Cognition, Vol. 4, No. 1

Editorial

Hiroshi Faculty Tokyo Hongo, Tokyo,

Azuma of Education, University, Bunkyo-ku, Japan

Richard Cromer MRC Develovmental Psychoiogi Unit, Drayton House, Gordon Street, London, WCIH OAN, Gt. Btitain

Paul Bertelson Laboratoire de Psychologie Exphimentale, UniversitP Libre de Bruxeiies I1 7, Au. Adolphe Buyl, B-IO.50 Bruwelles, Belgique

Peter D. Eimas Waiter S. Hunter Laboratory of Psychology, Brown University, Providence, R.I. 02912, U.S.A.

Ned Block Dept. of Philosophy, M.I. T., Cambridge, Mass. 02139, U.S.A.

Gunnar Fant Lab. of Speech Transmission, Royal Institute of Technology, S-10044 Stockholm 70, Sweden

T. G. R. Bower Dept. of Psychology, University of Edinburgh, 60, Pleasance, Edinburgh EH8 9 TJ, Gt. Bn’tain Franqois Bresson Laboratoire de Psychologie 54, bud. Raspail, F- 75006 Paris, France Roger Brown Dept. of Psychology, Harvard University, Cambridge, Mass. 02138,

U.S.A.

Jerry Fodor Dept. of Psychology, M.I.T. EIO-34 Cambridge, Mass. 02139, Kenneth Dept. of Monash Clayton,

U.S.A.

Forster Psychology, University, Vie. 3168, Australia

Merrill Garrett Department of Psychology, M.I.T. EIO-034, Cambridge, Mass. 02139, U.S.A.

board

David T. Hakes Department of Psychology, University of Texas, Austin, Tex. 78712, U.S.A. Henry Hecaen Directeur d’Etudes. Ecole Pratique des Hautes Etudes, Unit6 de Recherches Neuropsychologiques. I.N.S.E.R.M., 2, rue d ‘Al.&ia, F-75014 Paris, France Michel Imbert Laboratoire de Neuropl~~~siologie. College de France, I I, place Marcelin Berthelot, F- 75005 Paris, France Barbel Inhelder Fact&P de Psychologie et des Sciences de I’Education, Universitt de GenPve, CH-1211 GenPve 14, Suisse Marc Jeannerod Laboratoire de Neuropsychologie ExpCrimentale, Dojien L&pine, F-69500 Bron, France James Jenkins Center for Research and Human Learning, University of Minnesota, Minneapolis, Minn. 55455,

Jerome S. Bruner Dept. of Experimental Psychology, University of Oxford, South Parks Road, Oxford, OXI 3UD, Gt. Britain

Lila Gleitman Graduate SchoolofEducation, University of Pennsylvania, 3700 Walnut Street. Philadelphia, Pa. I9 104. U.S.A.

Peter W. Carey 2730 Dwight Way, Berkeley, Calil: 94704,

Pierre Greco Laboratoire de Psychologie, 54, bud. Raspail, F- 75006 Paris, France

Philip Johnson-Laird Laboratory of Experimental Psychology, Centre for Research on Perception and Cognition, Sussex University, Brighton BNI 9QC, Gt. Britain

Jean-Blaise Grize I, Chantemerle, Neuchdtel, Suisse

Daniel Kahneman Dept. of Psychology, The Hebrew Universitv of Jerusalem, Jerusalem,-Is&e1

U.S.A

Noam Chomsky Dept. Modern Languages and Linguistics, M. I. T., Cambridge, Mass. 02139. U.S.A.

U.S.A

Jerrold J. Katz Dept. of Philosophy, M. I. T., Cambridge, Mass. 02139, U.S.A. Edward Klima Dept. of Linguistics, La Jolla, University of California, San Diego, Calif 92037, U.S.A. Alexei Leontiev Faculty of Psychology, University of Moscow, 13, Frunze Street, Moscow G.19, U.S.S.R. Wilhelm Levelt Psychological Laboratory, Numegen University, Erasmuslaan 16, Nijmegen, Netherlands A. R. Luria Faculty of Psychology, University of Moscow, 13, Frunze Street, Moscow G.19, U.S.S.R. John Lyons Dept. of Linguistics, Adam Ferguson Building, Edinburgh EH8 9LL, Gt. Britain Humberto Maturana Escuela de Medicina, Universidad de Chile, A. Sanartu 1042, Santiago, Chile David McNeil1 Department of Behavioral Sciences, Committee on Cognition and Communication, University of Chicago, 5848 South University Avenue, Chicago, IN. 60637, U.S.A. John Morton Applied Psychology Unit, 15, Chaucer Road, Cambridge CB2 2EF. Gt. Britain George Noizet Laboratoire de Psychologie Experimentale, F-I 3 Aix en Provence, France

Daniel Osherson Department of Psychology, University of Pennsylvania, 3813-15 Walnut Street, Philadelphia, Pa. 19174, U.S.A.

Robert Shaw Center for Research and Human Learning, University of Minnesota, Minneapolis, Minn. 55455,

Domenico Parisi Istituto di Psicologia, Consiglio Nazionale delle Richerche, Piazzale delle scienze 7, Rome, Italy

Dan I. Slobin Department of Psychology, University of California, Berkeley, Calif 94720, U.S.A.

Michael Posner Dept. of Psychology, University of Oregon, Eugene, Ore. 9 7403, U.S.A. Zenon Pylyshyn Departnient of Psychology, The University of Western Ontario, London 72, Ont., Canada Martin Richards Unit for Research on Medical Applications of Psychology, University of Cambridge, 5, Salisbury Villas, Station Road, Cambridge CBI 2JQ? Gt. Bn’tain Steven Rose Biology Department, ‘The Open University, Walton Hall, Milton Keynes MK7 6AA, Gt. Britain Nicolas Ruwet Dept. de Linguistique, Centre Univ. de Vincennes, 12, Rue de Tourelle, F- 75012 Paris. France

U.S.A.

Jan Smedslund Institute of Psychology, Universitet i Oslo, Box 1094, Oslo 3, Norway Sidney Strauss Department of Educational Sciences, Tel Aviv University, Ramat Aviv, Israel Alina Szeminska Olesiska 513, Warsaw, Poland Yoshihisi Tanaka Dept. of Psychology, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan Hans-Lukas Teuber Dept. of Psychology, M.I.T. EIO-034, Cambridge, Mass, 02139,

U.S.A.

Virginia Valian Ph.D. Program in Psychology, C.U.N. Y. Graduate Center, 33 West 42nd Street, New York, N. Y. 10036, U.S.A.

Harris B. Savin Dept. of Psychology, University of Pennsylvania, Philadelphia, Pa. 19104, U.S.A.

Peter Wason Psycholinguistics University College London, Research Unit, 4, Stephenson Way, London NW1 2HE, Gt. Britain

Tim Shallice Psychology Department, The National Hospital for Nervous Diseases, Queen Square, London, WCI, Gt. Britain

Hermina Sinclair de Zwart Centre d’EpistCmoZogie Gtnetique, Universite de GenPve, CH-121 I Geneve, Suisse.

Editorial For the past four years we have been editing this journal both to print the most interesting advances in the field, and to discuss the epistemological, social and political stresses that are the scientists’s lot today. In the course of this endeavor, we have run into many situations which have obliged us to reflect on issues explicitly that we ordinarily might have ignored. For example, we have had to decide whether or not to publish manuscripts that originate in places where the political chaos would hardly appear to allow anyone to concentrate on affairs as abstract as those which generally interest us; we have had to reject and accept papers on the basis of criteria, which though stringently adhered to in our reviewing procedures, are not very clear in substance. Finally, we have had to develop a position as to what Cognition as a field is and what it should be. In doing such things as these, we have accumulated so many unsolved questions that we believe it best to come straight out and talk candidly about our perception of the problems that hound us as editors and scientists. It is obvious that we work within the framework of a cultural setting that obliges us to define what we think is valid and what is not. In our case, for example, we have tried to present papers that are theoretically relevant rather than papers that emanate from interesting observations but which are not presented in a wider frame of reference. Consequently, we have had to juggle such notions as interesting, relevant, theory and frame of reference: Our practical problem is how best to evaluate such notions. Of course, we use the usual methods for doing this, such as having referees who seem to us to be the most appropriate. We adhere strictly to the procedure of soliciting two reviews for each manuscript - two positive reviews are necessary for acceptance of a manuscript regardless of who the author may be or the nature of the subject matter. We also send a paper out to a third reviewer whenever the first two readers do not agree. All in all, we apply the standard methods which should allow us to proceed with some semblance of good conscience. Yet we remain uneasy. The problem that we are raising concerns the implicit nature of the procedures that have been with us all for a great many years. Psychology, like other human sciences, has been influenced greatly by the cultural background in which it has developed. For instance, it is now clear that during the years when positivism and behaviorism dominated the field, other approaches were by and large ignored in publication. Clearly, explicit censorship was rarely used, nor was it needed. Rather, rigorous format monetary correlation accompanies this phenomenon. What is “good”

8

Editorial

standards were established by editors in such a way as to favor manuscripts that originated in the “acceptable” scientific schools. A cultural provincialism developed rendering manuscripts that followed the accepted standards apparently more professional, better written and more comprehensible. As a result, the better known journals all began to look very much alike. Even after many scientists had given up the behaviorist dogma, they continued to mold their papers in the standard way, often masking their philosophical maturation behind a facade of useless procedures and empty statistics. We embarked on our venture, Cognition, largely in reaction to this state of affairs at a time when the pattern was already beginning to change. In fact, the rigidity of the standards and the monotony of the many atheoretical studies led to a reaction which began in the sixties and has been emerging ever since. Cognition has succeeded in part thanks to this reaction. Our problem now is how to avoid being imprisoned by our success. We now go through the same motions as did harassed editors before us. We have our own hobbyhorses and reviewing procedures to protect us. Our dilemma is this: We directly rejected the monotonous format of the standard journals in favor of an open style that would allow us to focus on content. We have discovered in the process that much of what traditionally passed for content was, in fact, form. With so many cognitive “theories” and so few direct empirical implications of any theory, how do our reviewers and we decide what to publish? We must grapple openly with the fact that the atmosphere surrounding the judgment of the worth of an article has become so rarified that we had better go back and take another look at the whole issue of value judgments. Accordingly, we think it may be clarifying to consider the process of criticism in art. This is an area in which at least initially, utilitarian considerations were not at stake as they are in one way or another in most branches of science. If we manage to gain a little understanding of the situation currently existing in art, it is possible that we may shed a little light on what is occurring closer to home. It is highly probable, that if questioned concerning which single painting in the world should be saved from the hecatomb, anyone having finished a secondary education would opt for the Gioconda. It is also equally likely that no consensus would result if the same population was asked which painting in the world they would most like to have hanging on their own wall providing, of course, they had the money to afford it. This observation, though admittedly rather superficial, would appear to demonstrate the fact that society obliges us to have a set of official standards and value judgments often far removed from our personal ones. Naturally, a certain

Editorial

is unquestionably more expensive than what is not. However, it should be noted that there is generally very little a priori justification for our implicit value judgments. Our evaluations lead to no far-reaching theories on aesthetics and human nature. On the contrary, what generally occurs is the irrevocable division of society into those who produce art and those who consume it, with little interaction between the two. The producers quickly draw others into their group and the public soon follows, either to be in the vanguard of a new cultural trend, or for more venal considerations such as potential investment value. Thus, self defined market value determines a good part of the motivation for preserving certain works of art and not others. Our own feeling is that there may well be an independent notion of “good” and “bad” art, based on a covert, aesthetic theory. Nonetheless, this does not affect our argument in one way or another. Most of the time our own feelings need not be closely related to the value judgments reigning in our community. Given the expertise and generally elitist properties required in order to appreciate the more dominant forms of art, there is very little relation between what we like personally, and the value that we as members of society attribute to works of art. Obviously, as academics, it would be hard for us to manifest the same disapproval or contempt for the dominant forms of art that we bestow on certain forms of popular art regardless of what our subjective judgment may be. In a wholly unalienated society it is possible that some form of universal aesthetic theory might be helpful in understanding the dynamics of this situation. However, in a system as plurally determined as ours, we doubt that it would do much good. What should be retained, however, is that aside from putative universal theory, a cultural pattern exists which controls us by fabricating a hierarchy of “good” and “bad” artists whom we select or reject in economic terms, thereby participating openly in the meritocratic system. That a meritocracy serves certain purposes cannot be denied. It automatically limits active participation in any one field to a chosen few while relegating the majority to the role of a passive audience. However, in so doing it also often defeats the fundamental purpose of art, which is to give pleasure by expanding and gratifying our aesthetic senses. This state of affairs is particularly effective in societies where there is a need to use the production and consumption of art as a symbol for the distinctions that are the form and content of elitist political and economic structure. But our point is not to enter into a discussion of social injustice so we will not pursue this point any further. What we would like to stress is that artistic production and criticism have from the start been a form of value judgment aimed at drawing a clear distinction between those in control and those controlled.

10

Editorial

With the passage of time, of course, the concept of what art is has undergone severe changes. It is possible, for example, to claim that the Beatles made an artistic contribution every bit as real as that of any composer working in the modern musical tradition. Schools and trends and their accompanying sets of value judgments are breaking up in all directions. Slicing one’s ear with a razor blade is called “body art”, while leaving a canvas blank is described as “minimal” art. This change in criteria1 attributes can be traced to the fact that society, with the dominance of technology in the last century, has shifted from emphasizing one symbol of social structure (art), to stressing another (science). Art has been freed of its social function and has in consequence been freed of the roots that constrained it. The artist today has few of the characteristics of his or her predecessors. Leonardo, or even, for that matter, the “pointillistes”, were exploring domains which were to be collectively shared. Today, artists generally restrict themselves to the expression of idiosyncratic regularities of form and structure: It is the person of knowledge who has the power of representing our aims and delusions. We are the cultural heros and it is our production that is subjected to societal ratings of “good” and “bad”. In that sense, then, it is far from surprising that the current unconscious race for primacy of value-systems is being fought out in the sciences rather than in the arts. In the light of this shift, it is necessary to explore the way in which we, as scientists, proceed in our daily endeavors, while at the same time acting as standard bearers for the characteristics that symbolize and ensure the hierarchical structure of our society. Science deals with the way in which we construct models of the world. This, in turn, allows us to predict and to control our environment. In this sense, therefore, it should be relatively easy to distinguish between those who contribute productively to such knowledge and those who do not. Although this was an acknowledged fact, the institutionalization of science has resulted in the proliferation of competitive domains, each of which has produced a long list of articles which will fall into oblivion without leaving any imprint on the environment or on the edifice of science. It is increasingly apparent that the selection of the domains to be studied can only partially be attributed to those who conduct the studies. Many intermediary organizations and social structures exert at least as much influence in determining research as any group or individual scientist. Finally, journals and other means of disseminating information have a great influence in shaping the direction of research. These considerations lead us full circle back to our concern about current behavioral science and our role in it as teachers, researchers, authors and editors. The main difficulty that we see is that the true intended product of

Editorial

11

behavioral science, knowledge about humanity, has become clearly alienated from the means which are supposed to produce this product. That is, scientific publications are presumed to lead to increased knowledge; but they serve so many other purposes that are more directly manifest, that their production and consumption has become an independent value. This is reflected in a variety of facts some of which are well-known, but some of which have surprised us as we have discovered them. It is obvious that this editorial has raised issues related mostly to judgments of value. If this is so, it is because it is easy to discern a certain unrest in the scientific community over the dichotomy which prevails between the ways in which we think and feel and the ways in which we actually proceed in everyday life. It would be possible to demonstrate that our comments apply not only to publications and editors but also to the conditions of our employment, promotion or income, the treatment meted out to us by funding agencies, our selection to committees, the pressures of specialization, etc., etc. In fact, although it is rarely discussed as such, the whole issue of value judgments underlies much of our everyday life. It is never apparent how the scientist’s desire to produce new knowledge and the more down to earth considerations such as advancement, power needs, the desire to secure the highest possible evaluation and the most praise, interact to determine our behavior. This being said, we have not raised the questions we have, in order then to be able to enumerate one hundred and eleven rules and regulations for solving the problems of the scientific community, nor to pass judgment on motivations or lack of them. It is our contention that no individual or single organism can do this without playing into the hands of the existing system. Alternatively we feel that it is only by raising a general awareness to the ways in which our feelings and desires diverge from our practices and how the latter are shaped and molded, that a collective form of change that does not contain the vices of the current system may be defined. It is only natural then, that we should invite you to join in a discussion and debate of all or some of these issues in the coming numbers of this journal. J. MEHLER T. BEVER S. FRANCK

Cognition, 4 (1976) 13-30 0 Elsevier Sequoia S.A., Lausanne

- Printed

1

in the Netherlands

Some attributes of adjectives used by young children*

KATHERINE Yale

NELSON

University

Abstract Analysis of 24 spontaneous speech samples from children at 24 and 30 months revealed a correlated progression in the form, function and meaning of modifiers used with increased language development. Predicate adjectives were used to comment on transitory states of objects and prcdominated in the early speech samples. Attributive adjectives were used to subdivide classes or to specify among particular instances within classes on the basis of physical characteristics such as size, by use of a conventional type name, or by) use of a possessive. Attributives predominated in the later speech samples. Differential use of a particular form appears to be a function of its cognitive and communicational usefulness rather than itssyntactic derivation. Adjectives in adult language serve a variety of functions: They describe something about an aspect of a situation, specify which of a number of alternatives is being referred to, or sub-classify within a larger class of things. For example, (1) Elizabeth

was a beautiful

child.

(2) It was the tall blond man who pulled the gun. (3) Dan bought an electric typewriter. We can see that in the first case the adjective describes a particular referent (Elizabeth), while in the second case it specifies which of a number of possible particular referents (several different men) is meant. In the third case, however, the adjective does not modify a particular *This research was supported by a grant from the Carnegie Corp’oration of New York. Data collection was made possible by a NICHD postdoctoral fellowship to the author. Requests for reprints should be sent to the author, Department of Psychology, Yale University, New Haven, Connecticut 06520.

14

Katherine Nelson

referent but a referent class (typewriter) which can be subdivided into the classes of manual and electric typewriters (Bolinger, 1967). Do young children learn and use adjectives in this way? Although children acquire a number of adjective types very early - some appear among their earliest words (e.g., ‘hot’,‘broken’) - there is little research on how they are used, with the exception of a small set of spatial adjectives (e.g., Donaldson and Wales, 1970; Sinclair-de-Zwart, 1969). The use of modifiers (adverbs and possessives as well as adjectives) by young children is of interest for several reasons, however, in addition to the fact that they constitute an early form of predication. (According to Brown’s [ 19731 summary of Stage I speech, when mean length of utterance - MLU - equals approximately 1.75 morphemes, attributives accounted for 6% of the multi-word utterances and possessives for 870.) First, they serve an important conceptual service. They enable the child to make distinctions among referent objects and classes of objects on the basis of both general and specific properties. That is, they provide him with a linguistic means to generate new reference classes. They are thus basic to lexical productivity. In addition, the adjectives used by the child specify the properties of objects, people and other referent entities, properties which are important to studies of concept formation, attention, memory and logical thought. The question of which properties are coded in the language of the young child is therefore a potentially revealing one. The research described here is an attempt to begin the investigation of this problem by examining the use of modifiers in the spontaneous speech of beginning speakers. In this analysis, the syntactic roles, meanings, and functional relations to modified nouns of early modifiers will be considered. Both adjectives and possessives are included here as modifiers. While possessives relate people to things possessed, and adjectives refer to states and properties of things, they share similar semanticsyntactic relations to nouns, and both are used by young children to make distinctions among objects. Adverbs (except for locatives) were rare in the present samples and were not included in the analysis. Special problems arise when spontaneous speech samples are analyzed for the purpose of revealing functional use that can be overlooked when the emphasis is only upon linguistic analysis. When children are learning language speech may contain a good deal of ‘form practice’ and elicited or imitated utterances in addition to functional statements. These learning forms may be revealing of linguistic structure (see Slobin, 1968; Weir, 1962, for example) but are misleading for a functional analysis in that whether they are representative of the child’s own knowledge or interest

Some attributes of adjectives used by young children

15

cannot be known.* Learning episodes have their own conceptual importance, but they are different in kind from the functional use of speech; they cannot be considered true speech acts. A second problem in attempting to determine what the child knows from what he says is that negative evidence cannot be evaluated. If a child does not say something we cannot conclude that he could not. This is especially true for relatively brief samples of speech. At the extreme, it is exemplified in those children who understand what is said to them but do not themselves produce language, whether for physical or psychological reasons. The antithesis of the point is that we cannot conclude that the child means what he says, even when he says it. That is, the child’s meaning may differ from the adult’s even when the same forms are used. At the extreme this is exemplified by the use of ‘empty’ or rote forms that have no reference for the child. ‘Why’ is used by many children in the pre-school period before they acquire an adequate understanding of causation, for example (Blank, 1974). Much recent analysis of developmental semantics (e.g., Clark, 1973) has been devoted to the exploration of the ways in which the child’s meaning for certain terms differs systematically from the adult’s. Still, except for empty forms, it can be argued that most terms used by the child share some of the central meaning structure of the adult terms; if they did not, they would not be acquired, understood or survive in the child’s speech (cf., Fodor, 1972 on this point). Therefore, if detailed claims are not made about the meaning structure of particular terms, the analyst is on safe grounds in assuming that terms and forms that are used frequently and by many child speakers reflect, at least in part, the knowledge that we would normally attribute to an adult speaker who used the same term in the same context. These preliminaries indicate the problems encountered in this type of analysis and serve as a cautionary guide to the interpretations presented here. Analysis of spontaneous speech is necessary to identify phenomena of interest in language use, but it will need to be followed by more systematic verification methods.

*This learning factor probably accounts for some of the discrepancy in the frequency use of object labels (see Bloom, 1973; Nelson, 1973). Children, when first learning (or any other) term tend to repeat it endlessly, but once it is practiced thoroughly, it out of use, if it is not of continuing importance to the child, while remaining in the vocabulary and available as a label when needed. Much the same thing happens later color terms are acquired, for example.

of early an object may drop receptive on, when

16

Katherine Nelson

Sampling Method The data for this analysis were derived from speech samples of one or two hours in length taken from 18 children at 24 months and from 16 of the same children at 30 months (the data are described in more detail in Nelson, 1973). For the present analysis, the first 100 utterances from each of 24 of the resulting transcripts were used. Some of the children whose speech will be analyzed here were earlier identified as primarily Expressive (E) rather than Referential (R) speakers, on the basis of the words that they learned and used at the one-word level (see Nelson, 1973 for details). R speakers learned predominately object-words. seeming to indicate an interest in things and their categorization, while E speakers used more social-personal forms. A question of interest in the present analysis was whether the differences in speech patterns between E and R children would persist at more advanced levels. To the extent that adjective use is related to categorization of objects it might be expected that R children, whose earliest language was more object-oriented and thus appeared to be related more directly to the child’s categorization of the world, would tend to use more adjectives than E children, or to use them in a more advanced way. The average age at which the identification of R versus E speech types was made for each child (based on the percent of object words in the first 50 words learned) was 19 months, 5 months prior to the 24-month speech samples and 11 months prior to the 30month samples used here. Transcripts were selected on the basis of the mean length of utterance (MLU), age, sex and vocabulary group (R or E) of the child. MLU was derived from the entire speech sample for each child and was based on Slobin’s (1967) rules. Seven MLU levels with a range of 0.5 morphemes were defined from 1.0 to 4.5. For most analyses, the levels were collapsed to two groups: Low proficiency (MLU = 1.0 - 2.5) and Hi& proficiency (MLU = 2.5 - 4.5). These cover roughly the. same levels as from less than Stage I to Stage II for the low group, and Stage III to V for the high group in terms of Brown’s (1973) Stages. The low group here may then be considered roughly equivalent to Stage I speech, for which there is a good deal of comparable data, although the range in this study is broader. Six transcripts from each MLU group (High or Low) and each vocabulary type (R or E) were analyzed. To the extent possible, sex of the child was balanced within each sub-group. Transcripts were analyzed from 10 boys’ and 14 girls’ records. Some but not all of the 18 children were represented at both ages, while a few of the total group of children were not represented at all. Age could not be separated from proficiency in this

Some attributes ofadjectives used bJ1young children

1I

sample; all of the High-E group were 30 months of age and all of the LowE group were 24 months, while only two of the younger R group were in the High level and two of the older were selected at the Low level. Thus, we cannot speculate to what extent age (and experience) rather than language or conceptual structure may account for the differences between High and Low proficiency speakers. The distribution across MLU levels was kept as balanced and evenly distributed as was possible within the total sample. Each of the seven levels was represented at least once for each vocabulary group and only one (R 2.0 - 2.5) was represented more than twice (3 times). The 2400 utterances selected in this manner (both single and multimorpheme) were analyzed in terms of noun-verb relations, presence of modifiers, pronouns, quantifiers, distinguishers and locatives. Aspects of the data other than modifier use will be reported eslewhere (Nelson, 1975). Of the total sample of utterances, 13.8% contained a nominal modifier; that is, a modifier was contained in every 7.21 utterances, on the average, for a total of 333 cases. As noted above, a modifier for this purpose might be a possessive noun or pronoun as well as an adjective. Terms that were used only in response to the mother’s elicitation (e.g., “what color is that?“) were not included in the final analysis.

Analysis and Discussion Several different analyses were undertaken; description of the method, results and discussion will be presented for each in turn. At the most general level, Table 1 presents the relation between language status and use of modifiers, in terms of the correlations of MLU (7 levels) with various’indices. It is surprising to note that there are few strong relations of modifier use with speech maturity apparent in this table. Although total number of adjectives and possessives show a positive relationship, they do not reach statistical significance. The number of different adjective and possessive types are, however, significantly related to MLU level; that is, with increasing maturity a greater variety of forms is used. Neither the type-token ratios nor the ratio of modifiers to nouns or to each other show any overall relation to MLU status, howev.er. This indicates that these classes occur in the same proportions with development - each increasing in use to the same degree. Differences within the class of modifiers were apparent, however, as shown below.

18

Katherine Nelson

Table 1

Correlation

of lnodificr

Total

number

of adtectivcs

Total

number

of possessives used

0.219

Total

number

of modifiers

0.35

of adjective

Number

of possessive types type-token

types

Adjcctive~-noun Modifiers-noun

correlations.

ad

0.245 --0.006

ratio

levels of MLU

Forms

ratio

0.1 09

ratio

*Seven

1

0.4 1o** --0.018

ratio

moment

0.335

0.436**

ratio

Possessive type-token Possessive-noun

used used

Number Adjective

Adjective

use variables with MLU level*

0. I29 from **p

I .Oto 4.5

were used. Correlations

are Parson

product

< 0.05.

I*‘urlctiorls

A basic distinction among adjective constructions is that between predicate adjectives and attributive or prenominal forms. Each of the examples above (1 - 3) involve attributive adjectives, that is, those appearing in prenominal position. The predicate forms (e.g., ‘Elizabeth is beautiful’) appear usually to be descriptive of an entity or comments on a topic. Such forms can be found even as single word utterances, for example, ‘broken’. In this case, some object which is understood from the prior linguistic or non-linguistic context is the topic and the comment refers to a property or state of the thing. Attributive forms precede the noun, as in ‘the big truck’, regardless of what role the noun phrase plays in the total utterance. It is obvious therefore that single-word modifiers can only be a reduced form of predicates and cannot be attributives, because these require the presence of a modified noun. Attributives have generally been analyzed as derived from the predicate adjective construction, thus involving an embedded sentence of the form ‘the truck is big ’ in the deep structure of ‘the big truck’. Several authors (for example, Bolinger, 1967 and Vendler, 1968) have shown, however, that for many adjectives this derivation will not work. Bolinger claims that the attributive, not the predicate, is the basic form, while Vendler has analyzed a number of different types of adjectives that are derived from different underlying predications. Brown (1973) has noted that attributives appear earlier than predicate adjectives in children’s sentences and therefore questions the developmental reality of the view that the former are derived from the latter. Although the question of the

Some attributes of adjectives used by young children

19

correct grammatical derivation is not of direct concern here, the related distinctions that have, been made in regard to different types of adjectives, as revealed in these linguistic analyses, are of interest. In the present analysis, all adjectives appearing in the pre-nominal position were classed as attributives, while all others, whether used alone (e.g., ‘broken’) or in a predicate phrase (e.g., ‘it’s broken’) were classed as predicates. This distinction relies on word order in child utterances in the same way that it does for adult utterances. Of the predicate adjectives, less than half (44%) were used in single word constructions. In this sample, the correlations of MLU with total number of attributive adjectives was positive and significant (Y = 0.54, p < 0.01) while the number of predicate adjectives was negatively related to MLU (Y = -0.375, n.s.). Further, the correlation of MLU with the proportion of all adjectives that occur in the attributive position was highly significant (Y = 0.667, p < 0.001). Table 2 s1rows that attributive modifiers occur more frequently overall, and this finding alone would tend to support Brown’s (1973) statement of the sequence of emergence of these forms. However, as indicated by the high correlation of MLU with proportion of attributives, attributives increase in relative frequency with development, regardless of language type, while predicates decrease. The two types are of approximately equal frequency in the Low samples, but attributives are about three times as frequent in the more mature samples. These facts tend to contradict the claim of attributive priority, although they might be explicable in terms of the prevalence of one-word utterances in the Low samples, which by definition are predicate types. However, while more of the predicate adjectives used by Low speakers were one-word utterances (62%) than those of the High speakers (21%), there was a proportional decline in use of predicate adjectives with language status whether all adjectives (including single words) were considered (from 62% to 27%) or only multi-word utterances were included (from 38% to 23%). While it appears from Table 2 that R speakers used more adjectives at the High levels than did E speakers, this comparison was not significant (Language Type main effect and Interaction with MLU were both greater than p = 0.15 in an Analysis of Variance on this variable). The MLU main effect was only marginally signficant (F = 3.649, p = 0.07). These are reflections of the fact that the inter-child variability in number of adjectives used was very great. Proportional analyses of adjective types were therefore deemed appropriate. It may also be noted in Table 2 that both the absolute numbers and the proportions of the two modifier types are very similar for the two types of language learners (Referential and Expressive). An analysis of variance

20

Katherine Nelson

Table 2.

Number and percent of attributive and predicate adjectives by language group. Language

Mean number of Mean number of Mean number of Percent predicate Percent attributive

adjectives predicates attributives adjectives adjectives

group

R-Lo

R-Hi

E-Lo

E-Hi

Total

6.17 4.00 2.17 66.7 33.3

13.33 2.83 10.33 23.3 75.3

6.33 3.17 3.17 56.6 43.4

7.50 1.50 6.00 13.7 86.3

8.33 2.88 5.42 40.1 59.6

on the proportion of attributive adjectives showed a significant main effect for MLU status (F,,20 = 13.299, p < 0.002) but no effect for language type or interaction of language type and MLU status (F < 1.OO). Clearly, while the number of predicate modifiers remains fairly stable with language development, the number of attributives increases. sharply. As noted above, predicate adjectives can be considered comments on a topic, whereas attributives may be considered to specify the referent or the referent class more completely. The predicate adjectives are apparently always descriptive of the referent in the traditional sense in which the adjective describes a person, place or thing. Attributives, even when used descriptively [see (1) above], perform a role subordinate to the modified noun. Bolinger (1967) has noted that attributives may specify something about the referent [see (2) above] or they may specify something about the reference class (3), in which case they serve as subclassifiers or types. He notes that, while the former can be derived from predicate adjectives, the latter cannot be. If an attributive modifies a particular referent, it can be easily related to the predicate adjective both in terms of its construction and in terms of its function. One can imagine a child saying, for example, ‘That is a bear. That bear is big. That is a big bear.’ Clearly, ‘big’ here refers to the referent, this particular bear, which is big for bears, or perhaps, big for pictures, or big for toys, if these constitute the immediate perceptual array. Consider, however, what looks like an equivalent sentence: ‘That’s a teddy bear’. Here the adjective serves to subclassify the referent into a new reference class. The adjective modifies the reference class, not the particular referent. One cannot say ‘That bear is (a) teddy’, as Bolinger points out. The general class term must be added: ‘That bear is a teddy bear’. Here the referent is not being described or specified, but identified by class or classified.

Some attributes of adjectives used by young children

Table 3.

21

Percent of referent and reference class modification by language group

Percent attributive referent Percent attributive class Total percent referent (Attributive and predicate)

R-Lo

R-Hi

E-Lo

E-Hi

5.0 28.3 71.7

21.8 48.5 51.1

13.0 30.4 69.6

10.1 16.2 23.8

In the adult language this distinction between referent and reference class modification can be made successfully according to type of adjective. Here, however, one of the issues of interest is what types of adjectives are used by young children for these different purposes. It is not always easy to determine which of these relations is being employed by the beginning talker. The young child may use size, for example, to subdivide a larger class of, say, dogs. Thus ‘big dog’ may be used as a classification as well as a comment on the size of a particular referent. Some uses of this kind are standard, for example, ‘little girl’. For this analysis, the context of the utterance, including emphasis and the preceding and following conversation, was used to determine whether the child appeared to be emphasizing the referent (e.g., ‘Where’s the big truck?‘) or utilizing the adjective to specify a sub-category (e.g., ‘Give me a big stick’). As this example shows, the preceding definite and indefinite articles are also indications of referent and class use, respectively. In addition, apparent intention and paraphrase were used to disambiguate the category. All classifications were made prior to further analysis. The results of this classification are shown in Table 3. An analysis of variance on the proportion of adjectives used as classifiers showed a significant main effect of MLU level (F,,20 = 6.92, p = 0.02) but neither the Language Type nor the interaction effects were significant. The distribution between the two types of attribution (reference classification and referent description) by language status clarifies the difference between the more advanced and less advanced speakers that was apparent in their differential use of predication and attribution as shown in Table 2. The difference in relative adjective functions as language develops is shown in summary form in Figure 1. Adjective

types

The referent-reference the types of words

class distinction used in different

just discussed constructions.

is apparent also in Consider first the

22

Katherine Nelson

Figure 1.

Comparative use of different 0

i

0’

Pred. Ref. Att

n

Class

of adjectives

at two MLU levels

Att

70 -

a

60

-

;

50

-

k k =

40

-

30

-

20

L

z w a

m

types

LOW

MLU

HIGH

MLU

predicatives. Each adjective was categorized into the following types (all words included in the category are listed in parentilcses): Descriptive properties of the object (big, bigger, white, colored); Evaluative terms (smarter, nice, no good); Transient object states (cold, hot, missing, dirty, all gone, whole, broke(n), tix(ed), open, on (sound), off (sound), stuck); and Animate states (hurt, dead, stinks, finished, sleepy). When crossclassified by MLU status the distribution according to these categories is shown in Table 4 with States collapsed into one category. (Because of the small numbers involved in some categories and because the proportions were similar for R and E speakers, the data for these distributions were grouped together.) It is obvious from Table 4 that the vast majority of predicate adjectives were used to comment on object or animate states, and that only the more advanced speakers use this form to comment on descriptive properties of objects, such as size or color. Consider now the distribution for attributive adjectives used for referents also shown in Table 4. Here the categories involved are: Descriptive properties (big, little, baby); Evaluative (bad, yummy, pretty, different); Object and animate states (new, bare, broken, whole). The small numbers for this group make the data difficult to interpret, particularly for the less advanced talkers. It appears, however, that more object properties are used in this position for description or specification of the referent than are used in the predicate form.

Some attributes

Table 4.

Number

of words

(percent

of total in paren theses)

Descriptive Properties Predicate

type

used by young

used as referent-specijjing

I
Object and Animate States

I (2.3) 3 (11.5) 4 (5.8)

42 (97.7) 14 (53.8) 56 (81.1)

2 (22.2) 5 (18.5) 7 (19.4)

4 (44.4) 4 (14.8) 8 (22.2)

children

23

adjectives

Adjectives

Low MLU High MLU Total Attributive-Referent

0 (0) 9 (34.6) 9 (13) Adjectives

Low MLU High MLU Total

3 (33.3) 18 (67.0) 21 (58.3)

Table 5.

Number (percent

Low MLU High MLU Total

of each

of adjectives

of’ words

of

each

type

used

as re,fkence-class

attributives

of total in paren theses)

Descriptive Properties

Evaluative

states

Human

15 (65.2) 37 (54.4) 52 (57.1)

0 (0) l(1.4) l(l.l)

2 (8.7) 0 (0) 2 (2.2)

l(4.3) 20 (29.4) 21 (23.1)

Types

Inanimate

Types

5 (21.7) 10 (14.7) 15 (16.5)

Finally, consider the attributives used to classify, that is to sub-divide the reference class. For this purpose, two new categories were necessary, specifying independent concepts, human or inanimate, which acting alone are generally nouns. This classification yielded the following with their instances: Descriptive properties (big, little, white, green, purple, red, brown, yellow, rubber); Evaluative (plain); Object states (new); Human concepts or types (Daddy, Mommy, Baby, Indian, Mother); Inanimate concepts or types (Christmas, teddy, pick-up, tree, dump, chocolate, Halloween, panda, school, steamer, spider, doll). They were distributed as shown in Table 5. Here we see that the most useful words for classifying for these children are the descriptive properties, mainly size and color, and for the younger group these are almost entirely size (big and little) terms. Next in frequency are the ‘human types’ which are used sometimes interchangeably with

24

Katherirw Nclso~~

the size terms in reference particularly to animals (e.g., ‘baby moose’). It is interesting, however, that these do not appear in the earlier or less advanced protocols - size terms seem to prccctfc the human analogs in the functional vocabularies of young children, And, finally, there are the type words with which the child will increasingly attain mastery over differentiated conceptual domains, distinguishing here for example, not just big trucks, but dump trucks, steamer trucks and pick-up trucks; panda bears and teddy bears; Halloween candy and Christmas trees. Even the less advanced talkers have by two years mastered some of these type distinctions. In summary, the most striking fact about Tables 4 and 5 is the finding that adjectives of different lexical meaning are used in different constructions (predicates and attributives) for different functional purposes (description and classification). A Chi-square analysis for the 3 different types of adjectives grouped into three Categories - Descriptive and Evaluative, States, and Types -- is highly significant (x2 = 112.83, #‘= 4, p < 0.001). These findings can be understood by noting that predications are used by beginning talkers to comment on the perceptible world around them.* What is worth commenting on at this point is obviously the transient states of objects. States of objects are not, however, useful for subcategorizing classes. For this, internal properties that distinguish one class from another are employed - size and color, or, and increasingly, a defined tJ*pc based on function or use. Therefore, as sub-categorizing and specifying become important, more attributives and fewer predicates are used.

An analysis of the types of nominals that are modified by these adjectives reveals some related facts. For this purpose, size adjectives (40 instances of ‘big’ and 21 ‘little’), and inanimate states (41 object states and 19 animate states) and human (2 1) and object (16) types were classified according to whether the nominal form in construction with them was included as the subject of a predicate adjective construction or the head of a modifier-noun phrase, or was omitted from the construction entirely. This analysis distinguished between the existence or non-existence of a nominal and its form, noun or pronoun. *Only ZIscattering comment

of true informative predications of one child, “baby mooses don’t have horns.”

are found

in these

records,

such

as the

Some attributes ofadjectives used by children

25

The great majority of the size adjectives (47/6 1 or 77%) were used in a construction with the modified noun. Only seven (1 1%) were used alone, (e.g., ‘big’) and of these, several were used in a construction in which the size term itself became nominalized, e.g., ‘a big’, and in one case, ‘a little little’. These latter anomalous cases appear to underscore the primarily classifying usage of the size terms, that is, their use in ways similar to that of the type terms, which can be used appropriately in both adjectival and nominal forms. As an incidental observation, it is interesting that both ‘big’ and ‘little’ were used to modify animal terms, but ‘little’ monopolized people terms (‘little girls’, ‘little boys’, ‘little kids’) with only one ‘big boys’, On the other hand, ‘big’ was used far more frequently than ‘little’ to modify physical objects (16 cases to 2), including airplanes, cars, trucks, houses, and also relatively small objects such as shovel, ball. book, and stick. This aspect of the distribution might reflect sampling characteristics, but it is intriguin g to speculate that children might use ‘little’ as a marked term for people (generally big) and ‘big’ as a marked term for objects (generally little). All of the Inanimate Type words were used in construction with their modified nouns. Of the 20 Human Type words two were used with the proform “one” where the noun had been previously specified, as in ‘the baby one’. All the other Human Type words were used with their nouns, and these were either human (‘baby boy’) or animal (‘baby deer’) or clearly related inanimate forms (doll , ghost). Thus the use of human analogs for the general size terms appears to have a well-understood animate implication for these young children. At the other extreme in terms of nominal use from the Type words are the State words. Of the 60 instances of these words, 34 (57%) were used without a nominal form of any kind in the construction, while 16 (27%) were used with the pronouns ‘it’ and ‘that’ and only 10 (17%) were used with an accompanying noun (including the nonspecific ‘thing’). Thus 83% of these terms were used without a specific referent in the sentence. Obviously, utterances of this kind can only be understood if the referent is clear from the prior linguistic or non-linguistic context, but this would usually be true if the child were commenting on a perceptually present referent. Although state terms are relatively more common in the speech of the less advanced speakers, as shown above, it was not the case that adjective-noun constructions became more common for these terms as language became more advanced. Five adjective-noun constructions were found at each developmental level (High or Low), and the only occasion out of 28 instances in which ‘broken’ occurred with a noun (‘broken light’) was found in the record of one of the Level 1 (1.0 - 1.5

26

Katherine Nelsorl

MLU) speakers. Thus the use of state terms by themselves was not only a function of the length of the constructions in which they were used. These differences in the degree of specification of the modified referent have thus been found to vary with the type of adjective used, just as the syntactic form and semantic function did. It has been seen that the state terms are used to comment on a readily apparent object, and for this purpose the name of the object need not be specified, while the type and descriptive terms are used to classify and to specify among objects, and for this purpose the name of the object to be specified must be included.

Possessives.

Consider now the use of possessives. Possession. whether alienable or inalienable (cf., Brown, 1973) has to do with distinguishing a particular referent, specifying for example, Mommy’s nose, rather than distinguishing among red noses and white noses as sub-classes of noses. The Possessive specifies a relation between a particular person and a particular thing possessed. Although we may speak loosely of the class of things that are mine, this is a heterogeneous category defined only by the relation ‘belonging to me’. Possessives may be used in the predicate position (and some of the children at even the lowest levels distinguished correctly between ‘mine’ and ‘my’ for these purposes) but they were used more frequently in the attributive position (91/l 11 or 82%’ of the time). In this they differ from the state terms which are used also as referent modifiers, but which are used primarily as predicates. Rather than describing, possessives specify among referents. as the size terms frequently do: they distinguish among particulars within a category or an array. Possession was used far more frequently for this function than any other kind of modifier (compare the percentage above with Tables 4 and 5). Table 6 shows the relative use of possessives, adjectives, and nouns by the children in language sub-groups. It appears from this Table that Expressive children used more possessive terms than the Referential group of children did, at least for the more advanced speakers, both in absolute numbers and in terms of the number of possessives in relation to the number of adjectives. Neither of these comparisons reaches an acceptable level of significance, however, because of small numbers and high between-S variability (the R-High vs. E-High comparison of PossessiveAdjective Ratios is of borderline significance, 0.05 < p < 0.10). If the direction of the difference is meaningful, however, it would indicate that the Expressive children were less oriented to properties of objects and

Some attributes of adjectives used by young children

Table 6.

Possessive-adjective

Mean number of possessives Possessive-adjective ratio Adjective-noun ratio __-

27

relations by language group R-Lo

R-Hi

E-Lo

E-Hi

Total

4.50 0.786 0.117

2.83 0.209 0.297

4.17 0.753 0.148

7.33 1.052 0.116

4.71 0.700 0.169

more oriented to relations with others, essentially what the original classification of the two groups was assumed to reflect. This interpretation is consistent with the interaction of language type by language development for adjective-noun (A-N) ratios shown also in Table 6. An Analysis of Variance of this ratio showed that the Interaction of Language type with MLU was significant at the 0.05 level (F, 20 = 4.361), although neither main effect was significant. The A-N ratio is lower for R speakers than E speakers at the Low level, primarily because more nouns were used by R speakers at this level. At the High level, the numbers of nouns used are approximately equal for both groups; however, the R speakers but not the E speakers have increased the number of adjectives used in relation to the number of nouns. This result provides some evidence then that R speakers are conceptually more advanced in terms of the object-property relations indexed by the adjective-noun ratio. One child characteristic that is relevant to this observation is that five of the six transcripts in the E-High group came from children with siblings, while only two of those in the R-High group did. It seems obvious that in the context where there is a potential dispute over possession among peers, possession as a dimension along which to distinguish among things will be far more salient than other possible dimensions. It should be noted, however, that the same functions are being served by the different types of modifiers - namely, to make distinctions among objects and classes of objects. Thus the E speakers may not be less conceptually advanced, only different in regard to the relative salience of properties and relations. The fact that E speakers actually used a greater proportion of adjectives as classifiers than did R speakers (see Table 3) tends to mute the importance of the possible cognitive difference. On the whole, the similarities between the speakers in regard to the variables considered here are more notable than the differences. Although different word types are used by different speakers, they are used for similar purposes of distinguishing among objects and they are used in similar constructions.

28

Katherine Nelson

Conclusions By

the spontaneous of adjectives possessives by who are developing language, have been to distinamong three of adjective and have that difkinds of are used these functions, that they used with frequency by at different levels. To The first of modifier is the adjective of type ‘it’s which is in reduced (‘broken’) even the onelevel. These predominate at levels of than 2.5. of the used in form are describing a state of object or They are that is, as descriptive to talk a within-object The focus on the referent without to others its class. referring both the particular and to class tend develop later predominate above MLU level 2.5. In cases, the is used as a of classification, subdivide a class or, the case the referent to make temporary distinction members in class or likely in perceptual array. these purposes, properties are primarily terms; and, classification purposes, terms may themselves nominals signifying concepts used as This conceptual as reflected use of forms is sistent with notion that child begins focusing on objects and changes in possible states relationships. At early point, child does focus on properties of nor on properties that useful for one member another, such color or but on characteristics and Moving beyond stage, the begins to his original classes by Type terms to distinguish particulars within class by descriptive property In both these uses, terms and little and their type analogs, a major at first. a few terms were for the of distinguishing things, they used far frequently than teams. Mother often engaged a teaching with color asking ‘what is that?’ as a most of children knew color terms, they used infrequently in speech. Size apparently a mode of among particular at two

Some attributes of adjectives used by young children

29

terms were used by children for purposes. likely, however, the permanent function will taken over and more type terms the child conceptually, size and terms will to be to distintemporarily among referents. The or classification is in a way naming new through wordHowever, at early stage, few general are used a similar in a non-specific way. relational properof the terms are being utilized distinguish big from little big trucks little ones as indicated the data, people from people. In similar way, child divides the world the basis the relation things to and others. are primitive of classification the concrete 2 they not permanent; they will increasingly replaced sub-categorized permanent as these names. On other hand, appears to only a awareness at level of potential usefulness grading or Apart from and ‘little’ one case ‘on’ and (for sound on and no contrasts Although ‘hot’ ‘cold’ were used, they not used the same Negatives applied the adjectives form a were rare exception: ‘That not colored’). observations, of are limited the nature the evidence; deserve special study. ‘There some indication those children focussed on early in language development (the Referential developed more modification schemes 2% years, indexed by possessive-adjective ratios the adjective-noun while the children appeared rely more on possessive tions to among objects. the overall from comments states of to distinguishing and sub-classifying was similar both groups. summary, by years, as have seen, have developed following possible for modifiers: Comment on states of by using stative verb predicate adjective. Distinguish among according to or color. Classify objects to possession use. 4. by combining another concept form a Even at earliest stages speech the makes use the first functions, and early adds fourth. Together enable the to express in the his increasingly concepts.

It

30

Katherine Nelson

References M (1974) Cognitive functions of language in the preschool years. Develop. Psychol., 10, 229-245. Bloom, L. (1973) One word af,a time. The Hague, Mouton. Bolinger, D. (1967) Adjectives in English: Attribution and predication. Lingua, 18, l-34. Brown, R. (1973) A first language: the early stages. Cambridge, Harvard University Press. Clark, E. V. (1973) What’s in a word? On the child’s acquisition of semantics in his first language. In T. E. Moore (Ed.), Cognitive development and the acquisition of language. New York, Academic Press. Donaldson, M. & Wales, R. J. (1970) On the acquisition of some relational terms. In J. R. lfayes (Ed.), Cognition and the development of language. New York, Wiley. Fodor, J. (1972) Some reflections on L. S. Vygotsky’s Thought and Language. Cog. I, 83-95. Nelson, K. (1973) Structure and strategy in learning to talk. Sot. Res. Child Devel. Monog., 38, (l-2, serial no. 149). Nelson, K. (1975) The nominal shift in early child language. Cog. Psycho/., 7, 461-479. Sinclair-de-Zwart, H. (1969) Developmental psycholinguistics. In D. Elkind and J. H. llavell (Eds.), Studies in cognitive development. New York’ Oxford University Press. Slobin, D. I. (1967) A field manual for cross-cultural study of the acquisition of communicative competence. University of California, Berkeley. Slobin, D. I. (1968) Imitation and grammatical development in children. In N. S. Endler, L. R. Boulten & H. Osser (Eds.), Contemporary issues in developmental psychology. New York, Holt, Rinehart and Winston. Vendler, 2. (1968) Adjectives and nominalizations. The Hague, Mouton. Weir, R. H. (1962) Language in the crib. The Hague, Mouton. Blank,

Resume L’analyse de 24 ichantillons du langage spontand d’enfants de 24 i 30 mois montrc une progression correlee dans la forme, la fonction et la signification des modificateurs utilisis. Lcs adjectifs predicats sont utilises pour exprimer les transitions dans les &tats des objets. 11s se rencontrent csscntiellemcnt dans les Cchantillons precoccs. Les adjectifs attributs sont utilises plus tard. 11s expriment les subdivisions interclasse ou ils specifient des elements i I’intdrieur d’unc m&me classe. Dans ce cas, ils caracterisent I’objet selon des proprietes physiques, telles que la taille, par I’utilisation d’un nom conventionnel ou d’un possessif. L’utilisation differenciee d’une forme particuliere semble davantage like i son utilite cognitive ou communicative qu’i sa derivation syntaxique.

Co~nitior?, 4 (1976) 31-43 0 Elsevicr Sequoia S.A., Lausanne

2 - Printed

in the Netherlands

Ocular motility and cognitive process*e **

SUSAN L. WEINER Educational

Testing Service

HOWARD EHRLICHMAN City University

of New York

Abstract

This study investigates the amount of ocular motility occurring in response to questions varying the kind of cognitive process required for answer. Fewer eye movements occur in response to questions designed to elicit visuo-spatial as compared to verbal-conceptual processes, a finding consistent with our earlier observation of more Stares’ occurring with the former than wifh the latter questions. The results are inconsistent with the traditional hypothesis that visual imagery involves an increase in scanning eJ)e movements. The findings are interpreted in ferlns of a model postulating an interaction of the form of visual information processing and the type of cognitive activity subjects engage in. Eye movements of the scanning or saccadic type have been studied in a variety of contexts. For example, such eye movements have frequently been used as guides to how and what people take in from their visual environment (e.g., Mackworth & Morandi, 1967; Yarbus, 1967). Eye movement patterns have also been studied in social contexts, particularly in dyadic interactions, in order to.provide information about a number of social variables involving the visual presence of another person, e.g., mutual attraction, deference, the taking or yielding of the floor in conversation, etc. (Argyle, Lalljee & Cook, 1968; Exline, 197 1; Kendon, 1967). Common observation suggests, in addition, that people may move their eyes a great deal in stable and familiar visual environments, where there may *This research was carried out while the authors were postdoctoral fellows supported jointly by NIH Grant MH08260 and Educational Testing Service. **The authors thank Donald R. Goodenough, David Owen and William C. Ward for their comments on earlier drafts of this manuscript.

32

Susan L. Weiner and Howard Ehrlichman

be no new visual information or where social interaction through eye-contact is not possible. For example, people often seem to move their eyes rapidly while on the telephone at home, an environment which is both visually familiar and devoid of the social cues provided by the visual presence of another person. Thus, it would appear that such eye movements may be related to internal cognitive or affective processes. A number of hypotheses have linked eye movements to cognitive activity. The most prominent of these has attempted to relate eye movements to the presence of visual imagery. The claim has been made that eye movements are a motoric manifestation of imaginal thought on the assumption that imagery (or an ‘internal visual system’) is analogous to the actual visual system (Hebb, 1968; Totten, 1935). Tests of the imagery-scanning hypothesis have produced equivocal results. Studies investigating the similarity in direction of eye movements during imagery to those which have occurred or might occur during the actual viewing of stimuli have either failed to produce clear isomorphic relations (Brown, 1968; Marks, 1972; Moskowitz & Berger, 1969) or have had limited success in cases in which rhythmic movement is the salient feature of the visual scene (Deckert, 1964; Zikmund, 1966). Further, the claim that increases in ocular motility (as opposed to ocular quiescence) reflect the scanning of imagery has been seriously questioned at least in dreaming (Jacobs, Feldman, & Bender, 1972; Koulack, 1972). In fact, relatively low levels of ocular motility have been associated with awake subjects’ reports of visual imagery (Antrobus, 1973). Thus, while there may indeed be a relation between ocular motility and visual imagery, the relations proposed by an imagery-scanning notion have received weak support. Other hypotheses attempting to relate eye movements and internal processes have relied on a variety of cognitive explanations other than (Amadeo & Shagass, 1963) and limited ‘channel imagery, e.g., attention capacity’ (Antrobus, 1973; Singer, Greenberg & Antrobus, 1971). These interpretations have come from studies in which mental activities other than visual imagery have been shown to be related to varying levels of ocular motility. For example, subjects showed greater ocular motility when they were asked to ‘suppress’ or put a secret wish out of mind than when they were asked to generate and ‘indulge’ such a wish (Antrobus, Antrobus, & Singer, 1964). The solving of arithmetic problems also produced greater ocular motility than did instructing subjects to let their minds ‘go blank’ (e.g., Lorens & Darrow, 1962). A number of factors have tended to restrict the generality of research on ocular motility and internal processes. First, the cognitive tasks which have been employed often have had ambiguous behavioral manifestations, forcing

Ocular motility and cognitive process

33

experimenters to rely on subjects’ self-reports for verification that the intended cognitive activity occurred. Second, ocular motility during task performance frequently has been compared with ‘baseline’ periods, where no specific cognitive activity was required. Since cognitive activity does not stop during such baseline periods but is simply unspecified, the meaning of such comparisons is ambiguous. Third, both the tasks and the physical situations have usually had little resemblance to everyday problem solving and memory activities. The present study attempts to avoid these limitations. In an investigation of the direction of the subject’s initial eye movement away from a fixation point at the end of an interviewer’s question (the ‘lateral eye movement’ phenomenon), we found that questions requiring different varieties of cognitive process for their solution were systematically related to the presence or absence of this first eye movement (Ehrlichman, Weiner, & Baker, 1974). Specifically, subjects were more likely to maintain continuous fixation (i.e., not to make any observable eye movement) in response to questions designed to elicit visuo-spatial processes as compared to questions designed to elicit verbal-conceptual processes. This finding has since been replicated by Galin and Ornstein, 1974. While it would be tempting to interpret this finding as evidence that different cognitive processes elicited during problem solving and memory tasks may systematically affect ocular motility, such a conclusion is not yet warranted. It is possible that the maintaining or breaking of fixation (often involving eye-contact) is a phenomenon in its own right not influenced by the same factors affecting later eye movements (for example, when serving an interpersonal cueing function, Kendon, 1967). From a different perspective, Galin and Ornstein (1974) seem to opt for such a view by claiming that ‘stares’ (cases where no initial eye movement occurs) reflect a balance or symmetry of hemispheric activation which prevents initial eye movements from occurring. We tested these alternatives by going back to the videotapes which were available’ from our earlier study (from which only the first eye movements had been scored) and counting the number of eye movements which occurred subsequent to this first movement. If it is true that these initial eye movements are unique events, there would be no reason to expect differences between verbal and spatial questions in the amount of ocular movement after the subject has looked away from fixation. However, if the presence or absence of initial eye movements reflects variations in ocular motility, one would expect fewer eye movements for spatial than for verbal questions after the subject has shifted away from fixation. Such a finding would demonstrate that changes in ocular motility are systematicallyrelated to varieties of everyday problem solving and memory activities. In addition,

34

Susan L. Weiner and Howard Ehrlichmm

it would provide further evidence against the imagery-scanning an explanation for such changes in level of ocular motility.

hypothesis

as

Method Subjects Subjects selected 1974.

were 15 right-handed adults, 8 males and 7 females, randomly from the sample in Experiment I reported in Ehrlichman et al.,

Design Each subject was asked 80 questions: 40 verbal and 40 spatial questions. (Forty neutral questions not relevant to the present study were also presented.) Each set of 40 questions was divided into two blocks of 20 questions each. These blocks were arranged in two orders: (a) verbal, spatial, spatial, verbal, or (b) spatial, verbal, verbal, spatial. Eight subjects received block order (a) and seven subjects received block order (b). The blocks of 20 verbal and 20 spatial questions were arranged in four random schedules and were balanced across block orders (a) and (b). A male voice and a female voice recorded duplicate orders of the questions, which were balanced within the sex of subjects. Questions* Verbal questions were designed to elicit a variety of linguistic and conceptual skills. For example, subjects had to supply particular words for which a definition was provided, e.g., a rule which guides conduct, a narrow cylinder with lenses at both ends or to define words which were given, e.g., charity, licorice. Subjects were also asked to interpret certain common proverbs, e.g., Too many cooks spoil the broth. In addition, judgments of semantically or grammatically malformed sentences were required, e.g., The man is six feet short, and some three-term series problems were presented, e.g., Jim is better than Ted; Ted is better than Bob; who is best? For spatial questions, an attempt was made to sample different types of visuo-spatial processes; for example, different kinds of imagery: ‘Try to form *The Service,

complete Princeton,

list of questions can New Jersey 08540.

be found

in RB-73-21

available

from

Educational

Testing

Ocular motility and cognitive process

35

a mental picture of what 1 am going to tell you and tell me when the picture is as clear as you can get it: A poodle; ‘What color is the top stripe of the American flag?; and ‘Describe the route by which you came to ETS today’; ‘If you cut a paper cylinder and lay it out flat, what shape is the resulting figure?’ Questions were also included which seemed to involve operations on spatial or visual representations, e.g., ‘How many windows are there in your house or apartment?’ ‘If you are the minister at a wedding, on which side of you does the bride stand?’ ‘How many angles are there in a Red Cross symbol?’ ‘What other letters do you get by rotating or flipping a lower case printed p?’ Procedure So as not to influence subjects’ eye movement behavior, automated interviewing procedures were used. The questions were prerecorded on audio tape, and subjects’ responses were video taped and monitored by the experimenter in another room. The subjects were tested in a small well-lighted symmetrically arranged room in which they sat at a table facing a video camera approximately 4 feet away. The camera was covered with a uniformly gray cardboard shield so that only the lens was visible to the subject. Subjects wore a lavalier microphone. The subject’s placement vis d vis the camera enabled a clear picture of his/her face to be recorded during the experimental session. Subjects were asked not to move their bodies to either side of the camera or to shift around in order to stay within camera range. They were informed generally of the kinds of questions they would be asked and were further instructed that the experimenter would say, ‘Ready’, before each question as a signal to look into the camera until the experimenter finished asking the question. They were told that this procedure provided some assurance that they were paying attention to the question, since no one would be in the room. The experimenter also emphasized that they need not look into the camera at any other time. Subjects were tested individually for about 40 minutes, which included a five-minute break after three of the six blocks of questions. The questions were self-paced by the subject. A post-experiment interview indicated that no subject suspected that his/her eye movements were being studied. From the resulting video tapes, two experimenters independently counted the number of eye movements that occurred during the interval between the last word of the question and the last word of the subject’s answer by creating with pencil and paper an analog of the direction and frequency of the subject’s shifts in gaze, if any occurred. Experimenters viewed the tapes

36

Susan L. Weiner and Howard Ehrlichman

until their counts were within five eye movements on a given trial. The mean of their two scores was used. In addition, the duration of the interval in centiseconds from the end of the question to the end of the subject’s answer was timed separately by two experimenters until agreement was reached within 30 centiseconds. The reliability of the counting procedure was assessed by having a third experimenter, blind to the purposes of the study, count the number of eye movements on each trial in which eye movements occurred for five randomly selected subjects. These counts were correlated with the mean counts of the two experimenters for spatial and verbal questions. The correlation for spatial questions was 0.86 and for verbal questions was 0.96.

Results Subjects’ ocular behavior was divided into trials in which eye movements occurred and sture trials, in which no observable eye movement occurred. The data are presented this way in Table 1 in order to analyze independently two potentially different ocular phenomena. Spatial questions elicited significantly fewer eye movements (excluding trials on which stares occurred) than did verbal questions (z = 2.50, p < 0.01; all analyses are Wilcoxon matched-pairs signed-ranks tests, and all probabilities are twotailed). This difference in number of eye movements for verbal and spatial questions could possibly have been an artifact of a difference in time taken to answer a particular type of question, since the longer the duration of an answer, the more opportunity there was to produce eye movements. This did not appear to be the case, however; although verbal questions tended to be somewhat longer than spatial questions, this difference was not significant

Table 1 Type of question

Verbal Spatial

Mean responses for trials with eye movements and trials with stares Type of response Trials with stares

Trials with eye movements ______ __Number of eye Time for eye Eye movements/ movements in time in seconds movements seconds

Number stares

7.33 5.12

6.07 8.93

9.84 9.36

0.84 0.66

of

Time for stares in seconds

3.34 4.24

Ocular motility and cognitive process

Table 2

37

Eye movements per second for answer categories Type of answer

Type of question

Verbal Spatial

One word

Lists

Extended

0.90 (24) 0.70 (26)

0.73 (2) 0.61 (5)

0.75 (14) 0.60 (6)

speech

Note: Numbers in parentheses represent number of questions category. Three spatial questions were ‘unclassifiable’.

in this

(z = 1.65, p < 0.11). Moreover, when time to answer was taken into account by using a proportion score of eye movements per second, spatial questions elicited significantly fewer eye movements per second than did verbal questions (z = 2.73, p < 0.01) (see Table 1). Because the time difference between verbal and spatial questions was not far from significant, the score of eye movements per second, which takes time to answer into account, is used below when discussing the question differences. It is also possible that the ocular motility differences for verbal and spatial questions were related to another characteristic of the response, viz., the amount of speech required to answer a question. In order to test this possibility, questions were further divided by the type of answer required: One word answers, lists of objects or people, extended speech, or unclassifiable. As can be seen in Table 2, comparable differences between verbal and spatial questions in eye movements per second emerged regardless‘ of the category of answer, with spatial questions eliciting significantly less ocular motility than verbal questions for one word answers (z = 3.24, p < O.OOl), and for extended speech (z = 2.84, p < 0.005). Although the difference for lists was in the same direction as the other answer categories, the failure of this difference to reach significance, z = 1.33, p < 0.20, may have been due to the small number of items in this category. As can be seen in Table 1, the pattern for trials on which continuous stares occurred was in accord with the pattern for trials in which eye movements occurred. As found in the larger group of subjects reported in Ehrlichman et al., (1974), when all the questions were considered, significantly more stares occurred for spatial than for verbal questions (z = 2.64, p < 0.01). In addition, stares for spatial questions tended to be longer than stares for verbal questions, although not significantly (z = 1.64, p < 0.11). The fact that the mean duration of stares tended to differ between verbal and spatial questions suggests that different kinds of stares may be associated

38

Susan L. Weiner and Howard Ehrlichmm

with each type of question. Stares of relatively short duration may be of the type observed by Duke (1968), which occurred for very easy questions with readily available answers. Stares of longer duration may reflect processes similar to those which affect ocular motility. Although none of the questions in the present experiment were designed to be of the very easy type Duke described, the possibility had to be examined that some of the questions had this effect. One way to test the hypothesis that different kinds of stares were associated with verbal and spatial questions was to test whether stares of long and short duration were distributed equally between the two types of questions. Trials on which stares occurred were classified as long or short in duration according to a median split (median = 2.58 seconds). The analysis was done on questions requiring one word answers in order to minimize the potential variability due to the length of the speech output. If the stares of long duration reflect the same processes as does ocular motility, then one would expect the number of such stares to be distributed differently for verbal and spatial questions. In particular, more stares of long duration should occur for spatial than for verbal questions, in accord with the observed ocular motility difference. On the other hand, if stares of short duration simply reflect the relative ease of the questions, there is no a priori reason to believe that the number of such stares should SySterhdtiCally differ between questions. In accordance with this prediction, the mean percentage of trials on which stares of long duration occurred was significantly greater for spatial than for verbal questions (13.54% for spatial, 8.61% for verbal; z = 2.10, p < O.OS), whereas this percentage did not differ significantly for stares of short duration (9.10% for spatial, 11.94% for verbal; z = 0.91). (Percent of trials was used because 24 verbal and 26 spatial questions required one word afiswers.) This analysis suggests further that stares of relatively long duration may be responsible for the difference between verbal and spatial questions in the number of stares elicited.

Discussion The observed differences in ocular motility between verbal and spatial questions have at least two implications. First, the fact that subjects stare more often during spatial than verbal questions can now be interpreted as an instance of subjects’ lower level of ocular motility during spatial questions rather than as a phenomenon having only to do with subjects initially maintaining or breaking fixation. Much research has dealt with eye movements towards or away from a specific locus (e.g., another person’s face, a

Ocular motility and cognitive process

39

video camera). By ignoring subjects’ overall pattern of ocular behavior, this research may be presenting a restricted view of the determinants of phenomena such as interpersonal gaze. For example, patterns of looking and looking away which occur during conversation (Kendon, 1967) may in part reflect variations in the level of ocular motility accompanying changes in cognitive functioning that occur during speaking and listening. Second, the present findings do not support an interpretation of the imagery-scanning hypothesis which associates relatively greater ocular motility with the presence of visual imagery. This conclusion is based, of course, on the assumption that the spatial questions evoked more visual imagery than did the verbal questions. It is possible to consider the notion that people scan mental images as they do visual displays as an hypothesis separate from one stating that imagery-scanning necessarily involves high levels of ocular motility. Indeed, because viewing a real scene can elicit any degree of ocular motility, depending on variables such as its distance, one’s perspective, etc., imagining scenes ought to elicit any degree of ocular motility, depending on the same sorts of variables. Thus a theory of the scanning of imagery need not be tied to predictions about specific levels of ocular motility. In addition, the imagery-scanning notion itself must eventually come to terms with the fact that subjects can transform and manipulate mental images in ways not applicable to real visual stimuli. Therefore, attempting to predict either the quantity or the direction of eye movements during imagery on the basis of what subjects have seen becomes extremely difficult, although these predictions become somewhat clearer in special cases of greater isomorphism between image and percept, e.g., in eidetic imagery (Haber & Haber, 1964) or in images of stimuli eliciting optokinetic nystagmus (Zikmund, 1966). If this analysis is correct, then both the usefulness of and the evidence for the notion that increases and decreases in ocular motility during most kinds of mental activity reflect the scanning of visual imagery are greatly weakened. For example, it becomes strained to explain increases in ocular motility during ‘mental multiplication’ by invoking the presence of imagery during such a task (Lorens & Dar-row, 1962). Rather, it would seem reasonable to try to account for changes in levels of ocular motility in more general terms which could be applied to many kinds of mental processes. One theory which provides a more general model of changes in ocular motility has been proposed by Singer, Antrobus and their colleagues (Antrobus, 1973; Singer et al., 1971). These investigators posit a single limited capacity central processor in which internal cognitive activities and impinging extermal stimuli compete for processing space. When subjects engage in tasks which take up this processing space, eye movements are

40

Susan L. Weiner and Howard Ehrlichman

inhibited in order to gate out visual and motoric input. When subjects’ active processing of a task is complete, inhibition is released and ocular motility increases. These sequences of processing and completion are described as cognitive change and are hypothesized to correspond to sequences of ocular motility and ocular quiescence respectively. It follows from this model that any task involving more frequent cognitive change will produce more eye movements. However, what constitutes equivalent units of cognitive change across tasks remains to be specified. While this general explanation of ocular motility has appeal, it is unclear how the model could account for the present data, since the questions do not appear to differ systematically in the rate of cognitive change that might be elicited. In addition, the assumption of a single channel central processor is challenged by the fact that other research has demonstrated multiple representational codes (e.g., Posner, 1973) and processing capacities (Brooks, 1968). The notion that external stimulation interferes with internal process can provide a mechanism relating ocular motility and cognitive activity, if it is reconceptualized to allow for multiple processing capacities. Such a theory would propose that when subjects perform tasks which engage particular capacities, their performance may be disrupted if they simultaneously carry out a second task engaging the same capacities, but may not be interfered with if the second task requires different capacities (see Brooks, 1968). To account for the present data, two further assumptions are needed: (a) interference can be caused by subjects’ mediating cognitive processes, regardless of the form of the stimuli or responses; (b) interference can occur when subjects engage in cognitive processing of task irrelevant stimuli, providing that such processing involves the same kind of cognitive processes as those required by the tasks. Applied to the present study, this ‘interference’ model can explain the obtained ocular motility differences by emphasizing imagery processes, verbal-conceptual processes, or both. When subjects are engaged in visual imagery, concomitant analysis of visual input can be interfered with (Segal & Gordon, 1969). Eye movements produce changes in visual simulation, resulting in visual analysis which may interfere with subjects’ imagery processes. Thus, subjects may move their eyes less during spatial than verbal questions to prevent the interference which might occur for spatial but not for verbal questions. If this is so, other responses which reduce visual information ought to occur more during spatial than verbal questions, e.g., closing or covering one’s eyes. We compared the number of times subjects closed their eyes for verbal and spatial questions in the three experiments reported in Ehrlichman et al. (1974). Instances of eye closing, though infrequent,

Ocular motility and cognitive process

41

occurred significantly more often for spatial (0.34 closings/trial) than for verbal questions (0.82 closings/trial) (z = 2.59, p < 0.01). While a hypothesis emphasizing interference between imagery and visual input accounts for the present ocular motility difference as a decrease during spatial questions, it seems equally reasonable to account for this difference as an increase during verbal questions. It is possible that moving one’s eyes in a static surround may be less interfering for verbal-conceptual processes than focusing on relatively few points. Looking too long at any particular stimulus may cause subjects to engage in conceptual elaborations (for example, by eliciting chains of associations). Thus subjects may move their eyes more during verbal than during spatial questions in order to avoid the may have on verbal-conceptual interfering effects such elaborations processes. Any theory of ocular motility which exclusively depends on an interaction of external stimulation and cognitive process is limited by the fact that systematic differences in ocular motility occur both during sleep and when awake subjects have their eyes closed or covered (Andreassi, 1973; Singer & Antrobus, 1965). However, alternative views of ocular motility which have been proposed are limited. Either variations in task demands (such as the degree of attention) have been correlated with changes in ocular motility without providing mechanisms responsible for those changes (Weitzenhoffer & Brockmeier, 1970), or potentially useful mechanisms (such as arousal) have been proposed without providing links with task variations (Andreassi, 1973). Moreover, for either attention or arousal to account for the present data, properties of the verbal and spatial questions would have to be shown to be related to these constructs. Current theories of visual information processing have progressed considerably by employing precise chronometric analyses (e.g., Cooper & Shepard, 1973; Posner, 1973). It seems likely that a number of the processing stages articulated for a variety of spatial and linguistic tasks, some of which are similar to questions used in the present study, could be related to changes in ocular motility. A detailed mapping of this sort might allow ocular motility to be used as a continuous response which is sensitive to stages in cognitive processing.

42

Susan L. Weiner and Howard Ehrlichman

REFERENCES

M., & Shagass, C. (1963) Eye movements, attention and hypnosis. J. nerv. ment. Dis., 136, 139-145. Andrcassi, J. L. (1973) Alpha and problem solving: A demonstration. Percept. Mot. Skills, 36, 905-906. Antrobus, J. S. (1973) Eye movements and nonvisual cognitive tasks. In V. Zikmund (Ed.), The oculomotor system and brain functions. London: Butterworth. Antrobus, J. S., Antrobus, J. S., & Singer, J. L. (1964) Eye movements accompanying daydreaming, visual imagery, and thought suppression. J. abn. sot. Psycho/.. 69, 244-252. Argyle, M., Lalljee, M., & Cook, M. (1968) The effects of visibility on interaction in a dyad. Hum. Rel., 21, 3-17. Brooks, L. R. (1968) Spatial and verbal components of the act of recall. Can. J. Psychol.. 22, 349.-368. Brown, B. B. (1968) Visual recall ability and eye movements. Psychoph_k*siol., 4, 300-306. Cooper, L. A., & Shepard, R. N. (1973) Chronometric studies of the rotation of mental images. In W. G. Chase (Ed.), Visual inftirmation processing. New York, Academic Press. Deckert, G. II. (1964) Pursuit eye movements in the absence of a moving stimulus. Science. 143, 1192-l 193. Duke, J. D. (1968) Lateral eye movement behavior. J. gen. Psychol., 78, 189-195. Ilhrlichman, H., Wcincr, S. L., & Baker, A. II. (1974) Effects of verbal and spatial questions on initial gaze shifts. Neuropsychologia, 12, 265-277. Exline, R. (1971) Visual interaction: The glances of power and preference. In J. K. Cole (Ed.), Nebraska symposium on motivation. Lincoln, University of Nebraska Press. Galin, D., & Ornstein, R. (1974) Individual differences in cognitive style: I. Rctlective eye movements. Neuropsychologia, 12, 361-316. Haber, R. N., & Habcr, R. B. (1964) Eidetic imagery: I. Frequency. Percept. mot. Skills, 19, 131-138. Hebb, D. 0. (1968) Concerning imagery. Psychol. Rev., 75, 466477. Jacobs, L., Feldman, M., & Bender, M. B. (1972) Are the eye movements of dreaming sleep related to the visual images of the dreams’? Psychophysiol., 9, 393-401. Kendon, A. (1967) Some functions of gaze-direction in social interaction. Acta Psychologica, 26, 2243. Koulack, D. (1972) Rapid eye movements and visual imagery during sleep. Psychol. Bull., 78, 155-158. Lorens, S. A., & Darrow, C. W. (1962) Eye movements, EEG, GSR, and EKG during mental multiplication. Electroencephalography and Clinical Neurology, 14, 739-746. Mackworth, N. H., & Morandi, A. J. (1967) The gaze selects informative details within pictures. Perception and Psychophysics, 2, 547-552. Marks, D. F. (1972) Individual differences in the vividness of visual imagery and their effect on function. In P. W. Sheehan (Ed.), The function and nature of imagery. New York, Academic Press. Moskowitz, E., & Berger, R. J. (1969) Rapid eye movements and dream imagery: Arc they related? Nature, 224, 613-614. Posner, M. I. (1973) Coordination of internal codes. In W. G. Chase (Ed.), Visual information processing. New York, Academic Press. Segal, S. J., & Gordon, P. E. (1969) The Perky effect revisited: Paradoxical thresholds or signal detection error? Percept. mot. Skills, 28, 791-797. Singer, J. L., & Antrobus, J. S. (1965) Eye movements during fantasies. AM.4 Arch. gen. Psych., 12, 71-76. Amadeo,

Ocular motility and cognitive process

43

Singer,

J. L., Greenberg, S., & Antrobus, J. S. (1971) Looking with the mind’s eye: Experimental studies of ocular motility during daydreaming and mental arithmetic. TrQnSQCtiOnS of‘fhe New York Academy of Sciences, 3, 694-709. Totten, E. (1935) Eye movement during visual imagery. Comp. psychol. Mono.. II, No. 3. Weitzenhoffer, A. M., & Brockmeier, J. D. (1970) Attention and eye movements. J. nerv. menf. Dis., 151, 130-142. Yarbus, A. L. (1967) Eye movementsand vision. New York, Plenum Press. Zikmund, V. (1966) Oculomotor activity during visual imagery of a moving stimulus pattern. Sfudia Psychologica, 8, 254-274.

RtsumP Le but de la presente etude est de rechercher le taux de mobilite oculaire accompagnant les reponses i des questions mettant cn jeu divers types de processus cognitifs. Les riponses i des questions reposant sur des processus visuo-spatiaux induisent moins de mouvements oculaires que celles qui reposent sur des processus conceptuels et vcrbaux. Ces r&hats vont dans le scns de nos observations ant&cures sur la ‘fixation’ oculaire; ils ne sont pas compatibles avec l’hypothese traditionnelle reliant I’augmentation du balayage oculaire i la presence d’images visuelles. Nous interpretons nos resultats en termes d’un modele d’interaction entre I’information visuelle et le type d’activite cognitive utilisee par les sujets pour son traitement.

Cognition, 4 (1976) 45-97 0 Elsevier Sequoia S.A., Lausanne

- Printed

3

in the Netherlands

Poetry and song in a language without sound*l**

EDWARDS.

KLIMA

University of California at San Diego URSULA

BELLUGI

The Salk Institute

for Biological Studies

Abstract We examine distinctive characteristics of a primary language (American Sign Language) which is manual-visual rather than oral-aural. It is proposed that attributes of the mode of this language predispose not only the language itself but also its art forms to certain special characteristics. An analysis is presented of some instances of ‘art-sign’, a poetic tradition developing within our own time, where gestural analogs are suggested to poetry and perhaps even to song.

1. Ihtroduction In the context of recent studies devoted to linguistics and the voice, it may appear perverse to offer a contribution which not only has nothing to do with the voice as a source of sound in the auditory channel of communication - but, what is more, has nothing to do with sound whatsoever. What we plan to discuss is some recent research on an art-form utilizing a type of primary human communication system - one of the sign languages of the deaf - which is articulated through manual and other non-vocal gestures and is perceived visually rather than auditorily. The sign language that we shall be examining is the so-called American Sign Language (abbreviated as ASL), and we shall refer to the heightened use of this silent language of signs as ‘art-sign’. *Contribution to seminar, “Musique et Linguistique”, organized by I.R.C.A.M. (April 29 - May 3, 1975). **This research was supported in part by National Science Foundation Grant #CSOC-7401780 and the National Institutes of Health Grant #NS-09811 to The Salk Institute for Biological Studies.

46

Edward S. Klima and Ursula Bellugi

Let us begin, however, with some general remarks about sign language. Although not acoustically based, the sign language we shall discuss certainly qualifies as a language - not only in the loose metaphorical sense in which we speak, e.g., of the language of the bees or the language of love, and not even with the extension which may be involved in speaking of music itself as a language. When we refer to sign languages as “languages”, we mean that they have sentential units which have a strict semantic-propositional interpretation (providing among other things for the possibility of paraphrase); that they also have a hierarchically organized syntax - open-ended in terms of possible messages - and furthermore, that at the formational level of the individual lexical units (the individual signs) as well as at the syntactic level, there are specific constraints as to well-formedness. What is more, there is a definite sense among those with a sign language as a native language (for example, the offspring, deaf and hearing, of deaf parents offspring who learned sign language as their first language) that the sign decidedly has a citation form - a form which exists out of any specific reallife context. That is, the sign is not situation-bound as are some affective units of communication. (We presume, for example, that a scream does not have a citation form in this sense; nor presumably would an element of free pantomime.) Thus, an ASL sign as such is no more bound to a particular context than is a word of spoken language. We must be careful to point out parenthetically, however, that in order for there to be ‘communication’ in the general sense of the term, there need not necessarily be syntax in the restricted sense in which human languages have syntax. A string of nominals can communicate admirably, as in the possible, essentially non-syntactic utterance: “The animals; the hunters; the kill.” By means of such utterances, relationships are suggested but - and this is a significant difference - not explicitly stated. A speech act of a totally different sort is involved. If all possible English utterances were of the nature of just such unstructured strings, it would be appropriate to describe English as not having a syntax. But in English as well as in all other human languages that have ever been examined there is a means not only of suggesting relationships but also of casting them in the form of explicit statements. The syntactically structured sentence: “The animals killed the hunters” (as contrasted with the syntactically different sentence: “The hunters killed the animals”) is a means for just such a statement and manifests the characteristics that we associate with syntax. ASL certainly has a syntax in this strict linguistic sense of the term. In referring to sign languages as primary languages we mean that they are not directly connected with (not based on) the units of vocal language. Thus written words, though perceived visually and produced manually, do not themselves represent a primary communication system but rather a

Poetry and song in a language without sound

47

secondary system based, of course, on such primary languages as English, French, etc. American Sign Language is the visual-gestural language that has evolved among the deaf in the United States. ASL, incidentally, was influenced strongly in the first quarter of the 19th century by contact with the French language of signs. Before that time there was undoubtedly a sign language in use among the deaf in the United States, but around 18 15 Gallaudet brought the teachings of the Abbe de 1’Epee to America. This importation undoubtedly enriched the indigenous sign language to a very great extent. In other publications we and our colleagues have reported on some of the changes that ASL has undergone since the 19th century (Frishberg, 1975). In short, the form of individual signs has tended to reduce itself to a combination of a limited set of formational parameters. These parameters include, according to our analysis and that first proposed by Stokoe, Casterline, and Croneberg (1965): Hand Configuration (of one or both hands), Relationship between the hands, Orientation of the hands, Place of Articulation with respect to the rest of the body, and type of Movement of the hands. There are definitely strict formational constraints on the form of actual (and possible) signs in ASL. There is, in other words, an analogue to “phonology” in sign languages like ASL. The mere fact that certain, but not all, gestures are recognized as possible signs, though they are not necessarily actual signs of ASL, is itself, of course, an indication that the formational level of the lexical items (the signs) represents a rule-governed system like that of the phonology of the word. But a sign is not in its internal constitution like a string of letters written in the air. In fact, there does exist a form of manual communication (distinct from ASL) used in certain situations by the deaf, and by the hearing in communication with the deaf, in which each letter of the English alphabet is represented by a certain configuration of the fingers. This is called fingerspelling and, of course, fingerspelling is not a primary system of communication any more than handwriting is. Let us emphasize that we are dealing here not with fingerspelling but with a totally different sort of system - sign language proper - and the organization of the basic lexical units of sign language is radically different from that of the word. Whereas the word - at least at one level of analysis - is constituted of a sequence of phonological segments; the simple sign is essentially a simultaneous occurrence of a particular value (a particular realization) of each of the formational parameters mentioned above. Let us take, for example, the English word “feeling” and its ASL translation equivalent represented in Figure l* and glossed as FEELING**. The *The illustrations **See overleaf.

for this paper were made by Frank

A. Paul.

48

Edward S. Klima and Ursula Bellugi

English word consists of a string of segments (the phonemes /f/,/?/,/l/,/$,/n/) arranged in a particular order in two syllables. By contrast, the sign with an equivalent meaning is articulated by one hand alone, the palm facing the body (Orientation); the hand is flat and spread with the middle finger bent in (Hand Configuration), contacting the middle of the chest (Place of Articulation) and stroking upward twice (representing the parameter of Movement). Incidentally, it is not surprising to find such a fundamental contrast. at the lexical level. between auditory symbols and visual symbols. Figure 1.

As Jakobson pointed out, there is a strong tendency for the auditory to be sequential, whereas the visual tends to be very rich in simultaneous components (Jakobson, 1967). And even at the morphological level, the modulation of meaning typically (but of course not exclusively) accomplished in spoken language by the addition of affixes (i.e. again predominantly sequential elements) finds its counterpart in sign language by simultaneous, super**In Notational Conventions: The English translation-equivalents of ASL signs are represented in capital letters, as in UNDERSTAND. (Naturally the form of the ASL sign need have no relation to the form of the English word.) If more than one English word is required to translate a single sign, (not signed) we hyphenate the two words - e.g., LOOK-AT. Words which have been fingerspelled appear with hyphens between each letter - e.g., B-O-Y-S. In order to convey descriptions of sign-plays discussed here, we have adopted further notations. Two overlapping signs, or two signs made simultaneously, are indicated by a slash ~ e.g., EXCITED/ DEPRESSED. If one sign is made and then blended with another sign, z represents the blending. If one value of a sign is changed to create a sign-play, the English translation-equivalent still appears in capital letters, but the modulation in meaning appears in lower case - e.g., UNDERSTAND-a-little. In addition, certain examples have two illustrations, one representing the standard form of the ASL sign, and the other the sign-play. When we refer to the sign-play, it is enclosed in quotation marks. The standard form of the sign appears in capital letters without quotation marks.

Poetry and song in a language without sound

49

imposed changes in, for example, the parameter of Movement. On the other hand, sign language does have its sequential aspects: The compounding of two signs, a sequential process, is one way of introducing new lexical items; and the sentence in sign language consists essentially of a sequential series of signs just as the sentences of spoken language consist of a series of words. Before touching on the poetics of ASL, let us add two more general remarks. You will notice that we have spoken about sign language as a type of system just as we referred to spoken language as a type of system; we have also referred to sign languages in the plural; for indeed there are different sign languages ‘as there are different spoken languages. It is quite erroneous to consider sign languages like ASL or Chinese Sign Language, which we have also studied to some extent, to be any sort of universuj shared system of gestural communication. Individual signs in such a language as ASL are not merely pantomimic gestures, although the source of many signs is indeed often in pantomime. But as such pantomimic gestures become ‘domesticated’ - to conform ultimately to the particular parameters that are part of the system - they tend to lose their iconic aspect and become more and more like arbitrary symbols - arbitrary in much the same way as words are arbitrary. There undoubtedly does, however, remain a difference in the degree to which these gestural, visual symbols - as opposed to vocal symbols - retain iconic, representational characteristics. In deaf communication, mimetic aspects are still very much alive - in the sense that when new objects enter into the culture and “look for” appropriate names in sign language, one method of creating a designation for such objects is to use pantomime. But we have noticed that very quickly this pantomimic way of symbolizing first becomes stylized and then often becomes totally conventionalized so as finally to partake of only the restricted parameters of the system. The objective of this paper, however, is not to make a detailed description of a sign language such as ASL as it is used in ordinary deaf conversation, how it is .perceived, or how it has changed historically. Elsewhere we have written about aspects of the linguistic structure of the language* and about various psycholinguistic experiments we and others have conducted which show the similarities and differences between sign language and spoken language* * (though to be precise most of these investigations deal with ASL and English). Similarly, it is not our intention here to deal with

*Bellugi and Klima (1975): Klima (1975). See also Frishberg (1975), Fischer (1975), Stokoe (1972). A collection of articles about sign language is now in preparation: Klima and Bellugi (to appear). **Bellugi and Siple (1974), Bellugi and Fischer (1972), Bellugi, Klima and Siple (1975).

50

Edward S. Klima and Ursula Bellugi

the innateness issue: Whether speech is special and whether the ‘linguistic’ faculty is specially tied to sound. Is there language without speech? Our own research shows that there most certainly is. Does the mode (the channel) make a difference? Our research suggests that it probably does. Is this difference an essential one - i.e., do the differences make sign language a qualitatively different system? This can be answered only after a great deal more research. Of course it seems obvious that speech constitutes part of the biological foundations of language. How else can one explain the relative ease with which any normal human being with the faculty for hearing can tune into the processes necessary for producing and perceiving such a very subtle signal. But if speech is specially selected for; if sound constitutes such a natural signal for language, then it is all the more miraculous how the human mind - or perhaps ‘mind’ makes the process seem too conscious, too much a matter of intellect - let’s say, rather, how the human disposition, when deprived of the faculty that makes sound accessible, seizes on and perfects an alternate form that enables the deeper linguistic faculties to give expression to ideas. In this paper, however, our objective is tp investigate certain other basic human faculties. Whether or not they are universal, they are decidedly very significant aspects of a great many cultures of the world. We are referring now to the heightened use of language in creating that special and complex type of symbol called poetry, where elements of the linguistic system are used to create new systems. In addition, we are asking a further question: In certain types of heightened, multiply structured signing that we have observed, is there not a silent-language analogue to that special blend of sound with sound - speech sound with musical sound - that, in the auditory channel constitutes song? While the study of poetry for its own sake requires no justification, it is important to remember that the analysis of such heightened uses of language can also inform us about the psychological reality of linguistic constructs. For spoken languages, the results of analysis of the poetic function have shown a sensitivity, on the part of native speakers, to grammatical elements of the language for their own sake: to sound as sound, to grammatical categories as grammatical categories - an awareness of these as more than just fleeting vehicles for the expression of meaning. And, of course, the combining of phonation and vocalization reveals among other things a sense of the play between these two types of signals - simultaneously occurring in the same modality and yet belonging to two distinct systems. From the very onset of our study of ASL, our research group has been interested in whether there are any analogues, in this language without sound, to such heightened uses of language that would similarly constitute

Poetry and song in a language without sound

independent support for the grammatical principles that have been proposed.

2. The poetic

function

constructs

51

and general structural

in wit

2.1. Plays on words Let us consider what is involved in the poetic function in general. In terms of propositional or referential content, it matters little that in English, for example, worst (the superlative of bad) and wurst meaning ‘sausage’ sound the same, or that June, moon, croon, swoon have the same vowel-sound and final consonant - i.e., that they rhyme. But there are functions of language outside of the purely referential for which such otherwise incidental similarities become significant in terms of the totality of what is communicated - in terms of the total import of an utterance. Among cases where this is obviously true for the English speech community are puns; such a punning ad-slogan for mustard as “It brings the best out of the worst (wurst)” derives its full import from the wurst/worst ambiguity in English. And, of course, just such rhymes as June, moors, swoon, croon provide the basis for a superimposed structure of sound whereby mere sentences take on, in addition, that special significance of the patterning embodying verse albeit the sentences may express inanities and the verse may be doggerel. What is special about plays on words, verse, and poetry in general is a heightened awareness of linguistic phenomena as linguistic phenomena. As Jakobson (1960, p. 356) puts it: “The set toward the message as such, focus on the message for its own sake, is the poetic function of language.” Like “art for art’s sake”, language for language’s sake would be pure poetic function. While the poetic function certainly dominates in various forms of language-based art - and certainly very much so in lyric poetry the poetic ‘function is also represented in everyday language use, though in a less structured way. In what follows, we shall show that the poetic function is represented also in the linguistic activities of the deaf using the primary visual-gestural language called American Sign Language (ASL). 2.2 Remarks

about wit and plays on signs in ASL

Uninformed “glottocentric” views of sign language as a collection of unstructured pantomimic gestures could lead some to wonder whether ,such structure-bound activities as punning and other linguistic play are possible -

52

Edward S. Klima and Ursula Bellugi

or at any rate, natural - to sign language, for such activity relies so heavily on subtle correspondences and quickly grasped associations of not only meaning but also, very significantly, ofform. One might even be led to ask if the general sort of structural properties present in all spoken languages those that provide the basis for such linguistic play - might not somehow be absent from sign language. Moreover, since such manipulation of linguistic units as appear in verse and linguistic play inform us about significant structural properties of languages, it is all the more relevant to ask whether comparable language-based phenomena occur in ASL. One form of linguistic play popular among some speakers of English involves the calculated use of a word in an utterance in such a way as to suggest, at the same time, the different meaning of another word or words having the same (or nearly the same) sound. When a critic talks of an uninspired TV series as “the bland leading the bland”, he is punning - playing on the similarity in sound between “bland” and “blind”, and on a cliche, although cliches are not essential to punning, as is seen in “Bad coffee every morning can be the grounds for divorce.” Linguistic play is not limited to cases where one and the same signal, in a single instance of uttering, independently represents two or more words, each with its own meaning. Sometimes parts of two words overlap, as when the Christmas season is referred to as “the alcoholidays”. Often in linguistic play no ambiguity is present at all. When some particular institution of higher learning is referred to as “more of a collage than a college”, the linguistic play therein involves highlighting the minimal differences in sound between the two otherwise very similar words “collage” and “college” and associating the differences in sound with less obvious differences in the connotations of the two terms with respect to the contrasting sorts of organization typically identified with each. In spontaneous ASL communication, linguistic play is, in fact, rich and varied. Such sign-play has the same general characteristics as the linguistic play described above: shared aspects of form - deliberately highlighted producing special (and, if successful, witty) associations of the meanings involved. Significantly, much of the sign-play we have observed reveals processes based on structural characteristics more special to sign language itself. Elsewhere (Klima and Bellugi, 1975) we have exemplified more fully the mechanisms involved in sign play. Here, we shall review those aspects that are also utilized - but in a less incidental, more structured way - in art-sign. 2.2.1. Overlapping of signs In signing, because of the existence of two autonomous articulators (namely, the two hands) there is the logical and physical possibility of producing two

Poetry and song in a language without sound

53

independent signs simultaneously. Moreover, such simultaneity would not be inconsistent with some of the special characteristics of ASL: The relative weakness of order of signs alone as a clue for grammatical function; the apparent tendency to compress information into single units (see Bellugi and Fischer, 1972); and the use of simultaneous (rather than sequential) modifications of signs to modulate meaning. In everyday signing, however, aside from the simultaneous production of a ‘pointing’ (pronominal-like) sign with one hand and a lexical sign with the other, we have found that there is no regular production of two independent lexical signs simultaneously. But overlapping of signs does occur - and quite frequently - in selfconscious signing of preplanned material, in the plays on signs that abound in communication among the deaf, and, as we shall later see, in poetic signing as well. Let us examine two cases of overlapping of signs in wit. a. Making two signs simultaneouslq~. Two signs are simultaneous when the movement of one sign coincides precisely in time with the movement of a second distinct sign. We captured one such example during a conversation with a deaf student. This particular play on signs involved the signs EXCITED and DEPRESSED, both two-handed symmetrical signs with the middle finger of each hand bent in and brushing the chest. The two signs differ in that in EXCITED the movement is upward and in DEPRESSED it is downward (the spatio-symbolic aspects of the difference in direction are obvious). When asked, during the conversation, how he felt about leaving a city he loved for a choice job in a less appealing location, the young man summed up his feelings by signing simultaneously EXCITED with the right hand and DEPRESSED with the left. The possibility of thus successfully compressing into a single new sign-creation such ambivalence of emotions depended crucially on the fact that the original signs involved had a regular semantic relationship and were also related formationally. Figure 2.

“EXCITED/DEPRESSED” (consult footnote of page 46 for notational conventions).

54

Edward S. Klima and Ursula Bellugi

b. Holding one sign while making another. This involves making one sign and holding the final position and handshape of that sign while making another sign with the other hand. The final positions of both signs are then held. Again, this depends either on choosing signs that are made with one hand only, or on changing a sign so as to make it as a one-handed sign. The following is an example. On seeing a young deaf man each day with a different girl, we commented that he had an eye for pretty girls. He, in turn, signed what amounted to: “Yes, I’m an expert (girl-) watcher.” But he signed it in a special way. He first made a sign for EYES, one that mimics the eyes with two hands. Then, holding the final position of half the sign with one hand, with a simple twist of the wrist he let his other hand slip into the sign for EXPERT (signing “eyes-pert”, as it were). The combination is particularly effective in ASL because the two signs he chose use the same handshape. Figure 3.

EYES and “EYES/EXPERT”

(consult footnote

of page 46 for notational

conventions).

c. Blending of two signs. There are various ways of blending two signs into one unit. One can make a sign, and then add the movement of another sign, so that the two signs become one unit. Or one can make a sign and continue its movement throughout what would normally be the transition to the next sign, allowing for slight changes in orientation or location, until one sign has been transformed into another. The following instance exemplifies such blending. One day when we were particularly inept in our attempts to sign, one of our deaf colleagues good-humoredly signed to us that we were clever, and then by blending formationally related, and semantically appropriate, signs went from CLEVER to DEFLATE to INFLATE to CLEVER once again suggesting that now the cleverness was diminished but that later it would return again. To accomplish this, she signed CLEVER in the normal way: the cupped hand contacting the forehead. Then without relaxing that contact (as would happen ordinarily in the transition between two distinct signs), she closed the cupped hand suggesting a displaced form of the sign

Poetry and song in a language without sound

55

DEFLATE - with active hand displaced from its normal position contacting the back of the base hand in the front of the signer’s chest. Then again without relaxation of the original contact with the forehead, the hand opened; in doing so, suggesting a similarly displaced INFLATE which then became the sign CLEVER once again. Figure 4.

“CLEVERnDEFLATE”.

2.2.2. Substitution of one regular ASL prime for another Another general process that is used in plays on signs - and also in art-sign is a change in one of the basic parameters of a sign so that there is simple substitution of one prime of that parameter for another. This is essentially a distortion of a sign so that all but one of the basic characteristics of the sign are retained. The result of this sort of linguistic play is a possible .but not an actual sign of ASL: neither a citation form nor a standard modulation of a sign. But it differs from an actual ASL sign in a way that is significant and meaningful - either in terms of ASL morphology, or in terms of more general spatial-gestural symbolism. Appreciating the wit (and often, in fact, recognizing the actual sign behind the distortion) usually depends on knowing the context in which the new sign was created. The following example shows a change in a prime of Hand Configuration that occurred in a play on signs during a discussion in ASL of a complex technical point about transformational grammar. When asked, “Do you understand?“, the deaf colleague replied with the sign for UNDERSTAND, but instead of flicking open the index finger, she made the sign with the little finger. The basis for this distortion is clear: The little finger occurs in a symbolic way in some signs where it conveys the notion of thinness (SPAGHETTI, THREAD, SKINNY-PERSON) or extreme smallness (TINY, INFINITESIMAL). The substitution of the little finger for the index finger in UNDERSTAND clearly carried the meaning “understand a little”.

56

Edward S. Klima and Ursula Bellugi

Figure 5.

UNDERSTAND

and “UNDERSTAND-a-little”

The processes of sign-play which we have been describing involve manipulations of signs which are, on the whole, special to the form of sign language itself. The overlapping of two signs, for example, depends on the existence of two independent articulators - the two hands. The blending of two signs involves the fusion of properties of the signs. Lewis Carroll attempted to achieve something similar in speech in his blends of, say, But the essentially sequential “furious” and “fuming” into “frumious”. nature of word segments does not lend itself in quite the same way to the creation of such fusions as does a visual-gestural language based on simultaneously occurring parameters. In any event, such sign play occurs frequently in spontaneous signing we have captured on video-tape.

3. The poetic function

in poetry

3. I. Remarks on poetic structure in English verse The poetic function figures most complexly, of course, in poetry itself, where linguistic form becomes the basis for the patterns constituting the multiple layers of structure underlying a poem. In spoken language, we can distinguish several major types of poetic structure. The first of these we shall call Internal Poetic Structure. By Internal Poetic Structure we mean structure which is constituted from elements that are completely internal to the linguistic system proper (i.e., in the case of sign language, constituted from the form of standard signs in ASL - constituted from parts of the grammatical code itself). The two sub-types of Internal Poetic Structure to which we shall address our attention here we refer to as ConventionaZ Poetic Structure (provided or even demanded by tradition) and Individual Poetic Structure (individual to the particular poem). In the English literary tradition, such metrical schemes as iambic pentameter constitute the basis for a kind of Conventional Poetic Structure. For this structure, the fact that

Poetry and song in a language without sound

57

a syllable has greater stress than the syllables immediately surrounding it becomes significant, as do the number of such significantly stressed syllables. Similarly, various end-rhyme schemes that establish recurring sound patterns (e.g. aabb, abab, abba) are part of Conventional Poetic Structure in the and English tradition,* as are larger designs like the Elizabethan sonnet-form, the haiku form borrowed from Japanese poetic tradition. In structurally complex poetry, however, Conventional Poetic Structure will be overlaid and interwoven with more innovative Individual Poetic Structure, consisting of more subtle patterning of not only sound texture but also of other linguistic elements - syntactic, lexical, semantic and thematic. The eight lines of Blake’s “Infant Sorrow”, analyzed thoroughly by Jakobson (1970), exemplify the distinction between Conventional and Individual poetic structure - both constructed from aspects of the grammatical code itself. INFANT

SORROW

My mother groan’d! my father wept. Into the dangerous world I leapt: t Helpless, naked, piping loud: Like a fiend hid in a cloud. Struggling in my father’s hands, Striving against my swaddling bands, Bound and weary I thought best To sulk upon my mother’s breast.

I t

verb pret. (c)loud

noun pl. b(r)est

The Conventional Poetic Structure of “Infant Sorrow” consists of four rhyming couplets: aa, bb, cc, dd (wept: leapt: : loud: cloud: : hands : bands: : best: breast). The individual lines are all equivalent to one another in being iambic tetrameter. For the purpose of illustrating the Individual Poetic Structure of the poem we shall restrict our attention to only certain limited aspects of the additional patterns manifested by the words occurring in linefinal and line-initial positions (though, as Jakobson’s analysis shows, the poem is replete with structurally significant equivalences). Not only do the line-final words constitute conventional rhymed couplets, they also reveal a special pattern of heightened rhyme that reinforces the division of the poem’s four couplets into two structurally equivalent quatrains. The “grammatical” rhyme wept: leapt (grammatically equivalent in both being *Though in the modern English literary tradition, rhyme may be one of the most common mechanisms involved in poetic structure, historically this was not always the case. In fact, rhyme was borrowed from Oriental poetic traditions.

58

Edward S. Klima and Ursula Bellugi

preterite verbs) in the first quatrain is paralleled by the “grammatical” rhyme hands: bands (both plural nouns) in the second quatrain; in addition, the end couplets of both quatrains are set equivalent in that each reveals “inclusive” rhyme: The sound-form of loud is literally “included” in the sound-form of cloud; the sound-form of best in that of breast. Further Individual Poetic Structure involving the reinforcement of the division of the poem into two quatrains is an equivalent alliterative pattern occurring in line-final words of both quatrains: Three instances of prevocalic !. in the linefinal words of the first quatrain (leapt, loud, cloud) paralleled by three instances of initial b in the line-final - i.e., structurally equivalent - words of the second quatrain (bands, best, breast). As a final illustrative example of Individual Poetic Structure in this poem we could cite the following: In the first quatrain one particular structural configuration consists of the repetition, at the beginning of a word in lineinitial position, of the dominant alliterative sound 1 (&ke repeating leapt, loud, cloud); this is paralleled in the final quatrain (line-initial Bound repeating the b in band, best, breast). What is critical in all these examples is that elements of the linguistic code (phonemes, word-classes) are used in structure-creating manner rather than just as incidental a patterned, ornaments. Let us stop for a moment to consider such structural devices in terms of their possible function in poetry. We assume that one of the technical problems being tackled in short lyric poetry is that of counterbalancing - in this particular complex symbol - the sequential, temporal aspects of language: The fact that one word necessarily either precedes or follows another in a sentence; one clause either precedes or follows another clause. If one of the technical problems of creating an appropriate structure for such a symbol is, for example, to remove the necessity for there to be only one definite direction in the progression of the grammatical and thematic units if one of the technical problems is to give what is essentially a line also the characteristics of a mass, then one approach to such an end is certainly the intermeshing of linguistic units by means of just such patterns of correspondences that can, for example, make the end equivalent to the beginning and at the same time, by perhaps a different strand in the fabric, equivalent to the middle. To a limited extent, this is reflected by the conventions, in cultures with a writing system, according to which poems are laid out in a block on the printed page. Except in very special cases, however, the printed or written form of the poem is irrelevant to the poem itself, and such misleading expressions as the ‘lines’ of a poem refer, in fact, to structural units defined by the signal/meaning structure of the poem itself. For example, it is not that a line in certain metrical pattern consists of such-and-such con-

Poetry and song in a language without sound

59

figurations of stressed and unstressed syllables but rather that such-and-such configuration may count as a line. When used to describe a unit of ASL poetry, the term ‘line’ can only be understood in its abstract sense. 3.2. Analysis

of poetic structure

in ASL art-sign

In sign language ‘art-sign’, we have identified the following types of poetic structure. We have found Internal Poetic Structure, corresponding to Internal Poetic Structure in the poetry of spoken languages, but the patterning of linguistic\ forms in art-sign is by-and-large Individual rather than Conventional. In addition, we have discovered two types of external structure, different from poetic structure in spoken language, and special to sign language poetry. One type we shall distinguish as External Poetic Structure, in which the basic principles include: a) creating a balance between the two hands; b) creating and maintaining a flow of movement between signs; and c) manipulating the parameters of the signs. Then there is yet another kind of external structure, an Imposed Superstructure: a kind of design in space along with rhythmic and temporal patterning which may be superimposed on the signs and the signing, just as in a song we may have melodic structure superimposed on the words of a poem. The sources for our discussion of poetic or art-sign structure are varied,* but our primary source is from deaf people who are or have been associated with National Theater of the Deaf, a remarkably talented group of deaf actors (with an occasional hearing person, often one who has had deaf parents). Several of the actors have worked with us in our research at one time or another, generously giving of their time and enormous creative talents. The members of the National Theater of the Deaf have been developing a poetic tradition in sign language within our own time. This blossoming tradition involving the heightened use of sign language is based, as we shall see, on the inherent structural properties of signs and on special characteristics of signing. Aside from formal poems, we have also videotaped “songs” that deaf children invented in sign language, lullabies, children’s sign games, and other aspects of what might be called folk art in sign language.

*Among those who have helped by creating and discussing art-sign for us on video-tape are: Bernard Bragg, Lou Fant, and Dorothy Miles, all of whom have spent many sessions with us in our work; also involved have been Jane Wilk, Linda Bove, Pat Graybill, Joe Castronova and Ed Waterstreet. The deaf mem&s of our research group have also shared in interpreting, performing, and discussing poems, especially Bonnie Gough, Carlene Pedersen, Ted Supalla, and Shanny Mow. For similar assistance we are also grateful to Sharon Solow, a hearing person of deaf parents.

60

Edward S. Klima and UrsulaBellugi

3.3. ASL poetic processes illustruted

in a transluted

line

In order to illuminate some of the basic principles of art-sign, we shall present first a resume of an earlier analysis in depth of a single line of poetry (Klima and Bellugi, 1975). Bernard Bragg, who is deaf and a master signer of the National Theater of the Deaf, has spent many days with us in our work and has greatly enriched our research. In order to study the creative process in the development of poetic signing, we gave him as a problem a poem which he had never worked on before. We asked him to translate it into everyday signing, and then to show us the process of changing it into poetic form in ASL until he found what was to him a satisfying solution. The poem was one by E. E. Cummings ~ “since feeling is first” - and was peculiary apt, we felt, for linguists and artists to work on together, since it juxtaposes ‘syntax’ and ‘feeling’. Comparison of art-sign and straight ASL by Bragg

Table 1.

STRAIGHT ASL

FIRST

ART-SIGN

BECAUSE

-/FEELING

A dash and a slash before or after tains the position and/or handshape

ITSELF/-

FOREMOST/-

a gloss indicates that dne hand involved in the preceding of that sign - one common poetic device in ASL.

sign main-

Poetry and song in a language without sound

61

The first four lines are: “since feeling is first, whoever pays any attention to the syntax of things will never wholly kiss you. ..” We shall present here only the first line, and study the change from ‘straight’ (i.e., everyday non-poetic) signing to poetic ‘art-sign’, in Mr. Bragg’s capable hands. The first row of drawings in Table 1 represents the signs that Mr. Bragg chose to represent the meaning of the first line in straight nonpoetic signing, in fact a direct literal translation of the English words into signs of ASL. The second row represents the art-sign re-creation that gradually evolved during the session. In the straight version, as in normal everyday signing, we find examples of three formational classes of signs: (a) Signs involving two hands, with both active and operating symmetrically: (SINCE); (b) Signs made with one hand only: (FEELING and TRUE); (c) Signs made with one active hand operating on the other as a base: (FIRST). Since Bragg is right-handed, the one-handed signs in the straight version are made with the right hand, and during those signs the left hand is by his side or otherwise not in use. In this sequence of four signs in the straight version, (Table 2) there are changes in handshape from one sign to the next as follows (using symbols from Stokoe et al., 1965 along with mnemonic designations)*. The right hand, then, starts with an ‘Index’ hand (‘G’), switches to a ‘Mid-finger’ hand (tr), and back to an ‘Index’ hand for the last two signs. The left hand starts with an ‘Index’ hand, then drops down toward the side of the body and returns with a ‘Fist’ hand (‘A’). Note that in the straight version the hands are not only involved in the movement proper to the signs themselves, but also move back and forth, up and down, in making the transition between signs, gradually changing handshape or at least relaxing during these transitions. For example, at the conclusion of the two-handed sign SINCE, the left hand relaxes and drops to the side, and the *The symbols used in the Dictionary of American Sign Language (also referred to as “Stokoe notation”) to represent the distinctive hand configurations of ASL signs are: (a) English letters whenever the particular handshape is the same as or similar to that occurring in the manual alphabet, used for fingerspelling English words; (b) Arabic numerals whenever the handshape is the same as that of the corresponding manual form of the numeral; (c) otherwise arbitrary invented symbols like the Hand Configuration of FEELING (Table 2).

62

Edward S. Klima and Ursula Bellugi

right hand changes from a ‘G’ to an ‘tl’ while it is moving in the transition from the final position of SINCE to the initial position of FEELING. Table 2.

Handshapes in Bragg’s straight version

1. 2. 3. 4.

SINCE FEELING TRUE FIRST

The ‘G’ hand (or ‘Index’)

Right hand

Left hand

G [active] 6 [active] G [active] G [active]

G [active] [unoccupied] [unoccupied] A [base]

The ‘8’ hand (‘Mid-finger’)

The ‘A’ hand (‘Fist’)

3.3.1. Internal Poetic Structure In moving from conversational style to the poetic style of art-sign, Bragg made special changes. While these changes are in fact interrelated, we shall consider them separately for purposes of analysis. Consider first those changes that are associated with the choice of signs, i.e., with Internal Poetic Structure. Bragg replaced three of the four signs in changing from a ‘straight’ to a ‘poetic’ version. In fact, the only sign that remained the same in the two interpretations is FEELING. In our view, one factor played heavily in motivating the replacement of so many of the signs represented in the sign-for-word rendition: The so-called literal translation of the English word ‘since’ renders in ASL only the temporal sense of the word and is thus correct ASL semantically inappropriate for the line. * The semantically sign is formationally very different from the sign for the English word ‘since’. The appropriate sign - glossed as BECAUSE - has as its Hand *This was fist

brought

to our attention

by Geoffrey

Coulter.

Poetry and song in a language without sound

63

Configuration the ‘Fist’ with thumb extended (A in Stokoe notation); furthermore, the sign moves from contact with the forehead to a final position off to the side of the head. The rest of the changes can certainly be thought of as at least in part motivated by the special characteristics of the sign BECAUSE since the other changes in the choice of signs vis-avis the original literal version all result in signs with that particular variation of the ‘Fist’. Instead of TRUE, Bragg chose the sign ITSELF; and instead of FIRST, he created a sign in which a one-handed rendition of MOST (normally a two-handed symmetrical sign) combines with the superlative marker -EST. He himself ‘re-translated’ the resultant blend as ‘mostest’ and we gloss this FOREMOST. While not precisely like the citation form of any single ASL sign, it is certainly interpretable by a deaf viewer. In this first line of the art-sign version of the poem, then, we have four signs, each of which is made with one hand only. The three made with the right hand share the same handshape - the Fist (A). We have come to feel that this notion of shared handshape similarity is analogous to such phenomena as consonance (alliteration) or assonance in the poetic tradition of spoken language. 3.3.2. Externul Poetic Structure External Poetic Structure, characterized not by the choice of signs, but rather by patterned aspects of their presentation, can be manifested in artsign a) by maintaining a balance between the two hands and b) by creating a flow of movement between signs. a. Balance between the two hands. Signers, like everyone else, are generally either right-handed or left-handed, and sign accordingly in everyday signing, using the dominant hand to make one-handed signs; the dominant hand is also the active hand in signs in which one hand acts on the other as a base. Thus in everyday signing there is an imbalance in the use of the two hands by any individual signer. * But whether, in the act of signing, it is the right hand that is active or the left is irrelevant to the grammatical code of ASL (and no two signs are distinguished by one being made with the right hand and the other with the left hand or one with the dominant hand and the other with the non-dominant). However, in the poetic tradition being developed by the National Theater of the Deaf one type of External Poetic Structure may be imposed consisting of a pattern of hand alternation that keeps both hands more equally in use. There are several ways in which *In a study of more than 2,000 signs of American Sign Language we found that only 35% involve the use of both hands where both hands are active. About 40% of the signs are made with one hand only, and another 25% are made with one hand acting on the other hand which remains stationary as a base. Thus, for almost two thirds of these signs, one hand is used as the dominant hand.

64

Edward S. Klima and Ursula Bellugi

ASL poets achieve this balance. One method is by ulternuting hands in making consecutive signs. In Bragg’s art-sign version, after signing BECAUSE with his right hand, instead of subsequently signing FEELING also with the right hand, as he would in ordinary conversation, Bragg uses his left (nondominant) hand for that sign, and leaves BECAUSE hanging in the air as it were. Another method of creating a balance is by overlupping or making parts of two distinct signs simultaneously, a mechanism which we previously referred to in discussing plays on signs. In this one line of art-sign, we note that Bragg engages both hands at all times after the first sign. He holds the sign BECAUSE, which he makes with the right hand, in its final position while making the sign FEELING with his left hand. He then, in turn, holds the sign FEELING (left hand) and in a way that would not occur in everyday colloquial signing, directs toward it the one-handed sign ITSELF, which he makes with his right hand - thus emphasizing the fact that ITSELF refers to FEELING. Continuing to hold the Hand Configuration and final position of the sign FEELING (still with the left hand), he makes the final sign FOREMOST, which he produces with his right hand active. It is in this sense that in art-sign (as opposed to straight signing) there is more of a balance in the use of the two hands - providing one basis for poetic structure external to the grammatical code proper. b. Flow of movement. Let us look at the use of the hands in Bragg’s art-sign version: Table 3.

Handshapes in Bragg’s art-sign version Right hand

Left hand

1. BECAUSE 2. FEELING 3. ITSELF

.“: A

4. FOREMOST

A

tl ::.:.* *..* :.:

Hand Configuration through subsequent

symbol signs.

written

in dotted

lines (.....) is being held

This reveals a second general process involved in External Poetic Structure: the creation of a flow of movement (a continuity) between signs. Creating a flow of movement goes beyond the general processes of Internal Poetic Structure whereby signs are chosen so that, for example, the handshapes (part of the grammatical code of ASL) of two consecutive signs are the same. Creating a flow of movement between signs is often accomplished

Poetry and song in a language without sound

65

by interesting sorts of distortions imposed on the form of the signs themselves, again going beyond the grammatical code proper. This, for the most part, is different from what we have found in plays on signs, in spoonerisms, in regular meaningful modulations of signs, or in the memory errors we collected from our short-term memory experiments, and is quite specific to art-sign. The distortions associated with flow of movement involve not only the form of the signs themselves, but also the manipulation of transitions between signs. An effort seems to be made to utilize the transitions between signs in such a way as to avoid ‘wasted’ movement. In distorting transitions between signs, the sign-poet seems to attempt to make every movement - even that involved in pure transitions - ‘meaningful’, displaying the formational properties of the preceding or following sign. Consider the sequence of signs SINCE and FEELING in straight signing (see Figure 6). The initial position of SINCE is represented by the broken lines near the shoulder, and the final position is represented by the hands in the solid lines in the space in front of the shoulder. Similarly, the initial position of FEELING is represented by the lower drawing of the hand, which is a broken line at the mid-line of the lower torso. The transition between SINCE and FEELING, then, involves dropping the left hand to the side since it is not in use; and, at the same time, moving the right hand from the final location of SINCE to the initial location of FEELING (as represented by the arrow in the second drawing) while changing the handshape from the ‘G’ hand to an ‘8 hand during-this movement. This is what we mean by the transition between signs.

Figure 6.

66

Edward S. Klima and UrsulaBell@

In the poetic version of the line, Bragg selects and manipulates the form of the signs so that the final position of the hand after making each sign is precisely the starting position of the next sign, as we have already shown. The final position of BECAUSE, which is held throughout the signing of FEELING, becomes the starting position of ITSELF, and the final position of ITSELF is also the starting position for FOREMOST. This would not be the case in the conversational style of signing the same sequence of signs. So we see that the internal and external structures of the line have been made to work together: a) There is a simple patterning (repetition) of an element of the grammatical code: the three signs made with the right hand all share the same handshape; and b) the continuity between the signs, already expressed in the similar handshape, is enhanced by making the final position of one sign coincide with the initial position of the sign following it, without the usual blurred transition or ‘wasted’ movement between signs. 3.3.3. External Kinetic Superstructure There is yet another type of external structure which we will consider, and that is Kinetic Superstructure. We consider this somewhat analogous to the combination of melodic and poetic structure which occurs in song, where melodic structure is superimposed on the words which may as a result undergo certain kinds of distortions from the point of view of the linguistic code, though aspects of melodic and poetic structure may coincide and interact as well. In the single line of poetic signing under consideration, it may be a little difficult at first to separate clearly devices producing the Kinetic Superstructure from some of the other devices we have discussed. However, if one looks at the flow charts (Figure 7) of the movement of the hands in the nonpoetic and poetic renderings of that one line, it becomes clear that in the poetic rendering there has been a further distortion of the signs which creates an enlarged pattern of movement. This is enhanced by other types of distortions we have discussed (such as those eliminating ‘wasted’ movement in transitions), but this further, grosser distortion clearly seems an aim in its own right as well. Bragg has superimposed a special design in space on the signs chosen for his ASL rendition of the poem: a design in space characterized by large, open, non-intersecting movement as is shown in the flow chart of his ASL poetic version of the line. Thus we have illustrated three aspects of poetic structure in examining closely one poetic line, and the way in which it has been molded, shaped and changed in passing from non-poetic straight signing to poetic art-sign. There is Internal Poetic Structure involving the choice of signs - in this case, perhaps an analogue of alliteration. There is External Poetic Structure, in-

Poetry and song in a language without sound

Figure 7.

FLOW CHARTS OF MOVEMENT: unlabeled the straight ASL version represent transitions.

67

sections of the arrows in

ART-SIGN

volvmg a balance bettieen the hands (by alternating hands in making one sign after the other and by holding one sign while making another) and involving a flow of movement, a continuity from one sign to another throughout the line (distortion of the transition between signs, in this case making the final position of one sign coincide with the initial position of the next). This merges into an external Kinetic Superstructure, having to do with creating a spatial, rhythmic design, superimposed on the signs themselves.

4. Analysis of original artsign in ASL Because of the basic difference in the mode, the ASL signed translation of the E. E. Cummings poem involves even more than the standard problems of translating poetry from one spoken language to another: A constant struggle to retain the meaning of the original, to capture some of its structural characteristics and at the same time to create poetic structure appropriate to the language of the translation. The analysis we ‘have presented shows how much Bragg was concerned with these aspects of

68

Edward S. Klima and UrsulaBellugi

translation, and his remarks about the various decisions he made in selecting signs indicate that he was very much aware of these problems.

4. I. ‘SUMMER

“from four haiku poems composed

in English and ASL

In this evolving ASL poetic tradition there has also been produced some original poetry in sign. “The Seasons” by Dorothy Miles is a special example - special in that it was composed simultaneously in ASL and in English. Dorothy Miles is a deaf woman, profoundly deaf since the age of eight, who has a brilliant command both of ASL and of English. She formerly acted in the National Theater of the Deaf and is now associated with California State University, Northridge, a university in California with a special program for deaf students. After analyzing the poem we had an opportunity to discuss the poetic process with Miss Miles. We were particularly interested in her decision to compose the poem simultaneously in ASL and English rather than in ASL alone. Her response was that such simultaneous composition was her own special style of poetic expression. The sequence is subtitled “Four Haiku Poems”. We have the feeling that the particular compression and rich imagery of poems in haiku style are especially suited to sign language. For our analysis of her poem, and long before our discussions with her, we had Miles’ ASL rendition recorded by her several times on video tape. We felt it important to have different recordings in order to see how much variation there would be that would be attributed simply to factors involved in any individual performance. We shall first restrict our structural analysis of Miles’ ASL and English versions to the verse entitled “Summer”. Later, we shall consider a different rendition of this verse as signed by another signer and also that signer’s rendition of the ASL version of “Winter”, another verse of Miles’ haiku sequence. Of course, our attention will focus on the ASL versions. The following remarks about the English version of “Summer” must suffice. The pure ‘text’ of that version of the verse goes as follows: “Green depths, green heights; clouds and quiet hours - slow, hot, heavy on the hands.” 4. I. 1. In ternal structure of English version As to conventional structure, Miles chose to cast the English version into standard haiku form. Accordingly, each verse has three lines at one level of structure; the first and the last line with five syllables each, and the middle line with seven syllables.

Poetry and song in a language without sound

Internal Poetic Structure Conventional (Haiku) Green depths, green heTghts, clouds Aiid quiet h&i-s, slow, hot, Heavy ijn the hands

69

Syllable count 5 7 5

Superimposed upon this Conventional Poetic Structure, however, is an Individual Internal Poetic Structure in the English version involving, among other things, repeated patterns of similar sounds. In terms of this level of structure, “Summer” is best analyzed as consisting of four structural ‘lines’ with the end of each line delineated by an alliterative word sharing an initial ‘h’ (heights, pours, hot, &znds - /JOUS, of course, constituting orthographic alliteration rather than phonetic alliteration [see Table 41). That the Table 4.

Individual Internal Poetic Structure B

A I areen II

qlouds

depths, and

Ill Slow, IV Heavy

Mreenaeights Quiet

<$jours,

@t on

the@ands

I = II = Ill = IV (based on line-final alliteration on /h/ or orthographic ‘8”) IA = IB (based on hemistich-initial alliteration on /g/j IIA = IIB (same as the preceding relationship but on /k/) Thus IA : IB :: IIA : IIB IVA parallels IVB (a similarity based on each hemistich simply containing a member of an alliterative pair: heavy and hands - a looser symmetry manifested also by I and II) Thus IVA : IVB :: IA : IB:: IIA : IIB

repeated velar /g/ in the two occurrences of green in the first line is structurally significant is highlighted by the fact the equivalent positions in the next line similarly contain an alliterative pair, in this case each with an initial /k/-sound (also velar): clouds and quiet. The two half-lines (hemistiches) of the first two lines are thus struct&ally equivalent. The first two lines also share a looser sort of symmetry with the final line whose two hemistiches are similar to those of the first two lines in that each contains an alliterative pair - based on /h/ in the final line, where, however, the alliterative pair does not occur in the same position in the hemistich. Table 4 presents a sketch of these aspects of the structure of the English version of this verse.

70

Edward S. Klima and Ursula Bellugi

4. I. 2. Internal structure of rendition A in ASL The ‘text’ (i.e., sequence of signs in their citation of “Summer” as signed by Miles runs as follows:

form) of the ASL version

“GREEN DEEP BELOW, GREEN HIGH ABOVE; WHITE AND QUIET HOUR - SLOW, HOT, HEAVY ON HANDS”. Miles’ ASL rendition suggests division in terms of verse structure: line I line II line III

GREEN WHITE SLOW,

DEEP CLOUDS HOT,

BELOW,

into three GREEN AND HEAVY

CLOUDS

basic units - ‘lines’ HIGH QUIET ON

ABOVE; HOUR HANDS.

Table 5 is an actual tracing from the video tape itself. From even a very cursory examination of the ‘text’ and restricting our analysis to one parameter alone ~- that of Hand Configuration in the overall manual arrangements used - it is immediately clear that this verse constrains itself to a very restricted number of the hand configurations occurring in the language, variously estimated at between 19 (Stokoe et al., 1965) and 40 (Woodward, 1973), depending on the basis of analysis. In the sixteen sign tokens occurring in the verse (only fifteen different signs since GREEN appears twice), the very similar handshapes notated here a: ‘Five-finger’ hand occur in the citation form of thirteen of the signs, sometimes as an active hand, sometimes as a base, sometimes as both. The variants of ‘Fivefinger’ hands that occur in this verse are the following in the notation developed by Stokoe et al. in the Dictionary of ASL.

Figure 8.

‘Five-finger hand’: Handshapes with all five fingers extended

.. .

5

5 (bent and spread)

(straight

B and spread,

(straight

and compact)

t=l n

12

EdwardS. Nimaand Ursula Bellugi

Table 5, (Rendition A)

DEEP

HOT

BELOW

HEAVY

GREEN

“SUMMER ” ,bv Dorothv Miles (Mtles Renditions

GREEN

SlC%!J

HIGH

ON

ABOVE

HANDS

Readers: Leave flap open for later reference

Poetry and song in a language without sound

73

In addition, through a distortion which is part of the External Poetic Structure of the verse, this same ‘Five-finger’ hand is found as some aspect of all the signs in the first line after the first GREEN. It occurs as a part of the sign, even in citation form, in the case of DEEP, BELOW, and ABOVE. In addition, although HIGH and GREEN are both normally one-handed signs occurring without another hand in ordinary signing, Miles keeps the left hand as a kind of reference base or “surface indicator” throughout the signing of DEEP BELOW, GREEN HIGH ABOVE. This provides a consistency of form to the signs of the first line, and is a poetic modification of the signs GREEN and HIGH. Thus, some aspect of the final form of every sign in the verse, with the exception of the first GREEN, involves a ‘Five-finger’ hand, and most of the signs are restricted to that handshape. In poetic structure, more significant than mere frequency is patterning - in this case the patterning of the restricted set of hand configurations used in the verse. The first line has two parallel halves, each beginning with the ‘Index’ hand (i.e., in the first and second GREEN) and each ending with an active ‘Five-finger’ hand operating respectively below and above a base ‘Five-finger’ hand (i.e., in BELOW and ABOVE), in similar arcs, as Table 5 reveals. The second signs of each half of the first line are DEEP and HIGH; DEEP uses an ‘Index’ hand as active as does GREEN, and HIGH uses a hand which we can call ‘Index + mid’. Those two handshapes are minimally different from one another - extension of the index finger as opposed to index finger along with middle finger. Figure 9.

‘Index’

(G)

As we have noted, the of DEEP is prolonged as during the signing of the ABOVE. This extension

‘Index-mid’

(H)

base ‘Five-finger’ hand proper to the citation form a surface indicator in BELOW, and then extended second GREEN and HIGH, and maintained during of the base ‘Five-finger’ hand through the signing

7’4

Edward S. Klima and Ursula Bellugi

Outline of the Individual Internal Poetic Structure

Table 6.

of “SUMMER”

(Miles’

rendition) in terms of Hand Configuration

II*

Five

B

0

Five

w/-//TE

:

EA. Five Five

/ Five

SLOW

HOT Five

B

: Five

Five

/ Five

ON

Five

/ Five

HANDS

HEOVY

The ‘Five-finger’ hands are abbreviated here as Five. A colon (:) indicates two symmetrical hands. A slash indicates a two-handed sign, with the active hand to the left. The boxing shows the parallelisms developed within each line.

Poetry and song in a language without sound

75

of HIGH constitutes an element of External Poetic Structure that further enhances the similarity between DEEP and HIGH. Of course, in addition, the two halves of the first line are semantically patterned as well. The first signs of each are the same - GREEN and GREEN; the second signs in each are opposite - DEEP and HIGH; as are the third signs in each - BELOW and ABOVE. In the second line, WHITE CLOUDS AND QUIET HOUR, there is further Internal Poetic Structure. WHITE and AND are each one-handed signs which have a ‘Five-finger’ hand closing to a ‘tapered 0’ and each is followed by a two-handed ‘Five-finger’ sign (CLOUDS and QUIET). This is all the more striking as a device resulting in Individual Internal Poetic Structure in the ASL version of the poem in that the sign WHITE, the first sign setting up this pattern, is not represented by a word in the English version. Finally, HOUR, the last sign of the second line, echoes in its active right hand the ‘Index’ hand motif characteristic of the first line and combines it with the ‘Five-finger’ hand that dominates the second line and, in fact, the whole verse. The third (and final) line of the stanza, SLOW, HOT, HEAVY ON HANDS, consists exclusively of uses and interactions of the ‘Five-finger’ hand in signs which are made in front of the chest with the hands touching or in close proximity and which vary in the relation of the hands’ movement, intensity, and orientation. Table 6 gives an outline of the Individual Internal Poetic Structure of the poem in terms of Hand Configuration. 4.1.3. Internal structure of rendition B in ASL In order to study the heightened use of sign language, we have sometimes asked several different signers to create individual poetic renditions starting from the same poem. In the case of Dorothy Miles’ haiku poems, for example, one of the ‘native’ signers, Lou Fant, who has also been with the National Theater of the Deaf, was videotaped while performing his rendition of the poems for us on different occasions. He began from the English version, and then created his own ASL rendition from that. In order to show the different aspects of ASL which can be used in art-sign, we will present an analysis of Fant’s rendition here, and then continue the investigation of different levels of poetic structure side-by-side in Fant’s and Miles’ renditions of the same verse. Table 7 shows Fant’s rendition in actual tracings of selected images from the screen of the videotape monitor. The choice of signs in Fant’s rendition is not radically different from that of Miles. However, Fant exploits one kind of handshape similarity in the first and second lines which is worthy of mention. But let us consider first the differences. Fant makes the title a part of the first line of the poem, as is

76

Edward S. Klima and Ursula Bellugl

indicated by his phrasing. He expresses ‘depths’ of the English version not by a separate sign but rather by extending the sign GREEN in a wide sweep of the arm which gives the impression of the sign GREEN moving into the horizontal distance away from him. ‘Green heights’ is expressed with two signs; first, the sign GREEN, this time moving upward with a sweep of the arm overhead; and then the sign HEIGHTS. Fant’s first line, then, is: SUMMER: GREEN-depths,* GREEN HEIGHTS. His rendition differs from Miles’ in the direction of more structural regularity: All four signs of his rendition of this ‘line’ involve an ‘Index’ hand as active. SUMMER begins with an ‘Index’ hand and closes as it moves across the forehead. GREEN both times involves an ‘Index’ hand with a slight variation in the thumb position. And finally, the sign HEIGHTS (instead of Miles’ HIGH, with ‘Index + mid’) is made with an ‘Index’ hand acting on a base. The dominant structural motif of the ‘Index’ hand in the first line is echoed by the ‘Index’ hand active in HOUR, the last sign of the second line of Fant’s rendition: WHITE CLOUDS AND QUIET HOUR. Finally, in keeping with the restricted set of hand configurations used in the rest of the verse (all variants of the ‘Five-finger’ hand), the sign HEIGHTS that Fant chose has a ‘Fivefinger’ hand as its base. In the second line, we find the same parallelism that was in Miles’ ASL rendition: The first signs of each hemistich, WHITE and AND, are both made with a ‘Five-finger’ hand closing to an ‘0’ hand, and each is followed by two-handed signs made with ‘Five-finger’ hands (CLOUDS and QUIET). The final line, as in Miles’ version, is composed of signs which use only ‘Five-finger’ hands: SLOW, HOT, HEAVY ON HANDS. Thus, we have again a sense of ‘alliteration’ in Fant’s version: The ‘Index’ hands are characteristic of the signs in the first line and echoed in the sign at the end of the second line; and the ‘Five-finger’ hands predominate throughout the second and third lines of the poem. 4.1.4. External structure of rendition A in ASL The type of external structural patterns that we noted in Bragg’s art-sign version of “since feeling is first” and that we shall discuss in Fant’s rendition of “Summer” are largely absent from Miles’ rendition of her own poem in ASL. This helps us to understand that the particular distortions and *GREEN-depths: The incorporation, into the meaning of the sign GREEN, of the extent of greenness is a modulation of the basic form of the sign; we symbolize such modulations by adding an English equivalent in minuscules to the base form in capitals, thus GREEN-depths. In fact, the second occurrence of the sign GREEN in this ‘line’ also includes a modulation indicating that the greenness extends also far up. A more precise glossing could have represented the hemistich by, for example, GREEN-soaring HEIGHTS.

I

I

I

I I

“SUMMER ” bv Dorothv Miles

(Fant Rendition)

Readers:

Leave flap open for later reference

Poetry and song in a language without sound

79

mechanisms characteristic of External Poetic Structure are by no means a necessary condition of art-sign or of poetic effect. As we shall see in the next section, however, Miles achieves an art-sign structure in other ways. In our conversations with Miles (as a matter of fact, after our analysis), we discovered that it was her intention to keep the signs as close to their normal form as possible. We find little spatial displacement, little extreme manipulation of the signs from their more nearly citation form. Miles does not, as some other sign-poets do, alternate in the use of the hands. Miles uses her right hand as active for all the one-handed signs (GREEN, HIGH, WHITE, AND, and HOT) and for all the signs involving one hand acting on another as a base (DEEP, HOUR, SLOW, and ON). There is very little overlapping of signs: During the one-handed signs there is no attempt to make use of the left hand; it is either by her side as in HOT, or it is off to the side and without a specific shape, as in WHITE and AND. Miles is a right-handed signer, even in her poetic rendition, and she does not alternate hands to impose a balance in the use of the two hands, nor does she make a special effort to overlap signs. We also find little evidence of other kinds of distortion. We did point out that Miles adds a ‘surface indicator’ (in the ‘Five-finger’ hand) to the signs GREEN and HIGH to create a continuity in the first line of the verse. One sign is definitely exaggerated: Her sign for SLOW starts as the normal sign does, moving along the back of the base hand, but then it becomes exaggerated and distorted, moving slowly up the whole length of her arm. This allows for a smooth transition between the sign SLOW and the sign HOT, for at the end of SLOW the hand is up near her shoulder and, thus, much nearer the starting position of HOT (at the mouth) than it normally would be. There is one other change from the citation form of a sign - a change that results in a special patterning in hand orientation. The sign HANDS is ordinarily made with hands compact and palms down. Miles modifies the form of the sign so that the orientation of the hands is palm upward. The structural motivation for the change becomes clear when we examine the palm orientations of the signs that precede HANDS in the ‘line’. SLOW, the first sign in the line, has palms down. The second sign in the line, HOT, has within itself a change in palm orientation from palm up to palm down. The next sign, HEAVY, has palms up, followed by ON with palms down. Finally, comes the sign HANDS, which with Miles’ alteration in palm orientation, here has palms up and thus continues the pattern of alternating palm orientations set up by the citation forms of the preceding signs: down - up down - .-up - down - up .* Actually, this change in the form of the sign ___ *In the Miles rendition, after the final sign HANDS the hands move slowly to a relaxed, neutral position - right hand on left, significantly with both palms facing down, of alternating hand orientations already established in the ‘line’.

thus continuing

the pattern

80

Edward S. Klima and Ursula Bellugi

HANDS is a substitution of one prime of the parameter of Orientation for another; since the change remains within the linguistic code of ASL, this particular motivated distortion should properly be considered as an aspect of the Internal Poetic Structure of the verse. Miles also makes one further change in the sign HANDS, making the hand loose and somewhat spread (rather than compact, as in the citation form). The result of this change is that all of the signs of the ‘line’ are uniformly loose. Even with these minor variations in the form of signs, Miles clearly does not intend to make major distortions in the interests of creating any semblance of what could appropriately be called a ‘design in space’. In Miles’ version of “Summer” (and this characterizes the other ASL verses by her as well) the signs are made within the normal signing space. This is quite different from the Kinetic Superstructure involving exaggerated spatial displacement that Bragg imposed. Miles does choose her signs carefully to create an Individual Internal Structure, and we shall see that she uses a very special type of External Superstructure, quite different from the other heightened forms we are presenting here. 4.1.5. Rhythmic Superstructure of rendition A in ASL A careful examination of Miles’ rendition of “Summer” reveals a special sort of Superstructure, not spatial but rather temporal and rhythmic. In the first place, each of the three ‘lines’ of this verse is of about equal temporal length; i.e., each takes 7.5 seconds (+ 0.3), although the length of the individual signs varies. The first and second hemistiches of the first ‘line’, as well as the last hemistich of the final ‘line’, each show a pattern of four accents and enclose a series of three internal hemistiches with fewer accents. Furthermore, there is a special rhythmic patterning to the three enclosing hemistiches with four accents each. The rhythm of the first hemistich of the first ‘line’ is repeated with only a slight variation in the following hemistich: there is syncopation on the second accent - the second hand twist - of GREEN. Then in the final hemistich of the verse, that variation is itself repeated, but again with a slight modification: the syncopation is broken down into two separate and distinct accents. Table 8 sketches this rhythmictemporal superstructure, which we are tempted to compare with certain characteristics of recitative as opposed to aria. * The slurred, half-staccato notes representing GREEN, CLOUDS, and HANDS indicate two distinct accents within one sign. Rest notes indicate long transitions between signs. The final long rest at the end of IIIB represents an exaggeratedly slow return of the hand to neutral position (not *This comparison I.R.C.A.M. seminar.

was first

suggested

to us by Linda

Vickerman,

one of the participants

of the

Poetry and song in a language without sound

81

represented in the sketch in Table 5). The relative length of the signs as represented by the notes was calculated by counting individual fields on the video tape (60 fields per second).* Table 8.

Rhythmic-temporal superstructure of ‘Summer” (Miles rendition) z3

A

,n .. 1

I

GREEN

cl-

DEEP BELOW

GREEN

sJ

II

J

cl. ol SLOW IA

HOT

parallels

H&40’ IB

parallels

1115

in

QUlET

rhythmic

7.2

sec.

7.7

sec.

HOUR

n,r.. ON

sec.

ABOVE

I J

AND

aI

HIGH

7.5

HANDS structure

I = II

= III

4.1.6. External Structure of rendition B in ASL While Miles’ rendition shows a very special type of External Rhythmic Superstructure that is not characterized by the modulations wrought on the citation form of signs by Bragg, Fant’s rendition is very much in keeping with the ,tradition of creating an external structure by modification in the form of signs themselves or in aspects of their presentation. a. Patterned alternation of the hands. As can be seen immediately from the tracings of the video tape presented in Table 7, in terms of External Poetic Structure, Fant makes a very decided use of the alternation of the hands. That is, he does not always use his right hand as the active hand in one-handed signs or signs which involve one hand acting on the other as a base. SUMMER and GREEN-depths are signed with the right hand active; *The Systems

individual fields were Design, Model #440-2)

numbered with the aid of a Video for ease of counting.

Frame

Number

Generator

(Data

82

Edward S. Klima and UrsulaBellugi

the second GREEN is signed with the left hand active; HEIGHTS returns to the right hand active. WHITE in the second line again changes to left hand active. CLOUDS is a double-handed sign, but AND, which follows it, is made with the left hand active. QUIET is a double-handed sign, and HOUR again has the left hand as active. In the third line, SLOW has the right hand active as does the sign that follows it, HOT, HEAVY is a two-handed sign; ON has the right hand active; and HANDS has within itself an alternation of hands: The right hand acts on the left and then they interchange, the left acting on the right. We thus see that Fant alternates the hands consistently in the first two lines. The third line is so heavy with two-handed signs that alternation of the hands does not occur. HOT, the only one-handed sign in the ‘line’, is even accompanied by the other hand (the left) in a definite hand shape. This is motivated by the sign which follows HOT as will be seen in lines lo- 12 in Table 9. From the alternation of the hands in the second ‘line’, a definite pattern emerges: WHITE(left)

CLOUDS(both)

AND(left)

QUIET(both)

HOUR(left

active)

where the first, third, and fifth signs have the left hand active, while the two intervening signs are made with both hands. This pervasive alternation of the hands is clearly part of a general pattern of alternations. Table 9 summarizes the overall scheme of hand alternations in Fant’s rendition of “Summer”. b. Overlapping of signs. The alternation of hands contributes to another poetic mechanism which is characteristic of Fant’s rendition of “Summer” and also of Bragg’s translation of the line from Cummings’ poem, for such hand alternation permits the overlapping of even one-handed signs occurring in sequence. By such overlapping - where the form of a just-executed sign is maintained with one hand while the next sign is made with the other hand - two signs are, in effect, presented simultaneously to the eye. That hand alternation does contribute to the possibility of overlapping signs becomes clear when we take into account the fact that a great many signs of ASL involve one hand only. (Of the more than 2,000 signs in the Dictionary of American Sign Language, Stokoe et al., 1965, about 40% are one-handed signs.) In everyday signing, one-handed signs are generally made with the dominant hand, and with these signs, the non-dominant hand is not part of the signing picture - either it is at the signer’s side or it is lax in shape and position. This is exemplified in the video tape tracings of the onehanded signs in Miles’ rendition of the verse “Summer” (Table 5). As Miles signs the first GREEN and as she signs HOT, her left hand is by her side. Both of these are one-handed signs. As she signs the one-handed signs

Table 9.

Alternation

of hands and overlapping signs in Fant’s rendition of ‘Y%.unmer”

by Dorothy Miles

f “B’‘-hand

“G’‘-hand

-(closing “X’‘-hand

The varieties

of

“5’‘-hand

the

“Five-finger”

hand

...

“5”-hand

to)“0” hand

A symbol written with dots indicates that the hand configuration of a previous one hand while subsequent signs are being made by the other hand.

sign is maintained

with

84

Edward S. Klima and Ursula Rellugi

WHITE and AND, her left hand is lax and not part of the signing picture. Whereas this laxness of the unengaged, non-signing hand is typical of the presentation of one-handed signs in everyday signing, Fant never has a lax, ‘unused’ hand after the first two signs of the verse (see Table 7). In fact, Fant emphasizes this modification of one aspect of the presentation of signs and raises it to the status of a major structure-creating mechanism. After the second sign of the verse, both hands are engaged through the verse, and there is much overlapping of signs, so that wherever possible two signs are present simultaneously to the eye. With the one-handed signs the other hand is engaged in the following ways: The second GREEN is made with the left hand, but the hand position and configuration of the first GREEN is still held in the right hand. The sign WHITE is made with the left hand, but the active right hand of HEIGHTS, which precedes it, retains its shape throughout the sign WHITE. The sign AND is made with the left hand, but the final position and shape of the right hand for CLOUDS remains through AND. In the final line, the sign HOT is made with the right hand, while the left hand anticipates the shape and position of the sign which will follow: HEAVY. Thus, Fant is overlapping signs (and images) by holding part of one sign while making another. (Recall that this occurred also in one type of simultaneity of signs used in plays on signs.) Both of these mechanisms - patterned alternation of the hands and overlapping of signs - contribute to keeping the use of the hands in balance throughout the poem. Table 9 shows the interaction of the two devices in Fant’s ASL rendition of “Summer”. c. Flow of movement. As was the case in Bragg’s translation of the Cummings poem, Fant’s rendition of “Summer” is characterized by pervasive manipulation of the transitions between signs. (Pertinent transitions can be interpolated from Table 7.) For example, the second sign of the verse (GREEN-depths) begins at the same level as the preceding sign rather than in the ‘neutral space’ in front of the signer’s chest, as would be the case in the citation form of the sign. While the left hand forms the next sign - the second occurrence of GREEN, even more distorted than the first in its sweeping movement and its exaggerated ascent to a final position way above the head, and, in fact, far above the normal ‘signing space’ - the right hand maintains the hand configuration and the final position of the previous sign (GREEN-depths). In that position, the right hand turns into the active hand forming its part of the sign HEIGHTS and sweeps up to contact the base hand of HEIGHTS, which itself has maintained the final position of the justprior sign (GREEN). Similarly, while the left hand signs WHITE, the right hand maintains the final exaggeratedly high position of HEIGHTS and from that position begins the sign CLOUDS. The sign AND, which in ordinary

Poetry and song in a language without sound

85

signing is made in the ‘neutral space’ in front of the signer’s chest, (compare, for example, Miles’ rendition in Table 5) is signed in Fant’s rendition at the same exaggeratedly high position above the head as the sign CLOUDS. The ordinary transition between the two signs from above head downward, at least toward the chest, has been eliminated. In the rest of the signs of the verse as well, the final position of each becomes the starting position of the next sign. A final remark along these lines is appropriate for the sign HOT. HOT itself is a one-handed sign, in this case signed by Fant with his right hand. The left hand in such a case would normally tend to move at least toward a neutral position, as it does in Miles’ rendition. But in this instance of art-sign, the free left hand instead assumes an orientation with palm upward more similar to that of the initial position of the left hand which is engaged in signing HOT. The free left hand, maintaining that neutral orientation and position, then simply turns into one of the hands of the symmetrical two-handed sign HEAVY, thus anticipating it, as mentioned earlier - without any superfluous movement at all (cf., Table 7). It is in this sense that the signs of certain types of art-sign ‘flow into one another’. 4. I. 7. Kinetic Superstructure of rendition B in ASL Very striking visually in the Fant rendition is its superstructure of space and movement, taking the signs not only out of the normal signing space but also creating a very obvious design in space consistent, in this instance, with the theme of the verse: heaviness. I.e., beginning with the second verse the signs slowly descend from far above the signer’s head (a location which is not used in everyday signing) to below the waist. In terms of this particular superstructure, Fant’s deviations from the ordinary place of articulation are considerable. In fact, in the first two ‘lines’ all the signs except SUMMER and WHITE are signed much higher than they would be in ordinary signing. The upward displacement is particularly striking in AND and QUIET. As was mentioned above, AND is normally signed in the neutral space in front of the chest and QUIET normally begins in front of the mouth. (cf., Table 5 for the normal form of these signs.) In fact, even the sign CLOUDS, which is normally signed just above forehead level and would, in everyday signing, have the highest place of articulation of the fourteen signs found in this rendition, is raised. A further distortion associated with this generally.descending line involves the two-handed sign HOUR, which has one hand active and uses the other hand as a base. In everyday signing the base would stay in the same location (in front of the chest). In Fant’s rendition, the sign is higher than it ordinarily would be, and the sign as a whole slowly moves down - base and active hand both.

86

Edward S. Klima and Ursula Bellugi

Given that the first GREEN sweeps horizontally across the area in front of the signer’s face from the far left to the far right, we might consider that the general superstructure has taken the shape of a cross, sweeping horizontally at a rather high level from one side to another, then moving directly upward to a position high above the signer’s head, and gradually moving each sign heavily and slowly downward until the final sign is made with the body bent over, shoulders haunched, and the hands low in the signing space. Table 10 shows the signs involved in the dominant descending line.

Table 10.

Kinetic Superstructure

oj’ “SUMMER *’

HOUR

ON

Poe@ and song in a language without sound

4.2. External

Poetic Structure

87

in the ASL verse “WINTER ”

The final verse of the Miles haiku follows: Winter: Contrast: Black and white Bare trees, covered ground, Soft snow. Birth in death.

sequence

is, in the English version,

as

hard ice,

Again, Miles has constructed the English version so that it maintains the haiku 5 - 7 - 5 syllable-per-line pattern. We have studied a number of different renditions of this verse by different signers. Here we shall present the verse in Lou Fant’s ASL rendition because of some of the interesting manipulations of signs which characterize its poetic form. The rhythm of his signing and the internal patterns suggest four structural ‘lines’ with the title WINTER included in the first line. The glosses for the Fant rendition are as follows: line I line II line III line IV

WINTER: BARE HARD BIRTH

CONTRAST TREES, ICE, INTO

WHITE; BLACK, GROUND; COVERED SNOW; SOFT DEATH.

We shall restrict our considerations here to the external structure of the verse. (Table 12 presents tracings from the screen of the video tape.) One distinguishing feature of the External Structure of the art-sign rendition of “Winter” by Fant is clearly the constant involvement of both hands. In their citation form, all but two of the signs in Fant’s rendition of “Winter” are two-handed signs. The exceptions are BLACK and WHITE. By alternating hands and overlapping, Fant brings these two one-handed signs also into the pattern. As can be seen from Table 12, in Fant’s rendition the final position of the right hand of the symmetrical, two-handed sign CONTRAST is held while BLACK is made with the left hand; then the final position of BLACK itself is held with the left hand while WHITE is signed with the right hand. A comparison of this sequence of three signs in Fant’s rendition with the same sequence in ‘straight’ signing (Table 11) reveals that there is more involved in the External Structure of this rendition than merely the overlapping of signs. The signs also undergo a distortion that is part of a pattern of distortions characteristic of the other signs in the verse and is consistent not only with the general theme of the verse, winter’s contrasts, but also with thematic oppositions that are developed in the verse: ‘black’ versus ‘white’, ‘bare’ versus ‘covered’, ‘hard’ versus. ‘soft’, ‘birth’ versus ‘death’. It is not the case in this verse that Fant develops an

88

Edward S. Klima and Ursula Bellugi

External Poetic Structure that creates a flow of movement to thread all the signs together in a single consistent way. Rather, he manipulates certain sequences of signs so that either their contrast in form or their similarity is exaggerated. Let us consider first the pair of signs BLACK and WHITE, which constitute thematic opposites in the verse. Whereas, in citation form the movement of the sign BLACK is a brush along a short line across the forehead (cf., Table 1 l), in Fant’s art-sign rendition the hand makes an exaggerated, large, sweeping motion from the forehead in an arc downward and toward the extreme Zeft of the signer, where it is held while WHITE is being made. The sign WHITE itself is not only exaggerated; it is, in fact, in a literal sense, diametrically separated from BLACK and opposed to it in its position in space. For whereas the citation form of WHITE carries the hand from contact with the center of the chest straight out to the neutral space directly in front of the chest, in Fant’s art-sign rendition the hand shoots out, this time to the extreme right of the signer - where its final exaggerated position forms a real spatial ‘contrast’ to that of the left hand still held, at the extreme left, in the final exaggerated position of BLACK. Table 11.

Signs from beginning of Miles’ “Winter” (Fant Rendition)

STRAlGHT ASL

BLACK

CONTFtAST

WHITE

ART-SIGN

;&_*&f__

I

1 CONTRPST

1 &cu

WHITE

The form assumed by the pair of signs BLACK and WHITE of the second hemistich of the first ‘line’ in Fant’s poetic rendition bears a special relation to the just prior sign CONTRAST in the first hemistich. As Table 11 indicates, the form of the two-handed sign CONTRAST in the poetic version

Table 12. ” VWVJER

I

’

” by Dorothy

WNTER

BARE

Mks

HARD

(Fant

Rendbon)

CONTRAST

TRkS

(transrtm)

(trarls,tm)

-

GROUND

WHITE/

-

Poetry ami song in a language without sound

/BLACK

COVERED

h

>

-

-

89

aaualajal Aale JOJ uado de14 aneal mapeal

,

Poetry and song in a language without sound

91

is itself exaggerated with respect to the spatial displacement of the two hands to the extreme left and extreme right of the signer. Significantly, the exaggerated sweep of BLACK, the sign that follows next, ends in a final sustained position that is an echo of the final position of the corresponding hand in CONTRAST. Next, the right hand, as it shoots out to the extreme right in the sign WHITE, reaches a final position that echoes the final position of the right hand of CONTRAST. In the other ‘lines’ of Fant’s art-sign rendition of “Winter”, there is similar interplay between the form of the signs within hemistiches and between hemistiches. Let us consider first the pair of signs BARE and TREES constituting the first hemistich of the second ‘line’ (cf. Table 12). In citation form, BARE would involve a straight, short movement along the back of the base hand, by an active hand that is flat and spread, the middle finger bent inward. The same handshape occurs in FEELING discussed in relation to Bragg’s translation of the Cummings line and exemplified in Figure 1 and Table 2. In citation form, TREES is made with a ‘Five-finger’ hand, the fingers straight and spread. In Fant’s art-sign rendition of the sequence, the sign BARE is extended in an upward-sweeping arc until the handshape of BARE has ‘become’ the active hand of TREES, with one handshape substituted for another. Similarly, the base hand of BARE slides under the elbow of the active right hand and with minimum transitional movement ‘becomes’ the base of the sign TREES. Through these distortions, the similarity between the two signs of the hemistich has been enhanced and a continuity between them has been superimposed. (In addition, the substitution of the handshape proper to BARE for that of TREES directly associates the notion of ‘bareness’ with that of ‘trees’. In terms of form, this change does not go beyond the linguistic system itself and thus actually figures in the Internal Poetic Structure.) Likewise, Fant’s art-sign rendition superimposes a continuity between the members of the hemistich pair COVERED and GROUND. An alternation of the hands with overlapping of signs occurs with this pair. The left hand already takes the configuration and the location of the sign GROUND as the sign COVERED is made with the right hand above it. The sign GROUND is generally made with two hands, but here it is one-handed in form, fingertips rubbing together below the hand that has maintained the handshape of the sign COVERED. Further continuity of signs within the hemistich is manifested by the pair HARD and ICE, both involving extra tension in the hands and arms in Fant’s poetic rendition. SOFT and SNOW, the next pair of signs, differ markedly in their formation from the two signs of the preceding hemistich in that SOFT and SNOW both involve an extra laxness of the muscles of the hands and arms.

92

Edward S. Klima and Ursula Bellugi

The transition also between the two pairs of signs forming the two hemistiches of the third ‘line’ is manipulated to play a part in the External Poetic Structure of the verse (i.e., the transition between HARD ICE and SOFT SNOW). In everyday signing, the final position and hand configuration of the sign ICE in a similar context would be held briefly with muscles tense, but then the muscles and hands would relax before the hand turned upward and progressed through its transition to the sign SOFT. By contrast, in this art-sign rendition the hands maintain the muscle tension and the bent ‘Fivefinger’ handshape of ICE throughout the transition to the sign SOFT, only relaxing at the last moment, when the ‘soft’ downward movement of SOFT begins. (The effect of this transition is represented in line III of Table 12.) As a matter of fact, the exaggeration of the transition between the two hemistiches of the third ‘line’ turns out to be part of a definite pattern of transition manipulation between the two halves of each of the first three ‘lines’. For in the preceding line, the transition between TREES and COVERED is also exaggerated. After the movement appropriate to the sign TREES (a twist from the wrist of the vertically oriented hand and forearm), the transition to the next sign begins. But rather than the active hand and forearm becoming lax and the elbow slipping from its contact with the back of the base hand (characteristic of the transition between these signs in everyday signing), the whole forearm retains a quality of tense articulation and, maintaining elbow contact with the base hand, slowly descends from vertical to horizontal position as it describes a 45-degree arc. (See line II of Table 12). The pattern of exaggerated transition is also represented between the hemistiches of the first ‘line’: When the sign CONTRAST is completed, the index fingers of both hands (the symmetrical ‘Index’ hand configuration appropriate to CONTRAST) are kept stiff throughout the transition to BLACK and the left arm moves in an exaggeratedly broad and sweeping curve up to where BLACK begins on the forehead. (See line I in Table 12.) By contrast, it is once again continuity that characterizes the three signs comprising the final ‘line’ of this art-sign rendition of the verse. The three signs BIRTH INTO DEATH are blended together into one continuous flow. The signs all have the same hand configuration, are all two-handed, and are all made in the same location in front of the torso, with the hands close together or touching. The main difference among the three is in the orientation of the hands and their movement. In fact, the sign INTO is indicated in so minimal a way that it is almost lost, nearly becoming just part of the transition between BIRTH and DEATH.

Poetry and song in a language without sound

4.3. Revitalization

of iconic-pantomimic

93

associations

In concluding, mention should be made of an additional phenomenon that is particularly prominent in art sign (though apparently not creating definite patterns): the intensification of representational aspects of signs. In all of the ASL renditions of the verses of Miles’ haiku sequence, there are signs whose representational aspects are exaggerated. One means we employed for assessing the degree and effect of such intensification was the following: we asked native signers who had not read the English version to record their impressions of the various renditions after they had viewed them several times on video tape. These reviewers were asked to include in their remarks a discussion of any exaggerations or distortions they had observed. Let us consider first the title SUMMER, the first sign in Fant’s rendition of that haiku verse. The citation form of SUMMER involves the bent index finger (the ‘X7-hand) brushing across the central part of the forehead. But, in the words of Mr. Shanny Mow, a deaf signer reviewing the video tapes, Fant elaborates the sign by “increasing its length . . . thus producing a more pantomime-like action”, with an outstretched index finger (a ‘G’-hand) that gradually bends into an ‘X’-hand, “Fant ‘wipes’ the entire length of his forehead”. The ‘wiping’ is presumably the wiping, from the forehead, of the sweat from summer’s heat. In Fant’s rendition of the verse, the sign CLOUDS also undergoes iconic elaboration. Mow describes this in the following words: “[Fant] modulates CLOUDS by a loose balling action and by rotation of the hands slowly across the space overhead - and the clouds even move.” In other words, the movement of the sign itself directly portrays the drifting of the clouds. Finally, Mow comments on the representational aspects of the exaggeration in the formation of the sign HEAVY in Fant’s rendition: “HEAVY certainly looks heavy, so heavy that the bottom drops. In this action, there is no [actual (author’s interpolation)] suddenness; yet one gets the feeling there is. This doesn’t occur in the regular form of the sign but it surely gives finality to the sinking effect Fant has produced as he moved his signs downward. One begins to feel the oppressive claustrophobic heat and time standing still as the long summer drags on.” In all renditions of “Summer”, (Tables 5 and 7) the sign SLOW is treated iconically - lengthened both in duration and in space. In the citation form of SLOW, the fingertips of the active flat ‘B’-hand brush once over the back of the base ‘B’-hand from the fingertips to the wrist of the latter. In Miles’ rendition of her own poem, SLOW takes more time than most of the other signs of this particular verse (2.0 seconds, whereas the average duration of signs in this verse is 0.9 seconds). Furthermore, the active hand, as it brushes over the base hand, continues well up onto the upper arm.

94

Edward S. Klima and Ursula Bellugi

A final example of iconic elaboration in art-sign is the sign SNOW in Fant’s rendition of “Winter” (Table 11). As Mow puts it: “The length of SNOW is longer than usual” representing the movement of the snow “from sky to ground” and the fingers, instead of “wiggling sloppily and irregularly” as in everyday signing, are “gently organized”, imitating the fluttering of the snow.

5. Conclusion In his article “Linguistics and the study of poetic language” Stankiewics ( 1960) characterizes poetic organization as “completely embedded in language and fully determined by its possibilities”. The purpose of the present study has been to examine the form that the poetic function assumes in American Sign Language (ASL), a language which itself has a structural organization fundamentally different from that of oral languages and where, accordingly, the possibilities for poetic organization are also radically different. In this study, we have analyzed several examples of a complex type of composition that we call ‘art-sign’, distinguish4 by three levels of structure. We recognize that in any given instance of art-sign!, mechdnisms associated with the different levels may be interrelated. One level of structure in art-sign is Internal Poetic Structure, constituted from elements that are completely internal to the ASL linguistic system proper - constituted from parts of the grammatical code itself: realizations of the regular formational parameters of ASL, actual signs, regular grammatical processes. Elsewhere in linguistic and psychological studies of the constitution of the sign in ASL, we and others found evidence that signs consist of simultaneously occurring parameters - among them, Hand Configuration, Place of Articulation, Movement, and Orientation. In the present analysis of Internal Poetic Structure in art-sign, we chose to concentrate our attention on the patterned occurrence of signs with similar Hand Configurations. In other sign poems that we have recorded, we found patterned uses of similarities in other parameters. In some instances of art-sign analyzed in this study, one of the mechanisms that results in the similarities that form the basis for Internal Poetic Structure is a restricted type of distortion that involves only parts of the linguistic code: the substitution of one regular ASL formational prime (e.g., one particular Hand Configuration or one particular Orientation) for another - resulting in a form that is no longer the normal form of the sign but which is, nonetheless, a possible sign of ASL.

Poetry and song in a language without sound

95

At the other extreme from Internal Poetic Structure in ASL art-sign is a level of structure which we call Superstructure - analogous, we suggested, to the melodic line superimposed upon the words in song. The Superstructure is superimposed on the form of the signs themselves, and the signs may, as a result, undergo drastic distortions from the point of view of the linguistic code of ASL. In this study, we described Superstructures of two kinds: kinetic and rhythmic. Kinetic Superstructure consists of a special ‘design in space’ superimposed on the signs of the poem. In one instance, the ‘design in space’ was characterized by large, open, non-intersecting movement; in another, it took the shape of an enlarged cross. In another artsign composition examined here, a special temporal-rhythmic pattern was superimposed on the signs. Between these two extremes - Internal Poetic Structure constituted exclusively of elements of the linguistic code and a Superstructure constituted of spatial or rhythmic effects that are not otherwise characteristic of signs or of signing - there is a third level of structure, which we refer to as External Poetic Structure. This intermediate level is created not by choosing signs such that elements of the grammatical code can be used as the basis for poetical effects, but rather by playing, in a structured way, on aspects of the presentation of signs. One mechanism involved in External Poetic Structure is patterned alternations of the hands. In casual signing, signers do occasionally switch hands, but such switching is done to create special effects - e.g., to emphasize the distinction between the predicate that goes with one noun sign and the predicate that goes with another. By contrast, in certain examples of art-sign examined in this study, the alternation of the hands is pervasive and becomes an end in itself. Another mechanism used to create External Poetic Structure is the overlapping of the signs - maintaining the form of the just-executed sign with one hand while making the next sign with the other. This, in effect, presents two signs simultaneously to the eye. The deliberate manipulation of the transitions between signs is still another basis for External Poetic Structure. In ordinary signing, there is a specific movement proper to each sign; between signs the hands relax and change Hand Configuration while moving in transition from the terminal position of the sign just completed to the initial position of the next sign. In art-sign, not only may these transitions be obliterated, extended, or otherwise exaggerated, but such manipulation may even assume a regular pattern through the composition. Characteristic of all these cases of External Poetic Structure is the creation of structured effects - of patterns - through manipulation of what are otherwise ‘incidental’ aspects of the act of signing.

96

Edward

S. Klima

and Ursula Bellugi

In the introduction to the present study, we asked whether there was not, in certain types of heightened signing, a silent-language analogue to that special blend of sound with sound - phonation with vocalization - that, in the auditory channel, constitutes song. It seems to us that in what we have called art-sign we have the beginnings of a comparable development in the visual channel: the hands simultaneously involved in signing and designing in space.

REFERENCES U. and Siple, P. (1974) “Remembering With and Without Words.” in Current Problems in Psycholinguistics, F. Bresson (Ed.), Paris, Centre National de la Rechcrche Scientifiquc, 215-236. Bellugi, U. and Fischer, S. (1972) “A Comparison of Sign Language and Spoken Language,” Cog.. I, 173-200. Bellugi, U., Klima, ES., and Siplc, P. (1975) “Remembering in Signs,” Cog., 3(2), 93-125. Bellugi, U. and Klima, E. S. (1975) “Aspects of Sign Language and its Structure,” in The Role of Speech in Language, J. Kavanagh & J. E. Cutting (Eds.), Cambridge, MIT Press, 171-203. Cummings, E. E. (1923) “Since Feeling is First,” in A Selection of Poems, New York, Harcourt, Brace & World. Fischer, S. (1975) “Influences on Word Order Change in American Sign Language,” in Word Order and Word Order Change, C. Li (Ed.) Austin, U. of Texas Press. Frishherg, N. (1975) “Arbitrariness and Iconicity: Historical Change in American Sign Language,” Lang., 51, 3, 696-719. Jakobson, R. (1960) “Linguistics and Poetics, ” in Style in Language, T. A. Seheok (Ed.), Cambridge, Mass, MIT Press. Jakohson, R. (1970) “On the Verbal Art of William Blake and Other Poet-Painters,” Linguistic Inquiry, 1(l), 3-23. Jakobson, R. (1967) “On the Relations Between Visual and Auditory Signs,” in Selecred Wirings II. The Hague, Mouton, 33X-344. Klima, E. S. (1975) ‘Sound and its Absence in the Linguistic Symbol” in The Role ofSpeech in Language, J. Kavanagh & J. E. Cutting (Eds.) Cambridge, MIT Press, 249-270. Klima, E. S. and Bell@, U. The Signs of Language. Cambridge, Mass, Harvard University Press, to appear. Klima, E. S. and Bellugi, U. (1975) “Wit and Poetry in American Sign Language,” in Sign Language Studies, 8, 203-224. Stankiewics, E. (1960) “Linguistics and the Study of Poetic Language,” in Sty/e in Language, T. A. Seheok (Ed.), Cambridge, Mass, MIT Press. Stokoe, W., Casterline, D., and Croneberg, C. (1965) A Dictionary of American Sign Language, Washington, DC., Gallaudet College Press. Stokoe, W. C., Jr (1972) Semiotics and Human Sign Language, The Hague, Mouton. Woodward, James C., Jr. (1973) Implicational lects on the Deaf Diglossic Continuum. Georgetown University, Washington D.C. (Unpublished Doctoral Dissertation). Bellugi,

Poetry and song in a language without sound

97

Nous exarninons ici les caracteristiques distinctives d’un langage primaire (A.S.L. - Langage Americain par signes). Ce langage est plutot manuel-visuel qu’oral-auditif. Nous faisons I’hypothese que les modahtds de ce langage predisposent, non seulement ce langage lui-meme, mais ses formes artistiques i certaines caracteristiques spdcifiques. Nous presentons une analyse de quelques exernples de ‘signes artistiques’ et proposons le ddveloppement actuel d’une tradition poetique et peut 6tre lyrique i base de gestes analogues.

Cognition, 4 (1976) 999122 0 Elsevier Sequoia S.A., Lausanne

Generative

- Printed

4

in the Netherlands

theories in language and music descriptions*

JOHAN SUNDBERG Royal institute

of Technology, Stockholm BJh3N

LINDBLOM

Stockholm University Abstract The style of a set of Swedish nursery tunes is described in terms of a generative rule system. A generative rule system producing melodically similar versions of an old Swedish folk song is also presented. Examples of melodies generated by these two rule systems are given. Both these rule systems are similar in several respects. Thus, the marking of the hierarchical constituent structure seerns to be one of the important principles in composing simple melodies. The rule systems also show a number of similarities with the Chomsky & Halle (1968) generative phonology of English. For instance, the procedures used for deriving a stress contour from a tree diagram are almost identical. Moreover, in sentences as in melodies this stress, or prominence contour is of decisive importance to the generation of the surface structure, such as meter, harmony, and sequences of pitches. It is believed that such parallels between language and music reflect characteristics of man s perceptual and cognitive capacities.

Introduction Traditionally, music theory works with impressionistic, non-formalizing methods. In describing musical styles, musicologists generally content themselves with mentioning the more remarkable and apparent stylistic features, while other features which are taken as self-evident are considered as not worth mentioning and disregarded. A similar attitude was *The authors are indebted Slrn Mahnstrom, Stockholm National Institute of Health Foundation Grant No. 67/48.

to Professor Ingmar Bengtsson, Uppsala University, and Professor University for valuable discussions. The work was supported by the Research Grant No. NB 04003-07 and a National Bank of Sweden

100

Johan Surzdberg and Bjtirn Lindblom

Figure

I.

Block diagram illustrating procedure of scientific inquiry.

KNOWLEDGE

T SCIENTIST

A

I

+

#O > HYPOTHESES

t DATA

THEORY PREDICTION

,.

-

DATA

OBSERVATION

+-

REALITY

PREDICTION)

adopted in linguistics some decades ago but has now been abandoned. Instead, generative methods are used, largely as a consequence of the work of N. Chomsky. The generative method may be illustrated as in Figure 1. Observations are made and data are collected serving as the basis for the scientist’s formulation of hypotheses. These hypotheses are integrated into a tentative rule system, or theory capable of generating data, or predictions. If the predictions agree with the observations, the rule system used for the predictions is a good description of reality, and knowledge is obtained. If not, the rule system contains errors which should be corrected. In this way, a generative rule system offers a unique possibility of testing hypothetical descriptions scientifically. Generative methods in descriptions of musical styles have recently been advocated, particularly by linguists (cf. e.g., Winograd 1968, Ruwet 1972, Nattiez 1974). The present paper attempts to describe facts in musical theory by means of generative rule systems. Two rule systems will be presented, one for some Swedish nursery tunes, another one for a set of different versions of an old Swedish folk song which are all felt to be melodically similar. The rule systems show striking similarities with the generative phonology of English as proposed by Chomsky & Halle (1968). The significance of these similarities will be discussed.

Generative theories in language and music descriptions

Figure 2.

Tree diagram of’ the constituent his sister”.

structure of’ the sentence

NOUN /\

101

“John beats

PHRASE

\ NOUN PHRASE

NOUN

/ VERB

/\ PRONOUN

NOUN

John

beats

his

sister

Formalization

of sentences

Let us start with recapitulating the major properties of generative grammar by considering the linguists’ description of certain aspects of spoken sentences. Take the utterance John beats his sister In a traditional grammatical description of this sentence the following features might be mentioned: First, the whole string of words makes up a sentence (S). beats his sister is a verb phrase (VP) which consists of the verb (V) beats, and the noun phrase (NP) his sister. The latter unit is made up by the pronoun (Proun) his and the noun (N) sister. John is an NP that consists of a noun. This information on the constituent structure of the sentence can be represented in the form of a tree diagram as shown in Figure 2. The constituent structure of an utterance is reflected in the prosodic aspects of pronunciation, i.e., in intonation and stress pattern. A procedure for relating prosody and constituent structure has been proposed by Chomsky & Halle (1968). It can be diagrammed as in Figure 3. Applied to the sentence: John beats his sister this procedure works as follows. We can represent the distribution of stresses that might be observed in speech under normal, idealized conditions, as Join

be&s his si&ter

where the heaviest, main or primary

stress = 1, secondary

= 2, and tertiary

102

Johan Sundberg and &G-n Lindblom

Block

Figure 3.

diagram of the Chomsky

CONSTITUENT

PROM I NENCE

DURATIONS,

FUNDAMENTAL

prosody.

CONTOUR

FREQUENCY,

BEATS

JOHN

of speech

STRUCTURE

Cyclic derivation of prominence

Figure 4.

& Halle model

HIS

AMPLITUDE,

VOWEL

QUALITY

contour of sentence. SISTER RULE:

OUTPUT:

= 3.

((

1

jN ((

1

jv

((

1

jN ((

’

)” (

((

’

IN (

2

I

(

2

3

1

2

3

I

(( 1 jPRON(

1

I

STRESS )N)NP)VP)SLEXICAL PLACEMENT

)NP)VP)S )VP’S

)s

RIGHT

PRIORITY

RIGHT

PRIORITY

Within the framework of Chomsky & Halle’s theory of generative phonology it has been suggested that a sequence of stresses such as that proposed for John beats his sister be generated in the manner illustrated in Figure 4. Here the constituent structure is represented in terms of 1abeEed brackets which is exactly equivalent to a tree diagram. In short, the computation of the stress contour begins inside the innermost pair of brackets. We start out by applying rules that assign stresses to the individual words. The deepest constituents are his and sister. The latter receives stress on the first syllable. The former is a pronoun and does not receive

Generative theories in language and music descriptions

103

stress. Next we erase the innermost pair of brackets and again look for the deepest constituents. This time we find (John)N(beats)v and (his sister)Np. At this level also John and beats are dealt with according to the rule for word stress. Erase innermost parentheses once more and we get the second line of the figure. The verb phrase, the VP, contains two main stresses. Since it is a phrase, a right-priority rule (Nuclear Stress Rule) is applied, that is, the rightmost stress remains and the other one is lowered by one degree. After this operation brackets are erased and we now have the sequence (1 2 1 )s_ The right priority rule gives 2 3 1 and that is exactly what we intended to get. Two things should be observed here. One is that the prominence contour is derived automatically from the constituent structure by means of a rule system. The second is that the prominence values determine a number of important properties in the acoustic code. Next it will be demonstrated that this seems to hold also in our melodies. Formalization

of melodies

Figure 5 shows an overview of the occurrence of various phenomena in meter, harmony and tone assignment in a series of Swedish 8-bar nursery tunes written by the composer Alice Tegner. These melodies were composed in the late 19th century and they are familiar to and sung by most Swedish children even today. With respect to meter we observe that in even-numbered bars long durations occur and short notes are avoided. The chords are divided into three groups according to their harmonic functions: Introductory chords which present the tonic, target chords, chords preceded by their dominants, and anticipatory chords, dominant chords followed by its own tonic. Introductory chords are found in bars I and V, and target chords may occur at the end of every bar. Anticipatory chords are not found at the end of bars IV and VIII. With melody we distinguish between chord notes and non-chord notes. The suspension is regarded as a special case of a non-chord note. Suspensions seem to be avoided in approximately the same places where short durations are avoided. By and large non-chord notes occur in the same positions as suspensions, but have a greater number of forbidden positions. These positions show a symmetrical and regular pattern. With each of these three kinds of data we see traces of a segmentation of the tune. Thus, bar VIII, particularly, but also bar IV, and bars II and VI, are treated in quite a special way. Periodicity occurs in all the diagrams. It is noteworthy that these regularities are associated with conceptions employed by musicologists for a long time: Beats, subphrase, phrase, and period.

104

Johan Sundberg and BjGrn Lindblom

Figure 5.

Schematical demonstration of the distribution of metric, harmonic, and tonal features within a corpus of Swedish nursery tunes. Shaded areas indicate the position in which the feature listed to the left has been observed to occur. Position is shown with reference to the bars of the period (vertical dotted lines). BAR XCURRENCE d AND

OF

J

,

1

HARMONY INTRODUCTORY CHORD

m

TARFET

-_I-

CHORD

ANTICIPATORY CHORD

NON

CHORD

,,,

----

MELODY NOTE

SUSPENSION

Model Figure 6 shows an attempt to integrate the observations made into a generative model. At the deepest level we find the syntactic structure. Through a prominence contour this structure determines the choice of durations, chords, and pitches. We now describe in more detail the most important properties of the generative model. A syntactic structure of the type shown in Figure 7 can be ascribed to a typical 8-bar melody. The entire melody is a period consisting of two phrases. Each of them consists of two subphrases made up of bars, feet, and beats. The problem is to relate this constituent structure to durations, and to harmonic and tonal data. In order to do this we transform the syntactic information to a string of prominence values. This procedure is exactly the same as in the case of the linguistic example mentioned before. The procedure is illustrated step by step in Figure 8. The topmost line shows the set of beats, each embedded in pairs of parentheses so that the syntactic structure is reflected in the distribution of parentheses. In erasing the parentheses, a left priority is applied up to the bar level and right priority at higher levels. The final result is given in the lowermost line in the figure. It gives the prominences of the beats along a scale of

Generative theories in language and music descriptions

Figure 6.

105

Block diagram of the generative model. I

I

L-l--J SYNTAX

CONSTITUENT

STRUCTURE

PROMINENCE

PROMINENCE

CONTOUR

TIMING RULES

CHORD RULES

TONAL RULES

HARMONIZED MELODY

integers where numbers refer to rank of prominence. This prominence contour provides a method of relating the syntactic structure to the metric, harmonic and tonal properties of the melody. It gives the raw material for the derivation of meter and harmony and for the assignment of pitches in a melody. We should now explain in more detail the nature of the various components of the rule system. Let us do this by starting with the following

:

JJ

14

/B\

J1

of prominence

J1

=6

2

71 =7

=g

PERIOD

/Y =9

((3

(4

PERIOD:

(((2

PHRASE:

((((1

FOOT:

5

4

3

2)

4

3

2

(1

5

4

3

2)

4

5

4

3

2

3

2)

(1

3

2

1

(1

5

4

3)

2))

4

3

(2

((1

/p\

41

5

4

3

2)

4

3

2

(1

=I1

/& =12

CLOSING

r\ =I3

11

J1

1

1

J

4

2)

5

4

3

4

3

2

(1

2)

5

4

3

2

14)

1

(1

5

3))

2)))

4

(3

((2

(((1

2)

5

4

3

4

3

2

(1

2)

5

4

3

4

3

2

(1

1

5

4

3

2)

4

3

2

1

(1

4

5

4

3)

2))

d

4

3

(2

((1

1

5

4

3

2)

J

4

3

2

(1

1

5

4

3

2)

4

4

3

2

(1

J

5

4

3

2)

J

15)

1

4))

/Y F16

RIGHT

RIGHT

RIGHT

LEFT

PRIORITY:

F15

2))))

d

4

1 3)))

(1

d

F14

lP\

PHRASE

values for the beats in the period shown in Fig. 7.

=I0

3

/\

ofan 8-bar period in 4/4 time.

JJJJJJJJJJJJJJJJJJJJJJJJ

F5

7i

/\ PHRASE

strucCure

(((((l)(l))((~)(l))((l)(l~)((~xl))x((l)(l))((~)(l))((l)(l))((l)(l))))((((lx~)x(l)(l)x(lxl)x(~xl)))(((l)(~)x(lxl)x(lxl)x(l)(~)))))

44

SUBPHRASE:

BEAT:

INPUT:

Cyclic derivation

Figure 8.

/\

JJJJJJJJ

:

OPENING

Tree diagram of the constituent

BEAT:

FOOT:

BAR:

SUB PHRASE

PHRASE

Figure 7.

Generative theories in language and music descriptions

107

108

Johan Sundberg and Bj&-n Lindblonz

question: Have we found that system of rules which automatically and completely enumerates Tegner’s nursery tunes? If we have, the rule system should be able to generate all of Tegner’s &bar melodies. How do we choose among the various alternatives, made available by the rules at every step of the generation, in such a way as to produce a melody that approximates as closely as possible a given Tegner melody? The first step is to generate the meter. An overview of the metric rules is given in the right column of Figure 9. These rules are applied in the order shown in the figure. In the Ieft part of the figure the derivation can be followed step by step. The uppermost line shows the input to the rules constituted by the string of prominence values that has just been developed in Figure 8. First, the sequence-pattern is determined. This shows which bars are to be metrically similar and which dissimilar. Two types of metrical pattern may occur, A and B, and the metrical pattern of the first four bars is obligatorily repeated in the last four bars. The next step offers three alternative choices: Delete prominence rank 5, (a) when it follows prominence rank 1, (b) when it follows prominence ranks 1 and 2, (c) when it follows prominence ranks 1, 2, and 3. We call this deletion procedure catalexis, borrowing a term from the theory of verse. In the next step we may insert additional prominences of rank 6 after prominence ranks 4 and 5 in the first bars marked A and B. Though not done in this case, it is possible at this point to dot notes immediately followed by minima in the prominence contour. After this, the insertions and dottings are transferred to the other bars in accordance with the sequence-pattern. The final step is to transform the modified prominence contour into durations. Every prominence except those of rank 6, and those which have been dotted, is assigned a duration of one beat. Figure 10 presents an overview of the rule system used for the chordal interpretations of prominence ranks. Here, too, the rules are applied in the order shown in the figure. Their function is to determine how often and in which way the chords change. The derivation is illustrated in Figure 11. Once again the topmost line shows the input material, i.e., the string of prominence values derived in Figure 8. An obligatory rule deletes all items of prominence rank 5. After this, some items of prominence rank 4 may be deleted. Next, introductory chords, I:s, and target chords, M:s, are distributed. It would be possible to return to the introductory chord at the beginning of the closing phrase, but this has not been done in the melody considered. As a fundamental principle of the harmonic function of chords we propose that anticipatory chords, DOM:s, must alternate with rest chords, V:s. Given the function of a chord, its harmonic distance from the tonic is determined by its relationship to the following target

Generative theories in language and music descriptions

Overview of the harmonic rules.

Figure 10. HARMONIC

1:

DELETION

OF 4’S

2:

DELETION

OF

TEMPO 5’S

___-__---------___--_----_-____-_

__-__-_-----3:

CADENCE

109

INTRODUCTORY

4:

TARGET

5:

INTRODUCTORY

6:

EXTENDED

CHORDS

(OBLIGATORY

CHORDS

(OBLIGATORY

RULE) RULE)

STRUCTURE CHORDS

INTRODUCTION

7:

HARMONIC

FUNCTION

8:

HARMONIC

DISTANCE

_____--------------------------REALI7ATION

RULE)

(OPTIONAL

RULE)

-_----------

________-__---__-_-__-----------ABSTRACT HARMONIC REPRESENTATION

(OPTIONAL

FROM --

9:

CHORD

THE

TONIC

-

SYMBOLS

chord. These distances lie along a scale of integer numbers and in the figure they are given as indices. The abstract representation of the harmony obtained is then transformed, step by step, to chord symbols in a completely automatic way: When the harmonic functions and distances from the tonic have been determined for a set of chords, the chords themselves can be chosen automatically. The output of the metric and harmonic rules provides the material required for the generation of pitches. The rules applied to determine melody are those given in Figure 12. First a sequence pattern is assigned, determining which bars are to be melodically similar. After this, pitches are assigned in an order corresponding to the order of the prominence ranks. The various lines in Figure 13 show, step by step, the stages of the derivation. After the sequence-pattern has been determined, the pitches for prominence ranks 1, 2, and 3 are assigned in turn. The main rule allows nothing but chord-notes in these cases. After this, pitches are assigned for prominence ranks of four. Here there are three guiding principles: (1) Harmonic implication (the pitch signals the underlying chord); (2) tonal adjacency or the “short-cut-principle” (when two pitches have been assigned intermediate notes will receive intermediate pitches) and (3) the sequence-pattern. At this point it is possible to insert suspensions, which represent a special case of the principles of harmonic implication and tonal adjacency. In the present example this possibility has not been used.

OUTPUT:

INPUT:

Figure 11.

!

:

;:

Sp

1

j

SP

i

T

I

I

1

V2

1

T

I

1

1

V2

I

I

I

V2

DOM!

)

V

I

1

I

D7

D7

D7

DOMl

DOMl

41

31

II

41

4

I

b

1-I

5 14 52

31

3

5 4

41

14

5 14 5 3 5 b

I

f4

4

14 5 4 2

I

(4

514

T7

IS

SP

SP

DOM3 (S

V2 V2

1

V2

V

31

31

3

1

1 D7

D7 1

1

I

D7

I

14

514

I

DOMl

DOM

4

4

4

5 4

I D-f1

/

1

14

5 3 514

DOM3 1

“OM3

(4

514

DOM3

4

5 4 5

T

T

T

T

M.

M

DISTANCE

FUNCTION

11 CHORD

11 CHORD

11 CHORD

SYMBOLS

SYMBOLS

11 CHORDSYMBOLS

11 HARMONIC

II HARMONIC

AND

4’S

5’S

TARGET

FROM

INTRODUCTION

11 INTRODUCTORY EXTENDED

11 DELETION

OF

OF

1

11 DELETION

M

1

1511

THE

TONIC

CHORDS

Generative theories in language and music descriptions

Figure 12.

Overview of the tonal rules. CONTENTS

OF

4

HARMONIC

IMPLICATION

4

HARMONIC

IMPLICATION

ORDER 1.

PATTERN

2.

PROMINENCES

3.

PROMINENCES

OF

RULES

SEQUENCING > =

TONAL

ADJACENCY (SHORT-CUT

4.

111

PROMINENCES

<

4

PRINCIPLE)

SEQUENCING

The next step is to choose chord-notes for the introductory tones in every phrase, observing the principle of tonal adjacency. Using the same principles while taking the sequence-pattern into account, pitches are determined for the remaining notes of prominence rank 4, as illustrated in the next line of the figure. For example, the sequence-pattern dictates that the tonal relations between the tones in bar III be the same as those in bar I. The last step produces the realization of lesser prominence ranks numbers greater than 4. For tones which have been assigned the role of chord notes, the rules already mentioned are applied. For non-chord notes the principle of tonal adjacency is applied. These notes are related by a scale-step to the following note. The derivation is now complete and the result is given in the lowermost line in the figure. The melody obtained differs in only a few details from “Dansa, min docka”. These differences can all be accounted for by our rigorous use of the sequence-pattern. Evidently TegnCr’s realization of the sequencepattern is less rigid. We conclude that our rules can indeed generate a close approximation to one of TegnCr’s melodies. As a further test of the rule system we can generate melodies by making a completely randomized choice among the alternatives available at the various steps in our rule system. A set of melodies derived in this manner is shown in Figure 14. The general reactions of Swedes listening to these melodies informally are that they are similar in style to those by Tegner.

Versions of a folk-song Moberg (1950) has collected a large number of folk-songs which he claims have a common origin. His assumption is based largely on melodical

OUT PUT:

INWT.

I{

HH

H

_

H

J-IJ J I,““~&’

14 564

5 146563

H

564

HH

I’,,nJ

5 h

5 k

H

HH

H

J 11 JYJJ J I(,“’

5 14 Si2

HH

564

t!

l-l

I l$Jl&

5 146563

H

HH

J I’,J-lJ

5 14 564

561

H

HH

J 14 LJJ

5 i4

11

5 1

umLImV

PASSIK;

?mJlJENClNG

’ ADJECENCV

TONAL

IMPL

1IIARM

SUSPENSION

I

P,L

,

: P=4

P
Derivation of the final melody. The derivation involves eight stages: Stage 1: determination of sequence pattern. Stage 2: assignment of chord notes to notes with higher rank of prominence than 4. Stage 3: introduction of suspensions (not used). Stage 4: harmonic implication (HI), tonal adjacency (TA), sequencing (S) to all sub-phrase initial notes with prominence rank of 4. Stage 5: HI, TA, and S applied to all remaining notes with prominence rank of 4. Stage 6: chord notes (H) or non-chord notes (N) selected for prominence ranks equal to 5 (only H:s chosen). Stage 7: HI, TA, and S applied to the noted interpreted as H at stage 6. Stage 8: passing rule and auxiliary rule applied to the remaining notes which now all have prominence ranks lower than 5. In the cases where two notes are shown the ones with upward lines were derived by rule, the others chosen by Tegner.

Figure 13.

Generative theories in language and music descriptions

Figure 14.

Examples of’ generated melodies.

Nr 2

07

Nr

T

D7

T

T

3

S

Nr L

Nr 5

D7

SP

D7

T

113

114

Johan Sundberg and Bjtirn Lindblom

Figure 15.

Eight versions of the Swedish fold-song ‘Ro, ro till jiskesktir” selected from Moberg’s (I 950) article on that song.

similarity and his material may in consequence be useful in an investigation of what actually makes two melodies sound melodically similar. The purpose of the study of Moberg’s collection differs therefore from that in the ,case of the nursery tunes. What concerns us here is the fact that the melodies are felt to be similar. Here, then, we will not try to formulate a rule system describing the musical style represented by the folk-song. Instead our rule system should generate only those melodies that are felt to be melodically similar to the versions given. Thus, it is not sufficient that they all belong to the same musical style. Data

Figure 15 shows the versions of the folk-song that will concern us here. They were selected from Moberg’s material on the criterion of apparent

Generative theories in language and music descriptions

115

melodical similarity. Versions with extra bars inserted, with major changes in the meter and so on were excluded. All the versions in Figure 15 have the function of a lullaby. An apparent common denominator of the songs is the preference for half-notes at the end of the even-numbered bars. Also, we note that all melodies end on the fundamental of the tonic, and that all even-numbered bars end on this note or the fifth of the dominant. A closer analysis of the songs reveals some additional common features. Figure 16 gives an overview of them. Specific pitches, pitches belonging to a specific chord, intervals between odd-numbered beats, and the direction of such intervals are invariable in certain positions within the period. Also, we see that the positions for these phenomena form a regular pattern in the period. Again, then, the position within the period seems to be a factor of significance. In view of our experience from the nursery tunes it seems reasonable to interpret this segmentation as an indication of a constituent structure.

Rule system In order to account for these and other observations we again need a generative rule system. As there seem to be good reasons for assuming that a constituent structure exists which determines permissible modifications, a mapping of this structure and the generation of its prominence contour are required. We propose that the melodies be regarded as an &bar structure in which bars 5 and 6 are repeated once. This agrees with the observation that, in fact, bars 7 and 8 generally repeat bars 5 and 6 metrically and melodically. The transformation needed is used so frequently in music that it has a sign of its own, II: :II. The asymmetrical tree we obtain is shown in Figure 17. The next step is to transform the tree diagram into a string of prominence values. For this, we use exactly the same procedure as in the case of the nursery tunes: Labeled bracketing, erasing innermost pairs of brackets, and left and right priority up to and above the level of the bar, respectively. The output is shown in the same figure. The rule system consists of two groups of rules, as seen in Figure 18. One group for the tonality and the tone assignment and one for the meter. Contrary to the nursery tunes, the harmony is rudimentary in this material. This is rather natural, since the song is normally performed without any accompaniment. However, certain rules are followed. The introduction of the opening and closing phrases and the beginning of

NUMBER:

following

. IMPLICATION OF TONALITY

SIGN OF INTERVAL

re

INTERVAL

PITCH

BAR

Figure 16.

I

i

I

I

1

I

I

1 1 I

2

I

I

I

I

I

i

I

I I I

I

I

, I

I

I

I I I I

I

I

5

6 I

8 9

I

1 1

I

i

1 I

I

I 1

I

1

I

I

I I I I

I

I

J J ;J J J J;J J J ;J J J JIJ J J ;J J J J I I I I I

L

I I

3

Occurrence of various phenomena in the versions of the folk-song shown in Fig. 15. 10

I

VARIES

CONSTANT

C ONSTANl VARIES

CONSTANT VARIES

CONSTANT VARIES

Generative theories in language and music descriptions

Figure 17.

117

Tree diagram of the folk-song. The bottom line shows the prominence values of the beats. PERIOD

PHRASE

OPENING

CLOSING

SUBPHRASE

FEET

BEAT

PROMINENCE

JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ ILSL5/(535~5~514525451514535)545/45)5454514515 I I I

I

I

the last subphrase give chord notes in the tonic. The end of each subphrase contains the fundamental of the tonic or the fifth of the dominant. The main function of these few rules seems to be to establish the tonality of the song. The prominence values play an important role in the assignment of other pitches as well. The notes are treated in an order determined by the prominence rank. With the prominence rank of four the distance to the following note of higher prominence is decisive to the order of tone assignment. The prominence rank of five is determined in relation to the following note of higher prominence. It may be observed that the rules for this prominence rank are ordered so that smaller intervals will occur more often than larger intervals. When the tone assignment is completed, the meter may be modified. No modifications are tolerated in the bar containing prominence 1, and no insertion of prominence 6 is allowed in the bar containing prominence 2. Prominences 3 and 4 may be divided into two notes if it is a half-note. If so, the extra note approaches the following by a scale tone step in accordance with the adjacency principle or it simply repeats the preceding note. Inversely, a sequence of two identical pitches, the first of which has the prominence of 4 may melt together to a half note. Remaining notes of prominences 4 and 5 may be divided into two eighth notes, provided that this gives a sequence of scale tone steps with or without one repetition.

118

Johan Sundberg and Bj6rn Lindblom

Figure 18.

TONALITY

The rule system used for generating the versions of the folk-song in Fig. 19. The notes constituting a chord are given by their step number (e.g. V denoting the fifth). T and D stand for the tonic and dominant chords, respective&. I refers to the interval re. the nearest following note of equal or higher prominence, and ID refers to the direction of such intervals. S = step along the scale, R = repetition. AND

MELODY

p=l

+

p=2

+

p=3

-+

V of D

p=4

A:

I__ 5 q.

B:

c:

1 ofT

l_545q,

q=

1

identical

q=

2

I identical

q=

1

identical

u=

2

q=

3

q=

fq5_

’

2

ID identical (KI~;tical)

chord note of T

i i identical q=

3

ID identical I identical

p =

5

Interval

re. following

is

S,

but delayed

adjacency

principle?

adjacency

principle?

adjacency

principle?

adjacency

principle?

by 1

S? O? 3, 3? 4, 4? 5, 5. METER second note =

p=(~]$ykq q Z

3:

S re. following, the first one

p + p6 if this gives a sequence of S k one R

adj. print.

Generative theories in language and music descriptions

Figure 19.

Examples of rule generated

119

versions of the folk-song.

Another allowed alternative is that the extra note fills in a fifth interval. These restrictions as regards insertion of eighth notes are the main reason why the meter is modified after the melody in the case of the folk-song. This seems logical in view of the fact that the melody is used for different texts. As was pointed out before, our rule system does not represent more than a mere hypothetical description of the melodical similarity between the folk-songs concerned. The test of the hypothesis is attained by generating versions and judging whether or not they are felt to be similar to the melodies studied. As in the case of the nursery tunes the generation only involves random choice between the alternatives allowed by each rule. The generated melodies are shown in Figure 19. Regardless of whether or not we agree on the extent to which these melodies are melodically similar, we may conclude that the rule system defines the melodical similarity observed. It is interesting that this similarity is so heavily dependent upon the notion of prominence. As in the case of the nursery tunes the principle seems to be: The higher the prominence, the fewer alternatives are made available by the rules. Conclusions Above it has been demonstrated that the formalization of melodies and the formalization of sentences are in part strikingly similar. These similarities can be summarized as follows:

120

Johan Sundberg and Bj6rn Lindblom

(1) Melodies as well as sentences exhibit hierarchical constituent structure. (2) This constituent structure is projected onto a level of prominence by a set of transformational rules that are identical for sentences and melodies. It seems reasonable to hypothesize that in the case of both music and speech the function of these rules is to facilitate the generation and reception of the physical signal. In the present paper we shall not pursue our search for analogies between music and language any further. This is not to imply that such analogies are present only with respect to structure and logical organization. No doubt, parallels could be found all the way down to the physical signal. The diagram in Figure 20 sums up our presentation so far. As you may observe it recapitulates features of Figure 1. The research paradigm is here applied to LANGUAGE on the left and MUSIC on the right. The data on music and the data on language are described in terms of two separate generative frameworks. In the construction of these theories certain principles are postulated. This parallel exploration of language and music can be expected to yield principles of varying generality. Some may be highly language specific and others may be valid only for certain styles of music. On the other hand, some features are shared by all languages. Similarly there may be universal features that are characteristic of all styles of music. In the entire class of universals there may be some which are common to both language and music. Examples of principles emerging from the present study are: - Catalexis - Harmonic implication - Tonal adjacency - Sequencing and other processes involving information reduction - Marking of constituent structure. These principles are tentative in nature. Further research is clearly needed to substantiate them or replace them by better alternatives. Nevertheless, we would claim that the hypothetic-deductive method advocated here should be substituted for traditional, impressionistic and non-formalizing approaches. In any case, it seems justifiable to suggest that the presence of hierarchical constituent structure is a phenomenon of considerable generality. It might be asserted that hierarchical constituent structure is a product of social and cultural conventions and largely independent of the human organism. There is, however, a large body of opinion and evidence that this interpretation is incorrect. For instance, hierarchical

Figure 20.

4 I

1

._

YPOTHESES

IDIOSYNCRATIC

PRINCIPLES

AND

_

.-._

SCiENTIST

THEORY

I

THEORY

UNIVERSAL

OF

THEORY

BEHAVIOR

UNIFIED

OBSERVATION

GENERATIVE

ION

I

EXPLANATIONS

MUSIC

GENERATIVE

PREDICT

OBSERVATION

LANGUAGE

Block diagram of the parallel exploration of structures in language and music, which is advocated in the text.

(

z a

Ei

122 Joharz Sundberg and B@rn Lirzdblorn

constituent structure appears to be a linguistic universal. The literature on music theory supports the contention that it is also widespread in various styles of music. Recent psychological research by Restle (1970) on serial pattern learning shows convincingly that some structures or events are learned more easily than others by humans. A case in point is also Lashley’s ( 19.5 1) classical discussion of the integrative functions of the cerebral cortex where he puts forward the speculation that hierarchical syntactic arrangements are not limited to language. Ile mentions “the coordination of leg movements in insects, the song of birds, the control of trotting and pacing in igaited horse, the rat running the maze, the architect designing a house, and the carpenter sawing a board”. Furthermore. according to Restle, “it seems overwhelmingly obvious that long and complex serial patterns are divided into natural subparts, and that mastery is facilitated if the incoming sequence of events is somehow marked off into natural subparts”. The present work may evidently serve as further support for this view.

REFERENCES Chomsky, N. and Halle, M. (1968) Sourrd Pattern of English, NW York, Harper & Row. Lashley, K. S. (1951) The Problem of Serial Order in Behavior. In L. A. Jeffres (Ed.) Cerebral Mechanisms in Behavior, New York, Wiley and Sons, Moberg, C. A. (1950) Tva kapitel om svensk folkmusik,Svensk Tidsk~ifi far Musikforshing (Swcd. J. Musical.), 32. pp. 5--49. Nattiez, J.J. (1974) SCmiologie Musicale, L’itat dc la Question, Acfa Musical. 46, 2, pp. 153-~171. Rcstle, 1:. (1970) Theory of Serial Pattern Learning: Structural Trees, Psychol. Rev., 77, 6, pp. 481 -495. Ruwet, N. (1972) Language, Musique, Po&ie. Pari,, Winograd, T. (1968) Linguistics and Computer Analysis of Tonal Harmony, J. MusIL. Th., 12, I, pp. 3-49. Tegnbr, Alice, Sjunr: med oss Mamma, Vol. I VII, Stockholm, A. B. Seelig & Co.

On d&rit un cnsemblc de comptines suddoises en termes dc syst&me gd&ratif de r2gles. On prCsentc un autrc systhmc g&&atif produisant des versions mClodiques analogues d’un chant folklorique s&dois. On donne divers exemples de mklodies g&&&es par les deux syst&mes. Ces deux systhmes ont des aspects similaires. Ainsi le marquage des composants hiirarchiques dc la structure semble