The development of literacy

Reading is a very multifaceted, flexible, and sophisticated cognitive activity in-volving many component skills. As Perfetti (1984, 48) asserts, word representa-tions and the processes of word identification that are determined by these repre-sentations are crucial in the reading process. Perfetti (1984) also underscores the role of lexical access, orthographic structure and knowledge of the alphabetic prin-ciple. Access to semantic and syntactic information is also important in the read-ing process (Samuels & Kamil, 1984).

Although there is extensive knowledge of the reading process, there is no com-plete and unambiguously accepted model of reading acquisition (Juel, 1991). A detailed discussion of the different models of reading is not necessary here (c.f.

Samuels & Kamil, 1984), but in order to get an overview of the factors influencing the reading process, three models will be briefly mentioned here.

Bottom-up theories (Samuels & Kamil, 1984) emphasise the importance of word recognition, and assume that letter features, letters, words and phrases are combined, from smaller to larger units, in order to develop meaning. Research on eye-movements and fixations provides support for a bottom-up theory of reading.

There is, however, evidence suggesting that a bottom-up theory is not sufficient in explaining the complex reading process; for instance, the notion that context plays a significant role in developing meaning is not easily accounted for in a bottom-up approach (Carpenter & Just, 1986). Neither does this theory take into account the background knowledge that even young children bring to the reading process (McKeown, Beck, Sinatra, & Loxterman, 1992; Pearson, Hansen, & Gordon, 1979).

Top-down theories reject the belief that the precise processing of letter, word and phrase perception and identification is the key to successful reading (Samuels

& Kamil, 1984). Rather, the focus is on word identification. The background knowl-edge the reader brings to the reading process is essential (Goodman, 1967). Pre-diction is also very important in reading. However, one weakness of this model is that it does not recognise processes such as decoding (King & Quigley, 1985).

Interactive models emphasise the reader’s active processing of the text, in-cluding the reader’s goal to construct a model of what the text means (Perfetti &

Curtis, 1986). Here, the role of background knowledge in constructing meaning

from text is of great importance. Background knowledge and story schemas are structures essential for text processing (Anderson & Pearson, 1984). Studies on deaf children’s schema development indicate not only that deaf children – natural-ly – bring their world knowledge to the reading process, but also that the story structures of deaf children are not equally well developed as those of hearing children (Akamatsu, 1988; Schirmer, 1993). An exception is deaf children of deaf parents: this particular group of deaf children perform equally well as hearing children (Schirmer, 1993). These results suggest that teaching methods that elaborate deaf pupils’ story schema may improve their reading comprehension (Schirmer, 1993).

Language modality and reading

Language modality influences the processing of linguistic stimuli. Research find-ings show that deaf children unsuccessfully try to map their own primary lan-guage on a written lanlan-guage (Fok, van Hoek, Klima, & Bellugi, 1991; Rodda &

Grove, 1987). The fact that deaf children, when learning to read and write, have to change modality, and that there is an unfit match between a sign and a written word probably accounts for many of the difficulties deaf children have in the learn-ing of readlearn-ing skills. Moreover, many features of signed languages add up to com-plicate the mapping of a written/spoken language onto a visual language even further. Differences in syntax and morphology do not independently explain why it is so hard for native signers to learn how to read. The fact that the deaf child has no support in his or her own language on how to – in a spoken language – express, for instance, time, how to ask questions, and how to understand tenses, may ac-count for some of the difficulties. So, taken together, the difference in visual and orthographic representations, as well as fundamental differences in the grammar of signed and spoken languages may to some extent explain why deaf children are poor readers and writers. In addition, the fact that spoken languages are sequen-tial, while signed languages are more spatially organised, may initially impede the deaf child in acquiring a spoken language.

However, a main reason for the difficulties seems to lie in educational instruc-tion. The instruction in the national language for deaf children has been poor, and has not served as a base for emerging literacy. Deaf children need to learn a verbal language as a second language (Kristina Svartholm, personal communication, April 3, 1998). Despite deaf children’s difficulties and the obvious differences in the reading process, a crucial question is whether reading truly is different for deaf people and for hearing people (Paul, 1998). The answer to this obviously has sig-nificant educational implications. As Paul (1998, 62) points out, even though the literacy development of deaf children and that of hearing pupils are qualitatively comparable, deaf pupils’ literacy development often lags behind the literacy de-velopment of hearing pupils. This, naturally, has consequences for the deaf chil-dren’s learning and educational performance.

Strategies in deaf readers

For hearing individuals speech-recoding is essential in the reading process, and this strategy is also available to some deaf readers, particularly to better deaf

read-ers (Hanson, 1985), and to those with intelligible speech (Conrad, 1970). For deaf people a speech-based code is, however, difficult to use, and therefore deaf read-ers also use manually coded strategies, such as fingread-erspelling-based and sign-based strategies (Hanson, 1985). The recoding strategy based on fingerspelling is fre-quently used by deaf individuals with unintelligible speech (Locke & Locke, 1971);

according to Hirsh-Pasek (1987) this strategy is also used by children, although not spontaneously.

The use of sign is another recoding strategy available to hearing-impaired chil-dren, i.e. printed information is transformed into sign representations (Stofoen-Fischer & Ae Lee, 1989). But since there is no relationship between printed Eng-lish words and ASL, it has been suggested that recoding into signs does not en-hance word identification (Hirsh-Pasek & Treiman, 1982).

Stofoen-Fischer and Ae Lee’s (1989) study on the effectiveness of graphic rep-resentation of sign in developing word identification skills for hearing-impaired readers, as well as Robbins’ (1983) study on the effects of signed text on the read-ing comprehension of hearread-ing-impaired children show that graphic representa-tions of sign have several advantages for the deaf beginning reader. The graphic representation enhances the association between the printed words and the con-cepts they represent.

In an experiment on recall of words with or without sign equivalents, deaf sub-jects performed well in the words with the corresponding sign but poorly on the words with no sign equivalent (King & Quigley, 1985). Furthermore, deaf subjects understood connected prose better when the syntax of the printed information had been changed to the syntactic order of ASL. These and similar results support the view that deaf people process information in Sign Language, and do not translate signs into corresponding English words. Moreover, the key to successful reading may in part depend on establishing a connection between the two language systems (Davey & King, 1991). Studies on skilled deaf readers may suggest ways to develop the ability to read in the larger deaf population (Kelly, 1992).

Deaf children do not learn to read through old-fashioned drills and therefore meaningful texts and teaching materials are necessary. Rather, storytelling, dis-cussions on good texts written for children (not easy to read texts), and compari-sons of signed and written texts, are the keys to successful instruction of the na-tional language. Both Inger Ahlgren (personal communication, April 3, 1998) and Kristina Svartholm (personal communication, April 3, 1998) stress that a neces-sary condition for successful education of deaf children is the use of Sign Lan-guage as the lanLan-guage of instruction. Equally important is the viewpoint that deaf children should be taught in the same fashion as hearing children. According to Ahlgren nothing indicates that deaf individuals differ from hearing individuals as to information processing or learning. They simply need to receive instruction in a Sign Language. For a further discussion on bilingual education in Sweden, see Svartholm 1993; on second language learning in deaf individuals, see Ahlgren, 1982, 1984, Svartholm 1984, 1990, 1994; and on Sign Language as the first lan-guage, see Ahlgren 1994.

According to both Inger Ahlgren and Kristina Svartholm (personal communi-cation, April 3, 1998), an interesting question is why Finland, with a long tradition

of linguistic minorities and experience of teaching Swedish as a foreign language to Finnish-speaking people, cannot utilise this knowledge of Swedish as a second language in the education of deaf children.

Metacognition

Finally, metacognition is also an important component in emerging literacy. Met-acognition is used to describe the control an individual has over his or her own cognitive functioning, such as thinking, learning and problem-solving (Brown, 1980, 453), but is also linked to the development of good reading skills (Strass-man, 1992). In fact, studies suggest that deaf pupils’ metacognitive awareness is lower than that of hearing pupils (Strassman, 1992). The teaching of metacogni-tive awareness may turn out to be an important component in the teaching of reading to deaf pupils as well as in improving deaf readers’ comprehension of written texts (Fox, 1994).