LINGUIST List 4.746

Thu 23 Sep 1993

Disc: The Linguistic Wars

Editor for this issue: <>

Directory

1. , Linguistic Wars and History of Linguistics
2. , Re: 4.738 The Linguistic Wars
3. Steven Schaufele, Re: outdated theories & data.

The recent discussion on this seems to me perhaps to pay a bit to
little attention to two points:

(a) The history of linguistics is really a separate branch of inquiry
from, say, the study of reciprocal constructions in Bangbubangu.
This cuts two ways: Those who decry the lack of historical awareness
among many linguistics students really must make a distinction between
the stuff which they think every linguist should know, and the stuff
which is to be required only of students of the history of linguistics.
And those who protest that they cannot possibly teach all the
history of linguistics in the course of teaching linguistics must see
that that is not what is being asked.

(b) More speculative, but it seems to me that the issue is really
not how much history is made available to students or in what form
but rather what version of history is offered to them. Roughly,
it seems that there are two on offer currently (and both were
reflected in recent discussions). Oversimplifying like crazy, I
would say that these are:

 (i) Linguistics has been evolving quickly, unidirectionally,
 and without looking back 1956 or so, and since all
 the issues that arose along the way (not to mention those
 which used to bother linguists for the preceding
 milennium or so) have long been settled.

 (ii) Linguistics has been moving in all sorts
 of directions, often making egregious errors
 and having to turn back, and is still doing
 so, and many (most?) of the old issues are
 not only unsettled but perhaps only now coming
 up again for serious discussion.

Message 2: Re: 4.738 The Linguistic Wars

Margaret Fleck and David Pesetsky have valid points about general
introductions to the history of linguistics in a syntax survey course.
What is *less* acceptable, in my view, is a grad student's ignoring
previous work on a specific topic they are working on; the data may well
be valid even if the theory is "out of date". This is partially the
responsibility of faculty working with the student -- the faculty should
know better. Without this perspective, we end up re-reinventing the wheel.
Susan Fischer

In LINGUIST 4-738, Rick Russom asks if anyone holds the following position:

>Somebody who finds a crucial case for a theory gets credit
>for that crucial case only with regard to the theory actually
>cited. If the same case turns out to be crucial for some
>newer theory (notational variants excluded), no credit is
>awarded the original finder of the case, and all credit is
>awarded to the person who cites the case as crucial for the
>newer theory. In other words, no special credit is assigned
>to someone who finds a case that turns out to be crucial for
>lots of theories, the general utility of the case being attributed
>to accident.

Count me out. Data are data. Perhaps a particular datum is more
*interesting* within one theoretical framework than another, but -- unless
and until it is empirically refuted -- any datum about linguistic
usage/behaviour is still part of what our science is supposed to be
explaining. It shouldn't have to be 'rediscovered' after a new theoretical
framework arises to absorb the interest of researchers. To take a
(perhaps) analogous example from the history of chemistry, the fact that
e.g. lead oxide is heavier than the equivalent amount of pure lead didn't
have to be rediscovered after the phlogiston theory was abandoned. If i'm
working in, say, LFG, and come across some early paper by e.g. Haj Ross
from which i learn of a body of data that is absolutely critical for the
LFG account i'm trying to develop of some phenomenon, i sure as Grimm am
going to cite that paper of Haj's and give him full credit for having
tipped me on to something. Doesn't matter a hill of beans that at the time
he wrote the paper in question LFG wasn't born or thought of.

In the same issue of LINGUIST, Esa Itkonen accuses David Pesetsky of
>an interesting view of progress in linguistics. According to him,
>linguistic theories get outdated by some uniform and ineluctable
>historical process. Thus, theories produced in 1967 have become
>outdated in 1983 to exactly the same degree as theories produced
>in 1951 had in 1967. As intersting as this view is, it flies in the
>face of the facts. For instance, Apollonius Dyscolus invented
>(rather than discovered) the so-called performative hypothesis
>some 1800 years before Ross did. This means that, rather than
>being outdated by, Apollonius outdid (on this issue) the theories
>produced during some 1799 years after his time. Similarly, Panini
>invented underlying (morpho)phonological forms and rule-ordering
>some 2400 years before Bloomfield and some 2420 before Chomsky
>& Halle (not to speak of the invention of the theta-roles).

Now, Esa is quite correct concerning Apollonius and Panini, but not about
Pesetsky. As i remember, David was saying that, between 1967 and 1983 as
between 1951 and 1967, 16 years had gone by and *16 years' worth of
linguistic scholarship had been published*. A lot of stuff that was 'up
and coming' when any of us were grad students has not stood up very well;
some has. I personally feel that any generation of linguists will produce
a few 'classics' that will retain their value to the general field (ergo,
their importance to new recruits) for a long time, and i consider Ross'
dissertation to belong to this class. But i agree with David to the extent
that we cannot expect and ought not to expect (i think there's an ethics
issue here) our students to have to read all the stuff we ourselves
consumed as MA-candidates. I don't know about other departments but here
at Illinois we have a course in History of Linguistics that is required of
all linguistics grad students. To some extent, that's where students get
exposed to Apollonius and Panini, and there may come a time when Ross is
included there too. At present, Ross' dissertation is still of such
immediate relevance that it gets covered in Introduction to Syntax courses,
as the Neogrammarians get covered (at least corporately) in Introduction to
Historical Linguistics courses.

By the way, forgive a poor Midwestern linguist who's never been near MIT,
but how did John Robert Ross come by the nickname (ekename) 'Haj'? Could
somebody clue me in?
------
Dr. Steven Schaufele 217-344-8240
712 West Washington Ave. fcoswsux1.cso.uiuc.edu
Urbana, IL 61801

*** O syntagmata linguarum liberemini humanarum! ***
*** Nihil vestris privari nisi obicibus potestis! ***