Morphological Structure and Segmental Awareness

Morais et aI. are surely correct in argUing that segmental awareness is all but essential for the skilled reading of an alphabetic orthography, and that such awareness usually develops as part of the experience of learning to read (defined fairly broadly), rather than being a precondition for learning to read. The second point could be made in a slightly different way. Suppose that the goal were not to teach children to read, but to teach them to be phonologists. Surely, one of the teacher's first steps would be to introduce his pupils to phonological transCriptions. An adequate notation is a great help in mastering any abstract system, and most children would not become very advanced phonologists without some exposure to the notation, even if they were trained in segmental analysis tasks. The actual goal, of course, is to teach children to decode easily texts written in a quasi-phonological notation, and this requires that they be phonologists enough to have awareness of the segmental system that underlies this notation. Such awareness is certainly greatly enhanced merely through exposure to the notation itself. But is such exposure absolutely essential for segmental awareness? For Morais et aI., this "remains an open question." Surely, however, the answer must be, "No," if we are to understand the existence, in ancient India, of an oral tradition of morphological and phonological analysis of Sanskrit, only later codified and written down by Panini (Holender, in preparation); and unless we wish to believe that alphabetic writing was not invented by a segmentally-aware individual, but instead "evolved" (Gelb, 1963), a proposal whose various difficulties I have discussed elsewhere (Mattingly, 1985). Moreover, there are certainly a few children whose acqUisition of reading is so rapid and effortless as to suggest that they are merely learning a set of graphic symbols to transcribe phonemic segments they are already cognitively aware of. It is perhaps more of an open question whether the cognitive states of Paninl's predecessors and the unknown genius who invented the alphabet. or of these exceptional children, are relevant to an understanding of the acquisition of reading in the general population. But I would argue that they are. If we wish to understand some cognitive process, it is at least as valuable to study those indiViduals in whom the process is exceptionally well-developed as to study those in whom it is normally developed or defeCtive. What factors might lead naturally to segmental awareness in an alphabetically naive but phonologically-curious person? As a native speaker, he has access to mental representations of spoken utterances in his language. These representations have many subtle properties resulting from modular linguistic processes that are themselves inaccessible (Fodor, 1983). What might lead the naive phonologist to