[Corpora-List] Looking for linguistic principles

[Rob Freeman]

On Wednesday 19 October 2005 10:40, Mike Maxwell wrote:
>
> Anderson talks about a couple such cases in his "Reflections on 'On the
> Phonetic Rules of Russian'" paper, where American structuralists like
> Bernard Bloch (this was in 1941) said roughly "Such-and-such might look
> like the right analysis, but our theory shows us that the right analysis is
> s.t. else."  Maybe that's a product of other scientific methodology of that
> era, but it still sounds to me like discovery procedures...

I don't care what you call this, but it is the sort of evidence I am 
interested in.

Were these kinds of "learning conflicts" ever resolved, except by Chomsky's 
proposal that something innate was selecting between the various possible 
analyses?

For every other point you raise, Mike, I can find good counter arguments. For 
instance a quick Web search gives me this for aux. verb fronting:

http://www.ecs.soton.ac.uk/~harnad/Papers/Py104/pullum.learn.html

But what about these early observations of conflicts in the learned analyses?

-Rob