The Critical Period Hypothesis
The crticial period hypothesis in essence contends that the ability to learn a language is limited to the years before puberty after which, as a result of neurological changes in the brain (see the impact of neglect on neurological development), the ability is lost.
She [Genie] was a very communicative person. But, despite trying, she never mastered the rules of grammar, never could use the little pieces — the word endings, for instance. She had a clear semantic ability but could not learn syntax. Rymer, Genie: A Scientific Tragedy.
Feral Children and the Critical Period
Although the critical period hypothesis was hotly debated for some years, there is now compelling evidence — including the evidence from feral, confined and isolated children — that, unless they are exposed to language in the early years of life, humans lose much of their innate ability to learn a language, and especially its grammatical system. However, read on…
Modern Feral Children
Even if they've missed out on the critical period for language acquisition (such as Genie), feral children can be taught a few words, and very simple grammatical constructions. However, feral children don't provide the best evidence in support of the critical period hypothesis (which is, any case, now generally accepted), partly because they may have been abandoned because of subnormality (Victor) or suffered emotional and physical trauma (Genie) that would affect their learning capacity.
Language acquisition after return to civilisation
The ability of feral children to learn language on their return to human society is very varied. For most feral children from history, we don't have enough information to judge exactly how much language, if any, they might have been able to learn, were they taught properly. For some children, the historical records don't even mention whether or not they could talk when they were found, presumably because the assumption is that they clearly wouldn't have been able to.
Some children (see Isabelle) acquire normal language ability, but only if found before the onset of puberty. Her progress was dramatic: in two years she covered the stages of learning that usually take six years. Others, such as Memmie LeBlanc), also learnt to speak normally, but we suppose that they could speak before their period of isolation.
…...the second claim ([that Genie could produce] 'no auxiliary forms') is simply false. Jones, Contradictions and Unanswered Questions in the Genie Case.
A grammatical puzzle
But the evidence from more recently-discovered children such as Genie is confusing. Although Genie is often quoted as evidence that there is a critical period, in fact, in Genie: A Pyscholingustic Study by Susan Curtiss, we read that Genie did start, and continue, to acquire gramatical ability.
Unfortunately, Genie's language regressed after legal and financial considerations put a stop to the nurturing scientific environment she enjoyed for the first several years after her release. But the original evidence is also thrown into further confusion by later publications about Genie, which suggest she acquired little or no grammatical capabilities: for more on this, you can read Peter Jones's paper Contradictions And Unanswered Questions In The Genie Case online.
A Sensitive Period for Language Acquisition
In A Theory of Neurolinguistic Development, John L Locke provides us with a possible answer to this puzzle. He suggests the term sensitive period rather than critical period: a period which is optimal for "tuning" that part of the brain best suited to the acquisition of grammatical analysis. However, even after this period, the considerable adaptability of the brain means all is not lost: other, less optimal, parts of the brain are pressed into service, and some grammatical abilities can be acquired, albeit slowly.
Further reading online
Read Steven Pinker online on the subject of language acquisition in Language Acquisition, a chapter of An Invitation to Cognitive Science.
There's a chapter in Roger Brown's Words and Things on the subject of language acquisition in isolated children. See also Speech and the Development of Mental Processes in the Child.
In a little over two months after her first vocalisation she [Isabelle] was putting sentences together. Nine months after that she could identify words and sentences on the printed page, could write well, could add to ten and could retell a story after hearing it. Davis, Human Society.