Re: THEORY: OT Syntax (Was: Re: THEORY: phonemes and Optimality Theory tutorial)

On Sat, 18 Nov 2000, Marcus Smith wrote:

> Yoon Ha Lee wrote:
>
> ><wry g>  From what I can tell the history of mathematics is filled with
> >examples of speculations and ideas that did indeed turn out to be
> >"rubbish."  Theoretical math involves a lot more intuition and guesswork
> >than I realized
>
> You get this in linguistics as well. If you read material from the early
> days of linguistics, there is quite a lot of rubbish in it. There is also
> quite a lot of insight that was simply formalized is wrong, so imcorrectly
> dismissed by later researchers.
That's quite interesting--someday, when I learn more linguistics, I'd
like to find a history of the field.  :-)  Right now, I know too little
about a lot of things for such a history to make much sense to me.  (I'm
working on phonetics/phonology right now....)

> >To my knowledge calculus stayed around 'cause it worked, and because
> >later mathematicians were able to find a much more solid theoretical
> >foundation for it.
>
> In generative linguistics, this happened with what is called the
> A-bar-trace. The theory stated that when you move a topic or wh-pronoun to
> the front of a sentence, it leaves behind a "trace" - a footprint in the
> structure, if you will. The evidence was weak but the intuition was good.
> During the late 70s or early 80s, they began running tests on brain
> activity during speech. Brainwave patterns seem to corroborate the
> existance of traces in a very dramatic way. Once the "moved" word has been
> pronounced, the brain goes into a flurry of activity which only goes away
> once the speaker reaches the point where the trace was hypothesized to be.
> That is actually very startling, and I really doubt any syntactician
> expected evidence of this kind. Similar evidence of the A-trace is lacking,
> but syntacticians ignore that fact.
Oddness!  Are brainwave patterns used to corroborate other kinds of
findings?  Is this a common way of researching things in linguistics, or...?

> >You'd think math would be an example of a discipline where "rationality"
> >would prevent "rubbish" from appearing, and it isn't true.  Personally,
> >I'm not surprised this kind of thing shows up in linguistics (or history,
> >or the field of your choice) as well!
>
> I have read linguists who never have anything reasonable to say. And yet
> they are quite famous in the field. They point out interesting and
> troublesome facts, but the way they account for them is non-sensical. Or
> rather, they are possible according to the technical capabilities of the
> theory they work under, but they are counter-intuitive as to why that
> should be the proper analysis. For me, Chomsky falls into that category
> most the time.
<wry g>  I started reading something that was *supposed* to be a layman's
intro to Chomsky and gave up after about 10 pages, partly because the
thing seemed abstruse, partly because my knowledge of grammatical
terminology was pretty poor (it's better now, but I don't claim it's good!).

I should try again now that I know a *little* more about linguistics, and
see if he makes any sense now.  Theoretical linguistics is probably going
to be beyond my grasp until I have time to do more catch-up reading,
unfortunately.