linguistics and language universals

From: Mari Broman Olsen (molsen@umiacs.umd.edu)
Date: Fri Mar 27 1998 - 15:26:12 EST


I'm about done with my annual blast of letters to b-greek, so I"ll
free up the bandwidth. I did however want to briefly highlight some
assumptions in theoretical linguistics that do not go without saying
in a list of this nature.

The assumption that there are language universals (or rules or
commonalities between languages) is due, in large part, to the
observation that every child, almost regardless of IQ, acquires
his/her (spoken) language rapidly and with no (or little) overt
instruction. Furthermore, the language acquired is productive rather
than imitative; my three year old is daily putting together novel
sentences, like:

        1. I wanna sit on the booster chair and I press ('pess') the
'P' and you pess the app-ow (recognized by Mac users as the print
sequence)
        2. My head like a circle. (last year)
        3. (at a recent bedtime)
        Hanne (sister): We aren't morning people. We're night people.
        Me: At least you and Kyrie and I are. Dad says he isn't anything.
        Hanne: He's a morning person.
        Kyrie: You [mom] and me are morning people. Me and Hanne are
        night-night people. [truer than she knew...]

This almost universal success, as well as strong similarities in the
acquistion path, suggests Chomsky (father of generative linguistics)
that there are aspects of language that are part of our innate
endowment (similar to the endowment that unfolds in learning to walk,
learning to eat solid food, etc.). And since we have the same innate
endowment and use it to acquire different languages (and none is
appreciably more "difficult" to acquire natively), we hypothesize that
there are similarities between languages that derive from this innate
endowment, which, if properly described, distinguish "possible"
languages from "impossible" human languages (example of the latter
being a language in which every third word has a certain agreement
marker, or one in which questions are formed by completely reversing
the word order: You are more hungry at lunch today => Today lunch at
hungry more are you?)

So the "rules" Rich loves to hate are simply working hypotheses about
i) universal grammar and ii) the grammar of a particular language
derived therefrom, described in such a way that that language is
learnable by the child (in that last statement, I'm describing a
narrower set of linguists, but certainly all would agree that a "true"
explanatory theory of language must address learnability).

So I guess I'm saying linguistics is a science (with art necessary, as
with any science), in which hypotheses (rules) are made in a formal
falsifiable way and then tested against the data. If enough rules are
"broken", it's time for a scientific paradigm shift, or a
rearrangement of theories, but not time to discard the scientific
approach to the study of language. If researchers in individual
languages say linguists miss the beauty in the individual variation
and nuances (and eschew ;-) the possibility of formalizing any
linguistic descriptions), I point them to see the amazing beauty in
the cross-linguistic regularities one sees climbing further up in the
forest of trees.

AND I get paid to look at the view. Such a deal. I'd call it a
vocation, in the old sense. Now I need to write journal articles, so
I can get a tenure-track job ;-)

Mari
(lowly but happy postdoc)

********
Mari Broman Olsen
Research Associate

University of Maryland Institute for Advanced Computer Studies
3141 A.V. Williams Building
University of Maryland
College Park, MD 20742

PHONE: (301) 405-6754 FAX: (301) 314-9658
WEB: http://www.umiacs.umd.edu/~molsen
*********



This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:39:19 EDT