> we must have some special structure in the brain that gives us this capacity, that we are born with
We have special faculties for things like recognizing faces. Correspondingly there are people that have some damage/abnormality with this mechanism but they are otherwise fully functional human beings. There doesn't seem to be an equivalent phenomena with language. I mean, you have things like dyslexia.. but that's something that only applies to the writing word (which is kinda bizarre if you think about it b/c writing is relatively new, evolutionary speaking). As far as I know, people that can't process language at all are deeply mentally retarded. If it was some complex hard-wired mechanism in the brain then you'd expect to see functionally intelligent people that can't process language at all - but as far as I know that doesn't happen. Or does it?
That could add to the idea that language is primarily the tool we use for thought. If you can't have abstract thought without language (in humans), as Chomsky claims, then that would fit perfectly with an observation that people who are incapable of language are also incapable of 'higher' thought.
I'm not personally sure whether this observation is accurate or not.
I'm not generally persuaded that we think linguistically, my experience on reflection seems to be that forming ideas linguistically is 'post processing'. On the other hand the rigor of trying to express a thought or idea linguistically can greatly aid the thinking process. If they are 'technically' separate faculties, I suspect that in humans the specifics of the implementations are extremely highly coupled and interdependent.
To me at least, anything longer than 1 'idea' must be put together in language. For example, I couldn't think of a mulți-step algorithm or a plan or even a recipe without putting it in language.
Still, I think Chomsky's idea, as far as I understand it, is that there is a level of language that is not immediately accessible to us, where basically your mind constructs a concept tree that can be 1:1 mapped with the syntax tree of the phrase that it could be used to represent. The strucutre itself is the same, but the thought process works with internal concepts, while the expression in external language works with English or Japanese words. The brain would work with this tree internally, applying well-defined transformations to it, merging it with other similar trees, embedding it etc. ; this would happen both consciously, semi-consciously and unconsciously to different degrees. In the end, when a plan is formulated, it can be acted upon directly, or it can be communicated to others, by encoding it in English or Japanese or Sign Language etc.
Note that I am not claiming to have studied this seriously, so this is at best a layman's understanding from watching a few lectures. Chomsky also is careful to mention that his idea of language as the mechanism primarily for thought is a fringe view in linguistics, though he strongly believes some version of it will come to be accepted as correct in the future, as the field matures.
We have special faculties for things like recognizing faces. Correspondingly there are people that have some damage/abnormality with this mechanism but they are otherwise fully functional human beings. There doesn't seem to be an equivalent phenomena with language. I mean, you have things like dyslexia.. but that's something that only applies to the writing word (which is kinda bizarre if you think about it b/c writing is relatively new, evolutionary speaking). As far as I know, people that can't process language at all are deeply mentally retarded. If it was some complex hard-wired mechanism in the brain then you'd expect to see functionally intelligent people that can't process language at all - but as far as I know that doesn't happen. Or does it?