Bob,

At the risk of taking us down a rabbit hole, the Chomskyan notion of competence is a puzzling one.  Not only is its content subject to the vagaries of theory development, but it's not clear what function the ability to judge well-formedness serves, beyond syntactic research.  Linguists differ in their judgments, and non-linguists often find themselves baffled by how linguists make these judgments.  We clearly don't use this ability in communication, because we are constantly interpreting utterances that are not well-formed.  With apologies to McCawley, a sentence like "I think over easy." is not well-formed, but in answer to the question "How does Herb like his eggs?" it turns out to be a straightforward ellipsis, just not one that can happen at the sentence level and therefore not subject to sentence-level syntactic description.  

Another problem with competence is that it can only be defined circularly.  Competence is, in part, the native speaker's ability to distinguish well-formed from ill-formed sentences.  This ability defines the set of sentences the grammar must generate.  And we know that a sentence belongs to this set if we judge it well-formed.  Competence as a concept doesn't gain us anything. 
The fact that the skill is not binary, that is, judgments aren't simply well-formed vs. ill-formed, means that the set is necessarily fuzzy, another problem for generative grammars, which do not define fuzzy sets.

Our ability to recognize well-formed and ill-formed sentences requires practice and critiquing.  Many of our undergraduate students have a difficult time determining grammaticality out of context because they can't imagine a context in which the sentence would be grammatical.  Since they can't make sense of the sentence they call it ungrammatical.  They haven't yet developed the ability to abstract structure from meaning.

I don't see a use for competence beyond justifying the claim of generative grammar that the domain of syntactic description is the sentence.  I don't have any problem with making the sentence the domain.  There's a lot of important, formal information that we have to know about sentences.  I agree with Givon that we can't do without formal syntactic description.  But sentence-level syntax isn't enough.  When I close this message, I'll go back to reading a dissertation draft by one of our students who's working on the grammar of discourse particles in Taiwanese.  She demonstrates very nicely that this rich array of particles in Taiwanese can't be described at the sentence level.  Even their placement in the sentence can't be predicted from sentence-level syntax.  Rather, conversational data is essential to understanding how they work and to predicting their occurrence.  About the only syntactic statement you can make about them is that they reflect a head-modifier structure, the particles serving as modifiers, but that's a functional statement as well as a formal one.

I don't think I've reached the bottom of the hole.

Herb Stahlke


<<< [log in to unmask] 11/ 6 11:10a >>>

Herb has mixed an objection to competence-performance with why it has any
usefulness
for understanding what it means to know a language.  I don't know of any theory of
language which attempts it has to account for false starts, lapses in memory, typos
in writing posts to
listserv discussions.  In this sense, Herb is right that the competence-performance
distinction does limit what data a theory of language has to account for.

However, in any number of domains, we make the distinction between what people know
(declarative knowledge) and what people actually do with that knowledge (procedural
knowledge).  Thus, one can talk about knowing the rules of chess (declarative
knowledge) and knowing how to use those rules for winning (procedural knowledge).

When it comes to knowledge about language, this is a very useful distinction which,
as I observed, some of the ATEG grammar teaching tips exploits.


> As theories have developed, the membership of the sets of
> competence and performance have changed.

Of course, this is right, but all domains of inquiry change what kinds of data can
or can not be accounted for.  This is not unique to the study of language.

Bob Yates, Central Missouri State University

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/

                        

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/