LINGUIST List 3.265

Thu 19 Mar 1992

Disc: The Reality of Rules

Editor for this issue: <>

Directory

1. , Where's the Grammar?
2. "Ellen F. Prince", Re: 3.261 Reality of Rules
3. Geoffrey Russom, Re: 3.261 Reality of Rules
4. Vicki Fromkin, (COPY) Re: 3.261 Reality of Rules
5. Martti Arnold Nyman, Rules and intuition

Here are some random thoughts on rules and psychology. They
are familiar from the literature, but may be unfamiliar to
certain Linguist readers.

1. The word "grammar" is, I take it, systematically ambiguous
in Chomsky's mouth. He sometimes means the speaker's competence;
he sometimes means the linguist's grammar. He is very much aware
of this ambiguity, and thinks that it's a good thing.

So Ellen Prince's question about where the grammar is is itself
ambiguous.

In the sense of a theory about competence, who knows
where the grammar is. If theories are anywhere, then grammar (in this
sense) is up there with the theory of evolution, Newtonian
mechanics, and so on.

In the sense of what the speaker knows, it's not unreasonable to
think that there is no grammar in anybody's head. Maybe there are
just causal mechanisms responsible for our linguistic abilities.
(Compare: where is the differential calculus in our heads, which
permits us to catch baseballs, etc.?) It could well be that no
system of grammatical rules is represented in the native
speaker's head. In that case, grammar in the sense of competence
does not exist at all.

2. For me, there's an interesting question about the interrelation
of grammar1 (call it linguist's grammar) and grammar2 (call it
competence): e.g. could a linguist's grammar be true, even if no one
has a mentally represented body of rules that guides their
linguistic behaviour? (I.e. if there is no competence) The
answer, I think, is that there could be. It would still describe
regularities observed in behaviour. But we wouldn't say that it was in
virtue of knowing some grammar that speakers are able to
use language. The native speaker "fits" the rules which our theory
contains, but she does not "follow" those rules.

So linguistics might be like mechanics: no one attributes *knowledge*
of the gravitational constant to rocks. But their behaviour conforms
to that law.

3. What if different native speakers have different causal
mechanisms? Or, what if competence is responsible for speech, but different
speakers of English have internalized different rules? Here
again, I think linguist's grammars are still possible. My
cherished analogy is a theory of the waltz. Even if different
people apply different rules when they dance the waltz, there
can still be a single theory which describes the steps and
how they are executed.

Psychology is irrelevant here.

4. Another issue: I take it that there is something responsible
for our collective ability to communicate linguistically.
And it would be wonderful to find out what it is which
allows people to do this. But the interest of this study
does not rule out another kind of more traditional linguistic
inquiry: about the nature of languages. Returning to the
waltz analogy, it's perfectly okay to ask what goes on it
people's heads when they dance. But one could easily enough
forgo that (very interesting) question, and just study the
waltz.

5. Undoubtedly there are a multitude of ways of describing
the waltz. Each theory must conform to certain facts. But once
a the theory is consistent with the facts, and shows real
promise of *remaining* consistent with future dancings of the
waltz, it is correct. The existence of other equally correct
theories is not relevant.

I would admit that there may be only one correct theory of how
speakers actually manage to dance. But there are a multitude
of theories of the waltz that get all past, present and future
dancings.

We decide between them not on the basis of truth or falsity,
but on usefulness. Which theory we choose will depend on what
we want to do with it.

Returning to the case of linguist's grammars, I think the same
holds true. There are, in principle, any number of grammars
that capture English (or some dialects thereof). Which one
we opt for depends on whether we want to
a. use the grammar for teaching English to German speakers
b. use the grammar to create a machine parser for English,
c. use the grammar for literary criticism,
or what have you.

Best,
Rob Stainton
MIT

Message 2: Re: 3.261 Reality of Rules

>From: jscmbeya.research.att.com (John S. Coleman)
>Subject: 3.250 Rules
>
>Ellen Prince asks
>
>> uh, where exactly *are* these grammars, if not in people's brains?
>
>In the Platonic World of Ideals, of course, along with pi, the square root of
2
> etc.

>From: Penni Sibun <sibunparc.xerox.com>
>Subject: Re: 3.250 Rules
>...
>well, grammars are indubitably in *some* people's
>heads/brains/minds---the heads/brains/minds of linguists. it's far
>less clear whether they are in those of people in general.

ok, i'll bite. what do *you* call whatever it is in people's heads that
enables them to (appear to) communicate by making noises with their upper
respiratory tracts (or by moving their hands and faces)? a list of sentences?

Message 3: Re: 3.261 Reality of Rules

Linguists' grammars can hardly be more elegant than the grammars
people use intuitively, since many systematic features of linguistic
behavior have yet to be captured as explicit rules. To compare any printed
grammar with a complete internalized grammar is a mug's game. An internalized
grammar could of course be elegant without being IN a brain,
since this grammar might be supervenient on brain function without
being reducible to a particular kind of functioning in any one person's
wetware. Inelegancies of data compression or whatever are hardly of
central interest when we talk about the way a program works. Why
assume that neurological quirks have linguistic significance? Or
any significance at all?

 -- Rick

Message 4: (COPY) Re: 3.261 Reality of Rules

re psychological reality of rules. It's interesting how the
anti-mentalist (behaviorist) / mentalist debate which flourished before
Chomsky put the mind back into the brain with Syntactic Structures
and the changed goals of lingistics which generative grammar ushered
in is now replaced by arguments re how can we be sure the rules or
principles posited for grammars are 'psychologically real'. Kiparsky's
now classic 1968 paper 'How abstract is phonology?' reopened theis
quest for Sapirian type of evidence. But as is true in all sciences
evidence is evidence is evidence -- and as NC pointed out, correctly
I believe, one kind of evidence is not necessarily better than another --
there's just good evidence and bad evidence to support a theory or
hypothesis in a theory. Anyway, it is an interesting issue which
is obviously related to one's philosophy of science and which has certainly
intrigued me considering the number of my own papers which deal with aspects
of the question. The issue has stimulated lots of research using speech
errors, aphasic language, language games, etc etc. as a means of
providing different kinds of evidence (which I believe are intriguing but
not intrinisically any more valuable than linguistic evidence itself
including speakers judgments and intuitions). Anyway, when these
messages get too long readers log off (at least I do) so if there is
anyone out there who is interested in my own thinking will be happy to
send you reprints. If noone asks, I won't be hurt. Vicki Fromkin

> From: ADA612csc1.anu.edu.au (AVERY D ANDREWS)
> Subject: Rules
>
> David Eddington writes (Linguist 3.250, Rules)
>
> >A wealth of literature exists which
> >demonstrates that the majority of what is done in the name of
> >'empirical linguistics' is not empirical at all but rather should
> >be categorized along with the non-empirical sciences such as formal
> >logic (see Derwing, Botha, Linell, Itkonen, Skousen).
>
> I think this literature fails on the whole to impress people like me
> [ .. deleted .. ]
> And some of it is just off the wall,
> such as Esa Itkonen, who seems to think that linguists have
> explicit intuitive knowledge of grammatical generalizations.

I must correct Avery Andrews's false witness to what Esa Itkonen
thinks. In _Grammatical_Theory_and_Metascience_ (Amsterdam: Benjamins
1978), Itkonen writes:
 "A linguist sets out to describe a language that he knows, i.e.,
 of which he possesses atheoretical, intuitive knowledge, but
 when his description proceeds, it produces new, theoretical
 knowledge about which he has no previous intuition. From
 the fact that I know_1 something, it by no means follows that
 I also know_2 how to describe this knowledge_1 of mine in the
 best possible way." (GTM, p.216).

Martti Nyman, Department of General Linguistics, University of Helsinki,
 Hallituskatu 11-13, SF-00100 Helsinki, Finland