SNG,
Yes, I understand that critical acquisition period. But the first part of your quote above almost seems almost silly. I take it as a given that 99.9999% of humans live in contact with other humans. So interaction is a given. The children were certainly not exposed to language - the point is that each generation created what did not exist before. But I think we are in agreement here. A single person in isolation would not create language. Again, to me this seems like a silly point, except for the fact that the poor guy would be deprived of a great tool for organizing and ordering his own thoughts.
Leolaia has already questioned the supposed novelty of the Nicaraguan case. I don't know much about it, but I know Pinker has a generativist agenda (well every inguist has one) so I'd follow Leolaia's advice in this regard. I guess I'd have to study the case to say anything more, so I acknowledge the info you've given and I'll try to look into it and see if it has any bearing on my current views. Thanks. As for my point about some cases where there is no interaction - I think statistically speaking it's just about as significant as the case of "creating languages from scratch". And it brings some insights which are relevant to langauge acquisition and the critical period hypothesis. In other words this point is as "silly" or as anegdotal as teh case of Nicaraguan children. (99.9999% if any langauages are not created out of thin air) I'll try to find a description of such a case and paste it here to clarify my point.
If it was a tropical island, would they ever develop the concept of snow? Or is there some determinism involved anyway?I really don't think this is a function of language. Let's say this tropical tribe has no word for snow, since they've never seen it and never spoken to anyone else who has seen it. Does this lack of vocabulary constrain their ability to think of snowy scenes? Or is it the mere fact that they haven't seen snow that constrains them? If one unusual day a tropical snowstorm dumps a meter of snow on the island, will the islanders somehow be disadvantaged in their ability to comprehend it? Or will they not rather invent new words for the new phenomena at that moment?
That's one of the biggest misconceptions of early Chomskian linguistics. No, the literal meanings may not be such a problem, but if you look at any language you'll see that any abstract meaning is based on some culture specific metaphor. It's a ubiquitous phenomenon. It renders the apparent formal equivalence between the grammars of any two languages useless.
The fallacy involved is best illustrated when you make an attempt to translate a piece of text such as the Bible into the language of, say, the Eskimo Indians. If Jesus if the Lamb of God in a MIddle Eastern culture, what will he become in the Eskimo culture? The "Seal of God"? Will such a translation really have the same connotative meaning?
I consider vocabulary to be fairly divorced from the low-level engine that enables language. Words are just snap-in data components. Obviously, the more real-world knowledge you have, the more easily you will be able to create metaphor to describe other concepts.
Contrary to what first Bloomfield and then Chomsky made one of their basic premises, the lexicon (vocabulary) is not just a repository of arbitrary irregularities and idosyncracies. It is extremely systematic, but it's systematicity reflects human cognition, and that's why generative armchair linguists tend to discredit it altogether. I totally disagree with this view of the lexicon. It's out of keeping with all the emiprircal research in language I've done so far. (Not that I have some emotional investment here )
But this is a function of human ability to compare, anthopomorphize, etc, rather than being a magical power bestowed by possessing a set of words.
As I said it probably works both ways. In order to separate the "ability to compare, anthopomorphize, etc," from language you have to conclude that language is more or less contained in syntax. This is a very risky assumption to make especially when it comes to dealing with real language data e.g. in translation.
Language is my thing. Up till now I've examined it only by study of specific languages. Now I'm looking at it from a more abstract view, and it is fascinating.
In the cognitive paradigm I suggest Lakoff and Johnson to begin with some criticism of Chomsky. Then, for a more empirical apporoach to the lexicon and grammar you may want to look at Corpus Linguistics. I specialize in Computational/Corpus Linguistics - in case you are interested in my agenda.
Maybe we should move this to a new thread?
I;m enjoying our exchange and that's fine by me if you dont' think it's not too late to move the discussion to another thread.
Cheers,
Pole