LINGUIST List 2.643

Thu 10 Oct 1991

Disc: What is a Linguist? Is Language Finite?

Editor for this issue: <>

Directory

1. bert peeters, 2.634 Is language infinite - Infinite polysemy
2. , 2.634 Is Language Finite?
3. "Michael Kac", Re: 2.593 What is a Linguist?
4. paul saka, RE: what is a linguist?
5. , Re: 2.580 What is a Linguist?

> Date: Tue, 8 Oct 91 14:12:42 EST
> From: j.guytrl.oz.au (Jacques Guy)
>
> To me, a finite lexicon (which I
> see as the tautology of tautologies) can express new ideas and
> referents, without additional words, even without new compounds, old
> words simply taking on new meanings, sometimes related to the old ones,
> sometimes not.
"Old words simply taking up new meanings" - infinite polysemy is rearing its
ugly head again... I'm not saying that new meanings cannot be added to old
words; there is absolutely no evidence whatsoever in support of such an
extreme view. However, lexicographers and linguists tend to posit polysemy
a little too easily, I believe, and consider as a separate meaning something
which is purely and solely the effect of the context in which a word is used.
If a word has always been used in contexts A, B, C until P, it has one or
more meanings (probably not as many as there are contexts); now, if that
word starts being used in a context Q, which is entirely new, it does not
automatically follow that it takes on a new meaning. The existing meaning
or one of the existing meanings may be sufficiently coloured by the new
context to create the impression that a new meaning has just come about.
Lexicographers and linguists alike tend to underestimate the power of con-
text. I for one was very impressed by Charles Ruhl's work on monosemy,
where he posits that words are necessarily monosemic (that's the working
hypothesis) until proven otherwise. Ruhl's thesis cannot yet be considered
as final: it does need tuning - but it is a hypothesis worth looking at.
Reference:
Charles Ruhl, *On monosemy. A study in linguistic semantics*, Albany: State
	University of New York Press, 1989.
(A review is forthcoming in the Canadian Journal of Linguistics; other reviews
I know of are by Barbara M. Birch in *Language* 66:4 (1990), pp. 881-882, and
by Adrienne Lehrer in *Journal of linguistics* 27:1 (1991), pp. 298-300.)
Dr Bert Peeters Tel: +61 02 202344
Department of Modern Languages 002 202344
University of Tasmania at Hobart Fax: 002 207813
GPO Box 252C Bert.Peetersmodlang.utas.edu.au
Hobart TAS 7001
Australia

I hate to differ with Avery Andrews, but I do not see that increasing
the quality of life-support and such justifies assuming that sentences
have no upper bound on their length. Unless that is we assume that
human life spans have no upper bound, and even then there are problems
connected with the fact that might have spend your entire (unbounded)
life span just saying ONE very long sentence. The assumption that
sentences have no upper bound on their lengths strikes me as a
reasonable idealization but in no way superior to the one that Langendoen
and Postal advocate.
What continues to puzzle me about this whole debate, I should add,
is that a lot of people seem to talk as though (a) something important
rides on these distinctions and even as though (b) they were things
you could decide on factual grounds. Whereas I believe that the
people who came up with these ways of modeling things (people like
Turing, for example) would have agreed that we are just dealing with
convenient idealizations. In general, infinity is a convenient
way of talking about very big sets and especially ones which are
difficult to list (but easy to define), like the set of sentences
of a language, for example. Not that I am advocating the idea
that the set of sentences of English is REALLY finite, as some have.
I would argue that it REALLY is not anything, for the whole notion
of the set of sentences of English is itself a convenient fiction.
(This last point incidentally is what Chomsky refers to in his
recent writings when he says that E-language is not a real object.)
What does everybody else think?
Alexis Manaster Ramer

Message 3: Re: 2.593 What is a Linguist?

Hmmmm ... As a sometime graduate student at the University of Pennsylvania,
a Harrisian to me is a follower of Zellig, nor Roy. Ironically, the former
would appear to represent the epitome of what the latter seems to despise.
What, indeed, is in a name?
Michael Kac

	Recent discussion motivates me to elaborate on some of
my earlier remarks.
	First, I claimed that a linguist is one who (saliently)
engages in the study of language. Randy LaPolla reports that
in Chinese and Japanese the closest translation of "linguist"
suggests some sort of prestige or fame. It would seem, then,
that languages (or cultures?) conventionalize the degree or
manner of salience that is appropriate for "-ist" words.
	It has also been observed -- I forget by who, and
apologize -- that "linguist" applies more readily to undergrads
in Britain that in the US. This again MIGHT indicated a
difference in conventionalization; but it might alternatively
reflect the fact that undergrads at British universities
are almost as specialized as grad students in America.
	Second, I did NOT say that a linguist is one who does
research in linguistics, which would indeed be virtually vacuous;
I claimed that a linguist is one who does research into language.
This, I think, is a perfectly straight-forward definition that
causes difficulties only if you think in terms of exclusive
categories. A psychologist or philosopher can be a linguist
(or not) just as easily as a breadbasket can simultaneously be
made of wicker (or not).
	The reason that you would hesitate to call, say, Donald
Davidson a linguist has to do with Gricean concerns. While it
is a salient fact that Davidson studies language, the coordinate
fact that Davidson is a philosopher is yet more salient. This is
because administratively defined departmental boundaries constantly
impose themselves on the every-day activities of academics.

Message 5: Re: 2.580 What is a Linguist?

I appreciated Bert Peeters' affirmation that to do semantics is to do
linguistics. Yet when he writes that "there is a place out there in
linguistics for the study of meaning", I sense a lingering apology
for semantics as the poor relation of linguistics.
The study of meaning is no poor relation -- if it is not a central
concern in linguistic analyses, it ought to be.
Phonology, morphology and syntax are core to linguistic study because
1) each studies explicit elements of language structure and hence
2) rigorous analytic frameworks are readily available.
The language phenomena we call semantics is elusive: distinguishing
word knowledge from world knowledge and pragmatic knowledge is,
to put it mildly, tricky. Development of a rigorous methodology for the
study of linguistic meaning is still in its early stages.
Yet, if we accept Saussure's notion of the sign as a pairing of
signifier and signified (form/content), and if we accept that
language is a system of signs then meaning is central to language.
Any study of language which omits examination of meaning is incomplete.
Thus, not merely is there 'a place out there in linguistics for the
study of meaning' but if we take seriously that language is a sign system
then the place of meaning ought be as central in linguistics as the
signified is to the construction of the sign.
Rebecca S. Wheeler
Utah State University
Logan, Utah
507E 100N
Smithfield, UT 84335
REBWHLRcc.usu.edu