Nyelvtudományi Közlemények 85. kötet (1983)

Tanulmányok - Gleason, Jean Berko: Insights from the Extraordinary: Some New Trends in American Psycholinguistics 140

NEW TRENDS IN AMERICAN PSYCHOLINGUISTICS 145 many affixes (e.g. -ba, -hol, -ban, -ra, etc.) are lost by even mildly aphasic people, and which are retained even by severely aphasic patients. Presumably, the robust affixes are more central to the language. Comparative and cross cultural work is needed to illuminate the basic features of aphasia in whatever language it occurs. Agrammatism in English, for instance, is marked by a loss, among other things, of articles ; we might ask what form agrammatism takes in languages that do not have obligatory articles. Another important kind of question that aphasia poses is what relationship there is between acquisition in childhood and loss in aphasia. Some work has been done on this, but despite the strong theoretical models that suggest that loss is a mirror-image of acquisition, it is not possible to show that this is generally the case. Again, more cross cultural evidence is needed. Speech to children The study of children's acquisition of language has seen dramatic changes in recent years. I shall limit myself to only two kinds of research here, since my own colleagues and I have been involved in them. The first is the description of input language to children, that is, the specialized language that adults use when addressing young children ; the second is the role that adults play in children's acquisition of the social and interactive components of language. Beginning about ten years ago, a number of investigators in various parts of the world began, quite independently, to study the speech that adults (usually mothers) address to children who are learning language. Most of these researchers were concerned with the theoretical models that had been put forth by the transformational linguists. The transformationalists held basically that every child arrives in the world equipped with a language acquisition device (LAD) that is set in motion by exposure to adult language. According to this model, the basic principles of language are by and large innate. One of the supporting arguments for this claim was the supposed fact that the adult language that children hear is ill-formed, difficult to segment, filled with disfluencies, and quite ungrammatical. Thus, the principles of such a language would not be at all obvious, and children could not learn them unless they already possessed a set of innate discovery rules and a powerful hypothesis-generating mechanism. The transformationalists thought that adults spoke to one another and to children in the same way. A great many studies showed that this was not the case. (See Snow and Ferguson 1977 for one of the first books on the subject.) It has been shown that adult-child speech is fluent, grammatical, repetitive, and relatively easy to segment because of the special way stress and juncture are used. Adults modify their speech at every level, phonological, morphological, syntactic, and semantic, when speaking to children. Clearly, less reliance on innateness is required, since the adult provides much of the structure the child must discover. Moreover, adults adjust the complexity of their speech to the level of the child, speaking in more sophisticated ways as the child grows older; and all older speakers, not just mothers, provide the characteristic modifications. The question of universale in input language has been posed, but as yet we have no answer as to that such universale might be. All of the varieties of input language that have been documented contain some features that simplify or clarify the language, as well as some indication of the affective bond that exists between 10 Nyelvtudományi Közlemények 85/1.