Exploring the richness of the stimulus

Abstract

The common-sense layman's assumption would be that the structure of the languages which people grow up speaking reflects the properties of the language-samples they hear. There is no good reason to dispute that.

Pullum and Scholz see their own discussion as open to the criticism that their examples are drawn from genres which are untypical of the speech encountered by young children. Accordingly, I examine the evidence of a four-million-word sample of spontaneous everyday conversation; this, too, refutes the claims about usage made by poverty-of-stimulus theorists.

Widespread current acceptance of the poverty-of-stimulus idea has apparently come about not because linguists have found the contrary view empirically unsatisfactory, but merely because poverty of the stimulus is for one reason or another treated as an unquestioned axiom.

A related issue is lack of negative evidence. It is true that children acquiring their mother tongue encounter only positive examples. But in this respect the logic of language-acquisition is no different from that of scientific discovery.

‘Poverty of the stimulus’ seems plausible to linguists who believe that human beings inherit detailed knowledge about language structure. But that belief is itself grossly implausible. The languages of the world are too diverse to be the product of an innate language faculty. If language structure is not in any substantial respect innate, then children's experience of their elders' language must be adequate for language-learning; the evidence suggests that that is so.