Re: The things one finds

John Cowan wrote:

> Matt Pearson wrote:
>
> > The claim that the "vast majority" of generative linguists work only on
> > English, are obsessed with theory, and never go out into the field to
> > gather new data, is just plain false.  Whatever legitimate criticisms you
> > might have about generative theory and methodology, you can't accuse
> > us of not respecting the importance of data.  To do so is nothing less
> > than an insult.
>
> Without any intention of insult, there is a clear methodological
> distinction between people who think that leaving aside difficult
> counterexamples for a future generation of workers is a legitimate
> practice (19th-century physicists didn't go into a tailspin because
> they couldn't explain the anomalous motion of Mercury, which requires
> general relativity, etc.), and people who think that this *does*
> constitute "not respecting the importance of data".
Agreed.  There's a definite methodological divide here.  The
approach to linguistic theory-building which I was taught was:
Try to explain as much of the data as you can (knowing ahead
of time that you won't be able to explain *everything*).  Then
look at whatever data is left over, and base your future research
on trying to account for that data, modifying (or scrapping) your
theory as necessary.  However, implementing this methodology
in an intelligent and useful fashion is difficult, because of the
constant pressure to keep publishing and presenting new stuff
which is 'relevant to the theory'.  So you see a lot of articles and
conference papers out there which basically amount to works in
progress, full of loose ends in the form of 'problematic' data.  The
less honest (or more stubborn) of the theorists will tend to gloss
over the loose ends, or explain them away as being somehow
'irrelevant'.  The more honest theorists will merely acknowledge
that there's still work to be done, and move on.

> To make things
> worse, the history of linguistics is very nearly an empty discipline:
> people are taught, at most, caricatures of what their predecessors
> thought.  (This is true of the natural sciences as well, alas.)
Agreed!  I certainly wish *I* knew more about the history of
linguistics...


> > (Simmering, and hoping people will go back to talking about
> > conlang-related stuff...)
>
> FWIW, *every* professional linguist of whatever stripe seems to get
> into this state from time to time on Conlang, with veiled or
> open threats of resignation.  "Shane!  Don't go!  Shane...."
> In plain fact, the rest of us need you professionals to keep us
> honest, not to say informed.
Believe me, I have *no* intention of leaving this list.  My little
rant was merely a non-so-gentle reminder not to make sweeping
generalisations.

Matt.