Bloomfield's Taxemes

[Dan Everett]

It does interest me, moderately, and I will indeed be 'going for it' as
part of my OUP book, just a brief section though on Bloomfield and then
a chapter on DM.

I was not asking the question with any motive or expectation other than
to know whether someone had indeed thought about this. In fact,
Bloomfield's concept of Taxeme is very complicated. The only in-depth
review of it at the time in fact, was Pike 1943. Pike found strong
inconsistencies in Bloomfield's use of the term. I cannot recall whether
T. Bever discussed taxemes in his 1967 MIT dissertation on Bloomfield.

It is pretty much always the case that empirical issues emerge when one
carefully considers similar alternatives. Historical research is not
necessary for current research, but knowledge of the literature is. One
cannot excuse poor literature review (and skipping Bloomfield always
falls under this) on the basis that 'one knows of no empirical
consequence'. That of course would be circular. Research always involves
the law of diminishing returns on reading, writing, and experimentation.
One has to stop sometime at each phase.

So I wasn't being critical, nor did I have any axe to grind. Just
looking for some information.

Peace,

Dan


-----Original Message-----
From: The Distributed Morphology List
[mailto:DM-LIST at LISTSERV.LINGUISTLIST.ORG] On Behalf Of Martha McGinnis
Sent: Thursday, February 20, 2003 4:22 PM
To: DM-LIST at LISTSERV.LINGUISTLIST.ORG
Subject: Re: Bloomfield's Taxemes


Dan,

I believe that morphologists in general view Bloomfield's research as
extremely valuable. Perhaps it's just because I've been working on
Algonquian morphology recently, but I think it's too strong to say that
his work is ignored.

My impression is that so far, work within DM has focused on working out
empirical problems, rather than on examining the historical origins of
the framework.  I think this is a reasonable choice, though of course
there are other reasonable choices.  Some of those choices fall within
the realm of linguistics: for example, demonstrating that Bloomfield's
theory of taxemes (etc.) makes different empirical predictions from the
theory of readjustment rules (etc.).  I don't know of anyone who has
taken on this task, but it might lead to interesting results.

Another reasonable choice falls within the realm of philosophy of
science: for example, identifying conceptual similarities and
differences between the two theories.  I'm personally less interested if
it turns out that nothing empirical hangs on the differences, but as you
point out, we won't know until someone seriously considers the question.
If it interests you, go for it!

-Martha
--
mcginnis at ucalgary.ca