No title

Let us first have a look at an example with disjoint foci. I assume that [G], [F] and [E ] stand for the same objects as in ( 3 5 ) above.

(44) only ate2 [an3 apple]F, only((J. T.[G](T([F])), [E])), � [A] John1, ;. p;. tJ.a.{wg : wgE P(tx)({uh :3k[ukE a 1\ k::s;;1h 1\ h1 = jw]})), = [B)

I

Uohn1]F, (J. T.T, [B])

�

even, .! S.even( S)

even Uohn1]p, even((.A. T.T, [B])) even Uohn1]F only ate2 [an3 apple]p, even((.A. T.T, [B]))([A]) Spelling out the meaning rules of even and only and performing lambda­ reductions, we get the following result:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

= ). Ua{wg :wgE B(F)(t)(a) 1\ VXE ALT.; (F)(B(F)(t)( a)

Manfred Kritka

29S

A QA tA.a{wg :wg E [B](Q)( t)( a) 1\ VX E ALT'! ([B])[[B]( Q)(( t)(a)

The second conjunct expresses that for every alternative X to John it holds throughout the input state a that it is less probable that John only ate an apple than that X only ate an apple. VX. E ALT'! ( (B])[

{uk E a:3h[k::;;;1,2,3h 1\ ea�(h2h1h3) 1\ h1 = jw 1\ applew(h3) 1\ Pas�(h2)] 1\ --.3X3k[X E ALT'! ([E]) 1\ 3s[wk E X (A5xyAa.{wg E a: ea�(sxy)})( sx)({uh :3k[uk E a 1\ k::;;;1,21.h 1\ Pastu(s) 1\ h1 = j w]})]]}

in summary, (44) has the following meaning: it is asserted thatJohn didn't eat any alternative to an apple. It is presupposed thatJohn ate an apple, and that it is less likely for John that he didn't eat any alternative to an apple than it would be for an alternative to John. Furthermore, discourse entities for John, an eating event by John and an apple that is eaten in the event are introduced. This seems to be the correct representation for a sentence like (44). For example, when the sentence is negated by it is not the case that, or dialogically by no , only the asser­ tion part will be negated, but not the presuppositions. Other cases with multiple focusing operators can be analyzed in a similar way. Let us have a look at a derivation with overlapping foci:

(45 ) even [only ate2 [an1,2 apple]F]F only ate2 [an3 apple]F, o nly(( A T.[G](T([F])), [E])), - [A]

fI V

only ate2 [an3 apple]F] F, (A.P.P, [A]) even, A S.even( S)

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

\fuh E a3k[h�1•2,3k 1\ eatu(k2k1k3) 1\ k1 = jw 1\ appleu(k3) 1\ Pastu(k2)] 1\ 3k[wk E a 1\ k::;1;; ,2,3g 1\ ea�(g2g1g3) 1\ g1 = jw 1\ applew(g 3) 1\ Pas�(g2)] 1\ --.3X3k[X E ALT'! [E ]) 1\ 3s[wk E X(A sxyAa.{wg E a:ea�( sxy)})(sx)({uh :3k[uk E a 1\ k�1•2,.h 1\ Pastu(s) 1\ h1 = jw]})]]

296 Focus and Presupposition in Dynamic Interpretation

even [only ate2 [an3 apple) F) F, even((A.P.P, [A))) = A xAa{wg : wg E [P) (x)(a) 1\ 'tfX E ALT1 ([A])[[A) (x)(a)

The second co�unct expresses that it is presupposed in a that it is less likely that x only ate an apple than that x did some alternative to only eating an apple. Finally; let us analyze-a-Ease in-whiGh one-focusing-operatocis in the_ fo_c_us gf another operator:

(46) ate2 [an3 apple) F, (A T.(G](T([F])), [E)) only, A S.only(S)

I

(only) F, (A R.R, A S.only(S)) [only]F ate2 [an3 apple]F, (A. RR, A S.only(S))((A. T. (G)(T((F])), [E))) = (J. R.R((A T.(G)(T([F])), [E ])), A S.only(S))

�

even, A S.even(S)

even [only]F ate2 [an3 apple) p, even((A.R.R((A. T . [G)(T((F])), (E])), A. S.only(S))) A xAa(wg : wg E only((A. T.(G)(T([F])), [E]))(x)(a) 1\ 'tfX E ALT1 (only)[only((A. T.[G) (T([F])), [E ])))(x)(a)

The first conjunct reduces to the same formula as the first conjunct in the preceding derivation: it presupposes that x ate an apple, and asserts that x didn't eat any alternative to an apple. As for the second co�unct, we have to know what the possible alternatives to only are. In Krifka (1 992a) I suggested that the only alternative to only is the meaning of the focusing operator also . One piece

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

'Vuh E a3k[h::::2: ,3k 1\ eatu(k2xk3) 1\ appleu(k3) 1\ Pastu(k2)) 1\ 3k[wk E a 1\ k::::2: ,3g 1\ eatw(g2xg3) 1\ applew(g3) 1\ Pastw(g2)] 1\ --.3X3k[X E ALT" ([E)) 1\ 3s[wk E X(A.sxya.(wg E s:eatw(sxy)})(sx)((uh : 3k[uk E a 1\ k::::2: 1.h 1\ Pastu(s)]J)]]

Manfred Krifka 297

of evidence for this is the common locution not only X, but also Y. Let us assume the following meaning for also as a VP operator:

(47) also((B, F)) �

A tAa{wg:wg E B( F)(t)(a)1\ Vuk E a3X3h[X E ALT'I (F) 1\ uh E B( X)( t)({uk})]}

(assertion) (presupp.)

That is, it is asserted that B( F) holds, and it is presupposed that for some alternative X to F, B( X) holds. If also is the only element in ALT'I (only), then we get the following interpretation of the second conjunct of (46):

After the meaning postulates for only and also are spelled out, we get the presupposition that throughout the input state a it is less likely that x ate an apple and no alternative to an apple, than that x ate an apple and some alternative to an apple. This captures the meaning of expressions like (46) correctly. IO

C O N C LU S IO N

In this article I have shown that structured meanings can be incorporated in a dynamic setting, and that the resulting framework allows for a sophisticated treatment of focusing operators. In particular, we have seen that we can distinguish between presuppositional content and assertional content, and that we can deal with discourse markers that are introduced within the scope of such operators. Received: 1 4. 1 2.92 Revised version received: 2 5.05.93

MANFRED KRIFKA Dept. ofLinguistics Calhoun Hall 5 0 1 University of Texas at Austin Austin, Texas 787 1 2- 1 1 96 USA

N OTE S 1 I thank two anonymous reviewers for helpful comments and criticism. The paper is a substantially revised version of a talk I gave at the Fourth Symposium on Logic and Language, Budapest, 1 992. A preliminary version was published in the proceedings of this conference, Krifka

(199 3). I thank the organizers and partici­ pants of this conference for the oppormn­ iry to present this paper and discuss its ISSUeS. 2 This analysis was suggested to me by Amim von Stechow. 3 There is one interesting difference

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

{uk E a: 3h[uh E only((A T.[G] (T([F) )), [E]))(x)( a)]}

298 Focus and Presupposition in Dynamic Interpretation between German and English: Contrary to English, focus on negative terms is possible in German: (I) a. Nur kein Madchen hat geweint. [only no girl cried] 'Everybody who is not a girl cried but no girl cried.' a. Wir haben nur keine Giraffen gesehen. [we have only no giraffes seen] 'We saw everything except giraffes.'

This differs from the treatment in Beaver (I992), for whom [91 ](a) would be undefined in such a case. Consequently, Beaver has to employ a semantic meta­ language that allows for truth-value gaps. This complication is unnecessary, I think. As the empty information state does not serve any essential function, we might make use of it to express presupposition failure. 6 This analysis of negation differs from the one given in Heim (I98 3a) and Beaver ( I992). According to their analysis, a negated sentence � restricts an input state a to those world-assignment pairs whose assignments cannot be extended to satisfy rp :

[ ....,rp ](a) - a (wf E a: 3g � £1wg - E - [¢](a)Jj -

If kein (e) has to be decomposed in this fashion, then the above examples would obtain the following interpretations:

(3) a. . . . nur [Madchen]F NEG geweim hat. [. . . only girls didn't cry.] a . . . . wir nur NEG [Giraffen] F gesehen haben. [. . . we only didn't see [(any) giraffes]F-] 4

The present analysis differs from the one given in Krifka (I993), where I assumed that it is presupposed throughout the input state that g. is defined and refers to g1 's bike. A problem with that analysis is that it cannot handle quantified sentences as the following one, as the dicourse marker 4 cannot be fixed to a particular object Every boy1 who likes his1,4 bike keeps it4 clean. Noun phrases like his bike are analyzed as 'weakly familiar' in the terminology of Condoravdi (1992). That is, although their index is not present in the input state yet, their descriptive content is presupposed.

The problem with this representation is that if ¢ expresses a presupposition that is not satisfied throughout a, then the substracted set will be reduced to 0 , and a 0 is a again. So we would predict that a negated sentence containing a presupposi­ tion that is not satisfied simply docs not change the input state but otherwise does no harm. Note that the result is different in Beaver's theory, where in such a case [rp ](a) will be undefined, and conse­ quently [--'¢](a) will be undefined. There is one problem of the proposed analysis, illustrated by the following sen­ tence: -

7

He1 did not see [his1,3 bike);. He1 suspected that i� was stolen. Note that it can be anaphorically related to his bike, even though this NP occurs within the scope of a negation, and hence its index should be inaccessible. A way out is the following. Note that the negation does not affect the presupposition that I has a unique bike. Assuming that it picks up the description his bike, it follows that it refers to the same entity as its antecedent. In Krifka (I993) I assumed that the index 3 itself is presupposed; however, this creates

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

This difference is due to the fact that the German negative quantifier has to be decomposed into a negation part and an indefinite part (cf. also Jacobs I 983). This was shown by Jacobs (I98o) with examples like the following: (;) J�der .i;zr fahn k.�in Auto. [every doctor drives no car] One reading: 'Not every doctor drives a car.'

5

Manfred Krifka 299 problems in quantificational cases, as indicated in note 4· 8 This treatment of foci containing ana­ phone elements differs from the one given in Krifka ( 1992b), where I assume that the alternatives directly refer to the parmers, and as those entities differ for different gentlemen under consideration, the set of alternatives is dependent on input assign-

ments. The present treatment, where the alternatives are · something like Skolem functions (for each gentleman x, they give x's left parrner and x's right parrner), allows us to give considerably simplified semantic rules for only . 9 But see Jacobs (1983), who points our problems with an analysis of even in terms of probability.

Beaver, D. I . (1 992), The Kinematics ofPresup­ position , ITI.J Prepublication Series LP-92os, University of Amsterdam. Also in the

Proceedings ofthe 8th Amsterdam Colloquium . Condoravdi, C. (I 992), 'Strong and weak novelty and familiarity', in SALT II,

Proceedings from the Second Conference on Semantics and Linguistic Theory, Working

Papers in Linguistics 40, Ohio State University, Department ofLinguistics, 1 738. Dahl, 0. ( 1 974), 'Topic-comment structure revisited', in Topic and Comment, Contextual Roundness and Focus, Buske, Hamburg, I 24. Groenendijk, J. & Stokhof, M. (I 990), 'Dynamic Montague Grammar', in L. Kalman & L. Polos (eds), Papers from the

Second Symposium on Logic and Longuage , Hajduszoboszio, Hungary, Academia! Kiad6, Budapest, 3-48. Groenendijk, J. & Stokhof, M. (I99I), 'Dynamic predicate logic', Linguistics and Philosophy, 14, 39- 1 0 1 . Heim, I . (1 982), 'The semantics of definite and indefinite noun phrases PR', Ph.D. Dissertation, University of Massachusetts at Amherst, ·md University of Konsranz, SFB-Papier 7 3· Heim, I. (198 3a), 'On the projection problem for presuppositions', in West Coast Con­ ference on Formal Linguistics 2 , I I 4-26. Heim, I. ( I 983b), 'File change semantics and the familiarity theory of definiteness', in

B

Meaning, Use and the Interpretation of Longuage, Berlin, Walter de Gruyter, 16490. Horn, L. {1 969), 'A presuppositional analysis of only and even ' in Papers from the fifth

regional meeting, Chicago Linguistic Society,

3 1 8-27. Horn, L. { I 98 5), 'Metalinguistic negation and pragmatic ambiguity', Longuage, 6 1 , 1 2 1 74· Jackendoff, R {1 972), Semantic Interpretation in Generative Grammar, MIT Press, Cam­ bridge, Mass. Jacobs, J. (1 980), 'Lexical decomposition in Montague Grammar', Theoretical Linguis­ tics, 7, I 2 1 -36. Jacobs, J. (1983), Fokus und Skalen, Zur Syntax

und Semantik von Gradpartikeln im Deutschen , Niemeyer, Tubingen. Jacobs,]. (I99 1 ), 'Focus ambiguities',journal of Semantics, 8, I -J6. Kamp, H. (1984), 'A theory of truth and semantic representation', in J. Groenen­ dijk, T. Janssen & M. Stokhof (eds), Truth, Interpretation and Itiformation, Foris, Dor­ drecht. Karttunen, L. (I 974), 'Presuppositions and linguistic context', Theoretical Linguistics, 1 , I 8 H )4· Kartcunen, L. & Peters, S. { I 979), 'Conven­ tional implicature', in C.-K. Oh & D. Dinneen (eds), Syntax and Semantics J: Presupposition , Academic Press, New York, I -56.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

R E F E RE N C E S

300 Focus and Presupposition in Dynamic Interpretation

-·

·

pretation', Natural Language Semantics, I , 75-1 1 6. Rooth, M. E. ( 1 98 5 ), 'Association with Focus', Ph.D. Dissertation, University of Massa­ chusetts at Amherst. Rooth, M. E. ( 1 99 1 ), 'A theory of focus inter­ pretation', Nawral Language Semantics, I, 75-1 1 6. Sandt, R. A. van der ( 1 988), Context and Presupposition , Croom Helm, London. Sandt, R. �- van der (I991), 'Denial', Papers from the Parasession on Negation , Chicago Linguistic Society, Chicago. Seuren, P. ( 1 988), 'Presupposition and nega­ tion', Journal ofSemantics, 6, I 7 5-226. Stalnaker, R. C. ( 1 974), 'Pragmatic presup­ positions', in M. K. Munitz & P. K. Unger (eds), Semantics and Philosophy, New York University Press,-New York, -1 97-2 I 4· Stechow, A. von (r 982), Structured Propositions (Arbeitspapiere des SFB 99), Universitat Konstanz, Konstanz. Stechow, A. von ( 1 990), 'Focusing and back­ grounding operators', in Discourse Particles , Pragmatics & Beyond , John Benjamins, Amsterdam, 37-84. Stechow, A. von & Uhmann, S. ( 1 986), 'Some remarks on focus projection', in W. Abraham & S. D. Meij (eds), Topic, Focus and Configuration , John Benjamins, Amsterdam. Taglichr, J. ( I 984), Message and Emphasis , Longman, London.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Krifka, M. (1 992a), 'A compositional seman­ tics for mul tiple focus constructions', in J. Jacobs (ed.), Informationsstruktur und Cram­ matik (Linguistische Berichte, Sonderheft 4). Krifka, M. ( 1 992b), 'A framework for focus­ sensitive quantification', in C. Barker & D. Dowry (eds), Proceedings of the Second Conference on Semantics and Linguistic Analy­ sis, Ohio State University, Columbus, Ohio, 2 I 5-36. Krifka, M. (I 993), 'Focus, presuppositiOI.J and dynamic interpretation: the case of focus­ sensi'tive panicles', in K. Bimbo & A. Mi1te (eds), Proceedings of the Fourth Symposium on Logic and Language, Budapest 1 992, Aron Publishers, Budapest. Lerner, J.-Y. & Zimmerman, Th. E. ( 1 98 3 ), 'Pre-suppositions and quantifiers', in R Baurle, C. Schwarze & A. V. Stechow (eds), MeaninJ;, Use and Interpretation oJL.anxuaxe, de Gruyter, Berlin, 290-30 1 . Lewis, D. (1 979), 'Scorekeeping i n a language game', journal ofPhilosophical Loxic 8, 33959· Partee, B. H. ( 1 984), 'Nominal and temporal anaphora', Linguistics and Philosophy , 7, 243-86. Rooth, M. ( 1 987). 'Noun phrase interpreta­ tion in Montague Grammar, file change semantics, and situation semantics', in Generalized Quantifiers: Linguistic and Logi­ cal Approaches, Reidel, Dordrecht, 2 3 7-68. Rooth, M. ( 1 992), 'A theory of focus inter-

Joumal oJSemantics

10: JOI-J I 8

© Oxford Universiry Press 1993

The Importance ofBeing 'Only': Testing the Neo-Gricean Versus Neo-Entailment Paradigms J A Y D A V I D ATLA S Pomona College, Claremont

This essay is dedicated to the memory of my teacher ( 1 9 1 7- 1 993). George Lyman Crosby, Professor of Philosophy, Emeritus, Amherst College.

JOSEPH EPSTEIN

In Atlas ( 1 991) I proposed a novel account of the logical form of statements having the form 'Only a is F' and the form 'Also a is F', an analysis of the entailments and of the implicarures of those statements, and a discussion of the effects of focal stress on implicatures. In this paper I discuss the merits of my account over those of a Gricean account offered by Peter Geach ( 1 962), Larry Horn (1 992), and James McCawley (198 1 ). In doing so I discuss several fundamental problems in Gricean Pragmatics: the nature of the cancellation of implicarures, the intrusion into truth-conditions of pragmatic inference, Negative Polariry Items, and the non­ monotoniciry of 'only a' as a Generalized Quantifier.

In 'The Said and Unsaid' Larry Horn (I 992) entertains an analysis of 'only' sentences that revises, in an interesting way, his presupposition analysis of Horn (I 969) and his conventional implicature analysis of Horn (I 979). As I examined the presupposition analysis in Atlas ( I 99 I ), I wish now ro examine those areas of agreement and disagreement that exist between us in light of his new, pragmatic analysis, contrasting the analyses of Horn (I 992) with Atlas (I 991 ). His (I 969) original analysis claimed that Only Muriel votedfor Hubert should be understood to have (i) an assertoric parr, by which "entailment ( as reflected in constraints on cancellation or suspension), polarity effects, and monotoniciry diagnostics (c£ Barwise & Cooper I 98 I) are determined . . . alone . . . and not by what is presupposed or implicated" (Horn I 992: 1 79), and (ii) a logically pre­ supposed part. Thus statement ( 1 ) has rhe analysis (2):

( 1) Only Muriel votedfor Hubert . (2) ASSERTION: No one distinctfrom Muriel votedfor Hubert . PR ESUPPOSITION: Muriel votedfor Hubert . Hom (1 969, I 992) adopts as the logical representation of the ASSERTION and PRESUPPOSITION in (2) the first-order formulae:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Abstract

302

The Importance of Being 'Only'

(3)

ASSERTION: - ( 3x )( - (x = PRESUPPOSITION: Fm .

In Atlas

m ) & Fx)

(1991) I suggested several difficulties with this logical presupposition

analysis, not the least of which is the fact that if no one voted for Hubert, the logical form of the assertion would be, vacuously, true while the logical form of the presupposition would be false. Yet in a case oflogical presupposition, a false presupposition should yield a truth-value gap, not a vacuous truth. Traditionally the 'only' sentences have been analyzed as conjunctions (see Horn

1 989: 248):

(4) Muriel votedfor Hubert & no one distinctfrom Muriel votedfor Hubert.

the earlier view by now suggesting that the second part of the analysis be a Gricean ( I97s}GENER:AHZED-GONVERSATIONAL IMPLICATUM (GCI), rather than a logical presupposition (Horn 1 969), or a Gricean CONVEN­ TIONAL IMPICATUM (Hom 1 979), in most, but not all, cases. Thus his ( 1 992) ; analysis is:

(2')

ASSSERTION:

GCI:

No one distinctfrom Muriel votedfor Hubert . Muriel votedfor Hubert.

with the same logical forms as in (3). In some cases Hom believes that propositions that play the same role as

Muriel votedfor Hubert are "explicata" in the sense of the Relevance Theory of Sperber and Wilson (1 986) and Carston (1 988). Carston (1988: 164-5) supposes that utterances like:

(5)

The park is some distancefrom where I live. b. It'll take us some time to get there,

a.

taken literally, expresses trivial truths far from the speaker's meaning. The "propositions" expressed by the speaker are more definite than the literal sense of the sentences. The propositional content of the utterance is constructed from the literal sense of the sentence and from the products of pragmatic inferences, which give a contextually definite interval of time or distance. The products of the contextual, pragmatic inferences are "explicata", not Gricean "implicata", since they are constitutive of the truth-conditions of the utterances.1 For example, Hom (1992: 1 79-80) believes that though Muriel votedfor Hubert is a generalized conversational implicatum, not an entailment, presupposition, or part of the assertion, of asserting the sentence 'Only Muriel voted for Hubert',

Kim passed the test is an "explicatum" of asserting the sentence 'Guess what: only Kim passed the test!' (Hom 1 992: 1 82). Hom (1992: 1 83) concludes:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Horn (1969, 1979, 1 992) rejects the conjunction analysis, committed as it is to Muriel votedfor Hubert being (i) asserted and (ii) entailed, and adopts the minimal content for the assertion that Peter Geach (1962) recommends. But he modifies

Jay David Atlas

303

I submit that an asymmetric theory of the conventional meaning of only in the spirit of William of Sherwood and of Geach, combined with a Gricean approach to the positive or existential component, and with a London-sryle account of the apparently recalcitrant cases, provides the most narural and least stipulative rreatment of the full range of data.

(6) It's not the case that Lenin belongs even in a museum . . . Lenin does not belong even in a museum . . . that for all contextually relevant locations it is more likely that Lenin does not belong in them than that he does not belong in a museum.2 It would then seem to be pragmatically inferable, in the same morally stringent context, from that conventional implicatum that, Lenin not belonging in a museum, he does not belong anywhere. If so, on Hom's account of the truth-conditions of 'only' statements, if Lenin does not belong even in a museum, he belongs only in a museum! (I take it that this is an absurd result.) In Adas (I 99 I ) I had offered the following analysis of the statement Only a is F: (a) Only a is F is semantically anomalous unless there is someone other than a ; so I suggested that There is someone other than a was "grammatically presupposed" by the statement (Atlas 1 99 1 : 1 28). After giving criteria for the Topicality of Noun Phrases, I pointed out that: (b) A statement is about a only if 'a ' is a Topical Noun Phrase in the statement. Then I claimed: (c) The statement Only a is F is not about a ; it is about those who F.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Supporter though I be of Radically Pragmatic analyses in logical and linguistic semantics, I have always been unhappy with Geach's (I962!I98o: 207) logical analysis of the truth-conditions of'only' sentences, which Hom also adopted in Hom (I 969, I 979, I 992). That in certain circumstances the assertoric content would be vacuously true though "uncooperative to assert" (Hom I 992: I So) has often struck me as unsatisfactory from a linguistic point of view as the judgment on a mass-murderer that he is innocent though "legally insane" is unsatisfactory from a legal point of view. To take one of Hom's (I992: 1 83, n. I 3) own examples, the newspaper headline 'Lenin Belongs Only in a Museum-or Does He?' would contain the assertion Lenin belongs nowhere other than in a museum . In case Lenin (i.e. his remains) belongs nowhere at all-and I can imagine someone holding such a morally stringent view-it is true that Lenin belongs only in a museum! Furthermore, since Hom took the headline to be raising the question whether Lenin belongs EVEN in a museum, on Karttunen and Peters's ( I 979: 2 3-3 3) and Hom's own account it would follow by conventional implicature from:

304

The Importance of Being 'Only'

On the principle (Atlas and Levinson I 98 I) that logical forms should preserve aboumess, my suggestion (d) preserves aboumess: (d) Only a is F asserts (and has the truth-conditions oD One individual, and no one other than a, is F. The appositive clause 'and no one other than a' does not have the properties of a simple coordinating conjunction. Thus the arguments of Atlas (I 99 I ) lead to the view that (8) means ( I I): (8) Only Muriel votedfor Hubert. (I I ) One individual, and no one other than Muriel, votedfor Hubert .

(12) (3x)(Vy ) [(x � y

=

Fy) & (Fy - y - m )].

This logical form is logically equivalent to, but not the same form as, one for: (I� ;) - Ex�ct� on�person is F and no one-o-ther than m- is F ' vrz.:

( I 2 ') (3x)(Vy )(x � y = Fy) & ( x)(Fx - x - m ), which is, unlike (I 2), a conjunction of two closed formulae. For reasons that I gave in Atlas ( 1 991 ), ( 1 1) is the appropriate analysis. The difference in the logical forms just displayed captures the difference between the appositive 'and' clause in (I I ) and the coordinate 'and' clause in ( I I '). With the peculiar consequences ofhis own view about the assertoric content of 'only' sentences, why does Horn resist the traditional conjunction/entail­ ment view? Besides the arguments of Horn ( I 969), criticized in Atlas (I 99 I ), Hom ( I 992) suggests three further problems. First, how docs the conjunction­ alist explain why: (7) Muriel and only Muriel votedfor Hubert . is distinct from: (8) Only Muriel votedfor Hubert and not simply redundant (Horn I 992: I 79)?3 After all, on the conjunctionalist account, (8) should be paraphrased by: (9) Muriel votedfor Hubert & no one other than Muriel votedfor Hubert , and so (7) should mean: ( r o) Muriel votedfor Hubert & Muriel votedfor Hubert & no one other than Muriel votedfor Hubert ,

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Its logical form is not the traditional conjunction, which is a conjunction of two closed formulae; it is instead (Atlas I 99 I : I 38):

Jay David Atlas

305

which, except for the redundancy of the first conjuncts, is just (8). The traditional conjunctionalistlentailmentist has no good answer to Horn's acute question, but the non-conjunctional entailment analysis of Atlas ( I 99 I ) does offer an answer. To repeat, the arguments of Atlas ( I 99 I ) lead to the view that (8) means ( I I ): (8)

Only Muriel votedfor Hubert .

( I I ) One individual, and no one other than Muriel, votedfor Hubert . Its logical form is not the traditional conjunction, which is a conjunction of two closed formulae; it is instead (Atlas I 99 I : I3 8): =

This logical form is logically equivalent to, but not the same logical form as, one for:

( I I ') Exactly one person is F and no one other than m is F, Vlz.:

( 1 2 ') (3x )(Vy )(x

=y=

Fy) & (\fx) (Fx -+ x = m ),

which is, unlike ( I 2), a conjunction of two closed formulae. For reasons that I gave in Atlas ( 1 99 I ) , ( 1 1 ) is the appropriate analysis. The difference in the logical forms ( 1 2) and ( 1 2 ') captures the difference between the appositive'and' clause in ( I I ) and the coordinate 'and' clause in ( I I '). Notice that on this view (7) is distinct from (8) and not simply redundant; (7) means ( 1 3), while (8) means ( I I ):

(7) Muriel and only Muriel votedfor Hubert . ( 1 3 ) Muriel votedfor Hubert, and one individual-no one other than Muriel-votedfor

Hubert . (8) Only Muriel votedfor Hubert . ( I I ) One individual, and no one other than Muriel, votedfor Hubert .

The theory of Atlas ( 1 99 1 ) can meet Horn's objection. Horn's second argument concerns the role of the proposition [HILLARY TRUSTS Bin] in the speech-act of asserting Only Hillary trusts Bill . Horn ( 1 992: 1 80) claims that the data of

( 1 4) a. b. c. d. e.

Ifonly Hillary trusts Bill, all is well . Ijust discovered that only Hillary trusts Bill. It's too bad that only Hillary trusts Bill . I know Hilla ry trusts Bill, but does ONLY Hillary trust Bill? #I know nobody besides Hillary trusts Bill, but does only Hilla ry trust Bill?

show that Only Hilla ry trusts Bill at most conventionally implicates, and does not "say" (assert), that Hillary trusts Bill. Horn ( 1 992: 1 80) writes:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

( 1 2) (3x )(Vy ) [(x = y Fy) & (Fy .... y = m )].

306 The Importance of Being 'Only' The fact that the positive proposition ['Hillary trusts Bill'] falls outside the scope of the assertion in each cases reinforces the view that we are dealing with a non-truth-conditional aspect of conventional meaning. But are we? Or is a more unconventional analysis called for?

The more unconventional analysis" that Horn will adopt appeals to generalized conversational rather than conventional implicature. But his objection is to the traditional conjunction/entailment analysis. This objection can be met. The entailment analysis of Atlas (I 99 I) shows how the positive proposition can fall outside the scope of assertion and yet be entailed, and so be a truth-conditional aspect of meaning. After giving the analysis of (I I )-( 1 2), I (Atlas I 99 I : I 39) wrote of ( I I)-( 1 2):

preserves Aquinas's and Geach's intuition that an "excluder" like only excludes everything other than what is named by the subject-term 'a ' from "sharing in the predicate 'F' " and need .not go on. to say something !_la!!le4 by the subject-term does �share in the predicate" (Geach 1 9621 ! 980: 208-9).

.

In my (Atlas I 99 1 : I 3 8) view the positive "proposition is a syncategorematic proposition, and it cannot be simplistically conjoined with No one other than a is F by '&' to yield a logical form for the sentence of the type P & Q ", where P, Q are closed formulae. The objection that Horn makes to the traditional conjunctionalist analysis does not apply to my entailment analysis. The third argument rhar Hom (1 992: 1 80, n. 1 0) considers is the following:

If the semantics of only if are compositional, they reinforce the conclusion that the positive proposition is nor entailed, or P only if Q would be equivalent to P if and only if Q , which it clearly is nor: I'llgo (* ifand} only ifyou do and maybe not even then . But the distinction berween only if and ifand only if is akin ro that between only linguists and linguists and only linguists.

It is not explicit in Horn's discussion what the "compositionality" of only if actually amounts to, but there are two plausible reconstructions consistent with Horn's argument. First, that P only ifQ means Only [P ifQ ] , and, second, that it means [P only] ifQ , i.e. [Only P] ifQ . On either of these alternatives, P only ifQ would presumably entail P ifQ , given Horn's assumption that the entailmentist holds that only a entails a . Thus P if Q and P only if Q would be logically equivalent to P only ifQ . This clever argument, if it were compelling, would be, and should be, immensely troubling to any entailmentist. Fortunately, it is not compelling. Horn (I 992: I So) says, apparently as solely a matter of intuition, that "the distinction between only if and ifand only if is akin to that between only linguists and linguists and only linguists ", which indicates that for Horn a in only a is a meta-variable that ranges over at least Ss and NPs. The trouble is that the principle of compositionality to which Hom appeals does not commit one to

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

From the point of view of a theory of speech-acts, in asserting Oniy a is F, we do not thereby

assert a is F, the way we would if the statement were to consist of a conjunction a is F & ¢. Rather, what we do assen entails a is F, but it does not �say" it. This feature of my analysis

Jay David Atlas 307

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

either of the possible analyses of P only if Q that I canvassed above. Any entailmentist will hold the logical equivalence of P only ifQ ; ifP, Q ; etc., so that a compositional analysis must be consistent with these equivalences. Since Hom { I 992: I 8 I) recognizes the traditional converse relationship between only and all, e.g. the equivalence of Only Democrats support Brown and All Brown supporters are Democrats , and he recognizes the conditional analysis of all statements found in nineteenth-century Idealist logic, e.g. F. H. Bradley's, as well as in Frege, Peano, and Russell, Hom's compositional analysis of Only ifQ P should be consistent with its logical equivalence with IjP, Q . But this, Only [P if Q] and [Only P] ifQ signally fail to be. The only remaining analysis of P only ifQ is P, ifonly Q . This is the compositional analysis that is consistent with the converse relationship between only and all and with its equivalence to Q if P. But this analysis, and the assumption that only a entails a , does not imply P if Q , so the thesis that on the entailmentist's view P only ifQ is equivalent to P if and only if Q fails to follow. Hom's reductio argument of the entailmentist's po�tion fails, as the two analyses that lead to his conclusion are inconsistent with obvious logical facts, and the analysis consistent with the obvious logical facts fails to lead to his conclusion. Larry Hom (personal communication) has pointed out to me arguments of James McCawley (I 98 I ) that apparently he had in mind in giving the reduction argument just discussed. McCawley's arguments are also of interest, so I shall discuss them briefly. First, McCawley (I 98 I : 5 I) offers a reason that the meaning of'only if' should be treated compositionally and not as an idiom. The argument is simple; McCawley (I 98 I : 5 I) writes that since "expressions such as only if, even if, except if, and especially if appear to be immediately intelligible to anyone who knows the words of which they are composed (i.e. they are in no sense idioms), an analysis in which only if is treated as ordinary only plus ordinary if seems to be inescapable". That's it. That argument for composi­ tionality offers no support to the reductio argument that Hom gives and that I just criticized. There is a second argument for compositionality offered by McCawley {I98 I : 5 I -2). It depends upon an analysis of if by Geis { I 973). If is analyzed as in cases in which ; e.g. IfBill comes tomorrow, I'll give him the books is analyzed as In (all ) cases in which Bill comes tomorrow, I'll give him the books . McCawley {I98 I : 5 2) then claims that Geis's {I973) analysis "allows only if to be analyzed as 'only in cases in which ' ". Now, that's an interesting claim; one wonders just how Geis's {I973) analysis allows that. Geis's analysis suggests that ifP means in (all ) cases in which P. Then how do we construe only ifP ? As only [ifP] ? As [only if] P ? As [if[only (P]]] ? McCawley suggests that only ifP means [only if] P. But one wonders by what application of the principle of composition he makes that choice? Does Geis's analysis tell us which we should choose, and why? Answer. no. This compositionality does not support Horn's reductio argument either. My argument against Hom's reductio does not depend

308 The Importance of Being 'Only'

upon objecting to Hom's claim that only can be combined with ifP just as it combines with NP; it depends upon the ill-defined character of the compositionality operation, whether the operand is ifP or is NP. So far, I have considered three arguments that Hom gives against the traditional, conjunction/entailment analysis of only statements. First, how does the conjunctionist explain why:

(7) Muriel and only Muriel votedfor Hubert . is distinct from:

(8) Only Muriel votedfor Hubert

The fact thaHhc-positive proposition -[HILLARY-TRUSTS Btu] falls outside the _scope of the assertion in each cases reinforces the view that we are dealing with a non-truth-conditional aspect of conventional meaning.

The nee-entailment analysis of Atlas ( r 99 I ) also meets the objection based on this observation about assertion. Thirdly, and finally, Hom adduces a clever reductio argument purporting to show that on the entailmentist's assumptions, P only ifQ turns out to be logically equivalent to P ifand only ifQ . This reductio argument turns out to be fallacious. Horn's first two arguments clear the board of the traditional, conjunction, entailment analyses, leaving only the entailment analysis of Atlas ( I 99 I ) still standing.4 His third argument, which would undermine any entailment analysis, fortunately for the view of Atlas ( I 99 I ), is fallacious. But his first two arguments have done a major service in exposing the inadequacies of the traditional conjunction analyses. With two views now left in the field, the neo-G ricean Horn ( I 992) and the nee-entailment Atlas ( I 99 I ) views, I want to consider the claims and consequences of Horn's view. Recall that on Hom's ( I 992) view, the statement: ( 1 s) Only Muriel votedfor Hubert is analyzed as: ( 1 6) a. ASSERTION: Nobody ( distinctfrom/but/except/other than} Muriel votedfor Hubert . b. GENERALIZED CONVERSATIONAL IMPLICATUM: Muriel voted for Hubert . and formalized as in:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

and not simply redundant? Of the entailment analyses available only the nee­ entailment analysis of Atlas ( I 99 I ) meets Hom's objection. Second, Hom ( I 992: I So) writes of the data in ( I 4):

Jay David Adas

309

(I7) a. ASSERTION: -(3x)(-(x = m ) & Fx ) b. GENERALIZED CONVERSATIONAL IMPLICATUM: Fm . Horn (I 992: I 8 I ) properly attempts to show how, assuming the assertion in (I 6/ I7a), one can infer the implicatum in (16!I 7b) by familiar Gricean, Radically Pragmatic arguments. The argument is this. The statement Only Fs are Gs entails · All Gs are Fs . But stating All Gs are Fs implicates Some Gs are Fs · . Thus, stating Only Fs are Gs implicates · Some Gs are Fs. How would this argument apply to (I s)? Using a Quinean predicate 'Muriel­ izes' whose extension is the singleton set {Muriel}, the argument goes: Only Murielizers votedfor Hubert . So, All Hubert-voters are Murielizers . Stating the latter then implicates · Some Hubert-voters are Murielizers · , which is equivalent to · Muriel voted for Hubert . Thus, stating Only Muriel votedfor Hubert implicates · Muriel voted for Hubert · . Of course, these are intuitively appealing arguments. The first, obvious, difficulty with them was mentioned years ago by William Lycan, and that is that Gricean arguments, in principle, appeal to statements/utterances, not to propositions. No implicatum is inferable from a proposition; it must be stated to generate an implicatum. Such a constraint on Gricean arguments can lead to the psycholinguist positing, beyond necessity, acts of mentally 'asserting'. Another difficulty, discussed at length in Atlas (1984: 372, n. 6), lies in the inheritance of implicature under entailment. I myself once entertained the thought that the schema: A entails B B implicates C ·

·

So, A implicates C was valid. Rob van der Sandt then, almost instantly, produced a counter­ example, and I retreated even from entertaining that thought let A = P, let B - P V Q, and let C = -(P & Q). Unfortunately, this schema is the one that Horn requires to generate · Muriel voted for Hubert as an implicatum of Only Muriel voted for Hubert . Thus Horn cannot satisfactorily generate the implicatum from the assertion. 5 Besides this argument Horn also appeals to considerations in James McCawley (I 98 I : 226). Those considerations are these: ·

If one of the persons enumerated in: (a) Only Muriel, Lyndon, and Ed votedfor Hubert. is known by the speaker not to have voted for Hubert, then the speaker is being misleading: he could have been more informative by leaving the person our of the list.

What?? The speaker is being MISLEADING!? The speaker is simply lying; he is notjust being misleading and uninformative.6 McCawley ( I 98 1 : 226) continues:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

•

310

The Importance ofBeing 'Only'

Thus (a) could be uttered cooperatively only if for each of the three persons enumerated, either the speaker knows that person voted for Hubert or he does not know whether that person voted for Hubert.

Well, far from being cooperative, the speaker, if he does not know whether that person voted for Hubert, has violated Grice's maxims of Quality in saying that for which he lacks adequate evidence. The speaker is being uncooperative. McCawley ( I 98 I : 226} continues:

If this speaker says and perhaps Ed, he is not being cooperative; he is being genuinely misleading, and hence uncooperative, given that he has no idea whether Ed voted for Hubert or not. If he were being cooperative, he would not mention Ed at all. McCawley_(1_9.8J_: 2�6)_ COJ!<Judes: Thus a speaker coulc utter (a) cooperatively only if he holds that Muriel, Lyndon, and Ed all voted for Hubert . . . These considerations suggest that the analysis of only . . . in which Only Muriel votedfor Hubert was analyzed as Muriel voted for Hubert, and no one other than Muriel voted for Hubert' is incorrect only the second conjunct is really part of the meaning of an only-sentence, with the first co�unct being conveyed by virtue of the assumption that the speaker is being cooperative.

In my view there is no chance that McCawley's argument as quoted here demonstrates the inference to any such implicatum. Even though these derivations of the implicatum fail, one could still have Muriel votedfor Hubert be an implicatum for which we have yet to discover a proper pragmatic argument. What suggests to McCawley and Horn that the proposition is an implicatum are alleged data of (i) suspension, (ii) cancellation, (iii) non-epistemic cancellation, e.g.:

( I 8} (i) Only Muriel votedfor Hubert, ifeven she did (ii) Only Muriel votedfor Hubert, and maybe even she didn't (iii) Only Muriel votedfor Hubert, # and even she didn't . In the cancellation of an implicatum i from what is carried by asserting a, the denial of the implicatum must be consistent with the assertion. Horn (I 992) observes that felicitous cancellation will require epistemic cancellation ( I 8 )(ii), rather than the infelicitous, non-epistemic non-cancellation of ( I S)(iii). But in the classical version ofGrice's ( I 97 s: 5 7) discussion, he describes the mechanism as follows: a generalized conversational implicature can be canceled in a particular case. It may be explictly canceled, by the addition of a clause that states or implies that the speaker has opted

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

But if he doesn't know, say, whether Ed voted for Hubert, cooperativeness would demand that he indicates that (say, by saying Only 'Muriel, Lyndon, and perhaps Ed . . . ), since it is so easy for him to i ndicate that his knowledge about Ed is incomplete, and since his addressee presumably cares who voted for Hubert . . .

Jay David Atlas 3 I I out, or it may be contextually canceled, if the form of utterance that usually carries it is used in a context that makes it clear that the speaker IS opting out.

For the speaker to "opt out" from the operation of a maxim or of the Cooperative Principle is "to say, indicate, or allow it to become plain that he is unwilling to cooperate in the way the maxim requires. He may say, for example, I cannot say more; my lips are sealed" (Grice 1 975: 49). In Grice's (r96 1/r965: 4467) earlier discussion of non-cancelabilicy, he had written:

The classical Gricean version of attempted cancellation in Only Muriel votedfor Hubert would be: (r9) #Only Muriel votedfor Hubert, but I do not mean to imply that Muriel votedfor Hubert . Correlatively, the assertion conjoined with the denial of the implicatum should be anomalous: (20) #Only Muriel votedfor Hubert, and Muriel did not. The evidence of the anomaly of(r9) and (2o) for the non-cancelability of Muriel votedfor Hubert seems compelling. It also seems evident that ( 1 8)(ii) is a clear withholding of commitment, not merely to the alleged implicatum but, to the assertion Only Muriel voted for Hubert. In Grice's sense of 'non-cancelabilicy', Muriel votedfor Hubert is not a cancelable implicatum of Only Muriel votedfor Hubert ? If one takes the epistemic, modal qualifier as essential in tests of cancellation, anything, including logical entailments, will, incoherently, tum out to be cancelable. Suppose, unbeknownst to me, P entails Q . Then P andfor all I know not Q will be perfectly felicitous. Does chat show chat Q is a cancelable implication of P ? Surely not.8 If Muriel votedfor Hubert were a generalized conversational implicatum of Only Muriel votedfor Hubert, it would be necessary chat the implicatum be, in Grice's sense, non-detachable. That is, excluding considerations of Grice's Maxim of Manner, there would be no other way of"saying the same thing" that would fail to offer the same implicatum. The problem here is that for the entailmentist, this cuts no ice. Non-detachability does not distinguish between what is implicated and what is entailed. Evidence of non-detachability is irrelevant to the crucial dispute. So far, I have considered evidence from cancelability and from derivability

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

one cannot take a form of words . . . and then add a further clause withholding commitment from what would otherwise be implied, with the idea of annulling the implication without annulling the asserrion. One cannot intelligibly say 'Smith has left offbeating his wife, but I do not mean to imply that he has been beating her'. I shall express this fact by saying that . . . the implication is not cancelable (without canceling the assertion).

'J I 2

The Importance ofBeing 'Only'

that purport to show that Muriel votedfor Hubert is a conversational implicatum of asserting Only Muriel votedfor Hubert. So far, I see no compelling reason to believe that Muriel voted for Hubert is a conversational implicatum. The two arguments that purport to show how to derive Muriel voted for Hubert as an implicatum, Horn's argument and McCawley's argument, fail to show such a derivation. There now remains the interesting question why Horn and McCawley think that the meaning of Only Muriel votedforHubert is given by the ASSERTION No one other than Muriel votedfor Hubert , whose logical form is given by Horn in (I 7). Horn (I 992: I 79) states that evidence for the ·

·

(2 1 ) -Did only Muriel vote for Hubert? -No, [Lyndon did too I *she didn't).

I don't �myself see the conipellingness of tliese data for the "riegadviry" ofthe 'only' sentences. In fact in Horn (I 969: 99) these data were used only to suggest the presuppositional character of the 'only' sentences, not to determine their negativity. McCawley (1981: so-1) offers two arguments in support of the negativity of only sentences. One argument relies on alleged facts of paraphrase. First, he states two alleged facts, "the fact that only can be paraphrased by expressions involving two negatives in combination such as No X other than Y", e.g.John read only thefirst chapter byJohn read nothing other than thefirst chapter, and Only Susan has a key to this room by No one other than Susan has a key to this room , and "the fact that NotA ifnotB is so much better a paraphrase of A only ifB than is IfA then B (or B ifA )", i.e. that only if is best paraphrased by an expression combining two negatives. These alleged facts suggest to McCawley ( 1 98 I : so), and to Horn, that the only of A only ifB is the same only that will combine with NPs and be ana­ lyzed as containing two negatives, e.g. no and other than . Let me begin with problems for the second alleged fact. Though it is true that IfA then B and A only ifB do not seem to be paraphrases in cases like (a) If Mike straightens his tie once more, I'll kill hmm and (b) Mike will straighten his tie once more only ifi'll kill him , McCawley (I 98 I : 49) fails to notice that what on his the­ ory would be the preferred paraphrase of (b) Mike will straighten his tie once more only ifI'll kill him , viz. (c) If! won't kill him, then Mike won't straighten his tie once more, is in fact a good paraphrase of (a) IfMike straightens his tie once more, I'll kill him . Thus we have the following: (a) is a bad paraphrase of(b). [McCawley's fact] (c) is a good paraphrase of (b). [McCawley's theory] (c) is a good paraphrase of(a). [Atlas's fact]

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

essentially negative character of sentences like 'Only Muriel voted for Hubert' was provided by the possible and impossible continuations . . .

Jay David Arias

3I3

But by the symmetry and transitivity of'x is a good paraphrase of y', it follows that: · (a) is both a good and a bad paraphrase of (b),

(22) a. Bill does not want Sam tofinish the report until Friday . b. • Bill wants Sam tofinish the report until Friday . Data like (23) suggest that only might be a trigger for Negative Polarity Items: (2 3) a. No one ever suspected David Alexander. b. *john ever suspected David Alexander. c. OnlyJohn ever suspected David Alexander. The difficulty with this hypothesis about only is the following data, which indicate that only is not, in general, a trigger for Negative Priority Items: (24) a. b. c. d. e. f.

•Bill wants Sam tofinish the report until Friday . *Only Bill wants Sam tofinish the report until Friday . •I was all that keen to go to the party . ?Only I was all that keen to go to the party . • Phil will give Lucy a red cent. • Only Phil will give Lucy a red cent .

Hence I do not find the syntactic observation that only triggers Negative Polarity Items sufficiently well grounded, and so I do not believe that there is yet sufficient ground to take only to be a negative lexical item. Horn ( 1 989: 248-9) has also pertinently raised the question of the negativity

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

which is absurd. So it is not obvious to me that, in general, we should accept McCawley's theory that Not A ifnot B is the preferred paraphrase of A only ifB . Thus it is not obvious that only if can be paraphrased by expressions involving rwo negatives as McCawley ( 1 98 I : so) claims. The problem with McCawley's first alleged fact, viz. that only NP can be paraphrased by No X other than Y, is that this so-called fact is just what is in dispute berween me and Horn and McCawley. I deny that their paraphrase even correctly gives the truth-conditions for only NP. So it avails Horn and McCawley nothing to appeal to this alleged fact in the current debate; they would just beg the question at dispute. McCawley's (198 1 : so) second argument in support of the negativity of only is the alleged syntactic fact that only triggers Negative Polarity Items, items that only occur acceptably in the scope of a negative item, like any, give a hoot , all that , and until Friday .9 For example, though the negative sentence (McCawley ( 1 98 1 : 1 32-3)) in (22a) is acceptable, the affirmative sentence in (22b) is unacceptable:

314

The Importance of Being 'Only'

(25) I love only you, but maybe I don't love you either.

According to Hom's theory, in saying (25) one would not withhold com­ mitment to, and so not undermine the truth of, the original assertion I love only you . But that consequence strikes me as so outrageously counter-intuitive as to be a reductio of the theory. The idea that simultaneously I love only you could be true while I love you is false just seems crazy to me; the following is, in my view,

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

of'only' in terms of the Barwise & Cooper (1981) classification of quantifiers as monotonically increasing, monotonically decreasing, or non-monotone. E.g. Singular Terms, 'many N', 'most N', 'all N', 'some N' are monotonically increasing, in the sense of the theory of Generalized Quantifiers, if the Quantifier contains every superset of every set that is an element of the Quantifier. Since 'Q VP' is true just in case Ext [ VP ] E Ext [Q], the monotoni­ cally increasing quantifiers are ones for which, if Ext[ VP1 ] is a subset of Ext [ VP2], and 'Q VP1 ' is true, then 'Q VP2' is true. Thus, since Some men [entered the race early] VPt Some men [entered the race] VPz • but not conversely, 'Some men' is a monotonically increasing quantifier. Since 'No men' shows exactly the opposite pattern, as No men entered the race No men entered the race early, but not conversely, it is a monotonically decreasing quantifier. If Hom were right that the meaning of'Only Muriel voted for Hubert' is essentially negative, as is 'No one other than Muriel', or 'At most Muriel', 'only' sentences should exhibit the semantic characteristics of monotonically decreasing quantifier sentences. It seems to me intuitively clear that Only Socrates entered the race� does NOT entail Only Socrates entered the race early . 'Only Socrates' is NOT a monotonically decreasing quantifier. Thus 'only' is not to be grouped with 'No one other than' or with 'At most', the English expressions for the negative ASSERTION that McCawley (1981) and Hom (1969, 1 992) have offered as the meaning of the 'only' sentence. Also, it seems to me clear that Only Socrates entered the race early does NOT entail Only Socrates entered the race . Hence 'Only Socrates' is NOT a monotonically increasing quantifier either. It is not to be grouped with Singular Terms, 'Some N', 'All N', etc. Consequently, 'Only Socrates' is a non­ monotonic quantifier. If so, Only Muriel votedfor Hubert does not mean what McCawley ( 1 9 8 1 : 5 1) and Hom ( 1 992) think it does: only does not mean no one other than ! 10 Neither the logical entailments nor the distributional facts of syntax support the conclusion that only a is monotone-decreasing or that only a is a negarive item. Finally, I want to consider some consequences of Hom's ( 1 992) neo-Gricean view. As Hom ( 1 992: 1 82} points out, on his theory I love only you implicates, and does not entail, I love you . As he puts it, "I love only you is not a declaration of love . . . but the recipient . . . is pragmatically licensed to hope for the best". Since, on Hom's theory, I love only you merely implicates I love you , it would be allegedly possible to cancel felicitously an assertion of fidelity by saying:

Jay David Atlas

31S

1 . The first two of the three objections that Hom makes to conjunction/ entailment accounts of Only Muriel voted for Hubert will dismantle the standard conjunction accounts but leave standing the neo-entailment account in Atlas (I99 I). The third objection fails. 2. The alleged data from cancelability and the two arguments for derivability of Muriel votedfor Hubert as an implicatum of asserting Only Muriel votedfor Hubert fail to justify the claim that Muriel votedfor Hubert is an implicatum. J. There are absurd consequences of Hom's (I 992) theory, e.g. that I love only you can be true while I love you is false, and that if Lenin does not belong even in a museum, he belongs only in a museum. 4· Hence, there is still a viable alternative to Hom's (I 992) neo-Gricean account, viz. Atlas's (I99 I ) entailment account, (1 1}-(1 2) above. JAY DAVID ATLAS Dept. ofPhilosophy Pomona College 551 College Avenue Claremont, CA 9 1 7 1 1-6355 USA

Received: 23.1 1 .92 Revised version received: 02.06.93

Acknowledgements An earlier version of this essay was read in the University of Groningen, The Netherlands, in March 1 993. I am indebted to Peter Blok, Jack Hoeksema, Sjaak de Mey, Jan Koster, John Nerbonne, Laurie Stowe, and Frans Zwarts. I am also grateful to Larry Horn for much discus­ sion of these problems.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

just semantically contradictory: I love only you but I do not love you . If any statement is necessarily false, that one is. 11 Further, Hom (I 992: I 82) admits that Guess what: only Kim passed the test! requires that Kim passed the test be a Sperber-Wilson ( I 986) "explicature" rather than a Gricean conversational implicature. On the Sperber-Wilson-Carston­ Kempson-Blakemore view, Kim passed the test then contributes to the truth­ conditons of Only Kim passed the test . I am not convinced, and Hom offers no argument in support of the claim, that Kim passed the test is just an explicature. What is missing from Hom's account is an explanation as to why these sentences are of the same semantic type as Carston's (sa) This park is some distance from where I live . But that aside, what's the difference between Guess what: only you are loved by me! and Guess what: only Kim passed the test!? Why doesn't I love you contribute to the truth-conditions of I love only you in the former case? Of course, I (Atlas I 991) do not think that the proposition is just an explicature; I regard it as an entailment, though not as a conjunct of a conjunction I love you and I love no one other than you . In sum:

3 I6

The Importance of Being 'Only'

N O TES r

A

4

I

T.

am assuming, with Horn, that the semantical presupposition analysis and the conventional implicature analysis are not viable. Horn (personal communication) has presented the following interesting argument:

A is logically equivalent to B . B implicates C . So, A implicates C. Horn's original argument was: The-statement OnlyFs are Gs entails All Gs are Fs . But stating All Gs are Fs implicates Some Gs are Fs . Thus, stating Only Fs are Gs implicates Some Gs are Fs ·

·

·

·

·

·

.

We can now amend this in light of Horn's argument: The statement Only fos are Gs is logically equivalent to All Gs are Fs . But stating All Gs are Fs implicates Some Gs are Fs . Thus, stating Only Fs are Gs implicates Some Gs are Fs •

·

·

·

·

6

·

.

The difficulty with this amended version of the argument was discussed in Atlas & Levinson ( I 98 I : 1-2, 8-9, 12-13 ), where it is shown that logical equivalence is not sufficient to give the same conversational implicata. As I also discussed in Atlas & Levinson (I 98 I), Hom's interpretation of Grice's ( 1 975) notion of "what is said" as truth-conditional content is not an accurate interpretation ofGrice (I 97 5 ). So I do not think that Hom's amended version will succeed in yielding his conclusions either. It is important to this argument that the speaker knows, in the case of the utter­ ance Only a, b, and c are F, that, e.g., c does not F. It is not merely that the speaker knows that no-one Fs in the case of an utterance of the form Only Gs are F. It is essential to this argument that proper names occur in the sentence uttered.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

2

Conversational imp!icatures are non­ detachable from 'what is said', and 'what is said' is, according to Grice (I975), the truth-conditional content of the asser­ tion. Thus, if A and B were logically equivalent, they should have the same implicata. So, the following schema should be valid:

In Atlas (1979) I had suggested that statements, as contrasted with sentences, would possess truth-conditions by virtue of Gticean inferences adding content to the semantically general (non-specific) semantic representations of the literal meanings of sentences. The semantically general sentences themselves did not possess well-defined truth-conditions. Versions of this point have now come to be called 'pragmatic intrusion into truth­ conditions' (see Levinson I 988). Grice (I975 ) never seems to have considered the possibility of this phenomenon. (6) are sentences that I, and, rm _sure, many other speakers would find accep�­ able in the interpretation: 'Even a museum is such that Lenin does not belong in it' (Horn I 989: I 49, I 5 2). There is a temptation to wonder why (7) is not "just a more emphatic form" of (8) rather than a sentence with a distinct meaning. Here's why one should not be so tempted: ( I ) Contrast the statement The king oJFrance is bald with The king oJFrance exists and he is bald . The latter statement does not presuppose the existence of the king of France, while the former state­ ment does presuppose it. That is a significant difference in meaning between the two statements; the latter is not "just a more emphatic form" of the former. (2) Let A be the axioms of Euclidean plan geometry, and let T be the theorem that the bisecting line segments of each side of a triangle meet at exactly one point. Clearly A & T is not just a more emphatic form of A even though

Jay David Atlas 3 I 7

8

9

10 II

poses a further, and subtle, problem for research. A referee for this Journal remarks, without amplification, that 'the argument [Atlas] gives . . . for abandoning the epistemically qualified test (on the grounds that even unrecognized entail­ ments will then pass the test) will, I think, fail if Gazdar's strategy is followed, whereby all implicatures are derived from assertions under epistemic qualification'. This vague but suggestive remark merits more extensive discussion than I can provide here. I offer it to the reader who has an interest in tests of implicature cancelation. One of the referees for this Journal explicitly raised the question of the significance of such an alleged fact for Horn's and McCawley's views. For further discussion of the logical semantics of 'only', see de Mey (I99 I ). I have taken a certain amount of'flak' for the alleged force of my brute intuition here. Those who know my work know that I have, with even startling regularity, held theoretical views, e.g. about the non­ existence of scope ambiguities in negative sentences, or about the topics of negative existence statements, that are hardly intuitive to many philosophers, though compelling to a number of Chomskyan linguists. I just have a very hard time with Horn's theory about the truth-conditions of I love only you .

RE FE RE N CE S Atlas,]. D. (I 979), 'How linguistics matters to philosophy: presupposition, truth, and meaning', in D. Dinneen & C.-K. Oh (eds), Syntax and Semantics 1 1 : Presupposition , Academic Press, New York, 265-8 1 . Atlas, J. D. (I 984), 'Comparative adjectives and adverbials of degree: an introduction to radically radical pragmatics', Linguistics and Philosophy, 7, 347-77. Atlas, ]. D. (I99 I), 'Topic/comment, presup­ position, logical form and focus stress

implicatures: the case of focal particles only and also ',journal ofSemantics, 8, I 2747· Atlas, J. D. & Levinson, S. C. ( I 98 I ), 'It­ Clefts, Informativeness, and Logical Form: Radical Pragmatics (Revised Standard Version)' in P. Cole (ed.), Radical Prag­ matics, Academic Press, New York, I -0 1 . Barwise, J . & Cooper, R. ( I 9 8 I ), 'Generalized quantifiers and natural language', Linguis­ tics and Philosophy, 4, I 59--2 I 9.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

7 A referee's comment has reminded me that I should mention that non­ cancelability does not, by itself, demon­ strate the existence of an entailment. Indeed, in Grice's (I96 I ) original view, non-cancelability does not distinguish between entailment, semantic presup­ position, and what he later called 'conventional implicature'. But Atlas (I99I) criticizes the semantical presup­ position view, while the later Horn ( I 979) abandons the semantical presup­ position view of Hom (I 969) in favor of the conventional implicature view, and Horn (I 992) abandons the conventional implicature view for the conversational implicature view. I do not here, nor do I in Atlas ( I 99I), offer critical arguments directly against the conventional impli­ cature view, as I have been accepting Horn's own rejection of the conventional implicature view. Horn (personal communication) has remarked, 'It does appear that, as [you] maintain, Muriel votedfor Hubert is not a cancelable implicatum of Only Muriel voted for Hubert, at least not without a significant revision of the original notion of cancelation'. He goes on to say, nevertheless, that 'there does appear to be an important asymmetry between the epistemically­ qualified suspension of the positive component of only sentences and the lack of any such suspension possibilities for the negative component'. Here Horn

3 I 8 The Importance of Being 'Only' Horn, Laurence R {I 992), 'The said and the unsaid', in C. Barker & D. Dowty (eds), Proceedings of the Second Conference on Semantics and Linguistic Theory (SALT II), May 1 -J, 1992 . Working Papers in Linguistics No. 4 0 . Columbus, Ohio State University, Dept. ofLinguistics, I 63-92. Kantunen, L. & Peters, S. ( 1 979), 'Conven­ tional implicature', in C.-K. Oh & D. Dinneen (eds), Syntax and Semantics 1 1: Pre­ supposition , Academic Press, New York, I 56. Levinson, S. C. { I 988), 'Generalized conversa­ tional implicature and the semantics/ pragmatics interface', Stanford, Stanford University, Dept. of Linguistics. McCawley, James {I98 I), Everything That Linguists Have Always Wanted to Know About Logic: But Were Ashamed to Ask, University of Chicago Press, Chicago. Mey, S. de (199 I ), ' "Only" as a determiner and as a generalized quantifier', Journal of Semantics, 8, 9 1 - 1 06. Sadock, Jerrold (I 984). 'Whither radical pragmatics?' in D. Schiffri n (ed.), George­ town University Round Table on Languages and Linguistics 1984 , Georgetown Univer­ sity Press, Washington, I 39-49· Sperber, D. & Wilson, D. {I 986), Relevance: Communication and Cognition , Harvard University Press, Cambridge.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Blakemore, D. {I 992), Understanding Utter­ ances, Blackwell, Oxford. Carston, Robyn (I 988), 'Implicature, explica­ ture, and truth-theoretic semantics', in R Kempson (ed.), Mental Representations, Cambridge University Press, Cambridge, I 5 s-8 I . Geach, Peter {I962II 98o), Reference and Generality , Cornell University Press, Ithaca. Geis, M. {1973), 'If and unless', in B. J. Kachru et a/ . (eds), Issues in Linguistics: Papers in Honor of Henry and Renee Kahane, Uni­ versity of Illinois Press, Urbana, 2 3 1 - 5 3 . Grice, Paul (196 I ), 'The causal theory of perception', Proceedings of the Aristotelian Society, 25, I 2 I -52. Reprinted in R.J. Swartz (ed.), Perceiving, Sensing, and Know­ ing, Anchor/Doubleday, Garden-City; NY ( 196 5). 43 8-72. Grice, Paul {I97 5), 'Logic and conversation', in P. Cole & J. L. Morgan (eds), Syntax and Semantics J: Speech Acts , Academic Press, New York, 41-58. Horn, Laurence R (1 969), 'A presupposi­ tional analysis of only and even ', CLS, s. 97-108. Horn, Laurence R {I 979), 'Only , even , and conventional implicature', paper presented at the annual meeting of the Linguistic Sociery of America, Los Angeles. A Natural History of Negation , Chicago Uni­ versity Press, Chicago.

Journal ojSI"tnnJntics

10: J I 9-J2S

© Oxford University Press 199 3

Book Review

Noel Sharkey (ed.),Connectionist Language Processing: Readingsfrom 'Connection Science', Kluwer Academic Publishers, Dordrecht, 1 992. 384 pages, £ s r.so (Hardbound. Also available in paperback from Intellect Books, Oxford.)

MARTIN HoELTER AND RoLF WILKENS

the generative approach contains large intellectual gaps-e.g., concerning the relationship between underlying competence and observable performance, or the nature of interactions

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Approaching connectionist systems from the perspective of a non-cognitive hardliner, one could begin-and end-with one of the more notorious quotes from Gazdar et a!. (1 985: s ) and state that grammatical theory is not 'a biological theory of the structure of an as-yet-unidentified mental organ'. Such a position w.ould render the major thrust of connectionism largely redundant for linguistics. On the other hand, or coast, if you wish, one could take up a cognitive stance that includes just this mental organ in the paradigm oflinguis­ tic research. Starting from the assumption that 'a person who speaks a language has developed a certain system of knowledge, represented somehow in the mind and, ultimately, in the brain in some physical configuration', Chomsky (1988: 3), for instance, put forth four essential questions of linguistic inquiry in his 'Managua lectures', the last of which was 'What are the physical mechan­ isms that serve as the material basis for this system ofknowledge and for the use of this knowledge?' From this point of view then, connectionism could defi­ nitely enrich research in (cognitive) linguistics. Even if one ignores the concrete biological aspects altogether, one cannot deny that connectionist systems nevertheless have certain properties which are potentially interesting for computational linguistics-independently of the fact that they are commonly associated with brain style or neuronal architecture and behavior. It's a rather crude analogy, but, among other things, connec­ tionism does to natural language processing-or any kind of information processing for that matter-what 'fuzzy' does to logical systems. Hence, from the point of view of cognitive as well as computational linguistics, connection­ ism bears the potential of a fruitful paradigm. The foundation of connectionism was certainly laid by the impressive two­ volume Parallel Distributed Processing: Explorations in the Microstructures of Cogni­ tion by Rumelhart et a/. ( r 986). These two massive volumes were acclaimed-and justifiably so-in a review article by Sampson ( r 987):

320

Book Review

between innate knowledge of language and primary l inguistic data through which language­ acquisition occurs. These gaps remain, despite the fact that, in reality, generativists have now had thirty-odd years to try to fill them. If PDP/connectionist researchers continue to make progress at their present rate, the challenge they offer will soon become irresistible. My instincts suggest ro me chat they will indeed prevail. (p. 886)

o o o o

For a given problem . . . choose a rule-based approach (or approaches) to the problem . . . and a target set of data that exemplify the problem. Based on the selected rules, identify a set of relevant fea tures for describing the structures appearing in the data. Based on the selected rules, identify which features can plausibly enter together into . . . constraints Embody these constraints as connections in a network.

Since the relevant constraints are already supplied by the theory to be implemented, Rager's & Berg's model simply is a 'synthesis of two important research trends: the development of principle-based . . . linguistic models and the renaissance of network or connectionist models of computation. The two come together in a connectionist model of movement in GB theory' (p. 28). It remains unclear, however, who the potential audience for such work is supposed to be, since its impact on linguistics as well as on connectionism is rather harmless. From a linguistic point of view, this article simply proves that the assumed processes can be implemented in a connectionist network and nothing more. It adds neither theoretical nor empirical insight to the theory. Connectionists, on the other hand, might have reservations about the approach since most 'pure' connectionists regard connectionist systems as typically, or even necessarily, showing the following properties: o o

They can process 'noisy' information, i.e. information that is incomplete or interfered by external sources. They employ a number of well-studied learning algorithms (backpropagation, Boltzmann­ machines, Kohonen-nets, Hebbian learning) that enable them to 'learn' from input data or

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Certainly, more linguists shared these high hopes, and the question is whether Connectionist Natural Language Processing can fulfil the expectations with which, six years after the publication of the groundbreaking work in the field, linguists might approach this collection ofseventeen articles that constitute 'some of the best recent work' (p. vii) on the subject, and 'represent much of the state of the art' (p. v). We will concentrate here on just a few articles that are representative and/or interesting under several aspects. Let us begin with 'A connectionist model of motion and government in Chomsky's government-binding theory' by Rager & Berg. The article can perhaps best be characterized as a venture in 'ill1pl_e��ent_atio11al connectionism'. The general strategy of this type of connectionism wa�- d-es�nbed by Legendre et a/. ( 1990: -6):

Book Review

o

321

generalize from examples. Also, stochastic regularities can be established across several modules. Due ro their architecture they are massively parallel. Hence they are ideally suited for the simulation of cognitive processing, which is generally assumed to employ massively parallel processes extensively.

·

·

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Due to the extreme complexity of the mapped process (move-a ), it is more than understandable that only one of the above characteristics-parallelism-is met by Rager's and Berg's network. Considering this work a first approach to an interesting field, the authors offer no further perspectives on how this network could be extended to become more connectionist or enrich the mapped theory. Clearly, the conception of physical realization of human thought, or more specific, language, and the representation of the relevant mechanisms in an adequate model is an important matter in our context. The blurb of the present volume tells us that 'connectionism is a new information-processing paradigm which attempts to imitate the architecture and process of the brain'. This hints at the feature that distinguishes the books and articles on 'Natural Language Processing' from the present collection: the connectionist approach to the well­ established discipline aims at opening a new frontier of linguistic research. If one accepts Chomsky's motivation for carrying the abstract and symbolic linguistic strategy of analysis that has generally been pursued so far to a concrete physical level, one must necessarily ask for the right methods by which this can be achieved. Kempen's and Vosse's outstanding article 'Incremental syntactic tree formation in human sentence processing: a cognitive architecture based on activation decay and simulated annealing' lucidly shows how biological processes can serve as a model for the design of a parser and, in a wider sense, for the design and biological interpretation of a unification-based grammar. Motivated by the process of protein biosynthesis, they implemented an architecture, the so-called unification space, 'which simulates well-known sentence understanding phenomena ' (p. 8 3 ) like right association, minimal attachment, and processing difficulties increased by lexical ambiguities and center-embedded clauses. Thus, this approach displays a 'computational model of a psychologically plausible parser' (p. 8 8) and covers a lot of the central issues of cognitive linguistics. Although the authors state that this architecture is not connectionist, they use several well-known techniques of connectionism, among them simulated annealing and activation decay. Furthermore, the system includes typical features of connectionist networks like a set of homogeneous processors only affected by irs neighbors and their state of activation. The parsing process can briefly be described as a combination, or procedural unification, of unary local trees, so-called segments, with attribute-value matrices as nodes. While scanning the input, the parser retrieves either one segment, or in the case of

322

Book Review

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

categories which require other constituents, a set of such segments from the lexicon. For example, the segment retrieved for a simple proper noun consists of a child node specifying the lexical head as an N, and a parent node specifying the complete category as an NP. For a transitive verb, the lexical access procedure yields a set-or a mobile-of three segments, defining the subject, head, and object, respectively. This mobile might be regarded as a 'tree' without any precedence relation. Hence word order is not specified. Each of these segments is assigned an activation value and is kept in the unification space. Two randomly-chosen nodes from different segments unify with a probability of P(U). Alternatively, two unified nodes are broken up with a probability of P (B ). The task of the system is to settle down in stable state when an optimal configuration of segments is encountered. This stable configuration is characterized by low values for both probabilities, meaning that almost no additional segments are combined or broken up. In many connectionist systems similar goals are achieved by simulated anne�ling. In such systems, �he likelihood of an action taking pl�ce is_controlled _ __ __ __ by a global temperature. At a high temperature, the system is in a chaotic state, at a low temperature very few actions take place, and hence the system is quite stable. Kempen & Vosse impressingly show that this probability can be computed in a linguistically plausible way by, for example, defining a function on the quality of the analysis. Additionally, modifications of the temperature can be employed to simulate special conditions like mental efforts or pathological behavior (aphasia). Consider now Lucas's & Damper's 'Syntactic neural networks' as an instance of the problems we have with the book under discussion. Don't let yourself be fooled by the title: it is not an article on natural language syntax. The network depicted here was trained to recognize hand-written digits. The authors claim that the 'new paradigm is applicable to a wide variety of pattern-processing tasks such as speech recognition and character recognition' (p. 56), but from our point of view, it just seems to be asking too much of readers to imagine the linguistic relevance of quasi-optical pattern recognition. The present model tries to achieve the desired results by inferring a strictly hierarchical, context­ free grammar. Intellectual effort and achievement are quite impressive here, but it appears to us that this kind of grammar might not be ideally suited-to put it mildly-for the application to natural language. Another problem from the point of view of a linguistic reader can be exemplified by Wermter's & Lehnert's 'A hybrid symbolic/connectionist model for noun phrase understanding'. While discussing the integration of semantic relationships with syntactic constraints in localise connectionist networks, the authors employ two syntactic constraints. These theoretical considerations are explained in full by eleven lines of text. We learn, for instance, that the 'no-crossing constrant [sic ] . . for noun phrases means that .

Book Review 32 3

branches for attachment do not cross. The following (constructed) example shows a violated no-crossing constraint: Influence of the temperature on the electrons in Fahrenheit' (p. 109). This constitutes the theoretical syntactic part. Now to semantics. Consider the following semantic features for nouns: ELEcrRIC-OBJECT, MECHANISM,

SCIENTIFIC-FIELD,

ORGANIZATION-FORM,

GAS,

(p. IOJ). These features were of course extracted from a technical corpus in the domain of the physical sciences, but for a linguist it is extremely difficult to see the theoretical claim behind an approach that works with this kind of ad hoc concepts.And again, this is still a good connectionist article. The bulk of the remaining articles in the present book more or less fall in the same category. Fortunately, there are highlights for linguists in the connectionist paradigm­ and volume under discussion, too. Apart from Kempen's & Vosse's excellent article, there is one more that clearly stands out from the collection. 'The role of similarity in Hungarian vowel harmony: a connectionist account' byMary Hare perfectly demonstrates what linguists should be justified in expecting from connectionist approaches to natural language processing or how they should proceed when adopting this new paradigm. While depicting a connectionist model, it still remains a genuine linguistic article-and this can definitely not be said of most of the other contributions. It brings theoretical insight, principled explanations of data, and generalization to the subject by applying a new methodological approach to it. Furthermore, the article is exemplary in lucidity, organization, and argumentation. Hare starts by presenting the relevant data from Hungarian, moves on by encapsulating the current theoretical analyses and giving an introduction to sequential processing in connectionist networks. She then states the conditions on assimilation in the sequential network, and, after reporting the results of the network training and a discussion, finishes with a modeling account of vowel harmony. The paper is rounded off by a wonderfully concise and straight­ forward conclusion. The present connectionist account of Hungarian vowel harmony is firmly rooted in the framework of autosegmental phonology-incidentally, the bibliography gives a complete overview of the central works in the nonlinear paradigm of phonology. The motivation for modeling harmony processes by way of a connectionist network resides in the fact that MECHANISM, MEASURING-EVENT, TIME, ENERGY, MATERIAL, EMPTY

This is where the connectionist approach comes in:

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

ideally, . . . we would like [sic ] locate [sic J general, independently motivated principle which predicts the harmonic or non-harmonic behavior of all classes of roots in Hungarian. This paper is an acrempt to establish such a principle. As I show below, the harmonic behavior of the Hungarian vowels can be predicted from the similariry relationships among the vowels, together with the temporal context in which it occurs. (p. 307)

324

Book Review

it offers a simpler and more explanatory account of the vowel harmony process. The vowels which will exhibit transparent behavior, and the environments in which this behavior will change, must be stipulated arbitrarily under the alternatives discussed in Section JII [i.e. the recent autosegmental analyses within non-connectionist linguistics], while both follow automatically from the account suggested in this paper. (p. po) In short, this model pulls the problem straight just beautifully. Especially in the present context, it is not without significance that Hare's article is virtually self-sufficient. Motivation and background information are rich enough to let readers approach it from arbitrary directions. The article is perfectly amenabl� for linguists as well as for connectionists not familiar with recent linguistic theory, and thus it is ideally suited for not only bringing

We cannot avoid a few remarks on editorial matters here. The volume under

review comprises seventeen articles on connectionist natural language process­ ing, all of which have already been published in the journal Cgn'!ection Science between

1 989 and 1 991. Eight

of these articles appeared in a special issue on

natural language processing, two of them in a special issue on so-called hybrid systems. The editorial of the former issue also serves as the introduction to the present collection. Each of the seventeen articles constitutes a separate chapter, and the editor offers no further organization except for stating that the book 'is laid out roughly in the traditional categories of language research starting with syntax and moving through question answering to knowledge application and speech processing' (p. v).

The

articles were apparently reproduced directly from the journal and

supplied with a different headline and page numbering. Since the present volume has a smaller format than the journal, pages appear irritatingly cramped (sometimes the text ends only one-tenth of an inch from the bottom of the page). Strangely, one article is set in a different typeface.

In combination with a half-page preface and a short index, which constitute the only original contributions to this volume, the collection makes the impression of having been compiled

en passant

and without much effort. The

one or two dozen typographical errors (for example, Sam pson, G. ( 1 987) is given as Samson, S. ( 1 987) (p.

27) and Behavioral and Brain Sciences becomes Behavior and Brain Sciences (p. 27), Pullyblank, D. is cited as Poullyblank, D. (p. 322)) were

forgivable in the journal, bur their second appearance in the present volume shoul d have been avoided. Let us conclude with a personal and subjective note. Connectionist approaches to linguistic or psycholinguistic topics are frequently confronted with stereotypical j udgement: 'unsound hacks for the hacks' sake', 'the zillionth net that learns to do something in only soo runs on a

VAX', or the like. While

there are numerous excellent and promising works in this framework, a vast

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

together both worlds but also enriching the inherent discussions within each discipline.

Book Review

32 S

number of connectionist systems simply lack motivation and fruitful impact on linguistic theory. Thus the outsider's view on the subject is in general distorted. From our point of view, most linguists still just don't know what to do with connectionism. Alas, Sharkey's book will not solve this sad dilemma. Sprachwissenschaftliches Institut Ruhr-Universitiit Bochum D-44780 Bochum Germany Email: hoelter@ linguistics.ruhr-uni-bochum.de wilkens®linguistics.ruhr-uni-bochum.de

Chomsky, N. ( 1 988), Language and Problems of Knowledge: The Managua Lectures , MIT Press, Cambridge, Mass. Gazdar, G., Klein, E., Pullum, G., & Sag, I. (198 s). Generalized Phrase Structure Gram­ mar, Basil Blackwell, Oxford. Legendre, G., Miyata, Y., & Smolensky, P. (1 990), 'Harmonic grammar-a formal multi-level connectionist theory of lin­ guistic well-formedness: an application', ICS Technical Report No. 90-4, Uni­ versity of Colorado at Boulder.

Rumelharr, D. E., McClelland, J. L., & the PDP Research Group ( 1 986), Parallel Distributed Processing: Explorations in the Microstructure ofCognition, Vol. 1 and 2 , MIT Press, Cambridge, Mass. Sampson, G. ( 1 987), Review article on 'Rumelhart, D. E., McClelland, J. L., and the PDP Research Group ( 1 986), Parallel distributed processing: explorations in the microstrucrure of cognition, Vol. 1 and 2', Language, 63, 871-86.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

REFERE N C E S

Journal '!f&mantics

10:

327-329

© Oxford University Press 1993

Book Review

Artikelworter im Deutschen: Semantische und prag­ matische Aspekte ihrer Verwendung (Linguistische Arbeiten 267), Max Niemeyer Verlag, Tiibingen, 1 99 1 , 1 72 pages.

Hansjorg Bisle-Miiller,

EMIEL KRAHMER Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Artikelworter im Deutschen contains four chapters. It starts off with a small chapter on methodological and theoretical issues. The second chapter concerns definiteness (Bestimmtheit ). Here Bisle-Miiller lays the foundation for the two final chapters. In the third chapter he discusses the meaning and use of the various articles in German, and the last chapter concerns their generic uses. The book contains a lot of interesting data and discusses a large part of the relevant (mostly German) literature. But the book is also interesting for linguists and philosophers interested in the articles in general. It seems to be a reasonable hypothesis that all languages code definiteness and related matters in one way or another, at least to some degree. And apart from that, German has a rich article paradigm as well as some well-known interesting features: there are dialects (Monchengladbach, North Frisian) which have two definite articles with different functions, and in more informal German there is an interesting difference between accentuated articles like zu dem ('to the') and cliticized ones like zum . Next to linguists and philosophers, Bisle-Miiller explicitly aims his study at people who want to learn German as a second language. He correctly points out that it is very difficult to illustrate the uses of, say, the definite article without also taking its alternatives into consideration. This motive of German as aforeign language also partly determines the methodology of the book. Bisle-Miiller aims to describe the use of articles in spoken discourse and to do this he makes use of what he calls Gricean reasoning (Rasonnement ). Mutual knowledge plays a crucial role in this reasoning. With the use of a certain article the speaker gives a clue to the hearer what kind of reasoning he has to do to find a referent in the mutual knowledge. In other words, the various articles function as co-ordinators of this mutual knowledge. Bisle-Miiller distinguishes two sorts of mutual knowledge: background knowledge (Dauerwissen ) and contextual knowledge (Laufwissen). The con­ textual knowledge is the only relevant part of mutual knowledge for understanding. It contains information about the context (in the broad sense), as well as activated pieces of information from the background knowledge

328

Book Review

-

-

According to Bisle-Miiller a script for 'hitting a nail in the wall' contains the elements 'hammer' and 'finger'. There is one hammer so there the definite article is expected, but there are ten fingers in this script. Now the reasoning goes as follows: When we exclude the five fingers of the hand that holds the hammer, we only have five fingers left to choose from. But since the number of fingers can be overlooked, and moreover the finger closest to the nail (i.e. the forefinger) is the most probable referent in question, the definiteness of the referent is unproblematic. And since the thumb d irectly next to the nail is usually referred to with the description 'thumb', this alternative can also be excluded (p. 54, my translation).

But if this kind of reasoning is allowed, we have a serious problem: how can we explain that in a lot of cases where we have ten possible referents a possessive (or some other non-indefinite article) is not suitable? Take for instance: (2) During an argument, Petra hit Hans so hard that she broke his tooth. One cannot say this out of the blue, neither in English nor in German. Still, a similar kind of reasoning can be applied here: the tooth in question is probably one of Hans's front teeth, it is very unlikely that it is one of his molars, and moreover we can describe these with another word, namely 'molar', etc. These examples show that although Gricean reasonings and the frames they trigger might answer some questions, they raise a number as well. How can we find frames for sentences? And what kinds ofGricean reasonings are allowed or not allowed? In general, how can we expell infolicitous article uses? Bisle-Miiller is correct in stating that mutual knowledge and context are

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

(world-knowledge, linguistic knowledge, information about communication patterns). Speaker and hearer use this mutual knowledge for generating frames or scripts (Rahmen ) which aid to determine referents. The presupposition behind all this is that, no matter what the mother-language is, Gricean reasonings about mutual knowledge can be carried out by anyone. Bisle-Miiller's approach is obviously related to the one presented in Hawkins (I 978). Hawkins also argues for a Gricean treatment of the various articles. This is not the place to dig deep in the differences in set-up and predictions between Bisle-Miiller and Hawkins, but, in a nutshell, I think it is fair to say that Bisle­ Miiller's criticism of Hawkins is largely correct, but the solutions he offers are not the final answers either. Gricean reasoning is a strong mechanism. To see this let us discuss one type of example which is interesting in that it requires a different article from what one would expect. In this case, we take a possessive (which normally require a 'definite' referent), where an indefinite might seem more appropriate. (I) -Bruno wanted to liii: -a-nail in the wall, out- iii doing so he rut -/ii:sjlnger with the hammer.

Book Review

3 29

important features for a theory of the articles, but we have already learned this from the writings of Searle and Hawkins. Bisle-Miiller applies his Gricean reasonings to a broad class of examples. But since he does not give a recipe for 'finding' the right frame, or for restricting fhe possible reasonings, I think non­ Germans will still have difficulties with the German article paradigm. And this is as good a sign as any that more work needs to be done. Institutefor Language Technology and Artificial Intelligence University ofTilburg P.O. Box g0153 sooo LE Tilburg The Netherlands

Hawkins, J . A. (1 978), Difiniteness and Indifiniteness: A Study in Reference and Grammaticality Prediction , Croom Helm, London.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

RE F ERE N C E S

J O U R NAL O F S E MA NT I C S Volume

10 (1993)

CONTENTS

Articles

The Importance of Being 'Only': Testing the Neo-Gricean Versus Neo-Entailment Paradigms

301

RE INHARD B LUTNER Dynamic Generalized Quamifiers and Existential Sentence in Natural Languages

33

CAROLA EscHENBACH Semantics of Number

I1

MANFRED KRIFKA Focus and Presupposition in Dynamic Interpretation

269

PETER LASERSOHN Existence Presuppositions and Background Knowledge

I I3

LAURA A. M ICHAELIS . 'Continuity' within Three Scalar Models: the Polysemy of Adverbial Still

I 93

UwE REYLE Dealing with Ambiguities by Underspecification: Construction, Representation and Deduction

I 23

jAN vAN EIJCK

The Dynamics of Description

239

jAN VAN VooRsT

A Localist Model for Event Semantics

c

65

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

jAY DAVI D ATLAS

C

o

m

p

u

t a t i

o n

a .1

.

L i

n

g

u

i s t i c s

foremost journal devoted exclusively ro computational analyses of natural language. It encompasses Al research in langtiage, linguistics, and the psychology of language processing and performance.

J a m e s F. A l l e n

With each issue providing applied and

Ed itor

theoretical papers, book reviews, technical correspondence, and letters ro the editor, the journal presents

a

stimulating forum for the exchange of ideas and trends in computational language.

Computational Linguistics is the

•

Ywly lrutlruaonal R=: sao.oo.

Add

$14.00 postage and handlina ouuidc

U.S.A. Canadians add additional 7% Gsr.

Prepayment U required. Send

check drswn ag>inst a U.S.

bank in U.S.

funds, MutcrCud or VISA number to: MIT Prcu Joum.W 55

Haywud Street

journal serving a rapidly expanding

Cambric!s<. MA 02142-1399

interdisciplinary field and should be

TEL: 617·253-2889 FAX: 617-25S-6n9

pan of every research library collection. Recommend CL to your librarian today.

[email protected]

Individual subscriptions arc: available only through membership in the A.uociation

fur Compuational l.inguistia. Pl=c conraa: Donald E Walker (ACL) Published quuwly by Th< MIT Pms for
Bdleor<. MRE 2A379

Auociation fur Compuational Unguistia

445 South Strcct. Bo• 1910 Morristown,

ISSN 0891-2017

NJ 07960-1910 U.S.A.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Computational Linguistics is the

J O URNAL O F SEMANT I C S Volumes

1-10

Abstracts of all articles in alphabetical order of author's name

AHLSEN, ELISABETH ALLAN' KEITH Interpreting English comparatives.

s: I -so

This paper attempts to clarify the way in which we interpret English comparatives. It shows that recognition o f a comparative depends primarily on the recognition o f the comparative operator c I. The c I has rwo constituents (I ) a comparative marker which, because there are less than a dozen of them, makes ci readily recognizable; and (2) a scale marker. I argue that comparisons are made on a particular scale, and th::: scales have a supra end and a suh end; the scale marker in c i identifi es which end. Thus the comb ination of scale marker and comparative marker determines the proper interpretation o f the comparative operator, and hence the comparative relation. This interpretation is :tf�ected by the 'committedness' (Cruse I976) and perhaps 'pull' (Rusiecki 1 985) of the scale marker. A comparison identifi es the relative locations o f the comparands X and Y on the scale named in the c i . X, the primum comparationis, is identified through the scope of c i . Y, the secundum comparationis, is recognized through the fact that it is normally a semantic-syntactic parallel to X in a clause introduced by the c2: c2 is normally than or as. The paper ends with detailed discussions o fthe means for translating English comparative constructions into an interpretative metalanguage.

AlLWOOD, j ENS Speech act classification. By T. Ballmer and W. Brennenstuhl Book review.

AI.twoon, J ENs; NivRE, JOAKIM and AHtSEN, ELISABETH On the semantics and pragmatics oflinguistic feedback.

9: 1-26

This paper is an exploration in the semantics and pragmatics of linguisticfeedback , i.e. linguistic mechanisms which enable the participants in spoken interaction to exchange information about basic communicativeJunctions, such as contact, perception, understanding, and attitudinal reactions to the communicated content. Special attention is given to the type of reaction conveyed by feedback utterances, the communicative status of the information conveyed (i.e. the level o f awareness and intentionaliry o f the communicating sender), and the context sensitivity of feedback expressions. With regard to context sensitiviry, which is one o f the most characteristic features o f feed back expressions, the discussion focuses on the way in which the type ofspeech act (mood), thefactualpolarity, and the information status o f the preceding utterance influence the interpretation of feedback utterances. The di fferent content dimensions are exempli fied by data from recorded dialogues and by data given through linguistic intuition. Finally, rwo different ways o f formalizing the analysis are examined, one using attri bute-value matrices and one based on the theory o f situation semantics.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

cf. Allwood, Jens; Nivre,Joakim and Ahlsen, Elisabeth.

2

Index of Abstracts

ARONSZAJN, MARK Thought and circumstance.

ASHER, NICHOLAS Meanings don't grow on trees.

In 'Meanings don't grow on trees' I invesr 6ate Lewis's proposal for using syntactical information to distinguish between intensions. Lewis's proposal, i f it succeeds, would eliminate certain deficiencies in the predictions made by possible world semantics concerning synonymy. I provide rwo criteria for judging semantic theories: descriptive adequacy and explanatory adequacy. I argue that Lewis's proposal concerning synonymy fails on both counts. I then offer a di fferent approach to problems with synonymy. Synonymy judgments involve rwo different kinds o f meaning: truth conditional content, provided by model theoretic semantics, and 'in formation content', provided by semantics in terms o f conceptual role. In developing the notion o f information content, I show how it solves some o f the problems Lewis's proposal addresses.

ASHER, NICHOLAS AND wADA, HAJIME A computational account of syntactic, semantic and discourse principles for anaphora resolution.

We present a unified framework for the computational implementation o fsyntactic, semanric, pragmatic and even 'stylistic' constraints on anaphora. We b uild on our BUILDRS implementation of Discourse Representation {DR) Theory and Lexical Functional Grammar {LFG) discussed in Wada & Asher (1 986). We develop and argue for a semantically based processing model for anaphora resolution that exploits a number o f desirable features: ( I ) the partial semantics provided by the discourse representation structures (DRSs) of DR theory, (z) the use o f syntactic and lexical features to filter out unacceptable potential anaphoric antecedents from the set o f logically possib le antecedents determined by the logical structure o f the DRS, (3) the use o f pragmatic or discourse constraints, noted by those working on focus, to impose a salience ordering o f the set o f grammatically acceptab le potential antecedents. Only where there is a marked difference in the degree o f salience among the possib le antecedents does the salience ranking allow us to make predictions on preferred readings. In cases where the di fference is extreme, we predict the discourse to be infelicitous i f, because o f other constraints, one of the markedly less salient antecedents must be linked with the pronoun. We also briefl y consider the application of our processing model to other defi nite noun phrases besides anaphoric pronouns.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

A long-standing logical and philosophical tradition holds that there are such things as objects of thought , things of the sort a person niay be said to be thinking-objects not only of doxastic thoughts (thoughts to the effect that something or other is the case), hut of wanderings, wishings, hopings and desirings, etc. Virtually all propotranents of this tradition have supposed that the objects o f thought are propositions, the (primary) bearers o f truth-value. There are various proposals within the tradition about what propositions are : but all standard conceptions hold, roughly, that a proposition is circumstantial in character-something akin to a state, or condition, a way things could be. I argue that objects of thought are nor circumstantial in character. So the view that they are propositions, standardly conceived, cannot be right. The argument centers on the case ofon­ doxastic thought-wonderings and wishings, in particular. The b ulk o f this paper, then, is devoted to laying out an alternative conception o f the objects of thought. This conception supports the traditional idea that objects of thought are what we express by our utterance of sentences. Moreover, on this new view, a partial account is afforded of what things are expressed by non-assertorir sentences-by sentences in moods other than the indicative.

Index of A bstracts

3

ATLAS, jAY DAVID 8: 1 27- 1 47 Topic/comment, presupposition, logical fonn and focus stress implicatures: The case of focal particles only and so .

ATLAS, jAY DAviD The importance of being only .

IO:ooo-ooo

I n Atlas ( I 99I) I proposed a novel account o f the logical form of statements having the form 'Only a is F' and the form 'Also a is F', an analysis o f the entailments and o f the implicatures of those statements, and a discussion o f the effects o f focal stress on implicatures. In this paper I discuss the merits of my account over those o fa Gricean account offered by Peter Geach ( I 962 ), Lary Horn ( 1 992) and James McCawley { I 98 1). In doing so I discuss several fundamental problems in Gricean Pragmatics: the nature o f the cancellation of implicatures, the intrusion into truth-conditions of pragmatic inference, Negative Polariry Items, and the non-monotoniciry of 'only a' as a Generalized Quanti fier.

VAN DER AUWERA,jOHAN Against 'Against conversational implicature'. A reaction to Kim Sterelny. BALLMER, THOMAS T. Semantic structures of text and discourse.

2:22 1-2 5 2

The paper starts out with a comparison o f the structural and the dynamic approach. It is maintained that when considering texts and discourses as full b lown linguistic entities, a structural and emically abstracting approach would not be su fficient. A text grammar has to

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In Chapter I 2 of the thirteenth-century Oxford logician William of Sherwood's Treatise on Syncategorematic Words (Syncategoremata ), Sherwood discusses the word only (iantum ), which in the example Only Socrates is running indicates, according to Sherwood, 'how much o f the subject is under the predicate-viz. that the subject Socrates and no more is under it. In that case it is an exclusive word' (Sherwood I 968: 8 I ). In Chapter 7 of the twentieth-century English logician Peter Geach's ( I 962/i 980) Reference and Generality , Geach discusses the words only and alone , remarking that medieval logicians 'were greatly interested in exclusive propositions, but their treatment of them was on the whole superficial. This comes out in their having generally accepted the idea that exclusive propositions were exponi ble as conjunctions-'Socrates alone is wise', say, as 'Socrates is wise and nobody besides (other than) Socrates is wise' . . . I f the force of the exclusive proposition is to exclude everything other than what is named in or by the su bject-term from 'sharing in the predicate', that is no reason for reading in an implication that something named by the subject-term does 'share in the predicate' (Geach I 9621i 98o: 208-9). This dispute between English logicians across seven centuries has been echoed in recent and influential work by the Anglo-American philosopher H. Paul Grice and by linguists, notab ly Laurence Horn in his ( I 969) 'A Presuppositional Analysis o f ONLY and EVEN', in his ( I 989) treatise A Natural History of Negation , Lauri Karttunen and Stanley Peters in their (I 979) 'Conven tiona! lmplicature', and Josef Taglich t in his ( I 984) book Message and Emphasis: On Focus and Scope in English . In this paper I shall argue that neither Sherwood, with his conjunction analysis o f Only x is F, nor Geach, with his non-conjunctive analysis, nor Horn, with his presuppositional analysis, nor Taglicht, with his conjunction analysis o f only and his conventional implicature analysis of also and even , have accounted for the semantic and pragmatic facts, for their analyses have failed to integrate linguistic facts about topic and focus, about entailments, and about Gricean ( I 975: I 989) 'implicatures'. By reconsidering their views I hope to show how a more coherent account can be achieved. In the course of this paper I will o ffer my own analysis, building on ' what I have learned from theirs and, I hope, improving on them.

4 Index of Abstracts account for dynamics, and specifically for the dynamics o f context change. Such a grammar has to ful fi ll a number o f requirements, specifi cally some concerning its formalizability. As an exampl e o f such a formal approach the Context Change Logic for a solution of the Bach-Peters Paradox is proposed. In a second and third part of the paper the missing lexical basis of Logical Language Analysis is criticized. A programme is then presented to give formal logical semantics of natural language a solid linguistic basis. The topology of the semantic spa�e of natural language is developed. This is achieved by reference to a comprehensive srudy of 2 I ,ooo German verb s, I J,ooo German adverb s and an indefi nite numb er of nouns. A last part of the paper demonsrrates how the semantic space of natural language impinges upon text and discourse structures. The expressive power of language is seen to be performed and severely restricted by the lexico-semantic findings presented in the paper.

This paper presents a theoretical foundation of German intonation. It describes the automatic recognition of focus accents and specifies a recognition algorithm. Results of the analyses o f complex utterances concerning their tonal characteristics are presented. This work is based on an acoustic-phonetic model of generating the Fo-contour in German. This generative model that contains phonetic and phonological rules is modifi ed in order to recognize.focal accents. .

BARKER, CHRIS Group terms in English: Representing groups as atoms. What do terms such as the committee, the league, and the group of women denote? Pre­ theoretically, group terms have a dual personality. On the one hand, the committee corresponds to an entity as ideosyncratic in its properties as any other object; for instance, two otherwise identical committees can vary with respect to the purpose for which they were formed. Call this aspect the group-as-individual. On the oth er hand, the identity of a group is at least partially determined by the properties of its members; for instance, a committee will be a committee of women just in case each of its members is a woman. Call this aspect the group­ as-set. Elaborating on suggestions in Link (I 984) and Lasersohn ( I 988), I propose that group terms in English denote atomic individuals, that is, entities lacking internal structure. In parti­ cular, it is not possi ble to determine the membership of a group by examining the denotation of a group term. The proposed account correctly predicts that group terms systematically behave differenrly semantically (as well as syntactically) from plurals such as the men and con­ junctions such as John and Bill. Thus the atomic analysis advocated here stands in sharp con­ trast to previous proposals, including Bennet ( I 97 5), Link (I 984), and Landman ( I 989), in which group terms are considered a piece semantically with plurals and conjunctions. Addi­ tional arguments come from the use o f names of groups as rigid designators, from the parallel between group nouns and measure nouns, and from the distri bution of group terms across two dialects of English.

BARTON, S. B. and SANFORD, A. J. 7 :8 1 -92 The control of attributional patterns by the focusing properties of quantifying expressions.

Recent evidence has shown that certain quantifiers lfew , only a Jew) and quantifying adverbs (seldom , rarely) when used tend to make people think o f reasons for the small proportions or low frequencies which they denote. Other expressions single out small proportions or low frequencies, b ut do not lead to a focus on reasons (e.g. a Jew; occasionally). In the present paper, these o bservations are applied to the attri bution of cause in short two-line vignettes which make reference to situations, and where su bjects have to say what is special in bringing out the

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

BANNERT, RoBERT Automatic recognition of focus accents in German.

Ind ex of A bstracts 5

BARTSCH, RENATE The construction of properties under perspectives.

Adjectives expre: s properties, is the trad itional opinion, be it properties an indivi d ual has itself (absolute properties) or properties that it has only in relation to others (relative properties or relations). I shall show in this paper that most properties are not expressed by adjectives, rather they are d enoted by them in what I call 'thematic d imensions'. Properties are expressed by thematic d imensions and adjectives together. Adj ectives can be used in pred icative, ad nominal, adverbial, or adsentential position and function. Besides elaborating the notion of 'thematic d imension' and explaining the relationship between properties and these dimensions, an aim of this paper is to assign suitable semantic types to adjectives and the 'thematic d imensions' they are used in. These serve for forming conjunctions and other combinations of thematic d imensions, and for forming conjunctions of expressions in several categories. These operations form the basis for the construction of properties und er perspectives, i.e. in thematic d imensions.

BATLINER, ANTON Deciding upon the relevancy of intonation features in the marking of focus: a statistical approach.

We present results on how focus is marked intonationally in German. Six untrained speakers prod uced a corpus of 360 sentences. The corpus was constructed in such a way that sentence mod ality and place of focus coul d only be d ifferentiated by intonational means. Acoustic features representing the parameters pitch, d uration, and intensity were extracted manually or automatically. The relevancy of these features and the effect of several transformations were tested with statistical method s (d iscriminant analysis). Perceptual experiments where the listeners had to deci de upon the place of the focal accent and to ju dge the naturalness and categories of the utterances were performed as well. By calculating average values for the (appropriately transformed) relevant features we found 'normal', prototypical cases; by looking at utterances where all listeners agreed on the naturalness and (intend ed) categories we arrived at coincid ing results. At the same time we found 'unusual' but regular prod uctions. Finally, the speaker-speci fi c use of the d ifferent parameters is d iscussed and the question is add ressed as to whether the parameters can be classi fied as relevant or irrelevant for the intonational marking of focus.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

state of affairs depicted . The proced ure is stand ard in the area of social psychology known as attribution theory, but the present experiment is concerned with the role of quantifYing descriptions in the process. Two theories are contrasted. The first, the frequency signalling theory, ascribes the peculiarity of an action to the frequency of that action in an ind ividual versus the frequency of it in the population at large. The second , the focus control account, says that contrasts are only important if one or more of the quanti fiers focuses attention on cause (i.e. serves as a comment on the frequency or proportion which is denoted ). The results support the second hypothesis, and suggest that frequency signalling alone is not enough to generate attri burional patterns. Apart from ind icating an important bound ary condition on attributional effects, the results show the important consequences of the non truth-functional aspects of the meaning of quantifi ers previously reported in Moxey :·nd Sanford ( 1 987). The attri butional effects are clearly depend ent upon linguistic phenomena, a point largely ignored by attrib ution theorists until recently.

6 Index of A bstracts

DE BEAUGRANDE, RoBERT Semantics and text meaning.

5:s9- r 2 r

The activity of 'doing semantics' is a specialized mode of being meaningful. The relation between that mode and the general mode o f ordinary d iscourse should be clarified within a framework that integrates past trends in semantic modelling with each other and with possible future ones. Th is paper proposes to view language as a complex system of control levels with characteristic distribution of determinacy. Here, meaning is described as a processing event whose structure is formed by the maintenance of control and the limiting of indeterminacy. Known properties of complex physical and biological systems offer some clues about how such operations migh t occur.

In this paper I will argue that evaluative adjectives, such as good, bad, clever and skilful, should be analysed as one-place predicates in logical translation. This approach, which is basically th e traditional logical treatment of 'absolute' adjectives, is to be contrasted with th e approach in Montague { I 974a) and Parsons { I 972), wherein all adjectives are translated as two-place predicates, i.e. as semantic attributives. The move away from the Montague-Parsons analysis is nor-new; Bartsch { I 972, -I 975).- McConnell-Ginet {I 973), Kamp { I 975), Siegel {I 976a, I 976b , I 979), Keenan & Faltz {I 978) and Klein {I 98o) have similarly advocated one-place predicate status, at least for fairly straigh tforward qualities (e.g. red, carnivorous , stony ) and even for degree adjectives {e.g. tall, short , heavy and old ). Evaluatives, however, remain trou blesome: Kamp concluded that th eir status was uncenain, and Siegel classified th em as two-place predicates a fter much argument. My remarks are directed primarily against Siegel's analysis; I intend to show that th ere are syntactic tests, some suggested by Siegel herself, which argue persuasively th at evaluative adjectives should be interpreted as one-place predicates.

VAN BENTHEM, jOHAN and VAN EIJCK, JAN The dynamics of interpretation.

1 :3-20

In current semantic theory compositional interpretations are assumed to go from linguistic items to their denotations in some model. This perspective still leaves room for a more dynamical account o f how such interpretations are actually created. One natural idea is to assume that each sentence in a discourse is understood through some representation, 'mediating' between the language and its models. Thus, th e old relation of interpretation splits up into two new ones, viz. that between linguistic items and their representations, and that between th ese representations and actual models. Now, at the Cleves conference it was clear that discourse representations are many things to many people. Some view them as syntactic constructs, some as psychological ones (yet others prefer to remain confused over th is issue). Again, one popular metaph or is that of the partial picture of reality, another th at of a procedural recipe for verification. Finally, these representations are supposed to explain such diverse phenomena as anaphora and progressive discourse information. It is not obvious that one coh erent notion could do all thesejobs. On the other h and, it is not obvious either that one need not try. Th e purpose o f th is paper is to clarify some logical issues concerning discourse representations, while trying to bring together two o f the main themes at the Cleves conference, viz. representation proper and the topic of partia l information. General considerations will be found in section I ; section 2 contains applications and illustrations drawn from the two best-developed formal paradigms o f discourse semantics (c£ Hintikka {I 979), Hintikk a & Carlson ( I 979), Kamp {I 98 I )). It is our contention that more clarity as to the narure and the purpose o f discourse representation wil l u nite, rather than divide the various currents in this developing area.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

BEESLEY, KENNETH REID Evaluative adjectives as one-place predicates in Montague Grammar.

Index of A bsrracrs

BEUKEMA, FRITS Chronological relations and free adjuncts in English.

7

4: 1 0 1 - 1 1 5

In this article I provide a discussion of chronological relations in rhe conrexr of English free adjuncts, which is based on rhe theory of tense proposed in Rigrer ( 1 982). After a b rief review of rhe main points of this theory, I address rhe problem of selecting rhe relevant domain in which rhe rime perspective o f the free adjunct and rhe modi fied clause are processed. I rhen discuss the availab le means ro express chronological relations in free adjunct. The final section deals with rhe influence of sequence indicators like before and after.

BrERwrscu, MANFRED Tools and explanations of comparison (Part I).

6: s 7-93

(a) The new theory accounts for a number o f relevant facts that have systematically been ignored by earlier analyses. (b) Ir relates these facts to those already analysed in a way which does not merely give a descriptive account, but rather an explanation in terms of a few underlying conditions from which the whole range of facts follow in a natural way. A detailed discussion of the various analyses proposed so far would by far exceed the limits set for the present paper.2 I will instead simply list, for the sake o f preliminary orientation, the main points that the present theory shares with some or all of its predecessors, and those in which it differs from them. In accordance with other approaches, I will make the following assumptions: (i) The Positive of relative adjectives must be analysed in close connection with the Comparative, the Equarive, and a number of related constructions. More speci fically, the constructions in question are all based on a single lexical representation o f the adjectives involved. (ii) The Positive of a relative adjective is interpreted with respect to a contextually determined class of comparison C. Within C, a standard, average, or norm NJC,Af is defined with respect to the property A speci fied by the adjective in question, so that, e.g., John is tall is interpreted roughly as john is taller than N1c,h,ighr( In the present paper, I will nor be concerned with rhe question how C and NJ C.Af are determined, but simply assume that N is available. (I will usually drop rhe index ( C, A J of N .) (iii) Relative adjectives assign ro an individual x a degree dA where d might be conceived as a class of individuals rhat are equivalent with respect to A . (This notion will be somewhat modi fied below.) Di ffering from all other approaches, I make rhe following assumptions: (iv) The lexical representation of a relational adjective is semantically a kind o f rhree-place predicate that relates an individual x , a standard comparison v, and a di fference c . With respect to their semantic type, both v and c are degrees, and rhe degree assigned to x is composed of rhe values of v and c .3 One o f the possible values o f v is N. (v) Comparative and Equative constructions are related to each other in roughly rhe following way: rhe complement clause of the Comparative specifies rhe value o f v, while that o f rhe Equarive specifies the value of c .• (vi) Relative adjectives belong to (ar least) rwo classes, which I will call dimensional adjectives

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In this paper, I will outline a theory of gradation that builds upon quite a number of previous analyses, preserving as far as possible rhe concepts thar have already been clari fied, but modifying rhe structure o f earlier proposals in crucial respects. The reason for adding a new theory to the ones already existing is rwofold:

8 Index of Abstracts

(tall, long , heavy etc.), and evaluative adjectives (clever, nice, good etc.). The degrees specified by D-adjectives are extents, the degrees specified by £-adjectives are grades.5 (vii) There is a small number of conditions on semantic representations that determine, among others, the value the standard of comparison v can assume in specified configurations. To conclude this preliminary outline, I should emphasize that more important than the list of individual points relating the present theory to or distinguishing it from other proposals is the general structure of the theory, which is different from its predecessors. This will become clear if we proceed.

BIERWISCH, MANFRED Tools and explanations of comparison (Part II).

6: 1 0 1 - 1 46

Although the term 'contrastive stress' has been incorrectly used in the past to describe default accent, contrastive intonation, or a combination of the rwo, English does have contrastive stress. However, contrastive stress is not used primarily to show a contrast. It is possible to have - coi1trastivenieaning wirh-6ut toiitfastive stress-and it is -also possib le-ro-have contrastive stress without contrastive meaning. The fact that English has contrastive stress constitutes furrher evidence for the existence of an unmarked or normal stress pattern.

BLOK, PETER I. Focus and presupposition. In this paper traditional approaches to the notion 'presupposition' are criticized. The relation berween the concepts 'topic' and 'presupposition' is discussed in a game-theoretical framework. It is shown that the concept presupposition has to be defined pragmatically with respect to its dialogical functions.

BLUTNER, REINHARD Dynantic generalized quantifiers and existential sentences in natural languages.

The central topic ro be discussed in this paper is rhe definiteness restriction in there -insertion contexts. Various attempts to explain this defi niteness restriction using the standard algebraic framework are discussed (Barwise & Cooper 1 98 1 ; Keenan 1 978; Milsark I 974; Higginbottham 1 987; Lappin 1 98 8) and the shortcomings of these attempts are demonstrated. Finally, a new approach to the interpretation of existential there he-sentences is developed within the framework of Groenendijk & StokhoCs ( 1 990) Dynamic Montague Grammar. This approach makes use of a variant of Partee's ( 1 986) 'rype-shifting'-operator BE and it overcomes the shortcomings of the rival analyses. The general conclusion is that Dynamic Montague Grammar has applications other than those which prompted it and advantages other than those Groenendijk & Srokhof claim for it.

BoLINGER, DWIGHT Where does intonation belong?

2: 1 0 1 - 1 20

Though intonation has many ties to the central arbitrary manifestations of human language­ to syntax, phonology, and to some extent lexicon-irs most intimate connecrions are with the general scheme of iconic nonverbal communication, particularly the now ;pontaneous, now simulated or ritualized, gestures of the face, head, hands and body. Its meanings are based on inferences from concepts of up and down-often associated with actual up-down movements

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

BING, jANET M. Contrastive stress, contrastive intonation, and contrastive meaning.

Index of A bsrraccs 9 in ocher pares of the gestural complex-plus metaphorical extensions of those concepts. Supposed grammaticizations are contradictory unless seen as intersections of two relatively autonomous systems: word-based language, and intonational and physical gesture. The study uses evidence from the intonation of English.

BoscH, PETER (ed.) Special issue on Anaphora.

1 :294-397

BoscH, PETER Words, worlds, and contexts.

3:26 1 -275

By Hans-Jiirgen Eikmeyer and Hannes Reiser Book review

By S. G. Pulman Book review

BoscH, PETER Pronouns under control? A reply to Liliane Tasmowski and Paul Verluyten.

Liliane Tasmowski and Paul Verluyten have recently expressed their misgivings about a proposal for a distinction between syntactically and referentially functioning anaphoric pronouns that was put forth in Bosch ( 1 980, 1 98 3 ) and have re-emphasized their ideas towards a uniform treatment of anaphoric pronouns, as originally published in Tasmowski and Verluyten ( 1 982). In the following pages I shall point out some limitations of the uniform pronoun treatment Tasmowski and Verluyten have in mind and I shall propose some amend­ ments and extensions of my earlier proposals in order to take phenomena of gender and num­ ber agreement of pronouns into account which were ignored in earlier versions.

BoscH, PETER and NooRDMAN, LEo G. M. (eds) Special issue on Semantics and Intonation. BREE, DAVID S. Counterfactuals and causality.

2: 1 00-2 1 7

There are three parts to chis paper. In the fi rst part the difference between hypothetical and counterfacrual conditionals is examined. Both Adams's argument that indicative and subjunctive conditionals differ in the degree to which they arejustified and Lewis's contention that counterfacruals differ from hypotheticals in that they fail to contrapose are both shown to be unfounded. Standard tests confirm Kartrunen's claim that the difference lies not in the truth conditions bur in the falsiry of the anrecedenr being presupposed and the truth value of the consequent being a conversational implicature. There is also a pragmatic difference: counterfaccuals are more difficult to verify. In the second part 4 solutions to the counterfactual problem are criticized partly in light of the differences found in the first part. Goldstick's extension of Goodman's classic solution is shown to be an incomplete algorithm. Lewis's possible world solution reduces the prob lem in finding a criterion by which to order possi ble worlds. Kratzer's solution, also in the possi ble worlds rradirion, requires a heuristic for partitioning propositions. It is only the older solution of Rescher and Simon, based on causal ordering, chat is adequate for causally based counrerfactuals with false consequents. In the last part of the paper Simon and Rescher's merhod is extended to accidental counterfacruals and counterfacruals with true consequents.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

BoscH, PETER Word meaning and belief.

10

Index of Abstracts

BREE, D. s. The durative temporal subordinating conjunctions since and

until .

The temporal subordinating conjunctions fall into two categories, durarive and non-durarive, depending on the length of time for which the main proposition is predicated to hold. Formally the two durative subordinating conjunctions in English, temporal since and until , are usually treated as though they were symmetric about the time of reference (henceforth the TOR). I examine this assumption from three points ofview: the time relationships between the main and subordinate propositions, the truth conditions of the two propositions, and the causal relationship that may be inferred between the states or events referred to by the main and sub propositions. The analysis is based on samples from rhe Brown Universiry corpus.

An event or state can be located on the time dimensions or given an extent of rime. This can be done by adding a temporal prepositional phrase or subordinate clause to a sentence. We give here an analysis of the different temporal prepositions and subordinate conjunctions rhar are found in English. This analysis has two parts. The first rakes the form of determining a set of rules for distinguishing the temporal use of such function words from other uses. The second gives rules ·for disringishing the different semantic meanings of r,hese words. These rules are then drawn together to produce a decision tree for selecting the appropriate function word, the function words appearing as the leaves of the tree. The English words are replaced by Dutch temporal function words, and the small differences between the two are noted. The criteria that are used to construct the selection trees provide the set of temporal Universal Linking Dimensions.

BREE, D. S. and VAN WERKHOVEN, J . P. Translating temporal prepositions between Dutch and English.

7: I-5 I

The decision trees for selecting rhe appropriate temporal conjunction or preposition in English and Dutch, developed in an earlier paper, are tested. Data are from the translations of parts of two books, one English and one Dutch, into Dutch and English respectively. The analysis of the data has led to the complete recasting of the original selection trees. The new trees are based on about a dozen different attributes that are needed to classify the normal, non-idiomatic use of almost all the temporal conj unctions and prepositions in both languages. Some of these attributes are: time point v. period; simultaneiry v. order; relative order of the matrix and sub events; whether or not the Time of Discourse is used to mark the end of a period, etc. Among other details we have been able to show how the cot� unction AS, in its temporal use, is not ambiguous between WHEN and WHILE as was originally thought. There is a large overlap between the selection trees for both languages, as well as some interesting differences. These trees have been specified in sufficient detail to be readily incorporated in natural language computer programs.

BROWN, GILLIAN Discourse markers. By Deborah Schiffrin Book re11iew

BROWN, GILLIAN and GARROD, SIMON C. (eds.) Special issue on practical aspects of semantics. CALIS, GE Intentionality. By John Searle Book re11iew

7: I-92

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

BREE, D. S. and SMIT, R. A. Temporal relations.

Index of A bstracts 1 1

CARTER, DAVID M. Common sense inference in a focus-guided anaphor resolver.

CARTER, DAVID M. Control issues in anaphor resolution. Much research in computational linguistics has concentrated on treating the individual phenomena of natural language rather than on how these treatments can be made to work together. This paper discusses four respects in which processes embodying individual treatments must be made to co-operate with one another i f anaphoric expressions are to be interpreted correctly. These are the co-ordination o f ( 1 ) treatments o f anaphoric and non­ anaphoric amb iguity in a sentence; (2) treatments of d ifferent kinds of knowledge relevant to interpretation; ( 3 ) the resolution o f several anaphors in one sentence; and (4) candidate referents arising from several sources. It is argued that a fl exible control structure based on numerical scoring allows the required co-operation to take place, whereas a more limited depth-first architecture seems not to allow this. The discussion is grounded in comparisons berween rwo very di fferent implemented systems that resolve anaphors. The fi rst, SPAR, is an experimental system embodying the control structure argued for. The second, the SRI Core Language Engine, is a wide-coverage system intended for practical applications. Its control structure is, at the time o f writing, depth-first, but planned enhancements will al low the problems discussed in the paper to be tackled, resulting in the development o f a powerful and widely applica b le natural processor.

CoATEs, JENNIFER Modal meaning: The semantic-pragmatic interface.

7 : 5 3-63

The spontaneous talk produced by real people in natural situations is an enormous challenge to any theory o f meaning. It is at the same time more complex (in terms o f the multiple conversational implicatures potentially borne by a single statement) and more simple (in terms, for example, of syntax and the use o f ellipsis) than the examples that regularly appear in textbooks. In order to say what such talk 'means', we need to draw on both semantic and pragmatic theories of meaning.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

This paper describes an invesrigation of the feasi b ility o f resolving anaphors in natural language texts by means of a 'shallow processing' approach which exploits knowledge of syntax, semantics and focal focussing as heavily as possible; it does not rely on the presence of large amounts o f world or domain knowledge, which are notoriously hard to process accurately. The ideas reponed are implemented in a program called SPAR (Shallow Processing Anaphor Resolver), which resolves anaphoric amb iguities in simple English stories and generates sentence-by-sentence paraphrases that show what interpretations have b een selected. To resolve anaphora, SPAR combines and develops several existing techniques, most notably Sidner's theory o f local focussing and Wilks' 'preference semantics' theory of semantics and common sense inference. Consideration of the need to resolve several anaphors in the same sentence results in Sidner's framework being modified and extended to allow focus-based processing to interact more flexi bly with processing based on other types of knowledge. Wilks' treatment of common sense inference is extended to incorporate a wider range o f types of inference without jeopardizing its uniformity and simplicty. In the absence o f large quantities of world knowledge, successful anaphor resolution is seen to depend on the coordination of predictions made by system components exploiting various knowledge sources. Such coordination normally allows anaphors to be resolved correctly even when no single source of predictions is sufficient on its own.

I 2 Index of A bstracts

COMRIE, BERNARD Reflections on subject and object control.

A recurrent problem in linguistic theory has b een trying to provide a principled b asis for the distinction between su bject control and o bject control verbs, where by 'su bject control verb' is understood a main clause verb that requires coreferente b etween its subject and the understood subject o f a dependent infinitive (e.g. I tried to leave , I promised you to leave), and by 'object control verb' a main clause verb that requires coreference between its object and the understood su bject of a dependent infinitive (e.g. I persuaded you to leave). A numb er of possible solutions are examined, in particular a purely formal principle (the Minimal Distance Principle) and a pragmatic principle rooted in Searle's theory o f speech acts. It is concluded that any explanatory account o f the subject/object control distinction must b e grounded in the pragmatics of speech acts, although in at least some languages, including English, these pragmatic principles have become grammaticalized.

CoRNISH, FRANcis s:2 3 3-26o Anaphoric pronouns: Under linguistic control or signalling particular discourse representations? A contribution to the debate between Peter Bosch, and Liliane Tasmowski and Paul Verluyten.

The article is a contrib ution to the deb ate between Tasmowski & Verluyten ( 1 982, 1 9 8 5) and Bosch ( 1 983, 1 984, 1 987) as to how the form as well as the interpretation o f anaphoric pronouns is determined. TV rightly criticize B's tests as to whether a particular third-person pronoun is functioning semantico-syntactically or referential-anaphorically; however, their examples and arguments do not warrant the conclusion that there is no substantive distinction to be drawn between the two types o f pronoun use. Many o f TV's examples in this connection merit further analysis, which leads to very different conclusions from the ones they arrive at. There is not a single dichotomy between two types o f pronoun use, but a cline, the crucial factor di fferentiating each position on the cline being the degree to which the pronoun's discourse referent or its intension is presupposed by the speaker. In section 2, I argue that the traditional notion 'antecedent', as espoused by TV, should be a bandoned, and that it is in terms o f the discourse model representation by means of which each discourse referent is encoded in the discourse model that anaphoric pronouns refer.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In this paper, I shall take the position that semantics is concerned with sen ten ce -meaning, and that pragmatics is concerned with uttera nce -meaning (see Leech 1 983; Levinson 198 3). A sentence is an abstract category defined within a theory of grammar; an utterance is a chunk of real live talk (which may or may not be a syntactic sentence). A semantic analysis o f the sentence it's cold in here would be concerned with its 'literal' or 'referential' meaning-in other words, the sentence would be taken as expressing the speaker's assessment of the temperature at the moment o f speaking in the location where the sentence was uttered. As such, the sentence can be labelled as true or false, depending on the actual state of the weather at that time and in that place. It is pragmatics, however, which enab les us to explain why the use of this sentence as an Utterance in a particular context can result in the addressee switching on the gas fire. A further characteristic o f sentence-based theories of meaning is their orientation to an isolated speaker. In fact, most branches o f linguistics have until recently focused on speakers in isolation, and have treated discourse as a sollipsistic product. Bur talk by its very nature normally requires more than one speaker. The dyadic nature of spoken interaction needs to be emphasized: meanings are created joinrly by speaker and hearer. Talk is dialogue, not a sequence of monologues.

Index o f Ab stracts 1 3 Finally, i n section 3 , the role of the 'agreement' of anaphoric pronouns in gender and number is examined , and the conclusion is d rawn that this is not a necessary cond ition for pronominal anaphora. Referenrial-anaphoric pronouns are relatively independent indexical expressions, and their gend er and number features may b e manipulated by the speaker to achieve a variery o f rypes o f reference to a particular d iscourse referent. Suggestions as to fruitful areas for furure research in rhe fiel d of pronominal anaphora are d erived from rhe foregoing d iscussion.

CoRNISH, FRANcis So be it : The discourse-semantic roles of so

and it .

DE CORNULIER, BENOIT Detachment is not the same as 'meaning detachment'. A reaction to the review by R. Hausser and C. Gerstner. CRESSWELL, MAX J. Comments on Von Stechow. CROFT, WILLIAM A conceptual framework for grammatical categories, or: A taxonomy for propositional acts.

The aim o f rhis paper is to provide a general conceptual framework that will accommodate all o f the non-protorypical lexical items and infl ectional morphemes, in addition to the protorype categories. In particular, broad parallels are d rawn between 'function' word s and inflections associated with nouns and those associated with verbs (and , to a lesser extent, adjectives). The purpose of this paper, then, is not to provid e a d irect account of linguistic structure and behavior, bur instead to provide an analysis of linguistic function that can in turn b e used to explain linguistic structure and behavior. This analysis has benefited from a large amount of prior research on the semantics o f grammatical categories, and on prior attempts to provid e an overall conceptual organization (e.g. Talmy 1 978, 1 988; Morrow 1 986; Jackend o ff 1 98 3), although it d i ffers in significant respects from their goals and results. It should go without saying that rhe framework sketched here is tentative. Nevertheless, irs purpose is ro provi d e a starring-point for the construction o f a conceptual framework that is su fficiently rich, comprehensive, and well d efined to use for semantic and pragmatic analyses o f grammatical phenomena.

CUTLER, ANNE c£ Levelt, Will em J. M. and Cutler, Anne.

DAHL, O STEN Some comments on 'The society of mind'.

Mu ltiple review o f 'The sociery o f mind ' by Marvin Minsky.

2:205-2 1 7

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

The aim of rhe article is to d etermine whether so and it fulfil distinct d iscourse-build ing roles, in their predicate- and proposirion-anaphoric guise. Is rhe choice berween them d etermined by the syntactic or semantic nature o f their 'antecedent', or of rhe context in which the choice berween them is to be mad e? Or d o they in fact themselves determine their 'anteced ent' (or at least superimpose upon it a characteristically d istinctive interpretation)? I shall argue for the latter analysis, indicating a range o f factors (of a mainly semantic nature) which co-d etermine both the choice berween these rwo anaphors, and their full utterance-level interpretation. A wid e range o f arrested and constructed examples will be presented as the basis for the discussion.

14

Index o f Ab stracts

DAHL, OsTEN The role of deduction rules in semantics.

6: 1 - 1 8

DAHL, O sTEN The-development ofEnglish aspectual-system. By Laurel J. Brinton Book review

VAN DEEMTER, KEES Forward references in natural language.

This paper deals with forward references (also called kataphoric references) in natural language. In order to calculate truth conditions for sentences that involve kataphoric references, an extension o f Discourse Representation Theory, PATIENT DRT, is proposed, inspired by so­ called b ackpatching techniques for the parsing of programming languages. The main idea is that a kataphoric element introduces an incomplete discourse entity, to b e completed by subsequent material under certain conditions. This approach is applicable to pronominal as well as complex Noun Phrases, and has no special difficulties with crossing co-references. The main virtue of this approach is that it allows parsing of kataphors from left to right, which makes it suita ble for on-line language processing by ccimpu rer and plausible as an element of a theory o f human language processing as well. However, the approach suggests that a left-to­ right treatment o f kataphoric constructions is hard to reconcile with the requirements of compositionality.

vAN DEEMTER, KEES Towards a general theory of anaphora. This paper deals with anaphoric properties of both pronominal and nonpronominal Noun Phrases within the framework of Discourse Representation Theory (DRT). A generalized notion o f anaphora is advocated, in which the notion of anaphora is extended to cover relations between anaphor and antecedent other than referential identity. This paper tries to make insights gained in DRT, as well as in other theories o f anaphora, applicable to a wide range of 'new' phenomena in the area of context dependent interpretation.

DELIN,JUDY Properties of it-cleft presupposition.

9:28<)-306

It is generally accepted that it -clefts convey logical presu ppositions. I n this paper, I examine the properties of those presnppositions with a view to shedding some light on what function they serve in discourse. First of all, an examination of naturally occurring data shows that

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

The distinction between 'partial' and 'total' interpretations (models) is d iscussed and related to the d istinction between proof-theoretical and model-theoretical treatments of logic. It is claimed that there is a parallel between the construction o f a proof based on a set o f premises and e.g. the production o f a natural-language text which is based on information in some kind o f data-b ase. The main part of the paper is devoted to a discussion of the relation between the deduction rules traditionally associated with the existential quantifier and notions pertaining to the theory of reference such as specificity and referentiality/attributivity. Two types of specificity are distinguished, which can be connected with 'Existential Elimination' and 'Existential Introduction', respectively. A distinction is further made between trivial and non­ trivial 'Existential Introduction', where only the latter kind involves erasure of 'coreference links.' I t is argued that an analogous treatment of the referential-attributive distinction is a way of making sense of Donnellan's suggestion that the latter may depend on the description's role in an argument. Finally, the notions of 'external anchoring' and 'stability of individual concepts' are related to the distinctions made earlier in the paper.

Index o f A bstracts 1 s presuppositions of it-clefts are not normally composed of information that is already entailed by the context: they are frequently used to communicate wholly or partly new in formation. In the main part o f the paper, I present an explanation of the function of it-cleft presupposition that is applicable to all defts regardless o f their information structure. The account appeals to the current notion o f presuppositions as anaphoric environments, motivating this view further with empirical evidence for anaphoricity. I turn first of all the Prince's ( 1 978) observation that it-cleft presuppositions mark information AS KNOWN FACT in the discourse. This o bservation, while useful, is not itsel f an explanation, since further factors can be shown to underlie the effect. First, I demonstrate that it -cleft presuppositions mark information as ANAPHORIC. Such marking is independent o f information structure, and has observable linguistic effects. The empirical ev�dence for the anaphoricity o f cleft presuppositions is of three types. 1.

3·

Arising out o f this anaphoricity is a second factor; presupposed information is in general I suggest that non-negotiability arises from anaphoricity because anaphora implies the existence o f prior reference to the same information. Participants in a discourse are, with each utterance, placing propositions 'on the table' for acceptance or rej ection by interlocutors. I f a proposition is placed on the table along with a marking to say that this is not the proposition's original appearance the speaker is indicating that the time for any negotiation-or, more speci fically, any rejection-is past. This gives rise to the 'known fact' effect o bserved by Prince. NON-NEGOTIABLE.

DENKEL, AR_DA The meaning of an utterance.

The fi rst target of the paper is to demonstrate that the Gricean explanation of the concept o f an utterance's occasion meaning by proposing an equivalence between what a speaker means by X on an occasion and what X means on the same occasion cannot be correct. An outline­ account o f utterance meaning that carefully avoids explaining this concept purely in terms of the speaker's intentions or purely in terms of the hearer's understanding is then developed. It is concluded that what determines the meaning o f an utterance (as well as that which a speaker can mean by the same utterance) is above all an objective connection of natural or conventional character between an aspect of the utterance and the state, attitude or fact about which something is meant by the speaker using the utterance.

ECKERT, H. Valences Ltd vs Valences Associated. Comments on Heringer's association experiment as a basis for valence theory.

Heringer claims that th� value o f existing theories of valence is limited as they have failed to give a clear account o f the crucial distinction between complements and supplements. He maintains that associations between verbs and question words can serve as a basis for valence theory. The results o f his association experiment, however, do not permit us to infer dependency relations, to distinguish clearly between optional and ob ligatory elements, to speci fY quantitative valence, or to distinguish between elements that are grammatically and semantical ly implied by the verb as opposed to merely contextual elements. I should also like

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

2.

Elements that are ambiguous between an anaphoric and an emphatic use take on their anaphoric reading when placed within an it-cleft presupposition; It-cleft presuppositions enable the anaphoric relation upon which contrast depends to be esta blished, in contexts where information that is simply given does not have the same effect; and Information placed within an it -cleft presupposition appears to remind rather than inform , regardless o f its objective status in the discourse.

1 6 Index of Ab stracts to argue th at the questions that are said to impose themselves upon the speakers need not necessarily do so because of the semantic power of the verb, and that the values for certain question words are partly influenced by the test meth od, wh ere each question word a ffects the values o f the subsequent ones. I feel th at while the association experiment yields supporting evidence for valences in a number of cases it cannot claim to function as a basis for valence theory.

VAN EIJCK, JAN c£ van Benthem,Johan and van Eijck,Jan.

VAN EIJCK,JAN The dynamics of description.

10:239-267

EscHENBACH, CAROLA Semantics of number.

10: 1 -J I

This paper presents an analysis of how number can be represented in a logical framework b ased on a semi-lattice universe. The features singular and plural of count nouns are treated in a uni form way, assuming that the meaning of nouns s hould generally be represented as independent o f number. This opposes the assumption, quite common in the current discussion o f plural. that plural should be analysed as an operator on the meaning of th e singular form o f count nouns. Based on a discussion o f relational nouns as son , sister, and owner it is shown that the predicate-operator approach is not able to handle the number phenomena systematically. In addition, an analysis of some determiners and quantifiers is presented in th e proposed framework to illustrate how to cope with their systematic interaction and restrictions with respect to number.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In a static approach to th e semantics o f natural language, the use of definite and indefinite descriptions encounters certain difficulties. Th is paper shows that these problems can to a large extent b e overcome by switching to a dynamic perspective, and demonstrates how description operators acquire a new lustre and attractiveness in a dynamic set-up. The starting point of this paper is dynamic predicate logic. The paper first extends the semantics o f dynamic predicate logic (Groenendijk & Stokhof 1 99 1 ) with a clause.for definite t assignment. The constructs for indefinite and definite assignment (I'J and t ) allow a very straigh tforward analysis o f indefinite and definite descriptions in natural language. It is shown how th e standard semantics for t assignment (van Eijck & de Vries 1 992) leads to a Russellian treatment of definite descriptions. A Hoare/Pratt-sryle calculus of assertions for dynamic predicate logic is presented, and it is demonstrated how the axiom schemata of th e calculus allow for the calculation o f success conditions (static truth conditions) or, equivalently, for the calculation offailure conditions (static falsity conditions) o f dynamic predicate logic programs. Next, th e dyn�mic semantics is enriched with error states, intended ro monitor failure of uniqueness presuppositions for definite descriptions. The semantic clause for t assignment can now take the presuppositions of the use of definite descriptions into account, whi ch gets us a Strawsonian treatment o f definites. It is indicated how a Hoare/Prarr-sryle calculus for the state semantics can be set up. Th e paper ends with a demonstration o f the use of th is calculus for finding success conditions, failure conditions, and error conditions (conditions for presupposition failure). It is also shown that the axiom system may serve as a calculus o f presupposition projection.

Index of Abstracts

EssER, JuRGEN Tone units in Functional Sentence Perspective.

17

2: 1 2 1 - 1 39

The phonological structure of the tone unit in terms of only one tonic element per tone unit is discussed in ( I . I ) and related to Brazil's theory of proclaiming and referring tones. It is argued chat certain claims ofBrazil's theory are too strong ( 1 .2). When describing discourse functions it is necessary to recognize besides the 'given'/'new' dichotomy, a distinction between 'foregroundworchy' and 'less foregroundworthy' elements (2.I ). This leads me to formulate two kinds of information presentation: rising and falling communication (2.2). It is also argued that there is a need for a distinction between a semantically and a formally defined theme. This makes it possible to incorporate the Prague notions of objective and subjective order of theme and rheme into the theory (2.3). Finally, it is shown that the concept of rising and falling communication can be used to differentiate different reading and text sryles (2.4).

By Joachim Jacobs Book review

FooLEN, An Sprache und Handlung. By Gunter Saile Book review

fOOLEN, An Focus, coherence, and emphasis.

5=79-86

By Paul Werth Book review

fRAURUD, KARl Definiteness and the processing ofNPs in natural discourse.

7= 395-4 3 3

Definiteness i s commonly seen as the watershed between chose noun phrases (NPs) that introduce new referents and those that refer to referents already familiar. Furthermore, for definite NPs, the anaphoric use is taken to be the paradigm case, while ocher, so-called first­ mention uses are regarded as secondary. The aim of the present paper is to challenge this view, and to argue for a more complex picture of the role of definiteness in the processing of NPs. The paper consists of two parts. The first part presents a corpus-based study of the co­ referential properties of definite and indefinite NPs in natural, unrestricted texts. The data bring into light several issues with regard to co-referentialiry in unrestricted discourse and the possible referential functions of indefinite and definite NPs. Particular attention is drawn to the fact that the most common function of definite NPs is not anaphoric b ut different rypes of first-mention uses. This is the point of departure for the second part of the paper, in which three different approaches to first-mention definites are discussed, and some preliminaries to an alternative model of the processing of first-mention definite NPs are presented.

fRrEDRICHSDORF, ULF and SCHROEDER-HEISTER, PETER A companion to modal logic. By G. E. Hughes and M. J. Cresswell Book review

GARCIA, ERICA c. A psycho-linguistic crossroads: Frequency of use.

The units postulated in linguistic analysis differ not only as to their structural properties, but also in their relative frequency of use in different contexts: the former should, ideally, shed light on the latter. Such a connection between linguistic analysis of the system and the use to which it is put requires an understanding of what constitutes 'appropriate' use of a form, and

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

fOOLEN, An Fokus und Skalen.

r

8 Index o f Abstracts

an appeal to psychologically plausib le cognitive processes. The need to make psychological sense o f linguistic frequency data is discussed on the b asis o f distrib utional skewings in the exploitation o f stronger vs. weaker deictic devices in Spanish, Portuguese and Dutch, with special attention for the problem posed by competing strategies.

GARNHAM, ALAN cf. Perner, Josef and Garnham, Alan.

GARNHAM, ALAN and 0AKHILL, JANE 7:379-393 Mental models as contexts for interpreting texts: Implications from studies of anaphora.

GARROD, SIMON c.

7= I -92

cf. Brown, Gillian and Garrod, Simon C. (eds).

GARROD, SIMON c. cf. San ford, Anthony,]., Garrod, Simon C., Lucas, Angela, and Henderson, R.

I :2 I -4 I GARROD, SIMON c. AND SANFORD, ANTHONY J The mental representation of discourse in a focussed memory system: Implications for the interpretation of anaphoric noun phrases. .

To a cognitive psychologist discourse comprehension poses a number of interesting problems both in terms of mental representation and mental operations. In this paper we suggest that certain of these problems can be brought into clear focus by employing a procedural approach to discourse description. In line with this approach a general framework for the mental representation of discourse is discussed in which distinctions berween di fferent types o f memory partitions are proposed. It is argued that one needs to distinguish both berween focussed representations available in immediate working memory and nonfocussed represen­ tations availab le in long-term memory and also berween representations arising from the asserted information in the discourse and those arising from what is presupposed by it. In the second hal f o f the paper a particular prob lem o f anaphoric reference is discussed within the context o f this framework. A general memory search procedure is outlined which contains three parameters for determining the search operation. We then attempt to descri be certain anaphoric expressions such as personal pronouns and full defi nite noun phrases in terms of the execution of this search procedure, where distinctions arise from the parameter speci fication derived from the expressions.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

One of the major tenets of the mental models theory of text comprehension Is that the model o f the text so far provides (part oD the context for understanding the current sentence. Using rwo sets of findings on the comprehension o f anaphoric expressions, we attempt to provide a more specific interpretation for this statement. We first consider the linguistic distinction between deep and surface anaphors, and the proposal that they are interpreted with reference to mental models and to representations of surface form, respectively. Although the linguistic distinction is reflected fairly directly in considered judgements, in on-line processing both aspects of representation are implicated in the interpretation of.both kinds of anaphora. The second set of findings shows that the interpretation of texts containing pronouns can b e incomplete-only part of the information in the model is used to interpret the anaphor. Readers may effect mappings berween role fillers in different clauses o f a text or they may effect mappings b erween names and role fi l lers. We discuss evidence that these rwo types o f mapping can be carried out separately and that, in certain circumstances, role-to-name mapping in particular may not take place at all.

Index o f Abstracts 1 9

GARROD, SIMON C. and SANFORD, ANTHONY J. Discourse models as interfaces between language and the world.

6: I 4 7- I 6o

GEERAERTS, DIRK On necessary and sufficient conditions.

5=275-29 1

This paper outlines an argument that the meaning of spatial term depends critically upon our mental models o f space. We argue that such models caprure the functional geometry o f spatial scenes to represent various control relations between the objects in the scene. The discussion centres around two analyses. First, an analysis of the spatial descriptions taken from task oriented dialogue, which seem to reflect a number of distinct mental models of the same visual scene, and secondly an analysis o f simple English spatial prepositions. We argue that these prepositions express control relations rather than simple spatial relations and depend for their interpretation on the model o f space assumed by speaker and listener. This analysis suggests that mental models should be seen as interfaces between the language and the world o f discourse rather than simply surrogates for that world.

It is often said that prototypically structured concepts are a counterexample with regard to the classical view o f categorization, viz. that categories always have a single definition in terms o f necessary and su fficient conditions. However, such statements about the nature o f prototypicality do not take into account the distinction between vagueness and amb iguity. In particular, they are never accompanied by a speci fication o f the criteria according to which the allegedly prototypical concepts are indeed a single semantic category. An exploration of ti.e distinction between analytic and introspective criteria for am biguity shows that proto­ typicality is part o f a larger class of phenomena exhibiting discrepancies between both approaches. Specifically, prototypicality and autohyponymy together make clear that analytic distinctness of meaning is neither a necessary nor a su fficient condition for introspectively recognized ambiguity, such as it determines the truth conditions of utterances.

GERSTNER, CLAUDIA cf. Hausser, Roland R. and Gerstner, Claudia.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

The cognitive psychology o f discourse i s concerned with the narure of the mental processes entailed in understanding what is written or spoken, and the prob lem o f how these processes might be realised in the mind o f the understander given the psychological constraints of limited attention and memory which we know to obtain. One very attractive line of attack is to view the many and various aspects of a discourse as having an instructional component, in the sense that the reader or listener is being instructed to assemble representations o f the elements of discourse in a particular way. An example o f this is to be found in a treatment of topic marking within the topic/comment distinction (Halliday, 1 976): topic identification may be thought of as instruction to implement a procedure in which the topic content is construed as an address in memory to which new (comment) information is to be affixed (e.g. Broadb ent, 1973; Haviland & Clark, 1 974). While any attempt at producing a process-model for comprehension inevitably makes use of such a procedural view, it is also sensible to consider a text as having a content, which is more directly interpretable as a set o f statements. In the present paper, we shall first consider the question o f text content. This immediately raises the problem of how to treat anaphoric reference, which is one of the key contrib utors to text cohesion. Finally, we shall attempt to illustrate how the instructional or procedural aspect of discourse interacts with the content aspect by reference to a speci fic problem o f anaphoric reference.

20

Index o f Abstracts

GEURTS, BART Generics.

There is a fairly general tendency to analyse generic statements as referring to the entities that are commonly identifi ed as 'kinds'. Apart from the fact that, rhus far, none of these analyses have proven to be satisfactory, there always remains the problem of explaining what a kind is. I propose, instead, to do away with kinds altogether, and to regard generics as expressing stereotypical assumptions. Although, at the moment, I have just the broad ourlines of a theory to offer, this approach seems to fi t the available data quite well. Moreover, it obviates the need to assume that generic descriptions are referential expressions.

GEURTS, BART Memory and context for language interpretation. GuNTHER, Uno

7=3 2 1 -345

cf. Sichelschmidt, Lorenz and Gunther, Udo.

GussENHOVEN, CARLOS three'-dimensional·scaling·of nine-English tones.

A

Dissimilarity judgements were obtained from 4 4 naive native speakers of British English on the semantics of five carrier-sentences. The aim of the experiment was to test the hypothesis that the nine tones form a linguistic paradigm of three sets of three tones, each o f which set represents a semantic continuum from 'special' to 'routine'. A three-dimensional scaling analysis was carried out on the data to see to what extent the con figuration predicted on the b asis o f the hypothesis marched the one actually o btained. The similarity turned out to be very satisfactory for the tones in the sets 'fall' and 'fall-rise', but poor in the set 'rise'. The results are reponed in such a way as to enable other researchers to test altcmative hypotheses concerning the relationship between the tones.

HABEL, CHRISTOPHER and RICKHEIT, GERT (eds) 7:32 1-43 3 Special Issue on referential and anaphoric processes in text comprehension. HAUENSCHILD, CHRISTA Definite and indefinite interpretation ofRussian noun phrases.

This paper aims at an illustration of a special type o f linguistic rule that has been developed for a text understanding model within a project of (partial) simulation o f the processes of understanding and translating natural-language texts. The rules in question are to reflect di fferent degrees of probability with respect to certain interpretation hypotheses on the basis o f the interaction of di fferent types of linguistic as well as non-linguistic in formation relevant to a given interpretation hypothesis. This type of relative rule for (partial) interpretations is exemplified with a paradigm problem of machine translation: the definite vs. indefi nite interpretation of Russian noun phrases, which normally do not contain any locally unambiguous expressions of defi niteness or indefiniteness (as e.g. articles are). Given the enormous complexity of the whole problem of interpreting such noun phrases (all kinds of linguistic and extra-linguistic criteria may have to be considered). I concentrate on the case of couples of noun phrases with identical heads. The interpretation hypothesis to be evaluated is the following: the two noun phrases in question are identical in reference, and thus the second is defi nite. The main task consists in determining the factors that yield relative positive or negative evidence for the given hypothesis. After a shorr discussion o f the underlying text understanding model and of the defi nitions

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

By Hiyan Alshawi Book review

Index of A bstracts 2 1 o f the basic concepts (anaphoricity, identity o f reference, definiteness/indefi niteness), sentential and textual themeness, I give the list o f single factors that are relevant for the interpretation task in question. Then the corresponding single-factor rules are combined in a glob al evaluation rule, the functioning o f which is illustrated with the aid o f a simple example. Finally, I discuss the merits and shortcomings o f the proposed format for complex evaluation rules as well as the potential role of such rules within a machine translation sysrem.

HAussER, RoLAND On vagueness.

HAussER, RoLAND R. and GERSTNER, CLAUDIA Meaning detachment.

2:3 50-3 54

By Benoit de Cornulier Book review

HEIM, IRENE Presupposition projection and the semantics of attitude verbs.

Karrtunen observed that, i f the complement of an attitude sentence presupposes p, then that sentence as a whole presupposes that the attitude-holder believes p. I attempt to derive some representative instances of this generalization from suitab le assumptions about the lexical semantics o f attitude predicates. The enterprise is carried out in a framework o f context change semantics, which incorporates Stalnaker's suggestion that presupposition projection results from the stepwise fashion in which information is updated in response to complex utterances. The empirical focus is on predicates of desire and on the contribution o f counterfactual mood.

HELLAN, LARS Note on some issues raised by Von Stechow. HENDERSON, R. c£ Sanford, Anthony,].; Garrod, Simon C.; Lucas, Angela and Henderson R.

HENGEVELD, KEES illocution, mood and modality in a functional grammar of Spanish.

In order to be able to account for the alternating and non-alternating uses o f mood in Spanish this paper explores the field of illocution and modality and argues for rwo elaborations of the Functional Grammar framework: (i) a representation of main clauses which distinguishes berween several layers, each representing a di fferent subact of the speech act, and (ii) a rep­ resentation o f noun clauses which distinguishes berween non-factive, factive, and semi factive complements.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

It seems to be a foregone conclusion that natural language meanings are vague. Much depends, however, on the way meaning is analyzed. For example, should vagueness of meaning be treated in terms of the truth- or denotation-conditions o f expressions? Rather than proposing yet another 'fuzzy' or multi-valued logic, the present paper investigates the nature o f riference and truth . We consider rwo possible interpretations of the formal model structures used in formal semantics. One is called the paradigm I approach, according to which the model structure is interpreted as a representation of reality (such that the speaker/hearer is part of the model structure). The other is called the paradigm II approach, according to which the model­ structure is interpreted as a representation of conceptual meaning structures (such char the model structure is parr o f the speaker/hearer). It is shown char the theoretical nature of vagueness is totally different in the rwo paradigms. In conclusion, a number o f standard examples o f vagueness are analyzed within the paradigm II approach, including the so-called Sorites paradox or paradox of the heap.

.22 Index of Abstracts

4:79-99 HERINGER, HANS JuRGEN The verb and its semantic power: Associations as a basis for valence theory.

Valence theory has been syntactically oriented; the fundamental distinction between complements and supplements has remained without justification. It is shown in this paper that (i) valence theory can be founded semantically and (ii) that the distinction between complements and supplements is a semantic, relational, and gradual distinction. These are conclusions from an association experiment we are reporting which gives the distance o f ques­ tion words from verbs by considering frequency, latency, and rank o f mention.

HERWEG, MICHAEL A critical examination of two classical approaches to aspect.

HEYER, GERHARD Semantics and knowledge representation in the analysis of generic descriptions.

In what follows I first want to clari fy the notion ofgeneric reference on a pre-theoretic level by collecting some of those paradigmatic cases as they have been proposed in the literature, and by mapping out some o f th eir systematic interrelations. Th is will lead to a characterisation of desiderata for a theory of generic reference. I shall then sketch a semantics of generic descriptions along the lines of Heyter (1985 and 1 987), and show how th is approach can be accommodated into a frame-based extension o f standard Discourse Representation Theory. Finally, the proposed semantics of generic descriptions is related to their use as difault rules in commonsense reasoning.

HIELSCHER, MARTINA and MussELER, JocHEN 7:347-364 Anaphoric resolution of singular and plural pronouns: The reference to persons being introduced by different co-ordinating structures.

For the resolution of plural pronouns referring to singularly introduced reference persons the plural antecedent has to be built up by the cognitive system itself (installing a plural complex, e.g. 'john wanted to have a picnic with Mary. They had . . '). For singular pronouns the antecedent is usually mentioned in the text explicitly. This contribution examined wh ich aspects of the pre­ pronominal sentence structure determine the installation of a plural antecedent and at which point o f time this process is initiated. Using the German pronoun 'sie ', which is ambiguous in respect to number, it was shown in a first experiment that subjects have a preference to continue a text by referring to both singularly introduced persons, i f they are combined by the conjunctions 'and', 'as well as' or 'neither/nor', or by the preposition 'with', i f the female person is in the verb phrase. Subjects prefer to refer to the female person only after the prepositions 'without' and 'instead of', and a fter 'with' i f the female person is in the noun phrase. .

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Th is paper discusses the merits and sh ortcomings of the two leading paradigms in the model­ theoretic account o f the perfective and the imperfective aspect: the proposition-based approach in the tradition of Tense Logic, and the eventuality-based approach in th e tradition o f Donald Davidson's semantics for event expressions. It is shown that neither approach may claim general validiry for their respective format o f analysis, as their theoretical means suit one particular aspect only: those of th e proposition-based approach are confined to the imperfective aspect, and those of the eventuality-based approach are confined to the perfective aspect. Contrary to what is suggested by their advocates, neither format of analysis can be generalized to account for the other aspect. Rather, it is imperative to have a theory which integrates the two complementary approaches to one unifi ed account. The basic features of such an integrated theory are outlined in the final part of the paper.

Index o f Abstracts

23

The reaction rime data o f the second experiment indicated that a t least these conjunctions and 'with' initiate a plural reference complex before a pronoun is read. This pre-pronominal installation of a complex serves to facilitate plural reference operations executed at a later point in processing. In our view, pronominal resolution is more than a mere recursive search-and­ match procedure initiated by reading the pronoun; the cognitive system i\s better prepared for processing further referential relations. This view is discussed in the context o f a 'pronominal occupation' hypothesis: ·

HINTIKKA, JAAKKO and KomAs,JACK 1 : 3 87-397 Russell vindicated: Towards a general theory of definite descriptions.

HIRST, DANIEL Interpreting intonation: a modular approach.

2: I 7 I - I 8 I

Intonation provides an apparent counter-example to the claim made by proponents of the Extended Standard Theory of generative grammar that there is no direct interaction between phonology and semantics. ln the light o f recent work on phonological representations, a phonological analysis o f intonation is proposed which breaks down an intonation contour into two component parts: the phonological structure and the underlying tone sequence. It is suggested that while the phonological structure is partly determined by the syntactic structure, the ronal sequence is assigned freely in the phonology. A further possible contribution to the intonation is the existence of tonal morphemes in the form o f floating tones. lt is argued that while the description of English intonation is simplified i f we assume that there is a tonal emphatic morpheme, a similar analysis for 'interrogative' intonation cannot be correct. It is suggested fi nally that the claim that phonology and semantics do not directly i nteract, rather than being disproved by the facts o f intonation, provides an essential clue to the composite syntactic/semantic/pragmatic nature of intonation meaning.

HOEKSEMA,jACK To be continued: The story of the comparative.

3:93- 1 07

Some aspects o f Arnim von Stechow's analysis of comparatives (this volume) are criticised, especially his failure to make a semantic distinction between oblique and sentential comparatives. Several arguments in favour of this distinction are provided and some prima facie di fficu l ties considered. In addition, von Stechow's analysis of the ambiguity o f comparison in intensional contexts is critically examined; a quantifying-in solution is argued to be inferior to an operator approach.

HoEKSEMA, JACK The semantics of non-Boolean and.

6: 1 9-40

The meaning of 'and' in noun phrase conjunctions differs from its ordinary Boolean interpretation in other cases o f conjunction, such as sentential and predicate conjunction.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Russell's so-called theory o f definite descriptions' has been called 'the paradigm of philosophy' (Ramsey 1 9 3 1 : 263). Its historical significance is due to its being a part of Russell's largely successful campaign to reduce many different kinds of discourse to what in our days would be called the language of first-order logic.2 ln view o f the overall success of the Russellian program, it is ironic that its paradigm case, Russell's theory o f definite descriptions, has come under heavy criticism, and that its applicabiliry looks quite limited. It seems to us that rime is ripe to rum the tables on Russell's critics. This paper is a first step towards a general theory o f definite descriptions which borrows its inspiration from Russell's theory, even though the conceptual tools it uses go beyond Russell's methods. Because o f this spirirual affinity, whatever success our theory may enjoy serves also as a partial vindication of Russell's original theory.

24 Index o f Abstracts More precisely, this is the case when the noun ph rases conjoined are referring terms. A regular Boolean interpretation is still possible whenever two or more quantificational NPs are conjoined. Disjunction is always a Boolean operation. A semantics based on the notion o f set formation is provided to deal with conjunctions o f referring terms and compared to other proposals in th is area, such as Link's lattice-theoretical approach. The present proposal has certain advantages, including the fact that it does not require conjunction to be an associative operation.

HoEKSEMA, jACK and ZwARTS, FRANS Some remarks on focus adverbs.

HoELTER, MARTIN and WILKENS, RoLF Connectionist language processing.

10:ooo-ooo

Edited by Noel Sharkey Book review

8: 1-275 HoEPELMAN, jAKOB and ScHNITZER, RuDOLF (eds) Special issue on Focus in phonetics, syntax, semantics and pragmatics. HoEPELMAN, JAKos; MAcHATE, JOACHIM and ScHNITZER, RuDOLF Intonational focusing and dialogue games.

8:25 3-275

The traditional conceptions concerning the problem of focus, which are known under names like 'Functional Senrence Perspective' and 'Topic Focus Articulation', leave unanswered many questions conceming the semantics and above all th e pragmatics of utterances where intonational focusing is involved. Bur precisely these questions of pragmatics such as the

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Much o f what we have to say in the present paper should be th ough t o f as an attempt to get a better u nderstanding of the intricate complexities surrounding focus adverbs. To this end, we discuss a number o f restri�tions which govern th� occurrence o f different classes o f focus adverbs in Dutch, German and English. We are aware that in doing so we limit ourselves primarily to the descriptive level-a decision which some may well disapprove of Indeed, when Pullum ( 1 98 8) speaks o f the 'slow agony o f empirical endeavor', he characterizes this position in the following way: 'Simply listing facts will do us no good . . . It is easy. It is boring. Anyone can do it. But ultimately it is useless to serious linguistic research .' It goes without saying th at we, too, completely disagree with this way o f.presenring the matter..When one tries to characterize focus phenomena, it is immediately clear th at there are numerous lexical differences wh ich interact in subtle ways with regular syntactic and semantic patterns. In our opinion, a linguistically interesting th eory of focus should be able to accounr for these differential patterns. The strategy followed explicitly by Alnnann ( 1 976, 1 978) and Jacobs (1983), and implicitly by many others, which is to concentrate a few common focus particles (often th e words for only and even ), and to hope th at the rest will somehow conform to the patterns exhibited by the selected items, may lead ro some initial progress, bur musr eventually be replaced by a more comprehensive e ffort. It is dangerous to rely too much on the assumed homogeneiry o f linguistic classes, especially in the closed-class systems. It is also important to gain a comparative perspective on focus adverbs by comparing items from di fferent languages i f a theory is to be constructed with rhe explanatory depth and broad empirical coverage o f current th eories o f WH-movement or anaphoric dependencies. We give a number o f examples w here it is useful to compare items taken from Englis h, German and Dutch. To summarize: this paper reviews some of the major problems wh ich a comprehensive th eory o f focus adverbs needs ro address, describes some o f the variations ro be found among focus adverbs, and places th is against the background of some of the available accounts o f focus adverbs.

Index of A bstracts 2 5 dialogue strategies o f the d ialogue participants, the usage of cond itions and d ialogical-func­ tions o f intonational focusing, and the dialogue context have to be answered to arrive at an adequate d escription of the complex phenomenon o f intonational focusing and its functioning in communication. In this paper we aim to discuss some o f these problems within the framework of the so­ called 'Dialogue Game Theory'-a game theoretically oriented discourse grammar-in ord er to develop a model for the interpretation o f intonational focusing in d ialogues.

VAN JAARSVELD, HENK J. and ScHREUDER, RoBERT Implicit quantification of temporal adverbials.

4:32 7-3 3 9

jACOBS,jOACHIM Focus ambiguities.

S: I -36

In what follows I will d iscuss amb iguities related to focus and stress in German sentences. Some of these ambiguities will be typical instances of what is now widely called 'focus projection', a term which was introd uced by T. Hohle in his seminal paper of 1 982. Focus projection arises in phrases with specific 'normal' stress patterns and consists in the possi bility o fassigning to such phrases several focus-background structures (FBS), d i ffering from each other in the size o f the constituent in focus. For example, in example ( 1 ), ( 1 ) Peter schlagt 1Erich.

with main stress on the object NP, this NP may be focus, but also larger units containing it, namely the predicate schlagt Erich or the whole sentence. What I will try to show is that within the theoretical framework sketched in Jacobs ( 1 988a) focus projection proves to be a non-general, non-homogeneous and , in a way, a non-essential phenomenon. Focus projection is non-general in the sense that its typical pattern d iscussed in the literature-'normal' stress, larger foci properly includ ing smaller foci-does not cover all instances of focus ambiguity. Focus projection is non-homogeneous because, even i f one confines oneself to the typical cases, it comprises several completely d i fferent kind s o f amb iguity. Finally, focus projection is non-essential i f-as the term suggests-it is interpreted as involving percolation o f focus features. In ord er to explain any o f the ambiguities to be d iscussed here, the grammar d oes not have to contain technical mechanisms that move focus features along certain paths in structural trees.1 Rather, all o f these ambiguities follow from ind ependently motivated rules relating d ifferent levels o f grammar or constraining grammatical representations. I wil l also try to show that within (a slightly mod ified version o0 the theoretical framework

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Vague temporal adverbials like soon , recent ,just, which are used to locate situations in time, do not set clear bound aries to the time-interval between the reference-point (usually the moment of speech) and the time at which the situation referred to occurs. In the context o f particular sentences these vague adverbials receive an interpretation that restricts the length o f the time­ interval to a more limited range of values (cf. John hasjust smoked a cigarette and John hasjust married). In two experiments the contextual inrerpretation o f vague temporal adverbials was investigated by means of quantifi cation judgement whereby subjects had t() provid e numerical estimates for the time-interval. The frequency and the d uration o f everyd ay human acts expressed by the verbal phrases of sentences were systematically varied and were shown to have an effect on the assumed length o f the time-interval. Larger estimates were obtained when adverbials were combined with verbal phrases that expressed in frequent acts and acts with longer d uration. In a final section some theoretical implications o f the experimental results are presented.

26 Index of A bstracts mentioned above some of the notorious empirical prob lems o f focus projection fi nd a solution, and that they do so without introduction o f additional theoretical machinery.2

KAc, MICHAEL B. A simplifies theory ofBoolean semantic types.

_

KASPER, �ALTER Presupp_osi�ions, composition, and simple subjunctives.

Th e traditional view on simple ���te-nces in subjunctive mood regards them as a kind of counterfactual conditional with a missing antecedent. This paper discusses the nature of these unexpressed antecedents by relating such sentences to corresponding sentences in indicative mood and full counterfactual conditionals. Usually it is assumed that context is the main I source o f retrieving these ! Unexpressed conditions. It is shown here that they can also be considered as presupposition-like entities induced by the semantic content of the simple subjunctive sentence itself. Subjunctive sentences also raise problems for standard assumptions of h ow the meanings o f expression contri bute to sentence meaning. For in simple su bjunctive sentences sentential constituents can play a di fferent role from th at in indicative senteno:s: in subjunctives th ey must be interpreted as contributing to the unexpressed antecedent o f the underlying conditional rather than to the consequent. Finally, a representation for simple subjunctive sentences as conditionals in Discourse Representation Theory is proposed together with a mechanism for deriving the antecedents of these conditionals from the content o f the sentence. The mechanism accounts for the different roles o f the constituents in simple indicative and subjunctive sentences without requiring special syntactic-semantic rules.

KLEIN, ULRICH F. G. Focus: An idea in motion.

Th is paper discusses stress patterns concerning the relation between intonation and the meaning o f an utterance. It is argued th at this should be done within a linguistic approach. W h ile the first part of this paper deals with differences in meaning that are caused by di fferences in intonation, the second part gives an overview over the focus theories o f Hohle, Rochemont, and Jacobs. With in this discussion the interrelation is shown between the syntactic feature (+F) (+ focus) and th e p honetic and semantic properties of an utterance.

KoULAs,JACK c£ Hinrikka,Jaakko and Koulas, Jack

I O:ooo-ooo KRAMER, EMIEL Artikelworter im Deutschen; Semantische und pragmatische Aspekte ihrer Verwendung. By Hansjorg Bisle-Miiller Book review

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Th e theory of semantic types in Keenan and Faltz ( 1 9 8 s) is insu fficiently constrained in the sense th at it requires denumerab le categories to be interpreted under certain conditions via nondenumerab le algeb ras. An ontologically more austere version of the theory is proposed in which expressions are always interpreted in terms of fi nite algeb ras and it is sh own how it is nonetheless possible to treat an infi nite language by providing an inductively defi ned hierarchy of such algebras, each representing a stage o f an expanding knowledge base. Some apparent o bstacles are considered and disposed of and some advantages discussed, having to do with th e alethic modalities and referential opacity induced by predicates of propositional attitude. Finally, it is shown that a weaker version of Keenan and Faltz's central mathematical result, the Justification Theorem, suffices for th e revised system and a simple, intuitive proof for it is given.

Index o f Abstracts 27

KRIFKA, MANFRED Generische Kenrizeichnungen.

6: 1 6 1 - 1 6 8

By Gerhard Heyer Book review

KRIFKA, MANFRED Focus and presupposition in dynamic interpretation.

I o:ooo-ooo

KuNo, SusuMo 1 :6 1 -93 Principles ofdiscourse deletion-Case studies &om English, Russian, and Japanese.

A syntactically optional constituent in a sentence can be deleted i f it is recoverable from the preceding context. This does not mean, howevt>r, that all such constituents are deletable. This paper hypothesizes that there is a pecking order of deletion, which dictates that deletion should proceed from less important to more important information. Evidence is drawn from English, Russian and Japanese in support of this hypothesis. Interaction of this constraint with various syiuactic rules in each individual language is examined, and it is hypothesized that unacceptabiliry does not result when the above pecking order of deletion principle is violated due to the structural pressure of the language. Further discourse deletion data from Russian and Japanese are introduced, and principles that control them are formulated and justi fied.

KuRODA, S. Y. Indexed predicate calculus.

A programme to construct an extension o f predicate calculus is proposed in which predicates and constants are indexed and interpreted with respect to di fferent (mini-)worlds referred to by indices. From another perspective the proposed system is an extension of the idea of indexing noun phrases in syntactic representations in generative grammar. Some applications are given. In particular, it is applied to the description of ambiguities in intensional contexts, and a comparison is made with a description recently given by Saarinen.

LADD, ROBERT D. Even, focus, and normal stress.

The rwo traditional schools o f thought about sentence-accent placement-the 'Nuclear Stress Rule approach' and the 'semantic highlighting approach'-do not really o ffer competing accounts of the same phenomenon, but emphasize different aspects of the overall problem. A good starting point for integrating these rwo approaches is provided by recent work by Gussenhoven. His work makes crucial use o f the notion that the focus of a sentence may extend over several constituents and may be divided up into one or more accent bearing domains, within which accent placement is structurally specified. By extending this notion beyond Gussenhoven's original use o f it, we arrive at a fundamental distinction berween 'information chunking' (which depends on 'given/new', semantic weight, etc.) and focus (which is a

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Structured meanings have evolved as a well-suited tool to describe the semantics o f focus constructions (c£ von Stechow 1 99o,Jacobs 1 99 1 , Krifka 1 992). In this paper, ! will show how structured meanings can be comb ined with a framework o f dynamic interpretation that allows for a cogent expression of anaphoric relations and presuppositions. I will concentrate in particular on the semantics of the focusing particle only and discuss several phenomena that were unnoticed or unsolved so far, for example the introduction of discourse markers in the scope o f only and alternatives that are anaphorically related to quantifiers. In particular, I will show that the proposed representation format can handle sentences with multiple occurrences of focusing particles. The paper also includes a discussion of the behavior of negation with respect to presuppositions, and of principles that govern focus on quantified NPs.

28 Index of Abstracts syntactic phenomenon). These represent two separate (though interrelated) functions of accent, and have distinguishable effects on where accents are located. Past descriptions emphasize one or the other of these functions, and can be reconciled with each other if the distinction between the two functions is recognized.

LANG, EWALD Primary perceptual space and inherent proportional schema: Two interacting categorization grids underlying the conceptualization of spatial objects.

LASERSOHN, PETER Existence presuppositions and background knowledge.

IO: I I J- 1 22

When a defi nite noun phrase fails to refer, the statement containing it is often felt to lack a tmth value, as in The king ofFrance is bald . In other examples, however, the statement seems intuitively false, and not truth-valueless: consider the case of a speaker who points at an obviously empry chair and says The king oJFrance is sitling in that chair. The difference appears to depend on the pragmatics of verification; we know the sentence is false because the chair is empty-the question of the existence of the king of France need not even come up. A semantics is sketched for assigning truth values to sentences relative to information states. A sentence containing a definite NP may be evaluated as false relative to a given information state rather than simply truth-valueless if, after removing the information that the NP fails to refer, the resulting information state still cannot be consistently extended to one making the sentence true. On this assumption, existing proposals for the semantics of negation in information-state semantics turn out to correspond to internal and external negation, respectively.

LEMMENS,jAN The game oflanguage. By Jaakko Hintikka in collaboration with Jack Kulas. Book review

LEvELT, WILLEM J. M. and CUTLER, ANNE Prosodic marking in speech repair.

2:205-2 1 7

Spontaneous self-corrections in speech pose a communication problem; the speaker must make clear to the listener not only that the original utterance was faulry, but where it was faulty and how the fault is to be corrected. Prosodic marking of corrections-making the prosody of the repair noticeably different from that of the original utterance-offers a resource which the speaker can exploit to provide the listener with such information. A corpus of more than 400 spontaneous speech repairs was analysed, and the prosodic characteristics compared

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Within the realms of cognitive studies, spacial structure is one of the few domains where attempts to trace mental representations from the level of sensory input conditions through conceptual structure to their lexical and grammatical organization seem to be feasi ble and revealing. Presenting a linguist's approach to the meaning and use of spatial dimensional terms, the paper aims to demonstrate why and how the semantic analysis of these linguistic items has to b e justified in terms of nonlinguistic conceptual structure formation, which in turn has to be shown to derive from categorized perceptual input. Regarding framework and approach, the paper supplements Manfred Bierwisch's recent article on comparison in ]S, 6:1.57-93 and 2. 1 0 1 - 1 46. As to substance, it is argued that the structure of conceptual knowledge of spatial objects can plausib ly be modelled by means of object schemata which result from two interacting categorization grids c:::alled Primary Perceptual Space and Inherent Proportion Schema . Offering an analysis which draws on linguistic theorizing, the paper is meant as an invitation to psycholinguists and psychologists for discussion and cooperation.

Index of Abstracts 29 with the syntactic and semantic characteristics of each repair. Prosodic marking showed no relationship at all with the syntactic characteristics of repairs. Instead, marking was associated with certain semantic factors: repairs were marked when the original utterance had been actually erroneous, rather rhan simply less appropriate than the repair; and repairs tend to be marked more often when the set of items encompassing the error and the repair was small rather than when it was large. These fi ndings lend further weight to the characterization of accent as essentially semantic in function.

L6BNER, SEBASTIAN On definiteness.

LOCKART, F. cf. Sanford, A. J. and Lockhart, F.

LoNGACRE, RoBERT E. The semantics ofthe storyline.

This paper starts out with the definition and exempli fication of storyline and non-storyline elements in narrative with emphasis on their structural and semantic functions as narrative universals. The schema thus developed is then confronted with languages of diverse word order rypologies: SOV (Ethiopia), VSO (certain Nilotic languages of Sudan), and SVO (a number of West African languages). The major storyline consideration in the Ethiopian languages which are represented is the matter of the storyline scheme status of various kinds of 'gerunds' (non- fi nal verbs) as opposed to fi nal verbs in chaining structures; this is seen to entail some basic semantic distinctions. Some Nilotic languages, of which Luwo of Sudan is representative, have strict VSO structures on the storyline but various kinds of NV structures off the storyline; here again various semantic distinctions are entailed. The function of consecutive tenses in various VSO and SVO languages is then considered along with the consideration of storyline schemes in SVO languages which do not have special consecutive tenses. A general parallelism of the medial and fi nal clauses in SOV languages to initial and consecutive clauses in VSO and SVO languages is noted. Several further parallelisms and differences among SOV, VSO, and SVO languages are noted.

LucAs, ANGELA

cf. Sanford, Anthony J.; Garrod, Simon C.; Lucas, Angela and Henderson, R.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

The paper develops a uniform analysis of the definite article in its various uses and of de fi nite NPs in general. Starting from the observation that singular and plural mass and count definites all refer to single, though possibly complex objects (Ch. 1 ), it is argued that the logical role of defi nites is that of individual expressions (terms). To treat definites as quantifiers is logically inadequate, but (referential) quantification involves defi nite reference to the domain of quantification (Ch. 2). In Ch. 2 a distinction berween functional, relational, and sortal concepts is intruduced along with a corresponding subcategorization of noun occurrences. This enables a systematic categorization of the uses of defi nites, the main distinction being that berween semantic and pragmatic definites (Ch. 4). Semantic definites represent functional concepts which exist independently of the particular situation referred to, whereas pragmatic defi nites depend on that situation for unambiguous reference. In sketching a more elaborated theory of anaphors it is finally argued that the definite article always indicates that the noun is to be taken as a functional concept. Defi nites in general receive functional concept interpretations, but the definite article is unique in its role: all other determiners require a sortal or relational concept interpretation of the noun.

30

Index: of Abstracts

LUDLOW, PETER The representation of (in)definiteness.

8:277-286

Edited by Eric Reuland and Alice ter Meulen Book review

LUTZEIER, PETER ROLF Varieties of formal semantics.

Edited by Fred Landman and Frank Veltman. Book review

MA.CHATE, j OACHIM

8:25 3-275

cf. Hoepelman,Jakob; Machare,Joachim and Schnitzer, Rudolf

The central topic of this article is the relationship between coreferential, non-identical, lexical NPs in texts (so-called NP;-NPj-pairs). The main question in this article is which interpretation and representation must be attributed to the relationship between NP; and NPi if they are to justify the (referential, cohesive and informative) functions and the adequacy of NPi in texts. Th; ;�s�er has been given In terms of text relations arid text intentions. If has been argued that the adequacy of NP;-NPj-pairs cannot simply be made dependent upon either rhe status of the entities involved in the reader's knowledge store, or the surface characteristics of sentences or texts, such as the distance or the structural relation between NP; and NPj. Rather, the adequacy is dependent on the question whether NP;-NPj-pairs are adequate actualizations of implicit text relations which, in turn, have to be adequate specifications of the intentions underlying the text. Such a treatment enables us to construct a representation for NP;-NPi­ pairs in which the different functions ofNPi arc incorporated.

MARsHALL,joHN C. Close enough for AI? Multiple review of 'The society of mind' by Marvin Minsky.

MARSLEN-WILSON, WILLIAM cf. Tyler, Lorraine K. and Marslen-Wilson, William

MARTIN, jOHN N. Negation, ambiguity, and the identity test.

1 :2 5 1 -274

Negation has been closely tied to semantic presupposition since the concept was first discussed. In most accounts there is a definition or a theorem to the effect that A presupposes B if and only if A and the negation of A, in one sense of negation, both entail B. The multiple senses of negation assumed by such principles hav.: been criticized and along with it the concept of presupposition. Irideed one of the most interesting arguments against semantic presupposition is the j oint claim that many-valued semantics for presupposition require ambiguous negation and that negation as found in English is not ambiguous. In this essay I propos� to discuss quire generally the idea of ambiguity and the role of negation in presupposition theory. Along the way I shall argue that it is quire difficult to explain precisely how the usual identity test for ambiguity employed by linguists should apply to a logical connective like negation, and that most versions of the rest when clarified do not yield the result that negation in English is ambiguous. I argue for these conclusions by attempting to clarify what the theoretical properties oflanguage would have to be if this critique of semantic presupposition were right. The kind of syntax and formal semantics needed to support the identity test when combined

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

MAEs, ALFoNs A. The interpretation and representation of coreferential lexical NPs in expository texts.

Index of Ab stracts

31

with the relevant data about natural l anguage usage does not yield the result that negation is ambiguous. The argument is based on details that are of some interest in themselves. An effort is made to formulate precisely what the identity test is, and in particular what the conditions are that must be met before a meaningful conjunctive abbreviation is permissible. Two different sorts of conditions are distinguished which really amount to rwo quite different versions of the test. Only one of these is really relevant to the issue of negation. This variety is also of interest because failure in this sense amounts to what philosophers have called zeugma. Both sons are d istinguished from a third version of the test, probably the most common, in which i t establishes syntactic but not semantic amb iguity.

MAYER, RoLF Motion imperatives.

MAYER, ROLF The release ofinformation in discourse: compactness, compression, and relevance. A family of formal concepts is developed to deal with the release of information in discourse. We show in particular how the effects of various temporal orderings can be properly described. The interaction with non-monotonicity is pointed out. Our results are pertinent to a linguistic theory of perspectivization.

McCAwLEY, jAMES D. Discourse semantics. By Pieter A. M. Seuren Book review

McDowEn,JovcE P. Quasi-assertion.

8:3 I I-3 3 1

Speech act theory recognizes several illocutionary acts which make the assertive point. This paper proposes a new member of this group, quasi-assertion. Epistemic modal sentences are examples. The force of quasi-assertion differs from full assertion with respect to the felicity conditions on these illocutionary acts. The propositional content condition on assertion is empty. but the propositional content condition on quasi-assertion is that the propositional content P represent a state of affairs in the actual world at utterance time. The preparatory condition on P requires that the speaker have grounds for the truth of P, but the preparatory condition on quasi-assertion only requires that the speaker infers or deduces P. The sincerity condition on assertion requires that the speaker believes that P, b ut the sincerity condition on quasi-assertion allows that the speaker's commitment to the truth of P is less than for full assertion. Under the hypothesis of constructab ility (Searle & Vanderveken 1 98 5 ), we show how quasi-assertion can be constructed out of assertion by the operations (�). (E>) and [-). Finally, truth conditions are given for quasi-assertion.

D

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In this paper a restricted sample of motion imperatives is treated within the framework of discourse representation theory. In order to pave the way for this treatment, the concept of path and linguistic aspects of path connection are discussed. The semantic analysis is then extended in the pragmatic direction: It is shown how semantic inferences may be filtered out via pragmatic considerations. We suggest that a level of execution structure is needed to supplement the level of ser.1antic representation. Motion imperatives are evaluated against maps, and aspects of executability are discussed. It is finally shown how the deontic function of motion imperatives can be fulfi lled by texts in the indicative mood and that the criteria of adequacy valid for motion imperatives then have to be met by motion indicatives.

32 Index of A bstracts

MERCER, ROBERT E. Default logic: Towards a common logical semantics for presupposition and entailments.

MERIN, ARTHUR Permission sentences stand in the way ofBoolean and other lattice-theoretic semantics.- - - -

Permission sentences undermine th e Fregean dogma of speech -act-independent propositional contents. I expound, develop and criticize little-read analyses o f this phenomenon by Lewis ( 1 979) and Karnp (1973, 1 979). Section 2 reformulates Lewis's state-transformation paradigm for atomic commands and permissions, related as direct and inverse, namely non-monotonic theory ch ange. Lewis's and Stalnaker's search for remedies inspired by conditional semantics is outlined. Section 3 develops Kamp's related analysis o f or-coordinated permissions in diverse ways and detail, and extends them to and -coordinated permissions. Boolean or intuitionistic or even semi-lattice:: interpretation of or and, more drastically yet, and turns out empirically non-predictive. Permission (unlike pure command) state transformations are not (semi-)lattice homomorphisms. Implicatural routines o f doubtful provenance would carry the whole explanatory b urden. The problem generalizes to similar analyses (Kamp 1 973) of determiners all, every , any, and some . In ferential di fferences b etween conditional antecedents and permission clauses are demonstrated. Section 4 b riefly notes prob lematic relations to axiomatic theories of non-monotonic theory change, and suggests desiderata for formal descriptive semantics. Hopes are pinned on a distinction between predictively overlapping theories of (rapid) understanding and (considered) inference, consistent with E. J. Lemmon's view that arguments in, not sentences o f, natural languages have logical forms.

DE MEY, SJAAK Only as a determiner and as a generalized quantifier.

Two rypes o f linguistic theories have been particularly concerned with the analysis of 'only': pragmatics, in particular focus theory and presupposition theory, and generalized quantifier (GQ) theory, the latter in the negative sense that it has been eager to show th at 'only' is not a GQ.Judging from such analyses, th en, it would appear that the analysis of'only' is nor at h ome in the grammar of natural language. The main negative point of the present article is to dispute this. The main positive point is the o bservation that there are strong relationships between 'all', 'the' and 'only'. We propose a way to account for them.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Presupposition and entailments play an important role in determining the meaning o f a natural language utterance. Considered as inferences, presuppositions and entailments can be derived from appropriate logical representations of the uttered sentence, the b ackground real world knowledge, and knowledge concerning conversational principles. Presuppositions are conjectural or defeasib le in nature, and entailments are deductive. In this paper we describe the application of Default Logic proof theory (which includes First Order Logic proof theory) to the generation o f presuppositions and entailments. Classical logic, wh ich can generate the entailments, is enhanced with default rules which capture the linguistic knowledge required to produce the presuppositions. The similarities and differences between presuppositions and entail ments when considered as inferences are discussed. We also show that the Default Logic paradigm, in addition to generating the appropriate presuppositions and entailments, has explanatory power.

Index of A bstracts 3 3

MICHAELIS, LAURA A. 10: 1 93-237 'Continuity' within three scalar models: The polysemy of adverbial still.

MaxEY, LINDA M. and SANFORD, ANTHONY J. Quantifiers and focus.

s: 1 89-206

This paper concerns a neglected but potentially important aspect of natural language quantifi ers. Certain quantifiers serve to identify various proportions of sets. Thus Jew , for example, identi fies a smaller proportion ofa set than many. However, different quantifiers may serve to identi fy similar proportions, yet produce somewh at different representations when they are used. The distinction between Jew and aJew is considered in some detail, along with related expressions. It is ciaimed th at these expressions serve to put into focus different su bsets of the supersets upon which they operate. It was suspected that while a Jew X do Y puts emphasis on the small su bset of X of which 'do Y' is true (the 'refset'), with Jew X do Y, emphasis is put on that large, complement, su bset of which 'do Y' is taken to be not true. This suspected difference in emphasis is revealed by considering the acceptability of pronominal reference in subsequent sentences to the set partitions which result from using these quantifi ers. In order to test these tentative observations, a large number ofsu bjects was examined ming a sentence continuation procedure. This allowed for a detailed analysis of the preferred patterns of anaphoric reference to subsets panitioned by quantifiers. The major conclusions were that Ji'w , veryJew , and 110t many put emphasis on the complement set, w hile aJew and only aJew put emphasis on the refset. Furthermore, the pattern of continuation content was markedly different for these two groups. Additional results are descri bed in wh ich it is shown how connectives influence this picture. It is argued that quanti fiers not only identify different proportions, b ur that they ditTerentially emphasize the subsets wh ich they generate, and can fur;ction as 'comments' on these proportions. Possi ble th eoretical relations with negativity and affectivity are discussed.

MUSSELER, J OCHEN c f Hielscher, Martina and Miissler,Jochen.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

This study represents an elaboration and revision of Konig's {1 977) account of the synchronic interrelations among three senses of the English adverbial still. These senses at issue are those in which still serves as a market of a state's continuation to a temporal reference point, as a concessive particle, and as an indicator of marginal memb ership within a graded category. I argue here th at the three semantically and grammatically distinct senses can be reconciled by the modern speaker: the lexeme still has an abstract meaning compatib le with three types of scalar models. In each of these models, still denotes the existence of effectively identical elements at two contiguous scalar loci. Still-bearing sentences code the existence of an element at th e more advanced of these loci, licensing the inference (via lexical presupposition or scalar entailment) that a like element can be found at (at least} one scalar point located closer to the origin of the scale. The three scalar models are ontologically distinct: th e scalar loci in question may be time points, worlds, or simply ranking within a property scale. The elements ordered may b e eventualities or entities. With respect to its role in discourse, still functions as a scalar operator in the sense of Kay (1 990): it serves to relate two propositions within a scalar model. The sense network described here, if it can be regarded as a plausible speaker generalization, provides evidence for the existence of an abstract conception of persistence, i.e. one not restricted to the temporal domain. Persistence can be defi ned for scales and via scalar inference in general.

34 Ind ex of Ab stracts

7:22 1 -244

NIINILUOTO, lLKKA

4:20<)--222

The cognitive resolution of references d epend s on whether an anaphor has previously been mentioned in the anteced ent text or whether a relation berween the anaphor and the anteced ent has to b e consrructed by means of an inference. This inference process in comparison with a mere concept repetition prod uces an i ncrease in read ing and comprehen­ sion time. In the present paper aspects of this processing di fference are examined. Recognition data are used in addition to reading and comprehension times in order to compare the mental representations generated by the d ifferent processes. Experiment 1 shows that the differences in processing time not only depend on the repetition of the concept, but are also produced by varying the semantic d istance berween concepts. Differences in recognition performance were not o bserved , ind icating that the resulting text representations were similar on a semantic level. In Experiment 2 the point of time of the inference process d uring text processing was determined by a word -by-word presentation. The results show that the inference process starts immed iately after the reception of the critical reference concept and ends with the completion of the proposition. Finally, Experiment 3 examines whether inference processes are d ifferent after specification or generalization of a previously mentioned concept. Here the inference effect occurs with specification -rather than generalisation. The-recognition d ata ind icate that it is questionab le wherher an inference process takes place when a concept is generalised . Imagination and fiction. This paper employs possible worlds semantics to develop a systematic framework for stud ying the syntax and the semantics of imagination sentences. Following Hintikka's treatment of propositional attitides like knowled ge and perception, the propositional construction 'a imagines that p' is taken as the basic form to which other sentences (such as 'a imagines b , 'a imagines an F', 'a imagines b as an F') are red uced through quantifi ers ranging over 'world lines', i.e., functions picking out individ uals from the relevant possi ble world s or scenes. This intensional analysis is compared and contrasted with Barwise and Perry's situation semantics. It is also suggested that the logic of imagination helps us to und erstand some peculiarities of fictional d iscourse. For example, acts of imagination can be d irected toward s functional entities (e.g. Donald Duck, Anna Karenina) as well as real ones. Further, fictional texts, like novels, can be thought of as occurring within the scope of an imagination operator, relative to the author or the reader. The author of a fictional text T can be viewed as performing an illocutionary act of recommendation of the form: Let us imagine that T! '

NIVRE, JOAKIM c£ Allwood ,Jens; Nivre,Joachim and Ahlsen, Elisab eth

NOORDMAN, LEO G. M.

o:ooo-ooo

c£ Bosch, Peter and Noord man, Leo G. M. (ed s)

NoRVIG, PETER

7: I 1 I - r r 3

Semantic interpretation and the resolution of ambiguity. By Graeme Hirst. Book review

0AKHILL, jANE cf. Garnham, Alan and Oakhill,Jane

7=379-393

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

MussELER,JocHEN and RicKHEIT, GERT The cognitive resolution of anaphoric noun references.

Index of Abstracts 3S

OATLEY, KEITH Conflict and control among mental agents. Multiple review of 'The society of mind' by Marvin Minsky.

vAN 0IRsouw, RoBERT T.

3:201 -227

Accessibility of deletion in Dutch. In this paper, I argue that the rules of Gapping, Right-node Raising, and Coordination Reduction in Dutch should be collapsed into one general rule which deletes material under identiry in coordinated structures; a rule which says: 'Delete under identity in coordinated structures'

OvER, DAviD E. Constructivity and relational belie£

Relational or de re b elief is normally contrasted with propositional or de dicto b elie£ A completely different distinction has been thought to be that between constructive and non­ constructive reasoning. But ifwe say that a constructive belief is one justified to some extent by constructive reasoning, then we can see that constructive and relational belief are connected. Constructive belief arises from some kind of direct contact (or at least possib le contact) with an object, and so does relational belie£ In more detail, these two concepts are joined by the pragmatic notion of effective reference: a person can only have a constructive belief for e relational belief about an object if he can effectively refer to it. A speaker is said to use a singu­ lar term with effective reference if and only if he has the ability to decide effectively which object it refers to. Effective reference is a vague concept which comes in degrees, but these are also properties of constructive and relational belie£ Effective reference is the basis of the most general theory or relational b elief, and should itsel fbe founded on an intensional semantics for the concept of ability. Quine puts forward an unacceptable analysis of ab ility, and any accept­ able one would imply that his scepticism about relational belief is unjusti fied.

OvER, DAvm E. Mind and meaning. By Brian Loar Book review

OvER, DAvm E. Inquiry.

By Robert Stalnaker Book review

PECHMANN, THOMAS Dynatnic aspects oflanguage processing. By J. Engelkamp and H. D. Zimmer Book review

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

This rule does not use structural information; it refers neither to 'deep' nor 'surface' structure , but only to the linear order of constituents. The rule i s su bject to one primary constraint: a deletion target site must be accessible according to at least one of three criteria: it must b e right­ peripheral to its clause, left-peripheral to it, or have a verb in it. An account is then given of what happens if a deletion target site is accessible according to more than one criterion. With these rules, it is possi ble to account for bidirectional deletions in Dutch su bordinate SVO clauses, and unacceptable deletions such as •svo and sv•, VSO and VO, in Dutch.

36 Index of Abstracts

PERNER, JOSEF and GARNHAM, ALLAN Conditions for mutuality.

6:369-3 8 5

PERRlG, WALTER J. and ZIMMER, HuBERT D. Understanding written language. - By A. J. Sanford and-S� C. Garrod Book review

PLANK, FRANs Verbs and objects in semantic agreement.

Even though not b oasting overt and systematically used noun classifi ers of the variety known as classificatory verbs, languages may still have predicates (co-)signalling particular categories of nominal classification (outside syntactic agreement). Standard examples are English verbs such as to barklne(gh !gallop requiring su bjects which refer to particular animals, or otherwise classify their su bject referems as being in the relevant respects comparable to the animals in question. I hope to demonstrate in this paper that semantic agreement of this kind, which has often figured in theoretical discussions about the structure of the lexicon and the interface of semantics and syntax, is not as unsystematic as is commonly assumed. Although there may be considerab le cross-linguistic variation, this variation at least appears to be quantitatively patterned insofar as some languages (such as German) have relatively more instances of semantic agreement between verbs and objects than others (such as English). I suggest further that the incidence of semantic verb-object agreement is not a minor, isolated, and emirely unpredictable difference between individual languages, b ur correlates with the rypology of the grammatical core relations of subjects and object, and in particular with the object­ differentiation characteristics of a language: verbs and o bjects seem to agree more commonly in languages which give morphosyntactic, and in fact lexical, recognition to at least two semantically relatively specific types of core objects (such as direct and indirect object).

DE RAAo, BOELE The language of social acts.

4:22 3-236

This paper descri bes the derivation of a taxonomy of interpersonal verbs which are used for the description of social acts. Starting from a theoretical delineation of the domain of social acts, a comprehensive set of 1 3 30 interpersonal verbs was selected from a Dutch d ictionary. Both the rationale of the selection criteria for this set, and the reliability of these criteria are discussed and the successive steps in the stocktaking-procedures and some adaptations of the original set are descri bed. Finally, the procedure for structuring the domain of verbs is explained. Suggestions are made for the use of the resulting method for the description of social acts, consisting of 4 s verb scales.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

We present a finite psychological decision procedure for determining whether a situation S provides a participant a in that siruarion with grounds G for assuming that a and b , the other participant, mutually know some proposition p indicated by S. Our criterion derives from analytic criteria proposed by Lewis ( 1 969) and Schiffer ( 1 972). We discuss how our criterion applies in a series of text examples, and compare it with Clark and Marshall's ( 1 98 1 ) triple copresence heuristic . We argue that triple copresence is empirically incorrect. It is neither a necessary nor a sufficienr condition for mutuality, and it fails on a wide variety of examples. We also consider Sperber and Wilson's ( 1 986) recent claim that the concept of mutual knowledge should b e replaced by those of murual manifesmess and mutual cognitive environmenrs, and argue that this move fails to solve the problem of mutuality. Finally we dis­ cuss how community membership produces mutualiry. We argue that mutuality can only be established if certain rules of common sense reasoning can be assumed, and discuss the sense in which these rules may be ;mutually' known.

Index of A bstracts

RAMsAY, ALLAN Presuppositions and wh-clauses.

37

,

We develop a formal framework for discussing presuppositions, based on the notion that meaning shoul d be seen as a rel ation between information states. We then provide treatments o f some specific presuppositional constructions within this framework in an attempt to show that these constructions can be described using rather simpler syntactic rules than are usually employed.

READ, STEPHEN Disjunction.

REICHGELT, HAN Mental models and discourse. In this paper I take the view that using language amounts to constructing 'mental models'. Accordingly, semantics has to explain the structure o f these mental models and the principles by which people construct them. The system proposed, which was developed jointly with Nigel Shadbolt, is called S-R Semantics. Among the fundamental features o f the system is a functional distinction drawn between two sorts of mental o bject epistemic objects, which are supposed to model the long-term estab lished knowledge a processor b rings to a discourse, and discourse objects, which model the o bjects introduced into a discourse and the predications made a bout them. The system is used to solve a number of problems which come up in connection with some uses of singular indefi nite noun phrases, and disputes.

REICHGELT, HAN Strategies of discourse comprehension. By T. van Dij k and W. Kitsch Book review

REYLE, UwE 10: 1 23- 1 79 Dealing with ambiguities by underspecification: Construction, representation and deduction.

In this paper we develop a theory o f language meaning that represents scope amb iguities by underspeci fied structures. The set of possi ble meanings o fa sentence, or text is determined by a set o f meta-level constraints that restricts the class o f semantic representations appropriately. Thus the way ambiguities are represented does not correspond to any o f the usual concepts of formalizing ambiguities by means of disjunctions (or completely specified structures). A sound and complete proof theory is provided that relates these structures directly, without considering cases.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In relevant logics one can formally distinguish two logical operators symbolized as 'A V B, and 'A + B,. Addition holds for 'V' and Disjunctive Syllogism for '+', but not vice versa. The question arises, whether this distinction between two different formal notions of disjunction can be found in natural reasoning. First it is necessary to reb ut Quinean objections to any rival to classical logic, Grice's claim that an intensional disjunction is not needed to explain the everyday uses of or, and Kempson and Cormack's argument that there can be no ambiguity between putative readings one of which entails the other (for A + B I- A V B ). Finally Jackson's use o f assertibiliry-conditions to defend the thesis that 'if A then B, is equivalent to 'A ::J B , is rejected, but the notion of rob ustness which he introduces is usefully adapted to show why Disjunctive Syllogism must fail for 'V'. Having cleared the ground in this way, two uses of or are considered which have in everyday reasoning the inferential properties of 'V' and '+', thus defending the relevantist claim that or is ambiguous.

3 8 Index of Abstracts

RHODES, RICHARD Lexical taxonomies.

RICHARDS, BARRY Discourse and deixis.

There is an inclination to think that pronouns, when they have singular indefi nite antecedents, aamit o f only two kinds o f interpretation. They can be seen - either as 'bound' by their antecedents or as coreferential with them. Which is appropriate on a given occasion will depend upon how the antecedent is used. We shall argue that neither account can be paradigmatic of the relation between pronouns and singular indefi nite antecedents, at least not as this relation is realized in discourse. 'Binding' would seem to misrepresent the typical structure of discourse and indefi nite noun phrases do not refer, despite the fact that people sometimes use them with something particular in mind. We shall suggest an alternative account which satisfies some of the intuitive desiderata.

RICKHEIT, GERT c£ Habel, Christopher and Rickheit, Gert (eds).

RICKHEIT, GERT c £ Miisseler,Jochen and Rickheit, Gert

10: 1 8 5- 1 9 1 RICKHEIT, MECHTHILD Grammatische und konzeptuelle Aspekte von Dimensionsadjektiven.

Edited b y M. Bierwisch and E. Lang Book review

RIGTER, BOB Intensional domains and the use of tense, perfect, and modals in English.

A theory for the use of tense and perfect in English should do three things: I . It should provide rules defining the phrase markers in which tense and perfect can occur; 2. It should speci fy what extralinguistic phenomena correlate with the occurrence of tense and perfect in the structures that underlie English sentences; 3· It should provide rules which are sensitive to these extralinguistic phenomena, and which place either + PAST or - PAST under tense nodes and either HAVE or 0 under perfect nodes, in such a way that the resulting surface realizations correctly march the conceptualizations which th�y are used to express. As regards point 1 , a set of rules of rhe type used in rhe autonomous syntax tradition (c£ Culicover et a/. 1977, Lapointe 1 980) is adopted here, and it is assumed that syntactic structures are paired with functional structures (c£ Bresnan 1 978) in such a way that any V1 node in a phrase marker is paired with a renseless proposition. As regards point 2, it is argued char for every tenseless proposition referring to a stare/event n , the speaker can make ; truth

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In the course of my work on a dictionary of Ojibwa, an Amerindian language of the Great Lakes region, I have come across a number of interesting lexical facts which constitute an important part o f what speakers know about the reference o f words but which are not generally included in semantic analysis as it is currently practiced. ' In particular, while working on plant names, it came to my attention that there exist lexical taxonomies which have an existence based on, but separate from, the conceptual taxonomies they label (and which, in turn, are separate entities from the Linnean taxonomies which the Western mind tends to believe are the 'real' truth about biology). Even a cursory examination of the literature on ethnobiology will quickly reveal the languages and culture specific nature o f these lexical taxonomies and the conceptual taxonomies which underlie them. Berlin, Breedlove, and Raven ( 1 973) (henceforth BBR} outline the basic principles governing the structure o f such taxonomies which they refer to as folk taxonomies .2

Index of Abstracts 39

RIJKHOFF, J. Nominal aspect.

8:29 1-309

In this article I argue rhar besides verbal aspect, which concerns the way a property or relation is represented in rhe temporal dimension, there is also nominal aspect, which relates to the way a property is represented in the spatial dimension. I will contend that certain elements, which are often believed to be number markers, are in fact nominal aspect markers. Evidence to support this will be taken from several generically unrelated languages. Additionally it is suggested rhar nominal aspect plays a role in connection with incorporated and predicate nouns.

RoHDENBURG, GUNTHER Dogs, bitches and other creatures.

In an attempt to uphold a speci fic constraint on sentential ambiguity Kempson ( 1 980) has proposed an elaborate framework for dealing with the semantic duality of items such as dog . Her analysis culminates in a duality principle relating the specific and the general interpretation o f the terms in question. These proposals are shown ro have a number o f serious shortcomings. First, they do not allow for rhe fact that the two relevant interpretations may vary in strength, and that independently o f each other. Second, the dualiry principle is inadequate in several respects. In a number of sets containing just two members like dog and bitch , the principle is incapable o f estab lishing precisely the speci fic interpretation of the general term. Nor does it correcrly predict the absence of such specific uses in certain cases. Worse still, the principle is incompatible with the fact that the kind o f duality under study is also found in a great variety of multiple sets. In addition, there are analogous phenomena associated with various types o f part-whole relations which are not captured by the principle. Finally, as was shown in Horn ( 1 98 3), a case in which a given term functions as irs own hyponym may originate in other ways than those predicted by Kempson's thesis.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

commitment to the effect that n holds true at a rime yo within a particular intensional domain (henceforth ID). It is argued rhar he does this by assigning two semantic tenses to that proposition on the basis of certain chronological orderings within that ID. These two semantic tenses will be called domain tenses. The two domain tenses are the F0-tense and the yo-tense. The notions yo and P are defined as follows: yo is the rime where n is located in the chronology o f the ID in which the domain tenses for the tenseless proposition referring to n are computed. P is the segment o f rime on wh ich the speaker focuses his attention when committing himsel f to the truth o f n at yo in that ID. pn is the present o f that ID. The tense is defined in terms of the chronological order o f P with respect to pn, and the yo-tense is defined in terms of the chronological order of yo with respect to P. It is argued that a third semantic tense, called a domain-shift tense (henceforth DOSH tense) is relevant for the surface selection o f tense and perfect. A DOSH tense is defined in terms of the relation between the present o f an embedding ID and the present of an embedded ID. Tense-representation rules are provided which match the three semantic tenses referred to above correctly with two nodes (i.e. tense and perfect) in finite S-structures, and wit� one node (i.e. perfect) in nonfinire V2 complements. The most complex examples of 'sequerice of tenses' are covered by these tense­ representation rules. Alleged exceptions to 'sequence o f tenses' are dealt with. It is argued that in free indirect style, in counterfacrual conditional statements, and also in children's games of pretend and in sel f-effacement strategies in conversational interaction (c£ Lodge 1979), the referent n of the renseless proposition that is paired with a nonembedded clause is presented as true, nor in the speaker/writer's primary ID, bur in an embedded ID. The idiosyncrasies o f the use o f tense perfect and modals in these nonembedded clauses are rhus also to be attrib uted to domain-shi ft phenomena.

40 Index of Abstracts

ROMMETVEIT, RAGNAR In search of a truly interdisciplinary semantics. A sermon on hopes of salvation &om hereditary sins.

ROMMETVEIT, RAGNAR On pronomes, pronouns and dyadic coordination of attention.

5: 1 77- 1 79

Multiple review of 'The sociery of mind' by Marvin Minsky

ROORYCK, j OHAN Restrictions on dative cliticization in French causatives. Causative constructions in French display restrictions as to the cliticization of lexical datives onto the causative. In altogether different frameworks, Fauconnier (1 98 3), Burzio (1 986) and Goodall ( 1 987) have related this restriction to the ergative-inergative distinction. However, the inabiliry to formally define ergative verbs in French, as well as further restrictions on the cliticization of datives in causative constructions show that this hypothesis fails ro account for the data observed. A thematic condition on dative cliticization in causatives adequately describes the restrictions noted.

SADOCK, JARROLD M. cf. Zwicky, Arnold M. and Sadock,Jarrold M.

vAN DER SANDT, RoB A. Presupposition projection and anaphora resolution. The present paper presents an anaphoric account of presupposition. It is argued that presuppositional expressions should not be seen as referring expressions, nor is presupposition ro be explicted in terms of some non-standard logic. The notion of presupposition should not be relegated to a pragmatic theory either. I nstead presuppositional expressions are claimed to be anaphoric expressions which have internal strUcture and semantic content. In fact they only differ from pronouns and other semantically less loaded anaphors in that they have more descriptive content. It is this fact which enables them to create an antecedent in case discourse does not provide one. If their capaciry to accommodate is taken into account they can be treated by basically the same mechanism which handles the resolution of pronouns. The

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Current expansions of linguistic research into pragmatic issues appear to be of a multi- rather than interdisciplinary nature, and novel mini-theories represent a mixture of different and partially competing auxiliary presuppositions about social-interactional features of language superimposed upon a shared heritage of Cartesian assumptions about its non-social essence. Hypothetical constructs from prestigious models of language as an idealized, static, and semantically closed system are introduced in various disguises as intervening variables into explanatory accounts of actual human discourse. Some hope of salvation from this hereditary sin emerges from an exegesis of Wittgenstein's aspect-theory, William James's notion of our 'trading on one another's truths' and a cryptic remark on natural language and realiry by Werner Heisenberg. A constructive alternative to current multi-disciplinary semantics may thus hopefully be developed on the basis of empirically founded constructivist theories of language and thought in conjunction with a systematic analysis of basic social-interactional features of ordinary language. The synthesis, I argue, yields a truly interdisciplinary and dynamic extension of semantics, and is finally illuminated by case analysis within the field of word semantics. Linguistically mediated meaning, within such an explicitly constructivist and social-interacnonist paradigm, is conceptualized in terms of orderly contextual specification of meaning potentials oflinguistic expressions under the constraints of the temporarily shared social realiry of conversation partners and their intuitive mastery of dialogue roles.

Index of Abstracts

41

theory is elaborated i n the framework o f discourse representation theory. It i s shown that pragmatic factors interfere in the resolution of presuppositional anaphors. The resulting account can neither be classified as wholly semantic nor wholly pragmatic. Section I presents a survey ofstanding problems in the theory of presupposition projection and discusses the major competing approaches. An argumentation for a purely anaphoric account of presupposition is given in section 2. Section 3 presents a coding of presuppositional expressions in an extension of discourse representation theory. The final section is devoted to a discussion of the constraints which govern the resolution of presuppositional anaphors.

vAN DER SANDT, RoB A. and ZEEVAT, HENK (eds) Special issue on Presupposition. Part I.

9: 1 79-286

SANFORD, A. J. cf Garrod, Simon C. and Sanford, Anthor..y J.

SANFORD, A. J. Some comments on •The Society of mind'. Multiple review of 'The sociery of Mind' by Marvin Minsky.

SANFORD, A. J.

s: 1 89-2o6

cf Moxey, Linda M. and Sanford, Anthony J.

SANFORD A. J.

6: 1 47- 1 60

cf Garrod, Simon C. and Sanford, AnthonyJ.

SANFORD A. J. cf Barton, S. B. and Sanford A. J.

SANFORD, ANTHONY J.; GARROD, SIMON C.; LucAs, ANGELA and HENDERSON, R. Pronouns without explicit antecedents? Yule { I 982) has argued that examples from speech show that pronouns may be interpreted nonreferentially. In the present paper, it is argued that pronouns elicit procedures for the identification of referents which are in explicit focus (Sanford and Garrod, I 98 I). Three experiments are offered in support of this view. The discussion centres on the need for carefully assessing the knowledge-states of listeners when pronouns are used in the absence of antecedents. It is proposed that felicitous use of pronouns without antecedents can occur only when listeners have particular things in mind which serve as 'effective antecedents'. If the listeners do not have these in mind, then it is argued that such usage is infelicitous. It is also argued that speakers may have particular antecedents in mind even if listeners do not.

SANFORD, A.J. and LOCKHART, F. 7:365-378 Description types and method of conjoining as factors influencing plural anaphora: A continuation study of focus. An experiment is reported which investigates the impact of two variables on the likelihood of obtaining plural pronoun anaphora in a continuation task. The first variable is syntactic: the use of and versus with as a means of relating two singular characters. Use of and enhances the likelihood of obtaining a plural anaphor in continuations, bur the incidence of plural is never

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

vAN DER SANDT, RoB A. and ZEEVAT, HENK (eds) Special issue on Presupposition. Part II.

42 Index of Abstracts as high as 6o%. The second variable is description type: whether the characters are introduced through proper names (e.g. Harry), or through a simple definition noun phrase (e.g. the doctor). When two descriptions are of the same type, plural reference is enhanced, regardless of syntactic condition, and regardless ofwhether the two descriptions are names or noun phrases. This effect is traced in part to the special status afforded to characters which are introduced through proper names. A discussion is presented within the framework of an account of plural anaphora based on singular individuals mapping into common role-slots in background knowledge (scenario).

ScHNITZER, RuDOLF

8: 1 -27 5

c£ Hoepelman,Jakob and Schnitzer, Rudolf (eds)

ScHNITZER, RuDoLF

8:2 5 3-27 5

ScHREUDER, RoBERT c£ van Jaarsveld, Henk J. and Schreuder, Robert

SCHROEDER-HEISTER, PETER c£ Friedrichsdorf;Ulf and Schraeder-Heister, Peter

ScHUTZELAARS, Ar.Ex J. H. c£ Smolenaars, Anton J. and Schutzelaars, Alex J. H.

SCHWARZE, CHRISTOPH Concept types and parts of speech. With special reference to the lexicalization of region concepts in French. It is a tradition to express overall structures in the lexicon of a language by classifying its words according to their formal as well as to their semantic properties. The problem treated here is how formal and semantic word classes relate to each other. The problem will first be discussed on a general level. Then some results of an empirical study on the lexicon of space in French will be presented. It will be shown how the concept of'region' is distributed across the parts of speech of that language. The study confirms and specifies the current assumption that there are typical relationships between formal word classes and concept types, bur it also raises the question of what happens within the domain of atypical lexicalization.

SEUREN, PIETER A. M. Thirty million theories of grammar. By James D. McCawley Review article

SEUREN, PIETER A M . The inheritance ofpresupposition. By John Dinsmore Book review

SEUREN, PIETER A M. (ed.) Special issue on the Comparative. SEUREN, PIETER, A M. The comparative revisited. Arnim von Stechow's critical appraisal of various theories of the comparative in the present issue of the journal ofSemantics provides a welcome opportunity for me to look at my 1 97 3 paper again and give my commems. It is clear that that paper falls short of what it wanted to

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

c£ Hoepelman,Jakob; Machate, Joachim and Schnitzer, Rudolf

Index of Abstracts 43 realize. Many extremely interesting observations were not made. Certain troublesome environments (such as counterfactual constructions) were not considered. The paper was definitely poor on the semantics of the comparative. And, finally, hardly any notice was taken of the various ways in which the semantic notion of comparative is expressed in the languages of the world. In the fol lowing I shall recapitulate my own position as given in 1 973. I shall then give my own comments and suggestions for improvement. Finally I shall discuss von Stechow's critique. On the whole, however, I shall maintain the principles on which the I 973 analysis was based.

SEUREN, PIETER A. M. Relevance, communication and cognition.

5: 1 2 3- 1 43

By Dan Sperber and Deirdre Wilson Review article

6: 1 69- 1 74

SEUREN, PIETER A. M. Presupposition and negation.

6: 1 7 5-226

This paper is an attempt to show that given the available observations on the behaviour of negation and presuppositions there is no simpler explanation than to assume that natural language has two distinct negation operators, the minimal negation which preserves presuppositions and the radical negation which does not. The three-valued logic emerging from this distinction, and especially its model-theory, are discussed in detail. It is, however, stressed that the logic itself is only epiphenomenal on the structures and processes involved in the interpretation of sentences. Horn ( 1 985) brings new observations to bear, related with metalinguistic uses of negation, and proposes a 'pragmatic' ambiguiry in negation to the effect that in descriptive (or 'straight'} use negation is the classical bivalent operator, whereas in metalinguistic use it is non­ truthfunctional but only pragmatic. Van der Sandt (to appear) accepts Horn's observations but proposes a different solution: he proposes an ambiguiry in the argument clause of the negation operator (which, for him, too, is classical and bivalent), according to whether the negation takes only the strictly asserted proposition or covers also the presuppositions, the (scalar) implicarures and other implications (in particular of sryle and register) of the sentence expressing that proposition. These theories are discussed at some length. The three-valued analysis is defended on the basis of partly new observations, which do not seem to tit either Horn's or Van der Sandt's solution. It is then placed in the context of incremental discourse semantics, where both negations are seen to do the job of keeping increments our of the discourse domain, though each does so in its own specific way. The metalinguistic character of the radical negation is accounted for in terms of the incremental apparatus. The metalinguistic use of negation in denials ofimplicatures or implications ofsryle and register is regarded as a particular form of minimal negation, where the negation denies not the proposition itself bur the appropriateness of the use of an expression in it. This appropriateness negation is truth-functional and not pragmatic, but it applies to a particular, independently motivated, analysis of the argument clause. The ambiguiry of negation in natural language is different from the ordinary rype of ambiguiry found in the lexicon. Normally, lexical ambiguities are idiosyncratic, highly contingent, and unpredictable from language to language. In the case of negation, however, the two meanings are closely related, both truth-conditionally and incrementally. Moreover, the mechanism of discourse incrementation automatically selects the right meaning. These properties are taken to provide a sufficient basis for discarding the, otherwise valid, objection

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

SEUREN, PIETER A. M. Collins Cobuild English Language Dictionary Book review

44

Index of Abstracts

that negation is unlikely to be ambiguous because no known language makes a lexical distinction between the two readings.

SGALL, PETR On the notion of the meaning of the sentence.

S_<;ALL,� PETR Focus and the levels oflanguage system.

8:37-49

Recent linguistic trends stress the necessity to investigate not only the system of language, but also discourse patterns. In this context it is important ro understand the sentence borh semantically and syntactically not just as 'an "assertion", being the linguistic counterpart of a fact', but rather as a component part of a discourse, which in the general case is not a monologue (Hoepelman & van Hoof I988: 250). The sentence structure should then be described in such a way that its properties imposed by the sentence's functioning in communication are not neglected. This meallS above all that one innsr not neglect the topic­ focus articulation (TFA), which is one of the hierarchies constituting the syntactic pattern of the sentence, and which exists due to the impact of the communicative function on the structure of language. The results of empirical research continuing the tradition of the Prague School and taking into account the methodological requirements of formal linguistics have made it possible to state that TFA (being not only pragmatically, but also semantically relevant, see Section I below) can be described in an economical way as one of the aspects of underlying strucrure (or of sentence meaning, of tectogrammatics), and that this hierarchy has its counterparts (means ofexpression) on the other levels (see Section 2). The i mportance of the phenomena now subsumed under TFA has been known since Weil ( I 844). Between the I 86os and I 920s, linguists such as G. von der Gabelentz, H. Paul, P. Wegner, A. Marry, H. Amman and 0. Jespersen introduced 'psychological subject and predicate' (or 'theme and rheme', later 'topic and comment') into the analysis of general properties of language, and discussed stress and word order as the means of expression of this dichotomy. In Prague, Mathesius ( I 929, I939) formulated a linguistic account of the dichotomy from the viewpoint of a structural comparison of Czech (with its 'free' word order) and English, pointing our that the subject typically expresses topic in English (whereas its primary function in Czech is being the Actor). Among his followers. Firbas ( I 9 5 7, I975) analysed the interplay of this 'functional sentence persective' and word order, showing that not only a dichotomy, but a whole scale of 'communicative dynamism' is present. After Halliday (I 967) brought TFA nearer to the centre of attention of the English-speaking linguistic world. the relevant issues started to be srudied also in the context of generative linguistics; see Chomsky (I 97 I ). more recentlyJacobs ( 1 98 3) and Rochemont (I 986). However,

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

Many linguists and logicians agree that two sentences often differ in their linguistic meaning, though they share their intensions {truth conditions), i.e. though they correspond to a single proposition (more exactly, to a single set of propositions with different reference assignments, see below). However, the research concerning the notion oflinguistic meaning is still scattered, and much remains to be done to clarifY this notion. Our approach is based on the existence of an operational criterion for synonymy, which has been presented elsewhere• and may be summarized as follows: Two expressions (lexical or grammatical morphs, or syntactic constructions) a and b are synonymous (i.e. share one of their meanings) if and only if in every sentence A containing a the substirution of b for a (if grammatically possible) yields a sentence B having the same intension as A (i.e. for every possible world and reference assignment both A and B are true, or both are false, or both are inappropriate).

Index of Abstracts 4S nor much continuity with continental research has been kept within this trend, so rhar even in the rare cases in which TFA is given due attention, the approaches used for its analysis do nor yield a possibility ro describe adequately its position in the system of language (see esp. Kokrova, 1 988a, b).

SHADBOLT, NIGEL Processing reference.

7:3 2 1 -345 SICHELSCHMIDT, LoRENz and GuNTHER, Uno Interpreting anaphoric relations during reading: Inspection time evidence. On-line mechanisms in the processing of anaphora were investigated in rwo reading experiments. Shorr German texts were presented incrementally for self-paced reading, and inspection times were recorded for every sing:e word. Each text contained a critical rwo-clause sentence with an elliptical gap and a personal pronoun. In the first experiment, varying the surface structure of the antecedent clause affected inspection times for the anaphor clause as a whole. In rhe second experiment, varying the semantic structure of the antecedent affected antecedent bur not anaphor inspection times. The findings are not in agreement with the view that anaphor resolution is an exclusively retroactive process triggered on encountering rhe anaphor. Rather, the results suggest that anaphor resolution includes proactive subprocesses: foundations for interpreting anaphora yet to come are laid during reading the antecedent.

SM!T, R. A. c£ Bree, D. S. and Smit, R. A.

SMOLENAARS, ANTON J. and ScHUTZELAARS, ALEX J. H. On cognitive semantics of emotion words.

s:207-23 I

An application of Solomon's semantic theory of emotion words ro 20 Dutch displeasure terms is empirically rested in a quasi-ecological way: the semantic specification of a word is systematically transformed into a real-life story. Subjects confronted wirh the story should recover the word that gave rise to it. The fir berween theory and data turns our to be only moderately satisfactory and this is attributed to the theory as such, rather than to specificities of the semantic description. The theory rends ro a semantics that is 'esoteric', over-meticulous and roo 'formal'.

SPERBER, DAN c£ Wilson, Deirdre and Sperber, Dan

5= 1 45- 1 62

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

A system of referential description is presented rhar attempts to represent crucial aspects of the process of performing and underst�nding referential acts. It is suggested that traditional logical accounts distract our attention from important properties concerning the use of referential expressions. The model proposed is consonant with a growing body of opinion amongst cognitive scientists that generating and interpreting natural language is best explained as a process of constructing cognitive models and procedures that represent and process the con­ tent of our utterances. If this position is taken seriously, there is a requirement rhar the state of language processors is the most important determinant of rh,: of the referential act. This leads to a process model of reference. The paper also touches on why language is in a sense 'radically opaque' and why this opacity does not consistently lead to failure in communicative acts. The theory predicts that using language is a 'risky' business and rhar misinterpretation will occur more often than other formal theories predict.

46 Index of Abstracts

STASSEN, LEON The comparative compared.

a) the question of how to explain the attested occurrence and non-occurrence of categories. That is, we will have to provide a principled answer to the question of why it is just these attested categories that figure in the typology, instead of other, equally thinkable, alternatives. b) the question of how to explain rhe distribution of languages over the a rrested categories in the typology. That is, we must ask ourselves why a given language L in the sample belongs to category X in the typology of comparatives, and nor to car�ory Y. In other words, we start from the assumption that the groupings of languages which emerges from our typology of comparatives is the reflection of a grouping into natural classes, and hence our task will be to identify rhe determining factor or factors of this naturalness. In section 3 of this paper I will present the typology of comparative constructions which I have constructed on the basis of the data in my language sample. Following that, rhe two explanatory questions mentioned above will be dealt with in section 4 and section S· However, before we can start out on the actual typological investigation of comparatives, there are a few preliminary points whch must be cleared beforehand. These points are both of a general methodological and of a merely practical nature.

STASSEN, LEON Composite predicates in English. By Ray Cattell Book review

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In the recent literature that has been devoted to the syntactic and semantic properties of the comparative construction, universal characteristics of this construction type have been largely left out of the discussion. In this paper, my aim is to argue that this is an unfortunate situation, and that from a universalist survey of comparatives valuable insights as to the nature of this construction type can be gained. In particular, I want to make a case for the claim that, from a universal point of view, the comparative construction should not be looked upon as a 'primitive' or 'autonomous' construction type; rather, its particular formal manifestation in a given natural language can, to a significant extent, be predicted from the formal manifestation of other syntactic patterns occurring in that language. In this way, the universal investigation of comparative constructions may be of interest to syntacricians, semanticists and cognitive psychologists alike. As far as formal syntactic theory is concerned, our claim has a consequence that, in universal grammatical theory, we do nor need construction-specific rules to generate comparative constructions; these constructions will be derived automatically by rules which are independently needed for the syntactic derivation of certain other construction types. From a semantic (or perhaps better: cognitive) point of view, our investigation leads to the contention that the mental operation by which two entities are compared to one another is not independent, 'primitive', mental operation; at least as far as its codification into natural languages is concerned, the mental act ofcomparison must be seen as a conceptual extension (a 'cognitive metaphor', so to speak) of certain other mental operations, which must be viewed as more 'basic' or 'fundamental' to the human cognitive system. The data upon which the present cross-linguistic survey of comparatives is based are gained from a sample of I 1 0 languages, chosen from genetically diverse language families. In each of these languages, the linguistic manifestation of the comparative construction is identified. A categorization of the various ways in which natural languages may encode the mental act of comparison gives rise to a typology of comparative constructions in which ar least five categories can be distinguished clearly. After this typology has been established, we face two interrelated questions of an explanatory nature, viz.

Index of Abstracts 47

VON STECHOW, ARNIM Comparing semantic theories of comparison. This is a critical and constructive view of recent semantic theories of the comparative. Ir is critical because I want co show that none of the existing theories gives an adequate account of what I consider relevant data. The article is constructive in so far as I try to show how deficiencies of existing theories can be overcome. At the end a picture emerges of what I believe co be a best theory of the comparative, given the actual state of research and the data discussed. I will discuss proposals by B. Russell, P. Postal, E. Williams, P. Seuren, E. Klein, D. Lewis, M. Cresswell, and L. Hellan. Some other approaches are briefly mentioned (Bansch-Vennemann, S. Wheeler). I try to evaluate the different accounts with respect co the question how certain linguistic phenomena are treated by them.

VON STECHOW, ARNIM Structured meanings. By Max]. Cresswell Review article

STERELNY, KIM Against conversational implicature. In this paper I shall attempt co show that a considerable amount of recent effort in linguistic theory has been devoted to a non-solution of an apparent problem.1 The problem arises through a communicative contrast. In some cases a sentence conveys its literal meaning. For instance, suppose a speaker (S) utters to someone (A ) { I ) Someone should teach oilmen's sons to shoot straight. How is it that A understands ( I )? A natural answer for those working within generative grammar runs as follows. S and A have internalised a common system of semantic, syntactic and phonological rules. In virtue of common possession of this apparatus, the message that S encodes A decodes.2 So our theory of the grammar of English provides an account of how English sentences are understood providing we take the theory co be a claim about the psychology of English speakers. But it seems that this account of understanding is insuffi­ ciently general. For consider the following quite standard examples. Imagine a group of people at a table, eating. One says to another:

(2) Can you pass the wine? The addressee, rather than answering, passes the wine. Another standard example: imagine a nobleman and his lackey alone in a room at the nobleman's residence. The nobleman says: {3) It's cold in here. The lackey closes the window. A further example. Two burglars are in a warehouse. One lifts his head saying:

(4) That's a police siren. The burglars leave hurriedly. Some writers on 'Indirect Speech Acts' have suggested that sentences like (2 ) are literally requests: that they are presented as requests in deep structure.3 But even if they are right, the account of understanding alluded to above will not generalise co all these cases. It is obvious

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

VON STECHOW, ARNIM My reaction to Cresswell's Hellan's, Hoeksema's and Seuren's comments.

48 Index of Abstracts thar (3), literally, is a statement about the temperature of a contextually definite space-rime region, while (4), literally is a statement about a continuing noise. Bur despite their literal meanings, (2) conveys a request, (3) a command, and (4) a suggestion. It seems therefore char some more general account needs to be given of how (2)-{4), and the indefinitely many similar cases, are understood. In this paper I will sketch Grice's mechanism for handling cases like (2)-(4). I then argue that: i. Grice's theory of how these cases are to be handled is nor easily integrated within a well motivated theory oflanguage. At best Grice's mechanisms are part of the heuristics of language use: rough rules of rhumb available to language users. ii. Grice's mechanism is not needed to explain how examples like (2)-(4) can be understood.

3=277-294

By Andrew Woodfield Review article

TASMOWSKI-DE RYCK, LILIANE and VERLUYTEN, PAuL Linguistic control of pronouns. We argue against the claim put forward by Lasnik ( 1976) that pronouns are, in all cases, - pragmatically controlled, i.e. .that they directly refer to objects or situations in the world. In fact, rhe generalization we defend is exactly the opposite: all pronouns are linguistically controlled, i.e. they have a linguistic antecedent in all cases. Even in chose instances where no antecedent is present in uttered discourse, the necessity of postulating such an antecedent, and the possibility of identifying it, can be demonstrated. An antecedent which is nor present in uttered discourse is subject ro particular recoverability conditions, both pragmatic (it must be control led by a salient object) and linguistic (the pronoun which is controlled by such an absentee antecedent can only occur in a restrained class of discourse conrexrs).

TASMOWSKJ, LILIANE and VERLUYTEN, S. PAUL Control mechanisms of anaphora. Taking as a starring point the hypothesis that all 'true' pronouns (in a well-defined sense) have a linguistic antecedenr even when neither the sentence nor the text provide one explicitly, we review some recent approaches where the contrary is defended or implied. It is shown that these approaches run counter to the facts at crucial points. The cases discussed lead to a further development of our own views with respect to a unified approach of pronominal anaphora.

TRAVIS, CHARLES Reference and essence. By Nathan U. Salmon Book review

TYLER, LORRAINE K. and MARSLEN-WILSON, WILLIAM 1:297-3 1 4 Processing utterances in discourse contexts: On-line resolution of anaphors. The on-line interpretation of utterances in discourse contexts was investigated by varying the type of dependency berween an utterance and its context. Listeners heard short sequences of utterances ending in incomplete fragments. The fragments varied in length and in whether their anaphoric linkage to the context (by repeated names, pronouns or zero anaphors) required inferences to be resolved. The subject's task was to name a visual continuation probe char appeared at the offset of the fragment. The differences berween naming latencies to appropriate versus inappropriate probes was constant across conditions, and irrespective of whether or nor inference-based processes were required to determine this preference. This was interpreted as showing chat on-line speech processing is not necessarily slowed down by the use of inference to link utterances to their contexts.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

STERELNY, KIM Thought and object.

Index of Abstracts +9

UHMANN, SusANNE On the tonal disambiguation of focus stress. In this article I will be concerned wirh cerrain well-known ambiguities of focus srrucrures arising from a special accenruation of phrases or sentences. This ambiguiry has been discussed under the heading focus projection 1 in the lirerarure. What is generally understood by focus projection goes back ro Chomsky ( I 97 I ) and maybe even ro Hermann Paul ( I 8 8o)-'avanr Ia lerrre'. Chomsky ( I 97 1 : 20 1 ) observed that a senrence such as { I ) was he (warned (to look our for (an exconvicr (with a red (SHIRT) ) ) ) )

1.

the internal structure of the intonational phrases; (i.e. the number of pitch accents and their choice from rhe inventory of German pitch accents); 2. the initial boundary of the intonational phrases; 3· the pitch range; 4· the association of syllables and pitch accents.

VERLUYTEN, PAUL cf. Tasmowski-De Ryck, Liliane and Verluyten, Paul

VERLUYTEN, S. PAUL cf. Tasmowski, Liliane and Verluyten, S. Paul

VET, Co Thematic relations (Syntax and Semantics, vol. 20).

1 0: ! 8 ! - 1 8 3

Edited by Wendy Wilkins.

VLIEGEN, MAURICE Die deutschen Verben des Sehens.

5:268-273

By Klaus Robering Book review

VAN VOORST,JAN A localist model for event semantics.

10:6 5- 1 1 1

This study presents a localisr model illuminating the semantics of transitive constructions. The model groups transitive argument strucrures inro a number of classes similar to those found in many aspecrual studies which focus primarily on temporal aspects of events. The building blocks of the model are based on linguistic evidence in the form of tests involving adverbials of different rypes. A crucial role is played by manner adverbials. These adverbials show that the entiry given by the subject has control over the entities given by the direct object of accomplishments and activities, bur not over those denoted by the prepositional object and by rhe direct object of achievements and states. The notions proposed are inclusion and cohesion which specify rhe spatial relationship berween rwo entities in terms of the event brought into existence by one of them. The localist model is a comprehensive conceptualization of events

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

any of the constiruents i n brackets may b e regarded a s a focus. The presence of the foci is indicated by an inronation centre placed on the constiruenr shirt. Although I in principle share Hohle's view,2 who underlines rhe importance offormulating the rules for focus projection, I am not going to deal here with the rules governing focus projection, the origin of focus, or the problem of focus assignment in detail. My main interests in this article are instead concerned with the tonal realization of senrences. The question I will rry to answer will be the following: does the ambiguiry stated on the accenrual level also show on the intonational level? To answer this question I will examine four aspects of the tonal realization more closely.

so Index of Abstracts expressed through transitive constructions and it explains why individuation has such an important role to play in aspect.

DE VUYST, jAN The present perfect in Dutch and English.

4: 1 3 7- 1 6 3

Certain differences between Dutch and English use of the present perfect are considered in relation to a more general difference between the two languages that involves temporal inclusion. It is shown that Dutch and English exploit different means of expressing a temporal inclusion relation between two events, notably where Vendler's activities and accomplish­ ments are concerned. Precisely in those cases Dutch and English use the present perfect in different ways. But when there are no differences in the expression of temporal inclusion, not­ ably where Vendler's states are concerned, the two languages use the present perfect in the same way.

WEKKER, HERMAN Structures and transformations. The Romance verb. By tJ Poutain-Book review

VAN WERKHOVEN, J. P. cf Bree, D. S. and van Werkhoven,J. P.

WESCHE, BIRGIT At ease with at.

The preposition at has a range of rather diverse meanings-locative, temporal, causal, etc.­ which would not seem to be captured by any common denominator that would still be strong enough to distinguish at from other prepositions. In order to clarify the relationship of the various meanings of at to each other, this paper assumes the notion of a semantic prototype and shows how other senses of at develop from a basic locative sense of at, and how these derived senses are motivated by the fact that their domain of application is conceptualized in an analogy to the domain of the basic spatial sense, or in analogy to another sense of at that is directly or indirectly related to the original locative sense along the same lines.

WEYTERS, ToN Discourse analysis. By Gillian Brown and George Yule Book review

WieHE, RoB T. P. External and verb phrase negations in actual dialogues.

8: ! 07- 1 2 5

Negations play a n important role i n actual dialogues. I f one participant o f a dialogue is negating an utterance of the other participant (or is uttering a sentence that entails a negation of an utterance of the other participant), there occurs a verbal conflict between the participants! This conflict is resolved as soon as a participant is forced (only by verbal means, of course) to give up a sentence so that there is no longer a conflict. In recent theories about negation a distinction is rather often made between two different kinds of negation.2 Gabbay & Moravcsik ( 1 978) and Hoepelman ( 1 979) make a distinction berween (sentence) negation and denial. Gabbay and Moravcsik motivate their distinction as fol lows: 'That the negation of a true proposition takes us to a false one, is one of the early lessons in elementary logic. Sentence negation is important for logic, for it gives us general

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

wADA, HAJIME cf. Asher, Nicholas and Wada, Hajime

Index of Abstracts

s1

WILDGEN, WOLFGANG Processual semantics ofthe verb.

s:321-344

Processual semantics opens a new field of research by looking at basic physical, organismic, and cognitive processes involved in meaning. Irs srrategy of model-building is directed towards rhe application of the theory of dynamic systems. We consider three rather specific domains of the lexicon ofverbs (using the classification of Ballmer and Brennenstuhl as an approximation of our descriptive goal): a) Verbs referring to bodily motions b) Verbs referring to actions (in space and rime) controlled by one agent c) Verbs referring to interactions (in space and rime) between different agents. Making use of major results of ecological psychology, we show that the dynamics of basic motions and actions (including their perception and control) can essentially be described with reference to fundamental physical systems (the simple and the double pendulum). Our main point is that those processual schemata relevant for the control and perception of bodily motions and actions are also cognirively and semantically basic. We claim that these schemata can describe and explain the geometrico-dynamic component in the meaning of verbs of motion and bodily action. These hypotheses are tested in the analysis ofseveral groups ofverbs, which are prototypical for the domain in question. When we add further classificatory schemes (e.g. auditory ones) and the internal evaluation of the process (e.g. using Osgood's semantic space), we can approximate the mental organization of meaning. The neurolinguisric aspects of our issue are discussed in the last section of the paper.

WILKENS, RoLF

o:ooo-ooo

c£ Hoelter, Marrin and Wilkens, Rolf

5: 1 4 5- 1 62 WILSON, DEIRDRE and SPERBER, DAN The self-appointment ofSeuren as censor. A reply to Pieter Seuren.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

ways of characterizing contradictoriness, and thus helps us formulating such basic laws as the law of non contradiction. In everyday discourse however, negative sentences are used to formulate denials of various sons. In fact, even the notion of a denial is too narrow; denial, objection, criticism, etc. are all everyday activities the point of which is to say: 'No, it is not like this; rather, it is like that." '3 Jacobs ( 1 982) distinguishes between 'kontrastierende Negation' and 'nicht kontrasrierende Negation', and Horn ( 198;, 1 989) between Truth functional (descriptive, logical) negation and Metalinguistic (non-descriptive, non-logical) negation. Barth & Wiehe (1 986) distinguish between three kinds of negation: Exclusion negation, Choice negation and Discrepancy negation. I will tty to combine all these approaches. In the style ofJacobs I will make a msrinction between Contrasting Negation (Cneg) and Non-Contrasting Negation (NCneg). Cneg has to be divided in Choice negation and Metalinguisric negation. NCneg has to be divided in Exclusion negation and Verb Phrase negation. Verb Phrase negation corresponds roughly with rhe Discrepancy negation from Barth & Wiehe. NCneg is characterized formally by the negation operator '-' from classical, two-valued logic, Cneg by the metalinguisric operator '-KORR'. 'KORR' is Jacobs's conecmess operator. Event logic will be used as a valuable instrument for characterizing formally the difference between the different types of negation that are discussed in rhis paper.

52

Index of Abstracts

WRIGHT, PETER Using constraints and making reference in task-oriented dialogue.

This paper; presents an analysis of linguistic data that stems from a task-oriented dialogue. \lV'e demonstra"te that certain referring expressions used in this setting are potentially ambiguous and indeterminate bur do not lead to referential errors such as under- or over-population of the discourse representation. Three studies are reported which show that it is the skilled use of non-linguistic constraints, present in the task which facilitates successful reference. When these non-linguistic constraints are removed, skilled speakers are able to make a compensatory adjustment in the precision of their referring expressions. The ability of people to make this compensatory adjustment is an aspec� of referential skill not revealed by more traditional task­ oriented assessments.

In this paper I will deal with one important aspect of syntactic structure and intonational structure. Syntactic structure is organized hierarchically and may involve some co-indexing between pans of it, whereas intonational structure is organized linearly from left to right. I shall argue that in matching these two different kinds of structure, one needs an interface level which I will call the lever of-contrast. The particular idea I propose is-one that on-this level syntactic information is used to form a structure of so-called contrast phrases which is purely right-branching, and that it is this constrast structure on which pitch assignment rules apply in order to yield the lay-our of intonational structure. Because in German syntax right-branching dominates over l eft-branching, the contrast structure often preserves the properties of syntactic structure. But there are also clear cases with different structures at the two levels. The empirical background comes from some observations made in a former experimental project on German inronarion (see Wunderlich 1 9RR). In that project intonation contours were systematically varied by resymhesis on the basis of notural utterances, and these synthesized stimuli were then judged in perception tests. We are planning to study the predictions outlined in this paper in a similar experiml"ntal framework. For several reasons, however, these experiments have nor yet been performed. Therefore, the study is still in a preliminary state. The most important observation concerns a particular type of intonation contour in German which may be called a 'bridge' (somewhat similar to the 'hat' in Dutch, which has been investigated by the Eindhoven group). It has been shown elsewhere that the bridge contour is composed of two successive pitch accents, the first one rising with a following high level, the second one falling. What has been puzzling are the conditions under which this bridge contour is realized. It can be produced under quire different circumstances. The idea put forth in this paper is that the two pitch accents whirh complement each other in the bridge signal a contrast between two focus domains on which the bearers of the accent can be projected.

YuLE, GEORGE Interpreting anaphora without identifYing reference. It is proposed that, by adopting an antecedent-determined account of the interpretation of anaphoric pronominals, we may misrepresent what is actually required in the interpretive process. If we adopt an antecedent plus predicate(s)-determined account, we may arrive at massively over-specified representations which would seem to create a substantial processing load. Since it can be observed that one of the characteristics of conversational speech is the occurrence of antecedendess pronominals, it is suggested that the analytically required referential identity for anaphora resolution may nor actually be an on-line processing requirement. What hearers may do is focus their attention on what is predicated of (at least some) pronominals, following the focus-marking of the speaker, and simply accept that there

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

WUNDERLICH, DIETER Intonation and contrast.

Index of Abstracts

s3

is, for the speaker, some referent or referential set for the pronominals encountered. Consequently, for the horer, identifying the reference of those pronominals need not be a requirement in the interpretation of the speaker's u tterance. That is, the interpretation of anaphora need nor be, in some circumstances, a referential issue at all.

ZADEH, LOTFI A. A fuzzy-settheoretic approach to the compositionality of meaningful proposi­ tions, dispositions, and canonical forms. ·

•

•

•

most ofher closefriends.

ZEEVAT, HENK Presupposition and accommodation in Update Semantics. A reconstruction is presented of van der Sandt's theory of presupposition in the framework of update semantics and extended to belief sentences. The resulting view is confronted with earlier approaches to presuppositions (especially Heim's) in update semantics, concentrating on the approach to accommodation. It is shown in some derail that the anaphoric view of presupposirion can be maintained for only a subclass of presuppositional triggers and must be given up for another class. The paper shows that the treatment of presuppositional anaphora and presupposirional accommodation is compositional with respect to stacks of information states. The brief development of the approach in section 7 shows, however, that, contrary to what one would expect, an approach in terms of stacks of informacion stares is a powerful method in the study of DRT and other dynamic systems.

ZEEVAT, HENK cf. van der Sandt, Rob A. and Zeevat, Henk (eds)

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

In irs traditional interpretation, Frege's principle of compositionality is nor sufficiently flexible to have a wide applicability to natural languages. In a fuzzy-set-theoretic setting which is outlined in this paper, Frege's principle is modified and broadened by allowing the meaning of a proposition, p , to be composed not from rhe meaning of the constituents of p , bur, more generally, from the meaning of a collection of fuzzy relations which form a so-called explanatory database that is associated with p. More specifically, through the application of rest-score: semantics, the meaning of p is represented as a procedure which rests, scores and aggregates rhe elasric constraints which are implicit in p . The employment of fuzzy sets in this semantics allows p to contain fuzzy predicates such as tall , kind, much richer ,etc.; fuzzy quanrifiers such as most, several, Jew, usually etc.; modifiers such as very , more or less, quite somewhat, etc.; and other types of semantic enriries which cannot be dealt with within the framework of classical logic. The approach described in the paper suggests a way of representing the meaning of dispositions, e.g. Overeating causes obesity, Icy roads are slippery, Young men like young women etc. Specifically, by viewing a disposition, d, as a proposition with implicit fuzzy quantifiers, the problem of representing the meaning of d may be decomposed into (a) restoring the suppressed fuzzy quantifiers and/or fuzzifying the nonfuzzy quantifiers in the body of d; and (b) representing rhe meaning of rhe resulting dispositional proposition through the use of rest­ score semantics. To place in evidence the logical structure of p and, at the same rime, provide a high-level description of the composition process, p may be expressed in the canonical form 'X is F' where X - (X1, , X.) is an explicit n-ary variable which is constrained by p , and F is a fuzzy n-ary relation which may be interpreted as an elastic constraint on X . This canonical form and rhe meaning-composition process for propositions and dispositions are illustrated by several examples among which is the proposirion p/1 Over the pastJew years Naomi earnedJar more than

54 Index of Abstracts

ZIMMER, HUBERT D. cf. Perrig, WalterJ. and Zimmer, Hubert D.

ZIMMERMANN, THOMAS EoE A note on transparency postulates.

ZwARTS, FRANS cf. Hoeksema, Jack and Zwarts, Frans

ZWICKY, ARNOLD M. and SADOCK,JARROLD M A reply to Martin on ambiguity.

3 :249--2 56

In 'Negation, Ambiguity, and the Identity Test' in this journal (r 982),]ohn N. Martin attempts an explication of the 'identity tests' used by linguists in deciding whether or not particular sentences are ambiguous, examines the application of these tests to negation by not in English, and concludes that claims by Atlas ( 1 977) and Kempson (1975) that there is no ambiguity berween an external and an internal interpretation of not have not been proven. We do not propose to reconsider the negation issue here (but see Blackburn ( r 98 3 ) for a response to Atlas rhar is generally consistent wirh our remarks); rather we are commenting on Martin's interpretations of(a) the function of identity tests in linguistic argumentation; (b) the notion of word in linguistic analysis; and (c) the use of the term understanding in Zwicky and Sadock ( 1 975; hereafter ZS). We content that M is mistaken on all three points and char in consequence his analysis fails as an explication of identity tests.

Downloaded from jos.oxfordjournals.org by guest on January 1, 2011

This paper deals with the role of meaning posrulates in Montague's theory of indirect interpretation. More specifically, it is concerned with the problem of finding suitable constraints on systems of posrulates. One plausible candidate for such a constraint is discussed in some detail. It is a principle to the effect that the logical complexity of any meaning posrulate must not exceed the expressive power of the narural language under description. It is argued that such a constraint would be too powerful because it rules out transparency (first­ order reducibility) posrulates because they are essentially second-order statements. A weaker alternative to the original constraint is then formulated. It is argued that this new constraint should not be the only restriction on the scrucrure of meaning posrulates.