Table Of Contents

1. Introduction

2. Speech & Writing

3. Sources of the Written Language Bias in Linguistics

4. On the Scope of our Claims

5. Linguistics & the Overall Theory of Language

6. Grammar

7. Semantics

8. Phonology

9. Linguistic Communication

10.Language Acquisition

11 Linguistic Variation

12 Epilogue




Our theories of language acquisition will naturally be dependent on what theories we have of language structure and language use. Accordingly, if we think of language as an autonomous structure which is relatively independent of other sign systems (non-verbal communication) and situational contexts, then we do not expect, of course, that the properties of language could be explained by reference to such phenomena. Exactly this predicament characterizes some structuralist approaches to language acquisition; they look at the ontogenesis of language as something which requires very special explanations, i.e. explanatory principles which are good only for language. Let me illustrate this by calling attention to some features of the theories of two great structuralists, Roman Jakobson and Noam Chomsky.


10.1 Phonological development

Roman Jakobson is famous for, among other things, his very clear-cut distinction between pre-phonological articulatory and perceptual developments in the child, which, in his opinion, has nothing to do with language, and phonological development, which is part of language acquisition. Thus, Jakobson argues that babbling is relatively unstructured, that it contains all sorts of phonetic effects, and that it has no relation whatsoever to the later phonological development. The latter, on the other hand, exhibits a universally valid structure in that children develop phonemic contrasts in very regular ways, from the most primitive system of just two or three units to the full-blown system of the adult language. Whereas phonological development is concerned with the acquisition of the distinctive features of the symbolic system, i.e. language, babbling has no such functions at all.

From a semiotic point of view, the distinction between phonological distinctive functions and the more or less continuous vocal behavior of babbling seems absolute. However, it may easily be wrongly stated, and it has de facto served to make linguists less eager to look at the structure of babbling. In fact, the child's phonetic achievements in the babbling period and the early stages of language acquisition are not unrelated. Babbling is not completely chaotic; rather it displays preferences for certain phonetic structures, at many points those which Jakobson predicted should appear early in the phonological development (Oller et al 1976). MacNeilage (1979:30) observes:

"From this viewpoint the child's first words can be seen as, at least partially, a matter of choosing from the babbling repertoire a set of approximations to adult word forms"

If, furthermore, we want to investigate young children's ways of phonetically distinguishing recognizable words belonging to the language they are surrounded by, then it turns out to be impossible to observe the early stages of phonological development that Jakobson postulated, since children have a considerable repertory of phonetic units, and hence many potentially phonemic distinctions, already at the stage when they have a vocabulary of say 25-30 words, i.e. at the earliest point in time that it is possible to perform any phonological systematization. On the other hand, some more primitive stages in the phonetic development may be observed in babbling. Thus, the development from pre-phonological to phonological stages is gradual rather than absolute.

Moreover, Jakobson's neglect of babbling and other early vocal and non-vocal behaviors is unfortunate, since such behavior has important emotive, social, evocative and expressive functions. These are functions which are still dominant, as the child goes on to develop language. If we want a comprehensive perspective on the child's communication capacities, the development from pre-linguistic to linguistic means of communication is gradual and far from linguistically uninteresting. Instead, we can see that the isolation of language as a communication system that is considered to be absolutely unique tout court is a reflection of the linguist's tradition of considering mainly or only the cognitive and referential functions in communication (IX.2).


10.2. The innate faculty of language

Since the mid 1970's there has been a vigorous movement in child language studies attempting to look for sources and similarities of spoken utterances and speech acts in the non-verbal communicative acts that occur very early in the child's life and develop into more and more diversified forms of interaction as the child grows older. Scholars (e.g. Werner & Kaplan 1963, Bruner 1975, Greenfield & Smith 1976, Bates et al 1979) have been able to show that there are interesting relationships between the non-verbal acts preceding language acquisition and the later communicative acts that comprise both verbal and nonverbal aspects. This applies both to the action-theoretical status (functions) of singular acts and to their internal semantic and formal structures. Certain fundamental features of syntactic organization seem to reflect properties of basic action patterns; Bruner has argued in this vein with regard to both the topic-comment structure and the role structure of sentences ("role" in the sense of Fillmore's deep case or the like).

Thus, the scientific approaches just mentioned serve to establish connections between language and various non-verbal communication systems. This goes against traditional linguistic views that assign an absolutely unique position to language. Perhaps the most extreme structuralist view in this regard is Chomsky's flat denial that there are any interesting relations at all between language and more "primitive" symbol systems (e.g. Chomsky 1968:70 ff.). Chomsky's position, however, is quite logical, given the long-time goal of establishing linguistics as an independent discipline with a purportedly well-defined object of study, i.e. language as an autonomous system. Chomsky in fact ascribes to the ideal speaker's language a very intricate structure, such as could be realistically described only by a generative transformational grammar. If language users have indeed "internalized" such grammars, then it seems inconceivable, says Chomsky, that all of this could be explained as developments of non-verbal symbol systems. Accordingly, Chomsky hypothesizes that man has an innate highly specialized capacity for language, a very specific faculty (faculte de langage) independent of other faculties of the mind. That is, he postulates a biological basis for language in a way that precludes the possibility of deriving properties of linguistic capacities from other cognitive abilities and social skills. This metaphysics is unfortunate at two levels. For one thing, the postulation of a specific capacity for only language is ad hoc and therefore methodologically inferior in that it persuades linguists not to seek for deeper explanations (Derwing 1973:63 ff.). Secondly, it is in fact at least partly unmotivated on empirical grounds, because the above-mentioned research on child language seems to show that there are lots of significant relationships between non-verbal and verbal means of communication. However, all this presupposes that we really explore into the dynamic interaction of spoken discourse. This is something that generative linguistics does not do. As we have seen (IX.2) Chomsky even argues that communication is an accidental property of the use of language!