M A R A I G N AC I A M A S S O N E A N D MNICA BAEZ

Deaf Childrens Construction of Writing

High illiteracy rates among the Argentine deaf population, even after long years of schooling (Massone, Simn, and Gutirrez 1999; Massone, Simn, and Druetta 2003), point to the need to revise certain approaches to deaf literacy, particularly in school settings. Qualitative change in deaf literacy requires the use of multiple conceptual tools if learners are to be able to tackle its complexity without reductionism or oversimplication. We dene illiteracy as the absence of knowledge that involves but is not conned to graphic marks. It has been contended that the term may also apply to the difculty one experiences in interpreting and using written materials in a variety of contexts, as well as the inability to take part in a literate culture despite having mastered its written symbols (Ferreiro 2000). Various studies have focused on different aspects of the conquest of the written language by deaf children and teenagers.1 Stressing their competence in and need for visual communication, this research therefore calls for the rejection of oralism in favor of the new ways of knowing made possible by todays essentially visual media and multimedia.
Mara Ignacia Massone is at the Centre for Research on Philosophical and Cultural Anthropology (CIAFIC), Linguistics Department. She is also a senior researcher at the Argentine National Council for Scientic and Technological Research (CONICET). Mnica Baez is at the Rosario Institute of Education Sciences Research (IRICE), which is afliated with the National Council for Scientic and Technological Research (CONICET). We wish to thank Mnica Descalzi for her assistance with the English version.

457
Sign Language Studies Vol. 9 No. 4 Summer 2009

458 | Sign Language Studie s

In speaking of writing as a language or a mode of language, we mean far more than simply communicating. We are also referring to the making of meaning, to the interpretation of cultural practices, and to the reconstruction of the representations that dene the family and culture in which every person is subjectively and socially embedded. Thus, literacy should be encouraged as a way to promote integration, and the processes that deaf children engage in to develop it deserve close attention. In order to account for the specic features of the cognitive and linguistic processes that deaf people utilize when dealing with written language, we draw on two perspectives that have revolutionized the traditional understanding of the factors at stake in deaf literacy. These are the socioanthropological view of deafness and the psycholinguistic theory of writing, which is based on psychogenetic studies (Ferreiro and Teberosky 1979). Written languages characterize the practices, representations, and discourses of the literate hearing societies in which deaf communities are embedded. Deaf children, to whom written language is a second language, are linguistically, communicatively, and pragmatically competent in their own natural sign language. We explore deaf childrens literacy and the role that sign language plays in the reconstruction of written language. The present study is part of a wider research project that focuses on how these children develop literacy skills. The article concentrates on the cognitive and linguistic processes involved in understanding written Spanish. It also discusses the conceptual schemata of language learners for whom sound does not constitute a source of information. They are already competent, nevertheless, in their own natural language (i.e., sign language), which presupposes a distinctive type of linguistic organization (Massone and Machado 1994). This article analyzes the interpretation of an illustrated text by deaf children who had had no oral training. The conclusions we draw are based on data obtained at the exploratory stage of this ongoing project. Although they are therefore provisional, we offer them here to encourage the revision of deaf literacy practices.

Literacy and the Deaf Community

In Latin America and other parts of the world (e.g., Spain and Portugal), the prevailing behaviorist and neocognitivist models regard oral-

Deaf Childrens Construction of Writing | 459

ity as the gateway to deaf literacy. They oversimplify the relationship between oral and written language, reinforcing ethnocentric and monocultural representations and ignoring the complexity of the dialectical process that occurs between the knowing subject and the written language as a specic object of knowledge. Some studies (Massone, Simn, and Gutirrez 1999; Domnguez Gutirrez 1999) focus on how best to teach deaf people to read and write by discussing educational experiences and suggesting various approaches and methods. Other studies consider deaf peoples written production from a descriptive point of view and focus on mistakes (e.g., Fernndez Viader and Pertusa 1995). Still others deal with the causes of mistakes (i.e., with the relationship between the levels of oral and written competence) (Newport and Meier 1985; Bells and Teberosky 1989). Bells and Teberosky examine the difference in verbal competence between deaf and hearing children and its inuence on early representation of the written language. To Massone, Simn, and Gutirrez (1999) and Massone, Simn, and Druetta (2003), this literature points to the need for a constructivist approach to deaf literacy. The former conducted an interesting experiment at a Mendoza, Argentina, school, where the constructivist approach was used to teach deaf children who had little reading and writing experience. Starting with simple sentences, the students learned how to write long texts within a year. Instruction was given in Argentine Sign Language (LSA). Particularly relevant to this article is Ferreiro and Teberoskys work (1979) with children who are learning Spanish and other alphabetic systems (e.g., Portuguese, French, Hebrew) and who come from a variety of sociocultural and economic backgrounds. Their research, grounded in Piagetian theory specied three objectives: (a) identify the cognitive processes that underlie literacy acquisition; (b) understand the nature of childrens hypotheses; and (c) ascertain the specic knowledge with which children begin their formal education (Ferreiro and Teberosky 1999). After more than ten years of study, they were able to describe the original and relevant hypothesizing processes that enable hearing children to reconstruct the alphabetic nature of the literacy system used in their social and cultural environment. Thus, they revealed the conceptual process by means of which individuals learn to make sense

460 | Sign Language Studie s

of written texts and to reconstruct and reformulate possible writing systems instead of isolated words. These nonconventional systems show an understanding of certain aspects of the writing system and the internal logic of childrens arguments.2 The heuristic value of Ferreiro and Teberoskys work has led us to view this learning process as a psycholinguistic theory. As such, it denes new issues, poses new questions, and provides the conceptual and methodological tools with which to tackle them. As Ferreiro herself wrote of Piagets production, the most profound contribution of a theory does not lie in the truths it establishes but in its capacity to generate new questions (Ferreiro 1999, 55). In line with these and other historical, linguistic, and discourse analysis ndings, we view literacy as a social, cultural, and historical ability that goes beyond the mere transcription of oral language. It is also a symbolic object insofar as the writing system constitutes a language representation system. Thus, literacy should be seen rst and foremost as an intellectual achievement rather than a trainable skill. Literate individuals acquire a particular way of understanding the world because they comprehend how graphic marks can represent speech; in other words, they know how to distinguish the written from the nonwritten. From this theoretical standpoint, literacy constitutes an intellectual achievement rather than a trainable skillthis latter perspective is sustained even nowadays by oralism in our country. It may also be interpreted as a semiotic function since it enables a specic comprehension that entails learning how to represent what is said by means of graphic marks while distinguishing them from what is not written. In this respect, we can speak of children who are capable not only of reading and writing according to their own conceptual schemata but also of reecting on what is read and written. Such writers make written language their own, a process that involves cognitive skills, linguistic competence, and the social conditions that determine access to literacy. We view Spanish literacy learning by deaf students as second-language learning. The interpretation of the linguistic and graphic features of written Spanish involves the recategorization and reformulation of the knowledge available to these students in the sign language they speak, which possesses specic syntactic characteristics (Massone and Machado 1994). We do not seek here to conrm the hypotheses formulated by

Deaf Childrens Construction of Writing | 461

previous researchers but to understand the conceptualization process that we assume both deaf and hearing children carry out. We are interested in the constructive activity of deaf learners, although we do not rule out the possibility of our coming across such writers categories.

Graphic Segmentation of Written Texts

The psychogenetic research mentioned earlier shows that one of childrens rst conceptual achievements lies in viewing writing as a substitute object that differs from drawing, which constitutes yet another representational system. In a literate culture, both systems share graphic spaces such as books, billboards, posters, and magazines, all of which are made out of what literate people identify as pages. These graphic surfaces are the privileged site of text inscription. The text appears graphically as a visual whole divided up into variously distributed parts of different lengths. Thus, headings, subheadings, paragraphs, sentences, and words constitute meaningful culturalnot naturalsegments. Among the different kinds of text segments that literate, hearing adults regard as natural, we should consider in more detail the spaces between words. These, together with other graphic elements, enable readers to interpret what has been written. Such textual conventions can be traced back to the institutionalization of text morphology in Western literacy tradition. They are the outcome of social and historical transformations that have affected the ways in which people read and write and knowledge is passed on, widening the gap between oral and written language. By considering that the meaning of a text is inherent in both our text organization patterns and language use structures, and that those patterns have, in general, a social origin, we understand the need to distinguish between literacy and textuality. This distinction makes it possible to approach the various problems children face in their acquisition of the former, as well as the historical processes that inuence the ways in which the latter manifests itself nowadays. An exhaustive discussion of these processes is beyond the scope of this article. Sufce it to say that the graphic division of information into words was introduced around the end of the seventh century by Irish and Anglo-Saxon monks who spoke Celtic and Germanic languages

462 | Sign Language Studie s

and recorded information by writing it down. Since, for them, Latin constituted a visible language (Parkes 2001), they followed Latin grammarians morphological criteria and thus altered contemporary graphic conventions and therefore texts, which until then had been characterized by scriptio continua. Another fundamental change also took place as a result: Silent reading became valued. Up to that time, reading out loud, in either a loud or a low voice (ruminatio), had been the norm. The slow and difcult institutionalization of blank spaces between words, then, modied not only the material appearance of texts but also the very notions of writing and reading, as well as of writer and reader. The theoretical units that linguists developed to identify words (e.g., morpheme, moneme, minimal syntactic unit) are no more than technical terms and are alien to ordinary language users. Literate people have intuitively adopted the denition of word as a set of letters between two spaces (Coulmas 1996), a concept that is based on a purely graphic criterion. It is this denition that teachers impart to students since the graphic fragmentation of written utterances by means of blank spaces provides a sort of explicit evidence in which the underlying concept remains implicit. This notion, which is transparent to literate adults, is opaque to those who do not master the particular features of writing; in fact, they need to put into play complex cognitive and linguistic processes to acquire it. Furthermore, Spanish literacy learning requires a different kind of linguistic analysis from that used in learning Argentine Sign Language. According to Massone and Machado (1994), LSA is a nongraphic language whose grammatical organization involves syntactic, morphemic, lexical, and pragmatic ways of representing linguistic categories that do not correspond to those of written Spanish. From the very beginning, readers-writers must distinguish between iconic and linguistic systems from a representational point of view. They must also command a conventional graphic repertoire and, at the same time, analyze and reinterpret graphic sequences in terms of written language categories (i.e., of meaning-making units). All of these processes point to different analytical dimensions vis--vis the subject of our study.

Deaf Childrens Construction of Writing | 463

Interpreting Illustrated Texts

Ferreiro and Teberosky (1999, 83) point out the following:
Since our writing system is alphabetic, it has been insistently claimed that it represents speech sounds, that it constitutes the phonetic transcription of language. We share others view (cf. Smith 1971) that this interpretation may be called into question . . . writing is very closely related to drawing and to language, but it is neither a transcription of (oral) language, nor a by-product of drawing. Writing constitutes a specic type of substitute object whose genesis we wish to explain.

Our goal is to show how children regard reading and writing and the problems they face, and, therefore, not all of our experimental situations are tests. For the study reported here we conducted interviews according to critical exploration guidelines. Each child was given a string of words accompanied by pictures, as well as an illustrated sequence of sentences. We examined their hypotheses about the relationship between drawing and writing and, in the case of the sentences, the value they attributed to sequential segments. Our results reveal the existence of a conceptualization process. At rst children are unable to distinguish between drawing and writing: At this stage meaning could be found in either one since they constitute an inseparable whole. Then children come to differentiate between the two, and nally they consider text properties in terms of segmentation and pointer letters that enable the ascription of meaning to each word or graphic segment. In line with these ndings, we hypothesize that, in approaching a written text in order to comprehend it, as in other knowledge domains, deaf children, like any other children, have to face and solve logical problems. In the present article we analyze the issues we dealt with in these studies and focus on illustrated sentences. We interpret deaf childrens hypotheses about what the text represents and thus about the relationship between the graphic segments of the syntagma (linguistic units) and the drawing, as well as about that between segments and the whole.

464 | Sign Language Studie s

Our Ongoing Research Project

Goals

In order to help challenge and revise theoretical standards of deaf literacy, we set out to explore the way in which deaf children acquire written language. In this article we do the following: (a) investigate deaf subjects constructive activity in this eld; (b) categorize deaf childrens ways of interpreting illustrated texts; and (c) determine the compatibility of the various processes through which hearing and deaf children learn written Spanish.

Method
Population

We set out to examine the processes that accompany the reconstruction of writing by fteen deaf children attending special schools in the cities of Rosario and San Nicols in Argentina (see table 1). We were particularly interested in the compatibility of hearing and deaf childrens learning processes vis--vis written Spanish. We took the learners analysis of the graphic features of the text as a fundamental, initial clue to the type of linguistic analysis entailed by the construction of the notion of words. The children we studied signed several LSA varieties and had been poorly trained in oral Spanish and not systematically taught to read and write, as oralism in our country has no method to promote literacy acquisition.3 Therefore, the children were practically illiterate, but had some knowledge of written texts such as their names and isolated words. Although age and grade were not considered as variables at the time of the interviews, they have been included in table 1 to illustrate the usual traits of the deaf population, as well as to highlight the scope of the results. Multiple social and family factors condition these childrens school attendance, and school entrance ages may therefore differ from the standard ages for hearing children. Older ages in the data do not necessarily mean the children have repeated a grade, and the criteria for promotion to the next grade vary from one special school to another but in general depend on the childrens speech abilities.

Deaf Childrens Construction of Writing | 465

It is important to note that, in Argentina, deaf public schools present a homogeneous population whose students generally come from lower class environments. Children arrive at school with poor or no speech but with sign language. These schools are hearing sociolinguistic settings that continue teaching oralism. If teachers of deaf students sign, they do so in Spanish word order. We believe the group of children under study is representative of the whole deaf population that attends school. Furthermore, it is very difcult to conduct research at deaf schools in our country, and in some schools it is impossible to do so since permission is refused, especially when the researchers are known to be nonoralist.

Table 1. Ages and Grades of Interviewees Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Name Anabella Ariel Dbora Ezequiel Isaac Leonardo Leonel Lorenzo Marcos Martn Milagros Silvina Jonathan Jonathan Roco Age (years and months) 8.3 9.6 9.1 8.5 7 8.2 5.8 7.5 8.6 10 4.4 5.1 8.8 8.11 8 Grade First Second Second Kindergarten Kindergarten Second Kindergarten Kindergarten Kindergarten Third Kindergarten Kindergarten First First Kindergarten

Instruments and Procedure

We carried out individual interviewees with nine cards, each containing an image and a string of written words. The images conformed to the prevailing standards for school texts.4 Ferreiros (1978) and Ferreiro and Teberoskys (1979) protocols were used, although we modied them to better reect the childrens communicative ability in LSA and the possibilities this offers for the translation of certain categories. The sentences on the cards were as follows:

466 | Sign Language Studie s

1. The duck is swimming. 2. The ducks are swimming. 3. The cat is eating. [The accompanying picture shows a cat eating sh.] 4. The cat is drinking milk. [The illustration depicts a cat and a container with the label milk.] 5. Nicols is having milk. [The image shows a boy drinking from a cup.] 6. The frog went out for a stroll. 7. The tortoise went out for a stroll. 8. Ral is rowing on the river. 9. The children are rowing on the river. The task was aimed at exploring the following areas: (a) the childrens hypotheses about the connection between the drawing and the writing, (b) the meaning they ascribe to the graphic segments that make up the latter, and (c) the possibility that children link text properties to what the images suggest. The interviews were conducted according to the critical exploration method (Piaget 1973; Ferreiro and Teberosky 1979) and mainly in LSA, although hearing interviewers sometimes spoke in Spanish, especially when they had to reproduce the sentences written on the cards in Spanish word order. The children expressed themselves in different sign language registers and varieties, depending on their ability to use sign language and interact with and within it. Children never answered in Spanish as they did not master oral Spanish. A literate adult deaf member of the research team was present in order to control and supervise the interviewers interventions and interpretations of the childrens utterances. Therefore, we transcribe the interviews in LSA glosses with their correspondent English translation and put between hyphens when the interviewee had to reproduce the Spanish text on each card. We expected that the methodological similarities between our study and those discussed earlier would make it possible to compare data on deaf and hearing children. Thus, we hoped to determine the ways in which written texts make sense to deaf children. We also wanted to understand the coordination and differentiation processes they engage in as they interact with the graphic features of both the drawings and our writing system. Finally, our research would, we hoped, shed light on the

Deaf Childrens Construction of Writing | 467

role of sign language in literacy learning, although an in-depth analysis is beyond the reach of this article.

Initial Results
The following fragments of the transcribed protocols illustrate certain types of responses consistently repeated by more than one child. 1. Lack of distinction between drawing and writing
Jonathan 2. Card 1

Interviewer:

THIS WHAT? What

is this? (She points to the pic-

ture.) DUCK. Duck. Jonathan: READ WHERE PRO2 READ? Where do you read? I: J points to the drawing. I: duck TO-SAY WHERE? Where does it say duck? (She points to the text.) J points to the image. HERE TO-SAY the duck is swimming TO-SAY. I: WHERE duck TO-SAY WHERE? Here it says the duck is swimming. Where does it say duck? J points to the picture of the duck. WHERE TO-SAY is swimming? Where does it I: say is swimming? J points to the picture of the duck. 2. Distinction between drawing and writing The picture complements and guarantees the interpretation of the text. Text and image constitute a meaningful whole and are construed as complementary.
2.1. The text provides the name of only one of the objects drawn.
Jonathan 1. Card 7

I: J: I:

THIS WHAT?

What is this? (She points to the picture.) TORTOISE. Tortoise. (He points to the picture.) READ WHERE PRO2 READ? Where do you read?

468 | Sign Language Studie s

J points to the text. I: tortoise TO-SAY WHERE? Where does it say tortoise? J points to the image. I: HERE (She points to the text). PRO2 TO-READ(imp) PRO2 the tortoise went out for a stroll. WHERE TO-WRITE tortoise WHERE? Here. (She points to the text.) Read. The tortoise went out for a stroll. Where is tortoise written? J points to the words: The tortoise went out. I: TOO WHERE stroll TO-SAY? And where does it say stroll? J points to the drawing. I: WHAT TO-WRITE HERE WHAT? What is written here? (She points to the words for a stroll.) J: TORTOISE. Tortoise.
2.2. Text and drawing constitute a complete utterance whose meaning is apportioned between text components and pictorial objects.
Leonel. Card 5

I: TO-READ WHERE? Where do we read? L points to the text. I: TO-SAY WHAT? What does it say? L: KID MILK TO-DRINK. Kid milk drinking.* I: WHAT TO-WRITE HERE? What is written here? (She points to the name Nicols.) L: CUP. Cup. (He points to the cup in the drawing.) I: HERE? Here? (She points to the words is drinking.) L: KID. Kid. (He points to the child in the picture.) I: TO-MAY TO-SAY is drinking TO-MAY? Could this say is drinking? (She points to the word milk.) L: TO-SAY-NO CHAIR. No, chair. (He points to the chair in the drawing.) I: HERE TO-SAY WHAT? What does it say here? (She points to the article.5) L: TO-NOT-KNOW. I dont know. I: WHAT TO-SAY1 BEFORE EVERYTHING WHAT? What did you say to me before all? (She points to the whole sentence.) L: KID TO-DRINK MILK. Kid drinking milk.

Deaf Childrens Construction of Writing | 469

*The syntactic sequence of Leonels utterance would correspond to the structure of LSA. In trying to repeat it, however, he transforms it into a new sequence that is closer to that of written Spanish. A similar situation appears in the next example:
Dbora. Card 3

I:

HERE TO-SAY WHAT? What

does it say here? (She points to

the text.) D: CAT FISH TO-EAT (cont). Cat sh eating. I: WHERE TO-SAY cat? Where does it say cat? D points to the word cat. I: Is eating? D points to the words is eating. I: Fish? D points to the words is eating and the sh in the drawing. I: HERE? Here? (She points to the article.) D: TO-NOT-KNOW. I dont know.
2.3. The text is divided into two names related to the image.
Roco. Card 4

I: TO-SAY WHAT? What does it say? R: CAT (she points to the words the cat) MILK (she points to the words is drinking milk). I: TO-SAY cat milk drinking? Does it say cat milk drinking? R: TO-SAY NO. No. (She points again to the same text segments and draws a cat and a milk bottle as she explains.) 3. Consideration of the graphic properties of the text 3.1. Each segment of the text corresponds to a pictorial object.
Ezequiel. Card 8 (The sentence is made up of ve graphic segments.6)

I: E:

TO-SAY WHAT?

BOAT FISH TORTOISE SNAIL BIRD.

What does it say? Boat sh tortoise snail

bird.

470 | Sign Language Studie s

Leonardo. Card 3 (The sentence consists of three graphic segments.7)

I: TO-SAY WHAT? What does it say? E: CAT MAN FISH ORANGE. I: WHERE TO-SAY cat? Where does it say cat? L points to the word cat. I: Man? E points to the article the.* I: sh TO-SAY WHERE? Where does it say sh? E hesitates and then points to the drawing. I: Orange? E points to the words is eating. *Even though articles do not constitute a sign language category, they are usually taught in special schools as gender markers and therefore construed as synonymous with man or woman.8 In this case and others, the interviewees response may have been prompted by such confusion. However, in view of the need to attribute each segment to a nominal category, this school interpretation is useful in that it favors the internal consistency of childrens answers. On the other hand, we should bear in mind the translation effort required of them. According to Jakobson (1985), if a language lacks a certain grammatical category, the meaning of the category may be rendered in the language by lexical means. These childrens mistakes would therefore be inherent in the translation process and due to the scarcity and simplication of the information available to them.
3.2. Each segment of the text corresponds to a possible segment of the utterance.
Anabella. Card 2 (The sentence comprises three graphic segments.9 The drawing shows a larger duck swimming with smaller ones.)

I: WHERE TO-READ PRO2? Where do you read? A ngerspells the sentence without being asked to.* I: TO-SAY WHAT? What does it say? A: DUCK. Duck. I: TO-SAY duck WHERE? Where does it say duck? A points to the words are swimming.

Deaf Childrens Construction of Writing | 471

I: HERE? Here? (She points to the word ducks.) A: LITTLE DUCK(pl). Little ducks. I: HERE? Here? (She points to the article the.) A shrugs her shoulders. I: TO-MAY TO-SAY ducks are swimming? Could this say ducks are swimming? A: TO-SWIM(cont). Swimming. (She hesitates.) I: WHAT EVERYTHING TO-SAY WHAT? What does it all say? (She points to the whole text.) A: TO-SWIM(cont) (she points to the article the). LITTLE DUCK (she points to the word ducks). DUCK (she points to the words are swimming). Thus, she construes the text as meaning swimming duck and little ducks, which would justify this segmentation. *Anabella ngerspelled all of the sentences we showed her of her own accord, which did not, however, guarantee conventional reading or her understanding of the text.
3.3. Everything is written except the articles.
Ariel. Card 3
I: TO-SAY WHAT? What does it say? A: CAT TO-EAT. Cat eating. I: WHERE TO-SAY cat WHERE? Where does it say cat? A points to the word cat. I: WHERE TO-SAY eating? Where does it say eating? A points to the words is eating. I: TO-SAY HERE WHAT? What does it say here? (She points to the article the.) A: TO-NOT-KNOW. I dont know.

4. Everything is written, even the articles

Martn. Card 3 His interpretation is similar to Leonardos:

I:

TO-SAY HERE WHAT? What does it say here? (She points to the whole text.)

472 | Sign Language Studie s

CAT TO-EAT FISH. Cat eating sh. (He points to the matching text segments.) I: WHERE TO-SAY cat WHERE? Where does it say cat? M points to the word cat. I: WHERE TO-SAY eating? Where does it say eating? M points to the words is eating. I: WHAT TO-SAY HERE? What does it say here? (She points to the article the.) M: MAN. Man.

M:

Martn. Card 4

I: M:

WHAT TO-SAY HERE? What

. does it say here? (She points drinking milk. (He points

to the whole text.)

CAT TO-DRINK(cont) MILK. Cat

to the whole text.) I: WHERE TO-SAY cat? Where does it say cat? M points to the word cat. I: WHERE TO-SAY drinking WHERE? Where does it say drinking? M points to the words is drinking. I: Milk? WHAT TO-SAY HERE WHAT? What does it say here? (She points to the article the.) M: He ngerspells the. To sum up, we categorized childrens responses according to the connection they established between image and text and to their interpretation of the latter in view of its graphic features. We have provisionally arranged the different types of responses in a developmental sequence that may be conrmed by a longitudinal study currently under way (table 2). In all of these cases, children drew on the pictures to predict the meaning of the text, just as their hearing counterparts do. However, the ways in which this relationship is established differs. In some instances, the childrens use of images may have been inuenced by their competence in LSA: They would point to illustrations to convey the lexical content of terms whose signs they had not yet mastered. Figure 1 illustrates the frequency and distribution of the types of responses (TR) discussed.

Deaf Childrens Construction of Writing | 473

Table 2. The Childrens Responses Type of Response Characterization
1. lack of distinction Meaning is provided by the image. Drawing and text form an between drawing undifferentiated whole. and writing 2. distinction between drawing and writing: The picture complements and guarantees the interpretation of the text. 2.1. The text provides the name of only one of the objects drawn. 2.2. Full meaning is attributed to the combination of text and picture. 2.3. The text is divided into two names related to the image. The text provides a name linked to the image. The graphic features of the former are not considered. The text constitutes a complete utterance whose meaning is apportioned between text components and pictorial objects. The text consists of two names linked to the picture. The division into two components is pointed out, and together the components comprise an utterance. A syntactic analysis seems to have been carried out. Inclusion in this category is determined by the importance ascribed to the drawing. The responses seem to seek a certain correspondence with the subject + verbal complement, which suggests a more advanced stage although the verb is not considered as actually written.

3. consideration of 3.1. Each segment of the The text constitutes a complete utterthe graphic proper- text corresponds to a pic- ance even though it is just a list of the ties of the text torial object. objects drawn. The graphic properties of the text are beginning to be considered although it is the image that guarantees meaning. 3.2. Each segment of the Each segment of the text is made to text corresponds to a correspond to a name linked to the impossible segment of the age but not determined by it. Children utterance. sought correspondence between the possible utterance and the text 3.3. Everything is writ- The text constitutes a complete utterten except the articles. ance. Children try to attribute meaning to each segment but cannot make sense of the articles. Two answers are possible: (a) the segment applies to a man or a woman,10 as taught in school; (b) the interviewees say they do not know. 4. Everything is written, even the articles. The text constitutes a complete utterance. All of the categories are written. Conventional reading takes place; the interviewees use sign language and ngerspelling.

474 | Sign Language Studie s

Figure 1. Distribution and frequency of the criteria (TR = types of responses)

expressed by the children.

As gure 1 shows, the predominant types of answers focus on the distinction between image and text and particularly on the analysis of the links between text and image. These are considered from different perspectives: In TR 3.1 the written words provide the starting point; the opposite occurs in TR 2.2. The results indicate that ve of these children are beginning to consider the graphic properties of the text since their responses reveal attention to the discontinuity of the written sequence. The idea that this discontinuity may provide a clue to the meaning of the text may initially be somewhat perplexing to children since, in earlier stages, they viewed the text as consisting of just a name or two. Such confusion, however, marks conceptual progress within Piagetian theory insofar as children become aware of an essential component of the intuitive notion of words, as well as of the conventional features of Spanish writing. The process is one of conict, in which children question their previous hypotheses and formulate new, more complex, ones.

Deaf Childrens Construction of Writing | 475

In brief, our preliminary results demonstrate the following: Both deaf and hearing children think about written texts and devise original concepts that are likely linked to the nature of written Spanish. A developmental progression takes place in both deaf and hearing children (Ferreiro and Teberosky 1979). The rst stage is characterized by the absence of a distinction between text and pictures, which does not imply an inability to perceive their graphic differences. Students view both systems as a whole, each of whose components enable the attribution of meaning. The nal stage involves an analysis in which graphic marks constitute linguistic indicators that are interpreted from the standpoint of sign language. There is, however, a substantial difference. In the last type of response, children do not see the text segments as equivalent to spoken Spanish utterances. They recognize the lexical, syntactic, and morphological components of the text and translate them into sign language; if no adequate term is already available in sign language, they resort to ngerspelling. The children we studied had rst learned ngerspelling at an early stage and then mastered it at school and in language therapy sessions. Nonetheless, our research suggests that literacy learners should reconstruct certain conceptual aspects of writing in order to use ngerspelling as a productive source of information (e.g., Anabella and Martn on card 4). Initially, both deaf and hearing children, unlike literate adults, seem to distinguish between what is written and what can be read. The inuence of formal education on childrens analysis of articles is evident in Leonardos and Martns responses. Sign language does not contain articles. Children who try to ascribe meaning to each written segment nd no comparable, segmentable category in sign language. At school, information about articles relates to gender rather than to morphological and syntactic functions. Consequently, the attribution of the meaning man to the masculine definite article is consistent with the need to assign each segment to a nominal category.10 Nevertheless, Martn later on (card 4) used ngerspelling to distinguish the segment.

476 | Sign Language Studie s

Final Comments
Although the widening of the sample, the analysis of cross-sectional data, and the longitudinal study are not yet completed, our ndings appear to coincide with those of Ferreiro and Teberoskys (1979) psychogenetically based research on writing. However, the nature of both processes cannot yet be determined since the inuence of our methodology cannot be ruled out. Nevertheless, we maintain the legitimacy of our data as to the features and originality of the processes these children engaged in. The similarity of the criteria used by deaf learners from different geographic, social, and school environments point, if not to the universality of such processes, at least to the linguistic and cognitive activity triggered by written text. As we have already stated, we view deaf literacy as second language learning, which thus involves a translation process. The children in our sample carried out an interlinguistic translation ( Jakobson 1985) insofar as they construed the meaning and organization of the written symbols from the standpoint of lexicon and structure of LSA. This kind of translation requires reinterpreting and recoding since the translator must nd equivalence in difference, which involves examining the mutual translatability of the languages concerned. The ability to speak a certain language entails the ability to speak about it, a metalinguistic operation that enables the revision and redenition of the lexicon employed. Therefore, access to literacy demands intense metalinguistic activity, which is greatly facilitated by sign language experience and diverse communicative contexts, and a variety of texts and textualization opportunities means more and better information. Thus, strictly speaking, the children we studied engaged in translators rather than writers processes. Data, therefore, show that deaf children, as every child, are active and constructive subjects that question and reconstruct the knowledge that context provides them with. In the particular case of deaf children the possibilities of interaction with written Spanish is a sine qua non condition so as to learn this language. However, this does not mean that they have rst to master oral Spanish. In fact, further research with deaf children that do not speak Spanish shows that they master the comprehension of the alphabetic principle.

Deaf Childrens Construction of Writing | 477

Appendix

Los patos nadan.

Nicols toma la leche.

478 | Sign Language Studie s

Notes
1. The studies were conducted in Argentina, Switzerland and Mexico. The ndings have been conrmed by research carried out in Brazil, Italy, Israel, and the United States within the same theoretical framework. In addition, for a detailed review of studies that have focused on different aspects of the conquest of the written language by deaf children and teenagers, see C. Lepot-Froment (1996), La conqute dune langue orale et crite, in Lenfant sourd: Communication et langage, by C. Lepot-Froment and M. Clerebaut, 83163 (Paris: De Boeck y Larcier, 1996). 2. For information on oralism, see B. Virole and D. Martenot (2000), Problmes de psychopdagogie, in Psychologie de la surdit, 41332 (Paris: De Boeck and Larcier). 3. Examples of this material appear in the appendix. 4. In the Spanish original, Nicols toma la leche (Nicholas is having milk), leche is preceded by the denite article la. 5. Ral rema en el ro (Ral is rowing on the river). 6. El gato come (the cat is eating). 7. See note 6. 8. Los patos nadan (the ducks are swimming). 9. In Spanish, articles are inected for gender. 10. See note 4.

References
Bells, R. M., and A. Teberosky. 1989. Produccin e interpretacin de textos escritos por nios sordos pequeos integrados en escuelas ordinarias. Convocatoria Proyectos de Investigacin Educativa en 1985. Coulmas, F. 1996. Superacin de la diglosia: El acercamiento del japons escrito y hablado en el siglo XIX. In Hacia una teora de la lengua escrita, ed. N. Catach, 24156. Barcelona: Gedisa. Domnguez Gutirrez, A. B. 1999. Lenguaje escrito y sordera: Sobre qu cuestiones es importante reexionar? In Lenguaje escrito y sordera: Enfoques tericos y derivaciones prcticas, ed. A. B. Domnguez Gutirrez and C. Velasco Alonso, 4758. Salamanca: Universidad Ponticia de Salamanca, Servicio de Publicaciones. Fernndez Viader, M. P., and E. Pertusa. 1995. Reexiones sobre la escritura y la alfabetizacin de los nios sordos. Revista Logopedia, Foniatra, y Audiologa 16(2): 7985. Ferreiro, E. 1978. Qu est escrito en una oracin escrita? Una respuesta evolutiva. Journal of Education 160 (4): 2539. . 1999. Dicotoma oralidad/escritura en la psicognesis contempornea In Vigencia de Jean Piaget, ed. E. Ferreiro, 5571. Mxico: Siglo XXI.

Deaf Childrens Construction of Writing | 479 , and A. Teberosky. 1979/1985. Los sistemas de escritura en el desarrollo del nio. Mxico: Siglo XXI. Jakobson, R. 1985. En torno a los aspectos lingsticos de la traduccin. In Ensayos de lingstica general, ed. R. Jakobson, 6777. Barcelona: PlanetaAgostini. Massone, M. I., and E. M. Machado. 1994. Lengua de seas Argentina:Anlisis y vocabulario bilinge. Buenos Aires: Edicial. Massone, M. I., M. Simn, and J. C. Druetta. 2003. Arquitectura de la escuela de sordos. http://www.librosenred.com. Massone, M. I., M. Simn, and C. Gutirrez. 1999. Aproximacin a la lengua escrita de la minora sorda. Lectura y Vida 20(3): 2433. Morrison, K. 1995. Fijacin del texto: La institucionalizacin del conocimiento en formas histricas y loscas de la argumentacin. In Cultura,pensamiento, escritura, ed. J. Bottero, B. Lonn, and P. Wilson, 134. Barcelona: Gedisa. Newport, E., and R. Meier. 1985. Acquisition of American Sign Language. In The Crosslinguistic Study of Language Acquisition, ed. D. I. Slobin, 881938. Hillsdale, N.J.: Erlbaum. Parkes, M. 2001. La alta edad media. In Historia de la lectura en el mundo occidental, ed. G. Cavalloy and R. Chartier, 15377. Madrid: Taurus Alfaguara. Piaget, J. 1973. La representacin del mundo en el nio. Madrid: Morata.