There are a lot of theories about first language acquisition: behavioural, innate, socio-cultural, cross-cultural, connectionism etcetc. Many of them corresponded to a certain method being particularly popular at a given time. Or a method appeared as a response to the theory. Apparently, the most controversial one is the innatism theory of Naom Chomsky, who managed to leave his mark on pretty much every aspect of English language study. Basically, what he is saying, is that a child has a built-in knowledge of language, in particular, grammar, so that this lucky child can magically distinguish grammatical sentences from not so grammatical. I don't agree with this theory at large, but once you read his theory attentively, you notice that he is just paraphrasing another theory.

So, Chomsky says, that children will learn the language, using this Universal Grammar thing, provided that adults offer communication to them. Isn't that exactly what behavioural theory is saying? talk to the child long enough, and they will talk back. I would totally support Chomsky's theory (I feel bad for not doing it, I loved his grammar/lexicology/structure works) if children would magically, out of the blue, acquire structures they never heard. If parents never say anything more complicated than present Simple, and the child is answering using present perfect Continuous, then - yes, innate grammar is the only explanation. Another thing that would persuade me would be if children, who were brought up outside of human interaction, still acquired languages. However, there is no research showing that the Mauglis ever achieve any level of language proficiency.

Another "loophole" of this theory is the critical age point. It is proven that if a child does not reach a certain stage of linguistic proficiency by certain age, they are never going to catch up. It does not make sense in lieu of the innate module. After all, nobody proved that grammar and the logic of language are unconditional reflexes.