Representations of Questioning and Answering in Children's...

Representations of Questioning and Answering in Children's First School BooksAuthor(s): Carolyn D. Baker and Peter FreebodySource: Language in Society, Vol. 15, No. 4 (Dec., 1986), pp. 451-483Published by: Cambridge University PressStable URL: http://www.jstor.org/stable/4167798 .Accessed: 11/04/2011 12:55

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=cup. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to Languagein Society.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=cup

http://www.jstor.org/stable/4167798?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/action/showPublisher?publisherCode=cup

Lang. Soc. 15, 451-484. Printed in the United States of America

Representations of questioning and answering in children's first school books

CAROLYN D. BAKER AND PETER FREEBODY

Centre for Behavioural Studies in Education University of New England

ABSTRACT

Children's first school books contain a considerable amount of reported talk among characters in the stories. This is a central aspect of the characterization of these books as transitional from the conventions of oral language to the conventions of written prose, that is, as introductions to literacy. The nature of the written representation of conversation in such books has not previously been examined. This paper presents a partial analysis of this feature of beginning school readers, focussing on "question-answer sequences." We show how these representations of talk compare with naturalistic research on child-adult interaction at home and in classrooms, and we propose that the model of child-adult talk portrayed in "home" and "school" scenes in the books appears to endorse some of the conventions for participation in instructional talk, and in this respect is implicitly a source of socialization into classroom culture. At the same time, we find that the texts give child speakers far more initiative in conversation than typically obtains in classroom talk, and this is seen also to be a feature of the social constitution of the child in these texts. Thus an image of childhood which combines conversational initiative and conversational competence as a member of the classroom community is conveyed. The paper also points out possible difficulties for child readers in interpreting the talk-on-paper, arising both from textual formats and from the particular version of the child as conversationalist which the books describe (Child-adult conversation, question-answer sequences, first school books, literacy acquisition).

INTRODUCTION

Recent theorizing on the topic of literacy has drawn attention to the notion that the written language used in schools is a special language (Halliday 1978; Olson I977; Perera 1982; Rubin I980). In particular, Olson has argued that, through this special language, written texts present readers with ""pictures of reality or forms of knowledge" which are different from those presented in the oral- conversational experiences of readers. Olson has further remarked that the areas

0047-4045/86/150451-34 $2.50 (D 1986 Cambridge University Press

451


of "conflict" between the languages of formal written text and oral conversation are clearly seen in the first books encountered by children in school (1977:76). It is our purpose to pursue particular aspects of this "'conflict" of languages through the detailed study of certain linguistic characteristics of a large collection of the texts of beginning school reading books.

We have documented elsewhere that these reading books - the first official "school" books encountered by children - contain a remarkable amount of reported talk (Freebody and Baker 1985). In fact much of what children first read at school is in the form of written reportage of oral talk. It has also been suggested that the nature and format of this reported talk make it different in certain ways from the actual everyday conversations children may be involved in. Our main objective in this paper is to describe the conversational practices of child and adult speakers in the children's reading texts, treating the participants as if they were members of a particular speech community. We will try to document how people talk in these books, focussing mainly on question-answer sequences. We regard this work as a necessary first step toward the more general study of how authors represent conversation and how school texts represent everyday life. Such an analysis is important both because representations of oral conversation in the children's texts are so prevalent, and because this material is central to our characterization of these texts as transitional from oral to written language.

Our second objective is to compare our descriptions of speaking within the texts with the findings of research on child-adult talk at home and at school. This will provide grounds for estimating the extent to which these books can be said to reflect oral language practices which occur in children's interactions with others prior to and at this age, or whether they provide a model of language practices which, in specific, identifiable ways, does not match children's oral language experiences. We would hope that our findings here could inform discussion of the use of these books in the teaching of reading by pointing to some previously overlooked sources of reading difficulty.

Also, in doing our analyses we have encountered particular difficulties in making sense of some of the conversations, and we wonder whether for many children a partial source of difficulty lies here too. To our knowledge, this possibility has not been addressed before: Attention has been given to vocabulary, sentence length, and to the use of pictures as indices of readability (see Klare 1975) but not to this aspect of the organization of these texts.'

A third objective relates to our interest in the portraits of children, child-adult relations, and family life provided in the texts. The way conversation is con- ducted between speakers in the texts contributes significantly to the images of children and adults which the books provide. The details of this contribution have rarely been examined. These are cultural images, built subtly into the texts in the representations of how people talk to each other. This is yet another possible source of children's difficulties with the texts, especially as the books

452

QUESTIONING AND ANSWERING IN FIRST SCHOOL BOOKS

are primarily stories about children and their conversational interactions with others, and are written as if from the perspective of children (see also Freebody & Baker I985). The children in the books are only one possible version of the children who read the books. This point is significant in that children could experience a form of cultural conflict with the reading materials if the image of childhood presented in the books does not capture or correspond to children's self-conceptions, experiences, and ways of construing the world. This problem of relating to the version of "childhood" embedded in the books could be a previously unrecognised source of reading difficulty. We examine the image of childhood and the various methods by which it is constructed textually in more depth elsewhere (Baker & Freebody in press). In the present paper, we detail the bases for our claim that the images of children, adults, and families are made available to the reader in part through the ways in which talk is represented. Our analyses reveal some of the conversational practices and routines upon which models of "children," "adults," and of child-adult interaction are built.

WHY QUESTION-ANSWER SEQUENCES

As part of our interest in the ways in which conversational interactions in children's first school books compare with descriptions of naturally occuring home and school talk between children and adults, we consider question-answer sequences to be especially salient. Recent research on child-adult conversation has included attention to various aspects of question-answer sequences as a category of linked "initiation-response exchanges." These are seen to be an important form of interaction for language learning in the home (Wells i98i), and such sequences are fundamental to the organization of instructional classroom talk (e.g., MacLure & French 1981; Mehan I979; Walker I98I).

We have characterized beginning school readers both as providing for transition from oral to written conventions of language, specifically to the language of school texts (Olson 1977), and as containing a form of adult-produced discourse in which the very nature of childhood is constituted. Question-answer sequences are central elements of both of these accounts: They can be viewed as displaying a particular version of coherent and "literate" discourse and can be seen to index important aspects of social relationships between speakers, in our case particularly between children and adults. As Edwards (I98I:303) has stated, "forms of social relationship are signalled, and are reproduced or modified or challenged, in the act of speaking."

MATERIALS

Included in the survey were readers commonly used in a representative educational division of New South Wales, Australia. This division contained 65 class- es of children in their first and second years of schooling. Also included were

453


readers published recently by the major relevant publishing houses in Australia. A total of 163 basal and supplementary readers comprised the corpus.2

There is a considerable volume of the reportage of verbatim talk in the reading materials we are examining: We have located over 1,700 conversational "turns" (i.e., speeches separately attributed to individual children or adults by the nar- rators in the texts). The corpus also contains conversations among animals, between animals and children, and among people in fairy tales and similar sources. These are not included in the analyses described here. We have attended initially only to the conversational representation in "real life" settings in the books - mainly "home" and "school" scenes and stories. We have also included only those in which the text provides quotation marks to identify the separation of speeches and speakers, due to the considerable interpretive difficulty associated with those few sources in which this punctuation was left out.

I YPES OF QUESTION-ANSWER SEQUENCES

With the constraints above applied, there remain over 300 instances of "questions" being asked and/or answered in the corpus.3 For selection of a "question" we have used the criterion that the utterance be expressed in interrogative form - although we appreciate that in naturally occuring conversation, utterances in noninterrogative forms can be heard and treated as questions or requests (cf. Schegloff I984). We appreciate that there are types of utterances other than 'questions" which also strongly invite a response or "second part" to form a

conversational pair. This amounts to an acknowledgement that we are inspecting only a portion of potentially interesting pairs of utterances in the corpus and that some of the work we do with our particular selection is relevant to a wider range of conversational couplets outside our selection (MacLure & French 1 198I 1, for example, also deal specifically and selectively with question and answer sequences in home and school talk). Even selecting only "question-answer" pairs as we have defined them here yields some 6oo conversational turns to be in- spected, a significant proportion of the total.

A point which refers both to our methodology, and to the child reader's task in interpreting text, is that in undertaking the task of identifying question-answer pairs, we have had to practise certain techniques of reading. We have noted, at some points in the paper, the various linguistic and format clues, and the kinds of cultural knowledge, that any reader would use to read adjacent utterances in this way. It is the case that "pairs" of this kind are more readily identifiable than some other two-part exchanges due to the use of markers in the punctuation, sometimes the use of address terms, and usually the adjacency of the pairs. Nonetheless, we have noted difficulties which could be encountered by the child reader on the basis of having encountered some of these difficulties ourselves, even given these textual markers. Similarly, the identification of the speaker or recipient of a question or answer is not always straightforward.4

454


To identify possible differences in the content, structure, and context of the textual representations of question-answer pairs, and with a practical interest in subdividing the varieties of cases observed, we created a typology of question- answer sequences.

Informaition Sequences. Here the questioner is heard genuinely not to know something and the question is put to find out. The sequence is deemed to be complete when the answer is provided.

(i) "Where are you going?" asked Dan. "I am going away," said Jack. "I am going to my house."

(2) "Will he let us help him milk?" she says. "Yes," says Peter, "he likes us to help him with his work."

Request Sequences. In some cases, we see these as intermediate between information and permission questions. Speakers use this question form to ask others to engage in an activity.

(3) "Can we see some more of the farm?" asks Peter. "Yes," says Pam, "'come on, then, let us go for a walk."

(4) "Father," said Jill. "Can you come up here?" "Yes I can," said Father. "I can come up and play with you. We can have fun.'"

Permission Sequenc es. Here the speaker requests the go-ahead to engage in an activity, or to be given something.

(5) "May we come again soon?" asked Robin. "We shall see," said Miss Brown.

(6) "Can we have some apples, please?" says Peter. "Yes," says Mummy. ''Apples do you good."

The task of classifying sequences into the categories of the typology also involves interpretive problems. Not surprisingly, the "request" category contains more of the ambiguous cases than the others since the ambiguity often pivots on the word can. It should also be noted that the "information," "request," and "permission" notions are applicable to sequences, to the two-part exchange as an integrity. Nonetheless, after we have classified sequences as, say, "permission" sequences, we then proceed to talk about the "'question" turns in them as "permission" questions. But, the form of words of the question alone does not always decide the classification. It is often only with the second part, the response, that we have clear grounds as readers for our decision.

While our discussion of the three categories of questions will involve s6me analysis unique to each section, we will be asking some parallel questions under each of the three sections: Who initiates the sequence, and who completes it? How are the turns within the sequence typically constructed? Besides describing individual utterances in this way, we will ask: How are or could the two parts of the sequences be connected'? This is the issue of linguistic coherence. In each

455


section we will be commenting on the contribution of sequences of that particular type to the overall portrait of childhood and of family life in the readers. Having treated each category of questions separately, we will briefly discuss a more lengthy passage, describing how these question-answer sequences are integrated with other kinds of conversational moves and with narrator's contributions to form a "story."

Information sequences

Under this category we have placed those question-answer pairs in which one speaker asks for information on some state of affairs from some other speaker(s). Questions in which information is sought and provided are the most numerous in the children's readers (204). Some exemplary exchanges classified under the heading of "information" sequences are presented below. These will be used in the development of our discussion of this category of question-answers.

(7) "Is Mr. Whiskers a new pet?" asked Dan. "You will see," said Jack.

(8) Then Peter says to Daddy, "Are you going to play with us this after- noon? You said so." "Yes," says Jane, "You said so." "Yes," says Daddy, "I said it, so I will play. What do you want to play?" "Let us play with the ball," says Peter. Jane says, "'Yes, that will be good fun."

(9) "Is it time for us to help?" asked Peter and Elizabeth. "Yes it is," answered Miss Long.

(io) "When will our bus be here'?" asked one of the girls. "Very soon," said Miss Brown. "Now, is everyone here?" "I do not see Peter," said Elizabeth.

(i i) "Where is Mother?" said Timothy. "I will get her," said Father, and away he went.

We now ask "who initiates informational question-answer sequences in the home- and school-located exchanges portrayed in the children's readers?" Table I shows the distribution of question initiations across the 204 instances of "information" sequences. As we are describing two-part tums, our interest is in who asked the question, not who initiates the stretch of talk. In example 8, the stretch of talk shown contains two initiations, one by Peter and one by Daddy; that is, we are not examining here who started that talk in the episode as a whole.

A number of overall aspects of Table i are noteworthy. The children portrayed in these readers ask more questions than do the adults (64% vs. 36%). Addi- tionally, children ask a larger proportion of the questions set in the home context than they do of those set in the school context (68% vs. 56%). In the "school" scenes, they ask considerably more than research on actual classroom lessons

456


TABLE I. Participants in information-sequence questions in home and school contexts

At home At school Total

To adult/s 31 13 44 Initiated To child/ren 53 14 67 by child/ren To other or unclear 13 7 20

Totals 97 34 131

To adult/s 6 2 8 Initiated To child/ren 31 24 55 by adult/s To other or unclear 9 1 10

Totals 46 27 73

Grand Totals 143 61 204

would predict. For example, Mehan (0979:80) found that students initiate less than i8 percent of sequences in lessons. McHoul (1978:208) estimates that 8o percent of lesson talk is teacher talk. However, it is not formal lessons but the planning and doing of other activities such as a play or a class excursion that are portrayed in the readers. Much of the classroom talk appears to be "organizational" talk rather than "instructional" talk. It is the special events of classroom life rather than the routine or mundane which are described in the children's first school books.

Most of the adults' questions are to children (75%). Only occasionally do mother and father, teacher and bus driver, or grandmother and farm helper ask questions of each other. Children's questions are directed more often to other children than to adults (5I% vs. 34%), especially in the "home" context. Overall, approximately half of these sequences involve crossgenerational talk, with proportionately more of it occuring in the "school" context.

This gives an initial indication of how representations of conversation in the texts give speaker roles to children and adults. While on the one hand, the adults appear to be onlookers and supervisors in a "child-centred" world (children's play, children's trips, etc.), we note that a central place is given to the representation of talking with adults. In this sense they are not backstage, but central figures in the communication which goes on. Their presence and perspective is continually sought and portrayed. In the children's texts the children are continually dealing with the adults in ways, we suggest, which endorse the conventional roles and responsibilities of adults as well as the conventional interests of "children. "

The questions in the corpus are usually answered in the next turn at talk, with very few questions left unresolved or unanswered. Table I serves as a reliable guide to the speakers of the second-part turns. On a very few occasions, the narrator supplies the resolution without having another character speak. There are some "open" queries framed to invite self-selection to answer, as in "Who

457

CAROLYN D. BAKER AND PETER FREEBO01Y

will help me'?" and in which some participant in the scene can selt-select as answerer. The majority of questions in this ("'information") category, however, are seen as being addressed to children (60%) and/or answered by children (62%).

The issue of who talks, to whom, and how often is a critical one in the study of these readers. Persons who get turns at talk, in these texts as in ordinary conversation, have a presence in the story which is more immediate than. say, mention or description by the narrator - they are made to speak for themselves. In the school scenes, for example, where we presume a class of the usual size, only those who have turns at talk are named and thereby become characters in the story. Giving the person a turn at talk is a narrator's way of bringing that person into the story and thus to the imaginal attention of the reader. Given that the organization of these texts seems to be based largely on a distribution of turns at talk interspersed with brief scene-setting or scene-moving interjections by the narrator, the turns at talk carry much of the story line. Hence, getting turns at talk marks a character's presence in the story and thus in the world it is supposed to represent.

Further, in looking at the types of questions asked within the information- sequence category, we apply Stubbs's (1983) distinction between what he terms X questions and yeslno questions. X-questions require some content to complete the proposition begun in the first part, as in (12).

(12) "Who should go first'?" asked Jill "'Robin should go first," said Jack. "He told us a good story."

Yes/no questions seek either confirmation or negation of the proposition contained in the first part, as in (13).

(13) "Hello, Jack," said Robin. "is this your ball'?" "No," said Jack. "It is Jill's ball. Your little dog ran away with it."

We now deal with this distinction more extensively, to point out some of the details which inform our interpretations.

X-Questions. X-Questions leave open the specific content which will complete the proposition begun in the query. Many of these are "wh" questions (where, when, who, why, what, which, how), as shown in Table 2. We see that questions beginning ""where" (location) and "'what" (identification) predomi- nate within this category in both the home and school settings. Many of the "'where" questions concern where some person or object is, has been, or is going. Missing children or objects figure in a number of the stories. The prevalence of ""what'" questions also indicates the insertion of puzzles into the stories. as in ""What is in the box'?" or "'What is that on the road'?" and including some ""What shall we/did you do'?" questions.

Taking Table 2 as a clue to the depiction of the intellectual world displayed in

458


TABLE 2. Distribution of "X-questions'' among various forms in school and home contexts

Where What When Who Which Why Other

School 10 8 1 3 - 4 Home 33 27 1 6 1 3 3

these readers, it seems that there is relatively little concern among the speakers in the books with when things did happen but somewhat more about when things will happen. Nor is there much interest in reasons for events (only three instances of why). "Why" talk may occur in the actual teacher's use of the book in the reading class, that is, the teacher makes causes and motives points of discussion to assess comprehension of the story (cf. Heap 1985). Teachers are also concerned with ensuring that the sequential frame is understood, as in the "what happened next?" questioning routine used in reading lessons. However, children (and adults) in the stories are not made to concern themselves overly with understanding or explaining the world. Rather, they are made to remark on it as it happens and seem to live in a perpetual present with half an eye to the short-term future. Further, their talk is largely describable as "organizational," a point picked up again in our discussion of the yes/no sequences in a later section.

We will comment briefly on the organization of turns and the linguistic relationship between initiations and responses in this category of "wh-" sequences. Given the number of these sequences it is not possible to be comprehensive, but we can show some of the more frequently occuring formats (see also examples (I), (IO), (I I), and (I 2)).

(I4) "This is fun," he said. "But where is Mr. Whiskers'? I want to see him! " "Here, Mr. Whiskers," Jack called. "Come and see Dan."

(I5) "What will your father do with the cans, Andrew?" asked Sue. "He will take them up the track to the road. Soon, a big milk truck will carry them to the milk factory. Every day the milk truck comes back with some cans from the factory."

(I6) "Oh, Ginger," said Mother. "Where did you find him'?" "I did not find him," said Ginger. "He came to see me. He looked in and he liked me. Now we are friends."

(17) "Where is Richard?" said Mother. "Look, he is playing with the hose," said Father. "No. Richard, no, no! We will get wet!"

(i8) "What is he going to do now?" asked Jack. "Look at him," said Grandmother.

459


We note first that these sequences occur largely within ongoing stretches of talk or narrative and that we find relatively few address terms in them. The initiation turns pose the first speaker's problem fairly succinctly; they tend to be done in one utterance, but on occasion, as in (14) and (i6), the initiation turn is split into two parts by the narrator's speaker-identification. The narrator rarely gives speakers two turns in a row, although this may occur where the speech is fairly long. Speakers typically do not hold the floor very long. Andrew's speech in (15) is one of the lengthiest in the corpus. But answer turns overall seem to be longer than initiation turns. The preferred format of splitting a turn with the speaker-identification by the narrator gives many speakers two parts to their turn, that is, an answer slot and a continue slot (see answers in (14) and (I6)).

We note also that answer turns usually contain complete sentences ("Look, he is playing with the hose" rather than "Playing with the hose" in (17)). Often some of the words in the initiation turn are incorporated into the answer. (In- terestingly, in (io), Miss Brown's answer contains one of the rare violations of the "complete sentence" rule.) We suggest that this use of sentences and the frequent incorporation of some of the words in the question into the answer endorses a classroom model of ""answering" - how to answer in "instructional" talk. We will elaborate on this point in a later section.

Yes/No Questions. In this section, we will treat ""information" questions of the yes/no type. Overall, this is the larger type of questions in the entire corpus, as most of the questions in the ""request" and "permission" categories below are also of this type.

In the "home" materials, the yes/no questions take the following main forms: first "'Can you/can we/may you . . .'" beginnings - read as asking whether it is possible (ii instances) or advisable (i) or permissible (i) that something be done. Many of these are of the form "can you see/make, etc. (something)," which may lead to looking and then announcing the results, as in (19) and (20):

(19) "Yes, Mother," laughed Jack. "'You have to look for it. Can you see something new in here?" Mother looked and looked. "No, Jack," she said. "I cannot see it."

(20) Mummy says, "Can you see the train?" "Yes," says Jane. "I can see it down there. It looks like a toy train."

Second, there are twenty-four instances of the form "Is/are (subject, predicate)," as illustrated by (21) and (22), and seven instances of the form "Will (subject, predicate)," as in (23):

(21) "Is Big Brown Bear in it?" asked Father. "I like him." "Oh yes," said Jack. "He plays with the bunny and the mouse."

(22) "Is Jane up, Mummy?" says Peter. "No, she is not," says Mummy. "I will go and get her up."

460


(23) "'Grandmother," said Jack, "will the horses jump?" "Yes," said Grandmother. "Some horses are going to jump, and some are going to run. The horses are ready, Jack. Here they come!"

Additionally, there are fourteen instances of the form of "Do you have/know/ want/like (something)."

The answer turn typically includes a "yes/no" term plus some further comment, as our examples show. This further comment expands the answer from the minimal required for a two-part sequence. At the same time, the format of the question-answer sequences contracts the exchange into two tums, where in ordinary conversation, a first pair of turns might be given over to establishing attention, it "opens up the channel and functions as what might be called a pre- sequence exchange" (Wells 198I:35). For example, (23) might occur as:

Jack: Grandmother? Grandmother: Yes?

Jack: Will the horses jump? Grandmother: Yes, some horses are going to jump, and some are going to run.

The horses are ready, Jack. Here they come!

Sacks (1974) has commented on the prevalence of children's uses of opening routines such as "You know what, Daddy?" in talk with adults, to which the "proper and recurrent answer is 'What?' ":

we may take it that kids take it that they have restricted rights which consist of a right to begin, to make a first statement and not much more. Thereafter they proceed only if requested to. And if that is their situation as they see it, they surely have evolved a nice solution to it (Sacks 1974:231).

Similarly, Schegloff (1972:IO9I) describes the "summons" (e.g., "Grand- mother?") as a "particularly powerful way of generating a conversation." Sche- gloff notes Sacks's point about the elegance in children's use of the Summons- Answer device to have speakership returned to them via the Answer. These features of conversational openers are familiar aspects of child-adult talk, but are not represented in these books. It is perhaps as if no one had restricted conversational rights, and hence no need to use such a strategy. Although many of the question-answer pairs described here occur within episodes, where conversational participation has been established, some do begin episodes or stretches of talk.

The reader can look over other examples and imagine the possible expansions which could be done to represent the problem of coordinated entry into a conversation. It would appear that the question-answer sequences in this corpus are written as if attention and coparticipation in conversation could be assumed, rather than being a conversational problem. This aspect of the conversation in the texts is one of a number which may contribute to a model of child-adult interac-

461

CAROLYN D. BAKER AND PETER FREEBoDY

tion different from that which children might recognize as more familiar or more true to their conversational position as children.

We note also that the recipient of the question is named (usually in first position, as in (23)) in about 20 percent of the "home" sequences and in about 15 percent of the "school" sequences. Thus to the extent that the inclusion of an address term in first position serves as a summons or notification to the intended hearer (Wootton ig98a) integrated into the turn at talk, we find this possibly underrepresented as well. As the narrator is constantly identifying speakers, the use of address terms within the turns just as often would lead to a chaos of names. Still, it is just this format, relying as it does on the narrator's voice to zoordinate speeches, that adds to the economy and orderliness of the conversations portrayed.

In the "school" scenes, there are thirty sequences in the category of yes/no information sequences. Many of the questions seem to be framed to establish what is about to happen or could happen (e.g., "are you/we going to . "would you like to comre/have/tell us . . ."). Along with the "where is (person)" questions and the "how many (of you/us)" questions noted under the X- questions, we see that these question sequences describe the process of locating, assembling, and organizing people for activity. This could be read as depicting classroom life realistically, as an instance of the management of large groups of people for collective activity.

We also note that child speakers are made to do a fair share of this work; they do not merely respond to teacher initiatives. Example (24) illustrates this and also provides for comment on the nature of the "turn-taking system" in which question-answer sequences are central.

(24) "Robin," said Miss Brown, "'have you something for show-and-tell time'?" "Yes," said Robin. "It is in this big box. Susan, can you guess what it is'?" "Is it a toy'?" asked Susan. "Yes," said Robin. "Is it a toy car'?" asked Jill. "No," said Robin, "but it runs just like a car." "We cannot guess," said Jack. "'Tell us what it is." Robin took a toy bus out of the box. "When we lived on the farm, we went to the show," said Robin. "We went on a big blue bus. It looked just like this one.'" "We like your bus," said Miss Brown. "Someday we shall all go for a ride on a bus." "Peter has something for show-and-tell time," said Jack. Peter said, "I have a magic hat at school . .

462


In this sequence we see Robin taking over from Miss Brown and having every second turn from then on in the way teachers usually do in instructional talk. The sequence continues until Miss Brown concludes it and Jack nominates Peter as next speaker. It is usually the teacher's prerogative to nominate next speaker (Edwards 198I; McHoul 1978; Mehan 1979). We note, then, the portrayal of conversational initiatives taken by the children here and their distribution of turns to each other. We note also that Jill and Jack self-select as next speaker in this passage. Both of these events are rare in classroom talk. The talk in the books is, however, orderly, and its "sense of formality" (McHoul 1978) rests primarily on the maximization of the potential for gap and pause and the minimization of the potential for overlap, through textual layout and narrator interventions. At the same time, there appears to be considerable "local management" of the turn- taking system, that is, turns do not appear overall to be "pre-allocated." Thus, the operation of an orderly but locally managed turn-taking system is assigned to the control of child speakers in this particular passage. Children direct speakership in the corpus as a whole in that they initiate approximately two-thirds of all sequences; next speaker is often decided by the current speaker's directing a question to some specific other person. We do not examine the turn-taking system in detail here, but make these points to indicate the importance of the question-answer routines for the turn-taking system described in the children's readers. We also make these points in line with our concern to compare the versions of "child as speaker" in the books with the versions available through research on classroom and home talk.

Request sequences

This category includes those sequences which contain a request for assistance, favours, or specific action by the listener. Wootton's (I981a, X98Ib, X98lc)

work on address terms and request sequences involving four-year-old children and their parents is relevant to our analysis of "request" sequences and to our discussion of "permission" routines in the next section. First, we should point out that Wootton uses "request sequence" to refer to the whole range of request formats, while we have distinguished between "request" (usually "can/will/ you/we") sequences and "permission" (usually "can/may we/I") sequences (where "we" does not include the addressee). In our "request" sequences, we read speakers to be seeking to involve other(s) in some activity; while in our "permission" category, we read speakers (almost always children) to be asking that they be allowed to do something. We have noted enough of these latter, which to us are especially expressive of child-adult relations in the texts, to warrant treating them as a category in themselves for our particular purposes. Also, Wootton's materials contain exclusively children's request to parents (our "permission" category is largely this), while our materials contain a proportion of sequences involving the reverse, and also intragenerational request sequences.

463


Additionally some of our sequences are situated in "school" settings. While we will be drawing on Wootton's work more directly in the next section, some points with reference to his work on naturally occuring child-parent request sequences will be introduced in this section.

We have located twenty-one "request" sequences in the corpus, only two of which appear in "school" scenes. Hence, we will not distinguish in this section between "school" and "home" scenes. The three question forms we find in this category are: a) "can you (come, help, etc.)" (i i instances), b) "can we (have, come, look, etc.)" (5 instances), and c) "'will you (tell, get, etc.)" (5 instances). Some of the instances we have put into this category appear to be intermediate between "information" sequences and "permission" sequences to be discussed in the next section.

(25) "I like milking time. Can we come again?" asked Sue. Andrew said, "We will come every morning."

In (25), for example, we hear Sue making a suggestion in the form of request (i.e., the sense of "let's") rather than asking about permissibility or approval. This category, also includes five cases of ""will you . . ." initiations, such as (26):

(26) "Will you tell us some more stories about Rinaldo'?" asked Robin. "Next Saturday we may have another story about him," said Miss Long. "Now it's time for the other surprise."

There are also eleven instances of "can you come/help . . ." routines, such as (4).

Of the twenty-one initiations, eighteen are by children, two by parents, and one by the family dog. Thirteen are requests directed to adults, seven to children, and one indeterminate. Of the twenty-one sequences, twelve are crossgenerational sequences. Most of the initiations (I8/2I) are successful; that is, the request is granted.

In four of the sequences, the request is "'answered" by a descriptive sentence from the narrator, as in (27)-:

(27) "Please will you hose me?" said Helen. "Please will you hose me?" said Mark. Father hosed Helen and Mark. "Please will you hose me?" barked Boxer. Father hosed Boxer too.

In most of the sequences, the request is "answered" verbally by the speaker, as in (28):

(28) "Will you tell us something about it?" asked Susan. "It is a place where there are trees and birds and animals," said Miss Brown.

464


Usually, however, there is a yes term included in initial position, and some additional comments, as in (29):

(29) "Mother," said Jack. "I cannot take my green fish. It is too big. Can you come and help me?" Mother laughed. "Yes, Jack," she said. "Here is a big box for you."

The yes term is included in ten of the fourteen grantings done verbally by the second speaker, while the no term is avoided in two of the three nongrantings. This parallels Wootton's (198ib:6i-62) finding that "granting tokens" (yes or the equivalent) are stated, usually in initial position, while no terms are omitted: "refusals of requests are frequently softened, delayed, and accompanied by accounts" (see (26) for an example). Thus this aspect of natural conversation is preserved here. However, in the "permission" sequences to be discussed later, seven of the nine nongrantings contain a no term. Wootton's notion of a "preference for granting," which includes the idea that " nongrantings are often done in formats that avoid stating a refusal component" is overall not reproduced in these books, even though the sheer number of grantings far surpasses the number of refusals or nongrantings. The "preference for granting" is shown when "grantings and nongrantings are nonequivalently displayed" (Ig8Ib:63): In these texts, the same formats are employed in nongranting answers.

In most of these request sequences, children are asking another person to help or attend to them in some way. They do this far more often than adults do in these books. Still, in the context of our concern with the images of childhood and child-adult relations available in the texts, and even combined with the larger category of "permission" sequences discussed in the next section, they are not numerous compared with "information" sequences. It would therefore be inap- propriate to suggest that the primary model of childhood presented in these texts is a "dependency" model. However, these sequences of request and permission occur often enough to impute the presence of such a model as a secondary portrait of childhood. We will return to this discussion of how childhood is constituted in the texts.

The most frequent type of sequence within the category of "request" sequences is the one we have called the "can you come/help" routine. There are ten of these routines, all in the "home" scenes, six of them put by children to adults, three by children to other children, and one mother-father exchange. Examples (29) and (30) illustrate this type. All of the ten instances contain a positive outcome, that is, the request was granted.

These are interesting as the only sequences within this category in which an address term is contained in the initiation turn, usually in the first position, as in (30):

(30) "Jill! Jill!" said Jack. "Can you come and help me?" "Yes," said Jill. "The pets can play here. I can come with you."

465

CAROLYN D. BAKER AND PETER FREEBOD)Y

Eight of the ten routines, and eight of the twenty-one request sequences, contain an address term in the first part of the pair. We have noted that address terms are relatively rare in "information" sequences, and we will show later that they are about as frequent in "permission" sequences as in "request" sequences. About 30 percent of answers in "request" sequences contain an address term.

First-position address terms are understood as devices for securing attention or to "instruct recipient to monitor the words that follow" (Wootton i 98 ia: 153). In these textual representations, the narrator's voice (e.g., saidJ ack) usually occurs between the address term and the request proper, such that the "pause" so effected in the speech might mimic the speaker's check on whether the listener is attending or can hear. Hence the splitting of speeches with the narrator's speaker- identification suits the linguistic and interactional aspects of the speech. The listener, over the period of the "'pause" for the narrator's contribution, is assumed to have become ready for the request proper. This refers back to our previous comments on summons-answer sequences.

It is also the case that children position address terms in other than first position (as summons or notification to listen), such as in final position, where they behave like question tags to solicit a next turn from the named recipient, that is, to keep conversation going (Wootton 198 ia:143-44). It is rare in this corpus for an address term to appear other than in first position. This could be taken to suggest that (child) speakers in the books do not face the kinds of conversational problems for which, for example, the final-position address term can be seen to be a solution. It seems that the compacted format of the exchanges prevents the representation of an almost infinite set of conversational practices which characterize real child-adult talk, and in this presents a curiously distorted view of the child as speaker. Children may find these representations of talk artificial. Thus, in detailing these ways that the texts represent child-adult talk, we are attempting not only to show fine points of contrast or similarity with specific findings from naturalistic research, but also to suggest the possible significance of these for the overall model of child-adult talk which comprises much of the content of the books which are written to ease children through the transition from oral to written modes of language use, particularly the language of school texts. The portrait of the child as speaker and as social participant in the books nlay be a significant factor in some children's difficulties with relating to these texts.

The organization of the answers indicates that the attention was secured and the request heard. While we hear these requests essentially as requests for action (come, help), most respondents supply a more-than-minimal verbal answer about their agreement - a kind of verbal reflection on agreement.

(31) "Mother! Can you come in here'?" called Father. "'Jack and I have a surprise for you." 'Yes, Father, I can come now,"' said Mother.

466


In most cases, the words in the answer turn recycle those in the initiation turn. Thus the two speakers are made to share the same vocabulary, to take into account the precise wording of the initiator's talk. A community of speakers, children and adults, is thus fashioned, and a common language described. The possible linguistic distance between "children" and "adults" is minimized. This recycling may result in part from a need to repeat a limited number of words for purposes related to early reading instruction. Nonetheless, as we elaborate in the next section, the conversational turn supplied by the initiator to the respondent is used by the respondent to do more than provide a minimal answer to the query.

Speakers in the reading materials seem to attend carefully and consistently to each other. Answers always seem to display an understanding of what the question was about (see Walker [I98I] on mutual orientation in the practical accomplishment of orderliness and sequentiality in classroom question-answer sequences), even though they may appear to be producing compact, economical versions of real-life talk. Verbal messiness, hesitation, interruption, and mis- hearings never occur between speakers. The textual representation is a "tidied- up," textbook version of ordinary conversation, in which utterances express whole thoughts in grammatically well-formed sentences. The statements can be read to follow each other through repetition or clear pronomial reference to preceding statements. We suggest, then, that the number of turns it takes to accomplish a given interaction is reduced in textbook talk, but that each tum is packed with material which might comprise several turns in ordinary talk. This description, including our observations on the explicit repetition of the wording of the first-part utterance in the second-part utterance, is reminiscent of the claim that in classroom talk teachers encourage students to answer in complete sentences. This often means incorporating the phrasing of the question into the phrasing of the answer as well as being factually correct. Mehan (1974:88-X14) has described this as the "complete correct response." The strong implication is that the earliest school texts introduce a model of instructional talk in the guise of representing oral conversation.

Permission sequences

Permission sequences are those which typically take the form of a "can/ may, we/I" first part, which is usually a request to proceed with some idea for action, followed, usually always immediately in the text, by a "'yes/no" opening by the next speaker, plus some continuation of the topic within the turn. Examples are shown in (32) and (33).

(32) "Can we go to the hospital for them?" said Timothy. "Yes," said Father, "we will go for them in the car. Susan, you get ready. I will look after Timothy."

467


TABLE 3. Sources and rec ipients of permission questions in

nonschool settings

Number

Bv children to mother 15 father 14 grandmother 10

Total to grand/parents 39 to other adults 2

Total to adults 41 to other nonadults 4

Total asked by children 45 Bv others 2 Total permission questions 47

(33) "Father," said Jill. "May I come down and play?" "No, Jill," said Father. "You have red spots."

Not surprisingly, the overwhelming majority of sequences we describe as "permission" sequences contain a query addressed by a child to an adult. Table 3 presents the distribution of permission-seeking in the home and other nonschool scenes in our corpus.

In the "school" scenes in our corpus, we have located an additional eighteen instances of such questions, all from child speakers to the teachers. As this section proceeds we will provide comparative figures for the "'home" and "school" materials.

We note that no permission questions are asked of children by adults, although we can imagine that this does occur in family life. Only two permission questions are found in child-child talk, for example, may-l-join-you-(in-this-play) type. From other sources (Corsaro 1979, 1981) we might expect that this underrepre- sents the prevalence of the problem of access in actual play settings, although it appears that entry into play with other children is not usually attempted through "direct, verbal access strategies" which Corsaro (1979:323) suggests are more "adult-like." He also points out that in attempting access, "the probability of being ignored or receiving a negative response is much higher than that of receiving a positive response." Thus it appears that this interactional problem within child culture, perhaps along with many others, is not represented in these texts that appear to be about children and their world.

In forty instances of children seeking permission from adults in the "'home" materials, and in the additional eighteen instances in the "school" materials, the forms in which the requests are cast are summarized in Table 4.

Wootton (I98Ic) reports that the declarative ("I want") and interrogative ("can I") request forms are both prominent among four year olds, though it has been suggested that the use of the declarative decreases from ages three to six. As

468


TABLE 4. Forms of permission questions in home and school settings

May/Can I May/Can we May/Can (he/she)

Home settings 19 15 6 School settings 6 12

our analysis has been concerned only with question-answer pairs expressed in interrogative form, we have reported here only on the uses of the more indirect form. We find that "I want" is rarely used as a direct cross-generational request form within conversational sequences.

The speakers of the "may/can we" queries are almost always identified as individual children (14/15 in the "home" and 11 /12 in the "school"). In school scenes, the preference for the "we" form, which we interpret as introducing questions about what the class may or may not do, indicates that the children in school are shown to fashion themselves into a cohort, a community of interests, through this method of framing questions. They ask questions for some collec- tivity of children more often than for themselves as individuals, while in the "home" materials the children are more likely to speak as individuals or for other individuals. Appeals on behalf of another individual do not appear at all in the "school" materials. Where "we" is used in the "home" materials, of course, we find a similar fashioning of a cohort, typically of siblings in relations to parents. In the "home" materials, the preference for the "I" form could be accounted for in part by the absence of other child participants in the conversational setting. (This could apply in I3 of the I9 instances.)

An additional aspect of the form of the question part of the sequences we are examining is that in about 40 percent of "home" scene questions in this category, children name their adult listener usually in the first part of their turn, as in (33). However, we find no instances of the teacher being addressed by name in the permission sequence itself. Thus a model of permission-seeking in school is described in which the addressee can be assumed, while at home one often has a choice of parents. However, in some of the "home" sequences, only one parent is present but is still named in the query. In the corpus as a whole, address terms occur considerably more frequently in "home" scenes, accounted for particularly in "request" and "permission" sequences, the majority of which occur in "home" scenes.

We now turn to some observations on the nature of the adult responses to children's permission-seeking turns. Table 5 addresses the issue of the balance of positive and negative answers to children's permission-seeking attempts in the children's reading materials.

In Table 5 are cases where the sense of the response is "yes" or "no" even

469


TABLE 5. Positive and negative answers to childreni's permission seeking

Yes No Equivocal/other

Home

Father 9 4 Mother I1 2 I Grandmother 9 1 Other adults 2

School Teacher 10 3 2

Total 41 10 3

though the words ves or no are not used. This classification problem arose especially in the "school" materials, as in (34):

(34) "Can we go for one more swim'?" they said. Miss Pennyfeather said, "I want to go for a swim myself, but it is too cold. Look, here is a boat. The water will be in the pool for one more day. We cannot swim, but we can sail boats. Tomorrow we will have a boat day. Remember to bring your boats tomorrow."

Also in the "school" materials, we find questions (presumably) addressed to the teacher which are not directly answered, as in (35):

(35) "What are we going to do now'?" asked Peter. "May we eat now'?" "We are going to take a walk," said Miss Brown. "We are going to look for some of the things on our plan.'"

Miss Brown is heard here to have chosen to respond to Peter's information question in preference to his permission question.

In addition, one "'equivocal" answer by mother is the only instance in the corpus of a challenge from a child, shown in (36):

(36) Max ran to his mom and said, "I am six and I am big. Can I get a gun and caps'?" "Not yet, Max," said Mom. "Why not'?" said Max. "Run into the den, Max," said Mom. "It is not fun to go into the den," said Max. "Can I get a cap gun'?" But Mom said, "No, go into the den."

Apart from (36), we find very little indication of bargaining or challenging from children over adults' decisions throughout the corpus. In this respect it is instructive to note Wootton's (9g81b:72) summary that

470


requests are handled quickly by most parents, in one or two tums . . . in standard grantings, though it is not the case for nongranting sequences. In the latter, there is greater variability in length, some sequences being terminated relatively abruptly, others extending through many pages of transcript. A feature of many of these nongranting sequences is that children continue to make various appeals against the initial parental line on the issue, and in some cases of course, they win concessions.

With the exception of Max in (36), our child characters seem satisfied with "you have red spots" (example 33) or other rationales to be presented later, and do not engage in further negotiation. It is interesting, however, that Mom's "not yet" is unique in the corpus, and is an example of what Wootton would call a "weaker refusal form" than a no. This would seem to be a type of nongranting which would invite further discussion (perhaps more than a no) in the otherwise clear- cut world of permission seeking in these readers.

Permission questions result in positive answers much more often than in negative ones, the best chance of success lying with Grandmother. It should also be noted that many of the children's permission requests are framed, or more often placed, in such a way that a negative response would cast the adult as a rather mean and somewhat churlish character, as in (37), which is preceded by Grandmother and the children making toy puppets and talking about their activity.

(37) "Grandmother," said Jack. "May I make a little puppet too'?" "Yes, you may," said Grandmother. "Look in the box."

While the treatment of these exchanges as if they were actual conversations is illuminating, it must be kept in mind that these events are the products of an author writing for children. The maintenance of this realization helps to emphasize the role of these readers in portraying a version of the child-adult relationship.

It is the children who ask adults for permission in these readers. At the most general level, we have seen that adults tend to respond favourably. However, there are other aspects of the adults' responses to these initiations to exercise authority which inform the model of family life displayed in the conversational exchanges. We can see from the examples shown already that a format of the type "Yes/No," (Child), said Adult. "(Further talk)'" is most common. Mini- mal answers, for example "yes" or "no" alone, are rare. The permission- seeking move often occasions further speech by the adult in which: (a) the child's identity (name) is acknowledged; (b) the child's words are recycled in the follow- up to the "yes" or "no"; (c) the child's topic is pursued or elaborated; and/or (d) a reason is given for the "no" answer. The thirty-eight complete sequences found in the "home" and fourteen complete sequences in the "school" materials are summarized in Table 6, which shows the proportions that contain the various components.

471


TABLE 6. Percentages of adults' answers to permission sequences in home and school contexts containing various components

Yes/No Child's name Echo Elaboration Rationale

Home 97 32 50 66 1 I School 70 07 42 42 14

We characterize the components of the responses in the following ways: Yes/No. A clear decision is announced at the beginning of the response. In the

"school" materials there are a further three instances where the sense of the answer "yes" or "no" is inferable from the context and content of the response, as in (34).

Child's name. The inclusion of the child's name could be interpreted as evidence that the adult is attuned closely to the particular child who put forward the request. The answer is given back explicitly to that individual. However, teachers' answers to questions from individual children are almost always designed in such a way that they could be heard to be addressed to the whole class. Individual children are identified by name in doing the asking (question turn), but the absence of this term in the teacher's answer suggests that the query has been heard as being put on behalf of the group ("May we") and that the answer is designed such that all children could be the recipients. This is an aspect of the organization of the class as a cohort through the design of teacher talk (cf. Payne & Hustler 1980).

Echo. By "echo" we mean that some portion of the child's wording in the query is worked into the answer. The adult models a response, usually early in this part of the turn, on the child's choice of words. We consider examples (32),

(34), and (37) to contain this recycling of the child's wording, and would treat this as possible confirmation of having listened carefully to the precise vocabulary used by the child and found it suitable as a basis for the adult's choice of words. This device is discussed by Wells (I981:30) as "'the making of cohesive ties through repetition and pronominal reference."

Elaboration. By this term we mean that new material relevant to the topic initiated by the child is introduced in the adult's response. We comment further on what can be seen to be done in "elaboration" components below.

Rationale. In (33), we read Father's "You have red spots" as a reason for having just said "no"; we have heard no echo or continuation of the topic in that statement. Rationales will also be discussed further below.

There is only one example which we would describe as containing a topic change by an adult during the answer turn in a permission sequence: this is Mom's "go into the den" in (36).

Using our sketch of five possible terms in the answer turn, we have examined

472


TABLE 7. Number of components in adults' respontses to permission questions in home and school contexts

I term 2 terms 3 terms 4 terms

Home 2 15 18 3 School 5 5 4

each question-answer pair. For example, we find three terms in the answer turn in (38):

(38) "Mother," called Jack. "May I take a bowl for Mr. Whiskers? He likes the red bowl. "Yes," said Mother. "Take the red bowl. Take the little bed, too."

The italicized segment is the "echo" term constructed out of words from both parts of the child's query in this case; the material following we see as the "elaboration" term.

Table 7 is a summary of the organization of adult answers based on the criterion of how many different terms are used in the answer. It appears that adults in the "home" materials build in rather more components more often than do teachers. As the "elaboration" and "rationale" terms tend to be the lengthiest terms in these responses, it is worth noting that ten of the fifteen two-term responses in the "home" materials contain one or both of these; and that sixteen of the eighteen three-term responses contain one or both.

We find then that in "home" scenes, the adults' (usually the parents') responses to permission queries are typically constructed so as to (i) indicate a clear decision initially, (2) acknowledge the child by name (12/38) and/or the child's choice of words (19/38), (3) offer new material in an elaboration and/or rationale as part of the response (29/38).

In observing these features of adult answers, we are reminded of the point made by Wells and Montgomery (I98I:229): Adult speech which encourages language development "contains a larger number of acknowledgements of the child's contributions and a larger number of directives and utterances related to the child's current activity or to the joint activity of child and adult." In this respect, the parents in our readers have been assigned some highly preferred qualities. We suggest, however, that this has relevance primarily as a model of how children should answer teachers' questions, as we develop further and summarize below.

If the representation of convelsational interact.on in these texts can be seen as providing a model for turn design within permission sequences, they can be seen as a source of continuing language socialization. They show how permission requests can be framed by children and offer an image of family life in which

473

CAROLYN D). BAKER AND PETER FREEBODY

parents respond in a way which acknowledges the child or the child's use of language. They also demonstrate how permission exchanges constructed in these ways usually amount to the request being successful. By seeking permission every so often, the children are portrayed as dependent on adult decision making, and at the same time they are assigning to parents and other adults the conventional position as persons who are (to be) approached in this way by children. Initiating the permission routine fashions an opportunity for the adult to exercise a conventionalized responsibility for children. Thus we find a clear underwriting of a model of social relations between children and adults.

To develop this point, we now look at what parents do in their turns when they opt for adding the "elaboration" or "rationale" terms to their answer turn. We would suggest that the "elaboration" term can be read not only as a continuation of the child's topic, but as an elaboration of the relations expressed in the very accomplishment of permission routines. We can read these routines as instances of children handing the adult an opportunity to exercise authority, and as instances of adults taking the opportunity to add something which is elaborative of the status implied in having been asked to decide in the first place. There are twenty-five "elaboration" terms in our home materials. In fifteen of these, the term contains some reference to what will be done next by the child or the adult, for example, "Make the big spot red and the little spot green" (Father to Jill's "May I help" make the puppet?) or "Take the yellow bowl and go down for some apples. You and I can make something good to eat" (Grandmother to Jack's "May I help'?"). In twelve of these, the adult uses the imperative form. In seventeen ot the elaborations, the adult's comment can be heard as organizational, for example, "He can sleep with you, Bill"; "Swim where I can see you"; "Now you can come and help me"; "Here is something for the mouse." The elaborations develop both the topic and the parents' rights to pronounce and organize. Addi- tionally, there are a few cases where the opportunity is fashioned out of this conversational option to make a further "adult" judgement, as in "Milk is good for you" or "Dan can play in the shed but he cannot jump and run," which might also be heard as examples of making a yes answer an accountable matter: "why I said yes" (cf. Wootton 1 98 1 b:72, who claims that in ordinary talk it is refusals and not grantings which are made accountable matters). In these materials we do not find a display of a "fundamental non-equivalence for participants of the actions of granting or rejection" (Wootton ig8ib:70); rather we find that very similar formats are used.

The ""rationales" are all pronouncements about a state of affairs which should, once made available to the child, count as adequate and reasonable grounds for saying "no." The "fact" produced in explanation is presumably knowledge to which either the child would not have prior access or has been inattentive. Besides the sequence in (33) we have three other "home" sequences which we find to contain a rationale:

474


(39) "1 like the one called Bumpity," said Jack. "May we have it again?" "No," said Grandmother. "it is time for you to go now."

(40) "Can we go fishing?" said Michael. "No, the sausages are ready," said Father. "You can go fishing tomorrow. "

(4I) "Father," said Jill. "May Sandy come up here and play'?" "No, Jill," said Father. "Sandy is too big."

Our general point is that children invite displays of adult competence from adults by initiating permission routines. Adults use those occasions to demonstrate decision-making and knowledge skills, to provide what the child has sought, to confirm the adult identity carried in having been asked, and to endorse the act of asking. Hence a particular version of the seeker's identity as child vis- a-vis adult is constituted.

SUMMARY

In the entire collection of question-answer pairs, children ask most of the questions: In these texts, children frequently initiate talk to each other and adults; they do not wait until they are spoken to. Through their question initiations they formulate problems, pursue their curiosity, make suggestions and seek help, organize their environment and approach others, and ask permission (thus recog- nizing their dependency). We do not, overall, see that this reflects the reactive speaker roles which have been noted for children in instructional talk in classrooms (i.e., teacher initiates, children respond, teacher selects next speaker). Children in the texts have a more active part in the generation of sequences of talk.

Also, the question-answer sequences in these readers reveal a good deal more crossgeneration talk than there is child-child talk in all categories of sequences, particularly "permission" sequences. That is, children talk with adults rather more than they do with other children. This can be seen as an aspect of socialization into adult culture, which is being effected throughout these texts in different ways, even though the texts are ostensibly "about" children and their world. Their world is populated by adults and especially parents. As we have pointed out elsewhere:

seven of twenty [most frequent common nouns referring to living creatures in the entire corpus] describe roles in the immediate family, and six of these seven refer to parents. A very large portion of the text, then, seems to deal with events set in the context of the child's immediate family . . .

. . . the prototypical family is portrayed as a small nuclear family. A grand total of 1,382 kinship terms appear in the entire sample. This, on

average, means that a family member, usually a parent or grandparent, ap-

475


pears about every sixty words across all of the texts . . . this is a strikingly high degree of frequency (Freebody & Baker 1985:388-89).

Our finding of the issue of crossgeneration dialogue thus supports the in- terpretation based on a simple word count of persons named or identified in the corpus; that is, that children are connected to the adult world through these texts. Adults are on the scene frequently, and they get talked to more frequently than do other children on the scene. An introduction to literacy through these texts, therefore, is simultaneously a strong linguistic engagement of children with adults in the texts, as shown particularly clearly by examining the conversational patterns. The children and the adults in the texts talk the same way, and they talk the "special language" of school texts (Olson 1977:76), a facet of which we have attempted to describe. In this way, we view the books as "transitional" between conventions of oral language and instructional prose.

It might be suggested more strongly that these texts are not "about" children at all. As Jenks (1982:13, 23) has written, concerning adult-produced discourse about children:

much of what is said about children and childhood is not really about children and childhood at all . . . in significant ways, the child within the social sci- ences is employed as a device through which to propound versions of social cohesion.

The idea of childhood is not a natural but a social construct; as such its status is constituted in particular socially located forms of discourse . . . the child is constituted purposively within theory; that is, the child is assembled intentionally to serve the purposes of supporting and perpetuating the fundamental grounds of and versions of man, action, order, language and rationality within particular theories . . . different "'theoretical" children serve the different theoretical models of social life from which they spring.

We have adapted this idea to the "theoretical child" and to the "'theoretical model of a speech community" available in the texts. On the point of what "child" is assembled in these texts, we have focussed here on the model of the child as conversationalist and have shown specific points of contrast between the textual portrait and details available from naturalistic research. We have suggested that differences between the child speaker as constructed by the texts and the child reader's experience of verbal interaction in real-life settings might be a barrier to children's ease of relating with the texts and possibly a subtle source of reading difficulty.

Our further suggestion is that the textual representation of the organization of talk, and in particular child-adult talk, endorses a model of turn construction and initiation-response sequences which is compatible with the instructional purposes of schooling. We noted that the number of speaker turns in question-answer sequences seem to be contracted in comparison with naturally occuring home

476


talk, and that speakers make maximal use of their turns. This parallels Mehan's point that in classrooms, most children have few turns, and once children get the floor, they need to use their chance to display competence in form and content (1979:137-39).

We also observed that answers in question-answer pairs are designed to correspond to the design of the questions: The answer incorporates words from the question and is expressed in complete sentences. We see this as illustrating the joint production of propositions, which Heap (1985) has discussed as one feature of doing reading lessons, and as illustrating the preference in instructional talk for answers to be expressed in whole sentences (cf. Mehan I974:84-87). In these ways, then, people ask and answer questions in ways which accommodate the "form" requirements of classroom questioning.

Further, Romaine (I984:177), overviewing a number of studies relating to children's acquisition of oral conventions for classroom purposes, points out that when children are relating events and experiences to the teacher or class, "the school narratives expected by the teacher are book-like in that details have to be fully lexicalized and explicit and are in certain respects far removed from the narratives of everyday life." We could take this to suggest that children are (implicitly) requested to talk in a way that prepares them for an acquaintance with written language. This would be the reverse of and complement to our point that children's texts represent talk in a way that suits oral language practices in the classroom.

With respect to the nature of the dialogue portrayed, we noted that speakers talk as if they were closely attending and highly attuned to each other; questions are answered immediately in the next turn of talk; conversational tasks are accomplished economically and tidily; there are few irrelevances, interruptions, or hesitations. Speakers seem to face few conversational problems such as getting attention, opening a conversation, or having cospeakers go off the topic. Indeed, the talk seems to be remarkably like Wells's (1981:29) description of "talking to a topic" (which occurs in more formal settings) and unlike "talking topically" (in which conversation "shifts and changes" as the talk continues).

The patterns of question initiations, however, do not support those found in research on teaching and classroom talk in that children do initiate talk by asking questions and often select the next speaker in their questions.

There are many other aspects of turn construction and the turn-taking system portrayed in the texts which would need to be examined further to come closer to a conclusion about possible differences between how people really talk (in and out of lessons) and how they talk in the children's books. For instance, we noted in these materials the absence of a third component to the question-answer sequences, that is, we observe the pattern Question-Answer, Question-Answer. Third components (parents' or teachers' comments on children's answers) have been found commonly in research on the organisation of adult-child talk: The pattern is Initiation-Response-Evaluation (Coulthard I977; Mehan 1979). The

477

C'ARoLYN D. BAKER AND PETER FREEBO[)Y

absence of the third component in the children's readers seems to be related to the almost complete absence of "pseudo questions," questions to which the asker already knows the answer and in which the point of the questions seems to be to establish whether somebody else knows it too. The third part signals authority (Edwards 1980:234-40) and effects control over the turn-allocation system (McHoul 1978). This aspect of child-adult talk is not reproduced in the readers. Its absence here bears on the portrait of child-adult relationships in the text. The implications of characteristics such as this need to be more fully pursued.

A final point about the portrait of child-adult relations in the children's texts: The integration, in the readers, of the talk of children and adults makes them, tor the most part, equal partners in conversation. Linguistic similarities and conversational cooperation are strong features of the conversations portrayed, and children get a large share of turns. This is not compatible with Speier's (1976) observations of adult conversational dominance over children in actual interactions. Even in "school" scenes in the children's books, children do much of the asking of questions, and everywhere adults are benevolent, attentive, and pre- pared to follow children's choices of topic and activity. The texts combine this image of the "child-centred" family with due recognition of conventional roles and responsibilities of parents, as in "permission" sequences. Thus the authority of the adult is prevalent, but benevolent and gentle. We might conclude that similarly, the authority of the text, in introducing a tacit model of language and interaction for schooling, is intricately constructed and subtly conveyed.

CONCLUDING NOTE

As our main purpose in the paper has been to analyze the nature of question- answer pairs in the corpus, we have given little attention to how such pairs are integrated with other dialogue into "stories." This is a substantial and distinct topic in itself, and we provide here only an outline of how some of our observations could be used in such analyses. Initially, we can build on our analyses of isolated question-answer pairs with an example of their use in combination with other forms of dialogue, and with the narrator's contribution. We will use "At the Camp" as our example, and present the text as it appears in the reader, where in the previous sections we have made each turn at talk a line of dialogue. Presenting this example in this way also points to the complication, for all our analyses, of textual layout, but we will not address that issue here.

At the Camp

(42) I) "Who will help me (2) with the tent'?" (3) said Father.

478

QUESTIONING AND ANSWERING IN FIRSTr SCHOOL BOOKS

(4) "I will," said Michael. (5) 'Penny and Brian (6) will help me make the beds," (7) said Mother. (8) The tent is up. (9) The beds are ready.

(lo) "Can we go fishing?" (i I) said Michael. (12) "No, the sausages are ready," (13) said Father. (I4) "You can go fishing tomorrow." (I5) "Is tea ready?" (i6) said Father. (17) "Yes," said Mother. (i8) "Where is Brian'?" (ig) "Here he is with Bimbo," (20) said Father.

In the first question-answer sequence, we read "I will" as the answer, after having scanned the two lines following, which could have been a continuation of the answer, and then line (7) to decide that this speech (lines (5) and (6)) must belong to Mother. Thus the identification of lines (I )-(W) as a question-answer pair is not settled conclusively until the reader arrives at line (7). The problem here tums on the narrator's device of identifying speakers after their turn has begun. We have pointed out this problem of speaker-identification elsewhere (Freebody & Baker 1985). A similar problem occurs later in the text: Who says "Where is Brian'?" (I8) and how is it decided? Retrospectively, we (and proba- bly the child reader) attribute that utterance to Mother on the basis of punctuation and format clues; it must be in the second part of Mother's speech, since (i) if Father (next speaker) asked and answered his own question, we might not have quotation marks between "'Brian" and "here" (there is one instance in the corpus where a speaker answers his own question); (2) writers of these texts seem to prefer two-part answers and utterances split by said Mother or equivalent; (3) answerers in our corpus usually do more than yes or no with their answer slot; (4) speakers are almost always identified somewhere, so it must be Mother or Father, who are the immediate speakers in this episode. Are these. punctuation and format clues made explicit in the teaching of reading? Actually, ignoring these clues which arise out of our familiarity with text, anyone on the scene could have said the lines "Where is Brian?" and "You can go fishing tomorrow."

Additional to the problem of speaker-identification and the identification of pairs is the work child readers must do to make the answer to a question sensible as an answer. The second question-answer sequence in the story is in lines (io) to

479


(04), which starts the next round of talk after the narrator has announced the results of the first. There is an implied connection between going fishing and sausages being ready, which requires the importation of a good deal of cultural knowledge. We can well imagine a teacher asking ""Why couldn't they go fishing today?" asking the children to use the text as source, or as a basis for inference (Heap 1985), to find or produce a connection between these items. Unpredictably, Father then asks in the third question-answer sequence (lines (15)-(17)) whether tea is ready. Hence, there is a need to distinguish between sausages being ready and tea being ready, that is, tea is more than and inclusive of sausages.

The final question-answer pair (lines (i 8)-(20)) overlaps with the second part of Mother's turn in lines (17) to (18): Mother's one turn here includes both a response and an initiation. Wells (I98 I:31) suggests that such "'linking" devices often accompany topical continuity, but here Mother uses her turn to conclude one topic and open another. Following stylistic rules we might have expected instead a continuation such as "Tea is ready now." Also following stylistic rules derived from our reading of the corpus, we would have expected line (1 5) to be said by someone other than Father; two turns in a row are relatively rare. We see then, how closely the text must be read to work out who said what to whom and how dependent we are on the narrator and on punctuation to cope with both regular and irregular formats.

Finally, how does a reading of the text make people talk to each other, rather than generate random utterances'? Since the narrator does not always provide explicit identification of speaker and intended respondent for each turn, readers need to rely on various clues to fashion utterances into pairs. The question- answer pairs are identified as pairs by readers following the points made earlier in the paper: through the repetition of words (who will? I will), through adjacent position in the text, and through the expectation that questions do get answered usually in a single next turn. However, the task of reading is further complicated by the observation that the utterances and conversations which make up the "story" could be assembled differently. If we were to extract Michael's "I will" from its position, and enter it somewhere else in this dialogue (two possible spots are easily recognizable), Mother's first turn could be heard as a response, but not an answer, to Father's first tum. In its present position, it is not clear (to us) to whom this remark is addressed: It is not tied through adjacent position to Father's first query, though it repeats the "help me" phrase. It could have come before Father's first turn. The conversational "'episodes" within our example have been put into a particular sequence of conversational turns which comprises the story line. Indeed we could mix up the sequence of these turns at talk in various ways and still ask children to make sense of them (cf. McHoul 1982).

We make these points to emphasize our claim that the reading of conversation

480


is a complex task, in which understanding the arrangement and coherence of utterances within a story or episode requires identification of adjacency pairs, such as question-answer sequences, and a preparation for announcements (such as Mother's) whose interactional target cannot be found in text alone.

The child's ability to bracket pairs of utterances into question-answer sequences (or other adjacency pairs) is crucial to the understanding and discussion of the story as any experienced reader (e.g., teacher) would be able to do. We have attempted to show how the representation of talk could variously clarify or confuse the child's grasp of the sequence and logic of this talk-on-paper.

Analyses such as these are critical, given that in the comprehension aspects of reading lessons, the principal concern seems to be with the understanding of what is going on among the characters rather than with the decoding of words. It seems to us that the representations of conversational interaction in the texts do not reflect how children talk with each other and with adults in everyday life. Therefore, to assume that children can recognize and follow these representations easily is unwarranted. We would suggest that putting conversation on paper reshapes the conversation, giving it a format and logic which are not identical to ordinary speech exchanges, such that the textual representation of conversation is transitional between conventions of oral conversation and conventions of written school materials. Children are asked to read and follow representations of possible oral conversations which are lodged in the language conventions and culture of the school and the literate world. This is both a powerful source of socialization into literacy and a possible source of reading difficulties.

Our analysis of the task of reading conversational exchanges suggests the possibility of making explicit to children what some of the conventions for representing talk are, to recognize this as a possible problem different from those usually considered, for example, word recognition. This in effect would entail making the relation between spoken and written language a topic in itself at the earliest levels of schooling.

The problem of differences between images of childhood and child-adult interactions portrayed in the books, and children's own self-conceptions and experiences as conversationalists is ultimately a problem of the relation between children's culture and the school-literate culture, as our analysis has sought to show. We have suggested that this aspect of beginning school texts contributes to the socialization of children into membership of the classroom language community. It might be argued from our analyses that early reading books could be written to more closely represent children's conversational practices and problems as naturalistic research has shown them to be, and hence to portray a different version of childhood. However, such a rewriting would have the effect of aligning the materials less closely with the particular ways of using written language to represent knowledge and experience which presently forrn the basis of school literacy.

481


NOTES

I. McHoul (in press) has recently proposed that the analysis of fictional materials become part of the program of conversational analysis, which to date has used (transcripts of) naturally occurring conversation as its material. This paper offers an illustration of this proposed application of conversational analysis to written materials. 2. The full Materials Reference List can be found in Freebody and Baker (1985). 3. The total number of sequences discussed in the paper is 290. This varies from the 300+ instances mentioned of questions being asked or answered. There are some questions which do not get answered or whose answerer is impossible to determine; there is one "answer" without a "iquestion" having been asked; and, on occasion, two "questions" receive one collective "answer" or one "question" generates more than one "answer." 4. In compiling figures for questioners and answerers we have noted that the initiator or recipient of the query is not always explicitly named in the text. In these cases, the reader has to work out who the speaker and listener must be from format, content, and/or context. In the case of questions where the intended answerer is not named, our coding has been based on observing who answers, and treating the question as having been addressed to that person. The same problem applies to identifying the recipient of the answer. In some cases, we were unable to decide, from the text alone, to which other speaker an utterance was directed. We have not included in our figures such instances where our cultural and linguistic resources did not give us grounds to make a good bet. 5. We have included in our materials a few instances of conversation with family pets where such conversation is integrated with other talk, as it is here through the parallel structure of the two parts of this scene.

REFERENCES

Baker, C. D., & Freebody, P. (in press). "Constituting the child" in beginning school reading books. British Journal of Sociology and Education 8.

Corsaro, W. A. (1979). "We're friends, right?": Children's use of access rituals in a nursery school." Language in Society 8:315-36.

(I98I). Entering the child's world - Research strategies for field entry and data collection ina preschool setting. In J. L. Green & C. Wallatt, (eds.), Ethnography and language in educational settings. Norwood, N.J.: Ablex. II 7-46.

Coulthard, M. (1977). An introduction to discourse analysis. Harlow, Essex: Longman. Edwards, A. D. (I980). Patterns of power and authority in classroom talk. In P. Woods (ed.),

Teacher strategies: Explorations in the sociology of the school. London: Croom Helm. 237-53. (I98I). Analysing classroom talk. In P. French & M. MacLure (eds.), Adult-child conver-

sation. London: Croom Helm. 289-306.

Freebody, P., & Baker, C. D. (I985). Children's first school books: Introductions to the culture of literacy. Harvard Educational Review 55(4):38I-98.

Halliday, M. A. K. (1978). Differences between spoken and written language: Some implications for literacy teaching. In G. Page, J. Elkins, & B. O'Connor (eds.), Communication through reading. Adelaide: Australian Reading Association. 37-52.

Heap, J. L. (I985). Discourse in the production of classroom knowledge: Reading lessons. Curricu- lum Inquiry. 15:245-79.

Jenks, C. (I982). Introduction: Constituting the child. In C. Jenks (ed.), The sociology of childhood: Essential readings. London: Batsford. 9-24.

Klare, G. R. (I975). Assessing readability. Reading Research Quarterly I0:62-I02. MacLure, M., & French, P. (I98I). A comparison of talk at home and at school. In G. Wells (ed.),

Learning through interaction. New York: Cambridge University Press. 205-39. McHoul, A. W. (0978). The organization of turns at formal talk in the classroom. Language in

Society 7:I83-2 13.

(I982). Telling how texts talk: Essays on reading and ethnomethodology. London: Rout- ledge & Kegan Paul.

(in press). An initial investigation of the usability of fictional conversation for doing conversational analysis. Semiotica.

482


Mehan, H. (1974). Accomplishing classroom lessons. In A. Cicourel et al. (eds.), Language use and school performance. New York: Academic. 76-1 42.

(1979). Learning lessons: Social organization in the classroom. Cambridge, Mass.: Har- vard University Press.

Olson, D. R. (1977). The language of instruction: The literate basis of schooling. In R. C. Anderson, R. J. Spiro, & W. Montague (eds.), Schooling and the acquisition of knowledge. Hillsdale, N.J.: Lawrence Erlbaum. 65-89.

Payne, G., & Hustler, D. (1980). Teaching the class: The practical management of a cohort. British Journal of Sociology of Education 0 ):49-66.

Perera, K. (1982). Some language problems in school learning. In N. Mercer (ed.), Language in school and community. London: Edward Arnold. 3-29.

Romaine, S. (1984). The language of children and adolescents: The acquisition of communicative competence. Oxford: Basil Blackwell.

Rubin, A. A. (1980). A theoretical taxonomy of the differences between oral and written language. In R. J. Spiro, B. C. Bruce, & W. F. Brewer (eds.), Theoretical issues in reading comprehension. Hillsdale, N.J.: Lawrence Erlbaum. 411-38.

Sacks, H. (1974). On the analysability of stories by children. In R. Turner (ed.), Ethnomethodologv. Harmondsworth: Penguin. 216-32.

Schegloff, E. A. (1972). Sequencing in conversational openings. American Anthropologist 70(6):1075-95.

(i984). On some questions and ambiguities in conversation. In J. M. Atkinson & J. Heritage (eds.), Structures of social action: Studies in conversation analysis. Cambridge Univer- sity Press. 28-52.

Speier, M. (1976). The child as conversationalist: Some culture contact features of conversational interactions between adults and children. In M. Hammersley & P. Woods (eds.), The process of schooling. London: Routledge & Kegan Paul. 98-103.

Stubbs, M. (1983). Discourse analysis: The sociolinguistic analysis of natural language. Oxford: Basil Blackwell.

Walker, S. (1981). Teacher-pupil question-answer sequences: Some problems of analysis. In L. Barton & S. Walker (eds.), Schools, teachers and teaching. Sussex: Falmer. 265-82.

Wells, G. (1981). Learning through interaction: The study of language development. Cambridge University Press.

Wells, G., & Montgomery, M. (1981). Adult-child interaction at home and at school. In P. French & M. MacLure (eds.), Adult-child conversation. London: Croom Helm. 210-43.

Wootton, A. J. (1 98 i a). Children's use of address terms. In P. French & M . MacLure (eds.), Adult- child conversation. London: Croom Helm. 142-58.

(1981b). The management of grantings and rejections by parents in request sequences. Semiotica 37(1 /2):59-89.

(1981c). Two request forms of four year olds. Journal of Pragmatics 5:511-23.

483

Representations of Questioning and Answering in Children's...

Documents

Transcript of Representations of Questioning and Answering in Children's...