LINGUIST List 5.224

Fri 25 Feb 1994

Disc: Universal grammar

Editor for this issue: <>


  1. David Powers, Re: 5.222 Universal grammar
  2. Richard Hudson UCL, UG

Message 1: Re: 5.222 Universal grammar

Date: Fri, 25 Feb 1994 18:42:07 Re: 5.222 Universal grammar
From: David Powers <>
Subject: Re: 5.222 Universal grammar

In linguist 5.222, Simon Kirby <> writes...
>In linguist 5.200, Joseph Stemberger writes...
>> I was wondering how this interacts with the term "Universal Grammar". It's
>> my impression that innatists tend to use the term "UG", while others just
>> talk about "universals". So, when I hear "UG", part of the meaning that I
>> get is "universals are innate".
>> So I'm wondering whether the term "UG" presupposes the notion of
>> innateness, or whether it's neutral on that issue.
>I think this is an important, and interesting, question. Certainly,
>within the currently dominant paradigm (Chomskian generative grammar),
>the use of the term Universal Grammar presupposes a particular approach
>to the explanation of language universals: the explanation from
>innateness. Researchers working within the functional-typological
>approach, however, often reject the innateness argument, but still
>refer to Universal Grammar. A small number (at least I think its a
>small number!) of linguists are seeking a more unified approach to the
>explanation of language universals. These researchers also refer to
>Universal Grammar.
>The question is: are these UGs the same thing?
>I think the main difference is that the innatist argument suggests
>that there is an innate mechanism for acquisition which more or less
>_is_ UG. This UG is apparent from single language studies, language
>pathology research, research on acquisition, and -- to a limited
>extent -- from cross-linguistic studies. So, _given_ UG, we should
>expect to see similarities between languages, ie. 'universals' in the
>traditional sense.

I was waiting to see what came in before having my $2 worth. I'll
try to limit it to $0.02 for now...

There is a difference between universals, or better language universals,
and UG. Pike was talking about Language in relation to Behavioural
Universals, back in the 60s and earlier (see e.g. Ken Pike, Language in
relation to a Unified Theory of the Structure of Human Behavior, Mouton,
2nd edn?, 1967).

The universals NEED not bear any relationship with any formal definition
of grammar, and therefore use of the term UG is prejudging the issue,
and is INTRINSICALLY nativist as it presupposes that we share a
universal GRAMMAR, as opposed to a set of Cognitive Mechanisms (some of)
which MAY lie deeper. The term UG was introduced as an axiom (actually,
it developed out of some so-called simplifying assumptions, which
then grew to be treated as axiomatic).

In fact, more and more evidence is coming to light of deeper,
not intrinsically linguistic, mechanisms for universal linguistic
behaviour, including mechanisms which can be derived in information
theoretic terms, mechanisms which are employed in another sensory-motor
modalites, mechanisms relating to the development of an ontology which
extends beyond language, etc. These mechanisms can explain (as opposed
to merely describe) certain phonological and syntactic effects
(references available).

We still have a way to go to explain all of language and language
acquisition, but the nativist perspective is not universal, and the
universals need not take the form of grammar.

Dr David Powers (Assoc. Prof. / SIGART Editor)
Discipline of Computer Science UniOffice: +61-8-201-3663
The Flinders University of South Australia Secretary: +61-8-201-2662
GPO Box 2100, Adelaide South Australia 5001 Facsimile: +61-8-201-3626
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: UG

Date: Wed, 23 Feb 94 08:26:11 +0UG
From: Richard Hudson UCL <>
Subject: UG

Larry Selinker's query about whether Catalan is suitable for UG throws
an interesting new light on Joe Stemberger's query about the meaning of
UG. Joe assumes that UG must at least mean `to do with universals', but
Larry's query makes me wonder. After all, if Catalan turned out not to
be suitable for UG, then we find languages of the world dividing into two
groups, UG and non-UG, with claims about UG applying only to those languages
to which they apply. Would anyone working on UG count that as a serious

Dick Hudson
Dept of Phonetics and Linguistics,
University College London,
Gower Street,
London WC1E 6BT
(071) 387 7050 ext 3152
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue