reply to post by dr_strangecraft
You learn to speak piecemeal. You can copy the sounds like a parrot would. But then you begin repeating the sounds back to people to see how
they respond. You learn the meaning of the word based SOLELY on how other people respond to it.
Is this really what you have observed?
Kids repeating words like parrots, trying them on different objects and seeing how the people around them respond?
I haven't seen that, and I doubt that many on this thread would say they had, either. If that was how children learnt language, they would take
thirty-six years, not thirty-six months, to do it.
Most of us have had the experience of hearing their child or toddling younger sibling utter a word for the very first time, applying it to precisely
the right thing - usually with such an air of satisfaction and certainty that it is clear they know beforehand that they are right. Which, of course,
they usually are; this kind of occurrence is the rule, not the exception. Compare the size of the average four-year-old's vocabulary and the number
of times the child had to be corrected about the meaning of a word in the course of acquiring it.
And vocabulary isn't even the point. Acquiring a vocabulary is the easiest part of language: a mere question of matching labels with objects or
concepts, easily done by rote. The hard part of language is making sentences out of those words you've learnt - sentences that will correctly convey
your intended meaning to others. This is the province of syntax and grammar, with their frightening complexity, their variable applicability, their
endless permutations and idiomatic exceptions. The number of possible combinations of words in grammatical sentences is so vast as to be effectively
infinite; rote-learning will get us nowhere here.
Yet learning to speak in grammatical sentences is something small children do rapidly and automatically. They do it so well that even their errors are
amusing to us, because when we hear one we instantly recognize how it arises from some innocently misapplied rule of syntax or grammar - which means
that they have already understood the rule correctly - understood it without ever having been taught, just by listening to other people
Children learn language like this because evolution, or God if you prefer, has designed the capacity to do so into them. They are born with something
you might think of as language-acquisition-and-generation software built into their brains. That software defines the universe of possible human
languages because it generates the rules of syntax and grammar for every last one of them. These rules vary from language to language but always
remain strictly within the parameters set by the software*. That sounds remarkably like Chomsky's 'generative grammar' to a lot of people.
But whether or not it corresponds to what Chomsky meant by that term or not, some neural facility of this kind clearly exists in children. They
emphatically do not
learn language like parrots - unless, as Irene Pepperberg
parrots actually learn language like children.
But that is another story.
*This kind of statement demands evidence; I can do no better than refer you to the chapter notes to The Language Instinct
more recent book by Pinker, The Stuff of Thought
. This second book looks in quite close detail at the way the brain turns thoughts - actually
not even thoughts, but the primitive components of thought - into words. I found the section on prepositions particularly fascinating. and intuitively
convincing as well. It is heavier going than The Language Instinct
, however; at least, I found it so.
[edit on 25-9-2008 by Astyanax]