harmonics . 427 Head/modifierstructures . 79. 82. 89, 96 Head-of-phrasefunction. 167- 168. 238 Head-of-phraseprocedure . 239. 247 lateralization . 22. 154 Hemispheric Henderson graph. 127 Hesitantspeech . 127- 128. 204 Hesitation.5- 6. 203- 204. 257- 259, 460. 484, 494 , 203, 204, 421 prelexical Homonym,208 , 44 Hyperbole

, 183 , 216,220 Hypemym , 201,204,207,210, Hypemym problem 212- 214,233

555

Subject Index I-marker, 23 Idea -motor, 94 sensory unit, 23, 36 Idealdelivery, 128, 384- 385, 391, 402

. SeeDeixis function Identifying - 187 Idiom. 186

Illocutionaryforce, 58- 65, 108, 123- 124,

151 . 157 . 398 . 402

andaccessibility , 264- 266 Imageability - 101 , 106 , , 63- 65, 67, 71, 100 Imperative 171 240 170 144 157 109 , , 403 , , , Implicature conventional, 44 conversational, 41- 44, 66- 67, 124, 138 Inclination of pitch, 404

Incremental , 157 , , 2, 23- 28, 109 production 245 253 235 239 162 , 365 , , - 254 , 284 , , - 384 371 , , 405 , 411 , 398 , 400 , 396 , 373 Incremental Production Grammar , 235,

249 . 282

field.46 Indexical Indignation . 30. 102. 171 Indirect speechact. SeeSpeechact Inferability of referents. 99- 100. Seealso Accessibility Infinitive . SeeTense

Inflection , 198 , 251, , 161 , 165 , 183 , 55, 104 286,288-290,316,322,344

Interference . 217- 218. 227. 229. 234 conceptual plan-internal. 222 Interjectionin repairs.482. 484 Interlinear-toneticnotation. 309- 310 Internalloop in self-monitoring. 472- 473 Internalspeech . 12- 15. 21. 27- 28. 469473. 498 InternationalPhoneticAlphabet. 12. 295. 500 . 63- 64. 67. 78. 100- 102. 106. Interrogative 109. 144. 157. 170- 171. 240. 403 . 28. 112- 113.459. Interruptionof speech 473. 478- 48. 489. 492 . 478. 480. 482. 484. 490- 491 delayed immediate . 479- 480. 495. 498 Main InterruptionRule. 478. 480- 482. 498 within-word. 479- 481. 484 Interview. 3- 5. 111- 112. 126. 158. 391 Intonation. intoning. 15. 18. 23- 24. 33. 62- 64. 102. 108. 165. 166. 171. 179- 180. 255. 290. 306- 307. 315- 317. 365. 368. 406- 407 generationof. 398- 405 Intonational . 102. 171. 312- 317. 366. 405 meaning lexicon. 171. 316 phrase . 259. 303- 304. 306- 311. 314- 317.

clitics versus . 378 asdiacritic feature . 183 . 191 . 256

Junctureof words, 302 Key, 315- 317, 366, 385, 398, 400, 411 Kinaestheticrepresentations , 73 Kinshipterms, 183, 207, 210, 212 Knowledged ~ larative, 10- II , 18, 72- 73, 182, 185, 236

Subject Index Knowledge(cont.) , 9- 10 encyclopedic mutual, shared , 8, 14, 29, 44- 45, 47, 69, 84, 114- 115, 123, 125, 146, 158 mode, 72- 74 , 9- 11, 72, 105, 124- 125, 149, procedural 185- 186, 236 , 72- 74 representation situational, 10, 72 store, II , 28

556

Lemma,6 access , 198- 213, 320 failure, 214- 221 access substitutionin self-correction , 489 , 134- 135 Legitimacyof request Levelsof processing , 231- 232, 234, 319, 321, 330, 343- 345, 361- 362, 367, 369, 468 Lexicalaccess , 186, 199, 200, 205- 206, 213- 214, 230, 234, 277, 468 failureof, 214- 221, 462 Labialspeech sounds theoriesof, 199, 201, 206, 211 , 295, 340, 442 Labio-dentalspeech sounds timecourseof, 181, 222- 232 , 433 Lexicalaction, 208- 210 Languages Cantonese Lexicalbiasin speech errors, 355- 356, 359, , 268 Danish, 307 363 Dutch, 71, 103, 122, 130, 140, 167, 220, editingaccountof, 465, 477 237, 238, 307, 350, 433, 483, 485 activationspreadingaccountof, 356, 359 Lexical Dyirbal, 55 Finnish, 185, 268, 289 , 167, 288 categories French, 231. 238, 297, 302, 307, 367- 368, , 186- 187, 233 encoding 387, 392, 429, 433 entry, 182- 183, 186- 188, 231, 234, 322, German,26, 45, 102, 165, 167, 179, 238, 344 248- 249, 254, 263, 265, 289, 350 error. SeeErrors Hausa, 50 form, 12, 231, 258, 285, 320- 321 Hebrew,268, 433 gap, 209 , 265, 268, 272- 273, 289 , 181, 235 Hungarian hypothesis inflectional intrusion, 184 , 289 Italian, 262, 272- 273 pointer, 165, 167, 174, 180, 193, 195, 344, 362, 373, 375- 378 , 103- 104, 186, 286, 272, 289, 329 Japanese Kannada,269 LexicalFunctionalGrammar, 162, 235 Lexicalrule(passive Kiliyila, 104 , dativeshift), 193 Lexicallydrivenformulation, 181, 198, Malay, 104 235- 236, 282 , 163, 167, 180, 270 Malayalam Lexicon Quiche,497 intonational, 171, 316 Serbo-Croatian, 268 lemma,9, 187, 199 , 53, 102- 104, 268, 392- 393, 433 Spanish mental, 6, 11- 13, 18, 22, 27, 162, 167, Swedish , 307 181- 234, 290, 302, 316 stresstimed~373, 392 393 form, 12, 163, 187, 258, 284- 301, 323, syllable-timed, 392- 393 329, 344, 352- 355, 363, 368. 373. 408 Thai, 497 Turkish, 185- 186, 268, 289 Liaison, 295. 302. 367- 368. 454 Linearization Tuyaluan,497 . 107. 138. 138- 144, 153. 159. 460 Warlpiri, 169 -specificconceptualizing content-relateddeterminants . 138- 139 , 103Language 105, 157 , 138. 140. 143 principles -relateddetenninants of thought, 71, 73- 74, 105 . 138- 144 process Language , 153. 159 ), 2- 3, 33 Lapse(pause strategies sounds , 295, 316 Lip. 295- 296. 413. 428- 430. 432- 434. 436. Laryngealspeech 440. 442, 446. 449- 452 , 295, 413, 423- 430 Larynx, laryngealsystem 30 /rounding. 409. 430. 434. 443 spreading Lecturing,

Subject Index Listingintonation. 386 in turn-taking. 34 Localmanagement Locationprogrammingin articulation. 436. 439- 440. 455 . 82 Logicaloperations Logogentheory. 201- 204. 214. 219. 230. 233 Long-termmemory. 10. 21. 72. 119. 121. 126, 351- 352 Longitudinalthyromuscularis. 424 Lookahead . preview. 18. 24- 25. 28. 235. 373- 385. 389- 390. 396. 398- 400. 402406. 411 Loudness . 7. 22. 108, 309, 364, 367, 369370, 396, 409- 411. 427, 455, 462, 495 Lungs,413, 422- 423 Lying. 40 , 5, 11. 107, 109- 110, 123Macroplanning 144. 152, 157 158 Main InterruptionRule. 478. 480- 482. 498 Main/sidestructure, 136- 137 . SeeErrors Malapropism Mandible,424 Manner , 74, 78- 79, 82- 83, category conceptual 92, 96 maxim. SeeMaxims Marked/unmarkedadjectives , 229- 231 Mass-springtheory, 438- 440, 448, 450, 452, 455- 456 Massetermuscles , 45I Mastication , 413, 430, 448 Maximizationof onset, 294, 335 Maxims, 39- 44, 69, 129, 158 manner,41 exploitationof, 43- 44 floutingof, 43, 124 quality, 40, 43, 60 quantity, 39- 40, 42, 123- 124, 129- 130, 133, 140, 148, 213 relation, 40, 42, 66, 112 Meaning core, 2.12- 213, 216 intended , 6- 7, 13 intonational, 102, 171, 308, 310- 311, 314- 317, 366, 373, 405 lexical, II , 13, 182- 188, 197, 200- 201, 212, 216- 217, 232, 236 , 364 propositional relationsin lexicon, 183- 184, 188, 215

557 Melody of segments , 290 of sentence , 301, , 171 , 180 , 18,63, 102 306- 307,311,317,398.Seealso Intonation Memory limitations , 10 long-term.SeeLong-termmemory , 121 for other-produced speech , 119 primary search , 158 , 126 - 122 , 121 for self-produced speech sentence , 74 . SeeWorkingmemory working Mental . SeeLexicon lexicon load, 259 model , 72, 113 - 154 tour, 142 , 153 , 16, 28, 70, 72, 107 Message generation 160 , 44 Metaphor Metrical grid. SeeGrid , 395 phonology , 304,317,373,397- 399,402, planning 408,410- 411 rules , 299,382 of connected structure , 303- 307, speech 364- 365,372,374,387,408 tier, 290,293,297- 299,301 - 110 , 126 , , 123 , 5, II , 107 MiC ;roplanning - 160 - 129 128 , 132 , 144 of linearization , MinimalLoadPrinciple 143 Minimalnegation , 212 Modalverb,62, 101 , 403 , 198 - 171 - 102 , 180 , 402. , 170 , 64, 100 Modality SeealsoAttitude . See Modeof knowledge representation Knowledge control,445- 447,451Model-referenced 452,456 - 170 , Modification , 79, 82- 83, 88, 96, 169 196 , 22 Modularity Monitoring connectionist theoryof, 459,474- 478 editingtheoryof, 459,467- 474 other-, 470-471,473 self-, 8- 10, 13- 14, 16,21, 28, 70, 458- 478, 498- 499

Subject Index , 126- 128, 147 Monologue Mood. SeeDeclarative , Interrogative , Imperative of verb, 63- 64 , 182, 185- 186, 285, 300- 301, Morpheme 319- 326, 329- 331, 344, 349, 353- 357, 362- 363, 370, 372, 375, 378, 406 affix, 104, 183, 185, 286- 290, 294, 299, 323- 326, 329- 331, 344, 349, 352- 353, 355, 362, 372 prefix, 285, 287, 353- 354, 362 root, 12, 185, 285- 290, 294, 299, 307, 329- 331, 344, 349, 354- 355 Morphology, 12, 15, 100, 102, 104- 106, 163, 169, 182, 193, 248, 271, 273, 286290, 316, 462, 470 /metricalspellout, 321- 324 Morphological , 187, 191, 301- 302 Morpho-phonology Motherese , 368 Motivationof speaker , 8, 20 Motoneurons , 437. SeealsoAlpha/gamma loop Motor controlof speech . SeeControl cyclicityof, 448 hierarchalpropertiesof, 448 Motor controltheories auditorytargets,445- 447, 456 coordinativestructures , 440, 447- 452, 456 locationprogramming , 436- 440, 455 -spring, 438- 440, 448, 450, 452, 455mass 456 orosensory goals, 442- 445, 456 Motor program, 28, 327, 396, 414, 416, 419, 421, 434- 435, 441- 442, 447, 449 Mouth, 13, 424, 429- 431, 435, 439, 445, 448, 451, 455- 456 Muscle-movementnodes , 475 Musclespindle,437, 443, 457 Mutual belief, 133 . SeeKnowledge knowledge Naming, 131, 158, 206, 222, 225- 232, 234, 277, 415, 463, 488 latency,207, 223, 225- 232, 277, 415 Narration, III , 126, 136 Narrativeability, I Narrowfocus, 177- 178 Nasal cavity, 296, 413, 423, 428- 430, 455 sounds , 290, 296- 297, 334, 340, speech 346, 350, 429, 432- 433, 441- 442

558 Nasopharyngeal port, 429 Naturalorder(in linearization ), 41, 138139, 142 principleof, 138, 159 Naturalreference points, 155 Negation,82, 212, 256, 314 minimal, 212 Negationtest, 212- 213 Network articulator, 450- 45I connectionist , 18- 19, 353- 354, 363, 474- 476, 498 . SeeDiscrimination discrimination , 75- 77, 79 propositional spatial, 140, 203 Newinformation, 5, 109, 114, 120, 151, 159, 172, 176. SeealsoGiveninformation Newsvalue, 71, 147, 149- 159, 171- 172, 400 Node-structuretheoryof monitoring, 474478 No-markingerrors. SeeErrors Nominative.SeeCase , 163, 168Nonconfigurationallanguage 169, 179- 180, 270 Nose, 428 Noun common, 196- 197, 199 count, II , 197, 238, 247 mass , 197 proper, 196, 200, 250 Nuanceediting, 482 Nuclear entities/elements , grammaticalencoding of, 97- 98, 152, 260- 271 , 391, 401- 403 pitch movement stress , 176, 179- 180, 306, 383- 385, 389391, 402 stressrule, 305, 308- 309, 382- 384 thematicstructure,96- 97 tone. SeeTone Nucleus of syllable,293, 325, 327- 328, 331- 335, 338, 344, 348- 349, 353- 354, 360- 362, 407- 410 of intonationalphrase , 309- 317, 386, 401, 403- 404 Number, 47, 105, 165, 170, 183, 191, 197, 233, 238, 240, 242, 247, 249- 250, 273, 275, 322 plural, 47, 168, 185, 187, 197, 241, 247249, 288, 301, 324, 326, 331, 344, 378

Index Subject singular,47, 168, 187, 197, 238- 240, 242, 289, 301, 344 Numerals , 104, 273 's 3 ' , , , speakers andaddressee Obligations 20, 31 32, 61, 63 65, 135 Object. SeealsoCase , 223 categorization naming, 130- 131, 158, 232. Seealso Naming orientation,49, 154- 155 , 230 recognition reference , 133- 134 size, 155, 228- 229, 276 Oblique. SeeCase Obstructionof pitch, 300. SeealsoPitch accent, movement Obstruents, 334, 434 Omissions. SeeErrors Onset

- 294 - 344 ofsyllable , 290 , 293 , 299 . 324 . - 349 347 353 354 357 359 362 , , , . 367 .

369, 407- 410 maximizationof, 294, 335 of word. SeeWord Oralcavity, 413, 428, 430, 445, 455 Orbicularisoris, 430 Orderingerrors. SeeErrors . SeeLinearization Orderingfor expression Origin, origo. SeeDeicticorigin Orosensory goals, 442- 445, 456 in objectnaming, 129, Overspecification 158 Overtone , 427 Overtspeech , 2, 9, 12- 14, 16, 21, 27, 469472, 474, 498 Palatalspeech sounds , 433 Palate(hard, soft), 423, 429- 431, 433 sounds Palato-alveolarspeech , 433 Parallelprocessing , 2, 20- 21, 24, 27- 28 in articulation, 449 in grammaticalencoding , 238, 240, 242243, 245- 246, 248- 251, 256, 259, 282 in lexicalaccess , 199- 202, 207- 211, 222, 233- 234, 372 in phonologicalencoding , 355, 363 Parenthetical , 308, 386, 408, 497 Pars-pro-toto principle, 124 Parsedspeech (in monitoring), 13, 26, 470 Particle,64, 375- 376 , 104 classificatory

559 Passive , 169, 181, 262, 268 elicitationof, 261- 264, 266, 268, 488 lemma,245 lexicalrule, 193- 194, 233 syntacticprimingof, 274- 275, 278 truncated , 275 Path, asconceptualcategory , 90- 91, 9496, 178, 189, 195, 237 Pathfunction, 90- 91 Patient. SeeThematicrole Pause , pausing , 23, 33, 35- 37, 126- 128, 203, 236, 257- 259, 305, 308, 364, 366367, 372, 381, 385- 387, 390- 391, 409410, 420, 459, 482 -loop theoryof monitoring, Perceptual 469- 474, 476, 498 Performativeverbs, 64- 68 Perseveration , 331, 347, 358- 359, 468 Personasconceptualcategory , 74- 98, 113- 115, 189, 241 asdiacriticvariable, 191, 197, 233, 238, 240- 242, 250, 273, 289, 322, 331 Persondeixis. SeeDeixis Perspective 's 52 addressee , informational/thematic,5, 11, 70, 96, 105, 109- 110, 132, 138, 145, 152- 159, 260, 269 ' speakers, 94 visual, 226 Pharynx,428- 430 Phonation , 424 Phonationperiod, 127 Phone,292, 294- 297, 328, 346, 362- 363, 438, 440- 441, 444, 472 Phoneme , 22, 211, 292, 298, 301, 320- 363 , 464, 475, 477 passim Phonemic clause , 23 code, 297, 300 similarityconstraint,350, 358 Phonetic feature,295- 297, 316, 350, 428, 432- 434, 440, 442, 455 , plan, 7- 8, 12, 24- 28, 284- 421passim 434- 436, 444, 449, 454- 457, 470, 473. SeealsoArticulatoryplan Phoneticspellout, 319, 321, 324, 326- 329, 332, 335, 338, 344, 348- 349, 361- 363 in connected , 364- 370, 375, 387, speech 389, 400, 405- 412, 417, 436 Phonicduration, 387- 392

560

Subject Index Phonological , 7, 11- 12, 18, 20, 24, 26, 102, encoding 161- 162 , 165, 166, 171, 179, 181, 187, 231, 233, 236, 239- 242, 246, 255, 259, 275- 276, 278- 279, 282- 284, 303, 316, 318- 412pa.rsim, 414, 416, 419- 422, 435- 436, 462, 468- 469 feature,23, 350 , 259, 303- 307, 317, 373, 378- 381, phrase 390- 391, 411, 419- 421, 454, 473 priming, 279- 280, 354 word, 302, 317, 361, 370- 380, 389- 392, 398, 406- 421passim , 454, 473 Phonotactic possibility, 294, 335- 338, 344, 362, 369, 482 rule, 294 PhrasalFocusRule, 174- 176, 178- 179 Phrase . SeeIntonational, Phonological , Phrasestructure Phrasestructure, 163, 166- 167, 169, 179180, 365, 369, 373, 384, 391, 408, 498 Pitchaccent , 7, 12, 100, 149- 151, 166, 174- 178, 237, 244, 299- 300, 306- 307, 310, 315, 317, 323, 364, 374, 382- 386, 389, 398- 400, 404- 405, 410. Seealso Accent, Accentuation andaccessibility of referent , 270, 273 contrastive , 172, 177, 247, 273, 307, 309, 401, 495- 496 default, 177 asdiacriticfeature, 165, 180, 191, 233, 366, 373, 382, 408 andfocus, 100, 174, 243, 246, 496 monitoring, 463 andprominence , 307, 386 andnuclearstress , 308- 309, 384- 385, 402 Pitch-AccentProminence Rule, 382 Pitch-AccentRule, 175 Pitch contour, 7, 171, 255, 300, 307- 310, 316, 364, 366, 373, 398, 405, 411, 427, 498 modulation,413 movement , 298, 309- 310, 327- 328, 367, 372, 381, 385, 391, 396, 399- 405, 409410, 427 obstruction , 300 raising, 113, 315 , 247, 282, 372 stranding withholding , 147 Placefeatures , 295, 316, 350, 443

Planning articulatory , 297, 420- 421 conceptual, S, II , 20, 24, 204. Seealso Macroplanning , Microplanning of connectedspeech, 364- 412 of discourse, 115, 119

, II , 119, 126, 259, 275, 467 message metrical, 304, 317, 373, 397- 399, 402, 4OS , 410- 411 motor, 420, 435, 443- 444 phonetic,296, 31S,321, 329, 341, 343346, 359, 365, 473 prosodic, IS, 364- 365, 396, 405, 410 - I09, 474 speechact, 10S andsyntax, 25S- 259 unit, 23, 256- 257, 259, 420 utterance , 3- 4, IS, 277, 2S2 , 112 Planningdiscourse Pleas , 61 , 61, 64 Pledging Plosivespeech sounds , 290, 413, 433, 441, 472 Plural. SeeNumber Politerequests , 66- 69, 135 Politeness , 3- 6, 21, 30, 37, 65- 69, 124,474 Posteriorspeechsounds , 296, 433 , 39 Pragmatics Prearticulatoryediting. SeeEditing Precisionof articulation, 22, 396, 410, 434, 462 Predication , 70- 71, S9, 90, 9S- 99, 113- 115, 150- 1S4 , 159, 171, 261, 30S Prefix. SeeMorpheme Prenuclear tune, 311- 312, 403- 405 , 4S, 122, 166- 167, 17S Preposition , 192 , 194- 195, 19S , 236, 251, 273, 2SS , 369, 371, 37S,421 Prepositionalobject, 23S,248, 264, 274 , II , 167, 169, 17S Prepositionalphrase , ISO, 192, 195, 23S,266, 26S, 304, 485, 487, 490 Presentatives , 272 in self-repairs,491 Prespecification , 102, 113, 117- 119, 171, Presupposition 175, 212, 216, 219. SeealsoBackward suppletion existential , liS factive, liS Pretonicaccent.SeePrenuclear tune Preverbalmessage , 9, II , 14- 15, 72- 74, 76, 7S, 93, 101, 105, 107- 110, 157, 160, 170, 179- ISI, IS6

561

Subject Index Priming andactivationspreading , 352, 354, 474477, 498 lexical, 221- 226, 230, 275, 476 , 279- 280, 354 phonological - 275 273 , syntactic Principle of connectivity , 140 of contrast, 200, 213 , 39- 40, 53 cooperative core, 213 linearization , 138, 140, 143 naturalorder, 138, 159 of specificity , 213- 214 , 213 uniqueness . SeeKnowledge Proceduralknowledge . SeeCategorial Procedure , Functional Processing capacity,20 , 1- 2, 8, 13- 15, 20- 24, 28, 70, components 365- 368, 468- 469 , 24, 234 stage Procliticization , 376- 377 Productions(condition/actionpairs), 17- 18, 82, 124, 207, 321 Pro-form, 148- 149 units(in turn-taking), 31, 35 Projectable Prominence , 147, 149- 151, 165, 172, 175conceptual 176, 178, 180, 236- 237, 243, 266, 280, 307, 398 contrastive , 158, 172, 176, 307, 398, 496 , 131, 147, 149, 152, 171, 180, prosodic 305- 306, 309, 311, 382, 384, 386, 401402 , 260- 261, 264- 266, 268, 275, syntactic 278 281 , 60- 61, 64 Promising Pronominalization , 99, 149, 271- 273, 283 Propositional format/modeof representation , 71, 105, 108, 110, 144- 145, 152- 155, 159, 188 network, 75 , 145 Propositionalization in articulation, 445, 451 Proprioception , , 323, 365- 412passim ProsodyGenerator 416- 417 Prosodic focus, 171- 172, 176, 271, 496 planning, 18, 364- 365, 396, 405, 410 stress , 132, 273. SeePitchaccent unit, 259, 373, 436

Prototype,84 attitude. SeeAttitude Psychological reality, 79, 124, 206, 257 Psychological Quantifier, 76, 196 , 30- 31, 35, 38, 60, 65- 68, 89, Question 118, 255 , 36, III , 122, ISO, 172, 179, answering 221, 254, 273 intonation, 102,403- 404 tacit, 78, 86 tag, 34, 308, 386 andtopic, 262- 266 wh-, 101, 404 yes/no, 101, 255

Radiotalk, 29, 112 verb, 190 , 250 Raising Rate(speed ) of activation , 35I, 359,366 spreading -

of errordetection , 466, 470- 473, 498 of lexicalaccess , 199- 20I , 206, 225, 228, 233- 234, 277 of grammaticalencoding , 246, 454 of processing , 26, 28 of speech , 22, 108, 199, 222, 306- 308, 316, 355- 359, 363, 366, 368- 370, 374- 375, 385- 386, 389- 391, 392- 393, 407, 410411, 413- 414, 422, 432, 447, 449, 453, 457 Rationality, 41, 67- 68, 158 , 259, 308, 390- 392, 394, 396, 400, Reading 405, 415- 416 Real , 15- 16, 18, 28, 206, -. -timeprocessing 210, 212 intonation, 313, 402 Reassuring Recall(verbatum ), 121- 122, 263- 264 . SeeThematicrole Recipient of conceptualstructure,82, Recursiveness 87 in objectnaming, 129- 132, Redundancy 158. SeealsoEndophoric , Exophoric Reduction , 166, 302, 364, 369- 371, phonological 376- 378, 388, 406, 409, 464 of referringexpressions , 5- 6, 147- 150, 176, 271- 272 Reference deictic. SeeDeixis demonstrative , 48, 53 , 51 geographic intrinsic. SeeIntrinsic

Index Subject

of contribution. SeeContribution ~ time time. SeeReferen Repair, 36, 220, 253, 304, 458- 499pa.rsim . 459, 461, 474, 481, 483, appropriateness 485, 490- 491, 495- 496, 499 covert, 478- 479, 483- 484, 492, 494 error, 459, 464, 490- 491, 496, 499 instant, 490- 492, 495 O(x:aSionof, 490- 491, 499 " on the ," 245, 497, 499 fly overt, 4 ) , 479- 480 prosodicmarkingof, 495- 496 syntaxof, 485- 489 of, 486- 487, 489, 491, well-formedness

562

basicclause , 257- 258, 282 sentence , 165, 166, ISO Rimeof syllable,290, 293- 294, 299, 324325, 332- 334 Ritual, 30, 37 Roaddi~ tion. SeeRoute Role. SeeThematicrole Root. SeeMorpheme Routedirection, 30, 109- 110, 112, 126, 136, 139, 152 Roundingof lips. SeeLip

Subject Index Selectinginformation for expression, S, 128- 137, 158 Selectionalrestriction, 189 Selectiveattention. SeeAttention Self-interruption , 459, 478, 480- 482, 484, 489, 498 Self-monitoring , 458- 478 Self-selectionin turn -taking , 32- 34, 37, III Semantic bias in monitoring , 465 categories, 79, 86, 234 cohort , 232 decomposition, 93- 94, 96 domain, 95, 209 entity, 74- 89 field, 96, 183, 219 function , 71, 87, 93, 102, 105 markedness, 229- 231 opposition, 219 relation , 74- 75, 161, 184, 207, 218- 22(J representation, 73- 74, 77- 78, 94, 105 test, 205, 207 type, 77, 86- 90, 253 unit , 35- 36 Semiotic extension, 94 Semi-vowel, 434 Sensory-analysisnodes, 475

Sentence boundary, 171 . SeeConstituent constituent memory,74, 263 mood, type. SeeMood topic, 98- 99, 194, 265 (versusserial) testing, 199, 206. Sequential 233 Serialorderin phonologicalencoding , 353- 354, 360- 363, 417 Setetfects in monitoring, 477 in naming, 134 Set-system , 96 . SeeKnowledge Sharedknowledge Shift. SeeErrors, Deictic Shriekyvoice, 102 Sideactivity, 37 Sideintention, 137, 159 Sidestructure, 136- 137 Signalingactivation, 359 Silence , 3, 5, 8, 32- 33, 38, 127- 128, 258, 387, 391, 398, 420. SeealsoPause Silent(demi) beat. SeeBeat Simplereaction,200, 214, 416- 418, 473

563 condition, 60- 62, 67 Sincerety . SingularSeeNumber . SeeKnowledge Situationalknowledge Skeletaltier. SeeTier

Subject Index activation. SeeActivation Spreading spreading Spurt, 23 Stackprinciplein linearization , 143 . SeeProcessing Stages Standard linguistic, 30, 410- 411, 418 of conversation , 31, 39, 41, 63, 61, 461 Statefunction, 80, 251 Statusof addressee , 45- 46, 69 Stem, 180, 183- 185, 288, 290, 323- 325, 329- 331, 344, 348- 349, 352, 354, 362, 312, 318 Stimulusonsetasynchrony , 226, 230, 234 Stops,433, 441- 442 Storytelling, 131 . SeeErrors Stranding StressS, 12, 298, 310, 403. SeealsoPitch accent alternation,299, 306, 314, 394- 396 , 298- 299, 306, 325, 382- 384, assignment 390 clash, 299, 306, 314, 381, 384, 394- 396 contrastive , 241 error. SeeErrors group, 420 nuclear.SeeNuclear pattern, 25, 186, 298, 301, 323, 344, 355, . 313, 382, 391- 398 . Prosodic See prosodic rules, 298- 299, 305- 309, 314, 382- 384, 390, 394- 396 strandingof, 241, 312, 282 syllable,291- 299, 306, 308- 310, 311- 318, 320, 325, 328- 329, 332, 341- 342, 344, 359, 369, 312- 314, 381- 390, 392, 391, 404, 408, 411 tertiary, 309- 310 word, 25, 255, 281, 293, 298- 301, 311, 324, 315, 381, 384, 389 Stridentspeechsounds , 296, 30I , 434 , 169- 110, 115, 180, Subcategorization 194- 196 , 399, 422, 421 Subglottalair pressure Submissiveness , 102 . SeeErrors Substitutions Subvocalspeech , 285. SeealsoInternal speech , 12, 295, 291, 316, system Supralaryngeal 413, 422- 423, 428, 434, 455 Surface . SeeClause clause

564 structure.7- 8. II . 15. 17- 18, 24, 26- 28. 161- 180. 191, 198. 217, 222, 235- 283, 284- 285. 299, 302- 303, 305- 308, 317318. 322, 343. 353, 362, 365, 368, 370371. 375- 379. 380- 381, 383- 384, 386. 398, 400- 406 Swallowingreflex, 448, 450 Syllabicity,292, 295, 323, 365. 370- 371,409 , 284, 294, 318, 365, 375, 405. Syllabification SeealsoResyllabification Syllable articulation, 452- 454 boundaries , 22. 292, 294, 300. 332, 481482 coda. SeeCoda . SeeConstituent constituent duration(length), 298, 318. 327, 364. 367. 369- 370. 373, 381, 387- 396. 398, 409411, 416- 417. 436, 449 heavy/weak, 299, 316 inventory, 318, 327 latencyeffect, 414- 419 node, 18- 19, 292, 357- 358, 475 nuclear,309, 313, 325, 402 nucleusof. SeeNucleus omission , 396- 397 onset. SeeOnset program, 319, 327, 361- 362, 417, 436 rime. SeeRime . SeeStress stressed / unstressed tier, 290, 292- 295, 298, 301, 325 unit, 292, 452 -timedlanguage . See / stress Syllable Languages , 447 Synergism , 183, 200, 213, 220, 234, 254 Synonym Syntactic bias281, 465 buffer, 12, 13, 26 , 168, 249, 275 specialist , 23 Syntagma Syntax,26, 35- 36, 182 cohesive , 271, 273- 275 andmeaning , 88

Index Subject Taskvariablesin motor theory, 45 Teeth, teethridge, 430, 433, 439, 445, 455 talk, 29, 37, 66 Telephone Tense function, 87, 157 future, 45, 57, 61, 64, 101, 103, 137 past, 46, 55- 56, 94, 182, 187, 241, 323, 326, 348 perfect,46, 55, 193, 197 , 55- 56 pluperfect 55 , present - 57, 182, 258, 289 presentperfect, 56- 57 primary, 55 Tensedverb, 171, 197, 242 Text-to-grid alignment , 382 Thematicroles, 89, 90- 95, 150, 189, 192, 233, 254, 266 actor, 11, 91- 94, 96, 105, 150, 265, 267 agent, 71, 91- 93, 95- 98, 105, 152, 189, 192- 194, 233, 241, 261, 262, 265- 269, 487 , 4, 6- 7, 92, 96, 190, 250 experiencer goal, 71, 90- 96, 98, 105, 178, 189- 190, 192- 195, 242, 266 instrument , 92, 96 89- 90, 92, 96, 149- 150, 192- 193, , patient . 253, 261- 263, 269 recipient, II , 21, 92, 95- 97, 190, 192, 261- 262, 266, 268, 487 source , 71, 90- 96, 105, 189, 192, 266 theme,90- 92, 94--96, 152, 172, 175- 176, 189, 192- 194, 241- 242, 245, 250, 266 Thematicstructure,70- 71, 90- 97, 105, 283 , 112 Therapeuticdiscourse Thorax, 422 Threshold(in logogentheory), 202- 204, 219 Throat (pharynx), 428- 430 , 424 Thyroidcartilage Tier (phonological ) feature,291, 297, 316 intonational,290, 299- 301 metrical, 290, 293, 297- 299, 301 , 290, 292- 297, 301 segment skeletal(or timing), 290- 294, 301, 316, 323 syllable,290, 292- 295, 298, 301, 325 Timbre, 413, 427- 430, 432, 455 Time adverbials , 58 calender , 58 deictical, 102

565 reference , 55- 56, 94 relevance , 56- 57 , 127 speaking utterance , 56- 57 Timecourseof lexicalaccess , 222- 231 , 405, 416, 436, Timingof motorexecution 439, 442- 444, 449, 473. Seea/soIntrinsic , Extrinsictiming , 188, 231, Tip- of-the-tonguephenomenon 319- 321, 323, 342, 355, 361 To-contraction , 376 Tone, 308, 310- 317, 366, 385, 390, 398405, 410- 411, 421 boundary.SeeBoundary fall-rise, 313- 314, 401 full-rise, 313 high-fall, 312, 404 level, 314 low-fall, 312- 313, 403, 404 low-rise, 313 nuclear,311- 312, 315, 317, 385, 398- 405 rise-fall, 314 Tonegroup, 23 Tonelanguage , 317 Toneunit, 23 Tongue,413, 423, 429- 430, 439- 440, 445446 blade, 296, 433 body, 430, 432, 453 extrinsic/intrinsicmuscles , 430 tip, 442 twister, 341, 343, 471- 472 Topic changeof, 33, 36- 37, 111, 398, 400, 404 discourse , 99, 112- 113, 115, 121, 152, 158- 159,460 encodingof, 260- 270, 283 of message , 11, 71, 97- 99, 108- 109, 120, 151, 238, 245 of sentence , 98- 99, 194 shift, 112- 113, 149 , 138, 144, 152, 233, 265 Topicalization Trachea , 423- 424 Transition-relevance place, 31- 33, 35- 36, 38 Tune, 311- 314, 316- 317, 366, 398, 403405, 410- 411 Turn allocation/assignment , 31, 33- 34 unit, 23, 31, 35 Turn-constructional Turn-taking, 29- 32, 34, 36- 39, 44, III , 133, 306, 373, 390 Turn-yieldingcues,34

566

Index Subject

, 79, 82- 83, 86- 9() Typetheory , 83- 86 Type/tokendistinction Underspecificationin object naming, 40, 130, 133 Uniquenessprinciple, 213 Unit Similarity Constraint , 329- 332, 334,

360- 362 Unitsof processing , 23- 26 Unpackingthemotor program,416- 421, 454 Utterance , passim -final lengthening Utterance , 373, 390 Uvula, 433 Uvularspeechsounds, 433

V-COMP, 169, 190- 191, 197- 198, 250- 253 Vectorsummationin articulation.444- 445 Velarspeech sounds , 433 Velum. SeeSoftpalate Verb, 192- 194,passim deictic, 54 , 64- 68 perfonnative raising, 190, 250

Word-initialconsonant andspeech errors , 321,338,341- 343,464 Wordorder

159 Writing.435

Vowel back, 431 errors. SeeErrors front , 431 lax/ tense, 350 lengthening - , 34, 36, 309. SeealsoDrawl semi-, 434 Vowel space, 431

WeU-formedness of conversation . 37- 38 of message 70 , monitoringof, 14, 498 Rule, 486, 489, 491, 497, 499 of self-repairs,485- 487 of syllable /cluster, 329, 369, 482 of syntax, 282, 468 Wh-movement , 253