LINGUIST List 4.824

Tue 12 Oct 1993

Disc: Linguistic History, Science

Editor for this issue: <>

Directory

1. Randy Allen Harris, Linguistic Pedagogy and History of Linguistics
2. Steven Schaufele, scientific linguistics? part 2

Supplementary textbooks are a great way to expose students to material
(like the history of the discipline) that you don't have time to take up
formally. Just put one on each syllabus in the "Recommended" column, have
the bookstore order them (roughly at about 50% of enrollment), and let the
students' curiosity do the rest. There are several well-written and
engaging books that can let the students follow historical developments
without taking up any course time. Putting a few such books on Ph.D. Comp
reading lists might also be useful.


Here are some that are (imho) particularly good for this purpose. You
might have to alert students to the ideology of one or the other, but
they're all valuable.

Anderson, Stephen. 1985. _Phonology in the twentieth
 century_. Chicago: University of Chicago Press.

Andresen, Julie Tetel. 1990. _Linguistics in America
 1769-1924_. London: Routledge.

Darnell, Regna. 1990. _Edward Sapir: Linguist,
 anthropologist, humanist_. Los Angeles:
 University of California Press.

Murray, Stephen O. Forthcoming. _American linguistic
 theories and theorists: A social history_.
 Philadelphia and Amsterdam: John Benjamins.

Newmeyer, Frederick J. 1986. _Linguistic theory in America_
 [1957-1986]. 2nd edition. New York: Academic.


Other suggestions? (In particular, mine are heavily biased toward American
work--for which I apologize, but that's the area I know--and there must be
work on the history of the discipline elsewhere).


While I'm posting suggestions for supplementary texts, I'll just add two
that aren't historical but do help convey the (often mislaid) sense that
linguistics is a fun and even adventurous pursuit.

Pullum, Geoffrey K. 1990. _The great Exkimo vocabulary
 hoax_. Chicago: University of Chicago Press.

Dixon, R. M. W. 1984. _Searching for Aboriginal languages_.
 Chicago: University of Chicago Press.

Suggestions in this category, too, would probably be appreciated by others
besides me.



Randy Allen Harris rahawatarts.uwaterloo.ca
Rhetoric and Professional Writing 519 885-1211, x5362
English, U of Waterloo FAX: 519 884-8995
Waterloo ON, CANADA, N2L 3G1

Philip Swann (LINGUIST 4-771) brings up three challenges to my earlier
rebuttal of his claim that linguistics does not qualify as a science. I
will here strive to answer each in turn.

>1. Similar claims have been made about the whole range of social "sciences",
>but they have generally been rejected by philosophers and historians of
>science.

I am tempted to challenge Swann to produce his list of authorities and i
will produce mine, since i suspect i could come up with approximately as
many 'philosophers and historians of science' who include the 'social
sciences' in general and 'linguistics' in particular among the sciences,
but i won't because that would be beside the point. With no context given,
an argument from authority is bootless. Science is not ruled by majority
or authoritative opinion, even among its professional practicioners. The
relevant questions should be, By what criteria does a given person, of
whatever authority, judge whether a given discipline is a science or not?
and, Are those criteria, in the final analysis, valid or adequate? I have
attempted to give my criteria, though as will be seen below they may be in
need of emendation.

Historically, my own method of wrestling with this question (ever since
Georgia Green, in her Introduction to Syntax class back in 1983, tried to
convince us that linguistics is a science) has been to make up a list of
disciplines that i would never doubt qualify as sciences, and to see what
characteristics they all had in common. My list was and is, in
alphabetical order: Astronomy, Biology, Chemistry, Geology, Mathematics,
and Physics. Note that most of these are non-laboratory disciplines
(Astronomy and Mathematics by definition, Geology insofar as it differs
from Chemistry, and a good deal of Biology as well) and that Astronomy in
particular cannot rely on any 'repeatability' criterion.

>2. A rational, methodical and data-driven investigative style
>is not enough to define an activity as "scientific". Otherwise,
>as Schaufele suggests, practically everything we do becomes
>science: Why not cooking and gardening?

I would regard cooking and gardening as applications of science
(biochemistry and botany, respectively, if i had to be specific); they are
not oriented toward acquiring new information (to use Swann's terms, they
are not 'investigative styles' because there is nothing inherently
'investigative' about them) but toward exploiting information already
available in order to achieve a predetermined, material end. They are
therefore more closely related to technology than to science per se.
Linguistics, on the other hand, is (presumably) oriented toward using a
suitable variant of the scientific method to acquire new information about
human language. In this respect, like the six disciplines listed above, it
is more closely related to science than to technology.

And i would deny that, given my definition, 'practically everything we do'
could qualify as a science (i don't remember suggesting any such thing), or
that this would be desirable. Theology cannot meet my definition because
although academic theologians may (very properly) use a variant of the
scientific method to develop deeper understanding of their data, their data
are not objectively verifiable; in order to even follow their arguments it
is occasionally necessary to accept a priori the validity of the
(empirically unverifiable) data on which these are based. Thus, i am able
to be swayed by the arguments of theologians in the Christian tradition
because i accept the validity of their data, but not by those of Hindu,
Jewish, or Moslem theologians because their data are inherently alien to
me.

On the other hand, literary criticism cannot be scientific, because
although the data are objectively verifiable the literary judgments that
are the end of literary criticism and the arguments that support them are
often subjective. (Textual criticism, the determinatioon of the most
'authoritative' version of a literary text, can however be done
scientifically, and has a good science-type name, 'paleography', to reflect
this.)

Note that both of the disciplines i have mentioned in the previous two
paragraphs are scholarly, academic disciplines (and, if you substitute
'music' for 'literary' in the last paragraph, they are both disciplines in
which i claim some experience and expertise -- as i do in cooking).
'Scientific' status is not essential to scholarly activity. Nevertheless,
as i indicated in my earlier posting, if it really is possible to pursue a
discipline scientifically and a large number of its practicioners are
striving to do so, why not call it such?

>3. Consider the stock exchange, a semiotic system at a level
>of complexity similar to that of language. An enormous
>amount of effort has gone into trying to build scientific
>theories of price movements in the market. These efforts
>have failed, so it seems, because the market performs a random
>walk driven by greed and fear in a space that is detached from
>underlying economic reality. All the retrospective studies
>confirm that there is no way to predict the stock market. In
>other words, it has been demonstrated scientifically that the
>market is not open to scientific description.

First of all, i have no brief for economics. Whether economics meets my
definition of a science is a matter of ignorance and, frankly, disinterest
to me. I would note, however, that on the face of it much of what Swann
says about the fluctuations of the stock exchange can be asserted, mutatis
mutandis, of elementary particles as well. Given Heisenberg's Uncertainty
Principle, 'there is no way to predict' the behaviour of individual
particles; there isn't even a way to know where they are and what they're
doing right now, never mind predicting. But Quantum Electrodynamics is a
supremely successful and useful theory, partly because it doesn't try to
pinpoint all the details but accepts that there is a level of detail below
which it is impossible to go and still maintain scientific rigour -- and
that an in-depth understanding of events below this level is not inherently
necessary or, perhaps, desirable. I submit that the same may be true of
economics and -- more important for the present context -- linguistics. I
am quite confident that we will eventually develop a body of linguistic
theory that makes accurate predictions at a suitable scale about human
linguistic behaviour, even though it won't be able to account for every
little detail. The important question in my own research, as it was behind
much of the development of quantum theory, is just what details are 'too
little' to be covered by the GUT (known in linguistics as UG) and which may
 *seem* 'trivial' but are in fact of vital importance.

Sincerely,
Steven
 ------
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! ****