UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A blind child's meaning for look : a replication of Landau & Gleitman Murphy, Cynthia Maureen 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1987_A6_7 M85.pdf [ 5.11MB ]
Metadata
JSON: 831-1.0097013.json
JSON-LD: 831-1.0097013-ld.json
RDF/XML (Pretty): 831-1.0097013-rdf.xml
RDF/JSON: 831-1.0097013-rdf.json
Turtle: 831-1.0097013-turtle.txt
N-Triples: 831-1.0097013-rdf-ntriples.txt
Original Record: 831-1.0097013-source.json
Full Text
831-1.0097013-fulltext.txt
Citation
831-1.0097013.ris

Full Text

A BLIND CHILD'S MEANING FOR LOOK: A REPLICATION OF LANDAU & GLEITMAN by Cynthia Maureen Murphy B A.., Simon Fraser University, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Faculty of Medicine School of Audiology and Speech Sciences We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1987 © Cynthia Maureen Murphy, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. department of fljudftfOQty 04AQ( SdfffA & The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date fief- / j T (?87 DE-6G/81) ABSTRACT Landau and Gleitman's experiments investigating a blind child's meaning for look, as it applied to herself, were replicated with a three year old boy who was totally blind, and had no concomitant disorders. Several commands to look were presented within informal play sessions. Responses to the look commands were compared with responses to instructions to touch, listen and taste. Experiments were video recorded for subsequent analysis. It was found that the blind child associated the haptic perceptual modality with the visual verb, in that an instruction to look at an object elicited manual exploration of the object. His meaning for look was distinct from his meanings for the other perceptual verbs. These findings were consistent with Landau and Gleitman's findings. Landau and Gleitman's interpretation, of how a blind child's mastery of visual terms bears on the word/meaning mapping problem, is critically discussed. i i i TABLE OF CONTENTS TITLE PAGE i ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES iv ACKNOWLEDGEMENTS v INTRODUCTION 1 CHAPTER ONE. BACKGROUND OF THE STUDY 3 What is Currently Known about Blind Language Learners 3 The Motivation for the Look Experiments 10 CHAPTER TWO. THE EXPERIMENTS 18 Method Overview 18 Experiment One 24 Experiment Two 38 Experiment Three 52 Experiment Four 61 Experimentation with the Control Subjects 69 Experiment One 69 Experiment Two 72 Experiment Five 76 CHAPTER THREE. DISCUSSION 80 Summary of Results of the Look Experiments Replicated with a Blind Child 80 Summary of Results of the Look Experiment Replicated with a Blindfolded Sighted Child 82 The Generalizability of Findings of the Original Look Experiments 83 Further Evidence to Delineate the Role of Experience within the Word/Meaning Mapping Task . 85 An Alternative Explanation for the Acquisition of Verbs 90 CHAPTER FOUR. CONCLUSION 96 BIBLIOGRAPHY 97 I V LIST OF TABLES TABLE I. Differentiation of Bryan's Responses to Look and Touch 49 According to the Exploratory/Nonexploratory Parameter TABLE II. Bryan's Response-Types for Each Experimental Condition 67 V ACKNOWLEDGEMENTS Grateful acknowledgement is extended to: Dr. Carolyn Johnson for her close involvement in the collection of data, and expert advice in composing the thesis; Elizabeth MacLeod for her time and helpful suggestions; my friends and family for their moral support throughout. 1 INTRODUCTION Through their research, Barbara Landau and Lila Gleitman have addressed a number of theoretical questions pertaining to first language acquisition. One issue which has received a considerable amount of their attention involves the mapping of meaning onto heard utterances. Landau and Gleitman have maintained that extralinguistic experience (i.e. an interpretive context) plays a requisite role in a child's language learning. However, while calling experience mandatory, they have also argued its insufficiency. According to these researchers, the interpretive context is supplemented by a specific innate predisposition which enables the child to use language itself in the initial determination of the meaning of utterances. Landau and Gleitman's theoretical orientation challenges the extreme view that language learning is explained by repeated associations of speech signals and extralinguistic experience. With the goal of constraining the role of experience within language learning, Landau and Gleitman investigated the linguistic development of blind children. These children were of particular interest to them because their access to an interpretive context for heard utterances was considered to be reduced. Landau and Gleitman reasoned that if language acquisition could proceed normally, given diminished extralinguistic experience, then a biological program for language learning would be indicated. Landau and Gleitman collected naturalistic data from three congenitally blind children for periods of two to four years. In addition, they conducted a series of experiments with one of their subjects, a little girl named Kelli. The experiments examined this blind child's meaning for words such as look, see and color terms which, in their literal sense, map exactly onto visual experience. Landau and Gleitman showed that in the absence of a conventional interpretive context, Kelli 2 acquired normal sophistication in the use and understanding of these terms. They asserted that this finding could best be accounted for by an innate capacity to use language itself as a cue to conveyed meaning. In their book, Language and Experience, Landau and Gleitman (1985) described their research with blind language learners, and detailed their explanation for these children and all children's success with linguistic development. This thesis reports a replication of Landau and Gleitman's experiments involving the term look. The primary objective of the thesis is to test the generalizability of the findings from their single subject investigation. The secondary objective is to evaluate Landau and Gleitman's interpretation of the relevance of their findings to language acquisition theory. Chapter one briefly reviews literature relevant to blind children's language development, in order to establish a context in which to consider Landau and Gleitman's research and my replication of their look experiments. It will also present the theoretical perspective which prompted Landau and Gleitman to conduct the studies they did, and which influenced the conclusions they drew. Presented in subsequent chapters will be the method used in the original and replicated look experiments, the results from each set of experiments, and a discussion of how the findings bear on language development theory. 3 CHAPTER ONE BACKGROUND OF THE STUDY WHAT IS CURRENTLY KNOWN ABOUT BLIND LANGUAGE LEARNERS To date, the course of language development in blind children has received relatively little research attention. However, due to a growing academic interest in this area, significant expansion in the current knowledge base relevant to the topic is anticipated. Early investigations of blind children's language tended to be subsumed within general developmental studies. Consequently, language reports were limited to accounts of linguistic milestone attainment. While superficial in depth, these investigations often yielded conflicting findings. Because published reports did not always provide detailed descriptions of the studied populations, some of the variance in findings might be attributed to such factors as heterogeneity among subjects, both within and across studies. Sources of heterogeneity would include (a) degree of blindness, (b) length of gestation term, and (c) concomitant disorders. Andersen, Kekelis and Dunlea (1984) studied the effect of variable visual impairment on linguistic development, and found that children who were had minimal residual vision were more similar to sighted children than to totally blind children in their exploration of the environment, and their early use of words. This study demonstrates the problems that would evolve out of the combining of 'legally blind' and 'totally blind' children within a single group of subjects, simply described as 'blind'. It also suggests the serious limitations on the formation of generalizations when researchers failed to define their populations. Others have indicated that variation in data collection procedures, and comparisons with different sighted norms might also account for incongruous findings. The outcomes of early investigations have been difficult to interpret and compare, 4 because of these potential sources of discrepancy. (See Mulford 1984, and Andersen et al. 1984 for discussions of the problems involved in interpreting early studies.) More recently, blind language learners have been studied intensively within longitudinal investigations. Data collection procedures have involved regular audio and video recordings of the children in various contexts. These rich data bases have sometimes been supplemented by lexical diaries kept by parents, and by experimental information. When reporting their studies, researchers have described their subjects extensively, enabling comparisons of findings and the formation of generalizations. Before outlining what has been found in relation to the word/meaning mapping problem, the global picture of blind children's linguistic development should be considered. A few acquisition deviancies have been reported in the literature with enough frequency to be viewed as characteristic of this population in general. 1) Blind children are commonly mildly delayed in language onset time, which is generally defined as the point when the child acquires its first two words (Landau & Gleitman 1985, Keeler 1958, Burlingham 1961, 1964). 2) They often demonstrate a marked propensity to imitate utterances directed at themselves and even others, sometimes to the point of sounding echolalic (Wood 1970, Urwin 1984). 3) Their acquisition of the I/you pronoun distinction is often delayed (Adelson & Fraiberg, 1974). Less pervasively reported linguistic irregularities, which will be outlined in the balance of this section, may also be widely characteristic of blind children. But because these irregularities have tended to be observed in single studies involving few subjects, it may be premature to use them to typify blind language learners in general. Nevertheless, in spite of obstacles which may or may not be encountered along the way, recent and early studies alike attest to the fact 5 that blind children having no concomitant disorders will attain full linguistic competence within the normal time frame. A review of the literature relevant to the role of vision in the assigning of meaning to words should begin with what is known about the development of the visual perceptual system. Studies have indicated that during the first several weeks of the neonatal period, infants can only focus well on objects about eight inches away. Because eight inches is the approximate distance between a breast feeding infant and the eyes of its mother (Robson 1967), Stern (1977:6) suggested that the natural design of anatomy and perception is adaptive towards the infant's "formation of his early human relatedness." Approximately six weeks after birth, the infant becomes capable of eye widening and brightening (Wolff 1963). By exercising this ability, the infant can more earnestly secure its mother's gaze, thereby inviting and reinforcing a myriad of social interchanges. The infant's visual motor system is essentially mature by the end of the third month, meaning that visual tracking from a distance can occur. Implicated in this maturation is the infant's volitional control over what will be maintained within view. Prior to, and during the onset of language, gaze control has been shown to be important for (a) establishing a shared focus of attention between child and caregiver (Stern 1974) , (b) enabling a child to follow a change in verbal topic (Collis & Schaffer 1975) , (c) allowing a caregiver to monitor the child's understanding of what is being talked about, and (d) helping the caregiver determine the referents of the child's first utterances. Studies of the speech input to young children have readily described its 'here and now' quality (e.g. Phillips 1973, Snow & Ferguson 1977, Shatz & Gelman 1973) wherein the tendency is to talk about "whatever is directly under the child's eyes", (Clark & Clark 1977:322). This speech attribute is widely believed to facilitate the child's task of determining the meaning of what is being heard. 6 Reaching to seen, and reaching to heard objects are each thought to be perceptual-motor responses which allow a child to explore its environment. Reaching towards an object within view first occurs for sighted children at approximately six months. On the other hand, reaching to a sounding object emerges later, during the last quarter of the first year for both blind and sighted children (Fraiberg, Seigel & Gibson 1966, Freedman, Fox-Kolenda, Margileth & Millar 1969). In effect, the blind child's opportunity to actively explore its surroundings is comparatively delayed. Landau and Gleitman (1985) have contended that this delay surpasses the time when reaching to sound first emerges, since haptic skills require additional time to develop to the point where they could reasonably be considered as a substitute for vision. Implicit in language acquisition theories which emphasize the importance of gaze control and visual access to an interpretive context, are predictions that the blind child approaches the word/meaning mapping task with a clear disadvantage. In response to such predictions, Urwin (1984:27) cautioned that developmental theories based on research with typical children "may be wrong or inadequate, even on their own terms." Therefore, pessimistic forecasts for blind language learners may be undue. Direct investigations of blind infants have shown that, prelinguistically, these children don't develop gestures such as pointing or reaching - gestures which for sighted children serve as early means of expressing desires, and drawing attention to faraway objects. Rowland (1983) reported that three blind children she studied were most successful prelinguistically at expressing affective states such as pleasure and dislike. Moreover, while all three children could attract attention to themselves, none had developed an ability to direct attention towards distant objects. The absence of prelinguistic communicative gestures further inhibits a 7 blind child's early exploration of its surroundings, and as a result may be partially accountable for a delay in language onset. Recognizing the difficulty in interpreting a blind infant's focus of attention, or contributions within early communicative interactions, Urwin (1978) explored the development of social relations between two blind children and their parents. She observed that touching and blowing were substituted for more conventional means of attaining these children's attention. Routines which are part of all parent/infant interactions appeared to be exceptionally important to the parents of the blind infants for sustaining social contact, and for building expectancies. The routines were gradually used in new situations, and underwent modifications to represent the children's growing range of interests. By the second year, the children were reported to be exploiting their own variations of the routines, and were lifting portions of them out of context to accompany solitary play. Modifiable routines serve as semantically familiar frameworks in which to consider new lexical items, thereby helping to organize the child's assigning of meaning to words. Blockberger and Johnson (1985) compared the use of routines with identical twins when one twin was blind. (These are the same children who participated in my replication of the look experiments.) They found that, initially, routines were used with both children with nearly equal frequency. As the sighted twin approached age two and was able to introduce his own topics, the use of routines declined. During the same period, the use of routines continued with the blind twin. Even at age three, he occasionally used routines to initiate conversational interactions. By virtue of the fact that their's was a twin study, Blockberger and Johnson's evidence strongly suggests that the high proportion of routines used with a blind child is related to the child's visual impairment as opposed to parenting 8 style. The outcome of this study serves as impetus for more seriously considering how the frequent use of routines might benefit the word/meaning mapping task. Several recent studies have addressed various aspects of blind children's emergence into the linguistic stage. (See Mulford 1984 for a review.) For example, Landau (1983) examined how two blind children's first fifty lexemes were distributed across six semantic categories. Those categories were: specific nouns, general nouns, action words, modifiers, personal social words, and function words. Landau compared the outcome of her analysis with the results of a similar analysis which Nelson (1973) performed with sighted children, and concluded that, when matched according to linguistic level, blind and sighted children talk about the same things in the same way. Andersen et al. (1984) studied the first one hundred lexemes of six children having varying degrees of visual ability, ranging from full vision to total blindness. In contrast to Landau, these researchers reported that blind and sighted children's early vocabularies share surface similarities only. By examining how all of the children used their vocabularies, they revealed striking differences contingent on visual ability. The blind children rarely extended the use of common nouns beyond the original contexts in which they were learned, and virtually never overextended the use of their early acquired words. Andersen et al. interpreted these findings to mean, the process that enables young sighted children to abstract criterial features of a referent and to extend the domain of application of early words is not functioning at the same level for blind children at the onset of language, (p.12) It was also reported that the blind children tended to use action words to refer solely to their own actions, at a time when sighted children used these words to refer to themselves and others. Regarding this finding, Mulford (1984:14) remarked that sighted children are typically prompted to comment on the actions of others, upon "seeing the agent in the act." She concluded that because a blind 9 child has limited access to the performances of others, it is not surprising that they are slow to extend the use of their early action words. Dunlea (1982) examined the illocutionary force of the early utterances of the same children observed in the Andersen et al. study. Her analysis spanned the time when the children were using single word utterances, up until two and three word utterances were being produced. Dunlea found a strong correlation between the relative proportion of requests and visual impairment, and a negative correlation between the relative proportion of assertions and visual impairment. Landau and Gleitman (1985) reported the same relative proportions of requests and assertions in the early utterances of two blind children. Dunlea also found that language was used for attention-getting purposes by the blind children only, and that just one of the sighted children used language to offer/show, and draw attention. This finding demonstrates the continuity of illocutionary force which characterizes a child's communication throughout the prelinguistic and early linguistic stages of development. Investigations of the nature of maternal linguistic input to blind children are also relevant to the word/meaning mapping problem. While interested in identifying input styles that were either beneficial or detrimental to successful language development, Kekelis and Andersen (1984) studied the maternal input to the six children observed in the Andersen et al. study. They found that the mothers of blind children provided more imperative sentence types (e.g. requests for action), and fewer descriptions of objects, persons, and events in the environment than did mothers of sighted children. Blind children's mothers also initiated a greater proportion of discourse topics. Kekelis and Andersen reported that frequent maternal requests for action were mirrored in some of the children's early language. They cautioned that blind children's reliance on directives can 10 harmfully limit their practice ,with other language functions, and can inhibit them from creatively solving their own problems when others readily respond to their requests. They also suggested that the paucity of detailed descriptions provided for blind children can forestall their discovery of the properties, functions, and relationships among objects and persons in their surroundings. These forestalled discoveries have obvious consequences affecting the selection of meaning for words. In summary, the reviewed literature clearly suggests that blindness provokes complications within the word/meaning mapping task. In general the complications stem from delayed and inhibited opportunities to explore the world. However, the literature also highlights language learning compensatory strategies which can be employed dually by blind children and their caregivers, to overcome the hurdles created by the absence of vision. Landau and Gleitman's explanation for blind children's ultimate linguistic competence would de-emphasize the role of compensatory strategies involving modifications of extralinguistic experience. Alternatively, their explanation stresses an innate language learning ability that prevails in spite of atypical environmental input. THE MOTIVATION FOR THE LOOK EXPERIMENTS With the objective of uncovering supporting evidence for a biologically based theory of language acquisition, Landau and Gleitman's research with blind children has been focused on the question, "How does the learner ... discover which of the many words she hears encodes which of the concepts she can understand?" (p.100). The scope of their interest has not encompassed an explanation for a child's initial concept attainment. Rather, it has been limited to the issue of how children assign meaning to words. 11 In accord with most language developmental theorists, Landau and Gleitman stated that experience plays a necessary role in language acquisition. We accept as self-evident that any explanation of this learning must take nonlinguistic experience as relevant: When children hear words spoken by adults, they also observe objects, scenes, and events, (p.vii) and, Some interpretive context, paired with speech events, is required if language learning is to get off the ground floor, (p.l) While different theorists have agreed on the necessity of experience, they have proceeded to debate its adequacy as the route to language learning. On one hand, many academics have assigned an emphatic role to experience, claiming that the interpretive context sufficiently enables the deduction of the correct meaning of an utterance. Extreme proponents of experience argue that all that the child must do is apply general cognitive skills in examining the elements of the interpretive context which are consistently present during repetitions of particular sequences of speech sounds. Stated otherwise, the meanings of utterances are to be inferred solely from the extralinguistic contexts in which they are heard, by means of nonspecific learning processes. Landau and Gleitman have contended that such an explanation for language learning is improbable. To underscore the implausibility, they explained, scenes relevant to the utterance "The cat is on the mat" are just as relevant to "The mat is under the cat" or "The cat and mat are on the floor." How is the learner to realize that the linguistic encoding (the particular sentence spoken) is of one of these descriptions of the scene and not the other? The real-world context available to the learner is apposite to all of them. ... The problem in short, is that there is always as much positive evidence in the external world for one of these interpretations as for the next. If no child chooses the false solutions, the question is how they all know enough to avoid them, (p.4) Landau and Gleitman argued that a language learning explanation, consisting of a distributional analysis of speech-signal/experience coordinates, assumes an exorbitant amount of memorial and computational ability. While acknowledging that a child might have the sophisticated capacities to be performing ongoing 12 distributional analyses, Landau and Gleitman contended that this experience-based theory of acquisition makes predictions about the course of learning which contradict empirical findings. For example, because all children's linguistic environments contain situation/utterance pairings which, when considered against each other, will yield more than just the correct word/meaning mapping, incorrect inductions should be commonly observable. Yet, studies have shown that wrong mappings rarely occur. Landau and Gleitman viewed unlikely predictions which follow from the distributional analysis theory as impetus for advocating an innate component to language learning. Alone, and in joint effort with a number of co-authors, Li la Gleitman has widely established her position in support of innatist acquisition theory, which de-emphasizes the role of experience. In a discussion of arguments commonly used to promote a biological predisposition for language, Gleitman (in press) stated that both empirical findings and logic are relevant to the defense. Derived from empiricism have been such arguments as uniformity in the course of acquisition, across cultures and in spite of variable linguistic input to the child. Citing Lenneberg (1967) as possibly the first to interpret the significance of this uniformity, Gleitman agreed that the relatively invariable course of language development is a rudimentary basis for considering a biological preprogramming. Empiricism has also demonstrated that there are systematic disparities between a child's linguistic input and output. Gleitman stated that these disparities are relevant to innatism since they, can be understood, but only by claiming the learner filters the input data through an emerging system of rules of grammar, rules to which he is never directly exposed, (p.4) To exemplify the application of innate rules, Gleitman referred to the situation in which a child may be exposed only to questions such as 'What can I eat?', yet early on will produce wordings such as 'What I can eat?'. Gleitman claimed that this 13 type of error is predictable only if it is supposed that "the child acquires formation rules that underlie the declarative first and countenances movement rules only later" (p.5). Asymmetries in input and output are also apparent as the child is acquiring closed class, or functor words. Gleitman and Wanner (1982) conjectured that a natural inclination of the language learner is to direct heightened attention to the stressed syllables of speech input. Because functor words are unstressed, subsyllabic units, they are subject to the filtering effect of "preexisting biases as to how to represent the sound wave" (p.6). As a result, the young language learner passes through the developmental stage characterized by 'telegraphic speech'. Gleitman has also indicated her support of logically based arguments used to promote innatism, such as Chomsky's claim that children acquire knowledge of certain language properties for which the environment provides no insight. As an example, children become competent users of the wanna contraction, when the rules for its usage are neither taught explicitly, nor evidenced in extralinguistic experience. Innatists assert that a child's errorless learning that contraction can occur only if want and to are contiguous in the deep structure of the sentence, is a product of natural biases about the structural properties of language. Gleitman has built up additional defenses for her theoretical orientation towards linguistic development, through direct research with language learners. In a study of the Motherese Hypothesis, Newport, Gleitman and Gleitman (1977) collected maternal speech samples, and assessed the language growth of fifteen young children across a six month period. The aim of their study was to identify aspects of maternal speech input that predicted rates of linguistic growth. These researchers found that, in general, differences in input were not highly correlated with variable rates of learning. This finding was viewed as support for the conclusion that certain language skills emerge at rriaturationally fixed times, and 14 are insensitive to environmental influences. The one aspect of maternal input that was correlated with learning rate involved closed class words. However, as was previously mentioned, Gleitman and Wanner offered an innatist explanation for this correlation, wherein a child, whose maternal input contains excessive use of functor words in stressed positions, will acquire these words more quickly due to a predisposition to attend to stressed units of speech. In another investigation focused on linguistic input to children, Feldman, Goldin-Meadow and Gleitman (1978) studied the development of 'home sign' in six deaf children born to hearing parents. Home sign is an informal gestural system known to emerge in deaf children whose hearing parents do not teach them any form of gestural language. Though grossly deprived of language stimulation, the children in this study were reported to (a) use single gestures at the same developmental time as single words are produced by hearing children, (b) use two and three gesture sequences when utterances of similar length are used by hearing children, and (c) to express the same semantic/relational roles in their early gesture sequences that are expressed in the two and three word utterances of hearing children. Feldman et al. concluded that an innate predisposition guiding language acquisition was accountable for the sophisticated communication skills that surfaced in the linguistic isolates. In the two studies just summarized, support for innatist language development theory was sought through investigations of v a r i a b l e , or a b s e n t l i n g u i s t i c i n p u t . Landau and Gleitman's research with blind children was intended to examine how r e d u c e d a c c e s s to t h e i n t e r p r e t i v e c o n t e x t would impact on language learning. It was reasoned that a comparison of the courses of linguistic development for blind and sighted children could be used to delineate the role of experience within language acquisition. If blind children are little affected by their reduced access to the interpretive context, then innatism would be supported. 15 Landau and Gleitman collected naturalistic language data from three congenitally blind children, who had no known dysfunctions outside of their visual impairment. As mentioned previously, the data was gathered at regular intervals across periods of two to four years. The findings which emerged from the comparison are summarized below. Although some of these findings were listed earlier in the literature review, they will be restated in order to provide a clearer picture of some of the compiled evidence on which Landau and Gleitman based their conclusions about the role of experience. - The onset of speech in blind children (i.e. the acquisition of two lexemes) tends to fall below the median onset time for sighted children, but is still within the normal limits. Like others, Landau and Gleitman have suggested that this tendency towards an onset delay is due to the late emergence of haptic exploratory skills, that enable a blind child to survey the interpretive context which accompanies speech. - The earliest expressed meanings (i.e. the early vocabularies) of blind children are very similar to those of sighted children. "In sum, blind children talk about what most young children talk about: mommies, daddies, dolls, cookies, and toys" (p.30). (Recall that Andersen et al. indicated that an alternative method of analysis revealed contradictory findings.) - Syntactic growth (i.e. the initial combining of two or more morphemes) occurs later in blind children, though still within the normal limits for sighted children. Despite a late onset for combining morphemes, the rate of syntactic growth for blind children parallels that of sighted children. By age 3;0 (three years;zero months), blind and sighted children are speaking in utterances of equal length. - When matched according to MLU, blind children are very similar to their sighted peers in terms of the semantic relations expressed in their early multiple-word utterances. 16 - When matched according to MLU, blind and sighted children's utterances are characterized by similar syntactic complexity. One exception to this statement pertains to the use of the verbal auxiliary. Blind children are slow to acquire this structure. In accord with Gleitman and Wanner's reasoning, Landau and Gleitman concluded that because a blind child's mother asks relatively fewer questions, and gives more commands, the child's maternal input is lacking in stressed verbal auxiliaries. Consequently, blind children acquire this structure more slowly. Beyond language onset time, and the use of verbal auxiliaries, Landau and Gleitman reported parity in the language development of blind and sighted children. The two populations are essentially indistinguishable from each other by the third birthday, including internal organization of syntax, thematic relations, and vocabulary. It seems remarkable that children who confront the real world as differently as blind and sighted children come to have so similar a perspective of it, as revealed by this first descriptive look at their language learning. Radically different sensory-perceptual bases for induction yield a similar pattern of language development, suggesting difficulties for any simple experiential account of such learning, (pp.49-50) In addition to the analyses performed on the naturalistic data, Landau and Gleitman conducted a series of experiments in an attempt to further demonstrate the constrained role of experience in the acquisition process. As mentioned earlier, the experiments were conducted with one of the participants in their longitudinal study, a little girl named Kelli. The experiments examined Kelli's meanings for vision-related terms such as the verbs look and see, and the color adjectives red and green. In pursuing their investigation, Landau and Gleitman reasoned that the vision-related terms "represent the case in which the experiences of the blind child and the sighted child appear to be maximally different" (p.viii). Therefore, if extralinguistic experience provides the route to learning, a blind child should have maximum trouble with these terms, for they seem to refer to the sighted world, (pp.2-3) 17 Landau and Gleitman's experiments with Kelli showed that a blind child can acquire considerable sophistication with the use of the sighted vocabulary. Regarding the outcome of their investigation, Landau and Gleitman stated, The findings from this inquiry ... give little support to the view that first language attainment is explainable as a straightforward derivative of information provided in the environment of the learner. Rather they suggest to us a learning procedure significantly modulated and constrained by the child's natural (innate) biases about the content and form of a natural language, (p.3) While advocating a distinct biological predisposition for linguistic development, Landau and Gleitman argued that a child initially uses language itself to determine the meaning of utterances. In their own words, a point often missed is that part of the child's experiential context is linguistic. Very early in the learning process, the systematicity of the language code becomes available to the learner at least in its rudiments. To the extent that the language system is organized in semantically relevant ways, it serves as an important bootstrap into ever more subtle discoveries about the linguistic means for communication. In short, a critical contextual cue to language learning is language itself, (p.20) Landau and Gleitman claimed that innatist language learning theory best accounts for the successful linguistic development of both blind and sighted children. They also asserted that innatism explains a blind child's acquisition of the sighted vocabulary, in spite of a diminution of experience. The details of how a child, preprogrammed for language learning, is thought to use language itself as an initial cue to the meaning of utterances, will be outlined in the discussion chapter, following the presentation of the look experiments. 18 CHAPTER TWO THE EXPERIMENTS METHOD OVERVIEW Subjects for the Original Look Experiments The experimental subject in Landau and Gleitman's look experiments, Kelli, was a child who lost her vision shortly after birth. She was the only survivor of a pair of twins born approximately three months prematurely. Her birth weight was 940 grams. As a tiny, ill infant, she was immediately placed in an isolete. Of great misfortune, Kelli's visual ability became victim to the oxygen rich environment of her isolete, which caused a condition called retinopathy of prematurity (ROP). ROP refers to vascular changes to the retinas of a premature infant. Approximately 20% of premature babies with birth weights less than 1500 grams suffer minor vascular changes in their retinas which cause minimal visual impairment, while approximately 1% suffer vascular changes that are severe enough to result in blindness. With stage 4 ROP in her left eye, and stage 5 ROP in her right eye, Kelli was diagnosed as totally blind. In later testing, it was thought that she might have some sensitivity to light in one eye, enough that she could tell bright light from darkness, but not enough to allow her to distinguish shadows or shapes. After birth, Kelli remained in the hospital for six months. Because ROP is often associated with concurrent neurological disorders, Kelli's development was monitored throughout the years that she was involved in Landau and Gleitman's longitudinal study. It was reported that no neurological defects were ever detected or suspected. Kelli was delayed in her motor development as compared with sighted norms, though this type of delay is common among blind and premature children. She walked with support at age 1;10, stood independently at age 2;0, and walked without support at age 2;2. Socially, "Kelli seemed happy, 19 active, and cooperative" (p.23). Her general cognitive and perceptual development was reported as normal according to the Bayley Scales of Infant Development, and the Stanford-Binet Tests of Intelligence. In the area of language acquisition, Kelli's rate of development was reported to be in the bottom quartile for intact blind children, according to norms derived from a study of 86 neurologically blind children (Norris, Spaulding & Brodie 1957). She said her first word at age 1;11, and combined two words in a sentence at age 2;5. With respect to the sighted vocabulary in particular, she was observed to use the words look and see before age 2;4, this being the normal time for the appearance of such words in the lexicons of sighted children (Bloom, Lightbrown & Hood, 1975). By age 3;0, when the look experiments commenced, Kelli was using look and see frequently and appropriately in her own spontaneous speech. Kelli's parents were of middle-class, nonprofessional status, and were described as being very accepting of her blindness, as they provided a nurturing environment that would best enable Kelli's normal development. Kelli also had a sighted sister who was one year younger than herself. The control subjects in Landau and Gleitman's look experiments were four sighted children ranging in age from 2;9 to 3;6. These children were blindfolded during the conduction of each experiment. No more than this was reported about the control subjects. Subjects for the Replicated Look Experiments The experimental subject in the replication study was a blind little boy named Bryan, and the control subject was his sighted, identical twin brother, Gregory. The language development of the two boys has been studied longitudinally since they were age 1;1, and is still being observed at present. At monthly intervals, the 20 twins have been video taped in their home, while engaged in daily play or eating-type activities. Bryan and Gregory were born approximately three months prematurely, with birth weights of 1080 grams and 1060 grams respectively. As is common for premature infants, both boys experienced many medical complications in the neonatal period, including ROP. Gregory, the firstborn, and smaller twin suffered stage 3 ROP, which left him with near normal vision in one eye, and ability to see only about 15 centimeters in front of the other eye. Bryan, the secondborn and slightly larger twin, suffered stage 4 ROP with bilateral retinal detachments, resulting in total blindness. Bryan is suspected to have some light sensitivity, since he will blink with the flash of a camera fired at very close range. The boys were hospitalized during the first three months of life and were discharged at around term. They were seen at regular intervals in a neonatal follow-up clinic, and are described as having undergone essentially normal neurological and medical development, in spite of their visual impairments. Bryan is still being followed by a multi-disciplinary team for the visually impaired at B.C.'s Children's Hospital. This team provides assessments of Bryan's development, and consultation and advice for his parents. Bryan and Gregory each said their first words at age 1;7. Their rates of linguistic growth have been normal according to sighted norms. At present, Bryan's linguistic developmental level is at age, not correcting for prematurity, and not taking into account his blindness. With respect to the visual vocabulary, both boys were using the words look and see at an age when sighted children typically include such terms in their lexicons. The twins are the only children of a young couple who, like Kelli's parents, have been very accepting of their children's visual impairments, and have constantly provided them with nurturing and stimulating environments. The 2 1 father is a blue collar worker, and the mother stays at home fulltime with the children. There is a large extended family including grandparents, great grandparents, aunts, uncles and cousins, who live nearby and assist in the twins' care. Procedure A set of five mini-experiments comprised Landau and Gleitman's look experiments. Each experiment was motivated by a particular question related to the investigation of Kelli's meaning for the word look, as it applied to herself.1 The stimuli for each experiment consisted of a subset of commands containing the verbs look, touch, taste and listen. Testing generally consisted of presenting Kelli and the control subjects with the commands and waiting for their behavioral responses. The children's understanding or interpretations of the stimuli were inferred from the behaviors which the commands elicited. Each mini-experiment was conducted temporally independent of the others, such that periods of two to eight weeks separated consecutive experiments. Kelli was involved in four of the five experiments, while the fifth experiment involved the sighted controls only. With Kelli, the experiments were conducted in the following order: one, two, four, three. Only Experiments One, Two and Five were conducted with the sighted controls. Order of presentation of these experiments was systematically varied for the control subjects to determine a potential learning effect. Landau and Gleitman later reported that no learning effect was observed. Commands for each experiment were presented during informal play sessions that took place in the homes of the children, or in a familiar laboratory play setting. Time intervals between the delivery of commands within a single session ranged 1 These experiments were meant to address four hypotheses (see Landau & Gleitman, 1985:54). Since the hypotheses are problematic and are not explicitly referred to in the authors' subsequent discussions of results, they are not presented here. 22 between one and thirty minutes. All experimentation was video taped, and results were gathered entirely from the video recordings. All analyzed response behaviors, including verbal responses, were coded according to relevant variables. Sample reliability checks for coding procedures were made by an independent observer. The reported inter-rater reliability was always at or above 0.90 (p.56). In the replication study, all experimentation was carried out within informal play sessions, in the home of Bryan and Gregory. Two experimenters were used in the data collection procedure. One person (usually myself) delivered the commands, while the second person operated the video equipment and assisted in entertaining Gregory away from the experimental play setting. During the early phase of testing, an attempt was made to keep Bryan and Gregory separated, so that Bryan's responses to commands would not become an obvious model for Gregory come time for him to serve as the control subject in the study. After a certain point in testing, endeavors to separate the boys were discontinued since it became apparent that Bryan had been a natural behavioral model for Gregory long before my experimentation began, and because the twins were uneasy when apart. Test format consisted of presenting the boys with the experimental commands, and awaiting their responses. Within the play context, the pacing of the delivery of test stimuli was varied. Since Landau and Gleitman reported that the order in which experiments were conducted did not influence the children's pattern of responding, the scheduling of command presentations in the present study was managed somewhat flexibly. At a few points in the overall investigation, for reasons of convenience, commands from different experiments were delivered within the same play session. Consonant with the original study, all experimentation was video recorded for subsequent analysis. Codings for all analyzed response behaviors were checked by an independent observer. Inter-rater reliability in the replication study was 1 0 0 % . 23 Each of the four look experiments which involved Bryan are presented in this chapter as self-contained units. The presentation of each experiment is sub-divided into the following sections: Objective of Experiment - This section contains a very brief statement of the motivation for each experiment. Landau and Gleitman's Procedure - This section presents available details of the experimental procedures used by these researchers. Critique - This section critically reviews the procedures used for each experiment in the original study. Included is a discussion of methodological problems that either, 1) existed and were overlooked in the original experiment, or 2) could not a priori be eliminated because of the sparseness of detail in Landau and Gleitman's published description of their procedures. Procedure Used in the Present Study - This section outlines procedures used for each experiment in the replication study, including a description of any implemented procedural modifications. For the first experiment, transcripts of the command presentations and analyzed responses are detailed in the procedure section, to provide an immediately clear image of how all experimentation was carried out. Landau and Gleitman's Results and Conclusions - This section summarizes the results of each of the original experiment, and Landau and Gleitman's interpretation of them. Results and Conclusions of the Present Study - This section reports the results for each experiment in the replication study, and my interpretation of them. The experiments conducted with Gregory and the control subjects are discussed in a separate section following the presentation of the experiments involving Bryan and Kelli. Although it is nonconventional to separate discussions of the control and experimental subjects in this manner, I judged that the segregation would 24 assist the reader in organizing the abundance of information reported in relation to each of the five look experiments. EXPERIMENT ONE Objective of Experiment Through the first experiment, Landau and Gleitman sought to answer the question, 'Does look mean touch to the blind child?'. Landau and Gleitman's Procedure In Experiment One, at age 3;0, Kelli was presented with the commands: Look up, Look down. Look behind you, Look in front of you, Look over here by me, and Look over there by mommy. The contexts for command presentations varied such that sometimes there was an object in the location mentioned in a command, and at other times not. When an object was situated in the named location, Kelli may or may not have been aware of this. Critique In reporting the above procedural facts for Experiment One, Landau and Gleitman omitted certain items of information, the absence of which precluded an exact reproduction of the original methodology. This omission of information also necessitated that assumptions be made in order to accept the validity of the reported causal relationship between experimental commands and elicited responses. As an illustration of where assumptions had to be made, consider that no details were provided about the types of objects used as props whenever an object was to be situated in the location named in a command. Although it was mentioned that a graduated ring tower was used in conjunction with the command 25 Look behind you (p.56), no more than this was indicated. Due to this lack of specificity, it had to be assumed that highly novel or unfamiliar items were not employed as props, since observed responses consisting of manual exploration of such items could plausibly be attributed to Kelli's curiosity about the objects themselves, as opposed to her interpretation of the heard commands. Also left unreported was specific information about the situational and verbal contexts in which Kelli was presented with the test commands. At least three approaches to delivering the commands could have been used: 1) the experimenters might have opted to wait for opportune moments, during which it would have been contextually appropriate and logical to instruct the child to look in a particular direction, or 2) they could have simply presented the commands 'out of the blue', irrespective of the verbal and situational context at hand, or 3) although unlikely, they could have delivered the commands within a play-type activity centered around the experimental instruction to look. The following hypothetical contexts for the presentation of commands demonstrate how response behaviors could have been influenced by factors unrelated to the experimental stimuli. 1) The delivery of commands within contexts that naturally called for an instruction to look. Child X could have been engaged in an activity such as playing with a puzzle, whereby the pieces for the puzzle would have been scattered about her position on the floor. While the child would have been using her hands to locate a puzzle piece, an experimenter might have said, Look in front of you. The child might then have oriented her hands in a forward direction on the floor, and have found the puzzle piece. Based on this series of events, it could have been concluded that for this child, look means seek and contact with the hands. The response-bias in this hypothetical context becomes clear when acknowledging that a statement such as, there's a puzzle piece in front of you, would likely have been followed by the same 26 behavior as that which followed, Look in front of you. Because the child was already seeking in order to contact before the command was given, it could not irrevocably have been said that the instruction to look had a causal effect on what Child X did with her hands. 2) Presentation of test stimuli 'out of the blue'. Landau and Gleitman's procedure for delivering commands might have involved any of the following types of verbal contexts: (A) Experimenter: Oh, there's a ball. Look behind you. (B) Experimenter: Oh, mommy has a ball. Look over there by mommy. (C) Experimenter: Oh, I see a ball. Look behind you. Within a context similar to (A) above, Child X might have been observed to orient her hands behind herself to successfully contact the ball. Based on the temporal proximity between the command presentation and the child's conduct with her hands, it might readily have been assumed that the manual exploration was elicited by the instruction to look. Yet, this assumption could reasonably have been challenged by the assertions that the child's response was a function of: - her own desire to apprehend the ball. - having been informed through the first utterance that the ball was in her presence. - having been informed through the second utterance that the ball was 'behind' her. Extraneous verbal input to the child such as that contained in contexts (A), (B), and (C) above, could strongly influence the behavior that would follow the presentation of a command. As a result, such input could undermine the validity of an inferred causal relationship between the stimulus command and any ensuing behavior. 3) Presentation of test stimuli in the context of a game. Although unlikely, Landau and Gleitman might have delivered the commands within a game-type 27 activity expressly designed for the delivery of instructions to look. Yet, the inherent goal (of seeking or finding) in such an activity would inevitably have biased the elicited responses. Because the presentation of any statement alluding to the location of a sought-after object in this type of context would likely have prompted a single response behavior, use of this context for the delivery of a look command, could also have undermined an inferred causal relationship between the the command and the ensuing response. Note that these examples all present look in its look for as opposed to look at meaning. Since Landau and Gleitman did not describe the situational contexts in which the test stimuli were delivered, it must be assumed that they avoided using response-biasing contexts similar to those described above. In summary, the sparseness of procedural information in Landau and Gleitman's published account leads to much speculation about both the method by which the reported findings were obtained the validity of these researchers' conclusions. Procedure Used in the Present Study In accord with all available methodological guidelines, an initial attempt to gather data for the first experiment was carried out with Bryan at age 3;1. This first attempt spanned four experimental play sessions conducted within three consecutive days. During the initial taping sessions, time was allowed for myself and my co-experimenter to acquaint ourselves with Bryan and Gregory, and they with us. Due to the children's high degree of sociability, little to no time was required before they were comfortable interacting with us, meaning that experimentation quickly got under way. However, adverse to the flow of experimentation, at the time that testing began, Bryan was recovering from a cold. As a result, the reliability of behaviors elicited by some of the commands fluctuated with the child's varying degree of alertness. Consequently, a small 28 portion of the data collected when testing commenced was rendered non-usable due to Bryan's reduced responsiveness. Because Landau and Gleitman's book did not describe their method for presenting commands in Experiment One, a number of situational and verbal contexts were tried and modified in the replication study, in order to arrive at a procedure that was minimally response-biasing. This process revealed potential problems in Landau and Gleitman's investigation, as discussed in the previous section. When determining which stimulus-response pairs from all accumulated data would be analyzed in each experiment, the following criteria were adhered to: 1. Bryan demonstrated some form of attending behavior towards the experimenter at the time the command was presented, e.g., sat still and silent as the experimenter spoke, or verbally responded to hearing his name called immediately before the command was delivered, or repeated the command out loud before responding to it. 2. Bryan was not already engaged in searching behavior at the time a command was presented. 3. Bryan's behavior which followed the command presentation was clearly an effort to fulfill the command, meaning that replies such as, "I don't want to," were not analyzed. This criterion also excluded from analysis, any behavior which demonstrated post-hoc that Bryan's attention had not actually been secured prior to the delivery of the command. 4. The command was presented no less than 30 seconds after the delivery of a previous command to which Bryan was criterially attending. This criterion was intended to prevent the establishment of a routine in which Bryan would repeatedly await the directional information contained in a command, then search in the named location. 5. A latency period of no more than 5 seconds existed between the presentation of a command and the initiation of a response. To follow are descriptions of the exact contexts out of which were derived analyzable data for the first experiment. Bryan's age at the time that each stimulus was presented is indicated adjacent to the commands. As will be apparent, two commands were delivered when Bryan was age 3;4, while all other stimuli were presented when he was age 3;1. 2 9 B: = Bryan C: = Cynthia 1. Look up. Bryan - age 3;1. While Gregory and his mother were busy in another room, Bryan and I sat and played on the floor in the children's bedroom. Our play activity consisted of matching circle blocks having variously textured surfaces, with round indentations on a board which had similarly textured surfaces. Several minutes into the activity, Bryan matched one of the circles with its corresponding indentation on the board, though the game was not yet complete. C: Another one? B: I got to get another puzzle, (referring to another play item) C: Okay sit still. Okay sit still and listen. Bryan remained sitting with the board for the game positioned directly in front of himself on the floor. He began to trace the circle indentations on the board with his right forefinger. B: I want to get another puzzle. C: Okay, are you listening? Bryan began feeling the textured indentations on the board with his left hand, while keeping his right hand still at his side. B: Where's another puzzle? C: Okay Bryan? Bryan? B: Uh. C: Are you listening to me? B: Yes. Bryan continued to sit and feel the indentations on the board with his left hand. His head was oriented in a forward position. While sitting beside him on the floor, I held one of the circle shapes above his head. 30 C: Okay, look up. Immediately after hearing the command, Bryan hyperextended his neck so that his face was oriented approximately towards the ceiling. While holding his head in this position, he continued to feel the indentations on the board with his left hand, as his right hand remained at his side. C: Look up Bryan. Bryan extended his head back even further and maintained the position for a total of 9 seconds. 2. Look down. Bryan - age 3;4. Bryan and I were alone in the living room, as Gregory and his mother were occupied in the kitchen. Bryan had been talking about wanting to play hockey, though he didn't really know what the game was. He stood in front of where I was kneeling on the floor, and with his feet anchored on the ground, he began to turn from side to side swinging his arms. B: Hockey, hockey, hockey, I'm playing hockey. C: No, I'm going to teach you how, okay Bryan? I then took hold of his arms in order to stop him from swinging from side to side. C: I'm gonna teach you how. You have to do what I tell you. C: You have to do what I tell you. B: Oh. While holding onto Bryan's hands, I began to sway him from side to side. C: You have to walk over here, then walk over here. 31 Bryan laughed in amusement. A few seconds later, I stopped swaying him and he stood still in front of me. C: And now this is what you have to do. Okay listen. Okay you have to listen. I released Bryan's hands, and he remained standing still. C: Okay, Bryan now look down. B: Okay, I'll look down. Bryan immediately bent at the waist and at the knees, extended both hands towards the floor, and touched the floor between his feet. Next, using each of his hands he felt the floor in front of, and to the sides of his feet, then swung his hands through the air along side his feet twice. Before standing upright again, he scraped the floor between his feet with each of his hands. 3. Look behind you. Bryan - age 3;1. Bryan and I were alone in the living room, as Gregory and his mother were playing in the children's bedroom. Bryan was standing near an end table where he was gathering all of he drink coasters he could find on the table. He proceeded to talk about the coasters, referring to them as cookies on a plate. I was sitting on the floor in the living room, approximately 6 feet away from Bryan. B: I got to put it in the plate. C: Where are you putting it? B: In the plate. C: In the plate? Hmm. I stood up from where I had been sitting to see if I could find another coaster somewhere in the room. C: Let me see if there's another one. I sighted an additional coaster. 32 C: Oh here's another one. I retrieved the coaster, then returned to my place on the floor. As if ignoring me, Bryan continued to fidget with his coasters at the end table. Without him knowing, I placed my coaster on the floor approximately 2 feet behind him. C: Bryan look behind you. B: Behind you? Simultaneous to saying, "behind you?", Bryan grasped the edge of the end table with both of his hands, remained facing the table, and extended his right foot backwards. He contacted the coaster with his foot. In a smooth motion, he let go of the edge of the table with his right hand, pivoted his body to the right approximately 90° , bent at the waist and slightly at the knees, and extended his right arm and hand to where his foot was touching the coaster. He took the coaster into his right hand, returned to his upright position facing the end table, and placed the coaster with the others. C: Where are you putting it? B: Oh. C: You're putting it in the ash tray. Although in the above context, Look behind you was preceded by the the statement Oh here's another one, behavioral evidence clearly indicated that this extraneous verbal input did not predispose Bryan to a searching response. Consequently, the command presentation was considered analyzable. 4. Look in front of you. Bryan - age 3;1. Bryan and I were sitting on the floor in his bedroom, as Gregory and his mother were playing in the living room. We continued to play with the game for which the objective was to match the variously textured circles with their similarly 33 textured, round indentations on a board. More than the criterial amount of time had passed since the command look up had been delivered in this context. With the board situated on the floor directly in front of him, Bryan began tracing the shapes of the indentations with each of his forefingers. I was sitting beside him and holding onto one of the circle blocks for the game. C: You need another piece. Bryan continued to feel the indentations on the board with his right forefinger. His left hand was held still at his side, and his head was oriented in a forward direction. C: Does that feel rough? B: Yes. C: Or smooth? B: Smooth. C: You need to put another piece in there. B: Yeah. C: Okay. Bryan was still tracing and feeling the circles on the board. Without him knowing, I placed the block which I had been holding onto, in front of the board for the game. C: Look in front of you Bryan. Bryan immediately reached as far forward as he could with both of his arms extended out in front of himself. He contacted the front of the board with his left hand, then leaned back and turned to his right. From there he rose to his knees, and leaned forward across the board, supporting himself with both hands placed on the board. C: Look in front of you. 34 Bryan then used his right hand to reach in a forward direction beyond the board. He located the circle which I had placed on the floor in front of the board. 5. Look over here by me. Bryan - age 3;1. Bryan and I were alone in the living room. We were sitting on the floor, facing each other with an approximate distance of two feet between us. Situated between Bryan's legs was a stacking toy which he had been playing with. Bryan was holding onto the final piece to be placed at the top of the stack with his right hand, while touching the bottom of the stack with his left hand. B: It's going to the bottom. I touched the piece that Bryan was holding onto. C: Not this piece. This piece goes on top. B: Oh on top? Bryan touched the top of the stack with his right hand. I also touched the top of the stack. C: Yeah, on there. C: Right there, that's where it goes. B: Uh. Bryan switched the piece he had been holding, into his right hand, and felt the middle of the stack with his left hand. I grasped the piece he was holding, and guided his hand up to the top of the stack. Bryan released his grasp of the piece, and I placed it in its position at the stack's peak. C: There we go. C: There, it's on. Bryan dropped his hands to his sides, then touched his left foot with his left hand. 35 B: I want the ball. I tapped Bryan's left foot with my hand. C: What do you need now? You need the ball. Bryan remained still with the stacking toy between his legs, his right hand at his side, and his left hand touching his left foot. Without Bryan knowing, I had been holding onto the ball that was used in a game played with the stacking toy. I placed the ball on the floor directly in front of me, and a couple of feet away from Bryan. C: Look over here by me. Bryan immediately extended his left arm and hand in my direction, and quickly located the ball which was within his reach on the floor. He took the ball into his hand and proceeded to place the ball at the top of the stacking toy. 6. Look over there by mommy. Bryan - age 3;4. Bryan, Gregory and I were playing a game of 'Simon Says' in the living room. I was kneeling on the floor, and the boys were standing in front of me. The boys' mother was sitting on the floor approximately 5 feet from where Bryan was standing. Bryan was aware of his mother's presence in the room. C: Simon says, "Sit down." Both of the boys fell quickly to the floor, and landed in a sitting position. C: Simon says, "Stand up." The boys both rose to their feet, and stood waiting for the next command. 36 C: Simon says, "Look over there by mommy." Bryan pivoted his feet 90° so that he was oriented in the direction of where his mother was sitting. He then extended his arms and hands out in front of himself and ran a few steps towards his mother. Upon contacting his mother with his hands, he proceeded to hug her. Meanwhile, Gregory stood back and pointed at his mother. Landau and Gleitman's Results and Conclusions In the first experiment, Landau and Gleitman found that when Kelli was instructed to look in a given direction, she consistently oriented her hands in the specified direction, and sometimes explored to find the referred-to object. Never did she tilt her head towards the presumed location of the object. Even when instructed to look behind herself, her head remained facing forward. Based on Kelli's responses to the commands for experiment one, Landau and Gleitman concluded that ''look means contact with the hands to the blind child', (p.56). Results and Conclusions of the Present Study Bryan's responses to the commands delivered for the first experiment, were outlined earlier in the transcripts. However, a summary of the experimental findings is presented below. STIMULUS COMMAND RESPONSE 1. Look up. Bryan hyperextended-extended his neck, to orient his face towards the ceiling. 2. Look down. Bryan bent over and oriented his hands towards the floor. He manually contacted the floor while exploring the area around his feet. No object was referred to in this command presentation. 3. Look behind you. Bryan extended his right foot behind himself and contacted the target object with his foot. Next he turned and extended his right hand to where his foot was touching the object. He then picked up the object. 4. Look in front of you. Bryan extended both of his arms out in front of himself so that his hands were oriented in a forward direction along the floor. He then propped himself up so that he could lean further forward. While extending out his right hand, he manually explored the floor in front of himself. Upon contacting the target object with his hand, he picked it up. 5. Look over here by me. Bryan extended his left arm out in front of himself, so that his hand was oriented in the direction of where I was sitting on the floor. He contacted the target object with his hand, and picked it up. 6. Look over there by mommy. Bryan turned so that he was facing his mother. He then extended his arms and hands out in front of himself and ran forward a few steps. He contacted his mother with both hands, and proceeded to hug her. Bryan demonstrated a similar type of behavior in response to all commands except Look up. Upon hearing this particular instruction, Bryan oriented his face in the named direction. Given the peculiarity of this behavior, it was decided that the same command should be delivered at later points in time, in order to test the reliability of the head-tilting response. Following repeated presentations of the command, it was found that Bryan's head-tilting behavior constituted a reliable response. The illogical nature of this response led to an inquiry about its occurrence. Bryan's mother was asked if she regularly instructed Bryan to Look up within a specific context. She reported that the command was often presented within the context of having him pose for a photograph. Furthermore, she indicated that the instruction was usually followed by, "Hold your head up." Within the picture-taking context, the implicature of Look up, was clearly for Bryan to orient his face towards the camera, which would tend to be situated in an 38 upward direction, relative to where he would be positioned. In effect, Bryan had been taught a specific response to the instruction Look up, such that he had come to interpret the command as an unanalyzed whole. Given this insight into Bryan's previous experience with the first command, I decided to exclude the idiosyncratic stimulus-response pair from the analysis of his meaning for look. As was already mentioned, all other commands for the first experiment elicited seeking behavior from Bryan, for which his goal was to apprehend the target object with his hands. When looking beyond a general description of Bryan's response behaviors, variation is seen with respect to the body part he used to explore his environment. Most often Bryan used his hands to locate the target object. Yet, when instructed to look behind himself, he used his foot for the exploratory purpose. Furthermore, some of the command presentations which had been deemed non-analyzable for experiment one, revealed that Bryan would alternatively use his back, or his arms, or his legs as the instruments of exploration, when seeking a target object. In conclusion, if Bryan's response to Look up, can be exempted from analysis for the first experiment, the findings as reported above indicate that for Bryan, look means 'seek through the haptic perceptual modality, in order to apprehend with the hands. EXPERIMENT TWO Objective of Experiment Observations made in Experiment One led Landau and Gleitman to consider the possibility that look and touch were synonymous for Kelli. Consequently, their objective in the second experiment was to determine whether or not Kelli's meanings for the two perceptual verbs were distinct. 39 Landau and Gleitman's Procedure Landau and Gleitman prefaced the second experiment by presenting Kelli with various instances of the instruction, Touch X but don't look at it. In response to this command, Kelli "would usually simply touch or stroke the object or bang it with her fist" (p.57). Next, Kelli was told, Now you can look at it. In response to this instruction, Kelli "would manually explore the object extensively, running her hands all over its surfaces" (p.58). Having observed this distinction in Kelli's interpretations of look and touch, Landau and Gleitman designed a task which they reasoned would illustrate a meaning distinction with even greater acuity. In Experiment Two, the researchers presented Kelli with thirteen pairs of commands, in which a variety of adverbial and prepositional phrases were used with each of the two perceptual verbs. In describing the twenty-six test stimuli, Landau and Gleitman reported, Using several toys, we asked Kelli to look or touch (1) with spatial modifiers: up, behind you, in that (container), under some (object), and here; (2) with intensity modifiers: real hard, gently, real good; and (3) with instruments of contact or perception: with your finger, fool, nose, mouth, ear. For each of these modifiers two commands were given, one using look, and the other using touch. (p.58-59) The commands were presented in a random order across three experimental play sessions, when Kelli was age 3;0 (but "several weeks" older than in Experiment One). Critique As with the previous experiment, Landau and Gleitman described their procedures for Experiment Two with a level of detail that precluded an exact reproduction of their methodology, and which made it difficult for the reader to independently evaluate their reported findings and conclusions. Landau and Gleitman did not describe the object referred to as "X" in the preliminary commands, Touch X but don't look at it, Now you can look at it. Given 40 that Kelli responded differentially to look and touch in these commands, it would be interesting to know if the objects used as props had any particular demand characteristics. Itemization of the objects employed could have eliminated concerns about the effects of the objects' inherent attributes on the responses associated with them. Furthermore, evidence of differential responding to the two verbs would be more compelling if the manipulated objects were intrinsically less apt to being handled in varied manners. Neither did Landau and Gleitman mention whether the "several toys" used in Experiment Two were familiar to Kelli. Because an unfamiliar toy would naturally elicit exploratory behavior from any child, it is presumed that Kelli was not instructed first to look at an object that was unfamiliar, then later told to touch the same object. Such ordering of command presentations could mean that a comparison would be made between what the child did when encountering a new toy, and what she did in response to a command to touch a toy which was no longer new to her. As a final comment regarding the props in Experiment Two, we do not know if the same object was used in conjunction with both commands to look or touch in any given manner. Inconsistent use of props would create two potential independent variables (the verb in the command, and the object referred to) in a research design that should have only one variable (the verb in the command). While the use of different objects might have had no effect on Kelli's responses to commands in this experiment, the above concerns cannot be eliminated a priori. Rather than specifying the verbs and prepositional or adjectival phrases used in the test commands for Experiment Two, Landau and Gleitman might have simply listed the exact wordings of the 26 test stimuli. Without a verbatim report of the presented commands, we do not know for certain whether or not the referred-to objects were named in the commands. Furthermore, if the object was 41 named, it remains unknown whether the child was instructed to look at, or look for this object in a specified manner when given a look command. Because in the previous experiment look was used in commands which implied look for, there is reason to suspect that either preposition could have been used in the commands for Experiment Two. Failure to specify which preposition was used is considered nontrivial since look for and look at have clearly different interpretations, with vision being much less central to the former than the latter. Without a verbatim list of the test stimuli, we also don't know whether the object, if named, was situated in the same sentential position for both the look and touch versions of a single command-type. For example, with respect to look/touch in some container, the only logical way to word the touch command would be, Touch the object in the container. On the other hand, the look command could logically have been worded, in either of two ways: Look in the container at/for the object, or Look at/for the object in the container. Although both wordings for the look command are grammatical, the latter wording is consistent with the sentence structure of the touch command. Using a single sentence structure in which only the verb varied, would be the most controlled way of comparing a child's responses to look and touch. Aside from leaving the reader uninformed about the controls which may or may not have been used in the original experiment, this omission of procedural information also contributed to the number of independent decisions to be made regarding the procedures to be used in the replication study. In the Critique section for Experiment One, a few issues were raised regarding the paucity of information provided about the verbal and situational contexts in which Landau and Gleitman presented their commands. Those same issues are relevant to the procedures which may have been used in Experiment Two. Without a description of the contexts in which touch and look commands were presented to 42 Kelli, it can only be assumed that response-biasing contexts were not inadvertently employed. As already noted, Landau and Gleitman presented Kelli with the command Touch X but don't look at it, before conducting Experiment Two. This juxtaposition of the two perceptual verbs in a single command may have jeopardized the validity of conclusions drawn about distinctions in Kelli's meanings for look and touch, prior to hearing the command. Upon being instructed to, touch but not look, a child who is 'eager to please' would be forced to make a distinction between the two verbs in order to successfully do as she was asked. In effect, the induced distinction might then be learned, and carried over into responses to commands for Experiment Two. In view of what Landau and Gleitman were trying to find out in the second experiment, their timing for the presentation of 'touch but don't look" was nonoptimal (cf. Werth, 1983). Procedure Used in the Present Study The replication of Experiment Two was carried out across two play sessions when Bryan was age 3;3, and 3;4. Throughout experimentation, all procedural guidelines provided by Landau and Gleitman were adhered to; commands in which look and touch were each paired with adverbial and prepositional phrases were randomly presented to Bryan in various play contexts. Because of Bryan's exceptional response to Look up in the previous experiment, I decided not to include Look/Touch up in the set of commands to be delivered in this experiment. Beyond the methodological guidelines provided by the original experimenters, the following procedural rules were decided upon and adhered to: 1. When an object was to be responded to in a command, that object would be named in the command. 2. Bryan would be familiar with all objects referred to in the commands. 3. The same object would be referred to in both the look and touch versions of a single command-type. 43 4. The look and touch commands containing the same adverbial or prepositional phrase would not be presented contiguously, regardless of the time span that separated their presentations. 5. When an object was to be named in a look command, the child would be instructed to look at, versus look for, the object. 6. A single phrase structure would be employed for all look and touch commands, such that when an object was to be mentioned in a command, it would be situated in the same sentential position for each of the look and touch versions of a command. 7. All objects used as props, including the object referred to as "X" in the command, Touch X but don't look at it, would have no particular demand characteristics in terms of how they should be handled. 8. The commands, Touch X but don't look at it, and Now you an look at it, would be delivered after the presentation of the 26 test commands for experiment two. In order to determine which command presentations would be analyzed for the second experiment, all data was screened according to the analyzability criteria outlined in the procedures section for Experiment One. In striving to meet these criteria, response-biasing contexts were avoided in the delivery of all test commands. The following is a verbatim list of all of the commands presented to Bryan in Experiment Two: 1. Look behind you. 2. Touch behind you. 3. Look at the tennis ball in the bag. 4. Touch the tennis ball in the bag. 5. Look at the apple under the box. 6. Touch the apple under the box. 7. Look at the alarm-clock here. 8. Touch here. 9. Look at the purse real hard. 10. Touch the apple real hard. 11. Look at the tennis ball gently. 12. Touch the ball gently. 13. Look at the purse real good. 14. Touch the purse real good. 15. Look at the Lego with your finger. 16. Touch the Lego with your finger. 17. Look at the monkey with your foot. 18. Touch the monkey with your foot. 19. Look at the Ogopogo with your nose. 20. Touch the Lego with your nose. 21. Look at the bunny with your mouth. 44 22. Touch the bunny with your mouth. 23. Look at the bunny with your ear. 24. Touch the bunny with your ear. * Touch the bunny, but don't look at it. Okay now you can look at it. ** Touch the puppet, but don't look at it. Okay now you can look at it. Mentioned in some of the commands were a bunny, a monkey, and an Ogopogo. These three objects were all stuffed animals. The purse referred to in a few of the commands was a play item for the children. Procedural rule three was broken for the following command pairs: 9-10, 11-12, and 21-22. Any effects of having broken this rule will be discussed in the results section for this experiment. Landau and Gleitman's Results and Conclusions When analyzing Kelli's responses to the stimuli for the second experiment, Landau and Gleitman coded all data according to two parameters. First, pairs of responses for a single command-type were described as either differentiated or nondifferentiated. Next, all responses were coded as to whether they were of an exploratory, or nonexploratory nature. To exemplify the difference between exploratory and nonexploratory response categories, the researchers indicated: extensive manipulation of an object with examination of all its parts was called an exploratory response; banging the object with a closed fist was called nonexploratory. (p.59) Landau and Gleitman reported that Kelli distinguished her responses to look from her responses to touch for eight of the thirteen command-types. Those command-types which elicited differential responding included: Look/Touch up; behind you; real hard; gently; with your foot; with your finger; with your nose; with your mouth. For the remaining five pairs of commands, Kelli did not respond differentially to look and touch. It was also found that, for seven out of the eight command-types for which Kelli differentiated her responses, the behavior associated with the touch 45 commands was a form of contact that involved banging, scratching, or tapping (p.60). In contrast, the behavior associated with the look commands for seven out of eight differentiated command-types was consistently an exploratory or apprehending behavior. For example, when instructed to look, Kelli would manipulate, feel all over, pretend to eat, or smell the object referred to in a command (p.60). Furthermore, it was noted that, in general, whenever a body part was specified in the command as the instrument with which looking or touching was to be carried out, e.g. Look/Touch with your foot, the responses to look commands involved moving the target object to the named body part. On the other hand, responses to touch commands involved moving the named body part to the object (P.61). Statistical analysis of Kelli's responses to the commands for Experiment Two revealed that her pattern of distinguishing between the verbs look and touch was highly significant, (p<.025, Fisher exact test) (p.61). Landau and Gleitman concluded that, for this blind child, touch implied physical contact. On the other hand, look meant to perceive or apprehend, perhaps amodally. Their inference that Kelli's meaning for look was potentially amodal was strongly influenced by her responses to the commands Look at the object with your mouth/nose, wherein she tasted and sniffed the target objects. Prior to these command presentations, it had appeared as if look was tied to the haptic perceptual modality. Yet, after the tasting and sniffing behaviors were observed, Landau and Gleitman were considering that Kelli's meaning for look was not unimodal. Results and Conclusions of the Present Study Bryan's responses to the stimuli for Experiment Two are summarized below. 46 STIMULUS COMMAND RESPONSE 1. Look behind you. (Taken from experiment one.) Bryan extended his right foot behind himself and contacted the target object with his foot. Then he turned and picked up the object. 2. Touch behind you. Bryan pivoted 9 0 ° to his left and swung both arms up in the air. He pivoted further to his left, and shouted, "Touch, touch, boom, boom", while still swinging his arms. 3. Look at the tennis ball in the bag. Bryan reached into the bag containing the ball, secured his grasp of the ball, and pulled it out of the bag. He then rolled the ball between both of his hands, while feeling the entire surface of the ball. 4. Touch the ball in the bag. Bryan reached into the bag and grasped the ball. While holding onto the ball he said, "The ball's in the bag." He then pulled the ball out of the bag and said, "There." 5. Look at the apple under the box. Bryan banged the top of the shoe box once with each hand, then lifted the box up and placed it right side up on the floor. He made no attempt to locate or contact the apple. 6. Touch the apple under the box. Bryan banged the shoe box once with each hand, then said, "Where is it?". When told that the apple was under the box, he turned the box over, and felt the inside of the box. He made no contact with the apple. 7. Look at the alarm clock here. Bryan took the clock into his hands, and turned it repeatedly between his hands, feeling all surfaces of the clock. 8. Touch here. Bryan clapped the fronts of his hands together once. He then slapped the front of his left hand against the back of his right hand five times. 9. Look at the purse real hard. While holding the purse with his left hand, Bryan banged the purse once with his right hand. 47 10. Touch the apple real hard. At first Bryan held the apple still between both hands. He then pressed the back of his left hand against the apple. 11. Look at the tennis ball gently. Bryan held the tennis ball against his chest with his left hand. 12. Touch the tennis ball gently. Bryan placed the ball on the floor, and flicked the three middle fingers of his right hand against the ball. 13. Look at the purse real good. Bryan held one of the handles for the purse with his right hand, and felt the surface of the purse with his left hand. He then used both hands to feel one of the handles on the purse. 14. Touch the purse real good. Bryan held onto the purse with his hands as he touched it with his feet. He then said, "Touch the purse with my foot." Next he slid his feet through the handles on the purse. 15. Look at the Lego with your finger. At first Bryan rolled the block between both of his hands. Then, he held it in his right hand, while contacting various surfaces of the block with his extended left index finger. 16. Touch the Lego with your finger. Bryan held the block with his right hand while pressing the three middle fingers of his left hand against the block. 17. Look at the monkey with your foot. Bryan picked up the monkey, and drew his feet in close to his body. He then tapped the monkey against his toes. Next, he placed the monkey underneath his feet and brushed the surface of the monkey with his feet. 18. Touch the monkey with your foot. Bryan held the monkey with both hands and tapped his right foot with it, three times. 19. Look at the Ogopogo with your nose. Using both hands, Bryan raised the Ogopogo up and held it against his nose. He then shook his head, thereby rubbing his nose over the surface of th object. 48 20. Touch the Lego with your nose. Bryan held the block in his right fist, and tapped his nose with the block three times. 21. Look at the bunny with your mouth. Bryan raised the bunny up towards his open mouth, and tapped it against his mouth four times. 22. Touch the bunny with your mouth. Bryan drew the bunny up and held it against his mouth. 23. Look at the bunny with your ear. Bryan raised the bunny up and pressed it against his right ear. 24. Touch the bunny with your ear. Bryan raised the bunny up to his left ear, and held it against his ear. * Touch the bunny, but don't look at it. Now you can look at it. ** Touch the puppet but don't look at it. Now you can look at it. No reliable responses obtained. Upon repeated presentations of the instructions (*, **), Bryan demonstrated neither looking nor touching behavior. Following Landau and Gleitman's data analysis procedure for Experiment Two, Bryan's responses to the look and touch versions of each command-type were initially judged as differentiated or nondifferentiated. All responses were then coded as either exploratory or nonexploratory. If the two responses to a single command-type, were coded similarly along the exploratory/nonexploratory parameter then the pair of responses were categorized as nondifferentiated. Likewise, if the responses were coded dissimilarly along the same parameter, they were described as differentiated. When assigning an exploratory coding to Bryan's responses, particular behavioral attributes were consistently watched for, including: physical contact with much of the surface area of an object, versus contact with a very restricted surface area; enduring contact with an object, versus either quick contact, or 49 repetitions of instantaneous contact with the object; and the presence of scanning or surveying of the environment, versus the absence of such behavior, when no prop was used during a command presentation. Analysis of Bryan's responses to the commands for Experiment Two revealed the results detailed in Table I below. T A B L E I Differentiation of Bryan's Responses to Look and Touch According to the Exploratorv/Nonexploratory Parameter.  C O M M A N D T Y P E S L O O K T O U C H D I F F E R E N T I A T E D behind you in the bag here real good with your finger with your foot with your nose N O N - D I F F E R E N T I A T E D under the box real hard gently with your mouth with your ear exploratory exploratory exploratory exploratory exploratory exploratory exploratory undetermined nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory nonexploratory undetermined nonexploratory nonexploratory nonexploratory nonexploratory As is indicated in Table I, Bryan differentiated his responses to look from his responses to touch for seven of the twelve command-types. It can also be seen that whenever differential responding was observed, the behavior associated with the verb look was always of an exploratory nature, while the behavior elicited by touch was always nonexploratory. Furthermore, with the exception of Look/Touch under the box, all command-types for which nondifferentiated responses were obtained elicited nonexploratory behavior, regardless of the verb contained in the command. Bryan's responses to Look/Touch the apple under the box suggested that he did not understand the preposition under. Although he always sought the apple referred to 50 in the two commands, he never succeeded in apprehending it. Consequently, the exploratory/nonexploratory parameter of his responses could not be determined. General similarities existed between Bryan and Kelli's response patterns in Experiment Two. For example, both children differentiated their responses for several command-types. Additionally, when differential responding occurred, each child tended to associate exploration with look, but not with touch. Yet, in spite of such broad similarities, some interesting variations also existed. For example, there were only four command-types for which both children contrasted their interpretations of look and touch. This meant that each child responded differentially to a number of command-types for which the other child did not. Moreover, Bryan's exploratory behaviors always involved the haptic perceptual modality, whereas Kelli additionally explored with olfaction and taste. Variation in the children's response behaviors also existed in relation to the nonexperimental commands: Touch X but don't look at it. Now you can look at it. Kelli's responses to the instructions suggested that she made a semantic distinction between the perceptual verbs. That distinction was consistent with the divergence which characterized her responses to the verbs within the test stimuli for the experiment. On the other hand, repeated presentations of the nonexperimental commands to Bryan elicited a variety of irrelevant behaviors. In effect, only unreliable responses could be obtained for the instruction to touch but not look. The unreliable nature of Bryan's behaviors provoked the suspicion that the child may have been uninterested in deciphering the meaning of the confusing instructions. Alternatively, it is possible that the unreliable responses may have reflected a metalinguistic inability to distinguish two words that did not have overtly different meanings for him. Regardless of how the behaviors might be explained, Bryan's responses to the nonexperimental commands contrasted with Kelli's. As a 51 result, variation in the children's response behaviors could be observed with respect to one more component of the second experiment. Before stating my conclusions for experiment two, a methodological issue relevant to the replication study must be addressed. It was mentioned in the Procedures section, that despite all efforts to be consistent in referring to the same object in both the look and touch versions of a single command-type, three of the thirteen command pairs were inconsistent. For commands 9 and 10, Look at the alarm clock here/Touch here, Bryan associated exploratory behavior with the instruction to look, but not with the instruction to touch. However, because an object was referred to in only one of these commands, the possibility arose that the referral/nonreferral to an object affected the obtained responses. In order to rule out this possibility, responses to additional command pairs were considered. In commands 3 and 4, Look behind you/Touch behind you, reference to an object was never made; yet, differential responding according to the verb in the command was observed. Such findings suggest that the absence of referral to an object could not have been associated with a particular type of response. Furthermore, when objects were referred to in both the look and touch versions of a command-type, (e.g. 15, 16, 17 and 18) differential responding according to the verb in the command was also observed. Hence, it would appear that explicit referral to an object did not influence the exploratory/nonexploratory aspect of a response. A similar line of reasoning was followed to determine that the inconsistency in commands 21 and 22 did not affect Bryan's responses to those stimuli. Different objects were also referred to in commands 11 and 12, Look at the purse real hard/ Touch the apple real hard. However, nondifferentiated responses were elicited by these commands. Evidence could be found within the experimental data, to suggest that the responses associated with commands 11 and 12 were not influenced by variable reference to an apple or purse. However, 52 because failure to validate the nondiffcrentiated responses would not have affected the outcome of the experiment, presentation of the validating evidence is not necessary. On the basis of Bryan's responses to the look and touch commands, I concluded that he interpreted the two verbs to be distinct in meaning. For this blind child, touch seemed to imply physical contact that is short in duration, and which involves a restricted surface area of the contacted object. In keeping with the previous experiment, Bryan interpreted look as meaning exploration through the haptic perceptual modality. EXPERIMENT THREE Objective of Experiment Having observed in the previous experiment that Kelli tasted and sniffed in response to the commands Look with your mouth/nose, Landau and Gleitman were considering the possibility that this child's interpretation of look could involve any perceptual modality. In Experiment Three, Landau and Gleitman investigated to see if Kelli had glosses for the perceptions tasting and listening by comparing her meaning for look with her meanings for these two perceptual verbs. Their goal was to determine whether Kelli's responses to Look with your mouth/nose, could be explained by a meaning for look which implicated nonspecific perceptual modalities, or if the responses had to be accounted for otherwise. Landau and Gleitman's Procedure Experiment Three was conducted when Kelli was age 3;3. The stimuli for the experiment consisted of a set of commands instructing Kelli to look at, touch, listen to, or taste various objects. There were three commands containing each verb, for 53 a total of twelve experimental commands. Landau and Gleitman was reported that the objects referred to in the commands varied in terms of the plausibility with which they could be acted upon in the specified manner (e.g. Look at the orange ring, Taste the radio, and Listen to the orange ring). The test commands were randomly presented to Kelli during a single play session. Landau and Gleitman indicated that one of the commands was inadvertently omitted, meaning that eleven stimuli were used in Experiment Three. Critique The first item to be raised in a critique of Experiment Three has to do with Landau and Gleitman's choice of stimuli, considering their objective for the experiment. Close inspection of the conditions under which look commands elicited nonhaptic perceptual behaviors revealed that amodal exploration occurred when the message within the stimulus was 'Look with a body part that has a perceptual capability which is not haptic' In general, when responding to this type of command, Kelli recruited the perceptual ability inherent in the body part specified to do the looking. Hence, it is clear that a particular form of request was correlated with the amodal responses to look. In view of their objective for this experiment, Landau and Gleitman should have examined this very correlation further. As an alternative to the stimuli presented in this experiment, instructions to listen to, taste and smell objects, with perceptual organs which would not normally be used for such purposes, might have provided more direct insight into Kelli's sniffing and tasting responses in Experiment Two. It is possible that a command such as Listen to the object with your mouth, would also have elicited a tasting response. Realization of this hypothetical stimulus-response pair, in addition to what Kelli did for Look with your mouth/nose, would have suggested that the perceptual modality used to carry out a request to look would likely be the 54 natural modality for the perceptual organ specified for the fulfillment of the command. Stimulus-response pairs such as this, would have served as compelling evidence to counter the suggestion that Kelli's meaning for look encompassed amodal perceptions. In summary, it is unclear why Landau and Gleitman chose to contrast Kelli's listening and tasting behaviors with her looking behaviors, under conditions which were not previously suspect of eliciting responses involving nonconventional perceptual modalities. Further critical comments regarding Landau and Gleitman's third experiment derive from bits of procedural information gleaned from their results and discussion section. In that section it was stated, "When asked to listen to some object, she (Kelli) became quiet and oriented her head to the sound but did not move toward it" (p.64). To supplement this report, Landau and Gleitman provided an illustration in which an experimenter was holding a radio a few inches away from the child's ear as the command Listen to the radio was presented. From all of this, it could reasonably be assumed that the object referred to in the instruction to listen, was being sounded concomitant to the command presentation. Although there was no reason to suspect that a command for Kelli to listen to a radio would elicit anything other than listening behavior, I am suspicious that the inferred context in which the command was delivered may have determined the response obtained. Because the radio was sounding and held at Kelli's ear as the instruction was given, it is highly possible that her listening behavior had nothing to do with the verbal command, but was a function of having had a noise-making object held at her ear. A more optimal way to determine Kelli's interpretation of listen, might have been to present her with an object which had the potential to produce noise, but was not sounding at the time the command was given. After hearing the command, Kelli might have attempted to affect the object so that it emitted sound, and then 55 have proceeded to listen to it. This series of events would have demonstrated that the child knew that sound was required in order for her to listen. Observed listening behavior within this context would more likely have been a product of the verbal instruction, than a product of the context itself. In view of the bias inherent in Landau and Gleitman's context for presenting Kelli with commands to listen, it is believed that little non-refutable insight was gained into the child's interpretation of the aural perceptual verb. Also contained in the results and discussion section for Experiment Three was the statement, "When asked to touch or look at objects, she (Kelli) reached out and took them" (p.64). The behaviors outlined in this statement, were considered as constituting Kelli's responses to the look and touch commands. Because no specific information was reported about the contexts in which commands were given, there is nothing to prevent the reader from considering that, when Kelli 'reached out and took' the objects she was instructed to either look at or touch, it may have been that she was not intentionally looking at or touching anything. Rather, she may simply have been accepting objects that she believed were being offered to her. In view of what look and touch elicited from Kelli in the previous experiment, the modestly defined and nondifferentiated responses to the same verbs in the present experiment might reasonably be doubted for their legitimacy as responses at all. All considered, it appears that there was significant variation in the conditions under which objects were presented to Kelli as commands for the third experiment were delivered. When asked to look at or touch an object, it seemed as if the objects were being handed to her. Yet, when asked to listen to an object, the object was sounded and held at her ear. Landau and Gleitman's general procedure for Experiment Three could have been improved by using a consistent manner of presenting objects to be acted upon in the fulfillment of each command. Without 56 this control, it can't be said that the reported findings were not significantly influenced by the varied manners in which objects were used as props. Procedure Used in the Present Study The replication of Landau and Gleitman's Experiment Three was carried out with Bryan at age 3;3. All twelve test commands involving the four perceptual verbs were presented. In keeping with the original study, the objects used in conjunction with each command varied in terms of the plausibility with which they could be acted upon in the named fashion. The specific objects employed as props included: a buzzer - this being a small box which emitted a "buzzing" sound when the button on the side of the box was depressed; an apple; a monkey (stuffed animal); an alarm clock; a bunny (stuffed animal); a doll which looked as if it were crying - referred to as "the baby"; a jewelry box which played music when wound up - referred to as "the music box"; and Bryan's own hand, which was referred to in one of the commands to taste. All objects used as props were familiar to Bryan. Consonant with Landau and Gleitman's procedure, the objects referred to in commands were varied in terms of where they were situated in relation to Bryan, and whether or not they were sounding as the commands were presented. In effect, simultaneous to the delivery of each look and touch command, the object referred to was placed in Bryan's lap, and was not sounding. Additionally, when a listen command was presented, if the object referred to was capable of producing noise, it was sounded and held at Bryan's ear. While replicating Landau and Gleitman's method for Experiment Three with as much exactitude as possible, two supplemental commands were presented to Bryan in order to monitor the effects of the varied location, and aural qualities of objects referred to in the stimuli delivered in the original study. The first supplemental 57 command was an instruction to look at a noise-making object. As this command was presented, the object was held near Bryan and was sounded. The second supplemental command was a listen command in which the object referred to was also capable of emitting noise. However, during this stimulus presentation, the object was placed in Bryan's lap and was not sounding. In accord with procedures used for the replication of Experiments One and Two, precautions were taken to avoid the situation in which Bryan's responses to test stimuli might have been undesirably biased by the verbal and situational contexts in which commands were delivered. The criteria used to determine which data would be analyzed for Experiment Three, were the same criteria outlined for this purpose in the Procedures section of Experiment One. Landau and Gleitman's Results and Conclusions Landau and Gleitman reported that Kel l i appropriately varied her responses according to the verb contained in a command for ten of the eleven stimuli presented. When told to touch or look at objects, Kel l i reached out and took them. When instructed to listen to an object, she became quiet and oriented her head to the sound but did not move towards it. Finally, whenever she was instructed to taste something, Kel l i responded by licking the object. The one command which elicited a questionably inappropriate response was Listen to the raisins. Upon hearing this command, Kel l i reportedly sat still, then reached out saying "raisins". The difference between appropriate and inappropriate responding according to the verb contained in a command was statistically significant (p = .005, Binomial Test). Landau and Gleitman concluded that Kelli's responses in Experiment Three indicated that for a blind child, "look is interpreted as calling for haptic exploration, while other specific verbs are used for apprehension by other means" (P-64). 58 Resu l ts and Conc lus ions of the R e p l i c a t i o n S tudy Bryan's responses to the commands for Experiment Three were as follows: STIMULUS COMMAND RESPONSE 1. L o o k at the apple . 2. L o o k at the monkey . 3. L o o k at the buzzer . At first Bryan held the apple in his right hand and banged it with his left hand. Next, he felt the surface of the apple with both hands. Using his hands, Bryan felt much of the surface of the monkey which had been placed in his lap. Bryan picked up the buzzer and immediately depressed the button to cause the object to emit its "buzzing" sound. 3. * L o o k at the music box (with the music box playing at the time the command was delivered). At first Bryan tried but failed to open the lid to the music box. He then held the sounding object at his ear. 4. T o u c h the bunny. 5. T o u c h the app le . 6. T o u c h the a l a r m c lock . 7. Tas te the apple . 8. Tas te your h a n d . 9. Tas te the monkey . Using the palms of both hands, Bryan repeatedly stroked the top surface of the bunny. Bryan picked up the apple that had been placed between his legs, and held it in both hands. Bryan placed both hands on the clock which was situated on the floor between his legs. He then picked up the clock and held it in the air. Bryan raised the apple to his mouth and held it against his closed lips. Bryan raised both hands up to his mouth and touched his lips with his fingers. He then licked his finger tips. Bryan raised the monkey up to his open mouth, and pressed it against his lips. 59 1 0 . Listen to the monkey. At first Bryan placed the monkey on top of his feet. After the command was repeated, he replied, "I can't." 1 1 . Listen to the baby. Bryan placed the doll aside and said, "The baby's gonna cry." 1 2 . Listen to the buzzer. Bryan became still and leaned his ear closer to the sounding object. He then reached up to grasp the buzzer, and held it against his ear. 12.'Listen to the buzzer (with the object not sounding at the time the command was delivered). Bryan took the buzzer into his hands, and immediately depressed the button to cause the object to sound. He didn't speak while the buzzer was sounding, and ignored all else that was going on around him. To restate the findings in more general terms, when instructed to taste an object, Bryan used his lips or tongue to contact the object. In effect, his interpretation of the perceptual verb taste was entirely conventional. In accord with the findings for the previous experiment, when instructed to touch an object, Bryan always made nonexploratory manual contact with the object. Hence, as was already suspected, Bryan's interpretation of touch was tied to the haptic perceptual modality. These findings were consistent with what Landau and Gleitman reported about Kelli. When instructed to listen to a sounding object, Bryan demonstrated listening behavior. However, when told to listen to an object which did not naturally produce sound, he replied, "I can't." Throughout all experimentation, this was the only instance in which Bryan stated that he couldn't do as he was asked. The exclusiveness of this reply, in addition to the fact that Bryan really could not listen to something that was incapable of producing noise, prompted the conclusion that the response to the command was reliable. In drawing this conclusion, it was inferred that Bryan knew that he couldn't listen without a sound source. 60 The instruction to listen to the baby (an object that was not realistically capable of producing sound), also elicited a response indicating that Bryan knew that sound was required before listening could occur. After hearing the command, Bryan stated, "the baby's gonna cry". In accepting this as a reliable response, it was judged that Bryan was prepared to listen with the pretending expectation that the doll was about to emit noise. Finally, Bryan's response to the supplemental command also suggested that he had a conventional interpretation for the verb listen. When told to listen to an object that was silent, but capable of emitting noise, Bryan caused the object to sound. All obtained responses to the listen commands were interpreted as support for the conclusion that Bryan, like Kelli, associated the aural perceptual modality with the aural verb. Bryan's responses to the look commands in the third experiment were somewhat discrepant with Kelli's reported responses to the same commands. When given an instruction to look at an object which did not naturally emit sound, Bryan responded by exploring the object manually. However, when told to look at an object that was capable of producing noise, Bryan did not explore the object manually, but did what was necessary to cause the object to sound. This response behavior was judged as having been influenced by the demand characteristics of the object referred to in the command. The supplemental command added as a control in the replication study also elicited an interesting response. When told to look at an object that was sounding concomitant to the command presentation, Bryan demonstrated listening behavior, as opposed to looking behavior. This stimulus-response pair also suggested that Bryan's response behaviors were affected by the demand characteristics of the object referred to in the command. The finding associated with the supplemental command supports the previous conjecture that Kelli's response to listen to the radio, may have had little to do 6 1 with the verb contained in the command, but instead was elicited by the aural characteristics of the object placed near her ear. Bryan's response to the supplemental command is also relevant to the suggestion that Kelli's responses to Look with your mouth/nose in Experiment Two were influenced by forces extraneous to her meaning for look. By listening to a sounding object when instructed to look at it, Bryan demonstrated that the inherent qualities of the object referred to in the command could successfully distract his attention from the literal meaning of the request given him. In an analogous fashion, Kelli's responses to Look with your mouth/nose, could have been confounded by the competitive forces of the inherent abilities of the body parts through which the looking was to occur. In summary, the response obtained to the supplemental command illustrated the logic by which Kelli's sniffing and tasting behaviors could be explained by something other than an amodal interpretation of look. In conclusion, Bryan's responses to the commands for Experiment Three were interpreted as additional evidence to indicate that his meaning for look is restricted to haptic exploration. In effect, the deductions made in the first two experiments were further supported. EXPERIMENT FOUR Objective of Experiment In the fourth experiment Landau and Gleitman further investigated the possibility that, for Kelli, look implied perception through nonspecific sensory modalities. Their objective was to fully ascertain the child's meaning for the perceptual verb as it applied to herself. 62 Landau and Gleitman's Procedure In this experiment, look commands were presented within contexts that either challenged the ease with which haptic exploration of objects could take place, or in contexts that made available auditory perceptual information, as an attempt to bias the modality through which Kelli would respond. Landau and Gleitman rationalized that if modality nonspecificity did characterize Kelli's meaning for look, then supporting evidence would be most apt to surface in the conditions under which some of the commands would be presented in this experiment. Twelve commands comprised the stimuli for Experiment Four. Through each stimulus, Kelli was instructed to look at an object that was capable of producing noise (e.g. a radio, or a squeaking toy fish, p.65). Four of the commands were presented under each of three conditions involving varying available perceptual information. In the first condition, identified as the tactile condition, tactile information was immediately available. Simultaneous to the delivery of a command, Kelli's hand was placed on the object referred to in the command. For two stimulus presentations the object was situated in front of the child, while for the other two presentations, the object was located behind her. The tactile condition was thought to be more conducive to haptic exploration than the experimental conditions which follow. The second condition, identified as the aural condition, involved the immediate availability of aural perceptual information. Simultaneous to a command presentation, the object referred to in the command was sounded. In contrast to the previously described condition, the aural condition did not make available tactile information. For the delivery of two commands, the sounding object was situated far from Kelli, i.e. 8 feet away, while for the remaining two stimuli, the object was located within arm's reach. The aural context was intended to influence Kelli to recruit the auditory perceptual modality when responding to the test commands. In the third experimental condition, identified as the visual condition, Kelli was provided with neither tactile nor auditory information simultaneous with the delivery of a command. For two trials, the object referred to in a command was not sounding and was held far from the child, while for the other two commands, the object was also silent but was situated near her. This condition required that Kelli choose between the perceptual modalities that she would draw on in order to fulfill the instruction to look. Kelli was age 3;1 at the time of experimentation. The number of play sessions required for the delivery of all test stimuli was not reported. Critique Landau and Gleitman's description of procedures used for Experiment Four were in keeping with what was reported for all previous experiments. That is to say that little detail was provided with respect to the situational and verbal contexts in which commands were delivered. Therefore, concerns raised in the Critique sections of the first three experiments regarding the inadvertent use of response-biasing procedures are equally relevant to this experiment. Beyond these general remarks, there are no issues to raise with respect to the procedures employed for the fourth experiment. Procedure Used in the Present Study The replication of Experiment Four was conducted across two play sessions. Bryan was age 3;3 at the time of the first play session and 3;4 at the time of the second session. All twelve commands were presented under the three experimental conditions described by Landau and Gleitman, such that four commands were delivered under each condition. The same set of objects was referred to in the commands presented under all three experimental conditions. Those objects were: the buzzer used in Experiment Three; the alarm-clock also used in Experiment Three; an elephant (stuffed animal which when wound up, played music); and a birdie (stuffed animal which when shaken, produced a whistling sound). Bryan was familiar with all of these objects prior to the conduction of the experiment. The situational and verbal contexts employed for the presentation of commands in Experiment Four were similar to what were used for the delivery of stimuli in all previous experiments. In effect, commands were never given at times when Bryan was already in the process of seeking, exploring or aurally focused on an object. Whether or not stimulus-response pairs would be accepted as analyzable was determined by the criteria outlined for this purpose in Experiment One. Landau and Gleitman's Results and Conclusions Before stating their results for Experiment Four, Landau and Gleitman presented a line of reasoning according to which they were prepared to interpret their findings. They explained, If look means 'perceive by hand,' then Kelli should find a way to explore the named object by hand under all (the experimental) conditions. If look means 'perceive using any modality,' Kelli still might seek to explore the objects manually under all conditions, but it is likely that in certain conditions, such as hearing a radio from afar, she might assume she is already looking and therefore need do nothing beyond quieting and orienting to the sound, (p.66) Given these implications that would be associated with particular patterns of responding, Landau and Gleitman reported that for the entire set of commands, irrespective of the experimental condition, Kelli exhibited a single type of behavior. At all times, when instructed to look the child retrieved the object referred to, and proceeded to explore it by hand. 65 Landau and Gleitman stated that the behaviors observed in the fourth experiment, served as compelling evidence to support the conclusion that, for Kelli, look could not be actualized by amodal means. Rather, this term meant explore or apprehend manually, i.e. through the haptic perceptual modality. Still to be accounted for were the behaviors elicited by the commands Look with your mouth/nose in Experiment Two. Landau and Gleitman suggested that Kelli's responses to these commands constituted rare events. As such, the tasting and sniffing behaviors were not to be explained by a meaning for look which implicated perception through any plausible perceptual modality. Results and Conclusions of the Present Study Bryan's responses to the test stimuli, presented under the three experimental conditions were as follows: STIMULUS COMMAND RESPONSE (Tactile Condition) 1. Look at the elephant, (in front) Bryan took the elephant into both hands, and immediately sought out the dial that he could wind, in order to cause the music to be emitted from the elephant. After winding the dial, he proceeded to sit and listen to the music. 2. Look at the alarm clock, (in front) Bryan took the clock into both hands, and felt much of the surface of the object with both hands. 3. Look at the buzzer, (behind) Bryan picked up the buzzer, depressed the button to cause the object to sound, then held the sounding object against his ear. 4. Look at the buzzer, (behind) Bryan picked up the buzzer, sought out the button on its side, and depressed the button. He then sat and listened to the "buzzing" sound. 66 ( A u r a l cond i t i on ) 5. L o o k at the buzzer , ( fa r ) B r y a n re t r ieved the buzzer f r o m the exper imen te r who was h o l d i n g i t , and con t i nued to cause the object to sound. 6. L o o k at the b i rd ie , ( fa r ) B r y a n re t r ieved the b i r d i e f r o m the exper imen te r who was h o l d i n g i t , then con t i nued to shake i t so that i t w o u l d sound. 7. L o o k at the buzzer , (near) B r y a n took the buzzer in to h is hands depressed the but ton to cause the " b u z z i n g " sound to con t inue . 8. L o o k at the e lephant , (near) B r y a n took the e lephant in to h is hands and fe l t m u c h of the sur face of the object. ( V i s u a l cond i t i on ) 9. L o o k at the e lephant , (near) B r y a n took the e lephant in to both hands , and fe l t m u c h of the sur face o f the object. 10. L o o k at the a l a r m c l ock , (near) B r y a n took the c lock in to h is hands and fe l t m u c h o f the sur face o f the object. 11. L o o k at the buzzer , ( fa r ) B r y a n re t r ieved the buzzer , then con t i nued to cause the object to sound. 12. L o o k at the b i rd ie , ( fa r ) B r y a n re t r i eved the b i r d i e f r o m the exper imen te r who was h o l d i n g i t , then shook the object so that i t w o u l d make its squeak ing sound. T w o genera l response-types ex is ted among a l l the behav io rs w h i c h the commands i n E x p e r i m e n t F o u r e l i c i t ed f r o m B r y a n . In one ins tance , he app rehended the object and proceeded to exp lore i t , wh i l e i n the other ins tance , he app rehended the object and caused the object to p roduce its noise. T h e d i s t r i b u t i o n of B r y a n ' s response-types across expe r imen ta l cond i t i ons is s u m m a r i z e d i n T a b l e II be low. 67 TABLE II Bryan's Response-Types for Each Experimental Condition  STIMULUS COMMAND RESPONSE TYPE  A P P R E H E N D & E X P L O R E A P P R E H E N D & L ISTEN ( T A C T I L E CONDITION) 1. L o o k at the e lephant . 2. L o o k at the a l a r m c lock 3. L o o k at the buzzer . 4. L o o k at the buzzer . ( A U R A L CONDITION) 5. L o o k at the buzzer . 6. L o o k at the b i r d i e . 7. L o o k at the buzzer . 8. L o o k at the e lephant . (V ISUAL CONDITION) 9. L o o k at the e lephant . 10. L o o k at the a l a r m c lock . 11. L o o k at the buzzer . 12. L o o k at the b i r d  + + + + + + + + + + + " + B r y a n ' s pa t te rn o f respond ing d i d not share the order l iness w h i c h cha rac te r i zed the set o f responses e l i c i t ed f r o m K e l l i i n the f o u r t h exper iment . Whereas K e l l i cons is tent ly app rehended a n d exp lo red the object re fe r red to i n a c o m m a n d , B r y a n responded i n this f a s h i o n to on l y f o u r o f the twe lve s t i m u l i . When a n a l y z i n g B r y a n ' s response pa t te rn , i t was in te res t ing to note that no cor re la t ion ex is ted between p a r t i c u l a r response behav io rs and the expe r imen ta l cond i t i ons . C o n v e r s e l y , each o f the two response-types were ra ther even ly d i s t r i bu ted across the three cond i t i ons . T h e imp l i ca t i ons of the ob ta ined f i n d i n g s are somewhat comp l i ca ted s ince B r y a n ' s pa t te rn o f respond ing d i d not c o n f o r m w i t h e i ther o f the postu la t ions w h i c h L a n d a u and G l e i t m a n made. T o re i terate, these authors hypo thes ized that i f a b l i n d c h i l d thought that looking c o u l d be ca r r i ed out on l y th rough the hap t i c percep tua l m o d a l i t y , then such a c h i l d shou ld respond to a l l commands i n the f o u r t h exper imen t by a p p r e h e n d i n g and e x p l o r i n g the target objects. If, on the other h a n d , looking cou l d occur t h rough e i ther the hap t i c or a u r a l percep tua l moda l i t i es , then listening behavior m ight be 68 associated w i t h any of the s t i m u l i presented i n the exper imen t , but w o u l d be most apt to occur under the a u r a l c o n d i t i o n . T h e ove ra l l h i gh i nc i dence o f B r y a n ' s listening responses to the look commands i n i t i a l l y suggested that, f o r h i m , a u r a l behav io r cou ld be pa i red w i t h the ve rb look. Y e t , i n l igh t o f this i m p l i c a t i o n , the i n s i g n i f i c a n t co r re la t i on between his a u r a l behav io rs and the au ra l exper imen ta l c o n d i t i o n , was ra ther cu r ious . B y re la t i ng B r y a n ' s response-types to the objects r e fe r red to i n the test s t i m u l i , i t can be observed that whenever i ns t ruc ted to look at the buzzer or birdie, regardless of the c o n d i t i o n under w h i c h a command was presented, B r y a n a lways caused the target object to sound . M o r e o v e r , whenever ins t ruc ted to look at the alarm clock, B r y a n a lways apprehended and h a p t i c a l l y exp lo red the object. A n d las t l y , when to ld to look at the elephant, e i ther response-type wou ld occur . V i e w i n g B r y a n ' s response-types f r o m the perspect ive o f the object re fe r red to i n a c o m m a n d , reveals some s t r i k i n g cor re la t ions . Based on those cor re la t ions , it is proposed that the listening behaviors observed i n the present expe r imen t were a f u n c t i o n of the inheren t a t t r ibu tes of the objects to w h i c h B r y a n ' s a t ten t ion was d i rec ted th rough the s t imu lus commands. Stated o therw ise , i t is thought that the demand charac te r is t i cs o f the no i se -mak ing objects served as compet i t i ve fo rces i n d e t e r m i n i n g how B r y a n w o u l d respond when ins t ruc ted to look at these objects. Whether or not the in te res t ing a t t r ibutes o f an object , w o u l d d is t rac t a ch i l d ' s a t ten t ion f r o m the w o r d i n g of a command re la ted to that object , shou ld depend on the d i spos i t i on of the c h i l d hea r ing the command . Wh i le se rv ing as the expe r imen ta l subject i n th is r ep l i ca t i on s tudy , B r y a n o f ten presented as a c h i l d w i t h h is o w n agenda, i.e. a c h i l d who was less "eager to please" than K e l l i m igh t have been. Consequen t l y , he may have been more i n c l i n e d than K e l l i to respond to the inheren t charac te r i s t i cs o f an object i n h is presence, than to a ve rba l i ns t ruc t i on g i ven i n re la t i on to that object . Wh i l e v i e w i n g B r y a n ' s listening responses i n th is exper imen t as h a v i n g been de te rm ined by the a u r a l qua l i t i es o f the objects to w h i c h he was asked to a t tend, I c o n c l u d e d , once a g a i n , that h is mean ing f o r look was to apprehend and explore through the haptic perceptual modality. E X P E R I M E N T A T I O N W I T H T H E C O N T R O L S U B J E C T S A t the t ime o f test ing, the age-matched, b l i n d f o l d e d s ighted con t ro l subjects were s i m i l a r to K e l l i i n that they were bar red f r o m v i s i on . Y e t , th roughou t tes t ing, these c h i l d r e n d i f f e r e d f r o m K e l l i in terms of the i r h i s t o r i c a l i n d u c t i v e bases f o r the ass ign ing o f mean ing to words. Whereas the cont ro ls had p rev ious l y l ea rned the verb look i n a v i sua l w o r l d , K e l l i had not. L a n d a u and G l e i t m a n cons ide red th is d i f f e r e n c e i n h i s to r i ca l i n d u c t i v e bases to be the on l y re levan t v a r i a b l e d i s t i n g u i s h i n g K e l l i f r o m the contro ls . Consequen t l y , any observed d i f f e r e n c e s i n the ch i l d ren ' s respect ive meanings f o r look w o u l d be a t t r i bu tab le to h i s t o r i c a l l y d i f f e r e n t i n d u c t i v e bases. Conve rse l y , s im i l a r i t i es i n mean ings w o u l d be usable as ev idence to demonstrate the cons t ra ined role o f ex t r a l i ngu i s t i c exper ience i n the ass ign ing o f mean ing to words. E X P E R I M E N T O N E L a n d a u and G l e i t m a n ' s P rocedure and Resul ts F o r the f i r s t exper imen t , L a n d a u and G l e i t m a n presented the con t ro l subjects w i t h the commands Look up, Look down, Look behind you, Look in front of you, Look over here by me, and Look over there by mommy. These are the same commands de l i ve red to K e l l i f o r the same exper iment . U p o n hea r i ng each o f the s ix commands , a l l con t ro l subjects d i s p l a y e d a s ing le response-type w i t h 100% cons is tency. Tha t response-type i n v o l v e d s imp le 70 orientation of the head in the named spatial direction. For example, when instructed to Look up, each child tilted her head back (p.56). Based on this pattern of responding, Landau and Gleitman made a preliminary conclusion that for the sighted blindfolded children, look meant 'turn one's eyes toward'' (p.56). Procedure and Results of the Present Study Commands for the first experiment were presented to Gregory, while he was blindfolded, across three play sessions. Gregory was age 3;2, 3;3, and 3;4 at the time of each successive play session. The situational and verbal contexts used for experimentation were very similar to contexts employed during testing with Bryan. It was reported in the Method Overview section that Gregory was initially separated from Bryan while Bryan was being presented with the look commands. To reiterate, the intent of the separation was to preserve Gregory's integrity as a control subject, whose conduct within testing would not have been influenced by how he had seen his brother respond to the experimental instructions to look. Following the initial separation, experimentation with Gregory was commenced. In spite of the separation, this blindfolded child's responses to the test commands indicated that his interpretation of look must previously have been influenced by all that he had ever seen Bryan do as he naturally attempted to perceive his environment. Therefore, it was decided that anything Gregory would witness during continued experimentation with Bryan could have no greater influence on his own response behaviors, than had the fact that he had spent the entirety of his three years of life with a blind twin. Consequently, Gregory may or may not have previously observed his brother's responses to any of the experimental commands with which he was being presented. It was found that no obvious correlation existed between Gregory's responses to commands, and whether or not he had previously observed his brother's responses to the same commands. Therefore, even 71 though Gregory was present during segments of experimentation with Bryan, it was confidently felt that his integrity as a control subject in the replication study had not been altered as a result. Gregory's responses to the stimuli for experiment one were as follows: STIMULUS COMMAND RESPONSE 1. L o o k up. Gregory oriented his face towards the ceiling while his hands remained still. 2. L o o k down . 3. L o o k beh ind you . 4. L o o k in f ron t of you . 5. L o o k over here by me. Gregory felt the floor in front of himself with his right hand, all the while keeping his head in a stationary position. Gregory turned around and oriented his face towards the floor behind himself while doing nothing with his hands. Gregory extended both hands out along the floor in front of himself while doing nothing with his head. Gregory extended his right hand towards me and touched my foot, but did nothing with his head. 6. L o o k over there by mommy. Gregory extended his left hand in the direction o f where his mother was sitting, and contacted her. His head remained in a stationary position all the while. In summary, Gregory responded to four of the six commands by orienting his hands in the named spatial direction. For the two remaining commands he oriented his head in the specified direction. Gregory's response tendency was markedly different from that of the sighted blindfolded children in Landau and Gleitman's study. Conversely, his behaviors greatly resembled those of his brother. Based on Gregory's responses to the stimuli for Experiment One, I concluded that when barred from visual input, this child's meaning for look involved orientation of 72 either the hands or eyes towards a target. In effect, his interpretation of look implicated either haptic or visual perception. EXPERIMENT TWO Landau and Gleitman's Procedure and Results For the second experiment, Landau and Gleitman presented their control subjects with 14 of the 26 commands that were delivered to Kelli in Experiment Two. Those commands were: Look/Touch up; behind you; with your foot; with your mouth; with your finger; with your nose; with your ear. It was reported that all commands were delivered to each child in a random order, within a single session. The analysis of responses obtained from the control subjects involved a procedure that was somewhat similar to that employed for the analysis of responses elicited from Kelli. Initially, the behaviors which each child associated with both the look and touch versions of a given command-type were coded as either differentiated or nondifferentiatcd. Next, all responses were assigned to one of six categories. Those categories were: MANUAL 1) exploratory 2) nonexploratory (contact) 3) orients hands only VISUAL 4) orients head only 5) orients head, plus hand movement 6) makes analogy to vision Landau and Gleitman found that on average, the sighted children differentiated their responses to 4.8 of the 7 command-types. Of those 4.8 command-types, on average 3.5 of the touch commands were paired with nonexploratory manual contact of the target object, whereas 2.5 of the look commands were paired with head-orienting behavior towards the target object. An additional 1.2 of the look commands were associated with an "analogy to vision" 73 response- type, w h e r e i n the c h i l d behaved as though the body par t named i n a c o m m a n d c o u l d see. F o r examp le , in response to Look with your foot, the c h i l d w o u l d b r i n g her foot close to but not touch ing the target object as i f her foot had eyes (p.62). Based on the pat tern o f responses observed in this exper imen t , L a n d a u and G l e i t m a n c o n c l u d e d that f o r the i r con t ro l subjects, look was a v i s u a l ve rb , whereas touch was a tac tua l ve rb mean ing contact. Procedure and Resu l ts of the Present S tudy C o m m a n d s f o r the second exper iment were presented to G r e g o r y , w h i l e he was b l i n d f o l d e d , across two p lay sessions. He was age 3;3 d u r i n g the f i r s t p l ay session and 3;4 at the t ime o f the second session. H i s responses to the commands were as fo l l ows : S T I M U L U S C O M M A N D R E S P O N S E 1. L o o k up. 2. T o u c h up. 3. L o o k beh ind you . 4. T o u c h beh ind you . (Taken f r o m exper iment one). G r e g o r y o r ien ted h is face towards the ce i l i ng . G r e g o r y ra ised his r igh t hand up i n the a i r and touched the top of his head. (Taken f r o m exper iment one). G r e g o r y tu rned a r o u n d and or ien ted his face towards the f l oo r b e h i n d h imse l f . G r e g o r y touched the f l oo r beh ind h imse l f w i t h both hands. H i s head remained f a c i n g f o r w a r d . 5. L o o k at the monkey w i t h your foot . G r e g o r y p laced the monkey in f r on t o f where he was s i t t i ng , and contacted it w i t h h is r igh t foo t f o r app rox ima te l y 3 seconds. 6. T o u c h the monkey w i t h your foot . G r e g o r y p laced the monkey under h is foot , and lowered his foot to contact the monkey . 7 4 7. Look at the elephant with your finger. Gregory pressed his left forefinger against the surface of the elephant. 8. Touch the apple with your finger. Gregory contacted the apple with his right index finger for approximately 2 seconds. 9. Look at the bunny with your mouth. Gregory raised the bunny up to his mouth, and pressed it against his lips for approximately 2 seconds. 10. Touch the bunny with your mouth. Gregory pressed the bunny against his lips for approximately 2 seconds. 11. Look at the apple with your nose. Gregory held the apple still against his nose for approximately 3 seconds. 12. Touch the apple with your nose. Gregory raised the apple up and contacted his nose with it, for approximately 2 seconds. 13. Look at Ogopogo with your ear. Gregory raised the object up and held it against his right car for approximately one second. 14. Touch the sail with your ear. Gregory raised the object up and held it against his left ear for approximately 3 seconds. Gregory's responses to the look and touch versions of each command-type were coded as either differentiated or nondifferentiated. Coding indicated that Gregory responded differentially according to the verb in the command for only two of the seven command-types. Those command-types were, Look/Touch up and behind you. For the look versions of these command-types, Gregory oriented his eyes in the named direction, whereas for the touch versions he made manual contact with anything he found situated in the specified direction. In response to all other stimuli, Gregory simply contacted the target object with the body part specified in the command to do the looking and touching. Gregory's responses to the touch commands in the present experiment were similar to the behaviors elicited by the same commands from the control subjects 75 in Landau and Gleitman's study. On the other hand, Gregory and the other control subjects varied greatly in terms of the behaviors they each associated with the instructions to look. Never did Gregory pretend that the body part named in a command had eyes. In effect, the "analogy to vision" response-type, common among Landau and Gleitman's controls, was never observed with Gregory. Moreover, while the other children rarely associated manual contact with the verb look, Gregory did so for five of his seven responses to the look commands. While contrastive with the other sighted children in terms of the behaviors he paired with look, Gregory was both similar to, and different from his brother in the same regard. In Experiment Two, when Bryan differentiated his responses for a command-type, he tended to associate looking with exploration through the haptic perceptual modality. Although exploratory behavior did not characterize Gregory's responses to the look commands in this experiment, perception through the haptic modality was observable. The results obtained from Gregory in the second experiment confirmed the suspicion that, when blindfolded, this three year old, sighted child had a meaning for look which reflected the fact that hc had continuously shared the company of a blind sibling. Unlike typical sighted three year olds (e.g. those in Landau and Gleitman's study), Gregory had a default perceptual modality which could be drawn on, when his visual modality was barred from use. It seems reasonable that his early awareness of the default modality would have arose from his constant observation of Bryan's dominant means of perceiving. As a result, the haptic component of Gregory's responses to the look commands might easily be accounted for. However, it is less clear why exploration was not also a component of Gregory's response-type as it was for his brother in this experiment. It is possible that because Gregory was previously familiar with, and had a visual image of the objects referred to in the commands, hc may have felt no need to explore them 76 f u r t he r . T h e a r t i f i c i a l context i n w h i c h he was commanded to look may i n i t se l f have p rec luded a na tu ra l mo t i va t i on to exp lore . In any event , G rego ry ' s response behav io rs suppor ted the conc lus ion that, when bar red f r o m v i s i o n , h is mean ing f o r the pe rcep tua l ve rb look i nvo l ves the rec ru i tment o f e i ther the v i s u a l or hap t i c pe rcep tua l moda l i t i es . E X P E R I M E N T F I V E L a n d a u and G l e i t m a n ' s P rocedure and Resu l ts In the p rev ious two exper iments , L a n d a u and G l e i t m a n observed that the i r s igh ted , b l i n d f o l d e d c h i l d r e n associated on ly v i sua l gestures w i t h the ve rb look. In the f i f t h exper imen t , these researchers quer ied whether this assoc ia t ion mere l y re f l ec ted a b ias i n the c h i l d r e n , "w i thout in p r i n c i p l e d e n y i n g that they c o u l d unde rs tand look to be in te rp re tab lc as explore haptically " (p.67). In an at tempt to de te rm ine i f the ch i l d ren ' s mean ing fo r look cou ld be b i m o d a l , L a n d a u a n d G l e i t m a n presented the i r subjects w i t h commands th rough w h i c h they were e x p l i c i t l y i ns t ruc ted to look w i t h the i r hands. M o d i f i e d vers ions o f the commands presented i n the f i r s t exper imen t were used as s t i m u l i f o r E x p e r i m e n t F i v e . S p e c i f i c a l l y , the c h i l d r e n were g i ven the ins t ruc t ions : Look up with your hands; Look down with your hands; Look behind you with your hands; Look in front of you with your hands; Look over here by me with your hands; and Look over there by mommy with your hands. L a n d a u and G l e i t m a n repor ted that the order i n w h i c h commands were de l i ve red f o r E x p e r i m e n t F i v e was consis tent w i t h the presentat ion order used i n the f i r s t exper iment . L a n d a u and G l e i t m a n f o u n d that the s ighted b l i n d f o l d e d c h i l d r e n became c o n f u s e d by the ins t ruc t ions to look w i th the i r hands. One t h i r d o f the tota l set o f ob ta ined responses, consisted o f the same behav iors w h i c h were e l i c i t ed by the look 7 7 commands de l i ve red i n the f i r s t exper iment . In other words , the c h i l d r e n o r ien ted the i r heads i n the named spat ia l d i r ec t i on wh i l e do ing no th ing w i t h the i r hands. F o r the r e m a i n i n g two th i rds o f the total set of responses, the c h i l d r e n "or ien ted the i r heads t owa rd the target object , and then t r i ed to do someth ing w i t h the i r hands as w e l l " (p.67). Bu t the behav io r i n v o l v i n g the i r hands was not at a l l o f an exp lo ra to ry or app rehend ing nature. R a t h e r it tended to be qu i te i r re levan t . F o r examp le , a t y p i c a l response to Look up with your hands, "was to t i l t the head u p w a r d ... and s imu l taneous ly to pos i t ion the hands near the face , pa lms u p w a r d " (p.67). N e v e r d i d the c h i l d r e n i nd i ca te that they thought the i r "hands cou ld serve as the exp lo ra to ry organs f o r l o o k i n g " (p.67). S ince L a n d a u and G l e i t m a n ' s s ighted b l i n d f o l d e d subjects cou ld not be i n d u c e d to associate the hap t i c moda l i t y w i t h the perceptua l verb look, i t was dec ided that f o r these young c h i l d r e n , look was t ied exc lus i ve l y to the v i sua l moda l i t y . P rocedu re and Resu l ts of the Present Expe r imen t C o n s i d e r i n g that the ob jec t ive of E x p e r i m e n t F i v e was to test the poss ib i l i t y that a s ighted c h i l d cou ld a r r i ve at a b imoda l i n te rp re ta t ion o f look, c o n d u c t i n g th is exper imen t w i t h G r e g o r y was redundan t . G r e g o r y had a l ready demonst ra ted his b i m o d a l i n te rp re ta t i on o f the perceptua l verb. Never the less , the exper imen t was conduc ted . T h e s ix look commands were presented to G r e g o r y , w h i l e he was b l i n d f o l d e d , across two p lay sessions. He was age 3;3 and 3;4 at the t ime o f each successive t ap ing session. T h e responses e l i c i t ed by the s t imu l i f o r E x p e r i m e n t F i v e were as fo l l ows : S T I M U L U S C O M M A N D R E S P O N S E 1. L o o k up w i t h your hands . G r e g o r y extended both hands up in the a i r above h is head , and waved them th rough the a i r as though exp lo r i ng the space above h i m . 78 2. L o o k down w i t h your hands . G r e g o r y extended both hands towards the f l oo r i n f r on t o f h imse l f , and s l i d them a long the f l oo r . 3. L o o k beh ind you w i t h your hands. G r e g o r y reached beh ind h imse l f w i t h both hands , and fe l t h is back. 4. L o o k i n f r o n t of you w i t h your hands. G r e g o r y extended his hands i n the a i r i n f r o n t o f h imse l f , d r a w i n g his hands together and apar t as though exp lo r i ng the space in f r on t o f h i m . 5. L o o k over here by me w i t h your hands. G r e g o r y extended his le f t hand towards me a n d grasped my hand . 6. L o o k over there by mommy w i th your hands. G r e g o r y extended both hands out towards h is mother , and reached in her d i r ec t i on un t i l he contac ted her w i t h his hands. In s u m m a r y , G r e g o r y responded to a l l of the commands by o r i en t i ng h is hands i n the named d i r ec t i on . F o r Look behind you, no exp lo ra to ry behav io r was i nco rpo ra ted in to the total response. H o w e v e r , fo r a l l other commands , G r e g o r y f o l l o w e d up his hand o r i en t i ng behav io r w i t h exp lo ra t i on o f the space associated w i t h the d i r e c t i o n named i n a command . A s i n E x p e r i m e n t s One and T w o , Grego ry ' s responses to the s t i m u l i f o r the f i f t h exper imen t set h i m apart f r o m the cont ro l subjects in L a n d a u and G l e i t m a n ' s s tudy. Whereas the other s ighted b l i n d f o l d e d c h i l d r e n cou ld not be i n d u c e d to exerc ise the hap t i c perceptua l moda l i t y when respond ing to ins t ruc t ions to look, G r e g o r y cou ld . F u r t h e r m o r e , G rego ry ' s responses i n E x p e r i m e n t F i v e were d i s t i nc t f r o m what he had been observed to do in the f i rs t exper iment when the i m p l i c a t u r e o f the commands was also to look for, his in te rp re ta t i on o f look i n E x p e r i m e n t F i v e i n v o l v e d a strong exp lo ra to ry component , perhaps because he expected that some u n k n o w n object wou ld be s i tuated i n the spa t ia l d i rec t ions i n w h i c h he was ins t ruc ted to look. 7 9 O n the basis of a l l da ta gathered in the inves t iga t ion of G r e g o r y ' s mean ing f o r look, I c o n c l u d e d that th is s ighted c h i l d in te rpre ted the percep tua l verb to mean exploration through the haptic modality when the v i sua l m o d a l i t y is ba r red f r o m use. H i s b i m o d a l i n te rp re ta t i on of the verb is be l ieved to be d i r ec t l y l i n k e d to the fac t that p r i o r to expe r imen ta t i on , he had been cons tan t ly exposed to someone else w h o in te rp re ted the w o r d in a nonconven t i ona l , yet e f f ec t i ve manner . B r y a n ' s p redom inan t means o f pe rce i v i ng the w o r l d had c lea r l y i n f l u e n c e d how his b ro ther thought he h imse l f cou ld also perce ive and exp lore . 80 C H A P T E R T H R E E D I S C U S S I O N S U M M A R Y O F R E S U L T S F O R T H E LOOK E X P E R I M E N T S  R E P L I C A T E D W I T H A B L I N D C H I L D E X P E R I M E N T O N E : In response to f i v e c o m m a n d s 2 to look i n va r i ous spa t ia l loca t ions (where in the i m p l i c a t u r e was to look for versus look at), B r y a n o r ien ted his hands or feet i n the appropr ia te d i r ec t i on w h i l e a t tempt ing to locate a target object . U p o n con tac t i ng the object , he apprehended it w i t h h is hands. These responses p rompted the conc lus ion that , f o r B r y a n , when the i m p l i c a t u r e is to look for, look means seek through the haptic perceptual modality in order to apprehend with the hands. E X P E R I M E N T T W O : In response to seven out o f twe lve commands to look at objects i n va r i ous spa t ia l d i rec t ions , i n pa r t i cu l a r manners , or w i t h va r ious body par ts , B r y a n demonst ra ted exp lo ra to ry behav io r . In response to the same seven command- types c o n t a i n i n g the verb touch, B r y a n demonst ra ted nonexp lo ra to ry m a n u a l behav io rs . F o u r o f the f i v e r e m a i n i n g pa i rs o f look and touch commands e l i c i t e d n o n d i f f e r e n t i a t e d , nonexp lo ra to ry behav io rs . Responses to the last pa i r o f look a n d touch commands cou ld not be de te rm ined . Based on B r y a n ' s pat tern o f respond ing i n th is exper imen t , I conc luded that h is meanings f o r look and touch are d is t inc t . F o r h i m , look (i.e. look at) means explore through the haptic perceptual modality, w h i l e touch means make physical contact which is short in duration, and involves a restricted surface area of the contacted object. 2 Recall that the sixth look command in this experiment was excluded from analysis. 81 E X P E R I M E N T T H R E E : In response to three commands to taste va r ious objects, B r y a n contac ted the objects w i t h h is l ips or tongue. Th ree commands to touch va r ious objects e l i c i t ed nonexp lo ra to ry m a n u a l contact . In response to three commands to listen to objects that were e i ther sound ing or capab le o f s o u n d i n g , B r y a n demonst ra ted l i s ten ing behav io r . When ins t ruc ted to listen to an object w h i c h cou ld not emi t sound , B r y a n rep l i ed , "I can' t . " F i n a l l y , w h e n commanded to look at an object w h i c h d i d not p roduce sound , B r y a n exp lo red i t by h a n d . O n the other h a n d , w h e n ins t ruc ted to look at objects that were e i ther sound ing or capab le o f em i t t i ng sound , B r y a n demonst ra ted l i s ten ing behav io r . T a k e n as a who le , th is set o f responses p rompted the conc lus ion that percept ions th rough nonhap t i c moda l i t i es (e.g. l i s ten ing and tast ing) have the i r o w n l e x i c a l encod ings f o r B r y a n . F u r t h e r m o r e , I c o n c l u d e d that h is mean ing f o r look does not i n v o l v e a m o d a l pe rcep t ion . A l t h o u g h B r y a n l is tened i n response to some of the look commands , I j udged those responses to be re f l ec t i ve o f the demand charac te r i s t i cs o f the objects r e f e r r ed to i n the commands. I d i d not cons ider the l i s ten ing behav io rs to be par t of B r y a n ' s mean ing f o r look. E X P E R I M E N T F O U R : B r y a n responded w i t h m a n u a l exp lo ra t i on to one o f f o u r commands to look at objects w h i c h were capab le o f em i t t i ng sound , and w h i c h he was a l ready con tac t ing . F o r the other three commands presented under the same c i r cumstances , B r y a n responded w i t h l i s ten ing behav io r . A n i d e n t i c a l response pat te rn was e l i c i t ed by f o u r commands to look at objects w h i c h were sound ing , and w h i c h B r y a n was not a l ready con tac t ing . In response to two o f f o u r commands to look at objects w h i c h were capab le o f emi t t i ng sound but were s i len t , and w h i c h were s i tua ted f a r f r o m h i m , B r y a n apprehended the objects and exp lo red them m a n u a l l y . In response to the other two commands de l i ve red under these cond i t i ons , B r y a n app rehended the objects, caused them to sound , and l is tened to them. T h i s 82 pat te rn o f respond ing c o n f i r m e d the p rev ious l y stated conc lus ion about the u n i m o d a l i t y of B r y a n ' s mean ing fo r look as i t app l ies to h imse l f . Once a g a i n , I cons ide red a l l o f the l i s ten ing responses w h i c h he pa i red w i t h some of the ins t ruc t i ons to look to be a f u n c t i o n of the demand charac te r i s t i cs o f the objects r e f e r r ed to i n the commands. T a k e n together, the results ob ta ined f r o m a l l f o u r exper iments suppor t the conc lus ions that , f o r B r y a n , to look at an object is to explore it by hand; and to look for an object is to seek it through the haptic perceptual modality in order to apprehend it with the hands. SUMMARY OF RESULTS FOR T H E LOOK EXPERIMENTS  REPLICATED WITH A BLINDFOLDED SIGHTED CHILD EXPERIMENT ONE: In response to f o u r o f s i x commands to look i n va r i ous spa t ia l loca t ions (where in the imp l i ca tu re o f the c o m m a n d was to look for versus look at), G r e g o r y o r i en ted his hands i n the spec i f i ed d i r ec t i on . In response to the r e m a i n i n g two commands , he or ien ted his head i n the named d i r e c t i o n . Based on th is set o f responses, I conc luded that , i n the absence of v i s u a l i npu t , G r e g o r y ' s mean ing f o r look is to orient the eyes or hands towards a target. EXPERIMENT TWO: G r e g o r y exh ib i t ed n o n d i f f e r e n t i a t e d responses to f i v e out of seven look and touch c o m m a n d pa i rs . T h e response behav io r e l i c i t ed by the each of the f i v e c o m m a n d pa i rs was s imp le contact o f an object w i t h the s p e c i f i e d body part . O n the two occasions w h e n he d i s t i ngu i shed h is look a n d touch responses, looking i n v o l v e d o r ien ta t i on o f the eyes in the named d i r e c t i o n , w h i l e touching i n v o l v e d m a n u a l contact w i t h a n y t h i n g s i tua ted i n the spec i f i ed loca t ion . Resu l t s ob ta ined i n this exper imen t f u r t he r suppor ted the conc lus ion that , i n the absence 83 of v i s u a l i npu t , G rego ry ' s mean ing fo r look can i n v o l v e e i ther the hap t i c or the v i s u a l pe rcep tua l moda l i t i es . E X P E R I M E N T F I V E : In response to s ix commands to look i n va r ious spa t ia l loca t ions w i t h h is hands (where in the i m p l i c a t u r e was to look for), G r e g o r y o r ien ted his hands i n the named d i r ec t i on . F o r f i v e o f the s ix commands , he f o l l o w e d up the hand -o r i en t i ng behav io r w i t h m a n u a l exp lo ra t i on o f the re levant l oca t i on . T h e s i x t h response d i d not i n c l u d e exp lo ra to ry behav io r . These responses f u r t h e r demonst ra ted that G rego ry ' s mean ing fo r look can i n v o l v e the hap t i c pe rcep tua l moda l i t y . T h e responses ob ta ined f r o m G r e g o r y i n a l l three exper iments co l l e c t i ve l y suppor ted the conc lus ion that , when bar red f r o m v i sua l i npu t , th is c h i l d ' s mean ing f o r look can i n v o l v e e i ther the v i sua l or hap t i c percep tua l moda l i t i es . F u r t h e r m o r e , i f the i m p l i c a t u r e o f the c o m m a n d is to look for, then hap t i c pe rcep t ion becomes a m e d i u m fo r exp lo ra to ry behav io r . T H E G E N E R A L I Z A B I L I T Y O F F I N D I N G S O F  T H E O R I G I N A L LOOK E X P E R I M E N T S T h e resul ts and conc lus ions w h i c h emerged f r o m the look exper iments conduc ted w i t h both B r y a n and K e l l i are genera l l y consistent w i t h each other. H o w e v e r , f i n d i n g s d i d v a r y i n that K e l l i ' s hap t i c behav io rs engaged her hands on l y , whereas B r y a n ' s hap t i c behav io rs somet imes engaged a d d i t i o n a l body parts. V a r i a n c e i n f i n d i n g s may also have ex is ted w i t h respect to the nature o f the hap t i c behav io rs w h i c h the commands e l i c i t ed f r o m each c h i l d . In the second exper imen t , L a n d a u and G l e i t m a n conc luded that K e l l i ' s mean ing fo r look was to exp lore or apprehend . It is not c lear what was meant by apprehend. H o w e v e r , i f th is verb 84 can be in te rp re ted as ' to take possession o f , or 'to seize ' , then i t w o u l d appear that K e l l i had a d u a l mean ing f o r look when the i m p l i c a t u r e is to look at. A l t h o u g h B r y a n responded to look commands by app rehend ing the re fe r red- to objects, he a lways proceeded to exp lo re the objects once they were i n his possession. Because, l o g i c a l l y , i t seemed that he cou ld not exp lore w i thou t f i r s t app rehend ing the objects, the app rehend ing was not cons idered a s i g n i f i c a n t component o f h is i n te rp re ta t i on o f the c o m m a n d . T h e r e f o r e , i f ' se i z i ng ' , or ' t a k i n g possession o f an object ' , somet imes cons t i tu ted K e l l i ' s looking response, then she and B r y a n were con t ras t i ve i n th is regard . T h e c h i l d r e n also cont rasted i n terms of the percep tua l moda l i t i es they each emp loyed when ins t ruc ted to look at objects that were capab le o f e m i t t i n g sound . Wh i l e K e l l i ' s responses to these s p e c i f i c i ns t ruc t i ons a lways i n v o l v e d hap t i c pe rcep t ion , B r y a n o f ten responded w i t h listening behavior. In a d iscuss ion o f language deve lopment i n c h i l d r e n over f i v e , K a r m i l o f f - S m i t h (1979) pe rsuas ive ly a rgued that ch i l d ren ' s unde rs tand ing o f many l i n g u i s t i c s t ructures is m u l t i d e t e r m i n e d . B r y a n ' s listening responses to ce r ta in look commands e x e m p l i f y just such a m i x t u r e o f de terminants . H i s listening was e l i c i t ed by the charac te r i s t i cs o f the objects re fe r red to i n the commands , i.e. someth ing ex t raneous to h is we l l es tab l ished in te rp re ta t i on o f look, w h i c h he demonst ra ted i n other responses. Consequen t l y , a l though B r y a n and K e l l i d i f f e r e d w i t h respect to the behav io rs they pa i red w i t h requests to look at sound-emi t t i ng objects, the d i f f e r e n c e was not t reated as representat ive o f con t ras t i ve in te rp re ta t ions o f the percep tua l verb . A s i d e f r o m v a r i a t i o n i n the range o f body parts used to perce ive h a p t i c a l l y , and possib le incons is tency i n the c o u p l i n g o f exp lo ra t i on w i t h app rehend ing w i t h i n the i r looking behav io rs , K e l l i and B r y a n seemed to in terpre t the v i s u a l te rm look i n a s i m i l a r f a s h i o n . In e f fec t , the f i n d i n g s f r o m the look exper iments that L a n d a u a n d G l e i t m a n conduc ted w i t h a s ing le b l i n d c h i l d were rep l i ca ted i n my study. 85 O n the other h a n d , the results w h i c h emerged f r o m L a n d a u and G l e i t m a n ' s expe r imen ta t i on w i t h the b l i n d f o l d e d s ighted c h i l d r e n were not rep l i ca ted w i t h such cons is tency i n my s tudy w i t h Grego ry . Whereas L a n d a u and G l e i t m a n ' s subjects cons t rued looking w i t h the v i sua l m o d a l i t y on l y , G r e g o r y demonst ra ted an awareness that h is hands cou ld a l t e rna t i ve l y serve i n the percep tua l ac t i v i t y . I c o n c l u d e d that G rego ry ' s b i m o d a l i n te rp re ta t ion o f look deve loped out o f h is ea r l y , constant exposure to a s i b l i n g whose dom inan t percep tua l m o d a l i t y was hap t i c . F U R T H E R E V I D E N C E T O D E L I N E A T E  T H E R E S T R I C T E D R O L E O F E X P E R I E N C E  W I T H I N T H E W O R D / M E A N I N G M A P P I N G T A S K L a n d a u and G l e i t m a n v i e w e d K e l l i ' s success at a c q u i r i n g a mean ing f o r the v i s u a l te rm look as suppor t f o r the c l a i m that the role o f exper ience w i t h i n language l e a r n i n g is more cons t ra ined than common l y be l i eved . T o re i tera te , these authors reasoned that b l i n d c h i l d r e n have reduced access to e x t r a l i n g u i s t i c exper ience , and that v i s ion - re la ted terms are l e x i c a l i tems f o r w h i c h b l i n d a n d s igh ted c h i l d r e n shou ld have m a x i m a l l y d i f f e r e n t i n te rp re t i ve contexts . T h e y f u r t h e r r a t i o n a l i z e d that , " i f ex t r a l i ngu i s t i c exper ience p rov ides the route to l e a r n i n g , a b l i n d c h i l d shou ld have m a x i m u m t roub le w i t h these terms" (pp.2-3). Because K e l l i had no t roub le i n ass ign ing a mean ing to look as i t app l i ed to herse l f , L a n d a u and G l e i t m a n conc luded that an exper ience -domina ted theory o f a c q u i s i t i o n was con t ra - i nd i ca ted . So f a r , L a n d a u and G l e i t m a n ' s case i n v o l v i n g the look exper iments , f o r c o n s t r a i n i n g the ro le o f exper ience i n language l ea rn i ng has been pu re l y l og i ca l l y based. A s w i t h any r a t i o n a l i z a t i o n , the st rength o f the i r a rgument is con t ingent on the p l a u s i b i l i t y o f the i r log ic . In e f fec t , K e l l i and B r y a n ' s mastery o f look e x e m p l i f i e s the cons t ra ined role of exper ience on ly to the extent that i t is agreed 86 that (a) b l i n d c h i l d r e n rea l l y do have reduced access to ex t r a l i ngu i s t i c exper ience , (b) the s ighted v o c a b u l a r y can be associated w i t h i n d u c t i v e bases that are m a x i m a l l y d i f f e r e n t f o r b l i n d and s ighted c h i l d r e n , and (c) that a hap t i ca l l y -based i n te rp re ta t i on of look does re f lec t t roub le - f ree m a p p i n g of m e a n i n g onto words . W i t h respect to po in t (a) above, the l i t e ra tu re r ev i ewed i n chapte r one supports the asser t ion that b l i n d ch i l d ren ' s access to an in te rp re t i ve context does not pa ra l l e l the access of a s ighted c h i l d d u r i n g the f i r s t year of l i f e . H o w e v e r , whe the r a b l i n d ch i l d ' s access is ' r e d u c e d ' ra ther than mere ly ' a l t e red ' by the t ime the c h i l d is app roach ing three years o f age is debatab le . B y s h o w i n g that b l i n d and s ighted ch i l d ren ' s expe r i en t i a l d i f f e rences are more qua l i t a t i ve than q u a n t i t a t i v e , one cou ld contend that a b l i n d ch i l d ' s l i n g u i s t i c deve lopment does not represent language l ea rn i ng i n "a s i tua t ion i n w h i c h the oppor tun i t i es to observe the w o r l d are d i m i n i s h e d " (p.v i i ) . In other words , i t cou l d be a rgued that the e lement o f exper ience is as present i n the l ea rn i ng s i t ua t i on o f a b l i n d c h i l d as it is f o r a s ighted c h i l d , the d i f f e rences rest ing on l y i n the substance o f the ch i l d ren ' s respect ive exper iences. Cons is tent w i t h th is a rgument w o u l d be the v i e w that ne i ther K e l l i nor B r y a n ' s l i ngu i s t i c competence demonstrates the cons t ra ined ro le o f exper ience . Because L a n d a u a n d G l e i t m a n d i d not present c o m p e l l i n g ev idence to show that b l i n d ch i l d ren ' s re levant ex t r a l i ngu i s t i c exper ience is ' d i m i n i s h e d ' versus s imp l y ' a l te red ' , the v u l n e r a b i l i t y o f the i r log ic is apparen t f r o m the outset. Po in t (b) o f L a n d a u and G l e i t m a n ' s a rgument states that the s ighted v o c a b u l a r y is the op t ima l test case f o r show ing up any d i f f i c u l t i e s that a b l i n d c h i l d m igh t have i n ass ign ing mean ing to words. A t a s u p e r f i c i a l l eve l , L a n d a u a n d G l e i t m a n ' s cho ice o f tests seems appropr ia te . B l i n d c h i l d r e n can ' t see. T h e r e f o r e , they shou ld have d i f f i c u l t y i n te rp re t i ng , f o r examp le , the w o r d look. H o w e v e r , even w i t h i n an exper ience -domina ted theory of language l e a r n i n g , the 87 ef fec t i veness o f th is v o c a b u l a r y becomes obscured when cons ide r i ng the abstract l eve l of mean ing o f v i s u a l terms. F o r ins tance, many uses o f the w o r d look i m p l y 'a t tend to ' or 'pe rce ive and ob ta in an image o f . Bu t a t tend ing to, or o b t a i n i n g an image of someth ing do not necessar i ly requ i re v i sua l percep t ion . Consequen t l y , i t is reasonable to expect that l ea rn i ng a v i sua l verb as i t app l ies to onesel f and as i t per ta ins to such mean ings , shou ld be no more d i f f i c u l t f o r a b l i n d c h i l d than l e a r n i n g any other perceptua l verb . One who contested component (a) o f L a n d a u a n d G l e i t m a n ' s a rgument , w o u l d cha l lenge component (b) w i t h the coun te ra rgument that b l i n d a n d s ighted c h i l d r e n have d i f f e r e n t , yet equa l l y cons is tent exper iences each t ime they hear ut terances con ta i n i ng v i sua l verbs. A s a resul t , both groups o f c h i l d r e n shou ld be equa l l y e f f e c t i v e at a r r i v i n g at meanings f o r the verbs , though the i r mean ings shou ld be consonant w i t h the i r o w n respect ive pe rcep tua l capab i l i t i es . P o i n t (c) o f L a n d a u and G l e i t m a n ' s argument can also be cha l l enged . R e c a l l that K e l l i ' s hap t i c i n te rp re ta t ion o f look was t reated as an ach ievement that e x e m p l i f i e d t roub le - f ree mapp ing i n spi te o f a d i m i n i s h e d i n te rp re t i ve context . In bo ld cont rast to L a n d a u and G l e i t m a n ' s v i e w o f K e l l i ' s success w i t h look is the o p i n i o n that the un iqueness o f her ex t r a l i ngu i s t i c exper ience is accoun tab le f o r the ways i n w h i c h her mean ing f o r the v i sua l te rm dev ia ted f r o m that o f her s ighted peers. A t v a r i a n c e w i t h the con t ro l subjects, K e l l i ' s i n te rp re ta t i on o f look made no a l lus ions to v i s i o n . R a t h e r , her mean ing i m p l i c a t e d her own dom inan t percep tua l moda l i t y . K e l l i ' s unde rs tand ing o f look as i t app l i ed to herse l f is ana logous to a p h y s i c a l l y h a n d i c a p p e d ch i l d ' s dec is ion that run means turn the wheels quickly. Consequen t l y , i t is unc lear how such w o r d / m e a n i n g m a p p i n g demonstrates reduced u t i l i t y o f the ex t r a l i ngu i s t i c context . L a n d a u and G l e i t m a n ' s see exper iments were conduc ted two years a f te r the look exper iments , when K e l l i was app rox ima te l y f i v e years o ld . In those 88 exper imen ts , i t was observed that K e l l i had a v i sua l l y -based i n te rp re ta t i on fo r see as i t a p p l i e d to others. H e r mastery o f see is a better examp le o f language l ea rn i ng i n the absence o f a r i c h i n te rp re t i ve context . H o w e v e r , because by age f i v e , a b l i n d c h i l d w o u l d have had s i g n i f i c a n t t u to r ing pe r t a i n i ng to the v i s u a l ab i l i t i es o f others, the w o r d / m e a n i n g m a p p i n g represented i n K e l l i ' s unde rs tand ing o f see is a ques t ionab le mode l o f what t y p i c a l l y occurs i n i n i t i a l language a c q u i s i t i o n . W i th respect to the outcome o f the look exper iments , K e l l i ' s unadu l t e ra ted mean ing o f look at age three cou ld be t reated as suppor t f o r an exper ience-based theory o f l i n g u i s t i c deve lopment , though L a n d a u and G l e i t m a n used it as ev idence f o r the con t ra ry . Wh i l e c o n f i d e n t about the l og i ca l i m p l i c a t i o n s o f K e l l i ' s success w i t h the te rm look, L a n d a u and G l e i t m a n sought to st rengthen the i r case f o r c o n s t r a i n i n g the ro le of exper ience . T h e y d i d th is by e x a m i n i n g the poss ib i l i t y that K e l l i ' s ma te rna l l i n g u i s t i c i npu t m igh t have con ta ined ut terances o f look that were cons is ten t ly pa i r ed w i t h p a r t i c u l a r con tex tua l c i rcumstances . T h e ex is tence o f such pa i r i ngs w o u l d have suppor ted the s imp l i s t i c theory o f language l e a r n i n g w h i c h they opposed. U s i n g aud io and v ideo record ings o f mate rna l speech to K e l l i be fore age three, L a n d a u and G l e i t m a n ana l yzed the con tex tua l corre lates o f look and see w i t h i n an e x a m i n a t i o n of the s i t ua t i ona l pa i r i ngs of a number of c o m m o n l y used verbs. T h e inves t i ga t i on was i n i t i a t e d w i t h the poss ib i l i t y i n m i n d that K e l l i associated hap t i c behav io rs w i t h look because her mother u t tered th is w o r d at t imes w h e n K e l l i was e i ther i n tac tua l contac t w i t h , or close to target objects. F o r the present case, we are ask i ng i f 'perce ive or exp lo re m a n u a l l y ' is the i n e v i t a b l e con jec ture about the mean ing o f look just because th is ve rb , as opposed to others, occurs i n mate rna l speech when the c h i l d is h o l d i n g and m a n i p u l a t i n g some object , or at least has i t w i t h i n arm's reach, (p. 102) 89 L a n d a u and G l e i t m a n set up the con tex tua l categor ies near ( w i t h i n arm's reach) , a n d far (out o f a rm 's reach) to re fe r to K e l l i ' s pos i t i on i n re la t i on to target objects a n d people, as her mother u t tered look. A p p r o x i m a t e l y 1500 ut terances c o n t a i n i n g the verbs look, see, come, get, give, go, have, hold, play, and put were assigned to one o f the two categor ies. T h i r t y - f o u r of the coded ut terances con ta ined the ve rb look, and e ighteen con ta ined the verb see. T h e ana lys is revea led that 73% o f the mate rna l ut terances of look occu r red w h e n K e l l i was near to, or h o l d i n g the target object. H o w e v e r , a lmost 100% of the ma te rna l u t terances o f give, hold and put were also p roduced under the near c o n d i t i o n . A l t h o u g h the near/far categor ies d i d d i f f e r e n t i a t e between some o f the verbs, the categor ies f a i l e d to d i s t i ngu i sh look. L a n d a u and G l e i t m a n t reated the outcome o f th is ana lys is as a d d i t i o n a l suppor t f o r the c l a i m that ex t r a l i ngu i s t i c exper ience is an i n s u f f i c i e n t source of mean ing . Because K e l l i ' s success at respond ing to the i m p l i c a t u r e look for i n E x p e r i m e n t O n e d i d not requ i re that an object be near at h a n d , I quest ion w h y any such uses o f the ve rb w o u l d n ' t have been t reated separate ly i n the ana lys is o f the con tex tua l corre la tes f o r look. L a n d a u and G l e i t m a n ' s o w n data p red ic t that ce r ta in mate rna l uses o f th is ve rb shou ld not be assigned a near spa t ia l cod ing . Consequen t l y , the i r f i n d i n g that the v i s u a l verb was incons is ten t l y g i ven a near c o d i n g comes as no surpr ise . F u r t h e r m o r e , because these researchers ' own expe r imen ta l des ign demonst ra ted that objects d i d not need to near, or even present i n o rder f o r K e l l i to respond to an i ns t ruc t i on to look, the i r cho ice o f cod ing schemes f o r an ana lys is of the con tex tua l corre la tes fo r common l y heard verbs must be ques t ioned. These authors a c k n o w l e d g e d that one m igh t contest the i r con tex tua l categor ies as inadequa te fo r the type o f ana lys is they a t tempted to p e r f o r m , and that such i n a d e q u a c y w o u l d u n d e r m i n e the re levance o f the i r resul ts to the con tex tua l l e a r n i n g hypothes is . Bu t th is acknow ledgement was f o l l o w e d up w i t h a defense (See L a n d a u & G l e i t m a n 1985:106-107) i n w h i c h it was ins is ted that "the cod ing categor ies chosen were re levant " (p.106). C o n f i d e n t about the adequacy o f the i r s i t ua t i ona l pa i r i ngs ana lys i s , L a n d a u and G l e i t m a n m a i n t a i n e d that the conc lus ions d e r i v e d f r o m the ana lys is were v a l i d . It is my o p i n i o n that the near/ far c o d i n g scheme was inadequa te . In a d d i t i o n to c o d i n g K e l l i ' s p r o x i m i t y to target objects associated w i t h va r i ous verbs, L a n d a u and G l e i t m a n cou ld also have noted whether or not K e l l i was a l ready a t tend ing to the targets as the verbs were heard . C o n s i d e r that a c o m m a n d to look cou ld be used to direct one's a t ten t ion towards someth ing w h i c h is nea rby , and a c o m m a n d to give m igh t be used to re fe r to a nearby object to w h i c h the person is already a t tend ing . In e f fec t , p r o x i m i t y to an object w o u l d not be e f f e c t i v e at d i f f e r e n t i a t i n g between look and give. O n the other h a n d , '+ p rev i ous l y a t t end ing to target ^object' cou l d d i s t i ngu i sh the verbs. L a n d a u and G l e i t m a n ' s a t tempt to de te rmine a consistent con tex tua l cor re la te f o r look shou ld have en ta i l ed a more de ta i l ed c o d i n g scheme. T h e genera l na ture of the categor ies emp loyed prec ludes these researchers f r o m j u s t i f i a b l y c l a i m i n g , that look cou l d not be d i s t i ngu i shed f r o m other verbs on the basis o f con tex tua l fac to rs w h i c h a c c o m p a n i e d the verb 's usage. AN ALTERNATIVE EXPLANATION  FOR THE ACQUISITION OF VERBS A f t e r p resent ing the i r ev idence to cons t ra in the ro le of exper ience i n language l e a r n i n g , L a n d a u and G l e i t m a n proceeded to e laborate on the i r e x p l a n a t i o n f o r how K e l l i , and any c h i l d assigns mean ing to verbs. R e c a l l that these au thors ' in terest d i d not encompass the o r i g i n of a ch i l d ' s concepts. R a t h e r i t was l i m i t e d to the a c c o u n t a b i l i t y f o r a ch i l d ' s success at m a p p i n g p rev ious l y conce i ved mean ings onto l e x i c a l i tems o c c u r r i n g in heard speech. 91 A t the base of L a n d a u and G l e i t m a n ' s exp lana t i on f o r the m a p p i n g task were a set o f c la ims . 1) P h r a s a l and l e x i c a l categor ies are i nna te l y k n o w n , or l ea rned by the c h i l d be fo re m a p p i n g occurs. 2) T h e c h i l d is b i o l o g i c a l l y endowed w i t h an a b i l i t y to p e r f o r m syn tac t i c analyses on ut terances heard i n her l i n g u i s t i c env i ronmen t . These analyses y i e l d syn tac t i c parse trees, c o n t a i n i n g h i e r a r c h i c a l l y a r ranged ph rasa l categor ies (e.g. N P , V P ) , w h i c h i n tu rn con ta in l e x i c a l categor ies such as nouns, p repos i t ions etc. T h e c h i l d uses sup rasegmen ta l / pe rcep tua l p roper t ies of the speech wave to make the cor rect phrasa l a n d l e x i c a l pars ings. (See G l e i t m a n & Wanner , 1982, f o r a d iscuss ion o f how this is so.) These pars ings i nd i ca te to the c h i l d w h i c h s p e c i f i c sound sequences i n the speech wave represent verbs. 3) T h e c h i l d has the concep tua l know ledge represented i n p red i ca te /a rgumen t sentence s t ruc ture , and knows that verbs encode pred icates , and that nomina l s encode arguments. 4) T h e c h i l d knows the type and number o f a rguments w h i c h s p e c i f i c verbs requ i re . L a n d a u and G l e i t m a n contended that the c h i l d examines the mean ing of an u t te rance w i t h i n the syn tac t i c representa t iona l f o rma t of a parse tree. These syn tac t i c analyses e f f e c t i v e l y d i s t i ngu i sh between d i f f e r e n t verbs, and p rov ide the c h i l d w i t h i n f o r m a t i o n about the verbs meanings. -L a n d a u and G l e i t m a n tested the i r exp lana t i on fo r verb l e a r n i n g , w h i l e hypo thes i z i ng that the verbs p rev ious l y a n a l y z e d i n the ma te rna l speech to K e l l i , c o u l d be i n d i v i d u a t e d i n terms of the syn tac t i c env i ronmen ts i n w h i c h they occu r red . These authors f u r t h e r hypo thes ized that , the spa t i a l - l oca t i ona l c o d i n g of the s i t ua t i on of use (near, f a r , no object) w o u l d p r o v i d e t rus twor thy i n f o r m a t i o n about the meanings, but now on ly as cons t ra ined by the syn tac t i c s t ruc ture of the mother 's sentences, (p. I l l ) L a n d a u and G l e i t m a n re -examined the 1500+ ut terances p roduced by K e l l i ' s mother i n K e l l i ' s presence be fore age three, w h i c h con ta ined the verbs look, see, give, put, hold, play, get, have, go, and come. U t te rances were coded a c c o r d i n g to 92 subca tegor i za t ion f rames , d e f i n e d as the phrasa l categor ies that f o l l o w e d a verb i n a sentence. T h i s ana lys is revea led pa r t i cu l a r syn tac t i c env i ronments i n w h i c h on l y look a n d see occu r red . A subset of these env i ronments was reserved f o r the use o f look. F o r examp le , on l y look appeared in de i c t i c in te r jec t i ve commands , e.g. "Look ! " and " L o o k , that 's a boot" (p. l 13). Look was also the on l y verb to appear d e i c t i c a l l y w i t h a f ree how re la t i ve , e.g. " L o o k how I do it" (p. l 13). The re was also a subset of subca tegor i za t i on f rames w h i c h pe rmi t ted on l y the ve rb see. L i k e w i s e , there were env i ronmen ts i n w h i c h ne i ther look nor see cou ld appear , yet other c o m m o n verbs cou ld . (See L a n d a u & G l e i t m a n 1985:112 fo r a summary o f the subca tegor i za t ion f rames associated w i t h each of the ten verbs.) L a n d a u and G l e i t m a n comb ined the spa t i a l - l oca t i ona l ana lys is f o r each use o f look and see w i t h an ana lys is of the subca tegor i za t ion f rames permi t ted by the two verbs. T h i s c o m b i n e d ana lys is revea led greater i n d i v i d u a t i o n o f look and see. Whereas the spa t ia l cod ings a lone had revea led that on l y 73% o f a l l uses o f look occu r red w h e n K e l l i was near to, or h o l d i n g the target object , the c o m b i n e d ana lys i s i n d i c a t e d that ce r ta in syn tac t i c env i ronments were associated w i t h the near spa t ia l category w i t h 100% cons is tency. L a n d a u and G l e i t m a n c l a i m e d that a c h i l d ' s d u a l i nspec t ion o f the spa t ia l and syn tac t i c env i ronmen ts i n w h i c h verbs are used w i l l l ead to c learer d i s t i nc t ions among them. L a n d a u and G l e i t m a n demonst ra ted the re levance o f syn tac t i c analyses to the i n fe rence of mean ing by show ing how d i s t i nc t i ve syn tac t i c encod ings f o r verbs are co r re la ted w i t h the i r semant ic proper t ies. T h e y began by s ta t ing the i r agreement w i t h G r u b e r ' s (1968) pos i t i on that verbs o f mo t ion and perceptua l exp lo ra t i on are re la ted , i n that look i nvo l ves the 'movement o f gaze ' , or i n the case o f a b l i n d c h i l d / m o v e m e n t of the hands ' . F u r t h e r m o r e , they expressed the i r agreement w i t h the c l a i m that the s i m i l a r i t y i n mean ing fo r the verb types is re f l ec ted i n the fac t that they each take loca t i ve prepos i t ions and par t ic les f ree ly . L a n d a u and 9 3 G l e i t m a n m a i n t a i n e d that i f K e l l i t reated haptic exploration as mo t i ona l , then she w o u l d k n o w that i ts l e x i c a l encod ing w o u l d be a verb w h i c h took loca t i ve prepos i t ions f ree l y . A p p l i c a t i o n o f th is k n o w l e d g e . w o u l d a u t o m a t i c a l l y exc lude verbs that are h i g h l y res t r i c ted in the l oca t i ve prepos i t ions they a l l o w (such as have, say and play), f r o m be ing cons idered as possib le l e x i c a l cand ida tes f o r the concept 'explore haptically'. T h e range of possib le cand ida tes f o r encod ing 'explore haptically' w o u l d be f u r t h e r reduced upon e x a m i n i n g the arguments requ i red to accompany th is p red ica te . 'Explore haptically' takes two arguments - a subject and an object , whereas some verbs such as give, put, get, and hold are three argument pred icates . S ince verbs w h i c h requ i re three arguments , or w h i c h are l i m i t e d i n the loca t i ve prepos i t ions they a l l ow , can ' t be cons idered as l e x i c a l encod ings f o r a p red ica te that takes on l y two arguments , and can take loca t i ve prepos i t ions f r e e l y , a l l that w o u l d r e m a i n i n the set o f verbs that were c o m m o n l y heard i n ma te rna l speech to K e l l i , w o u l d be come, go, look, and see. L a n d a u and G l e i t m a n e x p l a i n e d that syn tac t i c analyses cou ld also d i s t i n g u i s h look and see f r o m come and go. Wh i le a l l o f these verbs are m o t i o n a l , and each requ i res two arguments , on l y the perceptua l verbs can take sentent ia l complements , or how re la t i ves . G i v e n exposure to uses o f the percep tua l verbs on l y i n such syn tac t i c env i ronmen ts , the c h i l d w o u l d beg in to ru le out come and go as cand ida tes f o r the meanings explore and perceive haptically. F u r t h e r m o r e , by p e r f o r m i n g a c o m b i n e d ana lys is of the spat ia l cod ings and syn tac t i c env i ronments associated w i t h these f o u r verbs, K e l l i w o u l d have seen that whenever look and see shared syn tac t i c contexts w i t h come and go, there w o u l d have been a s t rong tendency f o r the f o r m e r pa i r o f verbs to be used when she was near to, or h o l d i n g the target object. C o n v e r s e l y , she w o u l d have noted that the la t ter pa i r o f verbs were p r e d o m i n a n t l y used when she was fa r f r o m the target objects. In e f f ec t , the 94 c o m b i n e d ana lys is w o u l d have f u r t h e r d i s t i ngu i shed the percep tua l verbs f r o m come and go. L a n d a u and G l e i t m a n c l a i m e d that once a c h i l d is aware o f the semant ic d i s t i n c t i o n between look and see, she shou ld be ab le to make the cor rect mapp ings between the two mean ings and the i r l e x i c a l representat ions. F o r examp le , i n k n o w i n g that see is s ta t ive , and look is ac t i ve , the c h i l d shou ld not expect that the f o r m e r w o u l d appear i n an impe ra t i ve sentence s t ruc ture , s ince a f te r a l l , one cannot c o m m a n d another to be i n a state. A s was ev idenced i n K e l l i ' s ma te rna l speech sample , on l y look, and not see, was used impe ra t i ve l y . F o l l o w i n g G r u b e r ' s ana lys is o f the mot ion verbs, L a n d a u and G l e i t m a n agreed that look and see are also d i s t i ngu ishab le i n terms o f the i r membersh ips to groups o f verbs that e i ther a l l o w the loca t i ve p repos i t i on to or the loca t i ve p repos i t i on toward. Look is a member o f the toward g roup (mot ion i n the d i r e c t i o n o f a goal) , and see is a member of the to g roup (mot ion to a goal). These loca t i ve prepos i t ions are syn tac t i c marke rs encod ing subt ly d i f f e r e n t no t ions , that can be used i n the deduc t i on of mean ing o f verbs w i t h w h i c h they are associated. G r u b e r ' s ana lys is took in to account the usages o f the v i sua l verbs to encode mo t ion to, or i n the d i r e c t i o n o f a goa l , i n sentences w h i c h do not con ta in l e x i c a l representat ions of the prepos i t ions . G r u b e r asserted that wh i l e the prepos i t ions are not present i n the su r face s t ruc tures , they exist i n the deep st ructures of the sentences. L a n d a u and G l e i t m a n expressed the i r conserva t i ve agreement w i t h th is e x p l a n a t i o n , w h i l e s ta t ing , We accept at least p r o v i s i o n a l l y the no t ion tha t verbs may have semant ic subs t ruc tu re not e x h i b i t e d i n the sur face syn tac t i c f r a m e ; i n p a r t i c u l a r , w i t h see the f u n c t i o n usua l l y hand led w i t h a separate p repos i t i on (to) is a component o f the mean ing o f the verb i tse l f , (p.145) L a n d a u and G l e i t m a n f u r t he r stated that such cover t encod ings fo r components o f mean ing demonst ra te the comp l i ca t i ons inheren t i n the w o r d / m e a n i n g m a p p i n g task, and that the re la t i ve d i f f i c u l t y of l ea rn i ng a ce r ta in ve rb shou ld be 9 5 pred ic tab le based on the s t ruc tu ra l leve l at w h i c h a l l components of i ts mean ing are represented. In s u m m a r y , a c c o r d i n g to L a n d a u and G l e i t m a n ' s exp lana t i on f o r ve rb l e a r n i n g , the c h i l d begins the w o r d / m e a n i n g m a p p i n g task w i t h a know ledge of the p red ica te -a rgument l og i ca l s t ruc ture of sentences; a know ledge (perhaps n a t u r a l l y endowed) about ph rasa l and l e x i c a l categor ies; and an a b i l i t y to p e r f o r m syn tac t i c analyses of the ut terances she hears. B y p e r f o r m i n g syn tac t i c ana lyses, the c h i l d dep ic ts ut terances i n terms of parse trees w h i c h serve as the representa t iona l f o r m a t against w h i c h she cons iders w o r d mean ing . Componen ts o f mean ing are encoded i n the syn tac t i c proper t ies of verbs w h i c h become d i sce rn ib le i n the parse trees. L a n d a u a n d G l e i t m a n con tend that the semant ic i n f o r m a t i o n represented in syn tac t i c marke rs that su r round a verb is exp lo i ted by the c h i l d who is endeavo r i ng to de te rmine the verb 's mean ing . C H A P T E R FOUR CONCLUSION A s proponents of inna t i s t language acqu i s i t i on theory , L a n d a u and G l e i t m a n agreed that a c h i l d uses the ex t ra l i ngu i s t i c context when ass ign ing mean ing to words . H o w e v e r , s imu l taneous ly they a rgued that the e x t r a l i n g u i s t i c contex t is an i n s u f f i c i e n t source of m e a n i n g , and that i t is supp lemented by the i n te rp re t i ve context p r o v i d e d w i t h i n language i tse l f . T o demonstrate the i n s u f f i c i e n c y o f e x t r a l i n g u i s t i c exper ience , and the necessi ty f o r an a l te rna t i ve source o f m e a n i n g , L a n d a u and G l e i t m a n showed how a b l i n d c h i l d deve loped a soph is t i ca ted a b i l i t y to use and unde rs tand the s ighted v o c a b u l a r y - terms fo r w h i c h she had no c o n v e n t i o n a l ex t r a l i ngu i s t i c exper ience. These authors v i e w e d K e l l i ' s success w i t h v i s u a l words as ev idence to suppor t an innate p red ispos i t i on in the c h i l d to s y n t a c t i c a l l y ana lyze heard ut terances, i n o rder t o ' ga in a d d i t i o n a l ins igh t about the i r mean ings . T h r o u g h my rep l i ca t i on s tudy , I also showed that a b l i n d c h i l d can master the v i sua l te rm look, as i t app l i ed to h imse l f . Wh i le not r e f u t i n g L a n d a u and G l e i t m a n ' s e x p l a n a t i o n fo r verb l e a r n i n g , I contend that the outcomes of our look exper iments do not necessar i ly suggest that exper ience p lays a cons t ra ined ro le i n language a c q u i s i t i o n . R a t h e r , I c l a i m that because the hap t i c basis f o r B r y a n and K e l l i ' s i n i t i a l mean ing f o r look u n i q u e l y re f lec ted these ch i l d ren ' s exper iences w i t h the w o r l d , the ch i l d ren ' s unde rs tand ing o f th is te rm cou ld be used to h i gh l i gh t the ro le p l ayed by the ex t r a l i ngu i s t i c context i n language l ea rn ing . L a n d a u and G l e i t m a n ' s i n te rp re ta t i on of the results f r o m the look exper iments demonstrates how theore t i ca l perspect ive determines the l igh t i n w h i c h we v i e w e m p i r i c a l f i n d i n g s . B I B L I O G R A P H Y A d e l s o n , E. , & F r a i b e r g , S. (1974). Gross motor deve lopment i n i n f a n t s b l i n d f r o m b i r t h . Child Development. 45. (114-126). A n d e r s e n , E . S., D u n l e a , A . & K e k e l i s , L.S. (1984). B l i n d ch i l d ren ' s language reso l v ing some d i f f e rences . Journal of Child Language 11. (645-664). B l o c k b e r g e r , S. J . & Johnson , C . E . (1986). "How big are you today?": interaction routines with twins when one is blind. Paper presented at the T e n t h A n n u a l Bos ton U n i v e r s i t y C o n f e r e n c e on Language Deve lopmen t , Oc tober . B l o o m , L., L i g h t b r o w n , P. & H o o d L. (1975). S t ruc tu re and v a r i a t i o n i n c h i l d language. Monographs for the Society for Research in Child Development. 160. B u r l i n g h a m , D. (1961). Some notes on the deve lopment o f the b l i n d . Psychoanalytic Study of the Child. 16. (121-145). B u r l i n g h a m , D. (1964). H e a r i n g and i ts ro le i n the deve lopment o f the b l i n d . Psychoanalytic Study of the Child. 26. (121-145). C l a r k , H . H . & C l a r k , E . V . (1977). Psychology and Language: an introduction to psycholinguistics. N e w Y o r k . H a r c o u r t B race J o v a n o v i c h . C o l l i s , G . M . & S c h a f f e r , M . R . (1975). S y n c h r o n i z a t i o n of v i s u a l a t ten t ion i n mo the r - i n f an t pa i rs . Journal of Child Psychology. 165. (315-320). D u n l e a , A . (1982). The role of visual information in the emergence of meaning: a comparison of blind and sighted children. U n p u b l i s h e d P h . D . d isser ta t ion . U n i v e r s i t y o f Sou thern C a l i f o r n i a . F e l d m a n , H . , G o l d i n - M e a d o w , S. & G l e i t m a n , L. R. (1978). B e y o n d Herodo tus : T h e c rea t ion o f language by l i n g u i s t i c a l l y dep r i ved dea f c h i l d r e n . In A . L o c k (ed), Action, Gesture, and Symbol: The emergence of language. L o n d o n . A c a d e m i c Press. F r a i b e r g , S., S iege l , B. & G i b s o n , R . (1966). T h e R o l e o f Sound i n the Search B e h a v i o r o f a B l i n d In fan t . Psychoanalytic Study of the Child. 21. (327-357). F r e e d m a n , D., F o x - L o k e n d a , B., M a r g i l e t h , D. & M i l l e r , D. (1969). T h e deve lopment o f the use o f sound as a gu ide to a f f e c t i v e and cogn i t i ve behav io r : a two-phase process. Child Development. 40. (1099-1105). G l e i t m a n , L . R . ( in press). Biological dispositions to learn language. G l e i t m a n , L . R. & Wanner , E. (1982). Language A c q u i s i t i o n : T h e state o f the state o f the art. In E . Wanner & L. R. G l e i t m a n (eds), Language acquisition: The state of the art. N e w Y o r k . C a m b r i d g e U n i v e r s i t y Press. K a r m i l o f f - S m i t h , A . (1979). Language Deve lopmen t A f t e r F i v e . In P. F le t che r & M . G a r m a n (eds), Language Acquisition. C a m b r i d g e . U n i v e r s i t y Press. 98 K e e l e r , W. (1958). A u t i s t i c pat terns and de fec t i ve c o m m u n i c a t i o n i n b l i n d c h i l d r e n w i t h re t ro len ta l f i b r o p l a s i a . In P. H o c h & J . Z u b i n (eds), Psychopathology of Communication. N e w Y o r k . G r u n e and St ra t ton , Inc. K e k e l i s , L . & A n d e r s e n , E . (1984). F a m i l y c o m m u n i c a t i o n styles and language deve lopment . Journal of Visual Impairment and Blindness. Feb. (54-65). L a n d a u , B. (1983). B l i n d C h i l d r e n ' s language is not 'meaning less ' . In A . M i l l s (ed), Language Acquisition in the Blind Child. L o n d o n . C r o o m H e l m . L a n d a u , B. & G l e i t m a n , L . R. (1985). Language and Experience: Evidence from the Blind Child. C a m b r i d g e , M A . H a r v a r d U n i v e r s i t y Press. M u l f o r d , R. ( in press). F i r s t words o f the b l i n d c h i l d . In M . S m i t h & J . L o c k e (eds), The Emergent Lexicon: the Child's Development of a Linguistic Vocabulary. N e w Y o r k . A c a d e m i c Press. N e l s o n , K . (1973). S t ruc tu re and strategy i n l ea rn i ng to ta lk . Monograph of the Society for Research in Child Development. 38. (1-2). N e w p o r t , E. , G l e i t m a n , H . & G l e i t m a n , L . R. (1977). M o t h e r , I'd ra ther do i t myse l f : some e f fec ts and non-e f fec ts o f mate rna l speech sty le. In C . E . Snow & C . A . Fe rguson (eds), Talking to Children: Language Input and Acquisition. C a m b r i d g e , E n g l a n d . C a m b r i d g e U n i v e r s i t y Press. P h i l l i p s , J . R. (1973). Syn tax and v o c a b u l a r y o f mothers ' speech to young c h i l d r e n : A g e and sex compar isons. Child Development. 44. (182-185). R o b s o n , K . S. (1967). T h e R o l e o f E y e to E y e Con tac t i n M a t e r n a l - I n f a n t A t t a c h m e n t . Journal of Child Psychology and Psychiatry. 8. (13-25). R o w l a n d , C . (1983). Pat terns o f i n te rac t i on between three b l i n d i n f a n t s and the i r mothers. In A . M i l l s (ed), Language Acquisition in the Blind Child. L o n d o n . C r o o m H e l m . Sha tz , M . , & G e l m a n , R . (1973). T h e deve lopment o f c o m m u n i c a t i o n sk i l l s : M o d i f i c a t i o n s i n the speech of young c h i l d r e n as a f u n c t i o n o f l is tener . Monographs of the Society for Research in Child Development. 38. (Ser ia l N o . 52). S n o w , C . E . & F e r g u s o n , C . A . (eds) (1977). Talking to children: Language input and acquisition. C a m b r i d g e . C a m b r i d g e U n i v e r s i t y Press. S te rn , D. (1974). M o t h e r and In fan t at P l a y : T h e D y a d i c In te rac t ion I n v o l v i n g F a c i a l , V o c a l and G a z e Behav io rs . In M . L e w i s and L. R o s e n b l u m (eds), The Effect of the Infant on Its Caregiver. N e w Y o r k . Wi ley . S te rn , D. (1977). T h e F i r s t R e l a t i o n s h i p . In W. D a m o n (ed), Social and Personality Development: Essays on the Growth of the Child. N e w Y o r k . W. W. N o r t o n & C o m p a n y , Inc. U r w i n , C . (1978). T h e deve lopment o f c o m m u n i c a t i o n between b l i n d i n f a n t s and the i r parents. In A . L o c k (ed), Action, Gesture and Symbol: the emergence of language. L o n d o n . A c a d e m i c Press. U r w i n , C . (1984). C o m m u n i c a t i o n i n i n f a n c y and the emergence of language i n b l i n d c h i l d r e n . In R. Sch ie fe lbusch & J . P i c k a r (eds), The acquisition of Communicative Competence. Ba l t imo re . U n i v e r s i t y P a r k Press. W o l f f , P. H . (1963). Observa t ions on the E a r l y Deve lopmen t of S m i l i n g . In B. M . Foss (ed), Determinants of Infant Behavior. 2. Wer th , P. (1983). M e a n i n g i n Language A c q u i s i t i o n . In A . M i l l s (ed), Language Acquisition in the Blind Child. L o n d o n . C r o o m H e l m . Wood, M . (1970). Problems in the development and home care of preschool blind children. U n p u b l i s h e d doc to ra l d isser ta t ion . U n i v e r s i t y o f N o t t i n g h a m , E n g l a n d . 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0097013/manifest

Comment

Related Items