The logical problem of language acquisition concerns a number of surprising aspects of our linguistic competence and how we come to achieve it.
Put simply, young children learn language in spite of poor linguistic input, and adults show sophisticated subconscious knowledge about their language which they cannot have been taught.
In other words:
People attain knowledge of the structure of their language for which no evidence is available in the data to which they are exposed as children.
(Hornstein & Lightfoot, 1981)
One aspect of this logical problem concerns the poverty of the stimulus; Chomsky gives a number of examples of underdetermination:
The poverty of the stimulus argument makes it difficult to explain how a child acquires language: if the language children hear is insufficient to determine the full language system, how it is that we all eventually achieve mature native speaker competence?
A related argument concerns our ability to distinguish grammatical and ungrammatical sentences, even for subtle language points which we cannot consciously articulate. How do we reach this competence?
Nativists interpret such arguments as evidence for an innate predisposition for language acquisition. This predisposition may be language-specific, as generativists would argue, or part of general cognition, according to emergentist accounts.
An important question for both schools concerns the role of evidence, or the language input which must be processed to develop language competence.
Read and reflect: the Competition Model