Chomsky's review of Skinner's Verbal Behavior is often credited with launching the cognitive revolution. His central point was that children produce sentences they have never heard and have never been reinforced for. Why is that observation so damaging to a purely behaviourist account of language?

C. Because reinforcement can only strengthen responses that occurred.

Reinforcement changes the probability of responses already in the repertoire, so every response it explains must already have occurred, and a child saying a sentence for the first time has no reinforcement history to grip. Input is plainly necessary; errors like 'goed' show children extracting rules rather than being handed them. The claim is that input alone is not sufficient.