<HTML dir=ltr><HEAD><TITLE>Re: Competence vs. Performance: Summary</TITLE>
<META http-equiv=Content-Type content="text/html; charset=unicode">
<META content="MSHTML 6.00.6000.16544" name=GENERATOR></HEAD>
<BODY>
<DIV id=idOWAReplyText59078 dir=ltr>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2>I would suggest that anyone who claims there is no competence/performance distinction in language is being terribly unfair to Miss Teen USA, South Carolina</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2></FONT> </DIV>
<DIV dir=ltr><A href="http://www.youtube.com/watch?v=lj3iNxZ8Dww">http://www.youtube.com/watch?v=lj3iNxZ8Dww</A></DIV>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2></FONT> </DIV></DIV>
<DIV id=idSignature86214 dir=ltr><PRE>Peter Gordon, Associate Professor
525 W 120th St. Box 180
Biobehavioral Sciences Department
Teachers College, Columbia University
New York, NY 10027
Office Phone: (212) 678-8162
FAX: (212) 678-8233
Web Page: www.tc.edu/faculty/index.htm?facid=pg328</PRE></DIV>
<DIV dir=ltr><BR>
<HR tabIndex=-1>
<FONT face=Tahoma size=2><B>From:</B> info-childes@mail.talkbank.org on behalf of Anat Ninio<BR><B>Sent:</B> Wed 10/17/2007 1:33 PM<BR><B>To:</B> Gary Marcus<BR><B>Cc:</B> info-childes@mail.talkbank.org<BR><B>Subject:</B> Re: Competence vs. Performance: Summary<BR></FONT><BR></DIV>
<DIV>
<P><FONT size=2>I'm sure one can come up with a well-articulated theory of psychology<BR>that won't scare us all with the dark shadow of behaviorism and still<BR>won't make an elementary mistake of identifying the rules of chess with<BR>a player's playing skill. This is what the competence/performance<BR>distinction is really about, isn't it? The pure rules versus the impure<BR>human behaviour?<BR><BR>Take note that even when a player is making a mistake, he is using --<BR>not terribly successfully -- his stored representation of the game's<BR>rules and his cumulated playing strategies. Why would you not see this<BR>as competence, I don't really understand. Stored competence can be<BR>partial, full of holes, context-dependent -- in fact, it is just (messy)<BR>potential for action, nothing more rule-like than that. I'm sure you'll<BR>agree to that, at the least.<BR><BR>Anat Ninio<BR><BR><BR>Gary Marcus wrote:<BR>> So far as I know, the only well-articulated theory of psychology in<BR>> which a competence-performance distinction does not arise is<BR>> behaviorism, and in that connection I still find Chomsky's 1959<BR>> arguments to be compelling. If we are to talk of internal<BR>> representations -- and I believe we must -- we must have a theory of<BR>> how those representations relate to behavior; I don't see how to do so<BR>> coherently without a competence-performance distinction.<BR>><BR>> Please note, by the way, that a commitment to distinguish competence<BR>> from performance per se does not commit one to any particular<BR>> theoretical apparatus; one need not, for example, be a nativist about<BR>> the rules of chess to think that a competence-performance distinction<BR>> applies -- and does useful work there (e.g., in distinguishing<BR>> "blunders" from more considered miscalculations).<BR>><BR>> When we discuss the merits of Chomsky's various proposals, it is<BR>> essential that we judge each of those hypotheses on its own merits.<BR>><BR>> -- gfm<BR>><BR>><BR>> On Oct 17, 2007, at 6:39 PM, Anat Ninio wrote:<BR>><BR>>> I disagree strongly, dear Gary, with your untested assumption that<BR>>> "in psychology, and in linguistics, we are continually left with the<BR>>> daunting challenge of inferring underlying representations from<BR>>> surface behavior". Just imagine for a minute that this rationalistic<BR>>> perception of psychology and linguistics is in effect incorrect and<BR>>> misleading -- as well as leading us to wasted decades of research on<BR>>> transformations, deep structures, and innate parameters. There are<BR>>> other ways to look at the organization of reality, and my only hope<BR>>> is that the next generation of theoreticians of both disciplines will<BR>>> not repeat the mistakes of the last one.<BR>>><BR>>> Anat Ninio<BR>>> Professor of Psychology<BR>>> The Hebrew University of Jerusalem<BR>>><BR>>><BR>>><BR>>> Gary Marcus wrote:<BR>>>><BR>>>> I couldn't disagree more strenuously with this last posting. To<BR>>>> ascribe both "spinned" and "spun" to competence is to miss the point<BR>>>> of the distinction, not to undermine it. "Spinned" and "spun" simply<BR>>>> aren't on a par; people accept the latter as correct and ascribe the<BR>>>> former to a mistake.<BR>>>><BR>>>> It makes perfect sense to collect data on both (I did just that in<BR>>>> 1992 SRCD Monograph on overregularizations, and I have cited<BR>>>> Stemberger's data in this connection), but no sense whatsoever to<BR>>>> treat the two as if they are on equal footing. And I say this not as<BR>>>> a matter of prescriptive linguistics (what the rules of language<BR>>>> "ought to be", according to some self-appointed grammarian) but as a<BR>>>> matter of descriptive psycholinguistics, as a characterization of<BR>>>> what speakers themselves believe.<BR>>>><BR>>>> When I inadvertently forget to carry a 1 and report that the sum of<BR>>>> 87 and 24 is 101, we need to able to distinguish my transient error<BR>>>> from my general understanding of what would constitute a correct sum<BR>>>> (viz. 111). A competence-performance distinction gives us a<BR>>>> theoretical tool with which to make that distinction; to say that<BR>>>> 101 and 111 are equally good answers ("since I have the 'competence'<BR>>>> to utter either number") would be nothing more than sophistry.<BR>>>><BR>>>> The competence-performance distinction does, as others have noted,<BR>>>> have any number of difficulties, but most are methodological ("how<BR>>>> can I tell which is which", a fact that will only reveal itself once<BR>>>> we have properly carved nature at it joints), not theoretical.<BR>>>><BR>>>> In psychology, and in linguistics, we are continually left with the<BR>>>> daunting challenge of inferring underlying representations from<BR>>>> surface behavior; there is no way we can conceivably succeed at that<BR>>>> task without a firm understanding of the fact that the mapping<BR>>>> between the two is often indirect: behavior is dictated not only by<BR>>>> underlying representations but by a host of other factors. Inasmuch<BR>>>> as the competence-performance distinction encapsulates that<BR>>>> fundamental truth, it is a valuable tool that we cannot afford to<BR>>>> dismiss.<BR>>>><BR>>>> -- Gary Marcus<BR>>>> Professor of Psychology<BR>>>> New York University<BR>>>><BR>>>><BR>>>><BR>>>> On Oct 17, 2007, at 1:26 PM, Anat Ninio wrote:<BR>>>><BR>>>>> Hello Everybody,<BR>>>>><BR>>>>> I have a feeling -- following Joe Stemberger's excellent<BR>>>>> contributions to this strand -- that we're all a little scared<BR>>>>> stating the obvious, which is that the notion of competence as a<BR>>>>> separate entity from performance is a philosophical error, pure and<BR>>>>> simple. If under time stress people say in 4% of the time "spinned"<BR>>>>> instead of "spun", then they possess a competence to do so.<BR>>>>> Competence is -- in plain English -- an a priori stored potential<BR>>>>> or ability to produce some behaviour, and any other treatment of<BR>>>>> it, by Chomsky or whoever, is at best a mystification of the obvious.<BR>>>>><BR>>>>> I agree with Joe that we should simply proceed with doing research<BR>>>>> and collect information on what people actually say, whether<BR>>>>> children or adults, and don't heed the voices saying that what we<BR>>>>> hear is "merely" this or "merely" that, when any philosopher major<BR>>>>> in their first year would tell us how confused the whole idea of<BR>>>>> competence/performance distinction is.<BR>>>>><BR>>>>> The best,<BR>>>>><BR>>>>> Anat Ninio<BR>>>>><BR>>>>><BR>>>>><BR>>>>><BR>>>>><BR>>>>> Joe Stemberger wrote:<BR>>>>>> Hello, everyone.<BR>>>>>><BR>>>>>> Just to start, I'd like to mention that I've spent all of my<BR>>>>>> career working on speech errors in adults (both spontaneous speech<BR>>>>>> and experimental tasks), and so I accept the notion that there are<BR>>>>>> outputs that are "correct" but that performance factors can lead<BR>>>>>> to "incorrect" outputs. But that doesn't mean that "the"<BR>>>>>> competence-performance distinction gets us anywhere practical.<BR>>>>>><BR>>>>>> Defenses of "the" competence-performance distinction are missing<BR>>>>>> two main points:<BR>>>>>> (1) that the exact division between what is competence and what is<BR>>>>>> performance, as well as the criteria that distinguish them, are<BR>>>>>> largely unknown after more than 40 years.<BR>>>>>> (2) There is no clear way to test competence except through<BR>>>>>> performance.<BR>>>>>><BR>>>>>> Consider John Limber's quote:<BR>>>>>> "Inferences about linguistic competence in children are typically<BR>>>>>> based on spontaneous speech.<BR>>>>>> This poses a problem since we know that other factors are also<BR>>>>>> involved in speech production."<BR>>>>>><BR>>>>>> Well, yes, there are performance factors in language production.<BR>>>>>> Just as there are performance factors in language perception,<BR>>>>>> language comprehension, grammaticality judgments, etc.<BR>>>>>> In the real world, we can observe only performance, and all acts<BR>>>>>> of any sort are "contaminated" by performance.<BR>>>>>> And it follows from that that inferences about competence must<BR>>>>>> come from performance, only and always.<BR>>>>>><BR>>>>>><BR>>>>>> OK, so now let's get concrete.<BR>>>>>> Observe a native speaker of English of any age for long enough,<BR>>>>>> and you will find them saying *SPINNED instead of SPUN.<BR>>>>>> Ask them about it, and they'll tell you that it was an error.<BR>>>>>> In typical speeded experimental situations, undergraduates produce<BR>>>>>> such errors about 4% of the time in neutral contexts.<BR>>>>>> (And in an experiment that I'm just finishing up, where they<BR>>>>>> produce coordinated verbs, the error GRINNED AND *SPINNED (with<BR>>>>>> rhyming regular in the first word) shoots up to about 25% of tokens.)<BR>>>>>> Young children produce such outputs from an early age, but every<BR>>>>>> study has shown that such overregularizations are in the minority<BR>>>>>> for most children (and most irregular verbs) from the beginning,<BR>>>>>> and that, like in adult speech, the frequency of the verb is one<BR>>>>>> of the predictors of error rate.<BR>>>>>> This is a generally viewed as a performance phenomenon at all ages.<BR>>>>>> But yet it is cited all the time as a nice deomstration that<BR>>>>>> patterns are being extracted and generalized.<BR>>>>>><BR>>>>>> At the same time, speakers will also occasionally produce, instead<BR>>>>>> of SPUN, *SPAN. In experimental situations, such errors are easy<BR>>>>>> to come by.<BR>>>>>> As well as things like *GRUN instead of regular GRINNED. (In that<BR>>>>>> experiment that I'm just finishing up, *GRUN occurs about 2% of<BR>>>>>> the time in neutral contexts,<BR>>>>>> rising to about 6% in SPUN AND *GRUN (with a rhyming irregular in<BR>>>>>> the first word).)<BR>>>>>> Bybee & Moder 1983 showed that irregular patterns generalize to<BR>>>>>> novel words at much greater rates.<BR>>>>>> So irregular patterns also show generalization.<BR>>>>>><BR>>>>>> Consider another of John's quotes:<BR>>>>>> "Does anyone really doubt that the language one observes is but a<BR>>>>>> subset of the language one might observe under such and such<BR>>>>>> conditions?"<BR>>>>>><BR>>>>>> So, if we MUST conclude that competence extracts and encapsulates<BR>>>>>> explicit procedures (such as rules) to occount for generalization,<BR>>>>>> then competence contains such procedures for creating both regular<BR>>>>>> and irregular forms, because both types of patterns<BR>>>>>> overgeneralize, right?<BR>>>>>><BR>>>>>> Steve Pinker, Harald Clausen, and others taking a similar<BR>>>>>> approach, have argued that the overgeneralization of the -/ed/<BR>>>>>> pattern involves the use of a rule after failure to access an<BR>>>>>> irregular form, but that overregularization of irregular patterns<BR>>>>>> is a performance phenomenon, based on the way that irregular forms<BR>>>>>> are stored in and accessed from the lexicon.<BR>>>>>> Which leads to the conclusion that generalization of patterns can<BR>>>>>> occur for performance reasons, even in approaches that accept<BR>>>>>> "the" competence-performance distinction.<BR>>>>>><BR>>>>>> Which leads to this possibility:<BR>>>>>> all inflected forms that a speaker has been been exposed to are<BR>>>>>> simply stored in the lexicon.<BR>>>>>> All generalization, even of regular patterns, is a performance<BR>>>>>> phenomenon.<BR>>>>>> And while it's true that the -/ED/ pattern overgeneralizes more<BR>>>>>> than any other pattern, we could attribute /that/ difference to<BR>>>>>> performance factors, right?<BR>>>>>><BR>>>>>> Or even<BR>>>>>> The purpose of a grammar is to enumerate the sentences of the<BR>>>>>> language.<BR>>>>>> Actual grammars consist of storing every sentence that has ever<BR>>>>>> been observed.<BR>>>>>> Generalization is just a performance phenomenon across stored<BR>>>>>> exemplars.<BR>>>>>> You want to know in detail how generalization occurs?<BR>>>>>> Hey, that's a performance phenomenon. It lies outside the proper<BR>>>>>> bounds of linguistic theory, and so it's not my responsibility to<BR>>>>>> show how it works.<BR>>>>>><BR>>>>>> And what I personally would want to see is some formal proof that<BR>>>>>> "the" competence-performance distinction couldn't lead us to that<BR>>>>>> sort of system just as easily as it has led us anywhere else.<BR>>>>>><BR>>>>>> >From a practical perspective, it seems to me that, since we have<BR>>>>>> to work with performance data anyway, we want to work with as wide<BR>>>>>> a range of types of data as possible.<BR>>>>>> And we want to develop theories that account for all of those<BR>>>>>> data, in detail.<BR>>>>>> If there's a distinction between competence and performance,<BR>>>>>> that's fine, but it has to be explicit, and our theories need to<BR>>>>>> explain exactly which phenomena are due to competence, which are<BR>>>>>> due to performance, and why. And if there are aspects of data that<BR>>>>>> aren't accounted for in detail, it means that we should be<BR>>>>>> uncomfortable, because we need to account for it all.<BR>>>>>><BR>>>>>> That isn't the way that "the" competence-performance distinction<BR>>>>>> has been used in the past.<BR>>>>>> If it had been, that would've been fine, as far as I'm concerned.<BR>>>>>> While some theoretical linguists feel it's fine to apply<BR>>>>>> linguistic theory to child language, few are open to using child<BR>>>>>> language to provide the tie-breaker for choosing between two<BR>>>>>> theoretical mechanisms.<BR>>>>>><BR>>>>>> In the meantime, I'll happily go on studying performance and<BR>>>>>> working on theories of how human language works, including claims<BR>>>>>> about the fundamental orgainization of the language system.<BR>>>>>><BR>>>>>> As should we all.<BR>>>>>><BR>>>>>><BR>>>>>> ---Joe Stemberger<BR>>>>>> Linguistics<BR>>>>>> UBC<BR>>>>>><BR>>>>>><BR>>>>>> john limber pravi:<BR>>>>>>><BR>>>>>>> On 10/16/07 6:46 AM, "Matthew Saxton" <M.Saxton@ioe.ac.uk><BR>>>>>>> <<A href="mailto:M.Saxton@ioe.ac.uk">mailto:M.Saxton@ioe.ac.uk</A>> wrote:<BR>>>>>>><BR>>>>>>> “No-one has stepped forward to defend the competence-performance<BR>>>>>>> distinction, or even to offer supportive references.”<BR>>>>>>><BR>>>>>>> OK— try this—with references too!<BR>>>>>>><BR>>>>>>> The basic idea is so widespread in science that it hardly needs<BR>>>>>>> defense— frictionless bodies, stimulus generalization, latent<BR>>>>>>> learning, stereotype bias.... and linguistic competence all are<BR>>>>>>> more or less scientific concepts designed to variously explain<BR>>>>>>> conditional performance.<BR>>>>>>><BR>>>>>>> Does anyone really doubt that the language one observes is but a<BR>>>>>>> subset of the language one might observe under such and such<BR>>>>>>> conditions? And that much of that observed language is<BR>>>>>>> fragmentary and ill-formed? While every case demands its own<BR>>>>>>> explanatory story, to toss out the whole idea of competence or<BR>>>>>>> similar concepts sounds like a lame return to behaviorism.<BR>>>>>>><BR>>>>>>> Years ago I spent a lot of time on the development of complex<BR>>>>>>> sentences (Limber, 1973). There was one gap in the thousands of<BR>>>>>>> two to three year old children’s utterances I observed -- a lack<BR>>>>>>> of relative clauses attached to subject NPs.<BR>>>>>>><BR>>>>>>> Did this mean these kids didn’t have the potential for those<BR>>>>>>> structures in their behavioral repertoire—their linguistic<BR>>>>>>> competence? Here’s the abstract of my answer (Limber, 1976)—which<BR>>>>>>> curiously in connection with this current discussion, involves<BR>>>>>>> pragmatics.<BR>>>>>>><BR>>>>>>> “Inferences about linguistic competence in children are typically<BR>>>>>>> based on spontaneous speech.<BR>>>>>>> This poses a problem since we know that other factors are also<BR>>>>>>> involved in speech production.<BR>>>>>>> Children who may use complex object and adverbial NPs do not use<BR>>>>>>> complex subject NPs. Is<BR>>>>>>> this a competence deficit, a performance problem, or simply a<BR>>>>>>> reflection of pragmatic factors?<BR>>>>>>> Evidence presented here suggests that children probably do not<BR>>>>>>> need complex subjects. An<BR>>>>>>> extensive use of pronouns in subject but not object position<BR>>>>>>> indicates that pragmatics may<BR>>>>>>> account for the distribution of clauses in their speech. A<BR>>>>>>> similar pattern in adult speech indicates there is no warrant to<BR>>>>>>> conclude children's lack of subject clauses reflects anything<BR>>>>>>> more than the nature of spontaneous speech.”<BR>>>>>>><BR>>>>>>><BR>>>>>>> In fact, in all my data of several thousand utterances of<BR>>>>>>> children and adults, only TWO subject NPs showed up—one shaky<BR>>>>>>> example from a three-year old and another from an adult. The<BR>>>>>>> probability that a child is exposed to a subject NP is, from my<BR>>>>>>> data, less than 1/1000. Here are the two:<BR>>>>>>><BR>>>>>>> Adult: “ Well these buses that I've had today have been really<BR>>>>>>> weird.”<BR>>>>>>><BR>>>>>>> Child: “I think that the girl ... that's here ... doesn't ... she<BR>>>>>>> doesn't want me to open it. “<BR>>>>>>><BR>>>>>>> I wonder how the Nuevo-Statistical approaches to language<BR>>>>>>> acquisition would handle this?<BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>>>> Limber, J. (1973). The genesis of complex sentences. In T. Moore<BR>>>>>>> (Ed.), Cognitive Development and the Acquisition of Language (pp.<BR>>>>>>> 169-186). New York: Academic Press.<BR>>>>>>> <A href="http://pubpages.unh.edu/~jel/JLimber/Genesis_complex_sentences.pdf">http://pubpages.unh.edu/~jel/JLimber/Genesis_complex_sentences.pdf</A><BR>>>>>>> <<A href="http://pubpages.unh.edu/~jel/JLimber/Genesis_complex_sentences.pdf">http://pubpages.unh.edu/%7Ejel/JLimber/Genesis_complex_sentences.pdf</A>><BR>>>>>>><BR>>>>>>><BR>>>>>>> Limber, J. (1976). Unravelling competence, performance, and<BR>>>>>>> pragmatics in the speech of young children. Journal of Child<BR>>>>>>> Language, 3, 309-318.<BR>>>>>>> <A href="http://pubpages.unh.edu/~jel/JLimber/pragmatics_performance.pdf">http://pubpages.unh.edu/~jel/JLimber/pragmatics_performance.pdf</A><BR>>>>>>> <<A href="http://pubpages.unh.edu/~jel/JLimber/pragmatics_performance.pdf">http://pubpages.unh.edu/%7Ejel/JLimber/pragmatics_performance.pdf</A>><BR>>>>>>><BR>>>>>>> John Limber<BR>>>>>>> University of New Hampshire<BR>>>>>>> Durham NH<BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>>>> I think that the girl ... that's here ...<BR>>>>>>> doesn't ... she doesn't want me to open it.<BR>>>>>>><BR>>>>>>> Well these<BR>>>>>>> buses that I've had today have been really weird.<BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>>>><BR>>>>><BR>>>>><BR>>>><BR>>>><BR>>>><BR>>><BR>><BR>><BR><BR><BR></FONT></P></DIV></BODY></HTML>