tag:blogger.com,1999:blog-5275657281509261156.post2640593462883766796..comments2024-03-28T04:04:55.806-07:00Comments on Faculty of Language: Guest Post: Jeffrey Watumull on Postal's Critique of BiolinguisticsNorberthttp://www.blogger.com/profile/15701059232144474269noreply@blogger.comBlogger64125tag:blogger.com,1999:blog-5275657281509261156.post-30086804204926788722013-04-04T06:53:41.299-07:002013-04-04T06:53:41.299-07:00This comment has been removed by a blog administrator.Anonymoushttps://www.blogger.com/profile/01506929629516149388noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-59177567300260479212013-04-02T17:11:29.306-07:002013-04-02T17:11:29.306-07:00I have just posted a paper on LingBuzz http://lin...I have just posted a paper on LingBuzz http://ling.auf.net/lingbuzz/001765<br /><br />dealing in some more detail with the issues raised by Watumull. If you read it you will see why i think that he has "resolved" the incoherence indeed - by essentially eliminating all biology from biolinguistics. And of course much of the debate here on the blog also has focussed on computational issues NOT on biological wet-ware...Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-71531013666151806942013-04-02T03:51:32.269-07:002013-04-02T03:51:32.269-07:00But, if the main aim is to understand language its...But, if the main aim is to understand language itself, with the mathematical investigations only a means to that end, that would be non-Platonistic. & even somebody like me is 'practicing Platonism', in accordance with what I'm saying, when they're trying to figure out if their analysis actually produces a given sentence with appropriate interpretation or not.AveryAndrewshttps://www.blogger.com/profile/17701162517596420514noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-38419880601062047512013-04-02T01:34:29.860-07:002013-04-02T01:34:29.860-07:00@Alex: I think I find mathematical Platonism less ...@Alex: I think I find mathematical Platonism less crazy than the other kinds, perhaps even plausible. At any rate, studying the mathematical properties of linguistic formalisms seems much closer to a Platonistic enterprise than what I do.AveryAndrewshttps://www.blogger.com/profile/17701162517596420514noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-39369102026643913422013-04-02T00:21:28.137-07:002013-04-02T00:21:28.137-07:00oops yes eric->edward. I will suggest as pathe...oops yes eric->edward. I will suggest as pathetic excuse the existence of a nephew called 'eric' ...<br /><br />A typical 'linguistics' question would be how well/badly a grammar fits the apparent data of a language, or whether a framework/theory can give good accounts of languages with the observed range of typological variation; a typical 'mathematical linguistics' question would be whether some formalism is mildly context sensitive, or, if not, usefully less than full context-sensitive.<br /><br />As for i-language, one possible place to get confused in KLT:57 mid, where the child is said to come to 'know an i-language', which certainly invites interpretation of i-language is something other than a brain state (which would be more naturally described as `acquired', I think. Then KLT:249-250 'two grammars of the I-language abstracted from this state'. On KLT:50 we read about a shift to mental representation *and computation* as part of the shift from E- to I-language, but without an (putatively infinite) structure set in the picture, what is there to compute? And if there is such a set in the picture, what is its name? I think that's enough to demonstrate a potential for confusion. (KLT = Knowledge of Language, Chomsky 1986).AveryAndrewshttps://www.blogger.com/profile/17701162517596420514noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-28229319300365607522013-04-02T00:00:52.799-07:002013-04-02T00:00:52.799-07:00I consider what I do to be theoretical linguistics...I consider what I do to be theoretical linguistics which bears the same relationship to linguistis that theoretical physics does to physics: given that it is theoretical it has to be mathematical otherwise it descends into idle pseudo-scientific speculation. The fact that it is mathematical doesn't mean it is Platonic: look at every branch of functioning science, which is in general highly mathematicized.<br /><br />If you are a Platonist then you don't care about acquisition at all, whereas for me that is the central problem. <br /><br />(Parenthetical aside: Indeed one of the points of the Postal critic, and many others over the years (.e.g Devitt etc) is that for all of the rhetoric of Chomskyan linguistics, which reaches new heights in biolinguistics, they have insulated themselves from all of the nonlinguistic data, which does make it look like Platonism.)Alex Clarkhttps://www.blogger.com/profile/04634767958690153584noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-88427456833893012742013-04-01T20:51:33.805-07:002013-04-01T20:51:33.805-07:00Avery writes: "there is a platonic subject as...Avery writes: "there is a platonic subject associated with linguistics, it's mathematical linguistics, as practiced by Eric Stabler, Alex Clarke and many others (but it's not what I and other descriptive/theoretical linguists do)"<br /><br />What is the real distinction here between mathematical linguistics and descriptive/theoretical linguistics? Is it really a difference in what's being studied (a platonic thing versus a non-platonic thing)? I would have thought that Ed Stabler and Alex Clark take themselves to be studying linguistics as a cognitive phenomenon in more or less the way Chomsky introduced it, though they may disagree with Chomsky on many details. To me the differences are in the methods, not the goals or object of study. Otherwise, are we going to conclude that Chomsky's early work on the Chomsky hierarchy was the study of a platonic linguistics?<br /><br />Of course you <b>can</b> also use mathematical methods to study "platonic linguistics", just as you can use the "normal" tools of descriptive/theoretical linguistics to. But I think the platonic/internalist issue is just orthogonal to the mathematical/non-mathematical one.<br /><br />Also: "distinguish between i-language as a description of the brain-state responsible for certain aspects of linguistic performance ... and i-language as the infinitary set produced by this state"<br /><br />I would have thought that I-language is clearly intended to be the former. Is there something Chomsky has said that suggests taking I-language to be a set of expressions? That seems to be almost exactly what he intends I-language <b>not</b> to be (i.e. almost E-language).<br />Tim Hunterhttps://www.blogger.com/profile/11810503425508055407noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-24855015580541509672013-04-01T19:12:40.524-07:002013-04-01T19:12:40.524-07:00"Eric Stabler" --- am I right to assume ..."Eric Stabler" --- am I right to assume you mean Ed(ward) Stabler? (smile)benjamin.boerschingerhttps://www.blogger.com/profile/00894608988488218285noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-10771140494185747802013-04-01T19:03:15.470-07:002013-04-01T19:03:15.470-07:00I have always thought that the solution to incoher...I have always thought that the solution to incoherence issue was to treat the platonic objects as a classification scheme for the real-world things we're studying; the sentence structures are classifications for linguistic performances, the grammars for the brain states (or states of whatever alternative organ one might wish to suggest) that are responsible for these performances. Then there is a platonic subject associated with linguistics, it's mathematical linguistics, as practiced by Eric Stabler, Alex Clarke and many others (but it's not what I and other descriptive/theoretical linguists do).<br /><br />Another contributor to the incoherence issue is Chomsky's systematic unwillingness or inability to distinguish between i-language as a description of the brain-state responsible for certain aspects of linguistic performance (the ones that he, me and Norbert are primarily but not exclusively interested in), in which case it can in a sense be a 'part of the brain' (more precisely, a description of an aspect of the brain, which may or may not correspond to a 'part' in the usual understanding of the term), and i-language as the infinitary set produced by this state, in which case it can't be part of the brain in any sense whatsoever (but could be described as an aspect of brain function).<br /><br />I hope that this or some similar formulation offers a way out of this mess, because I think it is completely unrealistic to expect Chomsky to say anything much different or clearer than what he's already said, as desireable as this might seem to be. (If I've forgotten about or never noticed some important clarification by Chomksy of what i-language is supposed to be, I'm sure somebody will point this out.)AveryAndrewshttps://www.blogger.com/profile/17701162517596420514noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-10032404618197608152013-04-01T06:16:52.516-07:002013-04-01T06:16:52.516-07:00Alas, I am unable to see that incoherence. Making ...Alas, I am unable to see that incoherence. Making of mathematical models, as in (ii), is routine practice in the natural sciences. Your supposed ontological incoherence would apply not only to biolinguistics but would demolish virtually all of physics -- to any use of infinitary mathematics (real numbers, Lie groups, fractals, ...) in the physical world. I doubt that is the intent.<br />Anonymoushttps://www.blogger.com/profile/11869582156654770141noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-40434107862819245912013-03-28T16:33:07.535-07:002013-03-28T16:33:07.535-07:00If you like to chase platonic unicorns that is of ...If you like to chase platonic unicorns that is of course entirely up to you but please do not expect i join you. The only sentence of your post that merits a reply is:<br /><br />I mention this line of thinking, ONLY because the platonist argument seems to be: biolinguistics is so incoherent the only reasonable alternative is platonism: language as formal, abstract object.<br /><br />It is extremely puzzling to me why you would think this. Postal holds Chomsky’s ontology to be incoherent because Chomsky assumes language to be both: (i) part of the human brain (that is a finite, physical object) and (ii) based on set-theoretic objects generated by the operation Merge (e.g., “a system of discrete infinity consisting of hierarchically organized objects” (Chomsky, 2008, p. 137)). Because (i) and (ii) cannot apply simultaneously to the same object (e.g. an I-language) Chomsky’s ontology is internally incoherent. This incoherence arises entirely independently of whether a critic defends linguistic Platonism as Postal does or linguistic naturalism as I do. In case it still is unclear read: http://ling.auf.net/lingbuzz/001573 where this point is explained in more detail.Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-87388185356328845122013-03-28T13:58:06.698-07:002013-03-28T13:58:06.698-07:00"After 45 years of biolinguistic research sur..."After 45 years of biolinguistic research surely we know something more about the innate structure?"<br /><br />Surely we do; do you need pointers to that literature?<br /><br />"You say: "We're rehashing the "psychological reality" debates in linguistics back in the 70s, 80s, with nothing new." There are really only 2 possibilities: either [i] LF is a biological 'object' part of the human brain or [ii] it is not. Since Chomsky has repeated [i] over and over again and since he also has rejected [ii] there is no debate. If you assume [ii], then maybe you are some kind of Platonist?"<br /><br />I think [i] is a good working hypothesis for scientists. Like all hypotheses, subject to verification, revision, refinement, etc., but not incoherent as stated.<br /><br />"As for why some physicists reject string theory I suggest you ask them. It is not clear to me why you seem to think that a LINGUISTIC Platonist like Postal should have a theory about strings. Do you expect your dentist to know how to fix your car?"<br /><br />What Paul Postal thinks is not really the issue here. <br /><br />The issue is that the platonist arguments for language as a formal, abstract object work just as well (poorly) for all sorts of other things, such as: vibrating strings in 11-dimensional spaces, every grammatical formalism ever conceived of, maybe even unicorns.<br /><br />Perhaps that is what platonists intend: that the constructs posited by discredited theories are just as real as those of currently held theories.<br /><br />If so, I don't understand how platonists square these criteria for realism, with the scientific method as usually carried out. The same problem does not arise in the biolinguistics programme. I mention this line of thinking, ONLY because the platonist argument seems to be: biolinguistics is so incoherent the only reasonable alternative is platonism: language as formal, abstract object.<br />Anonymoushttps://www.blogger.com/profile/11869582156654770141noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-65504741143866865562013-03-26T00:32:17.348-07:002013-03-26T00:32:17.348-07:00"So if you can enlighten me where Chomsky'..."So if you can enlighten me where Chomsky's scientific model of the language faculty in the human brai[n] is hidden I'd be much obliged."<br /><br />I think you are making a category mistake here; a scientific model is not "in the human brain", anymore than String theory is "in the universe". <br /><br />That was not a category but a grammar mistake. [You may have noticed that I asked Alex for clarification when he made a similar mistake and did not assume he really meant something stupid.] What I meant to ask is where the language faculty, stipulated by Chomsky's model, is located. The LF cannot be just Merge and as far as I know Chomsky has from early on hypothesized that we need to “formulate a hypothesis about innate structure that is rich enough to meet the condition of empirical adequacy” (Chomsky, 1967, p3). After 45 years of biolinguistic research surely we know something more about the innate structure?<br /><br />You say: "We're rehashing the "psychological reality" debates in linguistics back in the 70s, 80s, with nothing new." There are really only 2 possibilities: either [i] LF is a biological 'object' part of the human brain or [ii] it is not. Since Chomsky has repeated [i] over and over again and since he also has rejected [ii] there is no debate. If you assume [ii], then maybe you are some kind of Platonist?<br /><br />As for why some physicists reject string theory I suggest you ask them. It is not clear to me why you seem to think that a LINGUISTIC Platonist like Postal should have a theory about strings. Do you expect your dentist to know how to fix your car?Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-30453129899624722662013-03-25T13:04:46.748-07:002013-03-25T13:04:46.748-07:00CB quoting Chomsky: "talk about Merge and set...CB quoting Chomsky: "talk about Merge and sets is 'something metaphorical, and the metaphor has to be spelled out someday'"<br /><br />It seems to me this is standard operating procedure in any science, Chomsky is simply being more honest about it than some.<br /><br />How is formulating Merge and its set-theoretic properties as (part of) a model of the human language faculty any different than physicists positing String theory (more precisely M-theory) as (part of) a model of the universe?<br /><br />Both String theory and Merge are metaphors, idealized, abstract models.<br />No one has seen strings: 1-dimensional vibrating objects embedded in an 11-dimensional space to explain all of physics seems pretty far-fetched; must such work be dismissed for not being PHYSICAL enough?<br />What does it mean that Merge isn't BIOLOGICAL enough?<br />Whether String theory is physical enough, whether Merge is biological enough, are empirical matters: do they contribute to understanding the data, to good explanations, further work, etc.<br />Do platonists see these cases as different?<br /><br />"So if you can enlighten me where Chomsky's scientific model of the language faculty in the human brai[n] is hidden I'd be much obliged."<br /><br />I think you are making a category mistake here; a scientific model is not "in the human brain", anymore than String theory is "in the universe". <br /><br />We're rehashing the "psychological reality" debates in linguistics back in the 70s, 80s, with nothing new. <br />I suggest critiquing Rules and Representations, rather than the interview book, if you are really interested in this.<br /><br />Sidebar for any Platonists:<br />Do strings exist as abstract objects?<br />Would strings still "exist" if String Theory is abandoned?<br />If languages exist as abstract objects, does not Merge with its properties "exist" also, whether or not shown to be not a useful construct for explaining the language faculty? (Similarly ALL grammatical theories ever proposed)<br />Anonymoushttps://www.blogger.com/profile/11869582156654770141noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-39913143384923834962013-03-24T15:57:46.753-07:002013-03-24T15:57:46.753-07:00Much more serious, to me, are the problems that fl...Much more serious, to me, are the problems that flow from the severely undersupported assumption that Merge is binary. That headed, binary structure play a significant role in syntax seems to be true, but this falls massively short of being a case that they are all that is in play there.AveryAndrewshttps://www.blogger.com/profile/17701162517596420514noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-12083564138218065722013-03-24T15:03:46.764-07:002013-03-24T15:03:46.764-07:00I'm not worried about the 'category mistak...I'm not worried about the 'category mistake' because I take it to be a way of saying that the brain does something that can be usefully modelled as the formation of a set. Perhaps better modelled as the formation of something other than a set (hence C's disquisition of us not having sets in our heads). Scare quotes because I'm not convinced that it's any more of a category mistake than claiming that I ate breakfast sitting on a chair, as opposed to on a region of mostly empty space with some electrically charged stuff in it that can be usefully described as a chair.AveryAndrewshttps://www.blogger.com/profile/17701162517596420514noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-5458819520285580242013-03-24T13:53:26.236-07:002013-03-24T13:53:26.236-07:00Thank you for the clarification. Let me add a coup...Thank you for the clarification. Let me add a couple of clarifications as well<br /><br />1. You misunderstood my reference to Katz. It came up as response to a very specific point: having a computer [ex hypothesis we know its inner workings] calculating the F-numbers based on a program that was also known to us. You brought up Kripkensteinian worries when i said the computer does not run the program itself but a physical token. - only for this VERY LIMITED case did i claim Katz had a solution [and it might very well be he had it in more detail in one of his earlier books]. I would never claim Katz attempted to naturalize intentionality.<br /><br />2. i appreciate your pragmatic reasons. However, I would be very cautious about an implicit premise you seem to assume as unproblematic. You say you want an "insightful theory about some smaller aspect, say, our ability to produce and understand sentences (and our ability to acquire this ability)" that does not require to account for broader aspects of social cognition you worry [probably rightly] might be difficult if not impossibly accounted for, Now in order for this approach to work it must be [at least in principle] possible to isolate 'language' from 'general social cognition'. From my perspective the only way such isolation could work is if Postal is right and language is a Platonic object. Then we would have some reason to believe that whatever it is that allows us to acquire and use language is sufficiently distinct from 'social cognition' so that we can study it in isolation. <br /><br />But if you are not a Platonist about language and assume it is a biological organ [in whatever sense you want to use the term] i see no good reason for assuming that we CAN study language in isolation. Where would you draw a line between 'language' and 'social cognition'? Do you believe it is possible to give a full account of language acquisition that explains NOTHING about linguistic creativity? In other words do you think a child could acquire and hence understand English but not be able to use English 'creatively'? It should be possible if we can account for one without accounting for the other. <br /><br />To repeat: i agree at the moment no one has an account for translating talk about mental representations into talk about brains. i also doubt that the kind of 'eliminativism' of Churchland works. But i am worried about how little emphasis the biolinguists put on biology. Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-10474530012390702332013-03-24T13:32:07.781-07:002013-03-24T13:32:07.781-07:00You are assuming a Platonist view of language. Fro...You are assuming a Platonist view of language. From a cognitive view of language, the lexicon is just what some individual knows about the words in his or her native tongue. I speak English, but I don't know (or didn't anyway) that Frosch is a word in German. Frosch is not represented anywhere in my brain; there is no token of it, until I encounter it. So I certainly don't accept that there is a token of every word in every possible language, in my brain. That would be ridiculous, I agree.Alex Clarkhttps://www.blogger.com/profile/04634767958690153584noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-5124388067255086452013-03-24T12:46:11.461-07:002013-03-24T12:46:11.461-07:00Just a very brief reply; Postal is not objecting t...Just a very brief reply; Postal is not objecting to views that distinguish between language and knowledge of language [though he may object to specific proposals of some of those views]. <br /><br />The incoherence problem arises for a view like Chomsky's that claims there is no difference between language and knowledge of language and ALSO insists that language is set theoretic. Such a view could not even be salvaged by your 100,000 word proposal because we have no way to limit the lexicon for any language. A trivial example is the following sentence:<br /><br />[1] The German saying 'Sei kein Frosch' is translated into English literally 'Don't be a frog' but it means 'Don't be a spoilsport'.<br /><br />We probably agree that [1] is a sentence of English even though it contains a German clause. If we accept that we need to add all of German [and via parallel arguments every other language] to your 100,000 English words [because there is no a priori limit to words that can be used in those kinds of sentences]. And this is just one of several similar arguments [e.g., insertion of nonsense words, animal sounds and a host of other possibilities that one would need to account for] made by Postal [2004] also cited by Jackendoff & Culicover [2005]. So for Chomsky it would be impossible to accommodate such an extensive token collection and even for someone who is willing to move tokens out of the brain it may be awfully difficult if not impossible to set any size limit for 'amount of words'Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-63370607187793957192013-03-24T11:27:44.756-07:002013-03-24T11:27:44.756-07:00This comment has been removed by the author.Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-86754337487422310302013-03-24T11:22:08.836-07:002013-03-24T11:22:08.836-07:00One thing you asked that I didn't reply -- you...One thing you asked that I didn't reply -- you asked where the tokens were. So let's say that they are stored in the left frontal cortex (for the sake of argument). Given that there are only (say) 100,000 words and that the information that can be stored in the frontal cortex is maybe in the petabytes, this doesn't seem to be a problem. But that probably isn't the point you were making.<br /><br />As to Chomsky's claim that the whole lexicon is innate, I certainly do not support this claim which is bizarre, implausible and unsupported by any evidence whatsoever.<br />I don't know how widely shared it is. Most minimalists seem to keep quiet about it. But that is again a separate issue.Alex Clarkhttps://www.blogger.com/profile/04634767958690153584noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-9430280155922478572013-03-24T11:05:54.966-07:002013-03-24T11:05:54.966-07:00Sorry, FPT means 'fixed parameter tractability...Sorry, FPT means 'fixed parameter tractability' which is an update of the classic complexity theory to deal with problems that vary in size in different ways. Largely irrelevant to the main thrust of the argument. Alex Clarkhttps://www.blogger.com/profile/04634767958690153584noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-75807820122931783592013-03-24T11:04:25.184-07:002013-03-24T11:04:25.184-07:00I hope my continuing despite having called it quit...I hope my continuing despite having called it quits isn't already taken as the basis for a potential argument to undermine my credibility...<br /><br />"I had asked you based on what you prefer Chomsky's framework."<br />Then probably I didn't make this clear enough. The major attraction for me is pragmatic, so here goes in a nutshell: If there is any theory to be had about human cognition, it's not going to be a theory that accounts for "everything" (and establishing this - to me at least - always was the main reason for bringing up "creativity"). And if it's going to be an insightful theory about some smaller aspect, say, our ability to produce and understand sentences (and our ability to acquire this ability), we'd better not use too many notions that themselves stand in the need for explanation themselves, say, "theory of mind" or "rich social cognition" or "metaphorical extension", as the currently major alternative paradigm I'm aware of (usage-based approaches I think it's called) suggests.<br /><br />I prefer a theory that tries to do without recourse to these additional notions, and as I haven't seen strong evidence in favor of this being an utter failure, I still have my hopes up that this will get us somewhere. Whereas I'm very skeptical of there ever being insightful theories of, say, our "general social cognition", but yes, that's just my pragmatic reasons for preferring a "Chomsykan" (note the scare quotes; I'm no adherent of Minimalism but would probably qualify as a bio-linguist) approach.<br /><br />(Additional question: I dug up Katz 2004, and I don't find any discussion of Kripke (1982) (although it's in the bibliography, it doesn't seem to be cited at least once in the main text), and his discussion of Quinean indeterminancy (which you also might have had in min?) didn't strike me as providing a way out if one shares "Quine's empirical assumptions". And I didn't find anything about attempts to naturalize intentionality...<br />So in case I'm simply overlooking the relevant part, could you give me page numbers?)benjamin.boerschingerhttps://www.blogger.com/profile/00894608988488218285noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-44938842171296290832013-03-24T10:57:08.453-07:002013-03-24T10:57:08.453-07:00So *you* accept that the argument that merge opera...So *you* accept that the argument that merge operates on tokens of sets is a valid retort?<br /><br />I have no interest in arguing about what Chomsky or Postal might think of this argument; I am interested only in the issues themselves. I am also not interested in defending Chomsky against charges of failing to address criticisms. Rather I am interested in the validity of those criticisms.<br /><br />I care very much about the physical limitations -- I think though along with almost everyone else in the last 30 years that the best way of dealing with these physical limitations are through the tools of complexity analysis: namely asymptotic worst case bounds, perhaps modified by the FPT stuff, rather than by a simple fixed finite bound on the size of storage. So maybe there are arguments against using complexity analysis and in favour of a fixed bound, but I don't know of any -- do you?<br />Indeed I care so much about these issues that I recently wrote a paper on it (with Shalom Lappin) called "Complexity in Language Acquisition" that you can find here (http://onlinelibrary.wiley.com/doi/10.1111/tops.12001/abstract) and a copy is available freely.<br /><br />So yes, I agree that the brain is finite and cannot store numbers beyond a certain size. But that is a different argument which I thought we had resolved -- namely what sorts of idealisations are appropriate or not. What we are currently discussing is the ontological incoherence that you allege, but which seems to be soluble by considering that the sets of which Chomsky talks are in fact merely tokens in the same way that the numbers that I manipulate when paying for my coffee in the morning are.Alex Clarkhttps://www.blogger.com/profile/04634767958690153584noreply@blogger.comtag:blogger.com,1999:blog-5275657281509261156.post-64019397062679428762013-03-24T10:16:37.188-07:002013-03-24T10:16:37.188-07:00minor erratum:
"lack of promising approaches...minor erratum:<br /><br />"lack of promising approaches to translating talk about mental representations into talk about brains, and that this ought to be considered a problem for most of cognitive science as well, if you think bio-linguists ought to worry about there not being specific proposals in this direction, a point that you have been consistently ignoring (which is different from disagreeing with the point and giving reasons for your disagreement)"<br /><br />I had asked you based on what you prefer Chomsky's framework. Turning around and answering everyone has problems re translating talk about mental representations into talk about brains is not an answer to this question. You still have not given any. I happen to agree that at the moment no one has a satisfactory account - so why would I argue with you about that? I have given no reasons for disagreement because there is no disagreement.Anonymoushttps://www.blogger.com/profile/03443435257902276459noreply@blogger.com