Orthographic Knowledge

Beverley L. Zakaluk, Ph. D. 1982/96

E-Mail: [email protected]
Helen Bochonko
E-Mail: [email protected]
University of Manitoba
Created on June 24, 1996
Last Updated on October 9, 1998

One of the most significant facts about language is that it is structured: orthographic symbols are arranged according to a fixed set of rules and cannot be put together in simply any arbitrary form. At the graphemic level there are rules for permissable and nonpermissable graphemic sequences. For example it is appropriate to string together the letters /s/, /t/, and /r/ to form the sequence /str/, but inappropriate, in English, to combine them to form the sequence /tsr/. Eleanor Gibson (1973) states that of the more than 150 consonant clusters in frequent use, all but three are constrained to initial or final position. This feature of English spelling therefore provides a source of enormous redundancy which is useful to beginning readers in that there is more than one source of information beyond the single letter to suggest what a word might be. The fact that letters cluster in repetitive patterns thus serves as a generalization that can be applied to decipher new words.

All native speakers, with experience, acquire this orthographic knowledge. Where the difficulty in learning letters originates for some children however, is in developing the ability to select from among the many different letter features. DeStafano (1978) identifies the smallest possible perceptual features which contrast letters as:

1. straight lines as opposed to curved lines as in "l" and "c";

2. horizontal elements as opposed to vertical or diagonal elements as in "t", "h", and "z"; and
3. open formations in contrast to closed formations as in "c" and "o".

There are other problems which arise in distinguishing between letters. Gibson and Levin (1975) describe the letter characteristics of redundancy as in "w", and discontinuity as in "l" for example. Discriminating among these features is a source of disorientation for some children. Such letter features confuse the eye. While most objects remain the same no matter from which angle we view them, if we turn M upside down, it becomes W. Children must also learn two forms for each letter, upper and lower case.

Another important skill is transforming the graphic strings into phonemic representation. In English generally, one letter or group of letters represents one sound, but this relationship does not hold true one hundred percent of the time, when gh has the f sound for instance. This may be a source of confusion for some students. Others may have trouble or be unaware that segmenting long or complex words into their more manageable parts or syllables and then imposing their phonic knowledge, facilitates word identification.

In addition, there may be differences in idiolect or dialect which confound the perception of discrete sounds especially where the vowel system is involved. For example the word "pin" may be pronounced as both "pin" and "pen" in some jurisdictions. Children know the difference in meaning, but may be confused when efforts are made to teach them distinctions they themselves do not make, nor perceive.

Other examples cited (Savin, 1972) are that some pupils are able to tell that the sound of "cat" is different from the sound of "dog", but may not perceive that "cat" differs from "rat" nor that "cat" and "cow" begin with the same sound. These same children do not exhibit flexibility in manipulating sounds and are not competent in games involving language play. This suggests that in addition to the issue of dialect, there may also be a developmental component in the cognitive sense which interferes with the understanding of how letters are transformed into phonemic representations. Lack of experience in listening to rhymes and stories as a preschooler may also be a factor.

Dialect

Certain words may also be differently pronounced by dialect speakers. For example the idiomatic form "ness" may be substituted for "nest", "ress" may be substituted for "rest", and "han" for "hand". Similarly, as reported by Hall and Guthrie (1980), "test" may be pronounced "tess", "mend" as "men", "walked" as "walk", "cold" as "coal" and "find" as "fine". These differences could interfere with word identification in a phonics approach to reading instruction and in total, frustrate children forced to pronounce sounds and words in isolation. In teaching, it would appear more useful to concentrate on associating sounds with longer orthographic strings which serve as visual stimuli, rather than on pronunciation and sound alone. It should not be inferred, however, that a whole word approach to phonic instruction should be ruled out, instead an analytic approach using whole words as models in building new words is still viable. Letters simply need to be embedded in orthographically regular strings such as syllable units that can be viewed as gestalts.

One of the main assumptions in this discussion of orthographic knowledge is that dialect variations may cause reading obstacles over and above typical problems of discriminating between letter symbols and transforming letter symbols into their respective phonemic representations As Hall and Guthrie (1980) point out, it is obvious that phonological differences exist, but whether or not they actually interfere with a child's learning to read has not been demonstrated.

Hall and Guthrie review a number of studies on the effect of phonological interference to support their belief that dialect differences may not cause reading problems. One was a study by Melmed (1971) which compared the reading ability of third grade students who used vernacular Black English to that of third grade students who spoke Standard English. Vernacular Black English was characterized by vowel variations, weakening of final consonants, "r-less" and "l-less" speech, together with consonant cluster simplication. It was found that even though the third graders who used idiomatic speech often obtained low scores on isolated auditory discrimination tasks, and the production of selected phonological features, there was no difference in their ability to comprehend in reading. Similarly, Hall and Guthrie review research conducted by Rystrom (1970) which measured the effect of phonological interference on reading acquisition. Rystrom found that training in Standard English did not result in significant differences in reading achievement compared to a control group receiving the regular language arts program. Still another investigation using Appalachian dialect speakers as subjects (Rentel and Kennedy, 1971) studied the effects of pattern drill in Standard English on reading achievement, discovering no differences in reading scores between experimental and control groups.

Contrary evidence regarding the issue of phonological interference, however, is presented in research by Osterberg (1961) carried out in a dialect area of Sweden. In the first ten weeks of school, children were instructed in special materials which conformed to the phonological features of the dialect area in which they lived, while a control group received regular materials. Those children instructed in the special materials did acquire higher reading achievement performance both at the end of ten weeks and at the end of one year.

Numerous methodological flaws (Hall & Guthrie,1980) are inherent in research of this type. One fault in the American studies relates to the question of school setting. Hall and Guthrie suggest that in the school setting children may in fact not be using vernacular or idiomatic speech, but employing the school register in Bernstein's sense of the term. This would account for lack of phonological interference. The conflicting findings of the investigations summarized suggest that the question of phonological interference having an inhibiting effect on reading skill acquisition has no definitive answer as yet.

Summary

In terms of orthographic features of text, there is a degree of regularity in English graphemic structure which provides helpful redundant information to aid word recognition. Some children encounter difficulty, however, in either discriminating letter features, separating the sound components, or matching sounds to letters. Other children demonstrate insufficient flexibility simply in manipulating sounds. They lack language facility. Regarding the role of dialect and phonological interference in reading acquisition, research into the question suffers from methodological constraints. As yet there is no conclusive answer as to whether or not idiosyncratic speech interferes with reading progress. Over-reliance on letters and their sounds and on materials outside the realm of the child's experience which prevents the application of meaning to assist decoding are factors that inhibit word identification.

Implications for Instruction

Broadly speaking, the interactive model of reading suggests that orthographic knowledge has as important a role to play in reading acquisition as lexical, syntactic and semantic knowledge. Many experiences with oral language in the very early years which focus on language flexibility appear to be basic to reading acquisition. In the beginning school years a continued emphasis on teaching the visual regularity of orthographic strings, called the mechanics of reading, is essential, as is tailoring the text to the actual experiences of children. Adding a writing component to beginning reading instruction also helps children internalize the code.

semantic_knowledge