Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The transition to syntax : influences on children’s constructional complexity Lapadat, Judith Colleen 1983

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1983_A6_7 L36.pdf [ 4.56MB ]
Metadata
JSON: 831-1.0095805.json
JSON-LD: 831-1.0095805-ld.json
RDF/XML (Pretty): 831-1.0095805-rdf.xml
RDF/JSON: 831-1.0095805-rdf.json
Turtle: 831-1.0095805-turtle.txt
N-Triples: 831-1.0095805-rdf-ntriples.txt
Original Record: 831-1.0095805-source.json
Full Text
831-1.0095805-fulltext.txt
Citation
831-1.0095805.ris

Full Text

THE TRANSITION TO SYNTAX: INFLUENCES ON CHILDREN'S CONSTRUCTIONAL COMPLEXITY by JUDITH COLLEEN LAPADAT .A. (Honours), The Un i v e r s i t y of B r i t i s h Columbia, 197 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (School of Audiology and Speech Sciences) We accept t h i s t hesis as-conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA APRIL 1983 (c) Judith Colleen Lapadat I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e h e a d o f my d e p a r t m e n t o r by h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . D e p a r t m e n t o f The U n i v e r s i t y o f B r i t i s h C o l u m b i a 2075 Wesbrook P l a c e V a n c o u v e r , Canada V6T 1W5 D F - 6 f?/7<n i i Abstract The complexity of children's l i n g u i s t i c constructions i n conversation during the t r a n s i t i o n from single-word to multiword speech was measured i n s i x conditions produced by manipulating the adult i n t e r a c t i o n v a r i a b l e s , f a m i l i a r i t y , attention, and conversational c o n t r o l . Subjects were four c h i l d r e n , ages 1 ;9 to 2;3, and t h e i r mothers. The children's constructions were categorized according to three l e v e l s of complexity (single words, v e r t i c a l constructions, h o r i z o n t a l constructions), i n four tasks (high attention, low attention, high c o n t r o l , low c o n t r o l ) , during dialogues with t h e i r mothers, then with strangers. While t h e i r proportion of complex to l e s s complex constructions was not r e l a t e d to adult f a m i l i a r i t y or a t t ention, c h i l d r e n produced s i g n i f i c a n t l y more complex constructions (p = .005) when they c o n t r o l l e d the conversation. Further analyses of the children's frequency of constructions demonstrated that the c h i l d r e n talked more to t h e i r mothers than to strangers, and more given high adult attention than low adult attention. These findings have important implications for c l i n i c a l language sampling and therapy procedures. Table of Contents i i i page Abstract i i Table of Contents i i i L i s t of Tables i v Acknowledgements v Introduction .1 Chapter One: Review of Current Issues 5 Influence of Caretakers' Speech on Children's Language Development 5 Context and Language Functions 9 Conversational Constraints and Discourse Devices 16 Topic-Comment Structures 22 The T r a n s i t i o n to Syntax 36 The Model 42 Chapter Two: Method 50 Chapter Three: Results 55 Complexity of Utterances 56 Frequency of Utterances 60 Chapter Four: Discussion 66 F a m i l i a r i t y 67 Attention 69 Control 71 Frequency 75 Implications 77 For Further Research 79 References 83 Reference Note 88 Addenda 89 Appendix 90 L i s t of Tables Table 1 Comparison of D e f i n i t i o n s of V e r t i c a l Constructions Table 2 Children's Language Levels at Pretest Date Table 3 Summary of Variables By Session and Task Table 4 Total Number of Constructions Coded For Each Child by Task (High Attention, Low Attention, High Control, Low Control), Session (High F a m i l i a r i t y , Low F a m i l i a r i t y ) , and Construction Complexity (1,2,3) Table 5 Marginal Table Of F a m i l i a r i t y by Construction Complexity: Number of constructions at each l e v e l produced by a l l the c h i l d r e n for high and low values of the v a r i a b l e Table 6 Strength of E f f e c t s i n the Model To F i t Data on Relationship of F a m i l i a r i t y , Attention, and Control to Children's Construction Complexity Table 7 Marginal Table of Attention by Construction Complexity: Number of constructions at each l e v e l of complexity produced for high and low values of the va r i a b l e Table 8 Marginal Table of Control by Construction Complexity: Number of constructions at each l e v e l of complexity produced for high and low values of the v a r i a b l e Table 9 To t a l Number of Multiword Constructions ( V e r t i c a l , H orizontal, & H o r i z o n t a l - V e r t i c a l ) Produced by Each Child i n Each Condition Table 10 Strength of E f f e c t s i n the Model to F i t Frequency : . Data: Children's Number of Multiword, Horizontal, V e r t i c a l , Single-Word, & To t a l Constructions i n F a m i l i a r i t y , Attention, and Control Conditions Table 11 Total Number of Horizontal Constructions Produced by Each Child i n Each Condition Table 12 T o t a l Number of V e r t i c a l Constructions Produced by Each Child i n Each Condition Table 13 To t a l Number of Single-Word Constructions Produced by Each Ch i l d i n Each Condition Table 14 Total Number of Constructions of a l l Types Produced by Each Ch i l d i n Each Condition i v page 39 51 54 57 58 59 59 60 60 62 63 64 64 65 V Acknowledgements Many people have provided c r i t i c i s m , assistance, and encouragement over the course of t h i s research. I am deeply indebted to Daniel, Jon i , M i l l y , Cory, and t h e i r mothers f or volunteering t h e i r time to p a r t i c i p a t e i n t h i s study. In addition, I would l i k e to thank Carolyn Johnson, David Ingram, Malcolm Greig, Wendy Duke, Marshall Chasin, and Robert S t i r l i n g f o r providing c r i t i c a l advice and encouragement, and for a s s i s t i n g with taping sessions, research design, and s t a t i s t i c s . F i n a l l y , a s p e c i a l thank-you to Harold Janzen for h e l p f u l c r i t i c i s m at a l l phases of the study, for typing, and for u n f a i l i n g support. 1 The T r a n s i t i o n to Syntax: Influences on Children's Constructional Complexity Young chi l d r e n learning to t a l k begin by using one word at a time. Then sometime before t h e i r second birthday, the f i r s t two word utterances appear. These word combinations are viewed as the f i r s t "sentences" and the beginning of syntax. The t r a n s i t i o n from i s o l a t e d s i n g l e words, or holophrases, to syntac t i c two word sentences i s not abrupt, however. Structured language emerges gradually, and as I hope to demonstrate i n t h i s paper, c h i l d r e n produce constructions, or structured language, p r i o r to syntax. A model i s proposed to describe the stages i n constructional complexity from i s o l a t e d s i n g l e words to early two to three word sentences. While children's constructions develop from simple to complex, constructions of d i f f e r e n t l e v e l s of complexity can co-exist i n t h e i r speech at t h i s period (for example, a c h i l d may use some two word sentences along with some single-word utterances). In t h i s study, factors that influence the immediate complexity of children's constructions are considered. S p e c i f i c a l l y , the short term e f f e c t s of adult behaviours and manners of i n t e r a c t i n g with ch i l d r e n are investigated by s e t t i n g up d i f f e r e n t types of communicative s i t u a t i o n s and measuring the proportional complexity of children's constructions i n each condition. While long term e f f e c t s cannot be resolved from these measures, the adult behaviours that are rela t e d to immediate changes i n children's 2 constructional complexity are worthy of further longitudirial...inv.estigation'... The question of why ch i l d r e n move from si n g l e words to syntactic constructions, and the process.: through which t h i s occurs has been widely discussed i n the language a c q u i s i t i o n l i t e r a t u r e . An important observation i s that t h i s t r a n s i t i o n does not occur suddenly, or i n i s o l a t i o n from the people, routines, and s i t u a t i o n s of children's d a i l y environments. The f i r s t s y n t a c t i c two word utterances are often preceded by unanalyzed or formulaic phrases, or two word combinations with one element being a contentless place-holder. These are s i m i l a r i n form to the evolving two word sentences (see, for example, Dore, Fr a n k l i n , M i l l e r & Ramer, 1976; Lock, 1980). Once syntactic two word utterances make t h e i r appearance, they^represent ^ohiy.a small proportion of children's t o t a l output for a considerable period of time. Furthermore, there i s growing evidence that children's language i s already structured p r i o r to the emergence of syntactic utterances (Bates, 1976; Bloom, 1976; Scollon, 1976). Syntax does not represent the beginning of structure i n children's language, but rather the a c q u i s i t i o n of a new s t r u c t u r a l means of encoding pragmatic functions and semantic r e l a t i o n s h i p s which were previously, and continue simultaneously to be, encoded through other nonsyntactic structures. It can be argued that t h i s formal s t r u c t u r a l means of r e f e r r i n g to two or more things or to r e l a t i o n s h i p s between things i s more e f f i c i e n t than previous construction types and therefore more powerful. Increased communicative power provides an impetus for syn t a c t i c development. 3 Children's language learning i s embedded i n context. Children acquire language through i n t e r a c t i o n with other people, and each utterance i s an i n t e g r a l part of the event i n which i t occurs. Language i s used to get things done, to make s o c i a l contact, and to communicate meanings. These functions are served by d i f f e r e n t l i n g u i s t i c structures at d i f f e r e n t stages of development. Lock (1980) c i t e s an example of some constructions used by a c h i l d named Matthew. Deciding he wanted out of h i s highchair, Matthew used language f o r the function of getting h i s mother to l i f t him out. He communicated h i s desire to get down by whining, struggling, and reaching, and saying "down," then by using two singl e word utterances i n succession: "mommy1.1" "down." At a l a t e r stage he might have used a two word utterance: "mommy down." In each case h i s language accomplishes the same purpose (getting down from h i s c h a i r ) , and communicates the same meaning or proposition ("Mommy, I want you to get me down"), but the s t r u c t u r a l way of encoding the meaning changes. I w i l l argue that to understand how and why these structures evolve, we need to examine the discourse i n which they occur, and the conversational constraints acting upon ch i l d r e n and t h e i r partners. The r o l e that adults play as they i n t e r a c t with c h i l d r e n i n a conversation can have a major influence on discourse and on conversational constraints. This research addresses the question of what adult behaviours or att i t u d e s a l t e r the nature of the discourse, and whether there i s any r e l a t i o n s h i p between these adult behaviours and the complexity of constructions c h i l d r e n w i l l choose to use to encode meanings. In other words, are chi l d r e n at the onset of the two word stage more l i k e l y to produce syntactic constructions rather than s i n g l e words when adults i n t e r a c t with them i n a c e r t a i n way? To in v e s t i g a t e t h i s question, I w i l l f i r s t review relevant issues which have been discussed i n the language a c q u i s i t i o n l i t e r a t u r e . These issues include: the influence of the input language on children's language a c q u i s i t i o n , the r o l e of context i n language a c q u i s i t i o n , the r o l e of conversational constraints and discourse devices, topic-comment structures, and the t r a n s i t i o n to syntax. I w i l l then discuss Scollon's (1976) model of multiword development and propose some r e v i s i o n s and extensions to t h i s model. F i n a l l y , I w i l l report a study based on t h i s revised model, i n which behaviours of c o p a r t i c i p a t i n g adults are varied along the parameters of f a m i l i a r i t y , a ttention and c o n t r o l , and children's tendency to use synta c t i c or nonsyntactic constructions i n each of these d i f f e r e n t i n t e r a c t i o n conditions i s measured. 5 Chapter One: Review of Current Issues  Influence of Caretakers' Speech on Children's Language Development U n t i l recently, language a c q u i s i t i o n was assumed to occur .^primarily through the operation of a powerful innate "Language A c q u i s i t i o n Device". It seemed that adult speech was so rapid, complex and d i s f l u e n t that an inexperienced language learner could not possibly extract the necessary grammatical concepts from i t without a b i o l o g i c a l p r e d i s p o s i t i o n for language learning. Over the l a s t decade, t h i s assumption has been challenged i n a great number of studies (see Snow 1977a and 1979 for a discu s s i o n ) . A major find i n g i s that adults speak d i f f e r e n t l y to ch i l d r e n than they do to one another. These studies "revealed that the speech addressed to ch i l d r e n aged 18 to 36 months was much simpler and much more grammatical than the speech addressed to a d u l t s " (Snow, 1979, p. 364). Furthermore, speech adjustments that r e s u l t i n a s i m p l i f i e d language environment are not produced only by mothers, but also by men, women who are not mothers, and c h i l d r e n as young as four years who i n t e r a c t with one-and-a-half to three year-olds. We can assume that, at le a s t within a middle-class EuroAmerican culture, such speech i s a v a i l a b l e to language learning c h i l d r e n . As Cross pointscout, "most, i f not a l l , young c h i l d r e n have access to a highly s p e c i a l i z e d speech r e g i s t e r that i s s e n s i t i v e to t h e i r communicative immaturity" (1977, p. 152). The c h a r a c t e r i s t i c s of caretakers' speech to ch i l d r e n have been enumerated i n d e t a i l elsewhere (see Andersen 1975, Berko Gleason &.:.  Weintraub 1978; i€,£ossri sl977);7) . SnOwcvsuccinetlylsiinimaxizesci-hetfiindings -. 6 of the d e s c r i p t i v e studies: The broad outlines of mothers' speech to c h i l d r e n — t h a t i t i s simple and redundant, that i t contains many questions, many imperatives, few co- or subordinations, and few d i s f l u e n c i e s , and that i t i s pitched higher and has an exaggerated intonation p a t t e r n — a r e quite well established. (1977a, p. 36) Thus i t appears that a major r e s u l t of caretaker speech modifications i s a simple, well-formed, redundant input language. Does t h i s s p e c i a l i z e d caretaker r e g i s t e r i n fact help c h i l d r e n acquire language? There i s some evidence that the syntactic structure of caretaker output i s not ne c e s s a r i l y finely-tuned to children's l i n g u i s t i c l e v e l s (Ervin-Tripp & M i l l e r , 1977; Gelman &. Shatz, 1977; Newport, Gleitman & Gleitman, 1977), and that parents don't modify t h e i r t h e i r speech for the purpose of teaching language (Newport et a l . , 1977). Cross f a i l e d "to f i n d s i g n i f i c a n t differences i n the syntactic q u a l i t y of the inputs to groups of ch i l d r e n acquiring syntax at d i f f e r e n t r a t e s " (1978, p. 207), so her r e s u l t s do not support the view that children who are making rapid progress receive "inputs that are s y n t a c t i c a l l y simpler and better formed than children developing more slowly" (1978, p. 207). Snow reports that " no high c o r r e l a t i o n e x i s t s betweenithel.child.'jsllinguistic lev el:; and .most measures of the syntactic complexity of maternal input" (1979, p. 371). So i t appears that fine-tuning of input language syntax to children's developmental l e v e l s i s not a c r u c i a l factor i n children's rate.of. syntactic a c q u i s i t i o n . But while fine-tuning may not be necessary, the general s i m p l i f i c a t i o n t y p i c a l of caretaker speech might s t i l l have the e f f e c t of bringing the structure of language within children's processing range (Cross, 1978). 7 :~. Some inves t i g a t o r s have claimed that ' d e f i c i e n t ' input can have negative e f f e c t s on children's language development. There i s evidence that input i n s u f f i c i e n t l y adapted to the complexity l e v e l c h i l d r e n can process can hinder language development (Hess & Shipman, 1965). Nelson (1973) suggests that language a c q u i s i t i o n may be slower i f the input i s not matched to children's cognitive s t y l e . And Newport, Gleitman and Gleitman (1975) suggest that maternal u n i n t e l l i g i b i l i t y i s negatively correlated to children's rate of l i n g u i s t i c development. While the r e l a t i o n s h i p between synt a c t i c adjustments i n the input language and children's rate of s y n t a c t i c a c q u i s i t i o n i s equivocal, a f i r m r e l a t i o n s h i p between semantice and discourse c h a r a c t e r i s t i c s of the input language and children's rate of l i n g u i s t i c development has been proposed i n a number of studies. Adults t a l k i n g to young chil d r e n greatly r e s t r i c t the semantic content of t h e i r speech. "Mothers l i m i t t h e i r utterances to the present tense, to concrete nouns, to comments on what the c h i l d i s doing and on what i s happening around the c h i l d " (Snow, 1979, p. 370). Adults also use c e r t a i n discourse devices more frequently when t a l k i n g to c h i l d r e n , such as attention-getters, probes as to the effectiveness of the conversation (Snowy, 1977a), expansions, and r e p e t i t i o n s (Cross, 1977; Newport et a l . , 1977). Snow points out that "semantic i n t e r p r e t a b i l i t y and relevance are the c r u c i a l features i n f a c i l i t a t i n g , language a c q u i s i t i o n " (1979, p. 375), a point of view which i s also supported by Cross (1978). The r e l a t i o n s h i p between s p e c i f i c discourse devices and language l e v e l has also been 8 discussed i n studies by Cross (1977; 1978) and Newport et a l , (1977). Newport et a l . propose that " d e i c t i c usage might help b u i l d vocabulary, expansions might help b u i l d syntax and r e p e t i t i o n might influence both to the extent that i t could allow rehearsal or comparison among forms" (1977, p. 129), and they c i t e r e s u l t s which p a r t i a l l y support t h i s p o s i t i o n . Another important c h a r a c t e r i s t i c of caretaker speech i s that references tend to be l i m i t e d to the here-and-now. Cross notes that "the vast majority of the expressions the c h i l d hears encode events that are perceptually, c o g n i t i v e l y and semantically a v a i l a b l e and s a l i e n t to the c h i l d " (1977, p.169). While no causal r e l a t i o n s h i p between caretakers' semantic or discourse c h a r a c t e r i s t i c s and childeen^s rate of l i n g u i s t i c a c q u i s i t i o n has been c l e a r l y established, evidence that s p e c i f i c c h a r a c t e r i s t i c s have a f a c i l i t a t i v e e f f e c t continues to accumulate (for example, Newport et a l . ' s (1977) c o r r e l a t i o n of-mecher's yes/no questions with children's use of a u x i l i a r i e s , and of mothers' d e i x i s with children's noun i n f l e c t i o n s ) . While the causal r e l a t i o n s h i p has not been proven, i t i s c l e a r that adults' semantic, pragmatic, discourse, and possibly syntactic.modifications can have long term influences on children's language learning. Long term influences are generally measured i n terms of a growth score for the c h i l d (from language maturity at time 1 to language maturity at time 2 )as correlated with c h a r a c t e r i s t i c s of mother ':s language at time 1 . The r e l a t i o n s h i p of mothers' speech to rate of a c q u i s i t i o n i s summarized by Cross: 9 Mothers of accelerated c h i l d r e n provided an input that contained greater proportions of expansions, expansion-like utterances, and semantic extensions than did mothers of c h i l d r e n developing more slowly. They also used more p a r t i a l r e p e t i t i o n s and sequences which combine r e p e t i t i o n s with expansions and extensions of the c h i l d ' s preceding utterances. They produced far fewer utterances ""./that were new to the discourse. Their speech appeared to perceptually more s a l i e n t and analyzable; i t contained fewer u n i n t e l l i g i b l e and d i s f l u e n t utterances and fewer run-on sentences, a lower l e v e l of preverbal complexity, and fewer utterances per conversational turn. (1978, p..214) As well as the r e l a t i o n s h i p between input language features and rate of language a c q u i s i t i o n , another question of i n t e r e s t has been that of the r o l e of children's feedback i n e l i c i t i n g modifications c h a r a c t e r i s t i c of the caretaker r e g i s t e r or l i n g u i s t i c s t y l e (see von Raffler-Engel & Rea, 1980 for a discussion and review of recent research on t h i s t o p i c ) . There has been an increasing awareness that c h i l d r e n play an a c t i v e r o l e i n language a c q u i s i t i o n , and t h i s has led to widespread support for the " i n t e r a c t i o n hypothesis", a view that c h i l d r e n and adults influence each other's immediate speech s t y l e within the moment-by-moment unfolding of discourse (Berko Gleason, 1977; Lieven, 1978; Snow, 1979). However, features of caretaker•— c h i l d i n t e r a c t i o n s that have a s p e c i f i c immediate e f f e c t on children's output within the discourse have not been i s o l a t e d from the more general inventory of modifications c h a r a c t e r i s t i c of such i n t e r a c t i o n s . I d e n t i f i c a t i o n of these features -the goal of t h i s study - may be an e s s e n t i a l step on the way to understanding long term e f f e c t s of verbal i n t e r a c t i o n s on language development. Context and Language Functions In t h i s section "context"..will be defined, i t s r e l a t i o n s h i p to semantic and syntactic categories discussed, and i t s influence on 10 language a c q u i s i t i o n considered. The r e l a t i o n s h i p between context and language use w i l l also be reviewed, along with some s p e c i f i c examples of language functions used by c h i l d r e n . Children's utterances are imbedded i n context, so to analyze the complexity of constructions, how they evolve, and whether an.adult coparticipant's behaviour influences them, i t i s e s s e n t i a l to review t h i s l i t e r a t u r e . S i m i l a r i l y , language functions are expressed v i a more or le s s complex constructions depending on context and other coparticipant-based factors which I hope to elu c i d a t e i n t h i s study. Context includes the immediate physical environment and the>:verbal environment of the communicative act, as well as the s o c i a l and psychological world i n which the language user i s operating (Ochs, 1979). Each of these l e v e l s of context simultaneously influences speakers and hearers, a f f e c t i n g the formulation as well as the i n t e r p r e t a t i o n of t h e i r communicative acts. Aspects of the immediate physical environment or s e t t i n g , which a f f e c t communication include the i d e n t i t y and c h a r a c t e r i s t i c s of the speaker, hearer, and co-present others, the l o c a t i o n , the a c t i v i t y , and the inanimate objects present. These factors f i l t e r e d through the perception or world views of the conversational p a r t i c i p a n t s i n d i r e c t l y influence the discourse. The conversation i t s e l f provides the verbal environment, or l i n g u i s t i c context, which i s also f i l t e r e d through the perceptions of each speaker-hearer. Aspects of the verbal environment that influence communicative acts include choice of code (language, d i a l e c t , r e g i s t e r ) , choice of 11 speech event (joke, interview,eetc.), and l o c a t i o n of the utterance within the discourse ( t o p i c - i n i t i a l , response, etc.) (Ochs, 1979). The s o c i a l and psychological world of the language user functions as context. Successful communication involves mutual understanding of conventional routines and depends on the speaker-hearer's a b i l i t y to extract the underlying s o c i a l meanings of utterances. Communication depends on both w i t h i n - s i t u a t i o n context (such as the speaker-hearer's awareness of status r e l a t i o n s h i p s and r o l e s of cop a r t i c i p a n t s , f a m i l i a r i t y with politeness routines, e t c . ) , and e x t r a s i t u a t i o n a l context ("awareness of and assumptions about objects, events, and states of a f f a i r s outside the i n t e r a c t i o n a l s e t t i n g , and how these a f f e c t how language i s used and understood" (Ochs, 1979, p.-,4)). Shields emphasizes the importance of what she c a l l s "interpersonal cohesion" which involves: some conception, however elementary, that the perceptual and i n t e r a c t i o n a l f i e l d i s shared, that previous experience can be remembered and brought into play, and there i s some agreement •. '. onawhat is.:hereicalled.:the;;latent;.contextiwhichr.fr.amescthe^......::'.^.\. situation.," (1978, p. 317) Each of these v a r i a b l e s plays a r o l e i n comprehension and production of utterances within conversations. Ochs sums up the breadth of influence of context: It includes, minimally, language users' b e l i e f s and assumptions .-....•about .temporal,- .spatial.,;.aandcsocialjsettings;:'.prior.,. ongoing.,. and future actions (verbal, nonverbal), and the state of knowledge and attentiveness of those p a r t i c i p a t i n g i n the s o c i a l i n t e r a c t i o n at hand. (1979, p. 5) Furthermore, because context i s an i n t e g r a l part of every speech act, utterances cannot be extracted from t h e i r context for analysis 12 without a d i s t o r t i o n or los s of information, (cf. Bruner, 1975) How i s context, i n a l l i t s forms, rel a t e d to s p e c i f i c dimensions of l i n g u i s t i c competence such as syntax or •semantics? Lieven (1978) notes that before assigning a young c h i l d ' s utterance to any p a r t i c u l a r syntactic or semantic category, i t i s always necessary to know the context i n which that utterance occurs. Adults depend very much on context i n i n t e r p r e t i n g or analyzing children's utterances. Moerk (1977) claims that there i s a close interdependence of communicative structures and objective environmental structures. Ochs (1979) adds that the set of shared background assumptions a f f e c t s every dimension of language, including the syntax, semantics, l e x i c o n , and phonology. The a b i l i t y to interpret the context and to adjust utterances to f i t the context i s an i n t e g r a l part of communicative competence. L i n g u i s t i c and s i t u a t i o n a l context can be al t e r e d i n ways that influence children's a c q u i s i t i o n of communicative competence. I t i s c l e a r that caretakers l i m i t t h e i r t a l k to c h i l d r e n to s i t u a t i o n s which are a c c e s s i b l e to the c h i l d r e n ; that i s , the "here-and-now" which i s a minimally opaque context (Bruner, 1978). This i s p a r t i c u l a r l y true when the c h i l d i s very young. In a study of a mother's speech to her c h i l d between eight and twenty weeks of age, Sylvester-Bradley and Trevarthen (1978) found that 83.4% of a l l the references i n the mother's speech were to the c h i l d , or to the c h i l d ' s body, clothes, or general psychic state. The mother r e f l e c t e d back the baby's own state, moods, f a c i a l expressions, and actions as they happened. Snow (1977b) proposes that changes i n mothers' input language i s 13 r e l a t e d to babies' increasing a b i l i t i e s to function as conversational partners. For whatever reason caretakers encode t h e i r own and t h e i r children's behavior as i t i s proceeding, as well as s t a t i c environmental r e l a t i o n s h i p s , t h i s has the e f f e c t of r e l a t i n g context to l i n g u i s t i c structures i n a transparent way, and s i m p l i f i e s the task of learning language. Moerk points out that " i n t h i s manner she (the mother) models for her c h i l d the pairing of actions /perceptions/ conceptualizations and t h e i r l i n g u i s t i c codes, while both phenomena hold the a t t e n t i o n of the c h i l d " (1977, p. 179). Bates (1976); views ;.the. r e l a t i o n s h i p Lbetween^cohtext-andllanguage learning from a s l i g h t l y d i f f e r e n t perspective. She states that "the eventual commerce of propositions i s f i r s t c a r r i e d out with an exchange of concrete objects or an i n d i c a t i o n of v i s i b l e events. Words as symbolic vehicles with corresponding referents are then inserted into the prepared performative structures" (p. 72). She f e e l s that " i n i t i a l l y , the c h i l d does not understand or o b j e c t i f y the s p e c i a l v e h i c l e - r e f e r e n t r e l a t i o n s h i p with which adults l i n k words and things" (p. 90), but rather a l l sign r e l a t i o n s are procedures for acting on things or p a r t i c i p a t i n g i n events. That i s , the f i r s t words used are highly dependent on immediate context, e s p e c i a l l y ongoing a c t i o n events. These f i r s t words may be equivalent to what other investigators c a l l sensorimotor morphemes (Carter, 1978) or Phonetically Consistent Forms (Dore et a l . , 1976) which are not used independent of p a r t i c u l a r action or i n t e r a c t i o n contexts. 14 Children acquiring competence at t h i s l e v e l of using words as performatives then move towards using words to convey propositions. I n i t i a l l y construction and i n t e r p r e t a t i o n of these propositions require a great deal of r e l i a n c e on contextual cues; for example, chil d r e n may s i g n i f y topics by gazing at objects about which they are making comments. At a l a t e r step i n development c h i l d r e n wrestle with a more abstract l e v e l of context, that of shared background assumptions and l i n g u i s t i c conventions. They begin to r e a l i z e that a here-and-now context cannot always be assumed, and that coparticipants do not n e c e s s a r i l y have access to a l l the same background information as the speaker. At t h i s l e v e l c h i l d r e n are learning about "presuppositions"; which information they can assume coparticipants know and therefore does not need to be marked l i n g u i s t i c a l l y , and which information needs to be e x p l i c i t l y marked or introduced. As Bates points out: It should be c l e a r at t h i s point that pragmatic factors are not a sort of o v e r l a i d system, applied l a t e i n sentence d e r i v a t i o n and l a t e i n language a c q u i s i t i o n . The topic-comment, presupposition-proposition r e l a t i o n s h i p can be a powerful factor i n explaining the mapping system for any given language, and i n explaining the way that c h i l d r e n acquire the mapping system." (1976, p. 211) Context can also be examined on the basis of i t s r e l a t i o n s h i p to language use. The study of "the systematic use of language i n context" (Ochs, 1979, p. 7) i s known as "pragmatics" (one d e f i n i t i o n of t h i s i l l - d e f i n e d area), and includes i n v e s t i g a t i o n of "the many ways i n which context enters into the expression and understanding of propositions by language users i n a p a r t i c u l a r community" (Ochs, 1979, p. 1). Shugar states that "text i s always systematically r e l a t e d to 15 the context" (1978, p. 228), and points out that i t i s t h i s systematic r e l a t i o n s h i p that makes language learnable. Children learn to use language to get things done. Pragmatic information to be expressed includes the speaker's pragmatic i n t e n t i o n , the status r e l a t i o n s h i p s of c o p a r t i c i p a n t s , the speaker's a t t i t u d e towards information being communicated, the topic and comment, and the presuppositions (that i s , the background conditions necessary for an utterance to be understood). (See Bates, 1976; Bates & MacWhinney, 1979). In communicating an i n t e n t i o n or getting something done, Wells points out that "the same purpose can be achieved i n l i n g u i s t i c a l l y more or l e s s complex ways" (1978, p. 457). For example, a function such as getting someone's attention might be conveyed at a l i n g u i s t i c a l l y simple l e v e l , such as through a nonverbal channel or v i a a s i n g l e word, or i t might be encoded i n a l i n g u i s t i c a l l y more advanced structure such as a two word synt a c t i c utterance or a grammatic complete sentence. Keenan and S c h i e f f e l i n (1976) also discusseattention-getting. In p a r t i c u l a r , they inv e s t i g a t e the means by which c h i l d r e n draw a coparticipant's attention to referents located i n either physical space or i n memory space; To locate a referent i n physical space, c h i l d r e n can use nonverbal and/or verbal behaviours. Nonverbal means include looking at the object, holding i t , reaching for i t , o f f e r i n g i t , pointing at i t , or touching i t . The drawback to using nonverbal attention-getters i s that they are e f f e c t i v e only when the l i s t e n e r i s already v i s u a l l y attending to the speaker. Verbal 16 means include„the>,use;:djf .'.notice* verbs isuch-.as "look," "see," .expressive p a r t i c l e s ("oh," "hey"), d e i c t i c p a r t i c l e s ("that"), and d e s c r i p t i v e or i d e n t i f y i n g noun phrases ("my c a r " ) . Because nonverbal devices depend heavily upon immediate context, lack of adult attention r e s u l t s i n communicative i n e f f i c i e n c y which may motivate the c h i l d to replace nonverbal means with verbal means of drawing attention to objects and events. This factor of adult attention, and i t s r e l a t i o n s h i p to children's constructional complexity, i s one of the factors investigated i n t h i s study. Conversational Constraints and Discourse Devices It i s commonly acknowledged that language learning takes place within conversations (Berko Gleason, 1977; Bruner, 1978; Cross, 1977; Dore, 1979; Garvey & Hogan, 1973; Keenan, 1977; Keenan & S c h i e f f e l i n , 1976; Lieven, 1978; Ochs, Schiefelih.;& P i a t t , 1979; Shields, 1978, 1980; Shugar, 1978; Snow, 1977a, 1977b, 1979). Thus, by examining the nature of conversation, we can make some inferences about how c h i l d r e n learn language. For example, obtaining a coparticipant's attention i s a function fundamental to conversation and the necessity of carrying out t h i s function e f f i c i e n t l y may lead to the emergence of more complex constructions i n children's speech. Other conversational functions such as taking turns, introducing t o p i c s , maintaining topics, and r e f e r r i n g back to previous discourse may also lead to t h i s development, and conversational competence i n these areas w i l l c e r t a i n l y r e s u l t i n increased conversational-" c o n t r o l " as w e l l . 17 Conversation i s r e c i p r o c a l ; adults t a l k i n g to c h i l d r e n both communicate s p e c i f i c informationnand receive information, and the same two-way flow occurs for c h i l d r e n . In order for a conversation to be successful, the p a r t i c i p a n t s must follow turn-taking r u l e s , e s t a b l i s h a j o i n t frame of reference, provide information that i s relevant to t h e i r l i s t e n e r , and e s t a b l i s h some commonality of information. These conversational s k i l l s i n turn depend on the more basic a b i l i t y to simply keep the conversation going. As young c h i l d r e n are not yet p r o f i c i e n t at following the rules of conversation or holding up t h e i r end of the dialogue, the burden of sustaining the conversation f a l l s on the coparticipant adult. The f a c t that caretakers i n our culture treat c h i l d r e n as conversational partners, along with the conversational constraints created by i n t e r a c t i n g with u n s k i l l e d language users, r e s u l t s i n speech adjustments that are thought to be i d e a l f or c h i l d r e n learning language. For instance, Snow (1977b) reports that getting c h i l d r e n to take t h e i r turns i n the conversation i s a primary goal of mothers, and to t h i s end they w i l l follow-up on any conversational opener c h i l d r e n make, f i l l i i n f f o r . . . t h e m i l f necessary, or repeat and a l t e r t h e i r own utterances i n order to e l i c i t a response. She also notes that while mothers w i l l accept almost any infant behaviour as a conversational turn (including yawns, burps, sneezes, and v o c a l i z a t i o n s ) , they become much more stringent i n what they w i l l accept as t h e i r c h i l d r e n get older. Mothers' desire to communicate r e c i p r o c a l l y with t h e i r c h i l d r e n , which underlies t h e i r use of the conversational mode , may well be a c r u c i a l f actor i n l i m i t i n g the topics discussed and the semantic and syn t a c t i c complexity i n mothers' speech. (p. 20) 18 Because caretakers adhere to the turn-taking rules of conversation, "a large proportion of maternal utterances are responses to c h i l d utterances, and almost a l l maternal utterances are d i r e c t l y preceded and followed by c h i l d utterances" (Snow, .1979, p. 372). ^Cross (1977) notes that young language learning c h i l d r e n average j u s t under two utterances per turn. She points out that "the provision of l e s s than two maternal utterances between conversational turns may enhance the perceptual salience to the c h i l d of the l i n g u i s t i c information contained i n any sin g l e utterance" (p. 171). In addition, t h i s r a t i o of mother to c h i l d utterances indicates "that the c h i l d i s being given ample opportunity to p r a c t i c e the s k i l l s that he i s acquiring" (p. 171). Furthermore, the fact that conversing adults and c h i l d r e n are often focused on and ta l k i n g about the same thing at the same time makes i t p a r t i c u l a r l y l i k e l y that c h i l d r e n w i l l hear an adult expression of a meaning j u s t when they are entertaining that meaning. Thus the r e l a t i o n s h i p between context, meaning, and form i s c l a r i f i e d for the c h i l d through conversation. I n i t i a l l y coparticipant adults take on the primary r e s p o n s i b i l i t y for ensuring that conversations continue, and are successful. Adults impose a turn-taking structure on dialogues, ensure that t h e i r utterances are relevant to c h i l d r e n by l i m i t i n g topics to children's immediate environments, and use verbal and nonverbal means to e s t a b l i s h a j o i n t focus of attention. Then as c h i l d r e n begin to take t h e i r turns approprately and acquire other conversational s k i l l s , caretaker speech i s adjusted to complement t h i s increasing competence. At t h i s l e v e l , 19 caretakers use fewer utterances per turn (Cross, 1978) and make extensive use of repair mechanisms when ch i l d r e n have communicative d i f f i c u l t y (Keenan & S c h i e f f e l i n , 1976). They also use questions and r e p e t i t i o n s as discourse devices. Questions can be used to check understanding, to d i r e c t a t t e n tion, or to make a response obligatory thus maintaining a turn-taking structure (Ervin-Tripp & M i l l e r , 1977; Ochs et a l . , 1979). Repetitions can also be used to confirm understanding or to e s t a b l i s h that p a r t i c u l a r information can be considered "given" (Keenan, 1977). T y p i c a l l y , adults control the conversation j u s t enough to maintain children's communicative success, although some adults are not as s k i l l f u l as others. Children master turn-taking early and are adequate turn-takers by eighteen months (Lieven, 1978; Snow, 1977b). The a b i l i t y to e s t a b l i s h j o i n t focus of attention and to introduce discourse t o p i c s , on the other hand, are complex s k i l l s which gradually develop over an extended period of time. As pointed out previously, Keenan and S c h i e f f e l i n (1976) describe the means whereby c h i l d r e n locate a referent i n phys i c a l or memory space; they also diagram a dynamic model for e s t a b l i s h i n g discourse t o p i c s . They represent t h i s as an i n t e r a c t i o n a l process, including a t t e n t i o n - g e t t i n g , adequately clear a r t i c u l a t i o n , i d e n t i f i c a t i o n of referents i n the discourse t o p i c , and i d e n t i f i c a t i o n of semantic r e l a t i o n s between those referents on the part of the speaker, and p o s i t i v e rather than negative feedback from the hearer f o r each of these components (p. 353). Ochs et a l . (1979) extend t h i s work to propositions conveyed through a sequence of two or more utterances. 20 The f a m i l i a r i t y of the adult coparticipant to the c h i l d can influence how successful the c h i l d i s i n getting a topic established. The amount of shared experience i s c r i t i c a l i n t h i s reconstruction process. Someone who spends many hours a day knowing what the c h i l d has been doing can often understand an "out of context" utterance to a much greater extent than an i n v e s t i g a t o r making infrequent v i s i t s . (Keenan & S c h i e f f e l i n , 1976, p. 383) I w i l l return to t h i s discussion of topics i n the next section. An area of discourse where c h i l d r e n are slower to acquire competence i s that of providing information relevant to t h e i r l i s t e n e r . When ch i l d r e n l i m i t t h e i r utterances to events i n the here-and-now, adult coparticipants have a f a i r l y good chance of understanding through contextual cues. However, more wide-ranging topics often cannot be interpreted unless c h i l d r e n are able to t i e t h e i r utterances into p r i o r discourse and c o r r e c t l y separate information which can be presupposed from information that needs to be e x p l i c i t l y expressed. Among the requirements of l o c a l discourse between a turn and i t s sequel are a s e r i e s of obligatory and optional tying r e l a t i o n s . The tying r e l a t i o n s help the l i s t e n e r recognize the r e l a t i o n of what i s new to what was said before. (Ervin-Tripp & M i l l e r , 1977, p. 12) As c h i l d r e n begin to acquire these tying r e l a t i o n s , they tend to f i r s t mark taken-for-granted information through discourse rather than through syritax, because syntactic markers such as r e l a t i v e clauses, d e f i n i t e a r t i c l e s , and r e l a t i v e clause nominalizations are l a t e developments (Keenan, 1977). Through successful r e l a t i n g of present utterances to past and future discourse, commonality of information can be achieved. One way c h i l d r e n learn to produce cohesive discourse i s through producing text c o n j o i n t l y with adults. At the one word stage, the c h i l d "places h i s utterance i n systematic and meaningful r e l a t i o n s to the adult's utterances" (Shugar, 1978, p. 229). Ochs et a l . hold a si m i l a r view point: a c h i l d "may learn how to encode propositions by p a r t i c i p a t i n g i n a sequence i n which she contributes a component of the proposition" (1979, p. 267). They c i t e interrogative-response pairs (which can be reinterpreted as argument-predicate constructions) as a common example of jointly-produced constructions. M A l l i s o n I (holding cookies) What's Mommy have Cookie Cookie! O.K. Here's a cookie for you (1979, p. 265 c i t i n g Bloom, 1973) Ochs et a l . claim that, "What's Mommy have? cookie" i n t h i s example i s an interrogative-response pair i n which A l l i s o n and her mother each contribute a portion of the proposition. It i s c l e a r that c h i l d r e n have much to learn about language, and that they learn a good deal of i t i n conversation, "the immediate and primary context for a c q u i s i t i o n " (Dore, 1979, p. 337). In order to acquire communicative competence, chi l d r e n need to learn more than syntax and semantics; they also need to acquire p r o f i c i e n c y i n discourse. Adult i n t e r a c t i o n v a r i a b l e s , such as amount of attention paid to ch i l d r e n , c o n t r o l of the conversation, and assumptions of shared knowledge following from degree of f a m i l i a r i t y may influence the discourse and 22 thereby children's use or development of language structures. Topic-Comment Structures It has been argued that most of children's f i r s t word combinations have a topic-comment or comment-topic structure. That i s , t h e i r utterances encode propositions which involve mentioning a topic of i n t e r e s t , then making a comment about that t o p i c , or i n other s i t u a t i o n s , making a comment, then i d e n t i f y i n g the t o p i c to which i t r e f e r s . These two-part propositions can occur across utterances and speakers, or within single syntactic utterances. The r o l e that the coparticipant adult plays i n discourse can have a considerable immediate influence on how the c h i l d communicates topics and comments, with possible implications for future development. Discourse consists of e s t a b l i s h i n g topics and sustaining t o p i c s , and the speaker^s goal within discourse i s to make points (MacWhinney & P r i c e , 1980). Keenan and S c h i e f f e l i n define "discourse t o p i c " as "the proposition (or set of propositions) about which the speaker i s either providing or requesting new information" (1976, p.338). In continuous discourse, sequences of utterances may a l l have the same t o p i c , or utterances may integrate claims or presuppositions of immediately p r i o r utterances. When new topics are introduced, or old topics are reintroduced, the discourse becomes discontinuous. In these cases the topics are no longer l i n k e d — t h e y do not draw on the previously established presuppositions—and the attention of the l i s t e n e r has to be redirected. See a discussion of these matters i n Keenan and S c h i e f f e l i n (1976). 23 Topics and comments d i f f e r on the dimension of informativeness,. (Bates, 1976; Bates & MacWhihhey(,:/1979; :6reenf-.ieldl& Zukow, 1978; Keenan & S c h i e f f e l i n , 1976; MacWhinney & P r i c e , 1980). In general, comments convey new or important information, whereas topics encode given, taken-for-granted, or redundant information. That which i s c e r t a i n i s presupposed (the t o p i c ) , and that which provides information reducing the uncertainty among possible a l t e r n a t i v e s i s asserted (the comment) (Greenfield & Zukow, 1978). While comments usually represent new information, and topics represent o l d , or given information, t h i s i s not nec e s s a r i l y so. Bates and MacWhinney point out that "topic s e l e c t i o n i s motivated by some combination of at l e a s t these three i n t e n t i o n s : "giveness, perspective, and salience" (1979, p. 180). When elements of a proposition are of equivalent "giveness," speakers are l i k e l y to select the topic on the basis of either closeness to ego or inherent salience or att r a c t i v e n e s s . Comment s e l e c t i o n , on the other hand, " i s determined by 'newness,' 'distance from ego' (the r e c i p r o c a l of perspective taking), and 'salience'" (Bates &. MacWhinney, 1979, p. 186). At the s i n g l e word stage, c h i l d r e n because theyc.eanoonly^sayoone word at a time, are l i m i t e d to encoding either the topic of the comment i n any utterance. Greenfield and Zukow (1978) state that c h i l d r e n at the si n g l e word stage s e l e c t the element of an event to encode for verbal expression on the basis of the "informativeness p r i n c i p l e . " Comments, as they provide new information, are more informative and thus more l i k e l y to be v e r b a l l y encoded than topics, which can be presupposed. 24 (See Weisenburger ( 1 9 7 6 ) . She points out that c h i l d r e n l e x i c a l i z e constituents ".that are the l e a s t redundant, that i s , those that have not already been mentioned, or those that are not obvious i n the s i t u a t i o n . ) That which i s new i s also more l i k e l y to draw the c h i l d ' s attention. Under the constraints of one word speech, when the c h i l d must encode either the topiccor-the comment, a decision to encode ol d , given information would require suppressing or ignoring the p r i o r i t i e s established by the figure-ground and o r i e n t i n g mechanisms. (Bates, 1 9 7 6 , p . 1 6 0 ) So children's utterances p r i o r to combinatorial speech are more l i k e l y to be comments than topics. The problem with t h i s strategy of s t a t i n g s i n g l e word comments i s that an utterance w i l l be i n t e r p r e t a b l e as a comment only i f the coparticipant i s aware of the unstated topic to which i t r e f e r s . When the comment has the same topic as the utterances i n the immediately p r i o r discourse, or when the topic i s some object or event i n the present s i t u a t i o n to which the c h i l d i s d i r e c t i n g his or her attention, the coparticipant adult w i l l l i k e l y be able to r e t r i e v e the topic. However, comments r e f e r r i n g to wider-ranging events, such as those outside of the: here-and-now, frequently r e s u l t i n communicative f a i l u r e (Bates, 1 9 7 6 ) . Through such communicative m i s f i r e s , the c h i l d becomes aware that further topic s p e c i f i c a t i o n i s necessary. Bates gives the following example: "After repeated f r u s t r a t i o n i n t r y i n g to make a one-word comment work, the c h i l d may be forced to encode material that he previously took for granted (e.g., BALL! BALL! BALL...GIVE!)" ( 1 9 7 6 , p . 1 7 2 ) . Children l e a r n to t o p i c a l i z e information as they become 25 moreoable*to predict the l i s t e n e r ' s needs rather than automatically assuming that t h e i r communicative partner has access to the same presuppositions that they do. As Bates and MacWhinney point out, the c r i t i c a l problem f o r c h i l d r e n i s "learning not to presuppose or take old information for granted when such information i s not obvious to the listeners'" (1979, p. 190). Once the c h i l d becomes aware of the need to specify topics, however, topics and comments begin to compete for verbal expression. T o p i c a l i z a t i o n competes f o r channel access i n order to ensure that the l i s t e n e r has s u f f i c i e n t information for comments to work. Fo c i or comments compete for channel access by v i r t u e of t h e i r inherent importance, novelty or informativeness. (Bates, 1976, p.. 174) What s t r u c t u r a l options are open to the c h i l d ? The c h i l d can encode the comment only, and r i s k misunderstanding on the l i s t e n e r ' s part. Or the c h i l d can forgo expressing new, e x c i t i n g information by encoding the topic only. (Atkinson (1979) o f f e r s evidence that c h i l d r e n do i n fact sometimes state topics with no comment following, presumably f o r the purpose of e s t a b l i s h i n g an object of shared attention.) A l t e r n a t e l y , the comment (or i n some cases, the topic) can be expressed v e r b a l l y , whiletthe topic (or comment) i s indicated through gestural means. As Atkinson explains, however, pointing i s not e f f e c t i v e for references displaced from the here-and-now, or when the l i s t e n e r i s not already attending. Also " i t i s impossible, using gesture alone, for a speaker to make clear whether he i s d i r e c t i n g h i s addressee's .>.;:.'; atten t i o n to some i n d i v i d u a l , some property of an i n d i v i d u a l , or the l o c a t i o n occupied by an i n d i v i d u a l " (Atkinson, 1979, p. 235 c i t i n g Lyons, 1975). 26 Another p o s s i b i l i t y i s encoding topics and comment through dialogue. Bruner states that: an enormous amount of the communicaiton i s given to the managing of j o i n t a t t e n t i o n : achieving a common a t t e n t i o n a l focus and achieving some elaboration on the focus that begins i n c r e a s i n g l y toward the end of the c h i l d ' s f i r s t year to be i n the form of j o i n t topic-comment str u c t u r i n g . (1978, p. 252) Either the coparticipant adult can follow-up the c h i l d ' s topic with a comment, or the c h i l d ' s comment can re f e r to the presuppositions of the adult's immediately p r i o r utterance (which then acts as a t o p i c ) . Both s i t u a t i o n s r e s u l t i n a j o i n t l y produced topic-comment structure. Greenfield and Zukow (1978) c i t e an example of a j o i n t l y produced topic-comment structure: Child Adult TOPIC What are you doing with your shoe? COMMENT Off. (p. 291) In t h i s example, the adult's question functions as a topic, and the c h i l d ' s utterance functions as a comment on t h i s t o p i c . Children can also encode topics and comments across discourse i n successive s i n g l e word utterances. These sequences may but do not ne c e s s a r i l y , include conversational i n t e r j e c t i o n s by the coparticipant adult. This p a r t i c u l a r strategy i s discussed by Atkinson (1979), Bates (1976), Bloom (1976), Greenfield and Zukow (1978), Lock (1980), and Scollown (1976; 1979). Often t h i s type of topic-comment (or 27 comment-topic) structure makes i t s appearance because of pr i o r communicative f a i l u r e . In Bates' (1976) previously mentioned example, the c h i l d who started with a one word comment ("Ball.'..') f i n a l l y produced successive single-word utterances ("Ball. B a l l . B a l l . Give.") of the form comment-topic i n order to get the idea across. In other instances, ch i l d r e n may mention a to p i c , then wait for i n d i c a t i o n s from adults that they are attending and understand the reference, before proceeding with the comment. Atkinson (1979) provides an example of t h i s . Child Adult TOPIC Mummy. Mummy? COMMENT Gone, (p. 236) As Atkinson phrases i t , a p l a u s i b l e candidate for the function of the i n i t i a l mummy of the c h i l d i s that of drawing the father's attention to the i n d i v i d u a l , and only when the c h i l d gets some feedback to ind i c a t e that h i s addressee i s s u i t a b l y attending does he go on and predicate something of mummy. (p. 236) Bloom gives a number of examples allAll.isQnisoiSpe.ecn ofcsueeessivenslngle word utterances i n which no adult utterances intervene: " j u i c e , more," "door, open" (both topic-comment st r u c t u r e s ) , and "more, j u i c e " (comment-topic) (1976, p.. 46). The strategy of producing successive s i n g l e word utterances which express both the topic and the comment of a proposition across discourse, 28 then, represents a more advanced communicative l e v e l than the previously mentioned s t r a t e g i e s . I t i s more advanced because both topic and comment can be expressed, rather than j u s t one or the other, and because with t h i s strategy the c h i l d i s no longer dependent on i n e f f i c i e n t nonverbal means of conveying topics or comments. In addition, the c h i l d does not have to r e l y on the coparticipant adult to provide half of the proposition as i n j o i n t topic-comment structures. An argument can be made that i t i s p r e c i s e l y the communicative requirement of s p e c i f y i n g both topic and comment that propels c h i l d r e n from s i n g l e word speech into producing t h e i r f i r s t multiword constructions, the successive s i n g l e word utterances l a b e l l e d " v e r t i c a l constructions" by Scollon (1976). I have used his term to r e f e r to these constructions throughout t h i s paper. A final'; way that c h i l d r e n express topics and comments i s within the bounds of a s i n g l e utterance. These two word (or more than two word) utterances are often viewed as the f i r s t s yntactic utterances, because by producing two words together as a u n i t , the c h i l d formally recognizes the s t r u c t u r a l r e l a t i o n s h i p between them. This formalization represents a more advanced l e v e l of expressing topics and comments than producing successive s i n g l e word utterances. Some evidence for t h i s i s that two word syntactic utterances appear l a t e r than v e r t i c a l constructions (Lock,, 1980; Scollon, 1976), and i n i t i a l l y are produced only with a struggle. Scollon (1976), i n h i s analysis of Brenda's speech, gives several examples showing that her struggle to produce her f i r s t two word utterances often resulted i n phonetic s i m p l i f i c a t i o n s to the point of u n i n t e l i g i b i l i t y . Other evidences of d i f f i c u l t y that he c i t e s 29 are the great number of r e p e t i t i o n s associated with early s y n t a c t i c (horizontal) constructions, and the f a c t that they are often embedded in v e r t i c a l constructions. Scollon (1976; 1979) gives a number of examples of early syntactic (horizontal) constructions, for example "Ron t a l k " and "drink soup" (p. 167). Once c h i l d r e n begin putting two words together i n a s i n g l e syntactic utterance, how w i l l they order the topic and comment?''' This question has been addressed by Atkinson (1979), Bates (1976), Bates and MacWhinney (1979), and Greenfield and Zukow (1978). Atkinson (1979) takes the p o s i t i o n that children's utterances automatically assume the order topic-comment. He explains t h i s on the basis of the "naturalness" of f i r s t i d e n t i f y i n g what you are going to t a l k about, then mentioning what you are going to say about i t . He implies that putting the topic i n f i r s t p o s i t i o n may be a u n i v e r s a l ordering p r i n c i p l e across languages. This whole question has been widely discussed i n the general l i n g u i s t i c s l i t e r a t u r e . Greenfield and Zukow (1978) concur with Atkinson to a c e r t a i n extent. They claim that a c h i l d ' s sensorimbtor.'Structur:ing:of"aneeveht--has a topic-comment order. They explain t h i s to mean that the most basic and primary form of a topic i s an e n t i t y , whereas the most basic form '''Note that early two word utterances are not n e c e s s a r i l y of the form topic-comment or comment-topic. The c h i l d may just make a two word comment without e x p l i c i t l y s t a t i n g the t o p i c , or use a two word utterance to e s t a b l i s h a t o p i c , without including a comment. .See Lock (1982) for a discussion. Note also that there .isi-some disagreement about whether two word topic-comment structures are s y n t a c t i c . Givon (1979) refe r s to these as "loose pragmatic structures" as opposed to " t i g h t syntactic structures,:" See Givon for further discussion of .this point. 30 of a comment i s change of state. As a change of state cannot be perceived without f i r s t perceiving the e n t i t y that i s undergoing the change, the basic order (of how c h i l d r e n view events, or act out events) i s topic-comment. However, they point out that informativeness has an e f f e c t on how topics and comments are ordered at a l i n g u i s t i c l e v e l . That which i s more informative i s l i k e l y to be stated f i r s t . Greenfield and Zukow suggest that topic-comment sequences w i l l be more common than comment-topic sequences, as t h i s i s the basic cognitive order, but that comment-only utterances w i l l occur more often than topic-only utterances, 2 because they are more informative. Bates points out that there are two c o n f l i c t i n g demands on the c h i l d producing two-word utterances: 1. Encode the comment f i r s t , isomorphic r e l a t i o n to i t s attention value. 2. Suppress the attention p r i o r i t i e s and encode the topic f i r s t , to prepare the l i s t e n e r f or the comment. (1976, p;' 175) Semantic or syntactic constraints of the language (e.g., agent then action) can also influence ordering of elements within utterances. However, Bates and MacWhinney (1979) suggest that unless these constraints are c l e a r l y evident i n the adult input language, early ordering s t r a t e g i e s are more l i k e l y to be based on pragmatic f a c t o r s . They state topic i n i t i a l i z a t i o n i s presumably based on recognition of the l i s t e n e r ' s needs, while comment i n i t i a l i z a t i o n i s based on the salience and/or newness of information from the c h i l d ' s perspective. Hence we can 2Note that Greenfield and Zukow are p r i m a r i l y interested i n s i n g l e word utterances, and successions of s i n g l e word utterances. Thus t h e i r comments may not be f u l l y applicable to two word synt a c t i c constructions. 31 predict that the e a r l i e s t pragmatic ordering w i l l beccomment-topic. Later, when the c h i l d r e n have become aware of the need to a c t i v e l y specify topics for the l i s t e n e r , they may switch to a topic-comment ordering. (p. 190) Bates and MacWhinney found that the r e s u l t s of t h e i r research supported these p r e d i c t i o n s . When c h i l d r e n f i r s t encode a proposition, they comment-front, but i f there i s a communicative m i s f i r e , the c h i l d r e n often reorder the elements to produce a topic-comment structure. Thus, because of communicative m i s f i r e s , the c h i l d becomes aware that the l i s t e n e r ' s perspective i s d i f f e r e n t from h i s or her own and abandons the comment-fronting i n order to produce topic-comment ordered utterances, which are more appropriate to the l i s t e n e r ' s needs. In order to do t h i s , the c h i l d needs to learn to suppress the tendency of " b l u r t i n g out the novel or i n t e r e s t i n g information f i r s t , and adding other units on i n decreasing order of i n t e r e s t " (Bates, 1976, p. 174). I have argued here that learning to express topics as well as comments represents an important stage of children's communicative development. While topic s e l e c t i o n often j u s t involves topic continuation (Bates & MacWhinney, 1979), chi l d r e n must also learn how to e s t a b l i s h new topics i n discourse. What does t h i s e n t a i l ? B a s i c a l l y the c h i l d has to inform the l i s t e n e r that there i s something that he or she wants to t a l k about, then specify t h i s topic s u f f i c i e n t l y for i t to be i d e n t i f i e d (MacWhinney & P r i c e , 1980). Keenan and S c h i e f f e l i n (1976) have investigated t h i s question of how topics are established and have proposed four p r e r e q u i s i t e steps for e s t a b l i s h i n g a topic i n discourse. 32 Step 1: The speaker must secure the attention of the l i s t e n e r . Step 2: The speaker must a r t i c u l a t e h i s utterance c l e a r l y . Step 3: The speaker must provide s u f f i c i e n t information for the l i s t e n e r to i d e n t i f y objects, etc., included i n the discourse topic. Step 4: The speaker must provide s u f f i c i e n t information for the l i s t e n e r to reconstruct the semantic r e l a t i o n s obtaining between referents i n the discourse t o p i c . (p. 350) Children can succeed at each of these steps i n a v a r i e t y of ways. Es t a b l i s h i n g a topic can be as simple as holding up an object or repeating a previous utterance (Keenan, 1977) or as complex as the summons-answer routines described by Garvey and Hogan (1973). However, chi l d r e n j u s t beginning to use word combinations are l i m i t e d i n the types of topics they can e s t a b l i s h s u c c e s s f u l l y . "Before the age of three, c h i l d r e n experience enormous d i f f i c u l t y i n getting 'real-world' non-situated events, i n d i v i d u a l s , etc., established as a discourse t o p i c " (Keenan & S c h i e f f e l i n , 1976, p. 371). For a more comprehensive discussion of t o p i c s , see Keenan and S c h i e f f e l i n (1976). Children progress through each l e v e l of learning to communicate topics and comments of propositions l a r g e l y through i n t e r a c t i o n with t h e i r caretakers. C l e a r l y , the adult's r o l e i n dialogue has an important influence on t h i s aspect of communicative development, but the mechanics of t h i s i n t e r a c t i o n are not well-defined. Snow (1979) and Cross (1977) both hold that"-/the caretaker i s constrained to t a l k about what the c h i l d t a l k s about, and thus takes the r o l e of responding to c h i l d - i n i t i a t e d t o p i cs. Snow states that "a very common pattern i s for the c h i l d to introduce a t o p i c , and for the mother to make a comment on that t o p i c , or for the c h i l d to introduce a topic and make a comment, and for the mother to then expand that comment" 33 (p. 372). According to Snow and Cross, t h i s tendency of caretakers to follow-up on c h i l d - i n i t i a t e d topics i s what enables them to provide such semantically relevant and i n t e r p r e t a b l e input to c h i l d r e n . Keenan and S c h i e f f e l i n ' s (1976) point of view seems to d i r e c t l y c o n f l i c t with that of Snow (1979) and Cross (1977). They see topic establishment as a gradually acquired s k i l l and emphasize the number of steps the c h i l d has to s u c c e s s f u l l y complete i n order to i n i t i a t e a topic. They also point out the d i f f i c u l t y c h i l d r e n have i n getting most types of topics established, and accord much of the r e s p o n s i b i l i t y for s e t t i n g topics to the adult involved i n the i n t e r a c t i o n . "In many of the i n t e r a c t i o n s between adults and c h i l d r e n , for example, the adult controls the d i r e c t i o n of the conversation by repeatedly i n i t i a t i n g discourse topics which the c h i l d i s then expected Ito respond to" (Keenan & S c h i e f f e l i n , 1976, p..380). The question-answer t a c t i c as employed by adults i s an example of t h i s . So i t i s not c l e a r to what extent c h i l d r e n c o n t r o l the d i r e c t i o n of conversation and while i t seems that c h i l d control may p o s i t i v e l y a f f e c t communicative development, t h i s i s not c l e a r e i t h e r . Another way adult behaviour can influence how c h i l d r e n encode topics and comments stems from the e f f e c t of adults' attentiveness to c h i l d r e n within discourse. Atkinson (1979) points out that c h i l d r e n can e f f e c t i v e l y use nonverbal means of introducing topics i f the coparticipant adult i s already paying attention. Thus c h i l d r e n who i n t e r a c t with attentive adults may experience more communicative success, with p o s i t i v e e f f e c t s for a c q u i s i t i o n . On the other hand, he notes, a gesture combined with a s i n g l e word may not get the idea across to i n a t t e n t i v e adults, so adult i n a t t e n t i o n might act as an impetus for c h i l d r e n to produce a more complex multiword construction. For instance, c h i l d r e n may state a one-word t o p i c , wait for feedback from adults that they are attending then go on to make a comment. However, once c h i l d r e n are capable of producing ho r i z o n t a l constructions,' i n t e r a c t i o n with a t t e n t i v e adults may be h e l p f u l i n that the c h i l d r e n don't have to produce successive s i n g l e word utterances to check whether the adults are attending, but are free to produce multiword syntactic utterances instead. Thus adult a t t e n t i o n may have d i f f e r e n t influences on the discourse at d i f f e r e n t stages of children's development. The f a m i l i a r i t y of coparticipant adults may also influence how c h i l d r e n p a r t i c i p a t e i n the discourse. Keenan and S c h i e f f e l i n (1976) suggest that when an adult l i s t e n e r i s not intimate with a c h i l d , that the c h i l d has to work harder to accomplish each of the steps i n getting a topic established. This seems p l a u s i b l e , as nonfamiliar/ adults are l e s s l i k e l y to share the same set of presuppositions as c h i l d r e n might share with t h e i r parents, and thus would have more d i f f i c u l t y r e t r i e v i n g the c h i l d ' s references or topics. This seems to be the case i n a conversation between Catherine Snow and Meredith, anJ 18-month-old. C: Bandaid E: Where's your bandaid? C: Bandaid E: Do you have a bandaid? C: Bandaid 35 E: Did you f a l l down and hurt yourself? (Mother enters) C: Bandaid M: Who gave you the bandaid? C: Nurse M: Where did she .put i t ? C: Arm (Dore, 1979, p.340) Nonfamiliar adults may also comprehend l e s s of what c h i l d r e n say simply because of lack of f a m i l i a r i t y with these children's phonological systems. With both of these factors operating, c h i l d r e n wishingcto communicate s p e c i f i c information are l i k e l y to use l e s s developmentally advanced st r a t e g i e s of conveying the topic and comment with strangers than they would with f a m i l i a r adults. A f i n a l way i n which adult behaviour might influence children's communicative development i s i n the formulation of j o i n t topic-comment structures, as discussed e a r l i e r . I t can be argued that before children are competent to produce topic-comment propositions v e r b a l l y , they learn about p r o p o s i t i o n a l structure through dialogue, and are given the opportunity to p r a c t i s e encoding propositions j o i n t l y with an adult. These jointly-encoded propositions may be a precursor to v e r t i c a l constructions i n children's speech. In t h i s study, I investigate the influence of adult attention, c o n t r o l of the conversation, and f a m i l i a r i t y on children's communicative behaviour. ;::'—>.- rcivav*. _c L . i i a ^ i ; - c . ^ io.: {.•:•' j...e s e l e c t s * 36 The T r a n s i t i o n to Syntax The appearance of syntactic utterances i n children's speech i s a t t r i b u t e d to a v a r i e t y of developments: higher l e v e l processing (e.g., number of units held i n memory), removal of performance constraints (possibly at the motor-speech l e v e l ) , cognitive development (that i s , from the sensorimotor stage to a higher symbolic l e v e l ) , and semantic development (that i s , children begin..to encode semantic r e l a t i o n s h i p s ) (Bates, 1976; Bloom, 1976; Dore, 1979; Lock, 1980; Scollon, 1976). In considering children's increasing a b i l i t y to encode both topics and comments within propositions, i t also becomes clear that aspects of pragmatic development and children's early use of syntax are interdependent (Bates, 1976). As c h i l d r e n begin, .to communicate more complex ideas and r e l a t i o n s h i p s between ideas, i t can be hypothesized that s i n g l e words are no longer adequate to get t h e i r intended meaning across within discourse, so t h i s semantic and cognitive development could provide the impetus for developing two word synt a c t i c constructions. At the same time t h i s development could become possible because of the removal of previous processing and performance constraints, and the actual expression of syntactic constructions could be f a c i l i t a t e d by the nature of the i n t e r a c t i o n w i t h i n the discourse. The encoding of both topic and comment i n an utterance i s an example of an early s y n t a c t i c construction. Taking an approach based on those of Bates (1976), Branigan (1976), Dore (1979), Lock (1980),.-and Scollon (1976), I argue here, however, that c h i l d r e n do not move d i r e c t l y from i s o l a t e d s i n g l e words to 37 sy n t a c t i c constructions, and that t h e i r language i s already structured p r i o r to the emergence of two word syntactic utterances. Children communicate combinatorial meanings i n propositions using gestures paired with words, and v e r t i c a l c o n s t r u c t i o n s b e f o r e syntax. These constructions are structured but not s y n t a c t i c . They are also highly dependent on the discourse and place a heavy burden of i n t e r p r e t a t i o n on the coparticipant i n the discourse, so are often l e s s e f f i c i e n t . Bates (1976) points out that the syntactic system a r i s e s through an i n t e r a c t i o n between semantics and pragmatics, and suggests that, pragmatic and semantic structures w i l l appear f i r s t i n language development. Through his e f f o r t s to map these structures onto a l i m i t e d , l i n e a r channel,.the c h i l d w i l l gradually acquire syntactic structures and/or transformations as solutions to the mapping problem (p. 162). It can be argued that early two word utterances are not s y n t a c t i c , and that the appearance of grammatical i n f l e c t i o n s and a d u l t - l i k e grammatical categories (noun, verb, etc.) represent the actual beginning of syntax. However, Bloom defines syntax as, "the arrangement of words i n an utterance, r e l a t i v e to each other - an arrangement of words that i s determined by the r e l a t i o n s h i p among them" (1976, p. 37), and G r i f f i t h s (1979) concurs with t h i s point of view. He states that the e a r l i e s t syntactic constructions are two words long, and can be viewed asasentences i f , "a s i n g l e intonation contour spans both words," and i f , "the component words occur independently and i n other combinations" (p.V105). This approach taken by Bloom and G r i f f i t h s , among others, i s the approach I am taking as w e l l . Early two word utterances can be viewed as s y n t a c t i c because the two words are s t r u c t u r a l l y r e l a t e d to each other, and are combined according to rules of the c h i l d ' s grammar. Furthermore, assumptions should not be made about children's language on the basis of adult grammatical categories. How can sequential, or v e r t i c a l constructions be d i f f e r e n t i a t e d from s y n t a c t i c , or h o r i z o n t a l constructions? Scollon (1976) describes four defining c h a r a c t e r i s t i c s of v e r t i c a l constructions. Each word i s grouped under a separate intonation contour, a pause greater than t h r e e - f i f t h s of a second intervenes between each word, each word can occur independently, and the two or more words i n the construction show an appropriate semantic connection. Others who have discussed the phenomena of v e r t i c a l constructions agree with most of these c h a r a c t e r i s t i c s Bloom (1976) and Garman (1979) suggest that each word i n the construction may have r e l a t i v e l y equal stress as w e l l . See Table 1 for a comparison of d e f i n i t i o n s of v e r t i c a l constructions as defined by d i f f e r e n t i n v e s t i g a t o r s . Various types of v e r t i c a l constructions have been categorized. Bloom and Lahey (1978) d i s t i n g u i s h between chained successive utterances where each sUccesive utterance isLpairedcwith_a.:suocessive movement, and w h o l i s t i c successive utterances where utterances are not t i e d to p a r t i c u l a r movements of s h i f t s i n context. Scollon (1976) defines four types of v e r t i c a l constructions on the basis of presence or absence of word r e p e t i t i o n , and intervention of adult utterances. Greenfield and Zukow (1978), and Rodgon (1976) also d i f f e r e n t i a t e between sequences i n which an adult speaker intervenes, and those which ^Branigan argues that each word i n the sequence does not have a f u l l terminal p i t c h f a l l . However, he defines h i s sequential s i n g l e word utterances on the basis of a much shorter pause than does Scollon. Table 1 Comparison of De f i n i t i o n s of V e r t i c a l Constructions Bloom (1976) Branigan (1976),: Rodgon (1976) Scollon (1976) Greenfield & Zukow (1978) .'Garman (1979) G r i f f i t h s (1979) successive sin g l e word utterances sequential single word utterances sequences of single word utterances v e r t i c a l constructions (series of one word utterances) word successions successions of holophrases f a l l i n g p i t c h intervening pause terminal intonation contour only on f i n a l word i n sequence pause between 400 & 1100 msec. intonation d i f f e r e n t . than i n "comb inatort--^ i a l speech" intervening pause each morpheme grouped under separate intonation contour pause between words greater than .6 sec. ( ( 1 - 4 sec.) separate intonation contour intervening pause each word has a terminal intonation contour intervening pause separate intonation contour f o r each word usually intervening pause share topic & context encode semantic r e l a t i o n s separate morphemes occur independently appropriate semantic connection between words encodes two aspects of a s i n g l e event unitary topic and"context thematic unity r e l a t i v e l y equal stress on the 2 words utterance-type prosodic features of stress do not usually appear i n inverted order 40 which are i n t a c t . I discuss the categorization of v e r t i c a l construction types i n greater d e t a i l , with emphasis on Scollon's (1976) model i n p a r t i c u l a r , i n the next section. Children's presyntactic utterances can be considered to be l i n g u i s t i c a l l y structured f o r a number of reasons. G r i f f i t h s points outcthat, the two e a r l i e s t ways i n which separately s i g n i f i c a n t components enter into s i n g l e communicative acts are the combination of intonation patterns with sequences of sounds and the combination of a gesture and an utterance (1979, p. 118). So he views children's utterances as being structured i n some sense, even p r i o r to the use of v e r t i c a l constructions. Shugar notes that, "before the c h i l d i s able to put together two or more words spontaneously, he i s already connecting h i s utterances to those of other speakers i n a s t r u c t u r a l l y meaningful way" (1978, p. 250), a point which i s also made by Lock: One-word utterances have no s e n t e n t i a l structure, but they occur within structured acts. In other words, one-word utterances are  structured, but the word i s only a part of a greater structure, the act i n which i t i s produced. (1980, p. 182) Lock considers children's communicative behaviour to be structured from the age of one year and claims that, "the one-word and two-word stages d i f f e r only i n the form i n which that structure i s r e a l i z e d " (1980, p. 182). He r e l a t e s the c h i l d ' s previously acquired a b i l i t y to combine gestures to the learning of syntax: the processes underlying the c h i l d ' s a b i l i t y to combine his gestures i s the same process as that responsible for h i s being able to combine his words; and that s i m i l a r l y the roots of grammar go back beyond the two word period. (1980, p.1181) Bloom (1976) argues that because v e r t i c a l constructions are not 41 s y n t a c t i c , they cannot be considered " l i n g u i s t i c a l l y structured"^ but nevertheless f e e l s that a r e l a t i o n s h i p between a succession of words can be a t t r i b u t e d when they occur without a s h i f t i n t o p i c , and suggests that these successions of s i n g l e word utterances are, "evidence of the c h i l d ' s awareness of r e l a t i o n s h i p s among aspects of the s i t u a t i o n and h i s obvious i n a b i l i t y to code these r e l a t i o n s h i p s l i n g u i s t i c a l l y " (p//54). Branigan (1976) found i n one c h i l d ' s speech that only the f i n a l utterance i n a v e r t i c a l construction has a f u l l terminal p i t c h f a l l , and argues that t h i s shows that the component utterances of v e r t i c a l constructions are not independently planned. Scollon (1976) points out that the phonetic regression which occurs i n v e r t i c a l constructions i n d i c a t e s that words making up constructions are not merely accidental juxtapositions. Bloom (1976), Branigan (1976), Scollon (1976), and G r i f f i t h s (1979), among others, have noted that v e r t i c a l constructions are a phenomenon which appears at the end of the s i n g l e word period, j u s t p r i o r to, and overlapping with, early syntactic constructions. Thus the use of v e r t i c a l constructions can be seen as a t r a n s i t i o n from si n g l e word speech to the use of syntactic constructions. This t r a n s i t i o n i s characterized by f u n c t i o n a l c o n t i n u i t y ( G r i f f i t h s , 1979). Semantic r e l a t i o n s are i n i t i a l l y encoded i n successions of s i n g l e words then ^Unlike Bloom, I do not equate " l i n g u i s t i c a l l y structured" with " s y n t a c t i c j " and as I have argued above, I f e e l that many nonsyntactic sequences of words ( v e r t i c a l constructions) are l i n g u i s t i c a l l y structured. 42 later, as sy n t a c t i c two word utterances (Branigan, 1976; Bloom, 1976; Bloom Su.Lahey, 1978). The c h i l d f i r s t using a syntactic construction i s acquiring a new, l e s s context-dependent form for communicating the same meaning he or she can already express v i a a v e r t i c a l construction. The question of i n t e r e s t i n t h i s study i s which, i f any, adult coparticipant behaviours are l i k e l y to f a c i l i t a t e the c h i l d ' s use of sy n t a c t i c rather than v e r t i c a l constructions, and v e r t i c a l constructions rather than i s o l a t e d s i n g l e words. Can an adult i n t e r a c t i n g with a c h i l d influence the t r a n s i t i o n to syntax? While I examine only the immediate e f f e c t s of p a r t i c u l a r coparticipant behaviours on children's constructional complexity, implications f o r s y n t a c t i c development over the long term are also discussed. The Model Returning to the d e s c r i p t i o n of v e r t i c a l constructions touched on previously, Scollon (1976).^ defines v e r t i c a l constructions as sequences of s i n g l e words which have a semantic connection, but with each word having a separate intonation countour, and being separated by a pause greater than t h r e e - f i f t h s of a second. Each word can also occur independently; that i s , outside of a v e r t i c a l construction (see Table 1). Children begin to use v e r t i c a l constructions a f t e r they are well into the single-word stage, but p r i o r to t h e i r use of ho r i z o n t a l (syntactic) constructions. As t h e i r language develops, they continue §A11 references to Scollon i n t h i s section are to his book, Conversations with a One Year Old (1976), unless otherwise noted. 43 to use v e r t i c a l constructions along side of emerging h o r i z o n t a l constructions. Scollon defines h o r i z o n t a l constructions as follows: the two words^ are grouped under the same intonation contour, a pause of l e s s than t h r e e - f i f t h s of a second intervenes, each word can occur independently, and there i s a semantic r e l a t i o n s h i p between the words. Thus v e r t i c a l constructions are d i f f e r e n t i a t e d from h o r i z o n t a l constructions on the basis of a longer pause between words and terminal intonation contours on each component word (but see Branigan, 1976). Scollon describes four types of v e r t i c a l -.constructions. Type A i s the basic v e r t i c a l construction. It includes two one word utterances i n close succession with s i l e n c e before and a f t e r . They, .are 'no t i l i n k e d i n intonation contour, but there i s a d e f i n i t e semantic connection between the words, and the words together form a s i n g l e speech act. Type B i s l i k e Type A i n a l l ways except that either or both words may be ::.s repeated. Type C d i f f e r s from Type- A only i n that the speech of another speaker intervenes between the two components of the construction. However, neither word i s repeated. Type D i s a combination of Type B and Type C. Any of the words may be repeated, and there i s intervention by another speaker. As Scollon points out, Type D v e r t i c a l constructions are the weakest case and i t i s doubtful whether some Type D's are i n fac t constructions. Type A constructions provide the strongest evidence that the c h i l d i s vScollon, i n defining h o r i z o n t a l constructions, uses "word" and "morpheme" interchangeably.. I:\assume t h i s i s because free.morphemes (words) and bound morphemes ( i n f l e c t i o n s ) were not yet£differentiated i n h i s subject's speech,, so i n my discussion here, I w i l l only r e f e r to "words" as being components i n constructions. 44 communicating rel a t e d components of a proposition. Scollon also a t t r i b u t e s a developmental order, claiming that Type D constructions are the f i r s t v e r t i c a l constructions to emerge i n a c h i l d ' s speech (and the r e p e t i t i o n s and adult interventions, as well as the frequency of communicative f a i l u r e , show how d i f f i c u l t i t i s for the c h i l d to communicate rel a t e d ideas at t h i s p o i n t ) , while Type A constructions appear l a s t . Scollon (1979','1'p. o222) , 3 o f f.ersothe. f.o.lldw.ing.'diagram: As he says, "the elimination of r e p e t i t i o n i n the production of v e r t i c a l constructions i s a l a t e r development. The freedom from dependency on intervening discourse i s also a l a t e r development" (p. 222). So there i s a developmental convergence on Type A. Scollon claims that a c h i l d ' s a b i l i t y to use v e r t i c a l constructions, and then h o r i z o n t a l constructions, develops i n the context of conversations. The use of r e p e t i t i o n , and the dependence on intervening discourse are examples of how the use of constructions i s embedded i n conversation. Nor does the r o l e of discourse end with the emergence of h o r i z o n t a l constructions. Children continue to use v e r t i c a l constructions as well as h o r i z o n t a l constructions and, i n f a c t , many ho r i z o n t a l constructions are themselves components of larger v e r t i c a l constructions. The following i s an example from Scollon (1979) : Brenda Context wr i t i n g Brenda i s being read a story about a writing boy learning to write, and she turns read dat the pages to get to t h i s p i c ture. (p. 224) (- r e p e t i t i o n ) (+rrepetition) (- discourse) A B (+ discourse) C D 45 While Scollon was one of the f i r s t to i n v e s t i g a t e v e r t i c a l constructions, and h i s d e s c r i p t i o n and c l a s s i f i c a t i o n of t h i s phenomeonon i s u s e f u l , there are c e r t a i n l i m i t a t i o n s to the model he has described. One serious problem i s that Scollon's study involved only a s i n g l e subject, a l i t t l e g i r l named Brenda. Scollon's d e f i n i t i o n s and descriptions of v e r t i c a l and h o r i z o n t a l constructions are based s o l e l y on t r a n s c r i p t i o n s made of Brenda's speech from age 1;0.2 to 2;0.12. Moreover, Brenda was exposed to a v a r i e t y of languages and d i a l e c t s throughout t h i s period, including Standard English, Hawaiian English, Japanese Hawaiian English, Japanese, and Chinese. Therefore i t i s important to ask whether Scollon's findings can be generalized to other children's a c q u i s i t i o n of English. Use of v e r t i c a l constructions cannot be recognized as a normal stage i n language a c q u i s i t i o n on the basis of data from a s i n g l e c h i l d . However, descriptions of a s i m i l a r , i f not i d e n t i c a l phenomenon have been arr i v e d at separately by other researchers though not described i n the same d e t a i l (see Bloom's (1976) successive single-word utterances, and Branigan's (1976) sequential single-word utterances, for example), but again most of these were s i n g l e subject studies. Taken together, there i s good support for v e r t i c a l constructions as a t r a n s i t i o n a l phenomenon, but there i s l i t t l e information about the orderly progression from one v e r t i c a l construction type to another that Scollon suggests. Scollon's study was l o n g i t u d i n a l , covering a period of one year. However, during t h i s year there were two two-month gaps, and one four-month gap during which no data were c o l l e c t e d . Because of _ . these gapsoafe c r u c i a l points.'.ini.iihe.::acq.uisition idf -va.rt.ical and h o r i z o n t a l constructions, i t i s not always possible to trace the development of each type of construction, or to e s t a b l i s h a clear order of a c q u i s i t i o n . In Scollon's data, for example, the development of construction types B to A and D to C appeared to be p a r a l l e l developments. Another important problem i s the lack of a cohesive model. Scollon describes single-word speech, the use of v e r t i c a l constructions, and the use of h o r i z o n t a l constructions as chronological stages, and he b r i e f l y mentions a l a t e r development i n which Brenda uses h o r i z o n t a l constructions as components i n larger v e r t i c a l constructions, but he does not integrate each of these stages within a larger model or lay out a s t r i c t developmental progression from l e s s complex to more complex constructions. He also suggests that there i s some overlap i n the types of constructions used at any point, but does not attempt to r e l a t e the occurrence of s p e c i f i c constructions to s i t u a t i o n a l v a r i a b l e s . Scollon's categorizations meet the c r i t e r i o n of observational adequacy (that i s , Brenda's output i s accurately described), but i t can be questioned whether they meet the c r i t e r i o n of d i s c r i p t i v e adequacy (that i s , of making p r e d i c t i v e g e n e r a l i z a t i o n s ) . His concept of "constructions" and t h e i r d i v i s i o n into various types i s a r r i v e d at as a d e s c r i p t i o n of Brenda's l i n g u i s t i c behaviour; thus the d e f i n i t i o n s are not a p r i o r i but are derived following analysis of the data. I t would be c i r c u l a r to reapply these d e f i n i t i o n s to the same data from which they were derived i n order to explain l i n g u i s t i c behaviours. While a number of d i f f e r e n t i n v e s t i g a t o r s ' r e s u l t s converge on t h i s set of structures, the formulation of a more comprehensive, d e s c r i p t i v e model awaits r e p l i c a t i o n of Scollon's preliminary work, i n t e g r a t i o n with these other f i n d i n g s , and extension to a wider population. In t h i s study, I attempt to f u l f i l l three goals. The main goal, which goes behond Scollon's work, i s to show which of the s i t u a t i o n a l v a r i a b l e s f a m i l i a r i t y , a t t e n tion, and conversational c o n t r o l have an immediate influence on children's constructional complexity. Secondary goals on theyway to t h i s main goal, are to c l a r i f y Scollon's l e v e l s , r e s t a t i n g them i n a s t r i c t e r developmental progression, and to attempt to r e p l i c a t e Scollon's findings on a somewhat larger population. The model includes three types of verbal behaviour, nonparticipating si n g l e words, v e r t i c a l constructions, and h o r i z o n t a l constructions. Each type of construction evolves gradually. The construction types and l e v e l s have been recast below i n a cl e a r ( s t r i c t ) developmental progression, with the l e a s t complex construction type emerging f i r s t followed by increasingly more complex types. While some of the d e f i n i t i o n s are r e c a p i t u l a t i o n s of Scollon's o r i g i n a l d e f i n i t i o n s , the construction types have been r e l a b e l l e d to match the developmental sequence. 48 Level 1: Nonparticipating s i n g l e words. Holophrases with notstrong semantic l i n k s to words preceding or following. May be an i s o l a t e d word, or a word repeated several times (and each r e p e t i t i o n i s counted as a separate speech a c t ) . Level 2: V e r t i c a l Constructions. One word utterances i n succession, which have a semantic connection, but which are not linked under one intonation contour. Type I (Type D)^: V e r t i c a l construction i n which any of the words may be repeated and another person's speech intervenes. Type II (Type C): V e r t i c a l construction i n which neither word i s repeated, but the speech of another person intervenes. Type III (Type B): V e r t i c a l construction i n which either or both words are repeated, but no other speech intervenes. Type IV (Type A): Two one word utterances i n close succession, forming a sin g l e speech act. No r e p e t i t i o n or intervening speech. Type V: Three or more successive one word utterances forming a speech act, and with any combination of r e p e t i t i o n and intervening speech.8 Level 3: Horizontal Constructions. Two or more words with an appropriate semantic connection, and grouped under one intonation contour, with no pauses greater than t h r e e - f i f t h s of a second between words, and with no intervening utterances by other speakers. Words making up the h o r i z o n t a l construction also occur independently elsewhere. Type VI: Horizontal constructions with r e p e t i t i o n s of any p a r t i c i p a t i n g words immediately p r i o r to u t t e r i n g the h o r i z o n t a l construction, except for exact or more complete p r i o r r e p e t i t i o n s . The p r i o r i d e n t i c a l word i s not a r e p e t i t i o n i f i t can be determined from the context that i t i s c l e a r l y part of a separate speech act. Type VII: Horizontal constructions with no p r i o r r e p e t i t i o n s . Level 4: H o r i z o n t a l - v e r t i c a l constructions. V e r t i c a l constructions containing h o r i z o n t a l constructions as elements. Type VIII: H o r i z o n t a l - v e r t i c a l constructions with r e p e t i t i o n s and/or intervening speech. Type IX: H o r i z o n t a l - v e r t i c a l constructions with no r e p e t i t i o n s or intervening utterances. While I am proposing that construction types emerge i n children's Scollon's l a b e l i n brackets. §This type of v e r t i c a l construction may a c t u a l l y appear l a t e r than this.. 49 9 speech i n the order described above, e a r l i e r construction types t y p i c a l l y coexist with l a t e r developments as discussed previously. For example, a c h i l d may use some Type IV constructions, but the majority of h i s or her speech may s t i l l consist of nonparticipating s i n g l e words. The question of i n t e r e s t i s what motivates use of more or l e s s complex constructions. The hypotheses examined i n t h i s study are: Hypothesis 1: The complexity of a c h i l d ' s constructions i n terms of r e l a t i v e numbers of single-word utterances, v e r t i c a l constructions, and h o r i z o n t a l constructions w i l l not be affected by f a m i l i a r i t y c d f the coparticipant adult. Hypothesis 2: The complexity of a c h i l d ' s constructions i n terms of r e l a t i v e numbers of single-word utterances, v e r t i c a l constructions, and h o r i z o n t a l constructions w i l l not be affected by the amount of attention the coparticipant adult gives to the c h i l d . Hypothesis 3: The complexity of a c h i l d ' s constructions i n terms of r e l a t i v e numbers of single-word utterances, v e r t i c a l constructions, and h o r i z o n t a l constructions w i l l not be affected by the amount of c o n t r o l the coparticipant adult has over the discourse. ^While i t i s beyond the scope of the present study, t h i s model of the emergence of multiword speech awaits further t e s t i n g on a larger sample of c h i l d r e n , using a l o n g i t u d i n a l approach to v e r i f y the developmental sequence. Chapter Two;, Method Subjects. Subjects i n t h i s study were four middle-class c h i l d r e n learning English as t h e i r f i r s t language, and t h e i r mothers. Two of the c h i l d r e n were boys and two were g i r l s . They ranged i n age from 1;8 to 2;2 at the time of the pretest session, and from 1;10 to 2;3 at the l a s t session. Three of the c h i l d r e n were f i r s t - b o r n , and the fourth c h i l d had an older brother. The c h i l d r e n were selected from a group of volunteer subjects to p a r t i c i p a t e i n the study on the basis that each of them was j u s t enter.ings the two word stage of language use and i t was determined that they had roughly equivalent language l e v e l s . Language l e v e l s were measured i n a pretest by a) the Reynell Developmental Language  Scale (RDLS), Experimental E d i t i o n , expressive and receptive portions, b) mean length of utterance i n words (MLU) i n spontaneous play, and c) parental report. On the receptive portion of the RDLS, age-equivalent scores ranged from 2;1 to 2;6, and on the expressive portion, age-equivalent scores ranged from 2;0 to 2;4. MLU'1 s ranged from 1.10 to 1.43. Parents of each of the c h i l d r e n reported that t h e i r c h i l d had j u s t begun to put words together to form two word sentences within the previous month. On the basis of these test r e s u l t i t can beoseen that the four c h i l d r e n were at a s i m i l a r stage of l i n g u i s t i c maturity. See Table 2. The mothers of the c h i l d r e n were a l l native speakers of English. Each of the mothers had two or more years of education subsequent to Table 2 Children's Language Levels at Pretest Date Chi l d Daniel J o n i M i l l y a Cory Age at Pretest 1;8.19 1;10.9 2; 1.20 2;2.28 Age at end of study 1;10.1. 1;10.19 2;2.9 2;3.28 Sex .! ; M F F M MLU (words) 1.10b llv-43 1.24 1.37 RDLS Comprehension "A" Raw Score 24 20 25 17 Age equivalent 2;6 2;2 2;5 2;1 Standard Score 3.0 1.5 0.5 -0.5 RDLS Expressive Raw Score 20 22 25 25 Age Equivalent 2;0 2;0 2;2 2;4 Standard Score 1.0 0.9 -0.2 0.2 ahas an older brother. The older c h i l d r e n were f i r s t - b o r n s . MLU calculated on a second sample taken at age 1.;9.10. high school. Three of the mothers were f u l l t i m e parents, and the fourth worked outside the home on a part-time basis. Data C o l l e c t i o n . Each mother-child dyad p a r t i c i p a t e d f i r s t i n a pretest conducted i n t h e i r home which consisted of the administration of a standardized test (the RDLS), c o l l e c t i o n of a language sample, and a parent interview. Subsequent to t h i s , each mother and c h i l d p a r t i c i p a t e d i n two one-hour d a t a - c o l l e c t i n g sessions which were audiotaped and videotaped i n a c l i n i c a l observation room at the u n i v e r s i t y . Each c h i l d ' s i n i t i a l session involved i n t e r a c t i n g with his/her own mother, and the second session involved i n t e r a c t i n g with a stranger, who was the mother of one of the other c h i l d r e n p a r t i c i p a t i n g 52 i n the study. A d u l t - c h i l d i n t e r a c t i o n s were videotaped with the JVC Vidstar 6700 video cassette recording system and sessions were monitored from a separate room v i a closed c i r c u i t T.V. An a d d i t i o n a l audiotaped recording was made using a Sony Mono TC-106 r e e l to r e e l recorder, Ampex 631 quarter inch polyester audiotapes, and a Sony F-500 Dynamic microphone. Each data c o l l e c t i n g session consisted of four twelve-minute segments or "tasks". The remaining twelve minutes of each session were used for i n s t r u c t i n g the adult coparticipant, adjusting equipment, and c o l l e c t i n g a d d i t i o n a l data that was not used i n the f i n a l a n a l y sis. The f i r s t twelve minute segment, Task 1, was designed to e l i c i t a high l e v e l of adult attention to the c h i l d and as natural as possible a d u l t / c h i l d i n t e r a c t i o n and conversation, given the circumstances. Adults were instructed to play with the c h i l d as they normally would. (See Appendix I for the task i n s t r u c t i o n s given to the adult subjects.) During the next twelve minutes, Task 2, a second adult entered the room and interviewed the coparticipant adult, while the c h i l d continued playing with the toys. The purpose of the interview was to e l i c i t a low l e v e l of adult attention to the c h i l d by providing the adult coparticipant with a d i s t r a c t i o n . In each adult's f i r s t session, she was interviewed about her c h i l d ' s developmental h i s t o r y , and i n the second session each adult was asked general questions about r a i s i n g c h i l d r e n . Task 2 was the only portion of the session which involved the presence of a t h i r d person i n the room. During the t h i r d twelve minutes, Task 3, each adult was given a 53 complex toy and asked to v e r b a l l y Instruct the c h i l d how to use i t . This teaching task was designed to e l i c i t high adult control of the conversation. In each subject's f i r s t session, the toy used was a pegboard with various pieces that could be put together to make a v i l l a g e . In the second session, subjects used a magnetic board with v a r i o u s l y shaped coloured pieces that f i t together to make animals.' In the f i n a l twelve minutes, Task 4, the adults were instruc t e d to allow the c h i l d to play f r e e l y with the toys and to follow the c h i l d ' s lead i n play. The purpose of t h i s s i t u a t i o n was to e l i c i t a sample of low adult control of the conversation. Session one, i n which each c h i l d interacted with his/her own mother, and session two, i n which each c h i l d interacted with a stranger both followed the same format. Therefore Tasks 1 and 2 yielded the s i t u a t i o n a l (independent) variables of high and low attention r e s p e c t i v e l y , and Tasks 3 and 4 yielded the s i t u a t i o n a l v a r i a b l e s of high and low con t r o l . Each c h i l d ' s i n i t i a l session with his/her own mother i n i t s e n t i r e t y was the high f a m i l i a r i t y condition, and the second e n t i r e session with a stranger was the low f a m i l i a r i t y condition. Thus a high and a low l e v e l of each of the three v a r i a b l e s of i n t e r e s t — at t e n t i o n , c o n t r o l , and f a m i l i a r i t y — was e l i c i t e d . See Table 3 . T r a n s c r i p t i o n and Coding. Data were transcribed from the audio and videotapes i n a f a i r l y narrow phonetic t r a n s c r i p t i o n , and t r a n s c r i p t i o n s included three l e v e l s of stress and d e t a i l e d contextual notes. When semantically r e l a t e d words occurred i n succession i n the text, utterances 54 Table 3 Summary of Variables by Session and Task Variable Task Session Coparticipant 1 2 3 4 1 :' ..mother f a m i l i a r i t y • high high high high attention high low - -• control - - high low 2 stranger f a m i l i a r i t y low low low low attention high low - • -control high low were considered to be separate i f there was a terminal p i t c h f a l l on each word, and i f there was an intervening pause of at l e a s t t h r e e - f i f t h s of a second. Other c h a r a c t e r i s t i c s that supported a categorization of "separate utterance" included intervention of the coparticipant adult's speech, and primary stress on each word. The l i n g u i s t i c and s i t u a t i o n a l context was also carefully<.considered, to i n t e r p r e t the meaning of utterances and to determine semantic relatedness. If two or more words were judged to be semantically r e l a t e d , they were coded as a construction, and i f judged unrelated, each word was coded as a nonparticipating single-word. The s p e c i f i c c r i t e r i a used i n coding nonparticipating single-words, v e r t i c a l constructions, h o r i z o n t a l constructions, and h o r i z o n t a l - v e r t i c a l constructions are given i n the Model section of Chapter One, page 48. . 55 Chapter Three: Results Analysis i s based on a l l counted utterances f o r each c h i l d . The number of counted utterances ranged from 184 utterances from Joni i n session II to 737 utterances from Daniel i n session I. The mean number of utterances obtained per session was 402. These utterances were assigned to the three l e v e l s of construction complexity defined previously. The few h o r i z o n t a l - v e r t i c a l constructions that occurred were grouped with the ho r i z o n t a l constructions. The maximum amount of data discarded on the basis of u n i n t e l l i g i b i l i t y or n o n l i n g u i s t i c status f o r any c h i l d i n any session was 10%, although as l i t t l e as 1.5% was discarded i n some instances. Data were analyzed using a 2x2x2x3x4 f a c t o r i a l design (high f a m i l i a r i t y vs. low f a m i l i a r i t y X high attention vs. low attention X high control vs low con t r o l X three l e v e l s of construction complexity X 4 children) to determine whether there was a r e l a t i o n s h i p between high and low attention, c o n t r o l , and f a m i l i a r i t y conditions and the children's complexity or frequency of utterances. A computer program ^ was used to form and analyze multiway frequency tables, an approach which involves f i t t i n g a l o g - l i n e a r model to the c e l l frequencies i n order to obtain a d e s c r i p t i o n of the r e l a t i o n s h i p s between the factors of the table. W e l l - f i t t i n g models were obtained by including a l l s t a t i s t i c a l l y s i g n i f i c a n t r e l a t i o n s h i p s . The strength of each BMDP-77 Biomedical Computer Programs: P-Series. Dixon, W.J. & Brown, M.B. (Eds.). Berkeley: U n i v e r s i t y of C a l i f o r n i a Press, 1977. 56 p a r t i c u l a r r e l a t i o n s h i p was then determined by comparing the f i t of a model including that r e l a t i o n s h i p with f i t of one excluding i t , using the l i k e l i h o o d r a t i o c h i square. Table 4 i s a five-dimensional table r e l a t i n g the v a r i a b l e s f a m i l i a r i t y , a ttention, and control to three l e v e l s of construction complexity for four c h i l d r e n . Analyses were ca r r i e d out r e l a t i n g these v a r i a b l e s to the children's construction complexity, and also to t h e i r absolute frequency of contructions. Complexity of Utterances Table 5 gives the number of constructions at each l e v e l of complexity produced by a l l the chi l d r e n i n the high and low f a m i l i a r i t y conditions. The influence of familiar.ityooncchildreri!scconstructio.nccomplexity was not s i g n i f i c a n t , but the trend was for children to use a greater proportion of ho r i z o n t a l constructions to v e r t i c a l constructions with strangers, and a greater proportion of v e r t i c a l constructions to sin g l e words with t h e i r own mothers (see Table 6). The influence of high or low attention i^on the complexity of children's constructions was not s i g n i f i c a n t . Table 7 gives the number of constructions at each l e v e l of complexity produced by a l l the chi l d r e n i n the high and low attention conditions. While ch i l d r e n used considerably more utterances i n the high attention than i n the low attention condition, they did not use a proportionately greater amount of more complex constructions ( v e r t i c a l and horizontal) than l e s s complex s i n g l e word utterances. In the a n l y s i s of the attention v a r i a b l e , a l l the con t r o l data was omitted from the models, and i n the analysis of c o n t r o l , a l l the attention data was omitted. 57 Table 4 Total Number of Constructions Coded for Each C h i l d by Task (High Attention, Low Attention, High Control, Low Control), Session (High F a m i l i a r i t y , Low F a m i l i a r i t y ) and Construction Complexity (1,2,3) High F a m i l i a r i t y Low F a m i l i a r i t y Construction Attention Control Attention Control Child Complexity high low high low high low high low Daniel 1 176 45 158 120 152 63 127 100 2 16 0 7 29 16 2 5 6 3 36 3 27 22 15 6 25 18 Joni 1 85 4 60 78 19 13 28 25 2 8 0 4 11 1 1 2 5 3 27 0 20 33 14 12 19 24 M i l l y 1 56 13 26 51 23 16 23 14 2 12 4 9 6 4 2 3 1 3 42 11 34 36 40 . 47 27 15 Cory 1 34 13 29 34 41 7 30 40 2 3 3 3 4 3 0 2 6 3 64 35 72 75 48 8 40 48 Note: Construction; compiexitycgoes•_from l l a l e a s t o . c o m p l e x o t o ,3.,'omostocomplex, with 1--= s i n g l e words 2 = v e r t i c a l constructions 3 = h o r i z o n t a l and h o r i z o n t a l - v e r t i c a l constructions 58 Table 5 Marginal Table of F a m i l i a r i t y by Construction Complexity: Number of constructions at each l e v e l of complexity produced by a l l the c h i l d r e n for high and low values of the v a r i a b l e . F a m i l i a r i t y Construction Complexity High Low Single Words V e r t i c a l Constructions Horizontal Constructions 982 119 ',537 721 59 406 The influence of high or low control of the discourse by the adult speaker was highly s i g n i f i c a n t (X 2 :' (10.36,2>'. = .0.^ 006 (p<0.01).)':in the d i r e c t i o n predicted. See Table 8, which gives the number of constructions at each l e v e l of complexity produced by a l l the c h i l d r e n i n the high and low control conditions, and Table 6efof models compared to determine strength of e f f e c t . In the high c o n t r o l condition, the c h i l d r e n used seven v e r t i c a l constructions f o r every 100 sin g l e word utterances, and 55 ho r i z o n t a l constructions for every 100 s i n g l e word utterances, whereas i n the low control condition, the c h i l d r e n used 15 v e r t i c a l constructions and 59 ho r i z o n t a l constructions for every 100 s i n g l e -word utterances. Thus the grammatical constructions of the four c h i l d r e n were s i g n i f i c a n t l y ; more complex i n the low adult c o n t r o l condition than i n the high adult control condition. Of note i s the observation that the c h i l d r e n more than doubled t h e i r proportion of v e r t i c a l constructions to s i n g l e word utterances under the low con t r o l condition, compared to the high c o n t r o l condition. 59 Table 6 Strength of E f f e c t s i n the Model to F i t Data on Relationship of F a m i l i a r i t y , Attention, and Control to Children's Construction Complexity Complexity by vari a b l e ( s ) Term of in t e r e s t Models compared df LRX^ f a m i l i a r i t y CF KHAF, KHAC KHAF, KHAC, CF 2 5.51 0.06 at t e n t i o n c CH KHF, KFC KHF, KFC, CH 2 2.89 0.23 control 1- CH KHF, KFC KHF, KFC, CH 2 10.36 0.006 f a m i l i a r i t y , a t t e n t i o n 3 CEH„: KHF, KliC, KF.C KHF, KllCy KFC, CFH. 2 1.83 0.40 f a m i l i a r i t y , c o n t r o l ^ CFH KHF, KHC, KFC KHF, KHC, KFC, CFH 2 0.43 0.80 Note. Abbreviations: H = High - Low A;:= Attention - Control F = F a m i l i a r i t y C-— Complexity K = Chi l d a a t t e n t i o n data only ^ c o n t r o l data only Table 7 ;:--.-.:Mar.gihalbTable of tAttention byaConstructidn .C6mplexityl:.'u:.Number' of constructions at each l e v e l complexity produced for high and low values of the v a r i a b l e Attention  Construction Complexity High Low Single Words V e r t i c a l Constructions Horizontal Constructions 586 63 286 174 12 122 60 Table 8 Marginal Table of Control by Construction Complexity: Number of constructions at each l e v e l of complexity produced for high and low values of the v a r i a b l e Control Construction Complexity High Low Single Words 481 462 V e r t i c a l Constructions 35 68 Horizontal Constructions 264 271 Frequency of Utterances While an important area of i n v e s t i g a t i o n was the proportion of more complex utterances to l e s s complex utterances under various conditions, another question of i n t e r e s t was the o v e r a l l frequency with which chil d r e n produced each type of construction under each condition. One analysis was of the t o t a l number of multi-word constructions (including v e r t i c a l constructions, h o r i z o n t a l constructions, and h o r i z o n t a l - v e r t i c a l constructions) produced by the chi l d r e n i n high versus low f a m i l i a r i t y , a t t e n tion, and c o n t r o l conditions. Table 9 i s the data table analyzed. Table 9 Total Number of Multi-word Constructions ( V e r t i c a l , Horizontal, & Ho r i z o n t a l - V e r t i c a l ) Produced by Each Child i n each Condition. High F a m i l i a r i t y Low F a m i l i a r i t y Child Attention Control Attention Control high low high low high low high low Daniel 52 3 34 51 31 8 30 24 Joni 35 0 24 44 15 13 21 29 M i l l y 54 15 43 42 44 49 30 16 Cory 67 38 75 79 51 8 . 42 54 61 For f a m i l i a r i t y , the r e s u l t s were highly s i g n i f i c a n t (X (32.70, 1) = 0.1 E-07 (p<0.01)) that c h i l d r e n used more multi-word constructions when i n t e r a c t i n g with t h e i r mothers than when i n t e r a c t i n g with strangers. See Table 10 for models compared and strength of e f f e c t 12 values. For attention and co n t r o l , r e s u l t s were also s i g n i f i c a n t (X 2 (80.38, 1) =0.0 (p<0.01)). Across a l l the ch i l d r e n , more multi-word constructions were used i n the high attention than i n the low attention condition, and i n the low control than i n the high control condition. When the frequency of constructions r e s u l t s for high and low attention and control were examined separately, for mothers and strangers, the 2 same d i r e c t i o n of e f f e c t s was observed, and was s i g n i f i c a n t (X (9.13, 1) = 0.002 (p<0.01)). Anc.analysis of the frequency of ho r i z o n t a l constructions used i n the various conditions (Table 11) showed that for the f a m i l i a r i t y condition, c h i l d r e n used s i g n i f i c a n t l y more ho r i z o n t a l constructions with t h e i r own mothers than with strangers (X 2 (18.26, 1) = 0.2 E-04 (p<0.01)). Strength of e f f e c t s are given i n Table 10. This r e s u l t i s highly s i g n i f i c a n t i n the d i r e c t i o n predicted. The r e s u l t for attention and con t r o l i s also s i g n i f i c a n t (X 2 (47.72, 1) = 0.5 E - l l (p<0.01)). Children used more hori z o n t a l constructions i n the high attention than i n the low attention condition and somewhat more h o r i z o n t a l constructions i n the low control condition than i n the high control condition, as predicted. When the e f f e c t s of attention and cont r o l are examined separately for mothers and strangers, the r e s u l t s are s i g n i f i c a n t ^ 2 I n the analyses of frequency of utterances, attention and control were examined together, thus a s i g n i f i c a n c e l e v e l can only be assigned for these two conditions when considered together. However, the d i r e c t i o n of e f f e c t s for attention and control have been recovered separately. 62 (X 2 (6.81, 1) = 0.009 (p<0.01)) In the d i r e c t i o n s predicted except that with strangers, c h i l d r e n produced more h o r i z o n t a l constructions i n the high control condition than i n the low control condition (opposite d i r e c t i o n of r e s u l t predicted). Table 10... Strength of E f f e c t s i n the Model to F i t Frequency Data: Children's Number of Multi-word, Horizontal, V e r t i c a l , Single Word, and Total Constructions i n F a m i l i a r i t y , Attention and Control Conditions 9 „ . / N term of , . , constructions df LRX Variable(s) . ^  ^ models compared in t e r e s t r counted f a m i l i a r i t y HAK. , HAK, F Multiwd Horiz Vert S-wd Total 1 32.70 1 18.26 1 20.63 1 40.17 1 72.66 0.1E-07 0.2E-04 0.5E-05 0.2E-09 0.1E-16 attention and control HA HFK, AFK HFK, AFK, HA Multiwd Horiz Vert S-wd Tot a l 1 80.38 1 47.72 1 39.76 1 127.14 1 205.78 0.0 0.5E-11 0.2E-09 0.0 0.0 f a m i l i a r i t y , HAF attention and control HAK, HFK, AFK HAK, HFK, AFK, HAF Multiwd Horiz Vert S-wd Tot a l 1 9.13 1 6.81 1 1.72 1 13.69 1 0.002 0.009 0.19 0.2E-03 0.4E-06 Note. Abbreviations: H = High - Low A = Attention - Control F = F a m i l i a r i t y C = Complexity K = Ch i l d 63 Table 11 Total Number of Horizontal Constructions Produced by Each Child i n each Condition . . . . H i g h F a m i l i a r i t y Low F a m i l i a r i t y Attention Control Attention Control C h i l d high low high low high low high low Daniel 36 3 27 22 15 6 25 18 Joni 27 0 20 33 14 12 19 24 M i l l y 42 11 34 36 40 47 27 15 Cory 64 35 72 75 48 8 40 48 I also analyzed the frequency of v e r t i c a l constructions used i n each condition. Table 12 was analyzed. For the f a m i l i a r i t y condition, s i g n i f i c a n t l y more v e r t i c a l constructions were used with mothers than with strangers (X 2 (20.63, .1) = 0.6 E-05 (p<0.01)). For attention and c o n t r o l , r e s u l t s were also s i g n i f i c a n t (X (39.76, 1) = 0.0 E-09 (p<0.01)). These r e s u l t s were i n the d i r e c t i o n predicted for c o n t r o l , but opposite to the d i r e c t i o n predicted for attention. The c h i l d r e n used more v e r t i c a l constructions i n the high attention condition than i n the low attention condition, and more v e r t i c a l constructions i n the low control condition than i n the high control condition. The attention and control r e s u l t s , when examined separately f o r mothers and strangers, are not s i g n i f i c a n t . An examination of the frequency of sin g l e word utterances (see Table 13 and Table 10) showed that c h i l d r e n used a s i g n i f i c a n t l y greater number of s i n g l e word utterances with t h e i r own mothers than 64 Table 12 Total Number of V e r t i c a l Constructions Produced by Each Ch i l d i n each Condition High F a m i l i a r i t y Low F a m i l i a r i t y Child Attention high low Control high low Attention high low Control high low Daniel 16 . 0 7 29 16 2 5 6 Joni 8 0 4 11 1 1 2 5 M i l l y 12 4 9 6 4 2 3 1 Cory 3 3 3 4 3 0 2 6 Tota l Number Produced by Table 13 of Single Word Constructions Each Child i n each Condition High F a m i l i a r i t y Low F a m i l i a r i t y Child Attention high low Control high low Attention high low Control high low Daniel 228 48 192 171 183 71 157 124 Joni 120 4 .84 122 34 26 49 54 M i l l y 110 28 69 93 67 65 53 30 Cory 101 51 104 113 92 15 72 94 with strangers. They also used s i g n i f i c a n t l y more single word utterances i n the high attention condition than i n the low attention condition, but about the same number of sing l e word utterances i n the high and the low con t r o l conditions. From the above r e s u l t s , i t appears that c h i l d r e n may simply be more 65 verbose under c e r t a i n conditions. The influence of f a m i l i a r i t y , a t t e n tion, and control on verbosity was examined by comparing the t o t a l number of constructions of a l l types (including s i n g l e word utterances) that the c h i l d r e n produced i n d i f f e r e n t s i t u a t i o n s . See Table 14. Children produced s i g n i f i c a n t l y more constructions (X (72.66, 1) = 0.1 E-16 (p<0.01)), when i n t e r a c t i n g with t h e i r own mothers than with strangers. They produced more than three times as many constructions i n the high attention condition than i n the low att e n t i o n condition, but only s l i g h t l y more constructions i n the low control than i n the high control condition. The verbosity for attention and control conditions was s i m i l a r when r e s u l t s f o r mothers and strangers were examined separately, except for a s l i g h t trend for c h i l d r e n to t a l k to strangers more i n the high c o n t r o l condition than i n the low con t r o l condition. , Table 14 Total Number of Constructions of a l l Types Produced by Each Child i n each Condition High F a m i l i a r i t y Low F a m i l i a r i t y Attention Control Attention Cont r o l C hild high low high low high low high low Daniel 228 48 192 171 183 71 157 124 Joni 120 4 84 122 34 26 49 54 M i l l y 110 28 69 93 67 65 53 30 Cory 101 51 104 113 92 15 72 94 66 Chapter Four: Discussion In t h i s study, I investigated whether the complexity of children's constructions during the t r a n s i t i o n to multiword utterances i s re l a t e d to adult i n t e r a c t i o n v a r i a b l e s , s p e c i f i c a l l y , whether or not (1) the adult i s attending to the c h i l d , (2) the adult controls or leads the conversation, and (3) the adult i s a stranger or the mother. I defined three l e v e l s of children's utterance complexity, ranging from single words to v e r t i c a l constructions to h o r i z o n t a l constructions (least to most complex), then measured the r e l a t i v e proportions of these construction types i n the d i f f e r i n g adult i n t e r a c t i o n conditions. Given t h i s method of measuring children's speech complexity as a r e l a t i v e proportion of construction types, there i s not s u f f i c i e n t evidence to r e j e c t the hypothesis that the complexity of a c h i l d ' s constructions w i l l not be affected by the f a m i l i a r i t y of the coparticipant adult. In other words, the assumption that 9 c h i l d r e n use a higher proportion of complex constructions with t h e i r own mothers than they do with strangers i s not supported. S i m i l a r l y , the hypothesis that the complexity of a c h i l d ' s constructions w i l l not be affected by the amount of attention the coparticipant adult gives the c h i l d cannot be rejected. We cannot assume that c h i l d r e n form more complex constructions when someone i s paying f u l l a ttention to them than when they f i r s t have to capture someone's attention. On the other hand, the/hypothesis that the complexity of a c h i l d ' s constructions w i l l not be affected by the amount of control the adult 67 has over the conversation can be rejected on the basis of t h i s study. .. My r e s u l t s show that a.xhild uses a s i g n i f i c a n t l y higher proportion of complex constructions when he/she controls the conversation than when the adult controls the conversation. F a m i l i a r i t y The r e s u l t that the f a m i l i a r i t y of an adult i n t e r l o c u t o r was not shown to influence the c h i l d ' s complexity i s i n t e r e s t i n g , given research findings discussed previously: that mothers t a i l o r t h e i r speech for t h e i r c h i l d r e n and that children's s t y l e and rate of language learning may be influenced by t h i s l i n g u i s t i c i n t e r a c t i o n with t h e i r mothers. It seems p l a u s i b l e that c h i l d r e n might use more complex speech with t h e i r own mothers because they share a larger background of knowledge than with a stranger; thus there would be fewer d i f f i c u l t i e s i n s e t t i n g , defining, or maintaining t o p i c s , and because of the reduction i n these conversational pressures, the c h i l d r e n would have more opportunity to form more complex utterances (see Keenan & S c h i e f f e l i n , 1976). Another assumption i s that each mother would understand her own c h i l d ' s a r t i c u l a t i o n and i d i o s y n c r a t i c vocabulary better than a stranger would, and t h i s would reduce performance pressure, enabling the c h i l d to concentrate on forming more complex constructions. We could also suppose that the mother, knowing her ch i l d ' s production c a p a b i l i t i e s better than a stranger would, might provide more openings for the c h i l d to use h i s or her highest l e v e l constructions. 68 The r e s u l t s suggest that these f a c t o r s either have no e f f e c t on chi l d r e n at t h i s developmental stage, or are counterbalanced by competing pressures. For example, perhaps c h i l d r e n , when t a l k i n g with t h e i r mothers, assume that t h e i r mothers understand most of what they say, so they f e e l l i t t l e need to elaborate on t h e i r utterances and produce more complex constructions; whereas with strangers, they may be aware of the lack of shared background knowledge, so they expand t h e i r utterances i n a struggle to achieve understanding. There i s some support for t h i s explanation i n the c h i l d language l i t e r a t u r e . In describing the discourse context of children's syntactic growth, Ervin-Tripp (1977) has t r i e d to i d e n t i f y "bursts of syntax," instances when the c h i l d ' s production goes beyond that predicted by the rules governing preceding texts. In her study, one determinant of such bursts of syntax (s i m i l a r i n an important way to the s i t u a t i o n a l l y elevated construction complexity described i n my study) was "misunderstandings which pressure e x p l i c i t n e s s " (Ervin-Tripp, 1977, p. 17). I t would be i n t e r e s t i n g to discover i n p r e c i s e l y what ways c h i l d r e n at t h i s young age ( 2 - 2 1/2'years) do adjust t h e i r speech to strangers. A l t e r n a t i v e l y , the complexity of children's constructions may not be affected by the f a m i l i a r i t y of the adults they are t a l k i n g to. Perhaps they don't t r y to overcome the b a r r i e r of lack of shared knowledge by using more complex, informative constructions to strangers, nor to overcome the stranger's d i f f i c u l t y i n understanding t h e i r 69 p r o n u n c i a t i o n by u s i n g more e a s i l y u n d e r s t o o d s i n g l e words. And perhaps the s t r a n g e r s i n t h i s study (who were, of c o u r s e , a l s o mothers) were j u s t as c a p a b l e as t h e c h i l d r e n ' s own mothers a t r e d u c i n g communicative p r e s s u r e s , and s t r u c t u r i n g t h e i n t e r a c t i o n so t h a t t h e c h i l d r e n had many o p p o r t u n i t i e s t o use complex c o n s t r u c t i o n s . T h i s p o s s i b i l i t y i s s u p p o r t e d by f i n d i n g s r e p o r t e d i n t h e language i n p u t l i t e r a t u r e ( f o r example, Snow, 1972). The n e a r l y s i g n i f i c a n t (p=0.06) result l.thatc^chmldrencuse p r o p o r t i o n a t e l y more h o r i z o n t a l c o n s t r u c t i o n s w i t h s t r a n g e r s and p r o p o r t i o n a t e l y more v e r t i c a l c o n s t r u c t i o n s w i t h ; . t h e i r own mothers s u g g e s t s t h a t t h e r e may be some d i f f e r e n c e s i n c h i l d r e n ' s speech r e l a t e d t o f a m i l i a r i t y o f t h e a d u l t , however. Do c h i l d r e n p e r f o r m a t t h e i r h i g h e s t l e v e l of e x p r e s s i v e competence w i t h s t r a n g e r s i n an attempt t o achievecoommuriicatiari? Do they use more v e r t i c a l c o n s t r u c t i o n s w i t h t h e i r , mothers because t h i s i s how t h e y l e a r n t o put words t o g e t h e r i n t h e f i r s t p l a c e , and i n t e r a c t i o n w i t h a f a m i l i a r a d u l t b e t t e r f a c i l i t a t e s t h i s l e a r n i n g p r o c e s s ? A more f i n e - g r a i n e d a n a l y s i s .is .needdd.ini o r d e r to. .untangle t h e v a r i o u s i n f l u e n c e s of mothers and s t r a n g e r s on c h i l d r e n ' s c o n s t r u c t i o n a l c o m p l e x i t y . A t t e n t i o n The n o n s i g n i f i c a n t r e s u l t f o r a t t e n t i o n a l s o w a r r a n t s d i s c u s s i o n . As A t k i n s o n (1979) p r o p o s e s , i f an a d u l t i s a l r e a d y p a y i n g a t t e n t i o n , a c h i l d c o u l d e f f e c t i v e l y use a n o n v e r b a l means of i n t r o d u c i n g a t o p i c , t h e n f o l l o w - u p w i t h a s i n g l e word comment, whereas w i t h an i n a t t e n t i v e 70 adult, the c h i l d might be forced to also state the topic i n order to get the adult's attention, thereby producing .a. mor.e::complex.vertical construction. So we could assume that i n a t t e n t i o n stimulates the use of more complex constructions. On the other hand, at a more advanced stage of development, i n t e r a c t i o n with a more attentive adult may eliminate the need to produce successive s i n g l e words i n order to check that the adult i s l i s t e n i n g , thus freeing the c h i l d to produce a more complex ho r i z o n t a l construction. So i n t h i s way, increased attention could stimulate the use of more complex constructions. Neither of these suppositions i s supported by the r e s u l t s of t h i s study. This may be due to lack of s e n s i t i v i t y i n the measures (for example, opposite influences may have cancelled each other out, producing no e f f e c t ) , or coparticipant attention might not have a s i g n i f i c a n t influence on children's constructional complexity. For example, adult comprehension of the c h i l d ' s utterance, or the degree of comprehension that the c h i l d assumes the adult has, rather than adult a t t e n t i o n , may be the s i g n i f i c a n t factor that would account for Atkinson's observations. In t h i s study, the p a r t i c u l a r s i t u a t i o n that was used to obtain adult i n a t t e n t i o n may have reduced the e f f e c t of attention on construction complexity by d i s t r a c t i n g the adults so much that the children's usual ploys for a t t r a c t i n g a t t e n t i o n went unnoticed. The women seemed to view the interviewing s i t u a t i o n as important, so many of the children's s i n g l e word attention-getters were ignored, or responded to nonverbally. 71 When the adult subject did turn to the c h i l d , for example at a break i n the adult conversation, I observed that the ch i l d r e n tended to "b l u r t out" more complex constructions. This suggests that an analysis of the d i s t r i b u t i o n of construction types, rather than j u s t the o v e r a l l frequency, would be an i n t e r e s t i n g analysis to perform on these data. Also, most of the c h i l d r e n seemed to give up on conversing with the adult a f t e r a few unsuccessful attempts, so that there was much les s t a l k i n g i n t h i s s i t u a t i o n . Of the utterances that were produced, many were play monologues. Control The r e s u l t s of t h i s study, s i g n i f i c a n t to p = 0.005, showed that the c h i l d r e n used proportionately more complex constructions i n the low adult control s i t u a t i o n than i n the high adult control s i t u a t i o n . This suggests that when c h i l d r e n are permitted to "lead" or " c o n t r o l " the d i r e c t i o n of an i n t e r a c t i o n while the adult acts as a f a c i l i t a t o r , more complex language i s l i k e l y to emerge than i n structured play dominated by the adult. The notion of " c o n t r o l " i s d i f f i c u l t to define or describe, however. In t h i s study " c o n t r o l " has an operational d e f i n i t i o n : the s o c i a l and l i n g u i s t i c behaviours that emerge i n the coparticipant adult when t o l d to follow (the chil d ' s ) lead, and l e t him or her choose each toy and how i t should be played with (low c o n t r o l ) , compared to the adult's behaviour when t o l d to teach a c h i l d how to use a complex, unfamiliar toy and encouraged to teach by explaining v e r b a l l y rather than by demonstration (high c o n t r o l ) . In each of these s i t u a t i o n s 72 the ro l e s of the adult and c h i l d are d i f f e r e n t on the basis of how much power each person has i n defining the s i t u a t i o n and the d i r e c t i o n of the conversation, and yet we s t i l l don't know the s p e c i f i c v a r i a b l e s that, comprise " c o n t r o l " within an i n t e r a c t i o n . A p a r t i c i p a n t i n a conversational i n t e r a c t i o n has more con t r o l when he or she t a l k s more, introduces many of the t o p i c s , and i s able to e l i c i t responses from the other p a r t i c i p a n t that are relevant to previous remarks. Conversely, a person has l e s s c o n t r o l when he or she talks l e s s , responds to the other person's topics rather than introducing t o p i c s , and i s unable to e l i c i t relevant remarks from the : other speaker. I would propose that the i d e a l language learning s i t u a t i o n i s one i n which the conversational c o n t r o l i s shared or alternated between p a r t i c i p a n t s , because then c h i l d r e n can lea r n from more experienced speakers, while also having an opportunity to t r y out conversational techniques themselves. This a l t e r n a t i o n of conversational c o n t r o l seems to be t y p i c a l of the in t e r a c t i o n s between mothers and t h e i r young c h i l d r e n . The language input l i t e r a t u r e provides s u b s t a n t i a l evidence that mothers tend to l e t t h e i r c h i l d r e n lead both the a c t i v i t y and the conversation i n a free play s e t t i n g , and that t h i s noncontrolling q u a l i t y i s p o s i t i v e l y correlated with the c h i l d ' s language growth (for example, Nelson, 1973). Complementing such i n t e r a c t i o n s are conversations i n which the mother's primary purpose i s to teach or d i r e c t the c h i l d ' s behaviour (the s i t u a t i o n I attempted to duplicate experimentally i n my "high c o n t r o l " task). 73 In these i n t e r a c t i o n s , the mother i s obviously i n control of both a c t i v i t y and t a l k . Whatever the components of control are, the question remains: why do more complex language structures emerge when the c h i l d has greater l i n g u i s t i c and s i t u a t i o n a l control? S o c i o l i n g u i s t i c research suggests that conversational control i s r e l a t e d to s o c i a l power. A c h i l d who can introduce a topic that someone responds to, or give an i n s t r u c t i o n that someone c a r r i e s out, i s able to c o n t r o l h i s or her environment by manipulating other people. This evidence of personal power must be r e i n f o r c i n g to the c h i l d , providing motivation to obtain more of i t by becoming more competent at language, the means of obtaining t h i s power. It follows then, that a s i t u a t i o n involving low adult control and high c h i l d control of the conversation would f a c i l i t a t e the use of more complex language structures by the c h i l d . Another possible explanation draws on the notions of " s o c i a l speech" versus " c o l l e c t i v e monologue" (Piaget, 1959). In a previous unpublished case study (Lapadat, 1977), I found that a c h i l d at the same l e v e l of development (just entering the multiword stage) t y p i c a l l y produced longer, more grammatically complex utterances when " t a l k i n g to h e r s e l f " than when a c t i v e l y involved i n a two-way conversation. While i t i s d i f f i c u l t to o b j e c t i v e l y separate monologue or c o l l e c t i v e monologue utterances from s o c i a l utterances on the basis of utterance c h a r a c t e r i s t i c s or s i t u a t i o n a l d i f f e r e n c e s , I also informally observed a trend i n the present study for c h i l d r e n to formulate longer, often u n i n t e l l i g i b l e , utterances 7 while apparently " t a l k i n g to themselves" (for example, not making eye contact, playing u n i l a t e r a l l y with a toy, not waiting for a response to t h e i r utterances). One way to i n t e r p r e t t h i s i s to suppose that when adults becomes les s c o n t r o l l i n g and l e s s involved i n conversations, c h i l d r e n are freed from some of the demands of s o c i a l speech, such as turn-taking or making comments relevant to the established t o p i c . Thus they can d i r e c t more attention to length or grammatical complexity. Or perhaps as adults r e l i n q u i s h c o n t r o l , c h i l d r e n receive l e s s feedback about whether or not the l i s t e n e r understands t h e i r messages, so are l e s s l i k e l y to use simple constructions or to break down and repeat parts of longer constructions. However, t h i s contradicts research findings that c h i l d r e n use p r i o r adult utterances i n dialogues as a " s c a f f o l d , " allowing them to produce longer and more complex utterances than those predicted by t h e i r own production rules (Ervin-Tripp, 1977). Children produce a greater proportion of more complex constructions when they are d i r e c t i n g the conversation and the action and the adult coparticipant i s l e s s c o n t r o l l i n g . Further i n v e s t i g a t i o n i s needed to determine whether more complex speech occurs i n t h i s s i t u a t i o n because i t i s being s o c i a l l y r e i n f o r c e d , or because the nature of the s i t u a t i o n reduces other communicative pressures on c h i l d r e n , freeing them to concentrate on grammatical complexity. Or the explanation may l i e i n factors that I have not addressed here. 75 Frequency The count of the frequency with which c h i l d r e n spoke i n the d i f f e r e n t conditions showed that the c h i l d r e n talked much more to t h e i r own mothers than to strangers, and more i n the high attention condition than i n the low attention condition. Their amount of speech was not s i g n i f i c a n t l y d i f f e r e n t i n the high c o n t r o l condition than i n the low cont r o l condition. While the degree of f a m i l i a r i t y and attention did not s i g n i f i c a n t l y influence the complexity of children's constructions i n t h i s study, c h i l d r e n produced s i g n i f i c a n t l y more speech i n high f a m i l i a r i t y and high attention s i t u a t i o n s . These r e s u l t s have i n t e r e s t i n g implications. I suspect that the ch i l d r e n talked more i n these s i t u a t i o n s for the same reasons that I previously discussed with respect to more complex constructions i n these same s i t u a t i o n s (for example, greater background of shared knowledge, necessity f o r es t a b l i s h i n g same object of shared attention, etc. See ppp 67-71 )) Although the proportionate complexity of constructions did not change, j u s t t a l k i n g more means that the c h i l d r e n a c t u a l l y produced a s i g n i f i c a n t l y greater number of the more complex v e r t i c a l and h o r i z o n t a l constructions i n the high f a m i l i a r i t y and high attention conditions than i n the corresponding low f a m i l i a r i t y and low attention conditions. Thus the c h i l d r e n had more opportunity to p r a c t i c e producing more complex constructions when t a l k i n g to t h e i r mothers, and when someone was paying attention to them. Another i n t e r e s t i n g trend was that c h i l d r e n talked to strangers 76 somewhat more i n the high adult c o n t r o l condition than under low adult c o n t r o l ; i n p a r t i c u l a r , they produced more hori z o n t a l constructions. This i s contrary to the general trend for c h i l d r e n to t a l k s l i g h t l y more i n the low control condition than i n the high c o n t r o l condition and also contrary to the highly s i g n i f i c a n t r e s u l t that c h i l d r e n used more complex constructions i n the low con t r o l condition. Perhaps c h i l d r e n at t h i s age are reluctant to t a l k to s t r a n g e r s , ^ so that i f they are permitted to d i r e c t the conversation (low c o n t r o l ) , they tend rather to ignore the stranger and not t a l k at a l l . . On the other hand, they are sociable enough that i f the stranger c a r r i e s the burden of the conversation (high c o n t r o l ) , they w i l l take a passive conversational r o l e , and respond to d i r e c t questions etc The f i n d i n g that c h i l d r e n tend to use t h e i r most complex construction type (horizontal) with strangers matches Berko Gleason's (1975) observation that c h i l d r e n use more complex speech to strangers. Two explanations for t h i s follow. One p o s s i b i l i t y i s that c h i l d r e n have some awareness that the stranger w i l l be more d i f f i c u l t to communicate with, so they function at t h e i r highest l e v e l of grammatical competence i n order to get t h e i r message across. The other, almost d i r e c t l y contradictory, explanation i s that c h i l d r e n receive or interpret le s s of the feedback from strangers, so they do not s i m p l i f y or breakdown t h e i r messages u n t i l they receive confirmation of understanding as they would with a f a m i l i a r person. l^Berko Gleason (1973) observes that the f i r s t s o c i o l i n g u i s t i c choice c h i l d r e n make, at about one year, i s to speak or not to speak, with the l a t t e r related to u n f a m i l i a r i t y . 77 Implications I w i l l now discuss the c l i n i c a l and t h e o r e t i c a l implications of some of these r e s u l t s . This study shows that d i f f e r e n c e s i n adult i n t e r a c t i o n s t y l e s influence children's output i n p a r t i c u l a r conversational s e t t i n g s , either by a f f e c t i n g the proportionate complexity, or by increasing or decreasing the o v e r a l l amount of speech. The findings also support Scollon's (1976; 1979) claim that c h i l d r e n go through a stage of producing v e r t i c a l constructions as they make the t r a n s i t i o n from s i n g l e words to grammatical constructions. The r e s u l t that c h i l d r e n use more complex utterances when given more con t r o l of the i n t e r a c t i o n has important implications for our understanding of the a c q u i s i t i o n of language. The a v a i l a b i l i t y of t h i s type of conversational i n t e r a c t i o n i n which the adult takes a more passive, f a c i l i t a t i v . e . ' r o l e may have a long term e f f e c t on the c h i l d ' s rate .of language learning (Nelson, 1973). C e r t a i n l y i t appears that i n at l e a s t some s i t u a t i o n s , open-ended, f a c i l i t a t i v e i n t e r a c t i o n s lead to higher l e v e l language use i n the c h i l d than do structured, teaching types of i n t e r a c t i o n s . Furthermore, while I have not determined the components of control here, i t seems the f u n c t i o n a l value of t a l k i n g , what a person can get done, or what kind of response he or she can e l i c i t with language, i s r e l a t e d to constructional complexity i n language learning. The following c l i n i c a l implications for a c h i l d at t h i s developmental i ^ I n t h i s study, the contrived "low c o n t r o l " condition, i n which adults were asked to follow the c h i l d ' s lead, e l i c i t e d a great number of l i n g u i s t i c a l l y and s i t u a t i o n a l l y contingent responses from the adults, and resulted i n more complex language from the c h i l d . 78 l e v e l can be drawn. A c l i n i c i a n w i l l probably obtain a language sample closer to the c h i l d ' s constructional or grammatical l e v e l of competence by using a "low adult c o n t r o l " type of sampling procedure. This could be influenced by the degree of f a m i l i a r i t y of the coparticipant adult, however. Language therapy which increases the c h i l d ' s c o n t r o l , either by allowing the c h i l d to lead i n play, or by providing choices about s e l e c t i o n and ordering of other events, w i l l probably r e s u l t i n increased constructional complexity during that conversation. It i s not c e r t a i n whether rate of a c q u i s i t i o n over the long term w i l l be affected. The r e s u l t that c h i l d r e n t a l k more i n high f a m i l i a r i t y and high a t t e n t i o n conditions than i n low f a m i l i a r i t y and low att e n t i o n conditions, producing a correspondingly greater number of v e r t i c a l and ho r i z o n t a l constructions, suggests that children's rate of learning how to produce multiword constructions or grammatical constructions may be d i f f e r e n t i a l l y affected according to whether a f a m i l i a r person i s a v a i l a b l e for the c h i l d to t a l k to, and whether someone gives the c h i l d t h e i r f u l l a t t e n t i o n during the conversation. This does not necessar i l y follow, however, as for example, amount of p r a c t i c e may not be correlated with rate of learning. Confirmation of t h i s claim awaits further i n v e s t i g a t i o n . In addition, optimal amount of i n t e r a c t i o n i n d i f f e r e n t conditions can only be guessed at. How many minutes of "high attention i n t e r a c t i o n " a day i s best: f i f t e e n minutes, an hour, f i v e hours? And does i t make any d i f f e r e n c e whether the f a m i l i a r person i s the c h i l d ' s parent, a daycare worker, an older s i b l i n g , or a peer? Nelson (1973) notes 79 that children's language growth i s p o s i t i v e l y correlated with amount of time spent with adults, but negatively correlated with amount of time spent with other children. Also, the optimum configuration of adult i n t e r a c t i o n v a r i a b l e s could vary r a d i c a l l y depending on the c h i l d ' s l e v e l of development. The c l i n i c a l i m p l i c a t i o n , then, i s that for a c h i l d j ust entering the multiword stage, d a i l y one-to-one l i n g u i s t i c i n t e r a c t i o n with an a t t e n t i v e , f a m i l i a r adult would provide an opportunity for the c h i l d to p r a c t i c e h i s or her highest l e v e l constructions with a possible long term e f f e c t on rate of grammatical development. The reasons why c h i l d r e n go through a period of producing v e r t i c a l constructions in t h e i r t r a n s i t i o n from s i n g l e words to s y n t a c t i c constructions have been outlined i n Chapter One, and they w i l l not be r e i t e r a t e d here. A l l of the c h i l d r e n i n t h i s study used v e r t i c a l constructions, and i n high proportions, p a r t i c u l a r l y i n c e r t a i n s i t u a t i o n s . This suggests that while not n e c e s s a r i l y u n i v e r s a l , the stage of using v e r t i c a l constructions may be quite common and widespread. C l i n i c a l l y , t h i s has the i m p l ication that for language delayed c h i l d r e n at the one word stage, development of v e r t i c a l constructions may be a v a l i d therapy goal, preceding the goal of putting two or more words together into "sentences", and c l i n i c i a n s should know how to recognize them. For Further Research In the above discussion, I have pointed out a number of issues that 80 could be c l a r i f i e d through further i n v e s t i g a t i o n . I would l i k e to discuss here a few remaining important areas f o r a d d i t i o n a l research. A major question concerns the longt.term e f f e c t of adult i n t e r a c t i o n v a r i a b l e s on children's language development. This study shows that control i s rela t e d to constructional complexity within the immediate i n t e r a c t i o n a l s i t u a t i o n , but I cannot make any claims about the e f f e c t of control over the long term (but see Nelson, 1973). A longitudinal., approach i s required to trace e f f e c t s of any p a r t i c u l a r adult i n t e r a c t i o n v a r i a b l e on any p a r t i c u l a r aspect of language development. In t h i s study I have just examined the influence of three adult i n t e r a c t i o n v a r i a b l e s ( f a m i l i a r i t y , attention, and control) on one aspect of language development (constructional complexity) i n one environment (a u n i v e r s i t y c l i n i c ) and at one stage of development. A more meaningful i n t e r p r e t a t i o n of my r e s u l t s w i l l be possible once more i s known about these other facets. Another problem concerns the actual sequence of development once a c h i l d begins to move from using si n g l e words to using multiword constructions. I have proposed a model consisting of ordered l e v e l s to e s t a b l i s h the t h e o r e t i c a l framework of t h i s study, but l o n g i t u d i n a l data for a number of c h i l d r e n are needed to e s t a b l i s h the v a l i d i t y of the model. While with studies such as t h i s we can begin to answer the question of how chil d r e n move from one l e v e l to the next, f i n a l l y making the t r a n s i t i o n to syntax, supplementary in-depth analyses of i n t e r a c t i o n sequences are also e s s e n t i a l to round out the picture. A f i n a l question of i n t e r e s t involves the motivation f o r using constructions of a p a r t i c u l a r l e v e l of complexity. Throughout t h i s 81 study, I made an assumption that c h i l d r e n take a "build-up" approach to constructions. That i s , I supposed that they t r y a message at a l e s s complex l e v e l (for example, s i n g l e word) and :then.when^theccoparticipant does not understand or follow through, they provide more information by using a more c o n s t r u c t i o n a l l y complex utterance (f6r::example, v e r t i c a l or h o r i z o n t a l construction). Or I thought that c h i l d r e n may t r y a message at a simpler l e v e l as a step i n b u i l d i n g up to or formulating a more complex construction. This i s commonly reported i n the c h i l d language l i t e r a t u r e . An example of t h i s build-up process from my data follows: Daniel M. (Stranger) V II 1. out. (Daniel i s holding out. a box of b u i l d i n g 2. that. discs on h i s lap.) HWI 3. out that. However, another process, which I s h a l l c a l l "break-down" also appeared i n the data. I t seemed that often c h i l d r e n began by s t a t i n g an idea using a complex construction, then faced with noncomprehension on the part of the adult (perhaps because of poorer a r t i c u l a t i o n i n longer utterances), they broke the utterance down into component parts, repeating the message as a v e r t i c a l construction or as s i n g l e words u n t i l they received some i n d i c a t i o n of understanding from the adult, or gave up t r y i n g . An example follows: 82 M i l l y T. (Stranger) V IV H VII 1. look, bike. i t a bike. no response ( M i l l y has attached some discs and the the structure looks l i k e a b i c y c l e . ) The appearance of these break-down sequences suggest that c h i l d r e n may sometimes be forced to communicate at a simpler l e v e l because of communicative constraints (for example, poor a r t i c u l a t i o n or lack of shared experience). I t follows that once these constraints are removed, they would tend to t a l k at a complexity l e v e l approaching t h e i r l e v e l of competence. In order to discover the process c h i l d r e n go through to construct p a r t i c u l a r utterances, we must examine constructions embedded within t h e i r l i n g u i s t i c and s i t u a t i o n a l contexts. Build-up and break-down sequences may coexist with other processes, and by describing and tes t i n g these processes along with coparticipant i n t e r a c t i o n v a r i a b l e s , we w i l l be able to explain how and why ch i l d r e n move from sin g l e words to syntactic constructions. 83 References Andersen, E.S. A selected bibliography on language input to young chil d r e n . Papers and Reports on Child Language Development, 1975, 9, 75-86. Atkinson, M. Prerequisites for reference. In E. Ochs & B.B. S c h i e f f e l i n (Eds.), Developmental pragmatics. New York: Academic Press, 1979. Bates, E. Language and context: The a c q u i s i t i o n of pragmatics. New York: Academic Press, 1976. Bates, E. & MacWhinney, B. A f u n c t i o n a l i s t approach to the a c q u i s i t i o n of grammar. In E. Ochs & B.B. S c h i e f f e l i n (Eds.), Developmental  pragmatics. New York: Academic Press, 1979. Berko Gleason, J . Fathers and other strangers: Men's speech to young ch i l d r e n . In D. Dato (Ed.), Developmental p s y c h o l i n g u i s t i c s :  Theory and a p p l i c a t i o n . Washington, D.C: Georgetown Uni v e r s i t y Press, 1975. Berko Gleason, J . Talking to c h i l d r e n : Some notes on feedback. In C.E. Snow & C.A. Ferguson (Eds>)> Talking to c h i l d r e n : Language  input and a c q u i s i t i o n . Cambridge U n i v e r s i t y Press, 1977. Berko Gleason, J. & Weintraub, S. Input language and the a c q u i s i t i o n of communicative competence. In K.E. Nelson (Ed-), • Children's' language (Vol. 1). New York: Gardner Press, 1978. Bloom, L. One word at a time: The use of s i n g l e word utterances before  syntax. The Hague: Mouton, 1976. Bloom, L. & Lahey, M. Language Development and Language Disorders. New York: John Wiley & Sons, 1978. Branigan, G. Sequences of s i n g l e words as structured u n i t s . Papers  and Reports on Child Language Development,.1976, 60-70. Brown, R. A f i r s t language. Cambridge, Mass.: Harvard Uni v e r s i t y Press, 1973. Bruner, J.S. The ontogenesis of speech acts. Journal of Child Language, 1975, 2, 1-19. Bruner, J.S. The r o l e of dialogue i n language a c q u i s i t i o n . In A. S i n c l a i r ; .R.J. J a r v e l l a , & W.J.M. Levelt (Eds.), Children's  conception of language. B e r l i n : Springer & Verlag, 1978. 84 Carter, A. The transformation of sensorimotor morphemes into words: A case study of the development of "here" and "there." Papers  and Reports on Child Language Development, 1975, 1_0, 31-47. Cross, T.G. Mothers' speech adjustments: The contributions of selected c h i l d l i s t e n e r v a r i a b l e s . In C.E. Snow & CA. Ferguson (Eds.), Talking to c h i l d r e n : Language input and a c q u i s i t i o n . Cambridge Un i v e r s i t y Press, 1977. Cross , T.G. Mothers' speech and i t s association with rate of l i n g u i s t i c development i n young chi l d r e n . In N. Waterson & C. Snow (Eds.), The development of communication. New York: John Wiley & Sons, 1978. Dore, J . Conversation and preschool language development. In P. Fletcher & M. Garman (Eds.), Language a c q u i s i t i o n : Studies i n f i r s t language  development. Cambridge U n i v e r s i t y Press, 1979. Dore, J . , F r a n k l i n , M.B., M i l l e r , R.T. & Ramer, A.L.H. T r a n s i t i o n a l \.;phenomeria i n r early ...language acquisition,,'.-.. Journal! of iiChild. -.Language, 1976, 3_, 13-28. Ervin-Tripp, S. & M i l l e r , W. Early discourse: Some questions about questions. In M. Lewis & L.A. Rosenblum (Eds.), Interaction,  conversation, and the development of language. New York:•. John Wiley &>S6ns, 1977. Garman, M. E a r l y grammatical development. In P. 1'Eletcher & M. Garman (Eds.), Language a c q u i s i t i o n : Studies i n f i r s t language development. Cambridge Uni v e r s i t y Press, 1979. Garvey, C. & Hogan, R. S o c i a l speech and s o c i a l i n t e r a c t i o n : Egocentrism r e v i s i t e d . Child Development, 1973, 44_, 569-574. Gelman, R. & Shatz, M. Appropriate speech adjustments: The operation of conversational constraints on t a l k to two-year-olds. In M. Lewis & L.A. Rosenblum (Eds.), Interaction, conversation, and the development  of language. New York: John Wiley&& Sons, 1977. Givon, T. Understanding grammar. New York: Academic Press, 1979. Greenfield, P.M. & Zukow, P.G. Why do chi l d r e n say what they say when they say i t ? : An experimental approach to the psychogenesis of presupposition. In K.E. Nelson (Ed.), Children's language (Vol. 1). New York: Gardner Press, 1978. G r i f f i t h s , P. Speech acts and early sentences. In P. Fletcher & M. Garman (Eds.), Language a c q u i s i t i o n : Studies i n f i r s t language  development. Cambridge U n i v e r s i t y Press, 1979. 85 Hess, R. & Shipman, V. Early experience and the s o c i a l i z a t i o n of cognitive modes i n ch i l d r e n . C h i l d Development, 1965, _36, 869-886. Keenan, E. Ochs Making i t l a s t : Uses of r e p e t i t i o n i n children's discourse. In S. Ervin-Tripp & C. Mitchell-Kernan (Eds.), Child  discourse. New York: Academic Press, 1976. Keenan, E. Ochs & S c h i e f f e l i n B. Topic as a discourse notion: A study of topic i n the conversations of ch i l d r e n and adults. In C.N. L i (Ed.), Subject and topi c . New York: Academic Press, 1976. Lieven, E.V.M. Conversations between mothers and young c h i l d r e n : Individual differences and t h e i r possible i mplication for the study c:;: of language learning. In N. Waterson & C. Snow (Eds.), The development of communication. New York: John Wiley & Sons, 1978. Lock, A. The guided reinvention of..language^cnLondonicsAcademlc. Press;-. 1980/ MacWhinney, B. & P r i c e , D. The development of the comprehension of topic-comment marking. In D. Ingram,-.F.C. C. Peng & P. Dale (Eds.), Proceedings of the f i r s t i n t e r n a t i o n a l congress for the study of  c h i l d language. Un i v e r s i t y Press of America, 1980. Moerk, E.L. Pragmatic and semantic aspects of early language development. Univ e r s i t y Park Press, 1977. Nelson, K. Structure and strategy i n learning to t a l k . Monographs of  the Society for Research i n C h i l d Development, 1973, 38^ ( S e r i a l No. 149) Newport, E.L., Gleitman, H., & Gleitman, L.R. Mother, I'd rather do i t myself: Some e f f e c t s and non-effects of maternal speech s t y l e . In C.E. Snow & CA. Ferguson (Eds.), Talking to c h i l d r e n : Language  input and a c q u i s i t i o n . Cambridge U n i v e r s i t y Press, 1977. Ochs, E. Introduction: What c h i l d language can contribute to pragmatics. In E. Ochs & B.B. S c h i e f f e l i n . (Eds.), Developmental pragmatics. New York: Academic Press, 1979. Ochs, E., S c h i e f f e l i n , B.B., & P i a t t , M.L. Propositions across utterances and speakers. In E. Ochs & B.B. S c h i e f f e l i n (Eds.), Developmental pragmatics. New York: Academic Press, 1979. Piaget, J . The language and thought of the c h i l d (3rd ed.). London: Routledge, 1959. Rodgon, M.M. Single-word usage, cognitive development, and the beginnings of combinatorial speech. Cambridge U n i v e r s i t y Press, 1976. 86 Scollon, R. Conversations with a one year o l d . The Un i v e r s i t y Press of Hawaii, 1976. Scollon, R. A r e a l early stage: An unzippered condensation of a d i s s e r t a t i o n on c h i l d language. In E. Ochs & B.B. S c h i e f f e l i n (Eds.), Developmental pragmatics. New York: Academic Press, 1979. Shields, M.M. Some communicational s k i l l s of young c h i l d r e n : A study of dialogue i n the nursery school. In R.N. Campbell & P.T. Smith (Eds.), Recent advances i n the psychology of language: Language  development and mother-child i n t e r a c t i o n . New York: Plenum Press, 1978. Shields, M.M. The development of dialogue s k i l l s i n the preschool c h i l d . In D. Ingram, F.C.C. Peng, & P. Dale (Eds.), Proceedings  of the f i r s t i n t e r n a t i o n a l congress for the study of c h i l d language. Un i v e r s i t y Press of America, 1980. Shugar, G.W. Text analysis as an approach to the study of early l i n g u i s t i c operations. In N. Waterson & C. Snow (Eds.), The  development of communication. New York: John Wiley & Sons, 1978. Snow, C.E. Mothers' speech research: From input to i n t e r a c t i o n . In C.E. Snow & CA. Ferguson (Eds.), Talking to c h i l d r e n :  Language input and a c q u i s i t i o n . Cambridge Uni v e r s i t y Press, 1977a. Snow, C.E. The development of conversation between mothers and babies. Journal of Child Language, 1977b, 4 _ , 1-22. Snow, C.E. Conversations with c h i l d r e n . In P. Fletcher & M. Garman (Eds.), Language a c q u i s i t i o n : Studies i n f i r s t language development. Cambridge U n i v e r s i t y Press, 1979. Sylvester-Bradley, B. & Trevarthen, C. Baby t a l k as an adaptation to the infant's communication. In N. Waterson & C. Snow (Eds.), The development of communication. New York: John Wiley & Sons, 1978. von Ra f f l e r - E n g e l , W. & Rea, C. The influence of the c h i l d on the conversational behavior of the adult: The evolution of a new concept. In D. Ingram, F.C.C. Peng, & P. Dale (Eds.), Proceedings  of the f i r s t i n t e r n a t i o n a l congress for the study of c h i l d language. Un i v e r s i t y Press of America, 1980. Weisenburger, J.L. A choice of words: Two-year-old speech from a s i t u a t i o n a l point of view. Journal of Chi l d Language, 1976, _3,-275-281. 87 Wells, G> What makes for successful language development? In R.N. Campbell & P.T. Smith (Eds.), Recent advances i n the psychology  of language: Language development and mother-child i n t e r a c t i o n . New York: Plenum Press, 1978. Wells, G. V a r i a t i o n i n c h i l d language. In P. Fletcher & M. Garman (Eds.), Language a c q u i s i t i o n : Studies i n f i r s t language development. Cambridge U n i v e r s i t y Press, 1979. Reference Note 1. Lapadat, J.C. Two speech s t y l e s . Unpublished manuscript, 1977. (Available from 1913 St. John St., Regina, Saskatchewan). 89 Addenda Berko Gleason, J . Code switching i n children's language. In T.E. Moore (Ed.), Cognitive development and the a c q u i s i t i o n of language. New York: Academic Press, 1973. Ervin-Tripp, S. From conversation to syntax. Papers and Reports on  C h i l d Language Development, 1977, 13. Snow, C. Mothers' speech to c h i l d r e n learning language. Child  Development, 1972, 4_3, 549-565. 9Q Appendix One Instructions (for mothers) Task 1 (free play, high attention) "Now I'm going to leave you and (ch's name) i n t h i s room for about 10 or 15 minutes, j u s t to get used to the s i t u a t i o n . I'm interested i n the kind of language (ch's name) uses as he/she t a l k s , so what I want you to do i s ju s t play with (ch's name) as you normally would so I can hear the way he/she u s u a l l y t a l k s . There's a box of toys there that you can use. Wendy w i l l be i n i n 15 minutes to ask you a few questions." Task 2 (free play, low at t e n t i o n : interview) "Now I'd l i k e to ask you a few questions. If there are any that you'd prefer not toaanswer, you don't have to." INTERVIEW QUESTIONS Task 3 (free play, high c o n t r o l : teaching task) "Here's a toy I'd l i k e you to use for the next 15 minutes. I t ' s a v i l l a g e , and these pegs and blocks can be put together to make houses, trees, and people, and there are also cars and a t r a i n . I'd l i k e you to teach (ch's name) how to use i t . If possible, t r y to get him/her to put the pieces together him/herself, rather than doing i t for him/her. I ' l l be back i n 15 minutes." Task 4 (free play, low c o n t r o l : c h i l d - d i r e c t e d games) "Now I'd l i k e you and (ch's name) to play together with the other toys again. What I am p a r t i c u l a r l y interested i n now, i s seeing the ways i n which (ch's name) w i l l play with these toys, and the kinds of games he/she w i l l make up. If possible, I would l i k e you to follow (ch's name)'s lead, and l e t him/her choose each toy, and how i t should be played with. I ' l l be back i n 15 minutes, and at that point we'll f i n i s h up." 91 Instructions (for non-mothers) Task 1 (free play, high attention) "Now I'm going to leave you and (ch's name) i n t h i s room for 10 or 15 minutes to give you a chance to get to know each other. I'm interested i n the kind of language (ch's name) uses, so I'd l i k e you to play with (ch's name) the way you normally play with c h i l d r e n , so I can hear the way he/she u s u a l l y t a l k s . You can use that box of toys there. Wendy w i l l be i n i n 15 minutes to ask you a few questions." Task 2 (free play, low attent i o n : interview) "Now I'd l i k e to ask you a few questions. If there any that you'd prefer not to answer, you don't have to." INTERVIEW QUESTIONS Task 3 (free play, high c o n t r o l : teaching task) "Here's a toy I'd l i k e you to use for the next 15 minutes. I t ' s a magnetic board and d i f f e r e n t kinds of animals can be made using the coloured blocks. I'd l i k e you to teach (ch's name) how to make some of these animals. If possible, t r y to get him/her to put the pieces together him/herself, rather than doing i t for him/her. If you want, you can use these 2 pieces to stand the board upright. I ' l l be back i n 15 minutes." Task 4 (free play, low c o n t r o l : c h i l d - d i r e c t e d games) "Now I'd l i k e you and (ch's name) to play together with the other toys again. What I'm p a r t i c u l a r l y interested i n now i s seeing the ways i n which (ch's name) w i l l decide to play with these toys, and the kinds of games he/she w i l l make up. If possible, I would l i k e you to follow (ch's name)'s lead, and l e t him/her choose each toy and how i t should be played with. I ' l l be back i n 15 minutes, and at that point we'll f i n i s h up." 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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.831.1-0095805/manifest

Comment

Related Items