From kemmer at RUF.RICE.EDU Wed Dec 1 02:15:19 1999 From: kemmer at RUF.RICE.EDU (Suzanne E Kemmer) Date: Tue, 30 Nov 1999 20:15:19 -0600 Subject: Doctoral Fellowships at Rice University Message-ID: DOCTORAL FELLOWSHIPS IN LINGUISTICS AT RICE UNIVERSITY The Department of Linguistics at Rice University announces the opening of competition for its doctoral fellowships for 2000-2001. The Ph.D. program at Rice emphasizes the study of language use, the relation of language and mind, and functional approaches to linguistic theory and description. A strong component of the program is field studies in particular language areas. Areas of intensive research activity in the department include cognitive/functional linguistics, language universals and typology, language change and grammaticalization studies, lexical semantics, corpus linguistics, computational modelling, neurolinguistics, phonetics, sociolinguistics, and second language acquisition. Interdisciplinary opportunities are available with the Ph.D. programs in Cognitive Psychology, Philosophy, Anthropology, the interdisciplinary group in Cognitive Sciences, and the Center for Cultural Studies. The department hosts a distinguished speakers series, whose recent and imminent speakers include Marianne Mithun, Wallace Chafe, Tom Givon, Megan Crowhurst, and Knud Lambrecht. The department also sponsors a biennial Symposium on Language. The upcoming Symposium in April 2000 is "Causation and Interpersonal Manipulation in the Languages of Central and South America." The last two symposia were "The Interface between Comparative Linguistics and Grammaticalization: Languages of the Americas" (1997) and "Usage-Based Models of Language" (1995). Speakers have included Bernd Heine, Alexandra Aikhenvald, Berend Hoff, Ronald Langacker, Joan Bybee, Brian MacWhinney, Janet Pierrehumbert, Douglas Biber, Tom Givon, John Du Bois, Mira Ariel, and Arie Verhagen. FACULTY AND RESEARCH INTERESTS Michel Achard, Ph.D. Linguistics, University of California, San Diego. Cognitive linguistics, French syntax, second language acquisition. Michael Barlow, Ph.D. Linguistics, Stanford University. Grammatical theory, corpus linguistics, computational linguistics, second language acquisition, discourse. James Copeland, Chair, Ph.D. Linguistics, Cornell University. Functional linguistics, Germanic linguistics, grammaticalization, American Indian linguistics (Tarahumara). Philip W. Davis, Ph.D. Linguistics, Cornell University. Semantics and syntax, language and intelligence, Amerindian (Bella Coola; Alabama), Austronesian (Atayal, Ilokano, Yogad inter alia). Spike Gildea, Ph.D. Linguistics, University of Oregon. Diachronic syntax, field methods and ethics, phonology, typological/functional linguistics, Amazonian linguistics, Cariban languages (Akawayo, Arekuna, Tiriyo). Suzanne Kemmer, Ph.D. Linguistics, Stanford University. Cognitive linguistics, typology and universals, lexical semantics, semantics of grammar, syntactic and semantic change, Germanic, Austronesian, Nilo-Saharan (Luo). Sydney Lamb, Ph.D. Linguistics, University of California, Berkeley. Cognitive linguistics, neurolinguistics, neural network modelling, Amerindian (Monachi). E. Douglas Mitchell, Ph.D. Linguistics, University of Texas at Austin. Comparative Indo-European linguistics, historical linguistics, history of linguistics, early Germanic dialects, Sanskrit. Nancy Niedzielski, Ph.D. Linguistics, University of California, Santa Barbara. Phonetics, digital speech processing, language and society, American dialectology. Stephen A. Tyler, Ph.D. Anthropology, Stanford University. Cognitive studies, philosophy of language, anthropological linguistics, languages of India (Koya). FINANCIAL AID Graduate fellowships include tuition and a cash stipend. Fellowships are normally renewable for four years upon satisfactory performance, and students can apply for a fifth year of support. The department has so far been fortunate to be able to support all its graduate students. RICE UNIVERSITY Rice University, founded in 1912, is a private university dedicated to the promotion of arts and letters, science, and engineering. The university is highly selective, and departments are small and focused. The campus is spacious, tree-lined, and architecturally distinctive (a blend of Mediterranean and Renaissance). Rice is a close-knit academic community and the Department of Linguistics in particular offers opportunities for personalized interaction and collaboration with faculty. Current enrollment is ca. 2700 under- graduates and 1,200 graduate students; faculty:student ratio is 1:9. Houston is the America's fourth largest city and offers the full array of urban amenities (fine arts, large city parks, museums...). It is ethnically extremely diverse (affording not only excellent opportunities for working with linguistic consultants, but also a huge number of restaurants representing a wide spectrum of cuisines at all levels of affordability.) The university is 45 minutes from the Gulf Coast (Galveston Island). Rents are affordable on a graduate stipend. The university and department offer a wide range of computing facilities available to students. The library has an outstanding linguistics collection, including a vast array of reference grammars. Doctoral students are eligible for support for travel to conferences and for summer research funds. The department also supports linguistic consultant fees and photocopying accounts for its doctoral students. Both U.S. and international applicants are admitted on the same basis. Current graduate students include not only U.S. students but also students from Australia, Brazil, China, and Germany. Prospective students of diverse linguistic backgrounds are encouraged to apply. Visiting students with their own funding who would like to come to Rice for a limited time to work with an individual faculty member should contact that faculty member directly. APPLICATION DEADLINE: February 1, 2000. Prospective applicants for the Ph.D. program must take the Graduate Record Examination as soon as possible, and have the results sent to the university in time for consideration in February. Non-native speakers of English must also take the TOEFL test. Admission is competitive. For more information about the program and the application process, please contact: Department of Linguistics, MS 23 Rice University P.O. Box 1892 Houston TX 77251-1892 (713) 527-6010 Departmental Coordinator: Ursula Keierleber, ukeie at ruf.rice.edu Graduate Adviser: Philip Davis, pwd at ruf.rice.edu See also the home page at http://www.ruf.rice.edu/~ling From brucerichman at HOTMAIL.COM Wed Dec 8 15:38:29 1999 From: brucerichman at HOTMAIL.COM (bruce richman) Date: Wed, 8 Dec 1999 07:38:29 PST Subject: Announcement: Alternatives to Chomsky meeting Message-ID: ALTERNATIVES TO CHOMSKY A NEW PARADIGM FOR LANGUAGE STUDIES FOR A NEW MILLENIUM A Conference to be held on September 4, 2000 at Rutgers University in New Brunswick, New Jersey, in association with the Language Origins Society year 2000 meeting. The main obstacle that we have today to clearly understanding the nature and origins of language is the overly formalistic, anti-empirical, anti-historical influence of Chomsky's paradigm for doing linguistics. It has become very clear in recent years that Chomsky's generative calculus model of linguistics has no relevance at all to anything about actual language. It has also become clear that the main generative notion of the innateness of language is based on flimsy, non-existent evidence. Despite this, and despite the fact that many people in many different fields are now actually studying real language in realisitic, empirical ways, and despite the fact that many people intuitively understand the emptiness of generative claims and practices, it still remains true, that to the world at large, Chomsky's theories are somehow considered an important "scientific breakthrough." So, the time has come that those of us who want to start a new paradigm for language studies, who want to begin an empirical way of studying real language, should simply explain why the entire Chomskyan method must be thrown away. We are planning to invite 7 or 8 people from a wide range of fields to speak at our one-day seminar which will be held on Monday, September 4, just prior to the regular meetings of the Language Origins Society. (LOS 2000 will continue on through Saturday, September 9.) If you are interested in participating please send a brief abstract of what you would like to present to: Bruce Richman e-mail brucerichman at hotmail.com 3805 Woodridge Rd. Cleveland Hts., Ohio 44121 216-381-7510 ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From moorej at UCSD.EDU Wed Dec 8 23:34:51 1999 From: moorej at UCSD.EDU (John Moore) Date: Wed, 8 Dec 1999 15:34:51 -0800 Subject: Alternatives to Chomsky Message-ID: At 07:38 AM 12/8/99 PST, bruce richman wrote: > It has become very clear in recent years that Chomsky's generative >calculus model of linguistics has no relevance at all to anything about >actual language. It has also become clear that the main generative notion >of the innateness of language is based on flimsy, non-existent evidence. > Despite this, and despite the fact that many people in many different >fields are now actually studying real language in realisitic, empirical >ways, and despite the fact that many people intuitively understand the >emptiness of generative claims and practices, it still remains true, that to >the world at large, Chomsky's theories are somehow considered an important >"scientific breakthrough." > So, the time has come that those of us who want to start a new paradigm >for language studies, who want to begin an empirical way of studying real >language, should simply explain why the entire Chomskyan method must be >thrown away. I agree that there are severe problems with the way much work within the Chomskyan paradigm has been conducted, but I wouldn't agree that this means that it "has no relevance at all to anything about actual language." Furthermore, very good empirical work has come out of the generative tradition. I would stop and consider what would be lost by throwing the entire generative methodology away. Finally, there are a number of generative frameworks. Do the above comments apply equally to P&P, Minimalism, LFG, HPSG, OT., etc.? How much of this is going to be thrown away in this sweeping paradigm shift - do any of these approaches offer any insight that one might want to keep? John Moore http://ling.ucsd.edu/~moore/ From lakoff at COGSCI.BERKELEY.EDU Thu Dec 9 05:05:41 1999 From: lakoff at COGSCI.BERKELEY.EDU (George Lakoff) Date: Wed, 8 Dec 1999 22:05:41 -0700 Subject: Bruce Richman's announcement In-Reply-To: <19991208153829.48152.qmail@hotmail.com> Message-ID: In case anyone is interested, I'm attaching a copy of the overheads from an all-too-brief and oversimplified talk I gave last weekend at the UCSD Conference on Explanation in Linguistics. Jerry Feldman and I are currently writing a book on the emerging Neural Theory of Language that we are working on at Berkeley, with collaborators at the University of Chicago and SRI.This is intended to supplement the already existing field of Cognitive Linguistics. For a selected bibliography, see the references in Philosophy in the Flesh, by myself and Mark Johnson. Also take a look at Mike Tomasello's THE NEW PSYCHOLOGY OF LANGUAGE. At the International Cognitive Linguistics Conference in Stockholm last year, about 750 researchers from 34 countries attended. We don't need to start an alternatives to Chomsky movement. Such a worldwide movement already exists, with a huge body of work, a journal, a number of book series, an international society, regular conferences, and so on. I suspect that most ICLA members would probably agree with Richman's negative assessment of Chomskyan linguistics. But we are interested in positive research, not negative polemics. A huge amount of such research is going on at an extermely high level. It's an exciting field -- at least as exciting as generative linguistics was back in the 60's, but far more profound. Anyone can join in. But there is a bit of literature to catch up on. I'm reading as fast as I can to keep up. I wish Richman well with his conference and hope it takes a positive form. George Lakoff -------------- next part -------------- A non-text attachment was scrubbed... Name: UCSD_Overheads_Revised_2 Type: application/mac-binhex40 Size: 180164 bytes Desc: not available URL: From jlmendi at POSTA.UNIZAR.ES Thu Dec 9 11:38:50 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Thu, 9 Dec 1999 12:38:50 +0100 Subject: Alternatives to Chomsky In-Reply-To: <3.0.3.32.19991208153451.009c5610@ling.ucsd.edu> Message-ID: At 07:38 AM 12/8/99 PST, bruce richman wrote: > > It has become very clear in recent years that Chomsky's generative >calculus model of linguistics has no relevance at all to anything about >actual language. It has also become clear that the main generative notion >of the innateness of language is based on flimsy, non-existent evidence. > Despite this, and despite the fact that many people in many different >fields are now actually studying real language in realisitic, empirical >ways, and despite the fact that many people intuitively understand the >emptiness of generative claims and practices, it still remains true, that to >the world at large, Chomsky's theories are somehow considered an important >"scientific breakthrough." > So, the time has come that those of us who want to start a new paradigm >for language studies, who want to begin an empirical way of studying real >language, should simply explain why the entire Chomskyan method must be >thrown away. > Ok, the world is wrong and you're right! I should be grateful if you could define 'realistic' and 'empirical'. Thanks in advance. ***************************************** Dr. Jose-Luis Mendivil Giro Dept. of General and Hispanic Linguistics Universidad de Zaragoza C/ Pedro Cerbuna, 12 50009 Zaragoza (Spain) Phone: (+34) 976 761 000 Fax: (+34) 976 761 541 E-mail: jlmendi at posta.unizar.es From john at RESEARCH.HAIFA.AC.IL Thu Dec 9 13:42:11 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Thu, 9 Dec 1999 15:42:11 +0200 Subject: Alternatives to Chomsky Message-ID: While I agree fully with Bruce Richman's assessment of the value of Chomskyan linguistics, I disagree that we should spend much effort on `simply explain(ing) why the entire Chomskyan method must be thrown away.' Chomskyism is a religion, not an intellectual program. It is not something which can be rationally argued with. Chomsky himself does not even attempt to give rational arguments for the allegedly sweeping paradigm changes he seems compelled to make every decade or so to give the appearance of progress--he simply says that the new approach seems more attractive, or something like that, and that's enough to change the minds of his apostles. Several years ago I challenged funknetters to identify themselves if they had been convinced for or against a Chomskyan position after graduate school. Although I repeated this challenge several times in the context of a heated debate which many people were involved in, and although many people responded to other points in my messages, only a single person (Ellen Prince, towards Chomskyan linguistics) reported having had such an experience. I concluded that any convincing arguments against Chomsky must be directed at people who have not yet completed graduate school (in any case, Chomskyan devotees have a tendency to stop doing active research and detach from research affairs when they have been away from The Master for a few years, so I would not say 'conversions' like, e.g. Tom Wasow are worth anything). And since almost all graduate schools now are already clearly pro-Chomsky or anti-Chomsky (I include places such as Stanford as 'pro-Chomsky', although some of the faculty members have made names for themselves as professional gadflies), there is no realistic way to compete for the loyalties of graduate students either. Arguing with Chomsky is a waste of time. I'm all for alternative approaches, but Chomskyism is only going to die with Chomsky--and the man is 71 years old. John From bresnan at CSLI.STANFORD.EDU Thu Dec 9 17:46:32 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 9 Dec 1999 09:46:32 -0800 Subject: Alternatives to Chomsky In-Reply-To: Your message of Thu, 09 Dec 1999 15:42:11 +0200. Message-ID: In a message of 9 Dec 1999 John Myhill wrote: > ... Chomskyan devotees have a tendency to stop doing active research and > detach from research affairs when they have been away from The Master for a > few years, so I would not say 'conversions' like, e.g. Tom Wasow are worth > anything). And since almost all graduate schools now are already clearly > pro-Chomsky or anti-Chomsky (I include places such as Stanford as > 'pro-Chomsky', although some of the faculty members have made names for > themselves as professional gadflies), ... My, my. Let's not get personal. John, you may not be in touch any more with what's been going on at Stanford. We have been building intellectual bridges between formal and functional approaches. Elizabeth Traugott and Paul Kiparsky are working together on a historical project that centrally involves grammaticalization. Ivan Sag and some of his students have become deeply involved with the Fillmore-Kay Construction Grammar, which bears close family resemblances to HPSG/LFG. Bresnan (that's me) has a collaborative research project with Judith Aissen at Santa Cruz on "Optimal Typology" which focusses on using Optimality Theory as a formal tool to explore syntactic markedness hierarchies and explain some of their problematic properties (softness, variable expression, recurrence). We are arguing for using typologically motivated and functionally grounded constraints, much as has happened in phonology. Edward Flemming in phonology is one of the leading young exponents of the OT functionalist approach to phonology which argues that perception shapes the structure of language. Tom Wasow does corpus linguistic studies and natural language processing, so you're more likely to find his recent papers in _Cognitive Psychology_ and _Language Variation and Change_ than in _Linguistic Inquiry_. The corpus approach is very big at Stanford. Our new young faculty memember Chris Manning who holds a joint appointment with Computer Science argues that the new surge of statistical approaches to Natural Language can restore some of the balance in the field by strengthening areas such as historical linguistics and sociolinguistics that were somewhat marginalized by the Chomskyan revolution. Come spend a sabbatical year at Stanford, and see for yourself! Joan Bresnan http://www-ot.stanford.edu/bresnan/ From lakoff at COGSCI.BERKELEY.EDU Thu Dec 9 18:26:14 1999 From: lakoff at COGSCI.BERKELEY.EDU (George Lakoff) Date: Thu, 9 Dec 1999 11:26:14 -0700 Subject: Alternatives to Chomsky In-Reply-To: Message-ID: >I should be grateful if you could define 'realistic' and 'empirical'. >Thanks in advance. > > >***************************************** >Dr. Jose-Luis Mendivil Giro >Dept. of General and Hispanic Linguistics > >Universidad de Zaragoza >C/ Pedro Cerbuna, 12 >50009 Zaragoza (Spain) > >Phone: (+34) 976 761 000 >Fax: (+34) 976 761 541 >E-mail: jlmendi at posta.unizar.es That's easy to find out. Check with any member of the Cognitive Linguistic Society of Spain (AESLA). Their e-mail list address is: lingcog at fcu.um.es. Also take a look at the discussion of Chomsky's philosophy in Philosophy in the Flesh (obtainable via amazon.com). From nrude at ucinet.com Thu Dec 9 22:29:16 1999 From: nrude at ucinet.com (Noel Rude) Date: Thu, 9 Dec 1999 14:29:16 -0800 Subject: Announcement: Alternatives to Chomsky meeting Message-ID: Well, folks, Maybe there's even one more approach besides Chomsky's biological innateness vs. our supervenience upon a thousand-and-one physical phenomena. Like it or not, a large number (perhaps a majority) of mathematicians (and lots of other people in the nonbiological "hard sciences") are still Platonists (don't shoot me!) of the old-fashioned stripe. These blokes believe that the laws of physics lie further down in the hierarchy of reality than the primitive axioms from which mathematics (and logic) derive. Modern cosmological theory generally assumes that physical law traces from the Big Bang (however conceived), and today's theoreticians profit (they think!) from studying other possible worlds all the while assuming that the higher-arching principles of math and logic must obtain there too (and not necessarily the laws of physics). Whereas the formalists believe we invent math/logic, the mathematical Platonists believe we discover it. And--lest anyone gets too frightened--many maybe most of these Platonists are also hard core atheists. And they are reductionists. They just believe that "those things which could be no other way" have a reality of their own. So instead of throwing spit-wads at each other from these various camps within a strictly reductionist "physical law" materialism, why not keep our cool and at least acknowledge what the big boys in the hard sciences are thinking. Maybe Chomsky is on to something. Maybe some of the core of Language--maybe rather than being biologically "hard-wired"--maybe we as humans are equipped to "discover" it. No, let's avoid extremism (as Givon always warns against). Let's milk the biological and discourse-pragmatic sides of the coin for all they're worth. But maybe some folks who study the more abstractly logical side of things should hang in there too. Just what is there about Language "that could be no other way"? What does Natural Language share with other information systems? Is there something there that doesn't merely "emerge" from a swirl of atoms? Why is it we can't have both pragmatics AND Plato? Noel the Rude From fjn at U.WASHINGTON.EDU Thu Dec 9 22:55:45 1999 From: fjn at U.WASHINGTON.EDU (Frederick Newmeyer) Date: Thu, 9 Dec 1999 14:55:45 -0800 Subject: Alternatives to Chomsky meeting Message-ID: Joan Bresnan made an excellent point in her recent Funknet posting. The world of linguistic theorizing is a much bigger place than it was in the past. In my paper at the recent UCSD Conference on Explanation in Linguistics (the conference that George Lakoff referred to), I referred positively to work by John Haiman, Talmy Givon, and Jack Hawkins, and did not mention Chomsky's name even once. There are lot of people out there who are actively trying to synthesize the contributions of various linguistic traditions, including the generative. I do hope that the organizers of the 'Alternatives to Chomsky' conference will organize a meeting that is sensitive to that fact. Fritz Newmeyer From jmacfarl at unm.edu Fri Dec 10 01:06:20 1999 From: jmacfarl at unm.edu (jmacfarl at unm.edu) Date: Thu, 9 Dec 1999 18:06:20 -0700 Subject: How Functional is OT? In-Reply-To: <199912091745.JAA25025@hypatia.Stanford.EDU> Message-ID: On 9 Dec 99, at 9:46, Joan Bresnan wrote: > Bresnan (that's me) has a collaborative > research project with Judith Aissen at Santa Cruz on "Optimal > Typology" which focusses on using Optimality Theory as a formal tool > to explore syntactic markedness hierarchies and explain some of their > problematic properties (softness, variable expression, recurrence). We are > arguing for using typologically motivated and functionally grounded > constraints, much as has happened in phonology. Edward Flemming in > phonology is one of the leading young exponents of the OT functionalist > approach to phonology which argues that perception shapes the structure of > language. Dear Funknetters, As I understand it, OT posits innate or a-priori constraints on the well-formedness of linguistic structure. This leaves me confused when people refer to OT as a functional theory. For example, in Bresenan's description of her work.... We are arguing for using typologically motivated and functionally grounded constraints, much as has happened in phonology. My question is how does "function" inform "constraints" if these constraints are said to be innate or a-priori? Thanks, ************************* James MacFarlane University of New Mexico ************************* From bresnan at CSLI.STANFORD.EDU Fri Dec 10 03:05:03 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 9 Dec 1999 19:05:03 -0800 Subject: How Functional is OT? In-Reply-To: Your message of Thu, 09 Dec 1999 18:06:20 MST. <199912100113.TAA00954@listserv.rice.edu> Message-ID: >>>jmacfarl at unm.edu said: > > As I understand it, OT posits innate or a-priori constraints on the > well-formedness of linguistic structure. This leaves me confused > when people refer to OT as a functional theory. > > For example, in Bresenan's description of her work.... > > We are arguing for using typologically motivated and functionally > grounded constraints, much as has happened in phonology. > > My question is how does "function" inform "constraints" if these > constraints are said to be innate or a-priori? Oh, dear. There's so much confusion about this topic. I'm afraid of stepping into a FUNKpit. Well, here goes nothing... 1. Constraints can be innate without being a priori. For example, the possible inventories of phonological segments and the distribution of phonological contrasts within and across languages reflect constraints imposed by the limits of our articulatory and perceptual systems, which are innate in the sense that we are born with them. The work of Paul Boersma (http://fonsg3.let.uva.nl/paul/), Donca Steriade (http://www.humnet.ucla.edu/humnet/linguistics/people/steriade/steriade.htm), and Edward Flemming (http://www.stanford.edu/~flemming/) is representative. For example, Steriade's web page says: Donca Steriade works mainly on the interactions between phonological patterns and speech properties considered to be exclusively phonetic or even non-linguistic, such as perceptibility and avoidance of articulatory effort. The work is based on the hypothesis that these factors play a role in shaping sound patterns, not only in an evolutionary sense--as argued by John Ohala and Bjorn Lindblom--but also by defining the grammatical constraints whose interactions yield the phonologies of individual languages. In an analogous way, markedness hierarchies in syntax and semantics (Silverstein, Givon, Comrie et al.) have been postulated to reflect innate properties of the human perceptual and cognitive systems. --Not necessarily language-specific properties, but part of the perceptual/cognitive apparatus people are born with. For some VERY BEGINNINGS of OT work incorporating constraints of this kind, have a look at the background papers by Aissen and Bresnan on our Optimal Typology web site at Stanford: http://www-ot.stanford.edu/ot/. 2. Constraints can be innate without having language-specificity. (See the discussion on this very list some time ago, in which important distinctions about innateness and language specificity were made by Liz Bates, among others... don't you remember it?) 3. And constraints can be universal without being innate. Examples would be pragmatic constraints on reference, communicative constraints that influence discourse organization, cohesion, etc. -------- OT is really a theory of constraint INTERACTION, not a theory of what the substance of constraints must be, or where they must come from. You can certainly invent as many a priori, autonomous, artificial, mechanical, etc., constraints as you like (and many have...). But you don't have to. Explicit constraints within a precise (I'm afraid to use the word "formal" on this list!) theory of constraint interaction are useful in that they enable you to experiment with prioritizing constraints in order to generate typologies and test language-particular interactions of constraints. Boersma's model adds the nice refinement of probabilistically variable ranking to the constraints. See Boersma and Hayes' recent paper in the Rutgers Optimality Archive. ------- TTFN-- Joan From gentner at ILS.NWU.EDU Fri Dec 10 03:54:03 1999 From: gentner at ILS.NWU.EDU (Dedre Gentner) Date: Thu, 9 Dec 1999 21:54:03 -0600 Subject: language learning Message-ID: If the discussions of new approaches to grammar extend to issues of how it is learned, I'd like to suggest that analogy be reconsidered as an important mechanism in language learning. From haspelmath at EVA.MPG.DE Fri Dec 10 13:34:24 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Fri, 10 Dec 1999 13:34:24 +0000 Subject: How functional is OT? Message-ID: I found Joan Bresnan's remarks very interesting indeed, and I feel that FUNKNET subscribers should pay close attention to the work that is going on at Stanford and Santa Cruz. I completely agree that OT says nothing about the nature of the constraints, and that OT analyses with functionally-motivated constraints are possible. In fact, people in Natural Morphology (W. Wurzel, W. Dressler and colleagues) have been talking about the ranking of functional principles for almost two decades. But although especially in phonology more and more people seem to be interested in functional explanations of OT constraints, it is also true that most OT practitioners probably still think of constraints as abstract entities that are directly innate and need not and cannot be explained further. If the new wave of (post-MIT) functionalism comes from Stanford/Santa Cruz, that's great. And if you people there occasionally acknowledge that some people have said similar things about language function explaining language form before, that would be nice. Martin -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From geoffn at SIU.EDU Fri Dec 10 14:33:09 1999 From: geoffn at SIU.EDU (Geoffrey S. Nathan) Date: Fri, 10 Dec 1999 08:33:09 -0600 Subject: How Functional is OT? In-Reply-To: <199912100113.TAA00954@listserv.rice.edu> Message-ID: Let me chime in here with an endorsement of what Joan Bresnan and Martin Haspelmath have said about the functional nature of current phonological theory. It was interesting to note that at the Milwaukee conference on formalism and functionalism held a few years back there was a radical difference between the kinds of talks that the syntacticians and semanticists and typologists gave and those that the three phonologists gave. I felt that Joan Bresnan, Bruce Hayes and I all had things to say to each other, and listened and learned from each other, while the others seemed to be talking past each other. And it seems to me it was precisely because the 'formalists' in phonology have (in many cases) insisted that phonological structures be grounded (either in a technical sense, as used by Archangeli et al., or in the more general sense that functionalists use the term) in facts about production and perception. This, of course, is what Stampe and the Natural Phonologists proposed in the sixties and seventies. When Stampe used the term 'innate' he meant, explainable by the nature of the physiology, physics and psychophysics involved. Unfortunately, at the time, Chomsky was using 'innate' to mean unexplainable extrinsically, and NP's message got rejected by those rejecting Chomsky's version of innateness. All this is water under the bridge now, but I will second, (third?) the plea that theories founded purely on negative impulses are unlikely to gain us the insights we are looking for. And many valuable insights about the nature of language have arisen from purely formalist investigations, even if, as Karen van Hoek and Paul Deane (not to mention Lakoff, Langacker etc.) have shown, the insights have reasonable non-linguistic explanations after all. Geoff Geoffrey S. Nathan Department of Linguistics Southern Illinois University at Carbondale Carbondale, IL, 62901-4517 Phone: (618) 453-3421 (Office) / FAX (618) 453-6527 (618) 549-0106 (Home) geoffn at siu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From macw at CMU.EDU Fri Dec 10 15:47:06 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Fri, 10 Dec 1999 10:47:06 -0500 Subject: OT Message-ID: Dear Funknet, I think that a discussion on Funknet of the claims underlying OT could be extremely helpful. I agree with Joan Bresnan and Geffrey Nathan that there is nothing afunctional about grounding constraints on the facts of the perceptual and production apparatus. On the contrary, it would seem to me that the natural phonologies of Dressler, Bybee, Vennemann, Ohala, and others make an interesting functional statement when they trace the grounding of constraints on the speech production and perception apparatus. I agree that Hayes and others are doing a nice job of providing this type of grounding to OT. I promise to read the various sources that Joan points too, since I imagine that they further elaborate this important contribution. However, as a psycholinguist, I have been disappointed by five crucial "strategic" decisions in the development of OT that have tended to vitiate its potential for constructing a psychologically plausible linguistic theory of the type that Joan Bresnan and others have often sought. In particular, 1. Early on, OT was supposed to be linked to connectionist modeling. However, after the first few years, this linkage was largely dropped. Dedre Gentner's interest in analogy as an acquisition or production mechanism has pretty much suffered the same fate, I would guess. 2. Early on, OT constraints were supposed to have strength levels. However, later on this feature was eliminated. In our work on the Competition Model, Liz Bates and I learned how important strength levels are for describing and predicting psycholinguistic data. In fact, one has to go beyond strength levels for separate constraints and look at what we can conflict validity, but none of this could possibly fit in with current OT. 3. Early on, learning of a phonology (or grammar) could have involved the strengthening and weakening of constraints. Later on, it required the types of triggers used by G-B and P&P theories. 4. Even from the beginning, OT never questioned the need to provide a single abstract underlying structure for each lexical item. As far as I can tell, this commitment is the one that tends to lead linguistic theories away from being able to develop psychological reality. Trying to preserve this approach in OT models of syntax would be equally problematic. 5. At no point was OT really committed to an account of online processing. If problems 1-4 were not present, I would not consider this a fatal flaw, since the notion of constraint ordering has clear interest for typology and language change, at the very least. I am just a psycholinguist, so I am happy to have linguists explain to me how I have strayed in my judgments. But I would really like to hear some open discussion of these issues. If it has already occurred on some OT bulletin board, perhaps people can simply point me to an archived discussion of these issues. --Brian MacWhinney From Twright at ACCDVM.ACCD.EDU Fri Dec 10 15:53:36 1999 From: Twright at ACCDVM.ACCD.EDU (Tony Wright) Date: Fri, 10 Dec 1999 09:53:36 -0600 Subject: How Functional is OT? In-Reply-To: <4.2.0.58.19991210082140.009f5190@saluki-mail.siu.edu> Message-ID: At 08:33 AM 12/10/99 -0600, Geoffrey S. Nathan wrote: > Let me chime in here with an endorsement of what Joan Bresnan and > Martin Haspelmath have said about the functional nature of current > phonological theory. Indeed, this very trend is lamented by Mark Hale and Charles Reiss in _Substance Abuse and Dysfunctionalism: Current Trends in Phonology_ in which it is argued that phonology, particularly current implementations of OT, are drifting off toward functionalism, and that phonology qua phonology should be "all form, no substance." (This paper was once on the Rutgers Optimality Archive, but I find it has been removed, probably owing to its subsequent publication in Linguistic Inquiry (if memory serves)). --Tony Wright From nrude at ucinet.com Fri Dec 10 18:47:28 1999 From: nrude at ucinet.com (Noel Rude) Date: Fri, 10 Dec 1999 10:47:28 -0800 Subject: Phonology Message-ID: Folks! Have been away from the lit for quite a spell now and can't say I know anything about acronymic linguistics (nor much of anything, for that matter). What bothers me about letting linguistics degenerate into phonology is that this is what happens when we insist on grounding everything in neurology and physiology. Such was the error of the Bloomfeldians. Chomsky was right in so far as he argued for studying Language--syntax and semantics and universals and all--apart from any purely mechanistic theory (behaviorist or otherwise--is that what his "innateness" was meant to do?). We can still study Language and keep it empirical--grounding it in legitimate data (texts, etc.)--even if the Chomskians have tended not to do this. We can operate within a purely communicative theory of language. And yes as far as possible our functional explanations should be grounded in reality (biology, psychology, pragmatics). But we might remember that no one as yet has succeeded in defining information in purely physical terms (grams, centimeters, volts, etc.). If we don't want to deal with the logical/informational side of language then it will have to fall to the philosophers and mathematicians to do so. Of course I know most of us still believe in syntax and semantics. I just thought it would be good to remind ourselves that Chomsky did help in delivering us from the "biological extremism" of mid century America. Rude again From bresnan at CSLI.STANFORD.EDU Fri Dec 10 19:02:40 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Fri, 10 Dec 1999 11:02:40 -0800 Subject: OT, functionalism, etc. Message-ID: This is a very interesting discussion, and I will try to respond to specific questions raised, in this and a following msg. But remember that I am only a student of Optimality Theory and can't speak with the voice of Authority. James MacFarlane asked: >My question is how does "function" inform "constraints" if these >constraints are said to be innate or a-priori? My reply was that OT does not require the constraints to be a-priori, or even innate, but that innateness is quite compatible with functionalism, a point already made on this list by Liz Bates some time ago, when she distinguished the hypothesis of innateness from the hypothesis of specificity to language (or autonomy). I would like to add that OT does make an important claim of (near) universality of constraints that is used to derive certain learnability and typological results. This is not something that functionalists should find alien. Couldn't conflicting constraints such as iconicity and economy be universal, but prioritized differently across different domains and different languages? Sergio Meira asks: > ... I mean, any constraints which are derived >from the biology of our speech aparatus must be a-priori. Or do I >misunderstand the way this term is being used? Well, my epistemology has grown rusty since my time as a student of philosophy. What you say could be true in an Aristotelian sense of the a priori as that from a causal explanation can be logically derived. But I was thinking of the later philosophical uses of the terms "a posteriori" and "a priori" to mean, roughly, what is derived from experience (empirical truths) and what is derived from reason alone (logical truths). Functional OT phonology appeals to experimental work in human speech perception and articulation to support the constraints hypothesized, so I would not call these constraints a priori. Sergio Meira states: >If OT is only a theory of constraint interaction, then the sources >and natures of constraints probably aren't important... Please don't misunderstand me. OT does not claim that the sources and natures of its constraints are unimportant. My point was rather that as a theory of constraint interaction OT does not dictate the substance of the constraints. It does assume that constraints are (near) universal and violable, and it does urge simple and natural constraints (so that the interactions are not concealed inside the content of complex constraints), but these assumptions are compatible with functionalist views, I believe. It is true that a variety of researchers have been attracted to OT, and in the field of syntax these have included those who like to import their favorite constraints from the Minimalist Program (or whatever) into the new theory. But a number of us have argued against this, and see functional/typological work as the best source for well motivated constraints in OT syntax. Martin Haspelmath writes: >But although especially in phonology more and more people seem to be >interested in functional explanations of OT constraints, it is also true >that most OT practitioners probably still think of constraints as >abstract entities that are directly innate and need not and cannot be >explained further. I agree that this is true, especially in syntax. If I may quote from the conclusion of my paper "Optimal Syntax": Because OT per se is a theory of constraint interaction rather than a theory of substantive linguistic constraints, it is compatible with a wide range of substantive theoretical choices. (Some consider this an explanatory weakness of the framework, but it is also the source of its great integrative potential.) In phonology and to a lesser extent morphology, OT has led to a fundamental rethinking of the domain and to the widespread adoption of nonderivational theories. Syntax, in contrast, is still greatly influenced by the derivational frameworks advanced by Chomsky, and much of the initial work applying OT to syntax reflects this way of thinking by simulating derivational analyses. It is instructive to consider the history of architectural design, which shows that earlier designs, for example in bridge-building, persist long after the development of new materials with radically different engineering properties (e.g. steel compared to wood and stone). Martin Haspelmath adds: >If the new wave of (post-MIT) functionalism comes from Stanford/Santa >Cruz, that's great. And if you people there occasionally acknowledge >that some people have said similar things about language function >explaining language form before, that would be nice. I agree completely about this. Many of the most attractive and central ideas of OT about constraint interaction were anticipated by researchers in Natural Morpology, Natural Syntax, and pragmatics. That is one reason why OT seems to be a natural framework for incorporating and further developing and testing these ideas. The OT framework was initially developed by phonologists, who may be forgiven for not being aware of related work in different fields. But it would be unforgivable for us syntacticians to reinvent functional/typological theories in the guise of OT, without any acknowledgement... Geoffrey S. Nathan writes: > ... It was interesting to note that at the Milwaukee >conference on formalism and functionalism held a few years back there was a >radical difference between the kinds of talks that the syntacticians and >semanticists and typologists gave and those that the three phonologists >gave. I felt that Joan Bresnan, Bruce Hayes and I all had things to say to >each other, and listened and learned from each other, while the others >seemed to be talking past each other. I am pleased that we agree. But it must have been my agreeable spirit you were talking to in Milwaukee, because I wasn't there! TTFN-- Joan *---------------------------------------- ______ __o __o Joan Bresnan bresnan at stanford.edu ______ _`\<,_ _`\<,_ *---------------------------------------- ______ (*)/ (*) (*)/ (*) From dkp at U.ARIZONA.EDU Fri Dec 10 19:09:41 1999 From: dkp at U.ARIZONA.EDU (Dianne K. Patterson) Date: Fri, 10 Dec 1999 12:09:41 -0700 Subject: Phonology In-Reply-To: <38514AB9.4EFE@ucinet.com> Message-ID: On Fri, 10 Dec 1999, Noel Rude wrote: > Folks! > > Have been away from the lit for quite a spell now and can't say I know > anything about acronymic linguistics (nor much of anything, for that > matter). > > What bothers me about letting linguistics degenerate into phonology is > that this is what happens when we insist on grounding everything in > neurology and physiology. Perhaps I am missing something...but the standard position is that we do not want to confuse phonetics (which is mechanistic and mechanical) with phonology (which is a rule governed system and usually included as part of the grammar)...now people may wish to say phonology is not on the right track and should in fact be grounded in biology (e.g., John Ohala...and I have my own leanings in this direction)...but such views are outside the standard accepted views. > Such was the error of the Bloomfeldians. > Chomsky was right in so far as he argued for studying Language--syntax > and semantics and universals and all--apart from any purely mechanistic > theory (behaviorist or otherwise--is that what his "innateness" was > meant to do?). We can still study Language and keep it > empirical--grounding it in legitimate data (texts, etc.)--even if the > Chomskians have tended not to do this. We can operate within a purely > communicative theory of language. And yes as far as possible our > functional explanations should be grounded in reality (biology, > psychology, pragmatics). But we might remember that no one as yet has > succeeded in defining information in purely physical terms (grams, > centimeters, volts, etc.). There is a standard information theory. It is grounded in bits and underlies computer technology...but it is also used in standard ethology (the study of the behavior of communicating) which is generally used when looking at nonhuman behavior...but can, of course also be used to look at human behavior > If we don't want to deal with the > logical/informational side of language then it will have to fall to the > philosophers and mathematicians to do so. > > Of course I know most of us still believe in syntax and semantics. I > just thought it would be good to remind ourselves that Chomsky did help > in delivering us from the "biological extremism" I hope this is helpful, Dianne Patterson From geoffn at SIU.EDU Fri Dec 10 20:25:09 1999 From: geoffn at SIU.EDU (Geoffrey S. Nathan) Date: Fri, 10 Dec 1999 14:25:09 -0600 Subject: Phonology In-Reply-To: Message-ID: At 12:09 PM 12/10/1999 -0700, Dianne K. Patterson wrote: >On Fri, 10 Dec 1999, Noel Rude wrote: > > > Folks! > > > > Have been away from the lit for quite a spell now and can't say > I know > > anything about acronymic linguistics (nor much of anything, for that > > matter). > > > > What bothers me about letting linguistics degenerate into > phonology is > > that this is what happens when we insist on grounding everything in > > neurology and physiology. > >Perhaps I am missing something...but the standard position is that we do >not want to confuse phonetics (which is mechanistic and mechanical) with >phonology (which is a rule governed system and usually included as part of >the grammar)...now people may wish to say phonology is not on the right >track and should in fact be grounded in biology (e.g., John Ohala...and I >have my own leanings in this direction)...but such views are outside the >standard accepted views. Alas, the 'standard position' has become greatly confused. Many phonetician/phonologists now believe that there are language-specific (and thus not mechanistic and mechanical) phonetic implementation rules that are distinct from phonological rules, although perhaps both need to be grounded. These rules include the implementation of the feature [+voice] (which is voiceless in onset position in English, but voiced in French) and similar cases. Pat Keating has written on this issue, for example. And it should be made clear that Natural Phonology, for instance (and I think OT would agree with this way of expressing things) argued that while phonological constraints have physiological explanations, they are not mechanical in and of themselves, precisely because they are violable. To put things in a way that Hale and Reiss object to, it's easier to voice obstruents intervocalically, (for physiological reasons) and some languages give in to the temptation while others don't (i.e. suppress the process, as NP would say it, or rank Faithfulness higher, as OT would say it). Stampe always argued that processes are substitutions made on behalf of the vocal tract, not by it, and arguing that certain phonetically-motivated constraints are outranked by faithfulness constraints is saying the same thing. Geoff P.S. Nobody is arguing that all grammar is reduceable to physiology, but only that phonological processes (however interpreted these days) have physical explanations. Grammatical processes, according to Cognitive Grammar, at least, have extragrammatical cognitive explanations. GN Geoffrey S. Nathan Department of Linguistics Southern Illinois University at Carbondale Carbondale, IL, 62901-4517 Phone: (618) 453-3421 (Office) / FAX (618) 453-6527 (618) 549-0106 (Home) geoffn at siu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From bbergen at socrates.berkeley.edu Fri Dec 10 20:10:50 1999 From: bbergen at socrates.berkeley.edu (bbergen at socrates.berkeley.edu) Date: Fri, 10 Dec 1999 12:10:50 -0800 Subject: OT, functionalism, etc. In-Reply-To: <199912101901.LAA28017@hypatia.Stanford.EDU> Message-ID: Since the cents seem to be flying, I'd like to bring up a couple critiques of functionalist OT. I apologize if this commentary is on the long side, but I think that this position has been inadequately expressed in the recent discussion on this list. First, like other "natural" phonological models, it is very difficult for functional OT to explain " unnatural" phonological behavior. Juliette Blevins and Andrew Garret have recently shown for consonant harmony and metathesis that is is essentially impossible to explain why these processes have variant behavior on the basis of synchronic natural models. Instead, synchronic idiosyncracies result from regular, natural diachronic developments. The problem is not that there are no "natural" behaviors in synchronic phonology - the problem is that trying to explain all synchronic phonological properties by appealing to "natural" effects obscures an essential question: what properties of phonology are a direct result of (caused by, explained by) the synchronic grammar (including, perhaps functional constraints), and which are historically contingent (although historically functionally motivated)? Another questionable property of OT relates to issues brought up by Brian MacWhinney. OT has become an extremely powerful device. Such notions as Sympathy, Overlapping Constraints (Hayes), Output-Output correspondences (e.g. Benua), integration of phonological and extraphonological constraints, and levels of OT derivation have rendered OT essentially omnipotent. This makes it difficult to call it an explanatory theory. Of course, there are various attempts to reduce its power, but the real problem, I think, lies in the very notion that this OT grammar is actually explanatory of the language it models, independent of the "semantics" of the constraints. (Although I do recognize that Joan Bresnan disagrees with me here, I simply mean that there is no way in which the extra-constraint properties of functional constraints enter into their behavior within an OT model.). I would bet that most functionalists agree that whatever capacity for language individual speaker-hearers have, it is extremely powerful, and that certain restrictions are levied on what form a phonology can have by factors external to it. Some of these restrictions will come from the typical sources - processing constraints, neural architecture constraints, information processing constraints, physiological constraints, cognitive representation constraints, and so on. It might be a problematic step to reverse these roles as functional OT seems to - the grammar becomes explanatory, and the constraints have very little explanatory role. On the other hand, I couldn't be happier that the interface between phonetics and phonology is being closely scrutinized. Functional OT may very well provide the kind of theoretical framework required for good empirical work to get done. Ben Bergen ------------------------------------------- Benjamin K. Bergen Graduate Student Department of Linguistics U.C. Berkeley www.icsi.berkeley.edu/~bbergen ------------------------------------------- From chafe at HUMANITAS.UCSB.EDU Sat Dec 11 20:54:38 1999 From: chafe at HUMANITAS.UCSB.EDU (Wallace Chafe) Date: Sat, 11 Dec 1999 12:54:38 -0800 Subject: Inputs In-Reply-To: <3548217.3153811626@agate.psy.cmu.edu> Message-ID: I'd like to expand a little on the following point made by Brian MacWhinney: > 4. Even from the beginning, OT never questioned the need to provide a single > abstract underlying structure for each lexical item. As far as I can tell, > this commitment is the one that tends to lead linguistic theories away from > being able to develop psychological reality. I've worked a lot with two polysynthetic and highly "fusional" languages (Seneca and Caddo) that have obviously undergone a lot of phonological change. I'm convinced that the only way to explain/predict/understand the shapes of their words is by reconstructing earlier forms along with the changes that have made these forms what they are today. Now, earlier "derivational" approaches to phonology essentially mirrored what I've just described in "underlying forms" and "rules" (whatever deviations there were from historical shapes and processes were trivial). But it seems pretty silly to think that speakers of these languages "know" such systems in any realistic sense. Just how they do produce these words is a different and fascinating question, but it seems to involve very large memories combined with the ability to analogize. To put it all too briefly, OT has done away with the rules but kept the underlying forms in the guise of "inputs" to the constraints. So I'm left with the question of what those inputs could possibly mean when it comes to what's going on in the minds of speakers of these languages. It would be easy for me to give examples, which I've done a little in a paper called "How a Historical Linguist and a Native Speaker Understand a Complex Morphology" in the volume Historical Linguistics 1997: Selected Papers From the 13th International Conference on Historical Linguistics (Benjamins). I think this is the same problem that Brian was talking about. Wally Chafe From sylvester.osu at wanadoo.fr Sun Dec 12 08:24:50 1999 From: sylvester.osu at wanadoo.fr (Sylvester OSU) Date: Sun, 12 Dec 1999 09:24:50 +0100 Subject: Alternatives to Chomsky. A word to the initial Author Message-ID: Dear Bruce Richman, It's high time everyone realised that "killing" a man does not in any way "kill" his ideas. What does that really mean? Throwing Chomskyan method away? And what next? Alternatives to Chomsky. No. No. No. Chomsky has never been proclaimed the god of linguistic science, so why take him to be one. I thought we were doing some kind of objective work on languages. The truth is that Chomsky has made his contribution towards understanding the Human Language. If today we find a means of going somewhat further, then that's good. We push on. But we can't come back to square one everday 30 or 40 years and do as if nothing ever happened. Besides, I don't see anyway we can continue without making reference to his works (negatively or positively). So please, take it easy. Secondly, I'll like to remind you that other people have been working using other methods and proposing other frameworks. I refuse to believe that you are not aware of that anyway. Otherwise, why talk of proposing Alternatives to Chomsky? Or is it a provocative method of attracting people to your conference? Once again, Chomsky's method as far as I am concerned is just one among many others. Take for instance in Europe, many people have come out to propose some nice approaches to the study of Language and languages. Permit me here to simply name Antoine Culioli, a French linguist, and his research team, with their fantastic theory of predicative and enunciative operations. I bet you, it's something you should know about before going deeper in your conception of linguistic studies. Apart from Culiolian approach, you have Oswald Ducrot, you have Gustave Guillaume, Gross, Hagège, all French, who have proposed some different approaches. Think about it. Well, I just wish that over there, you don't think that the world outside is at a standstill. Try and get in touch, and you'll see that things are moving and that one approach has never and will never offer all the answers to all the problems that the study of Language raises. Should you need references from the Linguists I just mentioned, I'll be glad to submit them to you as quickly as possible. Wishing you well. Sylvester OSU. (France) ps: I'll like you to know that I'm not a Chomskyan and that I have only taken courses and read his works as any average linguist would do. Even, my mastery of his theories is just below what's expected of a contemporary linguist. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bresnan at CSLI.STANFORD.EDU Sun Dec 12 19:01:57 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Sun, 12 Dec 1999 11:01:57 -0800 Subject: OT--5 crucial strategic decisions that vitiate... In-Reply-To: Your message of Fri, 10 Dec 1999 10:47:06 EST. <3548217.3153811626@agate.psy.cmu.edu> Message-ID: Sorry I couldn't get to this till now. My real life called me.... This is a very long msg, because Brian MacWhinney has asked some very substantial questions. Brian MacWhinney wrote on Friday, 10 December: > However, as a psycholinguist, I have been disappointed by five crucial > "strategic" decisions in the development of OT that have tended to vitiate > its potential for constructing a psychologically plausible linguistic theory > of the type that Joan Bresnan and others have often sought. In particular, > 1. Early on, OT was supposed to be linked to connectionist modeling. > However, after the first few years, this linkage was largely dropped. Dedre > Gentner's interest in analogy as an acquisition or production mechanism has > pretty much suffered the same fate, I would guess. It is interesting to read what Prince and Smolensky say about this in their article "Optimality: From Neutral Networks to Universal Grammar" in the March 14, 1997 issue of Science Magazine. (available on-line: http://www.sciencemag.org/cgi/reprint/275/5306/1604): The principal empirical questions addressed by optimality theory, as by other theories of universal grammar, concern the characterization of linguistic forms in and across languages. A quite different question is, can we explicate at least some of the properties of optimality theory itself on the basis of more fundamental cognitive principles? A significant first step toward such an explanation, we will argue, derives from the theory of computation in neural networks. Linguistic research employing optimality theory does not, of course, involve explicit neural network modeling of language. The relation we seek to identify between optimality theory and neural computation must be of the type that holds between higher level and lower level systems of analysis in the physical sciences. ... Like thermodynamics, optimality theory is a self-contained higher-level theory; like statistical mechanics, we claim, neural computation ought to explain fundamental principles of the higher level theory by deriving them as large-scale consequences of interactions at a much lower level. They give some interesting examples where OT has been conceptually driven by the hypothesized underlying neural implementation, but point out that it doesn't explain the universal recurrence of linguistic constraints, so there's still a gap between the two approaches. I suspect that Paul and Alan have a deeper view of the relation to connectionist ideas than most generative linguists who have taken up OT, but *some* linguists have been curious enough about the underlying cognitive issues to walk part way across the bridge. I have heard Helen de Hoop, for example, start out a lecture on OT semantics by talking about tensor products... brave woman, intellectually valiant. > 2. Early on, OT constraints were supposed to have strength levels. However, > later on this feature was eliminated. In our work on the Competition Model, > Liz Bates and I learned how important strength levels are for describing and > predicting psycholinguistic data. In fact, one has to go beyond strength > levels for separate constraints and look at what we can conflict validity, > but none of this could possibly fit in with current OT. My colleague Edward Flemming (http://www.stanford.edu/~flemming/) makes a similar argument. He concludes his most recent paper, "Scalar Representations in a Unified Model of Phonetics and Phonology" as follows: The proposed model of phonetics and phonology is similar to Optimality Theoretic phonology in that outputs are selected so as to best satisfy conflicting, violable constraints. However, the constraints considered here (particularly implementations of minimization of effort and maximization of distinctiveness) trade-off against each other in an additive fashion, implying that these interactions are better modeled in a weighted constraint system rather than one which exclusively employs strict constraint dominance, as is the case for Optimality Theory. It is an interesting question whether the scalar-valued functions that play a role in assimilation and coarticulation are just what is needed for studying the typology and structure of, say, pronominal inventories or voice systems in syntax. But even if one doesn't go that far (into continuous modelling), the idea of optimizing symbolic structures using a discrete evaluation function defined over universal, violable constraints (like the markedness constraints of functional/typological linguistics) is a radical shift that seems promising, and creates an intellectual bridge where there wasn't one before. > 3. Early on, learning of a phonology (or grammar) could have involved the > strengthening and weakening of constraints. Later on, it required the types > of triggers used by G-B and P&P theories. ???? I find this a bit hard to understand in view of Tesar and Smolensky's 1998 article "Learnability in Optimality Theory" (Linguistic Inquiry 29, 229--68) (and their long technical report) which explicitly argues against the triggers learning models. (You can find a lot of these papers on Paul Smolensky's web page: http://www.cog.jhu.edu/faculty/smolensky.html.) However, I also like the new model of gradual learning of constraints whose ranking varies probablistically (see Boersma and Hayes, ROA, for references: http://ruccs.rutgers.edu/roa.html), even more. > 4. Even from the beginning, OT never questioned the need to provide a single > abstract underlying structure for each lexical item. As far as I can tell, > this commitment is the one that tends to lead linguistic theories away from > being able to develop psychological reality. Trying to preserve this > approach in OT models of syntax would be equally problematic. If you have a generative, derivational model of underlying structure, that is certainly true. It is hard to free ourselves from the kind of thinking we were originally trained in... and it may be that OT imported a bit too much of the derivational ways of thinking in generative grammar at the beginning. But we know much more now about nonderivational, constraint-based representations than we used to (especially those of us who have explored alternatives to Chomsky's grammatical architectures), and some OT work has a very different model of the nature and role of the input. (I would refer to several of my recent papers here, if I weren't so modest.) This is also true in phonology. I really like Rene Kager's new (1999) book, _Optimality Theory_ in the red textbook series of Cambridge University Press. It does a beautiful job of explaining some of the functionalist ideas and motivations for OT phonology, and showing some of the recent developments that do away with inputs in various ways... I would address this answer to Wally Chafe's message of December 11, as well. > 5. At no point was OT really committed to an account of online processing. > If problems 1-4 were not present, I would not consider this a fatal flaw, > since the notion of constraint ordering has clear interest for typology and > language change, at the very least. > This is a *very* difficult problem: how to do online computations comparing infinite sets of recursive symbolic structures? Here, the mathematically formalized representational theories of syntax (LFG, HPSG, categorial grammar, etc.) have some advantages, I think, over the Chomskyan syntactic approaches that get more press. Some new theoretical work has been done on this problem in OT-LFG: one, using conventional kinds of computation, is by Jonas Kuhn (http://www.ims.uni-stuttgart.de/~jonas/): Jonas Kuhn. 1999. Generation and Parsing in Optimality Theoretic Syntax---Issues in the Formalization of OT-LFG (draft version of November 1999. To appear in a CSLI volume _OT-LFG: Optimality Theory and LFG_, edited by Peter Sells. I can also point you to a remarkable paper on Optimality Theoretic LFG on Neuroidal Nets by Professor Tetsuro Nishino using a dynamic, parallel computational model to evaluate the OT-LFG candidate space: http://www.sw.cas.uec.ac.jp/tnlab/member/nishino.html. These works are specific to LFG, and the first I know of on generation and parsing in OT Syntax. But once a problem has been solved in LFG, it usually doesn't take too long to port it to HPSG and other constraint-based frameworks, ... :-) > I am just a psycholinguist, so I am happy to have linguists explain to me > how I have strayed in my judgments. But I would really like to hear some > open discussion of these issues. ... What irresistable modesty! And so unwarranted. You FUNKfolk are a pleasure to discuss things with. TTFN- Joan From macw at CMU.EDU Sun Dec 12 20:57:31 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Sun, 12 Dec 1999 15:57:31 -0500 Subject: rote, analogy and OT Message-ID: Dear Wally, Many thanks for your thoughtful response to my queries and concerns about OT. I believe I share your basic perspective. In my 1978 account of the acquisition of morphophonology in Hungarian, I relied largely on the mechanisms of rote and analogy which you (and Dedre Gentner) suggest. Of course, one of the big challenges facing connectionist theory is trying to cash out the meaning of an account based on analogy. It seems to me that OT might have some promise in this regard. However, there seem to be two core problems that have to be ironed out. I would guess that you and I would agree that rote and analogy play their largest role in the area of morphophonology. But, when we turn to issues in phonotactics, assimilatory processes, and metrical phonology, then basic OT-type constraints seem to move to the fore. So, how do we link up the mechanisms of rote and analogy with constraints in a single language system? We certainly want to avoid modularizing two systems that, in fact, far from separable. After all, many morphophonological processes are echoes of phonotactic and assimilatory processes. So, a better solution is to try to figure out how physiologically-grounded constraints impact complex word combinations. To do this, we need to have a system of graded (or scalar) constraints and a method for talking about their interaction. Ideally, OT would have provided this, but the jury is still out on whether strengths and interactions will be a part of OT. The second problem is the ongoing reliance on underlying form in OT. In my still not properly informed opinion (see my forthcoming email reply to Joan Bresnan), the basic problem in OT is its commitment to the doctrine of underlying form. If we build a system that is designed to work off of forms that have little relation to anything stored in the brain, we will be lacing our synchronic theories with diachronic commitments. I will read your paper from the Historical Linguistics volume as soon as I can. The question will be how to reshape OT in a way that solves these problems. Thanks again for the helpful comments. I will continue to think more about these issues. --Brian MacWhinney From pesetsk at MIT.EDU Sun Dec 12 21:03:23 1999 From: pesetsk at MIT.EDU (David Pesetsky) Date: Sun, 12 Dec 1999 16:03:23 -0500 Subject: Myhill's message Message-ID: Dear Funknet subscribers, We want to express our surprise and disappointment that no subscribers to this list have disassociated themselves from Myhill's optimistic anticipation of our colleague's death. We refer to the closing lines of his message sent three days ago: > Arguing with Chomsky is a waste of time. I'm all for alternative > approaches, but > Chomskyism is only going to die with Chomsky--and the man is 71 years old. The strongest rebuke that Myhill received was an invitation to visit Stanford on his sabbatical. Surely the readers of this list are not functionalists first and human beings second! Sincerely, Sabine Iatridou Michael Kenstowicz David Pesetsky From macw at CMU.EDU Sun Dec 12 21:37:13 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Sun, 12 Dec 1999 16:37:13 -0500 Subject: OT--5 crucial strategic decisions that vitiate... In-Reply-To: <199912121900.LAA00634@hypatia.Stanford.EDU> Message-ID: Dear Joan (and Funknet), Thank you for the extremely helpful reply. I have downloaded the various papers you mentioned from the OT archive. I had read the Tesar ones earlier, but I will reread them now. I realize that Tesar's learning algorithm for OT doesn't use triggers, but it does use single trial evidence to reorder constraints. The dominance ordering of constraints in his model is absolute, so there is no way in which the slow accumulation of data in favor of a particular pattern in a language can lead to the gradual accumulation in its strength. However, I believe that the real problem with the Tesar model is not the strict dominance assumption, but the commitment to requiring the child to learn an abstract underlying form. If the Boersma and Hayes model addresses this problem, as well as the strict dominance issue, that would be great. I recently finished Bernhardt and Stemberger's OT treatment of phonological development (primarily for English). Although the authors set up a nice framework for discussing phonological acquisition, I think that the mechanics of OT tend to make the actual presentation of constraint patterns a fairly tough job. At this point, I think it would be best for me to go "offline" and do my homework. I have contacted Flemming and Kuhn, asking for the relevant papers. I will read them, the OT files, and Kager (1999). In a sense, what is important is not whether OT has problems, but whether it is recognizing these problems and moving to address them. I found your letter extremely encouraging in this regard. However, it is now time for me to do some more reading. Until then. --Brian MacWhinney From bresnan at CSLI.STANFORD.EDU Sun Dec 12 21:45:50 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Sun, 12 Dec 1999 13:45:50 -0800 Subject: Myhill's message In-Reply-To: Your message of Sun, 12 Dec 1999 16:03:23 EST. Message-ID: Oh, come on, David! This is flame-bait, just like Myhill's post. Joan >>>David Pesetsky said: > Dear Funknet subscribers, > > We want to express our surprise and disappointment that no subscribers to > this list have disassociated themselves from Myhill's optimistic > anticipation of our colleague's death. > > We refer to the closing lines of his message sent three days ago: > > > Arguing with Chomsky is a waste of time. I'm all for alternative > > approaches, but > > Chomskyism is only going to die with Chomsky--and the man is 71 years old. > > The strongest rebuke that Myhill received was an invitation to visit > Stanford on his sabbatical. Surely the readers of this list are not > functionalists first and human beings second! > > Sincerely, > > Sabine Iatridou > Michael Kenstowicz > David Pesetsky > --------------------------------- Joan Bresnan From faucon at COGSCI.UCSD.EDU Sun Dec 12 22:37:01 1999 From: faucon at COGSCI.UCSD.EDU (Gilles Fauconnier) Date: Sun, 12 Dec 1999 14:37:01 -0800 Subject: Myhill's message Message-ID: Flame-bait or not, it's nothing for any of us to be proud of. Same goes for the post right after that from a long-time Chomsky admirer, boasting that he had given an entire talk without mentioning the Master's name. Gilles Fauconnier Center for Advanced Studies in the Behavioral Sciences 75 Alta Road Stanford CA 94305 USA tel. 650 - 321 2052 From bresnan at CSLI.STANFORD.EDU Sun Dec 12 23:19:46 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Sun, 12 Dec 1999 15:19:46 -0800 Subject: Myhill's message In-Reply-To: Your message of Sun, 12 Dec 1999 14:37:01 PST. <9912122243.AA10926@cogsci.UCSD.EDU> Message-ID: Gilles, I think you are wrongly accepting the truth of David's premise, that John Myhill's msg showed "optimistic anticipation of our colleague's death". This is ridiculous. I certainly didn't interpret Myhill's comment about the strength of Chomsky's ideas in this way, and I doubt that Fritz Newmeyer, Martin Maspelmath, Geoff Nathan, Tony Wright, Brian MacWhinney, or any of the other follow-up posters did, either. This is what I meant by "flame bait". It's not only ridiculous but deeply insulting to draw this inference from the substantive discussion about functionalism, OT, etc., that we were engaging in. I thought that Myhill was wrong to publicly disparage by name both a colleague of mine and by implication Stanford's entire linguistics department, and I directed my reply toward those points, trying to be good humored and constructive about it. It appears, though, that our friends at MIT are quite willing to believe that we all harbor death wishes for their colleague. That's a shame, and I find it hard to believe that it could be said in sincerity. If it was, then it shows what a long way we have to go to build some of the intellectual bridges I was talking about. Joan > > Flame-bait or not, it's nothing for any of us to > be proud of. > > Same goes for the post right after that from a > long-time Chomsky admirer, boasting that he had > given an entire talk without mentioning the > Master's name. > > Gilles Fauconnier > Center for Advanced Studies in the Behavioral Sciences > 75 Alta Road > Stanford CA 94305 > USA > > tel. 650 - 321 2052 > From macw at CMU.EDU Mon Dec 13 00:46:37 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Sun, 12 Dec 1999 19:46:37 -0500 Subject: Myhill's message Message-ID: Like others on FunkNet, I did not parse Myhill's message as an "optimistic anticipation of our colleague's death." However, I can see that it could be interpreted in this way. Perhaps we should wait until the end of the weekend and allow John to clarify his position. As my colleague Herb Simon would say, at 71, Chomsky is still a young man. Herb, who received the Nobel Prize in Economics back in his 40s, is now about 85 and is producing as much high quality intellectual output as ever. He comes in to work and teach everyday and seems to have no interest in ever really retiring. He is a great role model and a great thinker. I think we should also recognize the fact that the Chomskyan program will outlive the man. If we believe that there are reasonable alternatives to Chomsky's ideas, let's get on with the work of constructing them. --Brian MacWhinney From john at RESEARCH.HAIFA.AC.IL Mon Dec 13 08:00:28 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Mon, 13 Dec 1999 10:00:28 +0200 Subject: Myhill's message Message-ID: Thank you for defending me, Joan. I certainly did not intend my comment to be interpreted as an `optimistic anticipation of our colleague's death.' I was simply stating that the social phenomenon of Chomskyism--a phenomenon in which the beliefs of a single person, whose basic contribution to linguistics was made about 40 years ago and who, in my opinion, has not done much new of interest since then (with the exception of 'On wh-movement', which I think is a really great article with actual insights, although it is more than 20 years old), to a significant extent direct the course of the field, both in terms of blindly following on the one hand and serving as a rallying cry for 'alternatives to...' on the other hand--this phenomenon is not something which can be dealt with by a conference on 'alternatives'. There is obviously a magnetic appeal which the guy has on some people which can't be argued with rationally and which is going to be around as long as he is around. Let me draw a parallel with physics (SORRY, I know I've criticized other functionalists for doing things like this, but I'm talking to the formalists on this network now, and this is the kind of argument they like). Let's suppose that after 1910, `Einsteinism' developed as a phenomenon comparable to today's Chomskyism, AFTER Einstein had made what is now recognized as his contribution, but 48 years BEFORE he died. Since Einstein didn't care much for quantum mechanics ('God doesn't play dice'), quantum mechanics might not have developed. Everyone would have spent all their time arguing about Einstein's post-1910 theories, which don't seem to be leading to anything, or on organizing conferences on `alternatives to Einstein.' There wouldn't have been much progress. Fortunately that didn't happen. But that IS what's been happening to a large extent in syntax in the last 30 years. After spending my whole career as a linguist listening to arguments going nowhere and convincing no one on the merits or lack of such of the Chomskyan enterprise, I am firmly convinced that they are a waste of time. I did not intend what I wrote to be 'flame-bait'. I was addressing my message to functionalists, suggesting that we not waste time arguing with Chomskyists. Some formalists on the network got set off by this inter-functionalist dialogue and took what I said as provocative in a very different way than I had intended it. I mentioned Stanford because of the presence there of several people who have in one way or another directed their careers towards being 'alternatives to Chomsky' in a way which I have not found to be any better than the original. It was part of a general criticism of building a program around `alternatives to Chomsky.' Brian MacWhinney wrote: > However, as a psycholinguist, I have been disappointed by five crucial > "strategic" decisions in the development of OT that have tended to vitiate > its potential for constructing a psychologically plausible linguistic theory > of the type that Joan Bresnan and others have often sought. Well, gee, let's try to figure out why this may have happened. Could it possibly be that Chomsky isn't interested in such a theory, and his followers aren't interested in thinking for themselves? So when will this change? I guess when Chomsky isn't on the scene, right? Any other ideas when it will change? Any other ideas on why OT has made these decisions? And Joan's comment is quite to the point: It appears, though, that our friends at MIT are quite willing to believe that we all harbor death wishes for their colleague. That's a shame, and I find it hard to believe that it could be said in sincerity. If it was, then it shows what a long way we have to go to build some of the intellectual bridges I was talking about. Again, though, I'm afraid Joan is being overly optimistic. The problem (in this case) is not 'intellectual bridges.' The problem is the paranoid mindset of our friends at MIT and the degree of influence they have over the field. My statement touched a raw nerve, because, I would hypothesize, our friends at MIT fear they will be intellectually lost without Chomsky and can't imagine that they will one day have to think for themselves. Please note that I am by NO means saying that everyone to do with Chomsky should be excised from intellectual history, as other functionalists have suggested. By no means. Chomsky has made a great contribution with some of his articles, although this was a pretty long time ago. But the overall effect of Chomskyism, the personality cult which has built around him, has at the same time been intellectually debilating, if not devastating, to the field as a whole. And with regard to Brian's statement that: `I think we should also recognize the fact that the Chomskyan program will outlive the man'--it depends upon what is understood by 'the Chomskyan program'. Chomsky is not interested in `constructing a psychologically plausible linguistic theory', taking into consideration functional factors, and he never will be. He is resolutely, ideologically, intractably opposed to this. I consider this bias and blindless to be the essence of 'the Chomskyan program', as it has characterized this program from beginning to end. If Brian wants to make up his own version of 'the Chomskyan program' which leaves this bias out, then I would agree with his statement (though this definition of 'the Chomskyan program' seems bizarre to me). But if we understand 'the Chomskyan program' to mean a rejection of psychological plausibility, empirical data on usage, functional constraints, etc., as has been uniformly characteristic of Chomsky's writings, then I am not at all sure that this will survive him by much. I hope not. John From bresnan at CSLI.STANFORD.EDU Mon Dec 13 18:06:46 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 10:06:46 -0800 Subject: Myhill's message In-Reply-To: Your message of Mon, 13 Dec 1999 10:00:28 +0200. Message-ID: John, I apologize for characterizing your message as "flame bait". You are obviously sincere in your beliefs and not merely trying to elicit reactions from others. As I've tried to show, I think you may be mistaken in your judgements about the state of current linguistic research as being completely polarized into binary camps "pro-Chomsky" and "anti-Chomsky", or "formalist" and "functionalist". You seem to advocate simply writing off all work emanating from institutions or people that you characterize as "pro-Chomsky". (And there are those at MIT who have the same attitude toward you and the other "anti-Chomsky" posters to this list.) But your arguments do not refer to the substantive ideas and empirical claims that are made (as Brian MacWhinney's do, for example). Instead you refer to your judgments about the sociology of the field and even about the personalities and worth of individuals. I was pretty shocked when I saw how personal the remarks in your postings were. Neverthless, like it or not, there is a new functionalism attracting (among others) researchers who have been trained in formal methods and models, and there are new mathematical models for language making use of optimization and probability. The widespread availability of on-line corpora and powerful computers is also having its effect on attitudes toward data and methology: Stanford students working in syntax and semantics, for example, simply start doing corpus based research as a matter of course. I personally think that the most interesting work will hybridize the best from both worlds--research on symbolic representation (the work of the "formalists") and research on data driven cognitive modelling (the work of the "functionalists"), to put it very roughly. In other words, I think that substantive and serious discussions are possible between those on this list and those who you write off. There's no need for us to be so sectarian about our work! Joan From dryer at ACSU.BUFFALO.EDU Mon Dec 13 21:10:56 1999 From: dryer at ACSU.BUFFALO.EDU (Matthew S Dryer) Date: Mon, 13 Dec 1999 16:10:56 -0500 Subject: OT and functionalism Message-ID: The recent discussion of OT and functionalism misses what to me is an essential difference between OT and the relation of functional explanation, particularly competing motivations, to grammar. In OT, as I understand it, the constraints are "in" the grammar, either in the sense that the constraints are stipulated for particular languages, or in the sense that the grammar of the language makes reference to universal constraints, for example in specifying which constraints outrank which other constraints. But on my view, grammars of languages do not make reference to functional principles or motivations. For example, if two languages have a difference that one might describe in terms of economic motivation competing with iconic motivation, where economic motivation wins out in one language and iconic motivation in the other, I would not want to say that there are grammatical rules in these languages that refer to economy or to iconicity. To put the point another way, functional principles and motivations apply primarily at the level of the evolution of particular languages. If economic motivation has won out over iconic motivation with respect to some aspect of the grammar of a particular language, that means that over the past few thousand years, the particular grammatical changes that have led to the current state of the language reflect the influence of that functional principle. But once that has happened, the grammatical rules have an existence that is independent of the explanatory principles that have influenced them. OT shares with theories in the GB tradition of trying to build explanation for why languages are the way they are into the grammars themselves. From my perspective, this is deeply misguided. Languages are highly complex systems, whose properties reflect the interaction of diverse explanatory principles, and I believe that they can and should be described in terms that are independent of the explanatory principles that underlie them. While there are a wide variety of approaches which have been labeled "functionalist", OT as I understand it is fundamentally incompatible with the type of functionalism that I and many others espouse. Matthew Dryer From fjn at U.WASHINGTON.EDU Mon Dec 13 22:45:21 1999 From: fjn at U.WASHINGTON.EDU (Frederick Newmeyer) Date: Mon, 13 Dec 1999 14:45:21 -0800 Subject: OT and functionalism Message-ID: I feel that Matthew Dryer makes an excellent point when he writes: > functional principles and motivations apply > primarily at the level of the evolution of particular languages. If > economic motivation has won out over iconic motivation with respect to > some aspect of the grammar of a particular language, that means that over > the past few thousand years, the particular grammatical changes that have > led to the current state of the language reflect the influence of that > functional principle. But once that has happened, the grammatical rules > have an existence that is independent of the explanatory principles that > have influenced them. As I will try to argue now as concisely as possible, the kind of 'functionalist OT model' that Joan Bresnan has been advocating in her postings (if I understand her correctly), in which the constraints themselves are directly functionally motivated, falters when one tries to apply it to a wide variety of disparate phenomena within a particular language. OT analyses, particular in the realm of syntax, have, in general, been rather circumscribed in their domain of application. Typically, they focus on some little corner of the syntax, such as clitic order, auxiliary inversion, and so on. The machinery intrinsic to the OT approach works beautifully in such situations. But grammars are tightly integrated wholes, so decisions about how one process should be handled are likely to have repercussions for another, seemingly unrelated, process. This fact leads ultimately to a much greater disparity between grammatical principles and their functional roots than is posited by 'functionalist OT models'. Let's posit two functionally-motivated OT constraints for English, ICONICITY and HEAVY-LAST: ICONICITY: Syntactic constituents reflect semantic units. HEAVY-LAST: Heavy constituents follow light constituents. ICONICITY can be illustrated by the fact that, in every formal account, adjectives are generated under the same phrasal node as the noun that they modify. HEAVY-LAST can be illustrated by the fact that within the verb phrase, sentential complements are positioned after phrasal complements. The functional roots of these constraints are so obvious that no discussion is necessary. Now, then, which constraint is ranked higher for English, ICONICITY or HEAVY-LAST? Well, it depends. In some cases, we have identical grammatical elements in variant orders with no meaning difference, each option corresponding to a different ranking of the two constraints. For example, both of the following are grammatical English sentences: (1) a. A man who was wearing a silly-looking red hat dropped by today. b. A man dropped by today who was wearing a silly-looking red hat. Sentence (1a) reflects a ranking of ICONICITY over HEAVY-LAST; sentence (1b) the reverse ranking. In some cases, however, only a higher ranking of ICONICITY is possible. Simple adjective phrases cannot be extraposed from the nouns that they modify, no matter how heavy they are: (2) a. An extremely peculiar-looking man dropped by today. b. *A man dropped by today extremely peculiar-looking. And in other cases, HEAVY-LAST seems to be ranked over ICONICITY. When comparatives are used attributively, the adjective is separated from its complement by the head noun, despite the fact that together they serve to modify semantically that head noun (3a-b). Their structural unity can be obtained only in a manner that is consistent with HEAVY-LAST (3c): (3) a. That's a more boring book than any I have ever read. b. *That's a more boring than I have ever read book. c. That's a book more boring than any I have ever read. What is the solution to this ranking paradox? Only, I would say, to abandon ICONICITY and HEAVY-LAST as constraints internal to English grammar. Rather, what we need are something much more like the 'parochial' rules, principles, and constraints of standard models of grammar that interact to yield the grammatical sentences of the language. That move, however, is incompatible with 'functionalist OT models'. The relationship between detailed grammatical statements --- such as those that position the constituents of NP --- and their functional motivations can be extremely indirect. In other words, we are left with the sort of view advocated in Newmeyer (1998), in which, in a global / historical sense, grammars reflect external forces, but without each language-internal grammatical statement being tied to a particular functional motivation. Newmeyer, Frederick J. (1998). Language form and language function. Cambridge, MA: MIT Press. --fritz newmeyer From sstraigh at BINGHAMTON.EDU Tue Dec 14 04:17:32 1999 From: sstraigh at BINGHAMTON.EDU (H Stephen Straight) Date: Mon, 13 Dec 1999 23:17:32 -0500 Subject: OT, Functionalism, and the Myth of G Message-ID: The problem with Chomskyan (and post-Chomskyan, and the preponderance of pre-Chomskyan) theorizing derives not from over-formalizing, under-empiricizing, or most of the other errors identified in the current FUNKNET discussion. What it all boils down to is belief in the Myth of G. Brian Macwhinney points us in the right direction by urging us to reject the idea that each linguistic event derives from a single underlying form. Underlying forms derive in turn from a Grammar, in which each such form serves as the key to an actual or potential linguistic event, whether as a product of language production or an object of language perception. Let me attempt to debunk this pernicious Myth of G by critiquing Fritz Newmeyer's helpful contribution to this awesome FUNKNET thread. Fritz reasonably (and pro-functionally) asks us to "posit two functionally-motivated OT constraints for English [:] ICONICITY: Syntactic constituents reflect semantic units. HEAVY-LAST: Heavy constituents follow light constituents. ICONICITY can be illustrated by the fact that, in every formal account, adjectives are generated under the same phrasal node as the noun that they modify. HEAVY-LAST can be illustrated by the fact that within the verb phrase, sentential complements are positioned after phrasal complements." Fritz then poses the (false, OT-motivated) question, "Now, then, which constraint is ranked higher for English, ICONICITY or HEAVY-LAST?" Because Fritz, like OT, has only G to turn to, he concludes that "In some cases, we have identical grammatical elements in variant orders with no meaning difference, each option corresponding to a different ranking of the two constraints." But Fritz's supposedly conclusive examples mix receptive and expressive parameters: > For example, both of the following are grammatical English > sentences: > > (1) a. A man who was wearing a silly-looking red hat dropped by > today. > b. A man dropped by today who was wearing a silly-looking red hat. > > Sentence (1a) reflects a ranking of ICONICITY over HEAVY-LAST; sentence > (1b) the reverse ranking. In some cases, however, only a higher ranking of > ICONICITY is possible. Simple adjective phrases cannot be extraposed from > the nouns that they modify, no matter how heavy they are: > > (2) a. An extremely peculiar-looking man dropped by today. > b. *A man dropped by today extremely peculiar-looking. Omitted from Fritz's purview is the fact that 2b works just fine with a healthy, listener-warning pause: c. A man dropped by today ... extremely peculiar-looking. Also omitted is the fact that the explicit relativizer present in examples (1) is in fact optional: (1) c. A man wearing a ... d. A man dropped by today wearing ... So what can we make of these facts? Nothing about Grammar, if it is assumed to be neutral with regard to the demands of speakers and listeners. Fritz appeals to G-specific arbitrarinesses specifiable only in non-functional (i.e., formal) terms, though given the counter-examples I don't think this tactic will save him. However, if we instead see these competing demands as resulting in a gradient of preferences weighted differently for the domain of the receiver versus the domain of the producer, and perhaps even competing within each of these domains, we (rightly) conclude that one and the same "sentence" (a term already unfairly biased toward the G side of this issue) can be OK or problematic for the one but bad or differently problematic for the other. Take for example Fritz's frightful "*That's a more boring than I have ever read book." If we consider that this example violates the pragmatic (functional?) OLD-BEFORE-NEW principle because "book" is clearly topic rather than comment, and that it also ignores counterexamples such as "Thank you for the much needed by a teenager present" (attested in my own experience), we can conclude that Fritz has allowed his asterisks to lull him into complacency about the unidimensionality of grammatical judgments. Fritz rightly criticizes "functionalist OT models" for trying to use functionalist principles to predict grammaticality judgments, but what's wrong is not the principles but rather the judgments: They are a mixed and inconclusive source of evidence because language users employ a wide variety of different criteria to arrive at such (often demonstrably ephemeral) judgments. Fritz's conclusion that "grammars reflect external forces, but without each language-internal grammatical statement being tied to a particular functional motivation" sums up the Myth of G in a manner both ironic and poignant: "Language-internal" statements amount to no more than the unidimensional "prescriptive" judgments of popular grammarians of the likes of Kilpatrick, Newman, and Simon, judgments long since debunked by linguists from Sledd to Bolinger to Pinker. Why can't linguists admit once and for all the uncertainty and contradictoriness of their data on "grammaticality"? When will linguists stop living in armchairs and consider seriously the evidence provided by corpora, elicitation, and experiment? We say things no one can (easily) parse, and we (readily) parse things no one (hardly ever) says. Shouldn't these facts (along with a myriad of others) tell us something really basic about the (non-G) nature of construing and saying? Best. 'Bye. Steve ________________________________________________________________ H Stephen Straight - straight at binghamton.edu From bresnan at CSLI.STANFORD.EDU Tue Dec 14 05:22:25 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 21:22:25 -0800 Subject: OT and functionalism In-Reply-To: Your message of Mon, 13 Dec 1999 16:10:56 EST. Message-ID: In his posting of Monday, 13 Dec., Matthew Dryer makes the point that functional (principles, motivations, or) constraints are external to grammars, while OT constraints are internal to (OT) grammars. Hence, he believes: > OT shares with theories in the GB tradition of trying to build explanation > for why languages are the way they are into the grammars themselves. From > my perspective, this is deeply misguided. Languages are highly complex > systems, whose properties reflect the interaction of diverse explanatory > principles, and I believe that they can and should be described in terms > that are independent of the explanatory principles that underlie them. > > While there are a wide variety of approaches which have been labeled > "functionalist", OT as I understand it is fundamentally incompatible with > the type of functionalism that I and many others espouse. This certainly sounds like an unassailable position, but that is because it is true by definition. If you *define* a functional constraint to be something external to a "grammar", and you *define* an OT "grammar" as something that contains internal constraints, then how could his conclusion be other than true? But I think Matthew's point betrays a misconception about what exactly an OT grammar is. He elaborates: > But on my view, grammars of languages do not make reference to functional > principles or motivations. For example, if two languages have a > difference that one might describe in terms of economic motivation > competing with iconic motivation, where economic motivation wins out in > one language and iconic motivation in the other, I would not want to say > that there are grammatical rules in these languages that refer to economy > or to iconicity. OT has no grammatical rules that refer to economy or to iconicity. Language particularity is ultimately simply a harmonic function over the space of possible forms. If you look inside an OT "grammar", you find a representational basis (this is sometimes called "GEN") which specifies the set of possible structures, and an optimizing component (called "EVAL") which optimizes the candidate structures against the conflicting universal constraints in such a way as to minimize violations. The constraint component is "external" to the structure component. He writes: > To put the point another way, functional principles and motivations apply > primarily at the level of the evolution of particular languages. If > economic motivation has won out over iconic motivation with respect to > some aspect of the grammar of a particular language, that means that over > the past few thousand years, the particular grammatical changes that have > led to the current state of the language reflect the influence of that > functional principle. But once that has happened, the grammatical rules > have an existence that is independent of the explanatory principles that > have influenced them. As I indicated in my reply to J. MacFarlane's message, the work of "new functionalist" OT phonologists like Donca Steriade is based on the hypothesis that factors such as perceptibility and avoidance of articulatory effort "play a role in shaping sound patterns, not only in an evolutionary sense--as argued by John Ohala and Bjorn Lindblom--but also by defining the grammatical constraints whose interactions yield the phonologies of individual languages." Judith Aissen and I believe that the same may be true in syntax: The same functionally motivated constraints or "forces" that shape evolutionary properties and appear in crosslinguistic typological asymmetries also emerge within the grammars of particular languages. We have several papers making this point in the domain of syntax in some detail, and these are available from our Optimal Typology Project website (http://www-ot.stanford.edu/ot/), so I won't repeat the evidence and arguments here. This is what we are trying to do in our project. I will only give one very simple example (from Rene Kager's 1999 CUP textbook on Optimality Theory). Across languages, voiced obstruents are typologically marked: there are languages having only voiceless obstruents (e.g. Polynesian), and languages having both voiced and voiceless obstruents (e.g. English), but no (or hardly any) languages having only voiced obstruents. This typological asymmetry might be explained in terms of constraints on articulatory effort and perception that shaped the evolution of languages to explain why voiced obstruents are more restricted crosslinguistically. But within the phonological components of the grammars of particular languages we see the same thing: voiced obstruents are restricted in their *language-internal* distributions. For example, Dutch and German have a voicing contrast in obstruents, which appears in syllable onsets--a very salient position-- but voiced obstruents are devoiced in syllable codas. In contrast, there appear to be few or no languages whose language-internal phonologies have a voicing contrast in syllable codas and no voicing contrast in syllable onsets. OT can explain the positional neutralization of obstruent voicing in syllable codas and relate it to the typological of segment inventories across languages. The reason is simply that the *same* constraints reflecting ease of perception and articulatory effort are present in the grammars of all languages (as the universal constraint set). No other of grammatical theory that I know of does this. In my work--and in Judith's--we are developing analyses of syntactic analogues of these phonological phenomena. (I refer you to the papers at the website above.) I think it's extremely exciting, and has potential for truly integrative collaborations across the formal/functional divide. Of course, at the level of syntax it is much harder to find direct perceptual and cognitive evidence for the markedness constraints, compared to the work on phonetic perception and articulation that Steriade and others appeal to. There is much to be done. I hope that Matthew's statement of what a true functionalist believes isn't the last word on this topic. Joan From bresnan at CSLI.STANFORD.EDU Tue Dec 14 05:24:02 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 21:24:02 -0800 Subject: OT and functionalism In-Reply-To: Your message of Mon, 13 Dec 1999 14:45:21 PST. Message-ID: Isn't this fun? Too bad grades are due tomorrow. I will be brief in my reply to Fritz's message, since it piggybacks on Matthew's anyway, to which I replied at length. Fritz thinks that OT doesn't mesh with his conception of what functionalism is, namely-- "... the sort of view advocated in Newmeyer (1998), in which, in a global / historical sense, grammars reflect external forces, but without each language-internal grammatical statement being tied to a particular functional motivation. Newmeyer, Frederick J. (1998). Language form and language function. Cambridge, MA: MIT Press." My previous reply to Matthew Dryer applies just as well here, so I won't repeat it. But Fritz argues for this point in a peculiar way, by giving a failed OT analysis of his own design, and blaming it on the framework. (Hmmm... We certainly wouldn't accept this reasoning from a student, would we? ...sorry, I have grades on my mind. :-)) He also makes several very general characterizations about OT work, such as: "OT analyses, particular in the realm of syntax, have, in general, been rather circumscribed in their domain of application. Typically, they focus on some little corner of the syntax, such as clitic order, auxiliary inversion, and so on." ---and: "The problem with the framework seems to be that it "falters when one tries to apply it to a wide variety of disparate phenomena within a particular language." But Fritz hasn't really shown us this, has he? It's a bit odd for Fritz to talk about grammatical coverage, since OT syntax is still in its infancy, and the only formal grammatical frameworks that have produced genuine wide coverage grammars of real languages are the constraint-based lexicalist grammars in LFG, HPSG, etc. which are actively used in large-scale computational NLP projects and applications. (I am told that the Minimalist Program has advanced by shrinking its coverage of a wide variety of disparate phenomena.) It might be of interest to those concerned about wide-scale descriptive grammar coverage and OT that Jonas Kuhn (whose work I referred to in previous postings) has implemented the Grimshaw fragment of English syntax (on auxiliary inversion) in the XLE parser at Stuttgart. This is, of course, a toy fragment. But his formal analysis of the generation and recognition problem in OT-LFG now makes it possible to generate, parse, and study real grammars of the sort that have broad descriptive coverage. This is a remarkable advance. Have a look at: http://www.ims.uni-stuttgart.de/~jonas/ This is just the beginning, but the new work coupling the mathematically well defined group of constraint based grammars with the OT model of constraint interaction seems promising to me. TTFN-- Joan P.S. Lest anyone get the impression that I am an advocate for OT, let me repeat my warning that I am only a student of OT, and I'm only doing it in my spare time (of which there isn't much). *---------------------------------------- ______ __o __o Joan Bresnan bresnan at stanford.edu ______ _`\<,_ _`\<,_ *---------------------------------------- ______ (*)/ (*) (*)/ (*) From bresnan at CSLI.STANFORD.EDU Tue Dec 14 06:13:21 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 22:13:21 -0800 Subject: Chomsky bashing Message-ID: Several of my friends at MIT have privately told me that they consider it inappropriate that I spoke up against the public disparagement of Tom Wasow and Stanford by a post-er to this list, and yet said nothing about the disparagement of Noam Chomsky in the same message. It didn't occur to me that I should speak up for MIT as well as Stanford, since they are certainly an articulate bunch there at MIT and could reply even better than I. But they point out that since I am the current President of the LSA, my silence in regard to some of the Chomsky bashing that passes for discussion on this list seems signficant. I would therefore like to urge the members of this list to consider that we are all human beings, even the most famous (and the least famous!) among us. It is good fun to have a lively interplay of ideas and arguments, and we all enjoy witticisms (especially at others' expense)---but isn't it more enlightening to respect each other enough to listen and answer with civility? I have learned a lot from the intelligent postings and queries on this list in reponse to my recent messages, and I very much respect those of you who took the trouble to formulate such thoughtful and interesting questions and criticisms. I also have great respect for Noam Chomsky, who was my teacher. Even John Myhill's very high regard for Noam Chomsky is evident in his postings--comparing Chomsky in linguistics to Einstein in physics, referring to those of us who have explored alternative grammatical architectures as "not as good as the original". Chomsky is the original, and we are all of us his students, directly or indirectly. Joan P.S. That "we are most of us" construction in my last sentence surely shows that the V-movement analysis of Q-float under the internal-VP subject analysis can't be right. *---------------------------------------- ______ __o __o Joan Bresnan bresnan at stanford.edu ______ _`\<,_ _`\<,_ *---------------------------------------- ______ (*)/ (*) (*)/ (*) From beaumar at HOTMAIL.COM Tue Dec 14 06:44:12 1999 From: beaumar at HOTMAIL.COM (A R) Date: Mon, 13 Dec 1999 22:44:12 PST Subject: Chomsky Message-ID: You know, Funknetters, I don't know that I'd call myself either a formalist or a functionalist. But one of the most damning things I know about Chomsky is that he *never* goes to LSA, the one place where he might face genuine criticism. A friend of mine, a formalist, used to argue that he simply never went to academic conferences. This year, I discover that this is clearly untrue. Chomsky is going to talk at the MLA, to those who do literature. It's only his own that he refuses to talk to at conferences. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From john at RESEARCH.HAIFA.AC.IL Tue Dec 14 07:29:30 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Tue, 14 Dec 1999 09:29:30 +0200 Subject: common ground Message-ID: You may be right, Joan. You may be right. For the first 15 years of my linguistics career, I dutifully read formal linguistics articles of all types, in approaches such as your LFG or Simon Dik's functional grammar which claimed to be something new, something integrating formal and functional findings and results, to an extent I find hard to believe any other functionalist could stand (with the exception of Matthew Dryer). I have thought about formal issues enough that I can say that, e.g., I have been more influenced by 'On Wh-movement' than by anything else Chomsky has written, more by Burzio than by Rizzi, etc. I have never refrained from referring to formal articles or even arguments when they were relevant to my research. I even once gave a straight LFG paper on Indonesian at the LSA. But none of it, NONE of it, not LFG, not FG, not Construction Grammar, escaped from the same conceptual trap. So I gave up attempting to keep up; five years ago, after yet another 'Chomskyan revolution' which was more of the same, I deduced that it was a waste of time and decided to devote my research time to other things. I have been listening to you saying for nigh-on 20 years that you-all are doing something 'really different' from Chomsky, I spent a lot of time investigating whether it was true, and I came to the conclusion that it simply wasn't (which was what Chomsky himself maintained as well). But I have fallen behind. You may be right. This time there may really be a wolf. It is quite possible. But it will be hard for me to convince myself to take the time out from other things to check it out for myself now. But maybe some day... John >John, I apologize for characterizing your message as "flame bait". >You are obviously sincere in your beliefs and not merely trying to >elicit reactions from others. > >As I've tried to show, I think you may be mistaken in your judgements >about the state of current linguistic research as being completely >polarized into binary camps "pro-Chomsky" and "anti-Chomsky", or >"formalist" and "functionalist". You seem to advocate simply writing >off all work emanating from institutions or people that you >characterize as "pro-Chomsky". (And there are those at MIT who have >the same attitude toward you and the other "anti-Chomsky" posters to >this list.) But your arguments do not refer to the substantive ideas >and empirical claims that are made (as Brian MacWhinney's do, for >example). Instead you refer to your judgments about the sociology of >the field and even about the personalities and worth of individuals. >I was pretty shocked when I saw how personal the remarks in your >postings were. > >Neverthless, like it or not, there is a new functionalism attracting >(among others) researchers who have been trained in formal methods and >models, and there are new mathematical models for language making use >of optimization and probability. The widespread availability of >on-line corpora and powerful computers is also having its effect on >attitudes toward data and methology: Stanford students working in >syntax and semantics, for example, simply start doing corpus based >research as a matter of course. > >I personally think that the most interesting work will hybridize the >best from both worlds--research on symbolic representation (the work >of the "formalists") and research on data driven cognitive modelling >(the work of the "functionalists"), to put it very roughly. > >In other words, I think that substantive and serious discussions are >possible between those on this list and those who you write off. >There's no need for us to be so sectarian about our work! > >Joan From john at RESEARCH.HAIFA.AC.IL Tue Dec 14 07:42:32 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Tue, 14 Dec 1999 09:42:32 +0200 Subject: Chomsky Message-ID: Relating to the comment from AR (beaumar at HOTMAIL.COM, copied below), my father, a mathematical logician of the same name as me, met and spent a year around Chomsky at Princeton in the late 1950's, when Chomsky was attempting to pass himself off as a mathematician. The real mathematicians there observed with amusement that Chomsky would attempt to act like a mathematician, throwing around terms designed to impress, when speaking to philosophers, political scientists, etc., but when a real mathematician entered the conversation, Chomsky would either making an excuse to leave or change the topic to something like politics. John >You know, Funknetters, I don't know that I'd call myself either >a formalist or a functionalist. But one of the most damning things >I know about Chomsky is that he *never* goes to LSA, the one place where he >might face genuine criticism. A friend of mine, a formalist, >used to argue that he simply never went to academic conferences. > >This year, I discover that this is clearly untrue. Chomsky is going >to talk at the MLA, to those who do literature. It's only his own that he >refuses to talk to at conferences. > > >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com From haspelmath at EVA.MPG.DE Tue Dec 14 11:08:38 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Tue, 14 Dec 1999 11:08:38 +0000 Subject: optimality in synchrony and diachrony Message-ID: Joan, Matthew and Fritz, Thanks a lot for this truly illuminating discussion on the place of functional constraints in grammatical explanation. I think Matthew formulated the most coherent functionalist position very clearly: functional forces operate in language change, and should not be part of grammatical description (functionalist pronouncements have often been less clear, so that Fritz understandably citicizes functionalists for "stating functional forces in the grammar itself", 1998:141). In this debate, my tendency is to side with Matthew and Fritz. In my paper "Optimality and diachronic adaptation" (ROA-302-0399) I argued in particular that the diachronic dimension is necessary for true explanation. (This paper will appear in Zeitschrift fuer Sprachwissenschaft, with peer commentary by Newmeyer, Croft, Traugott, Wurzel, Dahl, Haider, Kirby, and others.) I am not so sure, however, that both sides could not be right, i.e. that functional factors are crucial for shaping grammars in diachronic change, AND that the best synchronic description is one that uses these constraints, perhaps in an OT-like fashion. Matthew, why would you exclude this possibility? (We have seen Fritz's answer, but as Joan pointed out, it is quite incomplete.) Joan, I must confess that I didn't understand your remarks that address Matthew's claim that OT "is trying to build explanation for why languages are the way they are into the grammars themselves": >OT has no grammatical rules that refer to economy or to iconicity. >Language particularity is ultimately simply a harmonic function over >the space of possible forms. If you look inside an OT "grammar", you >find a representational basis (this is sometimes called "GEN") which >specifies the set of possible structures, and an optimizing component >(called "EVAL") which optimizes the candidate structures against the >conflicting universal constraints in such a way as to minimize >violations. The constraint component is "external" to the structure >component. Are you saying that functional OT constraints are neither "built into the grammars themselves" nor are restricted to diachrony? Could you please clarify this? I think I didn't understand the thrust of Joan's example about the devoicing tendency across languages and in particular languages either. This is something that typologists have long been aware of (see e.g. Bill Croft's markedness chapter in his typology textbook). The usual functionalist explanation for both cross-linguistic patterns and language-particular regularities is that they show the effect of diachronic change. Martin -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From jlmendi at POSTA.UNIZAR.ES Tue Dec 14 12:24:57 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Tue, 14 Dec 1999 13:24:57 +0100 Subject: Myhill on Chomsky Message-ID: John Myhill gave us the following information: >Relating to the comment from AR (beaumar at HOTMAIL.COM, copied below), my >father, a mathematical logician of the same name as me, met and spent a >year around Chomsky at Princeton in the late 1950's, when Chomsky was >attempting to pass himself off as a mathematician. The real mathematicians >there observed with >amusement that Chomsky would attempt to act like a mathematician, throwing >around terms designed to impress, when speaking to philosophers, political >scientists, etc., but when a real mathematician entered the conversation, >Chomsky would either making an excuse to leave or change the topic to >something like politics. > >John If this kind of information is considered useful in order to evaluate the validity of the Chomskyan program, I think the following will serve as well: "For years now, Chomsky has been one of the very few scientists and philosophers who is widely read. The number of citations of his work in the Arts and Humanities Citation Index (nearly 4.000 between 1980 and early 1992) makes him the most cited living person and the eighth overall (...), and the citations of his work in the Social Science Citation Index (7.499 between 1972 and early 1992) are likely to make him the most cited living person there as well. Last, but certainly not least, from 1974 to 1992 he was cited 1.619 times in the Science Citation Index" (I quote from the Foreword to _Noam Chomsky. Critical Assessments_, Routledge, London & New York, 1994: p. xxii) It seems quite exaggerated for a false mathematician! Best Regards, Jose-Luis From john at RESEARCH.HAIFA.AC.IL Tue Dec 14 13:57:27 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Tue, 14 Dec 1999 15:57:27 +0200 Subject: corpus based research at Stanford Message-ID: Joan (Bresnan), you wrote: `Stanford students working in syntax and semantics, for example, simply start doing corpus based research as a matter of course.' Is this really true? Who's teaching these classes? Have you committed tenure-track faculty lines to people trained in corpus-based research? Can you give me references to relevant papers written by such people or their professors? This sounds interesting. John From reich at chass.utoronto.ca Tue Dec 14 18:20:36 1999 From: reich at chass.utoronto.ca (Peter A. Reich) Date: Tue, 14 Dec 1999 13:20:36 -0500 Subject: Myhill on Chomsky Message-ID: I'm sure God is cited more than Chomsky (I will not engage in a discussion as to whether God is Living), and probably Bill Gates is mentioned more than Chomsky as well in the popular media. And Clinton. So what? It is perfectly respectable in Canada, at least, to be an Atheist, and to prefer Linux or Macintosh. And, in the US, to be a Republican. Eat shit; 50 billion insects can't be wrong.--Peter Reich "For years now, Chomsky has been one of the very few scientists and philosophers who is widely read. The number of citations of his work in the Arts and Humanities Citation Index (nearly 4.000 between 1980 and early 1992) makes him the most cited living person and the eighth overall (...), and the citations of his work in the Social Science Citation Index (7.499 between 1972 and early 1992) are likely to make him the most cited living person there as well. Last, but certainly not least, from 1974 to 1992 he was cited 1.619 times in the Science Citation Index" (I quote from the Foreword to _Noam Chomsky. Critical Assessments_, Routledge, London & New York, 1994: p. xxii) It seems quite exaggerated for a false mathematician! Best Regards, Jose-Luis Mendivil Giro From bresnan at CSLI.STANFORD.EDU Wed Dec 15 01:58:52 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 17:58:52 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Tue, 14 Dec 1999 15:57:27 +0200. Message-ID: In addition to Tom Wasow's reply concerning his own corpus-based research, I would point you to Chris Manning, a new member of our department (http://www.stanford.edu/~manning/). Manning maintains a very useful and well-regarded web list of resources for statistical and corpus-based natural language processing: http://www.sultry.arts.usyd.edu.au/links/statnlp.html In addition, Anne Copestake, Senior Researcher at CSLI, provides a wealth of technical and practical experience available to our student researchers. Joan Bresnan > > `Stanford students working in syntax and semantics, for example, simply > start doing corpus based research as a matter of course.' > > Is this really true? Who's teaching these classes? Have you committed > tenure-track faculty lines to people trained in corpus-based research? Can > you give me references to relevant papers written by such people or their > professors? > This sounds interesting. > > John > From ellen at CENTRAL.CIS.UPENN.EDU Wed Dec 15 05:20:30 1999 From: ellen at CENTRAL.CIS.UPENN.EDU (Ellen F. Prince) Date: Wed, 15 Dec 1999 00:20:30 EST Subject: corpus based research at Stanford Message-ID: >In addition to Tom Wasow's reply concerning his own corpus-based >research, I would point you to Chris Manning, a new member of our >department (http://www.stanford.edu/~manning/). Manning maintains a >very useful and well-regarded web list of resources for statistical and > corpus-based natural language processing: > >http://www.sultry.arts.usyd.edu.au/links/statnlp.html > >In addition, Anne Copestake, Senior Researcher at CSLI, provides a >wealth of technical and practical experience available to our student >researchers. > >Joan Bresnan In addition, you've got John Rickford... :-) >> >> `Stanford students working in syntax and semantics, for example, simply >> start doing corpus based research as a matter of course.' >> >> Is this really true? Who's teaching these classes? Have you committed >> tenure-track faculty lines to people trained in corpus-based research? Can >> you give me references to relevant papers written by such people or their >> professors? >> This sounds interesting. >> >> John >> - --AAA10908.945235077/central.cis.upenn.edu-- ------- End of Forwarded Message From bresnan at CSLI.STANFORD.EDU Wed Dec 15 06:38:43 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 22:38:43 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Wed, 15 Dec 1999 00:20:30 EST. <199912150520.AAA10973@central.cis.upenn.edu> Message-ID: Yes, thank you. And we've got Penny Eckert, and we've got Beth Levin... I think that more corpus-based research from more perspectives on language is possible at Stanford than almost anywhere. (Penn being an exception to the rule... :-) J. >>>"Ellen F. Prince" said: > >In addition to Tom Wasow's reply concerning his own corpus-based > >research, I would point you to Chris Manning, a new member of our > >department (http://www.stanford.edu/~manning/). Manning maintains a > >very useful and well-regarded web list of resources for statistical and > > corpus-based natural language processing: > > > >http://www.sultry.arts.usyd.edu.au/links/statnlp.html > > > >In addition, Anne Copestake, Senior Researcher at CSLI, provides a > >wealth of technical and practical experience available to our student > >researchers. > > > >Joan Bresnan > > In addition, you've got John Rickford... > > :-) > > > >> > >> `Stanford students working in syntax and semantics, for example, simply > >> start doing corpus based research as a matter of course.' > >> > >> Is this really true? Who's teaching these classes? Have you committed > >> tenure-track faculty lines to people trained in corpus-based research? Can > >> you give me references to relevant papers written by such people or the ir > >> professors? > >> This sounds interesting. > >> > >> John > >> > > - --AAA10908.945235077/central.cis.upenn.edu-- > > > ------- End of Forwarded Message > --------------------------------- Joan Bresnan From bresnan at CSLI.STANFORD.EDU Wed Dec 15 06:57:22 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 22:57:22 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Wed, 15 Dec 1999 00:20:30 EST. <199912150520.AAA10973@central.cis.upenn.edu> Message-ID: Eve Clark! That brings us to: people who do corpus-based research in Stanford linguistics and CSLI: Tom Wasow, Chris Manning, John Rickford, Beth Levin, Penny Eckert, Eve Clark, Ann Copestake. representing syntax, computational linguistics, sociolinguistics, semantics, language acquisiton, and the lexicon... > > >In addition to Tom Wasow's reply concerning his own corpus-based > >research, I would point you to Chris Manning, a new member of our > >department (http://www.stanford.edu/~manning/). Manning maintains a > >very useful and well-regarded web list of resources for statistical and > > corpus-based natural language processing: > > > >http://www.sultry.arts.usyd.edu.au/links/statnlp.html > > > >In addition, Anne Copestake, Senior Researcher at CSLI, provides a > >wealth of technical and practical experience available to our student > >researchers. > > > >Joan Bresnan > > In addition, you've got John Rickford... > > :-) > > > >> > >> `Stanford students working in syntax and semantics, for example, simply > >> start doing corpus based research as a matter of course.' > >> > >> Is this really true? Who's teaching these classes? Have you committed > >> tenure-track faculty lines to people trained in corpus-based research? Can > >> you give me references to relevant papers written by such people or their > >> professors? > >> This sounds interesting. > >> > >> John > >> > > - --AAA10908.945235077/central.cis.upenn.edu-- > > > ------- End of Forwarded Message > From bresnan at CSLI.STANFORD.EDU Wed Dec 15 06:52:27 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 22:52:27 -0800 Subject: optimality in synchrony and diachrony In-Reply-To: Your message of Tue, 14 Dec 1999 11:08:38 GMT. <38562528.818756A1@eva.mpg.de> Message-ID: (long message, last message, please forgive) Martin Haspelmath asks me to clarify my reply to Matthew Dryer. Matthew argued (in essence) that OT can't be truly functionalist because functional constraints are external to grammar, shaping language in an evolutionary way, while OT constraints are internal to grammar. Matthew and Fritz Newmeyer believe that for this reason OT is fundamentally misguided, and Martin Haspelmath is inclined to agree with them. My reply was that this argument sounds plausible because it is a tautology, true only by definition. It rests on an erroneous definition of "grammar", which is not what an OT "grammar" is. What is a grammar and how exactly can it be "shaped" by evolutionary forces? Matthew's view of grammar may be shared by many on this list, but it one that has been completely abandoned among most phonologists and also among a number of OT syntacticians. It is the view expressed in the following quotations from Matthew's original msg (emphasis added): "For example, if two languages have a difference that one might describe in terms of economic motivation competing with iconic motivation, where economic motivation wins out in one language and iconic motivation in the other, I WOULD NOT WANT TO SAY THAT THERE ARE GRAMMATICAL RULES IN THESE LANGUAGES THAT REFER TO ECONOMY OR TO ICONICITY." and: "... over the past few thousand years, the particular grammatical changes that have led to the current state of the language reflect the influence of that functional principle. But once that has happened, THE GRAMMATICAL RULES HAVE AN EXISTENCE THAT IS INDEPENDENT of the explanatory principles that have influenced them." The conception of grammar that these passages invoke is one in which a grammar consists of RULES and/or PRINCIPLES AND PARAMETERS. Since the structure of a language is determined by the rules and/or parameter settings of its grammar, on this view, it doesn't make sense for there to be "internal" functional forces also shaping the language. [I leave aside the question of how exactly languages are supposed to be shaped by EXTERNAL forces on this view.] But an OT "grammar" is not a grammar in this sense. Nor is it a "grammar" in the Principles and Parameters sense. OT rejects Principles (because they are inviolable) and Parameters (because, among other reasons, they are categorically either on or off and therefore incapable of explaining the EMERGENCE of the UNMARKED) [see also the critique of parameter learnability in Tesar and Smolensky]. So OT is a much more radical departure from the conventional thinking about what grammars are than Matthew and Fritz seem to recognize. An OT "grammar" is a model of how a set of possible structures (the candidate set) is shaped by the prioritization of conflicting universal constraints, into a particular language. The constraints reflect (hypothesized) properties of the human articulatory, perceptual, and cognitive systems. They evaluate the markedness and the faithfulness of the structures as expressions of content. The faithfulness constraints require that features of the input content be preserved in the output expression. They thus serve the communicative function of expressing contrasts in content, protecting content against the eroding effects of markedness constraints on forms. Markedness constraints penalize complex or `difficult' structures, and so tend to erode contrasts. A particular language harmonizes the conflicting constraints by prioritizing (ranking) them in its own way. Note that the constraint component that does the selection is external to the structure-generating component. Unlike the older, conventional conceptions of grammar which it replaces, OT provides both an EXPLICIT way to model how languages can be shaped by functionalist pressures and also an explanation for why language-internal processes or "rules" (which it deconstructs completely in terms of markedness and faithfulness) reflect typological patterns. My example (from Kager's textbook): Final obstruent devoicing is a productive process in the grammar of Dutch among other languages. Presumably it applies to borrowed words, and affects Dutch second-language learners of English, etc. etc. It is not a historical relic, frozen in a few lexical forms, but a living process or "rule" of the sort that phonologists would write in their descriptions of the grammar of Dutch (in the broad sense of grammar used here). The OT analysis of this devoicing "rule" in Dutch derives its effects from ranking three constraints: (1) faithfulness to voicing contrasts, (2) one markedness constraint reflecting the difficulty of perceiving voicing contrasts in certain less-salient positions (syllable codas), and (3) another constraint reflecting the general markedness of voiced obstruents (compared to vowels, for example)--you've got a closure and you're vocalizing, and it's tough. [pardon my phonetics]. Now if you take this analysis and arbitrarily rerank these constraints in all possible ways, you create little obstruent-voicing/devoicing "grammars" for different languages. What you find, is that all these rankings give you only three different possibilities: voicing contrasts in obstruents everywhere (like English); voiceless obstruents everywhere (like Polynesian); and voicing contrasts present in the salient onset position of syllables and absent is the coda position (Dutch). In a nutshell, the analysis of final devoicing in Dutch predicts the existence of the well known typological asymmetry in obstruent voicing across languages. It is because the OT constraints are universal and not language-particular rules or parameter settings, that this deep connection between typology and language-internal "grammar" is possible. You literally cannot do the (real) OT grammar of any particular language without doing typology. Martin Haspelmath wrote: > I think I didn't understand the thrust of Joan's example about the > devoicing tendency across languages and in particular languages either. > This is something that typologists have long been aware of (see e.g. > Bill Croft's markedness chapter in his typology textbook). The usual > functionalist explanation for both cross-linguistic patterns and > language-particular regularities is that they show the effect of > diachronic change. > But precisely how? Joan Bresnan From Zylogy at AOL.COM Wed Dec 15 08:15:21 1999 From: Zylogy at AOL.COM (Jess Tauber) Date: Wed, 15 Dec 1999 03:15:21 EST Subject: optimality in synchrony and diachrony Message-ID: I've been very interested in this particular thread. I'll make my points and then get out. Storage forms of lexical items to be acted on by faithfulness and markedness constraints have to come from somewhere. Historical change shows that simplification of complex structures is the rule once structures have sufficient staying power and are in the lexicon proper. Combinatory processes create new structures of higher complexity and markedness, which can then undergo further simplification, etc., but these process are live, synchronic ones. Mastication metaphorically. Yet ultimately the origins of roots must be in something similar to expressive vocabulary such as ideophones, even if such origin is quite ancient, and the path traveled through the generations and between languages tortuous. And as morphology derives from free forms, it is probably safe to bet that everything boils down to old expressive vocabulary. It is of interest, then, that such vocabulary is often seemingly resistant to historical changes, if only because it isn't really in the lexicon, but created anew (if not de novo) with each use. And such vocabulary is hardly integrated syntactically, and as such is extraclausal and uninflected. My own research shows that expressive vocabulary obeys universal typological principles. But humanity just can't leave well enough alone. Combination is just too valuable a way to be precise, if not concise. As integration does start to occur the size and complexity of combinatory products starts getting in the way of communicative efficiency, and so we have the slippery slope to simplification. You've heard this all before, in other contexts. I bring all this up because I believe we need a theoretical treatment of how synchronic combination and historical change create inputs in the lexicon, as much as we need one for outputs. Something akin to antimarkedness (if ideophones are maximally unmarked already) to allow for the combinatory creation of marked structures, and antifaithfulness, which would allow forms to infringe/converge on structure already occupied by other forms. The system starts from a state of matrix-like crystallinity of form/meaning, and ends up with a fully mixed, seemingly randomly distributed state. It will be interesting to see what formalists and functionalists, and the folks trying to integrate the two viewpoints, have to say about it (assuming of course you buy the main thrusts). Best to all. Sincerely, Jess Tauber zylogy at aol.com From bresnan at CSLI.STANFORD.EDU Wed Dec 15 21:36:41 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Wed, 15 Dec 1999 13:36:41 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Tue, 14 Dec 1999 22:57:22 PST. <199912150655.WAA07293@hypatia.Stanford.EDU> Message-ID: For the record, I should have added Elizabeth Traugott, who does corpus work in both research and teaching, in all of her historical courses in her course on Discourse Analysis. The corrected list is now as follows. People who do corpus-based research in Stanford linguistics and CSLI: Elizabeth Traugott, Tom Wasow, Chris Manning, John Rickford, Beth Levin, Penny Eckert, Eve Clark, Ann Copestake. --representing historical linguistics, discourse analysis, syntax, computational linguistics, sociolinguistics, semantics, language acquisition, and the lexicon... I would have thought it was completely unnecessary to point out these obvious facts about linguistics at Stanford University, but apparently there is quite a lot of misunderstanding or misinformation about our program. This point is, data-driven methodologies in linguistics are a long-standing tradition of Stanford Linguistics, and they are being reinforced both by recent additions to our faculty and new research directions among some of our continuing faculty. At the same time, we probably offer a greater variety of the formal modelling and analytic techniques used in contemporary theoretical and computational linguistics than any other linguistics department. From nrude at ucinet.com Wed Dec 15 23:00:55 1999 From: nrude at ucinet.com (Noel Rude) Date: Wed, 15 Dec 1999 15:00:55 -0800 Subject: optimality in synchrony and diachrony Message-ID: Howdy! Don't really know what the folks are trying to say so maybe I should keep quiet, but then again--am I right that most of us believe that structure codes function, and that if function shapes grammaticalization then this structure-function relationship begins synchronically and persists until bleached out and/or changed? I thought we believed 1) in structure which includes segmantal sequences and prosodies and affixation and word order all of which is subject to various psycho-physical constraints, and which codes 2) simple and complex information 3) according to various principles (propositional structure and semantics and discourse/pragmatics and social stuff ...) which may or may not be related to structure iconically and which may be either "internal" or "external" (whatever this might mean) to Grammar or grannars and 4) which in practice is physically/psychically automated (or routinized) and 5) that anywhere along the line any rules (or "principles and parameters") which CAN be broken WILL be broken (we are not automatons and the rules are not the laws of physics). If we mostly agree on this, how far are we from mopping up the details? Noel From mithun at HUMANITAS.UCSB.EDU Wed Dec 15 23:16:17 1999 From: mithun at HUMANITAS.UCSB.EDU (Marianne Mithun) Date: Wed, 15 Dec 1999 15:16:17 -0800 Subject: corpus based research at Stanford In-Reply-To: <199912150637.WAA07099@hypatia.Stanford.EDU> Message-ID: > I think that more corpus-based research from more perspectives on language > is possible at Stanford than almost anywhere. (Penn being an > exception to the rule... :-) > > J. Actually, there may be considerably more active, corpus-based linguistics going on than many realize. There's been a long tradition among most Santa Barbara linguists of corpus-based analysis, in many cases dating from well before the formation of the department over a decade ago. It ranges from close work on phonetics and prosody through analysis of morphology, syntax, discourse, language change, and language acquisition. The corpora vary in size, but most are quite extensive. They generally consist of spontaneous spoken language, and considerable thought and discussion have gone into issues of content, collection, format, etc. Languages represented include Mohawk, Tuscarora, Cayuga, Seneca, Caddo, Central Pomo, Central Alaskan Yup'ik, Kapampangan, Mandarin, Korean (child language), Japanese (both adult and child language), American English, and others. Corpus-based linguistics is something most of our students simply do as a matter of course in much of their work. Marianne Mithun From macw at CMU.EDU Wed Dec 15 23:59:20 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Wed, 15 Dec 1999 18:59:20 -0500 Subject: corpus based research at Stanford In-Reply-To: Message-ID: Marianne, Joan, and other people interested in corpora, Mariane is right in saying that "there may be considerably more active, corpus-based linguistics going on than many realize." But why is this pivotal aspect of linguistics so low-profile? One way of increasing the profile of corpus-based work is for linguists and their allies to begin to develop more effective ways of sharing corpora, including spoken language corpora. One move in this direction is the new TalkBank project, which NSF (KDI/Linguistics/SBE) has recently funded. (see http://talkbank.org) The goal of TalkBank is to provide computational tools that support corpus-based linguistics and related efforts in about a dozen disciplines devoted to the study of spoken communication. On Dec 4-5, we held a first TalkBank workshop that explored the construction of a database for the study of language used in classrooms and tutorial interactions. http://www.talkbank.org/meetings.html The next TalkBank meeting is devoted to Linguistic Exploration (or what some people might call "field linguistics"). If you are attending LSA this year and are interested in sharing data on spoken communications, please take a look at the program for January 6 at http://www.talkbank.org/exploration.html My guess is that there is a wealth of fantastic spoken language data out there from languages such as "Mohawk, Tuscarora, Cayuga, Seneca, Caddo, Central Pomo, Central Alaskan Yup'ik, Kapampangan, Mandarin, Korean, and Japanese" to name just a few that would greatly benefit the progress of empirically-grounded research across linguistics and allied areas. With our new computational tools we can access these data directly over the Internet (while respecting confidentiality as required). Sounds can be directly linked to transcripts and data can be elaborated with collaborative commentary. TalkBank can provide us all with a way of gaining shared access to these data. In this way, we can also gain a better understanding of the actual data our colleagues are looking at. --Brian MacWhinney From haspelmath at EVA.MPG.DE Thu Dec 16 09:31:47 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Thu, 16 Dec 1999 09:31:47 +0000 Subject: diachronic functionalism Message-ID: Joan Bresnan asks an important question, referring to my earlier posting: >> The usual functionalist explanation for both cross-linguistic patterns and >> language-particular regularities is that they show the effect of >> diachronic change. > >But precisely how? The basic idea is that functional factors apply in performance. Take the devoicing example again: Voiced obstruents are harder to pronounce than voiceless ones, especially in (syllable- or word-)final position. Speakers are thus constantly "tempted" to devoice consonants (partially or completely) under these conditions, and sometimes they give in to the temptation. This may then spread throughout the speech community and result in a language change, such as that from Old High German (e.g. tag 'day') to Modern German (Ta[k] 'day'). Since the phonetic factors are universal, we find languages that have completely lost voiced obstruents (if they ever had any), as well as languages that have devoicing only in final position. We find no languages lacking voiceless obstruents, because there is no possible diachronic change that could give rise to such languages. The source of the universals is thus in performance and (hence) in diachrony, and they are reflected in competence only secondarily. The prediction that this view makes is that impossible languages (e.g. with initial devoicing, or with only voiced obstruents) should NOT be unlearnable, but that they should be diachronically unstable. Unfortunately, learnability experiments are impractical (and perhaps unethical), so the issue is so hard to resolve. Let me cite Joan Bresnan again: >In a nutshell, the analysis of final devoicing in Dutch predicts the >existence of the well known typological asymmetry in obstruent voicing >across languages. The same applies to the diachronic-functional view, except that we would talk in terms of "explanation" rather than "analysis". (Of course, in both approaches the prediction is not literal, because one would not have adopted this particular analysis/explanation for Dutch without the knowledge of the typological asymmetry.) >It is because the OT constraints are universal and not >language-particular rules or parameter settings, that this deep >connection between typology and language-internal "grammar" is >possible. You literally cannot do the (real) OT grammar of any >particular language without doing typology. This would be an advantage only if it turned out that that's the right way of "doing grammar". And diachronic functionalists (such as Matthew Dryer, Fritz Newmeyer, Joan Bybee, Joe Greenberg, Bill Croft) would reject the view that anything is gained by subsuming language-particular description and explanation of universal tendencies under a single theory. (Or in Newmeyer's case, he already seems to be convinced that this is impossible.) Martin P.S. Joan Bresnan seems to use the term "evolutionary" where I use "diachronic". I think this terminological usage should be avoided, because "evolutionary" is ambiguous between "phylogenetic (Darwinian) evolution" and "glossogenetic evolution" (i.e. diachronic change"). -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From john at RESEARCH.HAIFA.AC.IL Thu Dec 16 08:38:00 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Thu, 16 Dec 1999 10:38:00 +0200 Subject: Translation databases? Message-ID: This is terrific, Brian, but what I would really like to know is: Are there any TRANSLATION databases? That is, databases which have both originals and translations in other languages? Allowing for word searches (construction searches would be even better, but this is too much to hope for)? In a variety of genetically unrelated and geographically separated languages? If we want to do comparisons of functions of different structures, or meanings of different word, in different languages, translations are really helpful. For those of us who are seriously interested in language universals, translation data, like nothing else, force us to come to grips directly with differences between languages; we cannot, for example, so well blather about the `universal' or `cognitive' functions of voice alternations based exclusively on English data when confronted with translation data clearly showing that other languages use voice alternations in extremely different ways. I have applied for grants to develop such a translation database twice and been rejected both times. Wally Chafe tells me that the Pear Stories have never been rendered into a usable form (and in any case they are quite short). I have done a number of studies using the Bible, because at least there are a lot of texts with interlinear glosses in both languages, and there are concordances of particular words--but there aren't so many languages with such data, and Bible translations tend not to be into the most naturalistic language, if you know what I mean. There are of course many texts with interlinear glosses in, e.g. Native American or Australian languages and English, but each of these is in only two languages, and there's no concordance to help searches for individual words (in addition to difficulty in accessing native speakers for help). So, in order to get comparison between more than just two languages, I have been forced to do things by hand. I am presently doing a study of the comparative meanings of speech act verbs in Hebrew, English, Japanese, and Spanish by using novels and short stories by Gabriel Garcia Marquez, A.B. Yehoshua, and Banana Yoshimoto, with translations of each into each of the other languages, and let me tell you, it is pretty slow going. I have to search for each occurrence of a given word, then search to see how it is translated into each of the other three languages, without the use of concordances or interlinear glosses (it goes without saying that I read some of these languages more quickly that others). If I have only like 5 occurrences of a given word, the translation data often looks kind of chaotic, but if I can get 30 or 40 tokens, very clear patterns always emerge, but unless the word is pretty common it simply takes too long to get this number of tokens. I can do it, and the results are very interesting, but it takes a long time, and I more or less have to study only words which are pretty common (e.g., I would love to do a study of emotion words like 'angry', 'sad', etc. to see how they're translated, but it would take an enormous amount of time to locate enough tokens to use--I've tried). Any ideas or data sources which might speed things up? Hopefully, John From jlmendi at POSTA.UNIZAR.ES Thu Dec 16 09:14:49 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Thu, 16 Dec 1999 10:14:49 +0100 Subject: Myhill on Chomsky In-Reply-To: <38568478.C9732208@chass.utoronto.ca> Message-ID: This is my last message about this subject. Peter A. Reich wrote: >I'm sure God is cited more than Chomsky (I will not engage in a >discussion as to whether God is Living), and probably Bill Gates is >mentioned more than Chomsky as well in the popular media. And Clinton. >So what? It is perfectly respectable in Canada, at least, to be an >Atheist, and to prefer Linux or Macintosh. And, in the US, to be a >Republican. Eat shit; 50 billion insects can't be wrong.--Peter Reich Of course, when I used a 'quantitative' argument I was not trying to demonstrate that Chomsky's ideas about language and grammar are correct. This is an empirical matter, a scientific matter we are not dealing with here. In fact, someone here has said that the use of, say, 'qualitative' arguments would be a waste of time. I was only trying to show Myhill that his vision of Chomsky as a Pope indoctrinating an international herd of hundreds (or thousands) of icapable of thinking by themselves linguists and scientists is simply not solvent. That it is just an insult. Best Regards, Jose-Luis. From nc206 at HERMES.CAM.AC.UK Thu Dec 16 09:55:48 1999 From: nc206 at HERMES.CAM.AC.UK (N. Chipere) Date: Thu, 16 Dec 1999 09:55:48 +0000 Subject: use of corpora based In-Reply-To: <350474.3154273160@agate.psy.cmu.edu> Message-ID: On Wed, 15 Dec 1999, Brian MacWhinney wrote: > Marianne, Joan, and other people interested in corpora, > > Mariane is right in saying that "there may be considerably more active, > corpus-based linguistics going on than many realize." But why is this > pivotal aspect of linguistics so low-profile? > One way of increasing the profile of corpus-based work is for linguists and > their allies to begin to develop more effective ways of sharing corpora, > including spoken language corpora. One move in this direction is the new > TalkBank project, which NSF (KDI/Linguistics/SBE) has recently funded. (see > http://talkbank.org) Yet another way is, as far as possible, to base discussions of grammar on corpora rather than to rely purely on intuition. - Ngoni From david_tuggy at SIL.ORG Thu Dec 16 15:55:51 1999 From: david_tuggy at SIL.ORG (david_tuggy at SIL.ORG) Date: Thu, 16 Dec 1999 10:55:51 -0500 Subject: The rest of us on Chomsky Message-ID: In the interests of Chomsky-discussion rather than Chomsky-bashing: It seems clear that a number of people on this list think at least some of Chomsky's ideas have been problematic for linguistics rather than helpful. I'd enjoy hearing from any who'd like to respond to questions of the following sort. What do you see as *the* most problematic/pernicious/unhelpful of Chomsky's theoretical positions / basic analytical stances / posited grammatical mechanisms / etc. ? What's wrong with it? Why is it so bad? What's right with it? Why has it convinced so many intelligent people? How does your favorite alternative (functionalist or not) avoid what was wrong with it and keep what was right? --David Tuggy From MILLSCR at UCMAIL.UC.EDU Thu Dec 16 17:03:40 1999 From: MILLSCR at UCMAIL.UC.EDU (Mills, Carl (MILLSCR)) Date: Thu, 16 Dec 1999 12:03:40 -0500 Subject: Chomsky Message-ID: Actually, many of us who went to LSA in the late 1970s and '80s discovered that the "genuine criticism" our papers received, most of it from Chomskyans, was so ill-informed and narrow minded that we stopped going. So I don't blame Chomsky for not going. On the other hand, it is not surprising that Chomsky is addressing the MLA. Anyone who reads stuff from the sciences, e.g. biology, chemiostry, anthropology (the scientific, i.e. physical or biological, parts of it) is apt to discern that linguistics since the Chomskyan revolution (pick whichever one of them you want) has becoming less scientific and more humanistic over the years. Personally, I think that pre-Chomskyan structuralism was a sort of natural history, but what we see mostly today is on a par with literary theory (sic). Incidentally, my own work on how the sciences and a few other disciplines display their results indicates that there are a number of disciplines between the sciences and the humanities (psychology, sociology) that are not unlike linguistics in being unable or unwilling to decide whether they are sciences or humanities. Carl -----Original Message----- From: A R You know, Funknetters, I don't know that I'd call myself either a formalist or a functionalist. But one of the most damning things I know about Chomsky is that he *never* goes to LSA, the one place where he might face genuine criticism. A friend of mine, a formalist, used to argue that he simply never went to academic conferences. This year, I discover that this is clearly untrue. Chomsky is going to talk at the MLA, to those who do literature. It's only his own that he refuses to talk to at conferences. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From bresnan at CSLI.STANFORD.EDU Thu Dec 16 18:07:49 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 16 Dec 1999 10:07:49 -0800 Subject: diachronic functionalism In-Reply-To: Your message of Thu, 16 Dec 1999 09:31:47 GMT. <3858B17F.1D87F188@eva.mpg.de> Message-ID: Martin Haspelmath's response to my question ("But precisely how?") makes it clear to me that there might not be any real incompatibility between our views, once they are spelled out more precisely and terminological differences are clarified. >From his initial statement-- > The basic idea is that functional factors apply in performance. --I infer that he thinks that the differences in our views have something to do with "performance" vs. "competence". But I don't recall mentioning that chestnut in any of my postings. [In fact, I believe I once wrote that the competence/performance distinction has served as a convenient way for some linguists to insulate their theories from empirical disconfirmation.] Is Martin embracing the classic generative view of competence grammar as part of his historical explanation or just (incorrectly) presupposing that I own that view or that anyone who does OT must subscribe to that particular view articulated by Chomsky in 1965? It certainly isn't true. Let's simply drop it as a false barrier to communication. To explain the typological and language-internal patterns of voiceless obstruents, Martin supposes that > Speakers are thus constantly "tempted" to devoice consonants (partially > or completely) under these conditions, and sometimes they give in to the > temptation. Is the "constant" presence of this "phonetic temptation" any different the OT hypothesis that the contextual markedness of voiced obstruents is a universal constraint is present in every individual by virtue of the human articulatory and perceptual systems? Martin suggests that once speakers "give in to the temptation" to devoice, devoicing may spread throughout the speech community and result in language change. Yes, but when we try to model precisely what "giving in to the temptation" means, we may come up with the idea that the "phonetic temptation" becomes more dominant compared to the temptation to preserve constrasts. OT provides some very precise and nice ways to model this. One I like very much is the Boersma-Hayes model of probabilistically varying constraint rankings that I referred to recently. (See their recent ROA paper.) Constraints have continuous (not scalar) ranking values which vary according to a normal distribution. When two conflicting constraints very close in their ranking are applied to the same form, either one may dominate, yielding a certain frequency of variation in the forms produced. I said nothing in my postings about how variation of this kind spreads through populations, and that is certainly where diachronic and sociolinguistic explanations come in. We must look at all sorts of "external" issues such as the relative influence and power of social groups and networks of individuals. One way such factors can exert their influence on individual linguistic patterns, in OT terms, is by modifying the rankings of constraints. In some recent work I have done on negative auxiliary inversion in several English dialects including Scots and Hiberno-English, I show how through a sociolinguistically determined ranking of this kind, variations in both the syntactic distribution and the semantic scope of negation can be explained. One striking syntactic difference is this: Scots English: Amn't I your friend? *I amn't your friend? Hiberno English: Amn't I your friend? I amn't your friend? Note that OT does not determine the ranking of constraints, but only an explicit space of possible variation within which changes and their implications and interactions may be studied in precise detail. [Nigel Vincent has begun to apply this kind of model to historical change from Latin to Romance: see his recent posting to the LFG archive: http://www-lfg.stanford.edu/lfg/.] --Unfortunately, I must interrupt this conversation and go now. Best wishes to you functionalists and here's hoping for a productive and interactive new millenium! Joan > > P.S. Joan Bresnan seems to use the term "evolutionary" where I use > "diachronic". I think this terminological usage should be avoided, > because "evolutionary" is ambiguous between "phylogenetic (Darwinian) > evolution" and "glossogenetic evolution" (i.e. diachronic change"). I completely agree. Sorry, I just unthinkingly picked up some words from Matthew's msg, and realized later that they weren't what I really intended. From bresnan at CSLI.STANFORD.EDU Thu Dec 16 18:12:04 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 16 Dec 1999 10:12:04 -0800 Subject: diachronic functionalism In-Reply-To: Your message of Thu, 16 Dec 1999 10:07:49 PST. <199912161806.KAA12891@hypatia.Stanford.EDU> Message-ID: correction: > Constraints have > continuous (not scalar) ranking values .. I meant "continuous (scalar) ranking values". J.B. From chafe at HUMANITAS.UCSB.EDU Thu Dec 16 18:19:23 1999 From: chafe at HUMANITAS.UCSB.EDU (Wallace Chafe) Date: Thu, 16 Dec 1999 10:19:23 -0800 Subject: Corpora Message-ID: Evidently Joan wanted to make the valid point that Stanford is not just MIT West, but then it degenerated into "my department has more corpus linguists than your department" (unless you're Penn) (and I'm having a little trouble remembering who they are). More to the point might be a discussion of what corpus research involves, what kinds of corpora there are, how they can best be exploited, etc. Brian and others are providing an important service in this regard with the Talkbank project. As I see it, we're talking about an alternative to the popular kinds of data-gathering that have involved inventing sentences in isolation and asking people whether they "get" them, or measuring the reaction times of college sophomores who see them written on computer screens. The alternative is to examine how people actually use language, a process that necessarily involves confronting more than single sentences. In that sense it's part of what has come to be called the study of "discourse", which of course can be conducted in many different ways. It's worth noting that some people have been examining such data for a long time: one thinks, for example, of many language acquisition studies, of the analysis of conversation (from various points of view), and of the recording and analysis of "texts" collected by those who have been studying lesser known languages. This last kind of corpus work has been going on for well over a hundred years. It's worth noting, too, that this distinction between constructed language and what I like to call "real" language ("natural language" has been coopted with a different meaning) is orthogonal to the formalist-functionalist dichotomy, at least in the sense that while many functionalists do work with corpora, many do not. It might be worth discussing the problems that arise from the supposedly accidental nature of corpora, and the lack of the control and replicability that are so dear to the hearts of psychologists. One might actually find some significance, for example, in the fact that people rarely use a construction one might think easy to invent. And of course the problem tends to diminish with very large corpora. But very large corpora may introduce a problem of their own. Some of you may remember Zellig Harris's book Methods in Structural Linguistics, where he suggested we could get around the vexing problem of meaning by examining the distribution of linguistic forms in very large corpora. Machines to do that weren't available at the time (1950), but now they are, and it looks to me as if some people are doing what Harris had in mind, though so far as I know they don't refer to him. It makes me uncomfortable because I think it's more rewarding in the long run to confront semantics head-on, not trying to avoid it with big corpora and machines. Just one last reservation. Corpora make it easy to count things and come up with interesting findings regarding the frequency of this or that. But knowing exactly what you're counting may not be such a simple matter, and it's easy to come up with "operational definitions" that turn out in the end to be spurious. What I'm trying to say is that there's much of importance to learn from examining real language, but it shouldn't seduce us into thinking we can just crank out analyses mechanically. Understanding the nature of language is always going to require the intervention of perceptive human minds. Wally Chafe From MILLSCR at UCMAIL.UC.EDU Thu Dec 16 18:33:47 1999 From: MILLSCR at UCMAIL.UC.EDU (Mills, Carl (MILLSCR)) Date: Thu, 16 Dec 1999 13:33:47 -0500 Subject: Counting things Message-ID: Having spent three decades counting things, I want to second Wallace Chafe's "last reservation." Of course, there is always going to be a need for the intervention of human minds. But those minds tend to work best on inputs that do not originate within themselves. There is a real world, and it is, in some important sense, out there. Corpus linguistics offers the possibility of starting with something other than "Can you say S?" or worse "Can I say S?"and I don't think that is a small achievement. Carl -----Original Message----- From: Wallace Chafe To: FUNKNET at LISTSERV.RICE.EDU Just one last reservation. Corpora make it easy to count things and come up with interesting findings regarding the frequency of this or that. But knowing exactly what you're counting may not be such a simple matter, and it's easy to come up with "operational definitions" that turn out in the end to be spurious. What I'm trying to say is that there's much of importance to learn from examining real language, but it shouldn't seduce us into thinking we can just crank out analyses mechanically. Understanding the nature of language is always going to require the intervention of perceptive human minds. Wally Chafe -------------- next part -------------- An HTML attachment was scrubbed... URL: From wcmann at JUNO.COM Thu Dec 16 18:31:29 1999 From: wcmann at JUNO.COM (William Mann) Date: Thu, 16 Dec 1999 13:31:29 -0500 Subject: Translation databases? Message-ID: It seems timely to point out that there is a very active Corpora email list, and that discussions of multilingual databases appear there regularly. Participation is very international, at perhaps 5 messages per day. The control is through Majordomo at iub.no and a message body sent to it, containing subscribe corpora <> should begin to get you in. It also responds to the message body help with help. I don't know whether they have searchable archives. Happy processes. Bill Mann From jlmendi at POSTA.UNIZAR.ES Thu Dec 16 19:06:49 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Thu, 16 Dec 1999 20:06:49 +0100 Subject: The rest of us on Chomsky In-Reply-To: <9912169453.AA945364013@router-8.camnet.com> Message-ID: At 10:55 -0500 16/12/99, david_tuggy at SIL.ORG wrote: > In the interests of Chomsky-discussion rather than Chomsky-bashing: > > It seems clear that a number of people on this list think at least > some of Chomsky's ideas have been problematic for linguistics rather > than helpful. > > I'd enjoy hearing from any who'd like to respond to questions of the > following sort. > > What do you see as *the* most problematic/pernicious/unhelpful of > Chomsky's theoretical positions / basic analytical stances / posited > grammatical mechanisms / etc. ? > > What's wrong with it? Why is it so bad? Please forgive me for using the words of a member of this list, but I could not state better than him: "An undercurrent of hostility to generative grammar arises because many who identify themselves as linguists believe that Chomsky wishes to define them out of the field of linguistics" (Newmeyer, 1983: 138). Of course, this is not the only problematic affair of Chomsky's theories and attitudes at all, but I think it explains quite well some points of view showed here. In other words, it seems to me that the major problem is related, first, to the fact that Chomsky's program has not been properly understaken and, second, to the fact that Chomsky and, specially, Chomskyans tend to ignore what is not Chomskyan. There is mainly a problem of communication. Best regards, Jose-Luis. From dgolumbi at PANIX.COM Thu Dec 16 19:21:54 1999 From: dgolumbi at PANIX.COM (David Golumbia) Date: Thu, 16 Dec 1999 14:21:54 -0500 Subject: the rest of chomsky Message-ID: during this intense and often enlightening discussion, this sentence of recent vintage, by a contributor whose first language appears to me not to be English, strikes me with special force: > Chomsky's program has not been properly understaken rooting out the "problems" in Chomsky's program could (and should) be the object of even more informed comment than it has already (botha, r[andy] harris, r[oy] harris, huck/goldsmith, matthews, mccawley, newmeyer) - but you could take less productive starting points than this fun gem, which to start with, is*? a specimen of English? -- dgolumbi at panix.com David Golumbia From kfeld at CITRUS.UCR.EDU Thu Dec 16 20:48:08 1999 From: kfeld at CITRUS.UCR.EDU (David B. Kronenfeld) Date: Thu, 16 Dec 1999 12:48:08 -0800 Subject: Corpora In-Reply-To: Message-ID: Out of curiosity, how does the corpora vs. invented speech distinction relate to the old distinction between the way people normally talk vs. the way the talk which you ask them to slow down and be careful (or ask them what they meant to say--or some such)--particularly re how grammatical relations are approached ? And-- "amen" on the Z. Harris comment and what follows it. David Kronenfeld At 10:19 AM 12/16/99 -0800, you wrote: >More to the point might be a discussion of what corpus research involves, >what kinds of corpora there are, how they can best be exploited, etc. >Brian and others are providing an important service in this regard with >the Talkbank project. > >As I see it, we're talking about an alternative to the popular kinds of >data-gathering that have involved inventing sentences in isolation and >asking people whether they "get" them, or measuring the reaction times of >college sophomores who see them written on computer screens. The >alternative is to examine how people actually use language, a process that >necessarily involves confronting more than single sentences. In that >sense it's part of what has come to be called the study of "discourse", >which of course can be conducted in many different ways. It's worth >noting that some people have been examining such data for a long time: >one thinks, for example, of many language acquisition studies, of the >analysis of conversation (from various points of view), and of the >recording and analysis of "texts" collected by those who have been >studying lesser known languages. This last kind of corpus work has been >going on for well over a hundred years. > >It's worth noting, too, that this distinction between constructed language >and what I like to call "real" language ("natural language" has been >coopted with a different meaning) is orthogonal to the >formalist-functionalist dichotomy, at least in the sense that while many >functionalists do work with corpora, many do not. > >It might be worth discussing the problems that arise from the supposedly >accidental nature of corpora, and the lack of the control and >replicability that are so dear to the hearts of psychologists. One might >actually find some significance, for example, in the fact that people >rarely use a construction one might think easy to invent. And of course >the problem tends to diminish with very large corpora. > >But very large corpora may introduce a problem of their own. Some of you >may remember Zellig Harris's book Methods in Structural Linguistics, where >he suggested we could get around the vexing problem of meaning by >examining the distribution of linguistic forms in very large corpora. >Machines to do that weren't available at the time (1950), but now they >are, and it looks to me as if some people are doing what Harris had in >mind, though so far as I know they don't refer to him. It makes me >uncomfortable because I think it's more rewarding in the long run to >confront semantics head-on, not trying to avoid it with big corpora and >machines. > >Just one last reservation. Corpora make it easy to count things and come >up with interesting findings regarding the frequency of this or that. But >knowing exactly what you're counting may not be such a simple matter, and >it's easy to come up with "operational definitions" that turn out in the >end to be spurious. What I'm trying to say is that there's much of >importance to learn from examining real language, but it shouldn't seduce >us into thinking we can just crank out analyses mechanically. >Understanding the nature of language is always going to require the >intervention of perceptive human minds. > >Wally Chafe > David B. Kronenfeld Phone Office 909/787-4340 Department of Anthropology Message 909/787-5524 University of California Fax 909/787-5409 Riverside, CA 92521 email kfeld at citrus.ucr.edu http://www.ucr.edu/CHSS/depts/anthro/home.htm http://pweb.netcom.com/~fanti/david.html From fjn at U.WASHINGTON.EDU Thu Dec 16 22:05:38 1999 From: fjn at U.WASHINGTON.EDU (Frederick Newmeyer) Date: Thu, 16 Dec 1999 14:05:38 -0800 Subject: OT and functional explanation Message-ID: Joan Bresnan's clarification of her position with respect to Martin Haspelmath's was very helpful, but there may be differences that remain. I've been able to discern a number of views out there with respect to the 'functionality' of OT constraints. Here are three of them: (A) Constraints are universal, and may or may not be functionally motivated (e.g. Jane Grimshaw's LI paper, where the issue does not come up). (B) Constraints are universal and functionally motivated (Martin Haspelmath's forthcoming Z fuer S paper). (C) Constraints are universal and are the actual functional motivations themselves. That is how I interpret the following remark from Joan's December 10 posting: 'Couldn't conflicting constraints such as iconicity and economy be universal, but prioritized differently across different domains and different languages?' Martin and Joan: does that put the finger on the different ways that you view the constraints? I'm an utter novice at OT, so the issue I will now raise could be based on my own ignorance of that framework. If so, no doubt several of you will tell me so. But I have trouble seeing how either (B) or (C) could be made to work. My primary qualm comes from the fact that the forces (functional or otherwise) that bring a construction into a language are not necessarily the same ones that keep it there. An example: English is a primarily head-initial right-branching language. There are good functional reasons for a language to be 'consistent' in this regard (Dryer, Hawkins, et al.). So a constraint for English like HEAD-LEFT (or BRANCH-RIGHT) most certainly has a functional motivation. Now there are, in fact, left-branching constructions in English: the GEN-N construction is the best-known ('Mary's mother's uncle's lawyer'). A Grimshavian OT analysis could license this by a constraint called 'GEN-N' (or whatever) that would dominate HEAD-LEFT (the details are more complicated since English also has N-GEN, but we can put them aside). Now how might positions (B) or (C) handle this fact about English? I don't know. As I understand the history of English, 1000+ years ago it was largely left-branching, so the GEN-N construction was indeed functionally motivated at one time. For whatever reason (and many have been suggested), English has become largely right-branching. GEN-N survives as a conventionalized relic of the days when it had a real functional motivation. So, what does a functionally-oriented OT analysis do about this? Surely it would not want to say that the constraint licensing this left-branching structure is functionally motivated by parsing pressure, since that is manifestly false. In that respect this construction is counter-functional. Would the program necessitate finding some other functional explanation for its existence? Is 'convention' admissible as a category of functional motivation? (That would raise a host of problems, no?) I'm raising a number of questions, but don't mean to seem dogmatic about the answers. Languages are filled with constructions that arose in the course of history to respond to some functional pressure, but, as the language as a whole changes, cease to be very good responses to that original pressure. Such facts seem challenging to any theory (like versions of OT that have been suggested) in which the sentences of a language are a product of constraints that must be functionally motivated, or are the actual functional motivations themselves. --fritz newmeyer From macw at CMU.EDU Fri Dec 17 00:31:04 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Thu, 16 Dec 1999 19:31:04 -0500 Subject: Chomsky and deep structure Message-ID: David and FunkNet, Not to distract from the really fascinating issues being discussed in the functionalism-and-OT thread and the corpus linguistics thread, I thought I would take a shot at a quick response to David Tuggy's reasonable request for a statement regarding *the* most problematic of Chomsky's positions. As far as I can see, the core problem with Chomsky's generative grammar is the commitment to deep structure. All of the other problems derive from that commitment. "Underlying form" is a reflex of deep structure in that it allows us to map the many onto the few, thereby appearing to achieve "linguistically significant generalizations" along with abstractions that can be used for further abstract theory building. "Competence-performance" and the "ideal native speaker" are constructs that are required to support theories of deep structure. Without deep structure, the notion of universal grammar is hard to imagine. I realize that the importance of deep structure has waxed and waned in the various versions of the theory. But I see no evidence that, the core conceptual importance of deep or abstract structure has ever disappeared. What is wrong with deep structure? The problems have to do with the (1) the nature of empirical evidence and (2) the structure of human thought. People don't speak deep structures and it is essentially impossible to find them in natural interactions or experiments. Of course, one can argue that physicists rely on a particle and force-based deep structure in quantum dynamics. This is true, but the hypothesization of particles has been supported at each point by precise empirical measurements matched with detailed mathematical accounts that are remarkably precise. No such accuracy of measurement supports the constructs of linguistic deep structure. Psychologists like Sigmund Freud have shown us how dangerous it can be to construct theories of deep structure that cannot be clearly pegged to empirical tests. I am certainly not opposed to abstract theories in principle, but I am only willing to work with such theories when the individual components of the theory have demonstrable empirical grounding. Many of the constraints of certain versions of OT seem to fulfill this requirement, for example. My second reservation about deep structure has to do with what we know about other areas of perception and cognition. It is certainly true that the construction of a visual percept relies on low-level feature detection. In this sense, we could think of vision as building up constructs from a deep structure base. But this is not what Chomsky has in mind with the non-perceptually-grounded categories of universal grammar. In principle, one could imagine that a cognitive system could have arisen through evolution that was structured in this way. But the fact that language is a relatively recent event in primate evolution makes me wonder how something so abstract and ungrounded as deep structure could have arisen in such a short evolutionary time span. Calling language a sprandrel (ala Chomsky and Gould) is fine, but sprandrels are simple emergent structures and deep structure is not. So where would deep structure come from? One can argue that deep structure is just a hypothesis that should be pursued with the same tenacity with which we pursue any scientific hypothesis. I agree with that. But, just as I would not place all my funding resources on one approach to fusion research, I don't think that all of linguistics should be dedicated to pursuing theories grounded on deep structure. In reality, linguistics is far more diverse than the general public realizes. Some high profile media events have made it seem monolithic, but this is misleading. I think that Chomskyan approaches will never disappear, since the hypothesis of deep structure will always be tenable and worth exploring. Once we really understand how language is processed and stored by the brain, we may reinvent a empirically-grounded type of deep structure. But it will have only a vague family resemblance to the deep structures of "Cartesian linguistics" or Chomskyan linguistics. What is right about Chomsky? I think Chomsky's great contributions are his emphasis on generativity and mechanism, his treatment of language as a component of human cognition, his ideas about language and creativity, and his stimulating collaborations with George Miller and Eric Lenneberg. So, David, that's how I see it. --Brian MacWhinney From ward at MERLE.ACNS.NWU.EDU Fri Dec 17 01:08:40 1999 From: ward at MERLE.ACNS.NWU.EDU (Gregory L Ward) Date: Thu, 16 Dec 1999 19:08:40 -0600 Subject: LSA (bashing) Message-ID: Carl Mills writes: > Actually, many of us who went to LSA in the late 1970s and '80s > discovered that the "genuine criticism" our papers received, most of > it from Chomskyans, was so ill-informed and narrow minded that we > stopped going. So I don't blame Chomsky for not going. A R writes: > You know, Funknetters, I don't know that I'd call myself either > a formalist or a functionalist. But one of the most damning things > I know about Chomsky is that he *never* goes to LSA, the one place > where he might face genuine criticism. A friend of mine, a formalist, > used to argue that he simply never went to academic conferences. To sum up: Chomsky doesn't go to the LSA because he'd face "genuine criticism", and Mills doesn't go to the LSA because he faced 'genuine criticism' ("in the late 1970s and '80s"). So the LSA is a conference where the work of functionalists and formalists alike is criticized, presupposing the presence of both. Sounds like my kind of conference... ;-) Gregory _____ Gregory Ward Professor and Chair Department of Linguistics Northwestern University 2016 Sheridan Road Evanston IL 60208-4090 e-mail: gw at nwu.edu tel: 847-491-8055 fax: 847-491-3770 www: http://www.ling.nwu.edu/~ward From haspelmath at EVA.MPG.DE Fri Dec 17 10:15:41 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Fri, 17 Dec 1999 10:15:41 +0000 Subject: OT and diachronic functionalism Message-ID: I'm somewhat surprised to see that Joan Bresnan dissociates herself from the competence/performance terminology. OK, it probably has a lot of baggage attached to it that I may not be aware of, but my impression is that we all pretty much agree that there is a necessary conceptual distinction between language use and language structure, or speech and grammar, or more generally between processing and storage, or cognitive events and cognitive patterns. I used "competence/performance" as a convenient shorthand, following a well-known functionalists's (John Hawkins's) usage. True, the competence/performance distinction has often been attacked by functionalists, but I believe that this attack is misguided. The problem with the Chomskyan approach is not that it draws this distinction, but that all the ways in which performance influences competence are systematically ignored. (See also my review article on Fritz Newmeyer's 1998 book, soon to appear in Lingua). So in my view, three tasks of linguistics are: (i) describing competence, (ii) describing performance, and (iii) explaining competence, which must apparently often make reference to (ii). I understand the Chomskyan approach as claiming that (iii) is irrelevant, (ii) is moderately interesting, and (i) is the core task of linguists (especially concerning the innate aspects of competence). Joan Bresnan's OT functionalism draws the lines quite differently, it seems, and I haven't understood yet, how. So far my impression of OT had been that the Chomskyan program is virtually unchanged: The main goal is a maximally elegant description (or "characterization", the euphemism that is often used instead) of the competence grammar, but the archtecture of the formal grammar is different: The machinery makes use of abstract entities called "constraints" which often happen to correspond closely to the actual constraints on language use that speakers grapple with while speaking. (I call these "user constraints" in my ZS paper.) Joan Bresnan asks: >> Speakers are thus constantly "tempted" to devoice consonants (partially >> or completely) under these conditions, and sometimes they give in to the >> temptation. >Is the "constant" presence of this "phonetic temptation" any different from >the OT hypothesis that the contextual markedness of voiced obstruents >is a universal constraint that is present in every individual by virtue of >the human articulatory and perceptual systems? Until recently I thought yes, crucially, in that the OT constraint is part of the competence grammar (and is perhaps even innate as part of UG), whereas the "user constraint" is a constraint on the articulatory system and has nothing to do with the conventional language system (= competence). (But now Joan Bresnan's contributions to this discussion have confused me.) Joan continues: >Martin suggests that once speakers "give in to the temptation" to >devoice, devoicing may spread throughout the speech community and >result in language change. Yes, but when we try to model precisely >what "giving in to the temptation" means, we may come up with the idea >that the "phonetic temptation" becomes more dominant compared to the >temptation to preserve constrasts. Sure, intuitively that's what is going on, and linguists have been talking in such terms for centuries (though not as precisely as in OT, of course). But how can a "phonetic temptation" become more dominant than another temptation? After all, these temptations or constraints are universal ? identical for all speakers and all languages! Clearly, what happens is that the linguistic *conventions* change: It suddenly becomes socially acceptable to "give in to the temptation". Or in other words, the grammar has changed, and the functional motivation has left its mark on the competence system. Since the grammar is so clearly shaped by the performance constraint, it is possible (and elegant) to describe the grammar in terms of a competence constraint that mimics the performance constraint. My point in my ZS paper is that such descriptions are clearly better than arbitrary descriptions (so I welcome OT), but that we still need the link to the performance constraints if we want to explain grammar. Fritz Newmeyer asks: >(A) Constraints are universal, and may or may not be functionally >motivated (e.g. Jane Grimshaw's LI paper, where the issue does not come >up). >(B) Constraints are universal and functionally motivated (Martin >Haspelmath's forthcoming Z fuer S paper). >(C) Constraints are universal and are the actual functional >motivations themselves. That is how I interpret the following remark from >Joan's December 10 posting: 'Couldn't conflicting constraints such as >iconicity and economy be universal, but prioritized differently across >different domains and different languages?' >Martin and Joan: does that put the finger on the different ways that you >view the constraints? I thought so initially, but I haven't understood Joan yet. Fritz mentions some word order facts as difficult for an OT-functionalist position like Joan Bresnan's, and again Joan might reply that her general approach is only in its infancy, so it's too early to judge whether it might not work after all. But I agree with Fritz's general skepticism: Languages are so full of patterns that are there because of speakers' conservatism, not because they allow speakers to give in to some temptation. Fritz is absolutely right in pointing out that if one wanted to make these structures follow from an OT-functionalist grammar, one would have to introduce a functional constraint such as "conventionality". Another consideration that makes me skeptical about OT is that the descriptions rely so heavily on non-functional, language specific constraints such as alignment. HEAD-LEFT might be seen as a constraint in English grammar, but it surely cannot be functionally motivated, because HEAD-RIGHT is just as good from the language user's point of view. As John Hawkins has argued convincingly, the real motivation for the Greenbergian word order correlations is Early Immediate Constituents, and it seems difficult to integrate this into an OT-style grammar. But let's wait and see. Martin -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From david_tuggy at SIL.ORG Thu Dec 16 22:28:33 1999 From: david_tuggy at SIL.ORG (david_tuggy at SIL.ORG) Date: Thu, 16 Dec 1999 17:28:33 -0500 Subject: OT and functional explanation Message-ID: Fritz Newmeyer asks, **** Is 'convention' admissible as a category of functional motivation? (That would raise a host of problems, no?) **** I for one think it has to be, or else we can not begin to think of functionalist linguistics as covering anywhere near all of language. (It also seems prima facie obvious that doing things the same way you have done them before/heard others doing them is functionally advantageous!) --David Tuggy From wilcox at UNM.EDU Fri Dec 17 17:10:45 1999 From: wilcox at UNM.EDU (Sherman Wilcox) Date: Fri, 17 Dec 1999 10:10:45 -0700 Subject: Linguistic motivation Message-ID: A request for suggestions: I'll be teaching a new course this spring on linguistic motivation and I'm putting together the syllabus and reading list. I have a course outline, topics, and several readings in mind. I'd appreciate, though, any suggestions that list members might have regarding articles or topics you would consider important to include. I'll be happy to summarize and post the responses back to this list (I am posting this message on both cogling and funknet). Sherman Wilcox University of New Mexico From W.Schulze at LRZ.UNI-MUENCHEN.DE Fri Dec 17 18:58:15 1999 From: W.Schulze at LRZ.UNI-MUENCHEN.DE (Wolfgang Schulze) Date: Fri, 17 Dec 1999 19:58:15 +0100 Subject: Function and convention Message-ID: > David Tuggy quotes Fritz Newmeyer: > > **** > Is 'convention' admissible as a category of functional motivation? (That > would raise a host of problems, no?) > **** Perhaps we should replace 'convention' by 'routinization' (based on 'ritualization') in order to avoid the intentional connotation, the term 'convention' seems to imply. I think that the grammar of a language has to be explained on the basis of the acquired cognitive and (cognition based) communicative practise of an individual integrated into a collective. This practise is dominated by massive hypotheses about the self-attachment to a collective; it RE-presents a strongly ritualized but construing interaction of the individual with environmental or world stimuli which corresponds to the habitus of a collective and which takes place in form of the tacit (poiematic) and/or articulate (pragmatic) activation of an acquired (and tradi-tional) knowledge system. Linguistic practise can be thought to represent the individual reaction to a collective communicative and cognitive standard which itself is predominantly historical in nature. By this is meant that the linguistic knowledge system of an individual and its instantiation in a 'communicative community' always reflects strategies of linguistic adaptation that have been functionalized long before the individual has acquired a given system. Hence, it can be argued that language as a ‘metaphysical’ phenomenon owns strong anachronistic features: It hardly ever meets the immediate synchronic needs of information processing and communication. It follows that functional and semantic aspects of language architectures are mainly to be explained in a diachronic perspective, though the potential to adopt newly established communicative and cognitive routines plays an important role in the dynamic potential of language systems (what I call 'Pragmatic Intervention' (PI)). The assumption of an anachronistic ontology of language systems has an important consequence for linguistic explanation: Contrary to some other cognitive approaches, the framework underlying these assumptions ('Grammar of Scenes and Scenarios' (GSS) does not establish a direct synchronic relationship between language systems and cognition. Language systems and cognitive activities are thought to be structurally coupled on the basis of a mainly diachronic relationship. From this it follows that routinization (or if you want 'conventionalization') plays a crucial role not only in the dynamics of linguistic functions and in the functional architecture on language systems, but also in the motivation of these functions emerging from the cognition<>communication interface. Wolfgang ***************************** Prof. Dr. Wolfgang Schulze Institut fuer Allgemeine und Indogermanische Sprachwissenschaft Ludwig-Maximilians-Universitaet München Geschwister-Scholl-Platz 1 D-80539 München Tel.: +89-21805343 / Fax: +89-21805345 Email: W.Schulze at lrz.uni-muenchen.de http://www.lrz-muenchen.de/~wschulze/ ***************************** From bresnan at CSLI.STANFORD.EDU Fri Dec 17 19:57:46 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Fri, 17 Dec 1999 11:57:46 -0800 Subject: OT and diachronic functionalism In-Reply-To: Your message of Fri, 17 Dec 1999 10:15:41 GMT. <385A0D42.C59427A2@eva.mpg.de> Message-ID: 1. Fritz Newmeyer asks whether I subscribe to (a), (b), or (c): a. Constraints are universal, and may or may not be functionally motivated b. Constraints are universal and functionally motivated c. Constraints are universal and are functional motivations themselves. ???? I am subscribing to (b) as a hypothesis [not a religious tenet]. 2. Fritz asks how (b) or (c) can be reconciled with the existence of divergent word order patterns in English: predominant right branching coexisting with the left branching genitive N construction. If harmonic consistency in branching direction is functionally motivated (as argued by Dryer, Hawkins, etc.), how can we explain disharmonic patterns except as dysfunctional historical relics that belie (b) above? The mistake here, I think, is to talk about the functional motivation of *constructions* (or sentence types) instead of the functional motivation of *constraints* (which is what I am talking about). Constraints may be functionally motivated and yet conflict with each other, as for example constraints favoring ease of perception conflict with constraints favoring ease of production. There are many dimensions of harmony or markedness all at work at the same time, and a specific syntactic pattern which is optimal may in fact violate many well-motivated constraints. Consistency of branching direction is certainly not the only motivation at work in determining word order. As all good functional/typological linguists know, across languages possessive constructions often have special syntax, particularly when involving inalienable possession (but not exclusively so, if I am not mistaken). Hierarchies of animacy and topicality have often been invoked to account for this. (Cathy O'Connor has a very nice OT analysis of a case of this in Northern Pomo, a completely dependent-marking language everywhere except in the syntax of possession, which is both head-marking and strongly influenced by animacy and discourse topicality factors.) Many formalists would say that such things are merely historical wrinkles or accidents, reflecting nothing systematic about language. But I'm not so sure... Joan From gthomson at GPU.SRV.UALBERTA.CA Fri Dec 17 23:18:08 1999 From: gthomson at GPU.SRV.UALBERTA.CA (Greg Thomson) Date: Sat, 18 Dec 1999 04:48:08 +0530 Subject: competence/performance In-Reply-To: <385A0D42.C59427A2@eva.mpg.de> Message-ID: At 10:15 +0000 12-17-1999, Martin Haspelmath wrote: >I'm somewhat surprised to see that Joan Bresnan dissociates herself from >the competence/performance terminology. OK, it probably has a lot of >baggage attached to it that I may not be aware of, but my impression is >that we all pretty much agree that there is a necessary conceptual >distinction between language use and language structure, or speech and >grammar, or more generally between processing and storage, or cognitive >events and cognitive patterns. I believe Chomsky has used phraseology along the lines of "a disticntion between what one knows and what one does with what one knows". For a long time I couldn't imagine how anyone could take issue with the validity of such a distinction. However, suppose there are comprehension mechanisms that just sit there until they encounter speech, and at that point they process it. Suppose also that there are other mechanisms that just sit there until a to-be-verbalized concept gets kicked into them, and at that point they just verbalize it. In that case, competence is the performance systems. The only distinction to be made is a distinction between when they are running and when they aren't running. If there is a third system in people in addition to the comprehension and production systems--one that specifies or characterizes the sentences of their languages and the structural properties of those sentences--what is it for? In answer to Tuggy's question regarding harmful Chomskyan ideas, I would cite this idea as a damaging one. In my own area of special interest, Second Language Acquisition, it has lots of people thinking about acquiring grammars, rather than thinking about developing comprehension and production mechanisms. Greg Thomson From dubois at humanitas.ucsb.edu Fri Dec 17 22:25:12 1999 From: dubois at humanitas.ucsb.edu (John W. Du Bois) Date: Fri, 17 Dec 1999 14:25:12 -0800 Subject: CSDL 5 Conference Announcement Message-ID: CSDL 5 -- PRELIMINARY ANNOUNCEMENT The 5th conference on: "Conceptual Structure, Discourse, and Language" will be held at the University of California, Santa Barbara, on May 11-14, 2000. The conference is sponsored by UC Santa Barbara's Center for the Study of Discourse, Cognitive Science Program, Department of Linguistics, and Department of Geography. Papers in all areas of cognitive linguistics and related research areas are welcome, including research on conceptual structure, grammar, meaning, cognitive processing, acquisition, language use, discourse function, and other issues. Papers are especially encouraged bearing on, but not limited to, the special conference themes of: Metaphor Analogy Irony Space Grammar and Cognition Discourse and Cognition Learning and Acquisition Interactionally Distributed Cognition Abstracts are due February 14, 2000. An abstract of 500 words should be submitted via email to Patricia Clancy at: pclancy at humanitas.ucsb.edu Additional information about the conference will be made available shortly via the Internet. For additional information contact John Du Bois at dubois at humanitas.ucsb.edu. From jrubba at CALPOLY.EDU Fri Dec 17 22:43:08 1999 From: jrubba at CALPOLY.EDU (Johanna Rubba) Date: Fri, 17 Dec 1999 14:43:08 -0800 Subject: On Chomsky Message-ID: In response to Dave Tuggy's query: Autonomy of syntax certainly seems a most problematic aspect of Chomsky's view of language for those of the cognitive/functional/discourse analysis persuasion (not that I feel that any of these three labels are necessarily mutually exclusive). These lines of investigation seek motivation for syntactic structures in various places -- discourse, typology, the nature of the language-producing organism, history of languages and grammaticization, and synchronic semantics. Seeking motivations other than the 'innate universals in a syntax-specific module' posited by generativism seems to be a defining characteristic of many non-generative approaches to language. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Johanna Rubba Assistant Professor, Linguistics English Department, California Polytechnic State University One Grand Avenue • San Luis Obispo, CA 93407 Tel. (805)-756-2184 • Fax: (805)-756-6374 • Dept. Phone. 756-259 • E-mail: jrubba at calpoly.edu • Home page: http://www.calpoly.edu/~jrubba ** "Understanding is a lot like sex; it's got a practical purpose, but that's not why people do it normally" - Frank Oppenheimer ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From chafe at HUMANITAS.UCSB.EDU Sat Dec 18 20:34:09 1999 From: chafe at HUMANITAS.UCSB.EDU (Wallace Chafe) Date: Sat, 18 Dec 1999 12:34:09 -0800 Subject: Problems with Chomsky In-Reply-To: <9912169453.AA945364013@router-8.camnet.com> Message-ID: David Tuggy's questions invite so many different answers from so many points of view that I imagine most subscribers to this list will throw up their hands and not even attempt to answer them. "Let me count the ways." But we shouldn't leave things with the answer that the only thing wrong with Chomsky's program is deep structure. Many of us, I'm sure, could write books on this subject if we wanted to, and the books would be quite different. But let me mention several general points that immediately occur to me, without trying to go into the detail they deserve. (1) The nature of the data. I said a little about this in a recent message to this list, so won't repeat it here. But it's worth adding that an extraordinarily restricted set of phenomena have been involved. (2) The focus on the "sentence" as the preeminent unit of language. (3) The assignment of too many phenomena to innateness. (4) The refusal or inability to explain linguistic phenomena in the context of other human endowments, whether cognitive (including memory, consciousness, imagery, emotions, etc.); social (clearly language is an interactive phenomenon); or historical (one can hardly understand the shape of language without taking account of language change). (5) The disregard for the fundamental importance of meaning or content, which can be traced to Bloomfield's infatuation with behaviorism and logical positivism as subsequently exaggerated by Harris. Language organizes thoughts as much as sounds, and in some ways thoughts are more important. In even more general terms, one might say that this vast, wonderfully beautiful human endowment we call language has been reduced to a few rather mundane phenomena of limited interest, over which a house of cards (or series of them) has been erected. Postscript: It would be easy to question some of the positive contributions listed by Brian, but I'll mention just the notion of "creativity", which has always puzzled me. The ability of language users to produce and understand novel sentences (or whatever) doesn't come from recursion, but from the insertion of a vast lexicon into a relatively small set of patterns. Yesterday my wife said to me "Don't mist the ribbon." It had to do with squirting water on a Christmas wreath. I doubt that she had ever said that before, and I certainly had never heard it before, but there was no problem (and no recursion). Wally Chafe From wilcox at UNM.EDU Sat Dec 18 21:07:12 1999 From: wilcox at UNM.EDU (Sherman Wilcox) Date: Sat, 18 Dec 1999 14:07:12 -0700 Subject: Problems with Chomsky In-Reply-To: Message-ID: On 12/18/99 1:34 PM, Wallace Chafe said: > Postscript: It would be easy to question some of the positive > contributions listed by Brian, but I'll mention just the notion of > "creativity", which has always puzzled me. Hurray! I've always thought that the Chomskyan perspective on linguistic creativity was just plain bizarre. For me, language creativity is more akin to musical creativity than it is to mathematical recursion. So I find the words of a fine jazz musician such as Clark Terry more insightful than Chomsky on the subject. In an interview on a recent reissue of an old jazz album (with Oscar Peterson), Terry commented on his performance of "Mack the Knife" (PP is the interviewer): ***** CT: Well, all I can say is, I didn't make too many obvious mistakes. I got through that one okay. When you give vent to your feelings, playing jazz, some of your better solos spin off from what could have been catastrophes. This is something that we learned in the ghetto, where there's varmints -- we used to say, "Damn rat, I'm going to fix him -- I know where his hole is." So we would chink up the holes with a lid from a tin can, nail it down. But he finds another way. If you're attempting to get your idea through one way and it's closed, you have to get it out, so you take another way. Like a rat going from one rat hole to the other, if you're trapped, you can't stop and say, "Let's do that over again." You have to figure out ways and means, through the medium of your having mastered the instrument. It's not like the classics. It's extemporaneous composition. ... It's avoiding catastrophe. PP: But you can always base what you're playing on the standard. CT: Oh, yes. You're playing the changes, and you do what the old-timers did before they knew anything about theory or harmony or counterpoint. They used the melody as a guy wire to extemporaneously superimpose another melody. That's [still] how cats develop ears, and are able to play by ear. But this was all that the old-timers had to go by. PP: ... so they improvised by necessity, because they didn't have or couldn't read music? CT: Yes. Just finding a way to get through the song helped to develop their creativity. ***** Not only does Clark Terry hint at Wally's "insertion of a vast lexicon into a relatively small set of patterns" and the "real" language that he discussed in a previous message, but Terry provides what I'd like to offer as an alternative definition of linguistic creativity: avoiding catastrophe in extemporaneous composition. Sherman Wilcox Department of Linguistics University of New Mexico From bresnan at CSLI.STANFORD.EDU Mon Dec 20 18:46:24 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 20 Dec 1999 10:46:24 -0800 Subject: OT and functionalism In-Reply-To: Your message of Mon, 20 Dec 1999 11:22:32 +0100. <3.0.5.32.19991220112232.00797e80@mail.hum.uva.nl> Message-ID: Dear Harry Perridion, Thank you very much for your remarks on Dutch phonology, which are obviously intended for the FUNKNET audience, in view of the third person form of address to me. You make the point that it is not literally true that Dutch devoicing occurs in coda position, because in that position but before b or d voicing assimilation occurs (if there is no pause). You comment: > It is remarkable that OT is capable of explaining non-existent facts. I think you are correct that the generalization about the distribution of voiced obstruents in Dutch is overridden by anticipatory voicing assimilation in the particular context you note. But you are mistaken in concluding that this fact somehow undermines OT phonology or the point I was using the Dutch example to make. Phonological systems actually have more than the three constraints I used in my example. In particular, there are very well motivated constraints that favor voicing assimilation. In Dutch, such a constraint overrides the positional markedness constraint against voicing contrasts in codas. The reference I gave for the Dutch devoicing example in my original message is Rene' Kager's 1999 CUP book _Optimality Theory_. There and in the references you will find an excellent discussion, with many illustrations, of how the interactions of various phonological processes can be explained more insightfully in OT than in previous rule-based terms. Best wishes, Joan Bresnan xxxxxxxxxx > In one of her contributions to the interesting discussion on OT and > functionalism on this list Joan Brennan wrote the following on final > devoicing in Dutch: > > >>For example, Dutch and German have a voicing contrast in obstruents, > which appears in syllable onsets--a very salient position-- but voiced > obstruents are devoiced in syllable codas. << > > This is not correct. The facts are as follows. Voiced obstruents are not > possible before a pause, or a voiceless obstruent. The 'f' in 'leefde' > (lived) e.g. is both syllable- and morpheme-final, but voiced, as is the > 'b' in 'ebde weg' (ebbed away). In fact, syllable-, morpheme- and even > wordfinal obstruents are voiced before /d/ and /b/, if there is no > intervening pause, e.g. 'op' [p] (on) + 'doen' (do) -> 'opdoen' (to put > on) with [-bd-]. > What we have here is a rather simple case of voice-assimilation, that is, > if we assume that a pause functions like a voiceless obstruent. > CC-sequences in Dutch (C is either a stop or a fricative) are either voiced > or voiceless. The assimilation is regressive, except when the second > obstruent is a voiced fricative (/z/ or /v/), in which case it is progressive. > > >>OT can explain the > positional neutralization of obstruent voicing in syllable codas and > relate it to the typological of segment inventories across languages. > The reason is simply that the *same* constraints reflecting ease of > perception and articulatory effort are present in the grammars of all > languages (as the universal constraint set). No other of grammatical > theory that I know of does this.<< > > It is remarkable that OT is capable of explaining non-existent facts. > > >>In contrast, there appear to be few or no languages whose > language-internal phonologies have a voicing contrast in syllable > codas and no voicing contrast in syllable onsets.<< > > To a certain extent, American English is such a language: there is a > contrast in syllable codas between voiced and voiceless stops ('bed' vs. > 'bet') but not in onsets of syllables that follow a syllable ending in a > vowel ('latter' vs. 'ladder'). > Such voicing of previously voiceless stops in intervocalic (but > 'syllable-initial') position is well-attested in the history of a number of > languages. > > Another fact that might disturb the serene picture of voice in the > languages of the world sketched by some proponents of OT and/or > typologists, comes from Scandinavian: > > In Southern Scandinavian (Danish, southern Swedish and southern Norwegian) > there is no opposition between voiced and voiceless stops in final position > after a long vowel, but in S-Swedish and S-Norwegian the merger is a voiced > stop. Danish developed in most cases a voiced fricative (Old Danish 'ut' > (out) -> 'ud' -> 'udh' ('dh' stands for the fricative). > > It is possible that these Germanic languages are all typologically weird, > but it is more probable that the assumption that: > > >>Across languages, voiced obstruents > are typologically marked: there are languages having only voiceless > obstruents (e.g. Polynesian), and languages having both voiced and > voiceless obstruents (e.g. English), but no (or hardly any) languages > having only voiced obstruents. This typological asymmetry might be > explained in terms of constraints on articulatory effort and > perception that shaped the evolution of languages to explain why > voiced obstruents are more restricted crosslinguistically. But within > the phonological components of the grammars of particular languages we > see the same thing: voiced obstruents are restricted in their > *language-internal* distributions. << > > is simply wrong. In fact it surprises me that respected linguists like Joan > Bresnan (see the quotations above) and Martin Haspelmath feel obliged to > make sweeping statements like > > >>Voiced obstruents are harder to pronounce than > voiceless ones, especially in (syllable- or word-)final position.<< > > What evidence is there for such a statement? Is it a fact that voiced > consonants are harder to pronounce?, in what way? and why? > > Let us get the facts straight before arguing about their interpretation. > > > Harry Peeridon > University of Amsterdam > > From edith at CSD.UWM.EDU Mon Dec 20 18:56:54 1999 From: edith at CSD.UWM.EDU (Edith A Moravcsik) Date: Mon, 20 Dec 1999 12:56:54 -0600 Subject: antonomy of syntax Message-ID: Johanna Bubba has suggested that the most problematic aspect of Chomsky's view of language is the notion of autonomy of syntax. But I cannot see how this could be an important issue of divisive force. If the claim that syntax is autonomous means that it can be described without reference to meaning (and function), then the claim is almost true by definition and thus I do not believe even functionalists would take issue with it. The claim is necessarily true in the sense that any object in the world that includes recurrent parts and/or properties can be described in terms of the distribution (selection and arrangement) of these basic elements. Sentences are analyzable into recurrent parts - such as words - and thus sentences can be given a distributionally based description without reference to the meaning of these parts. To claim the opposite - that sentences can be described unless we consider meaning - would be like saying that, while the structure of a string of beads can be described in terms of the choice and order of the beads, as soon as this string serves the purposes of a rosary, the former structural description becomes invalid and a valid description can be constructed only if it is known which bead stands for which prayer. This is, of course, not so. As I think was suggested by several people earlier in this discussion, the difference between functionalism and formalism seems to exist not on the descriptive but on the explanatory level; but even there, it is less than categorical. Functionalists claim that most or all structural features of sentences can be explained in terms of meaning (or function). Formalists in turn claim that most - but not necessarily all - structural characteristics are subject to form-related (i.e., non-semantic and non-functional) explanations - which in turn may or may not be ultimately functionally-based themselves. If this is a correct characterization of the difference between the two persuasions, the differences is not a huge one. Edith M. ************************************************************************ Edith A. Moravcsik Department of Foreign Languages and Linguistics University of Wisconsin-Milwaukee Milwaukee, WI 53201-0413 USA E-mail: edith at uwm.edu Telephone: (414) 229-6794 /office/ (414) 332-0141 /home/ Fax: (414) 229-2741 From jrubba at CALPOLY.EDU Mon Dec 20 20:44:40 1999 From: jrubba at CALPOLY.EDU (Johanna Rubba) Date: Mon, 20 Dec 1999 12:44:40 -0800 Subject: antonomy of syntax Message-ID: To respond to Edith, I was writing about _explanation_ in linguistics, not description. It was clear from early in the program that generative linguistics had explanation as its goal, whether or not the program has ever achieved this. I think it is also clear that it has long been a goal of generativism to prove the existence of an innate, syntax-specific module, _physically_ distinct from other parts of the brain that process language (Chomsky's 'language organ'), to _explain_ the facts of syntactic structure, and thus to minimize to zero if possible the involvement of semantics in the explanation of syntax. The description of this module's workings would then constitute an explanation of why syntax is the way it is. I did some browsing in some foundational works of cognitive and functional linguistics, and several make explicit mention of autonomy of syntax being a central tenet of generative linguistics which they reject, replacing it with a tenet that claims the opposite. These ideas are laid out in parts of the work which are intended to summarize the core, central aspects of the cognitive/functional program, so I assume this means the autonomy thesis is a major difference between the research programs, at least in the view of some of the founders of these alternative theories. Some quotes: "Central to [Cognitive Grammar's] conception of grammatical structure are three closely related claims, which define the focal concern of this book ... 2. Grammar (or syntax) does not constitute an autonomous formal level of representation. Instead, grammar is sysmbolic in nature, consisting in the conventional symbolization of semantic structure." Langacker, _Foundations of Cognitive Grammar_ Vol. 1, p. 2. Givon, in _Syntax: A functional/typological introduction_, lists among 8 core components of generative grammar, the following one: "a. Structure and function: Language -- and syntax -- were conceived of as structure, existing and understandable independently of meaning or function. 'Autonomous syntax' then consituted its own explanation even within linguistics ..." (p. 7). He goes on to say at the end of this list: "The approach to the study of syntax adopted in this book developed gradually as a rejection of all the tenets of the transformational-generative tradition as listed above" (p. 9). George Lakoff, in _Women, Fire and Dangerous Things_, devotes a brief chapter to "The Formalist Enterprise", tracing the origin of formal theories of syntax to the history of mathematics, logic, and philosopy. I quote: "The idea that natural language syntax is independent of semantics derives from the attempt to impose the structure of mathematical logic on the study of human language and human thought in general" (p. 225). He writes a short bit later "The question of whether there is an independent syntax for natural language comes down to the question of whether the metaphorical definition that defines the enterprise of generative grammar is a reasonable way to comprehend natural language. Intuitively the idea that a natural language is made up of ininterpreted symbols is rather strange ... if language is a way of framing and expressing thought so it can be communicated, then one would expect that many (not nedessarily all) aspects of natural language syntax would be dependent in at least some way on the thoughts expressed" (p.228). The fact that both 'camps' have had to compromise a bit due to problematic data (as Lakoff notes in the quote given here) doesn't minimize the importance of autonomy vs. non-autonomy. There may be practitioners of each kind of linguistics who take on this particular 'article of faith' with more or less confidence/enthusiasm, but the salience of discussions of autonomy in these foundational works seems to me to be evidence of major disagreement on this principle. It also nearly always comes up in cross-theory discussions that I have with generative theorists. A favorite challenge of generativists is to name some favorite point of syntax (co-reference, for example) and ask 'How are you going to motivate _that_ semantically?' ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Johanna Rubba Assistant Professor, Linguistics English Department, California Polytechnic State University One Grand Avenue • San Luis Obispo, CA 93407 Tel. (805)-756-2184 • Fax: (805)-756-6374 • Dept. Phone. 756-259 • E-mail: jrubba at calpoly.edu • Home page: http://www.calpoly.edu/~jrubba ** "Understanding is a lot like sex; it's got a practical purpose, but that's not why people do it normally" - Frank Oppenheimer ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From moorej at UCSD.EDU Mon Dec 20 22:02:27 1999 From: moorej at UCSD.EDU (John Moore) Date: Mon, 20 Dec 1999 14:02:27 -0800 Subject: defining generative grammar Message-ID: Jo Rubba writes: >I think it is also clear that it has long been a >goal of generativism to prove the existence of an innate, >syntax-specific module, _physically_ distinct from other parts of the >brain that process language (Chomsky's 'language organ'), to _explain_ >the facts of syntactic structure, and thus to minimize to zero if >possible the involvement of semantics in the explanation of syntax. The >description of this module's workings would then constitute an >explanation of why syntax is the way it is. This has been a long-standing point of contention here at UCSD, but it does not take the form one might suspect. Rather, those of us who are identified as formalists are constantly hit over the head with this autonomy thing from the members of the cognitive science community, who tend to favor cognitive/functional approaches. We, putative formalists, on the other hand answer: "but we don't necessarily believe in the autonomy of syntax - perhaps some do, but we don't." Jo was a student here many years ago, and must remember these exchanges. I find it interesting that Jo uses quotes from cognitive/functional linguists to bolster the idea that autonomy is a major tenet of formal grammar. I suspect that this is a position that is often attributed to formal linguistics by others, and not an integral part of the paradigm. During my graduate training in formal syntax at UCSC in the late 80s I never heard anyone make the above claims. Of course, it is true that perhaps a majority of formal syntacticians in the GB/P&P tradition believe something along the lines of the autonomy hypothesis. It is also true that work in formal syntax tends to not make reference to anything other than formal devices to account for syntactic data. However, this is clearly changing, and has been for some time (I'm struck by a number of functionalist works that argue, in the 1980s and 1990s, against the generative tradition, and cite only _Aspects_). We recently had a seminar on explanation here - it was interesting that the psychological reality/autonomy of a generative grammar was never mentioned. As far as I remember, the closest any of the formalist participants came to Jo's characterization of generative grammar were: (i) the proposal that formal principles should be maintained as constants and pushed to their limits in a deductive research strategy (papers by Baker and Borer). (ii) the claim that syntactic accounts are preferable because they are cast in a well-understood algebra, whereas the nature of functional principles invoke a less well-understood domain. (iii) a similar point that generative (as some functional) accounts are of particular value when they are explicit and make clear predictions. I realize that these three points are controversial (and I don't fully subscribe to i-ii); however, they are a far cry from the above characterization of generative grammar. >A favorite challenge of generativists is to name >some favorite point of syntax (co-reference, for example) and ask 'How >are you going to motivate _that_ semantically?' I've heard (and expressed this) somewhat differently - generative research has uncovered a lot of generalizations about language - the challenge for a cognitive/functional approach is to either capture similar generalizations or show that the generalizations are epiphenomenal. Conversely, of course, formal linguists need to take generalizations discovered by functionalists seriously - this is what I understand Joan and Judith to be doing. John Moore P.S. Bach and Partee sketched a semantic account of some binding facts in a CLS paper almost 20 years ago. Aspects of their proposal have been very influential in subsequent RG, LFG, HPSG works, as well as work by Will imams and Reinhart and Reuland. http://ling.ucsd.edu/~moore/ From MILLSCR at UCMAIL.UC.EDU Mon Dec 20 23:06:14 1999 From: MILLSCR at UCMAIL.UC.EDU (Mills, Carl (MILLSCR)) Date: Mon, 20 Dec 1999 18:06:14 -0500 Subject: antonomy of syntax Message-ID: There are a few of us here on the fringes for whom the exchange between Edith Moravcsik and Johanna Rubba, with an interesting sidebar by John Moore, on the autonomy of syntax sounds like a debate on how many agels can dance on the head of a pin. Not that this thread is not fascinating, but at a recent conference another internationally known linguist said something like, "Nobody is trying to do away with rules of grammar." To which I replied, "Some of us are." Syd Lamb has published several papers on the "Unreality of Syntax." Vic Yngve has argued that the notion of language itself is a historical inheritance that impedes understanding what we are talking about. Steve Straight has posted his views on "The Myth of G" to this list recently. Paul Hopper continues to publish and present work on emergent grammars. I have no doubt that syntax can be described autonomously. The question is "Why bother?" or rather, "What is gained and what is lost in linguistic theories that seem to require an autonomous syntactic module?" It would appear that for most linguists today the gains associated with an autonomous syntax outweigh the losses. But some of us have concluded the opposite. Carl Mills University of Cincinnati - -------------- next part -------------- An HTML attachment was scrubbed... URL: From nrude at ucinet.com Tue Dec 21 00:12:35 1999 From: nrude at ucinet.com (Noel Rude) Date: Mon, 20 Dec 1999 16:12:35 -0800 Subject: autonomy of syntax Message-ID: Hmm, To me, as in all the mind sciences, "emergence" and "supervenience" seem like covers for ignorance, merely reductionist "promisory notes": When we know enough about neurons and whatnot, then language will be accounted for bottom-up. But until we do, how can we be so sure that there is no top-down phenomena like grammar? Noel From sstraigh at BINGHAMTON.EDU Tue Dec 21 04:03:43 1999 From: sstraigh at BINGHAMTON.EDU (H Stephen Straight) Date: Mon, 20 Dec 1999 23:03:43 -0500 Subject: autonomy of syntax In-Reply-To: <385EC5E1.6B1@ucinet.com> Message-ID: Noel Rude ruminates: > To me, as in all the mind sciences, "emergence" and "supervenience" > seem like covers for ignorance, merely reductionist "promisory notes": > When we know enough about neurons and whatnot, then language will be > accounted for bottom-up. But until we do, how can we be so sure that > there is no top-down phenomena like grammar? Granted, we can't be sure that there is no overarching set of linguistic "rules" (Grammar), but we can be sure that the positing of such a "supervenient" entity violates the rule of parsimony. To pursue the study of language scientifically we must avoid positing any theoretical entities beyond those necessary for the pursuit of our inquiry. In this case, we know we need separate accountings of how people interpret the linguistic events they perceive and of how they create linguistic events. The Myth of G would have us accept the existence of a Grammar as a "source" of such events independent of the processes of reception and expression, and would have us ignore the evidence for discrepancies, separateness, dissociability, cognitive multiplexity, and other (often highly "creative") interactivity of receptive, expressive, and other language processes. Arguments for the existence of G have considerable intuitive appeal, but so do arguments for numerous other common-sensical and supernatural entities that rightly play no role in scientific accounts of reality. H Stephen Straight From haspelmath at EVA.MPG.DE Tue Dec 21 09:06:40 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Tue, 21 Dec 1999 09:06:40 +0000 Subject: the pseudo-issue of autonomy Message-ID: I agree with Edith Moravcsik and John Moore that autonomy is not the central issue that divides Chomskyans and functionalists. Just look at any descriptive grammar (quite a few functionalists have published such works) and examine the morphology section: Inflectional classes, various stem forms, morphophonemic alternations ? all this is systematic arbirariness, i.e. autonomy (in Newmeyer's 1998 definition). In my view, what often divides the two camps is that Chomskyans are primarily interested in solving Plato's Problem ('How can we acquire language?'), whereas functionalists are primarily interested in explaining language structure. Jo Rubba is right that "it was clear from early in the program that generative linguistics had explanation as its goal", but not explanation of language structure. Chomsky's "explanatory adequacy" is only about explaining language acquisition. The practical goal of spelling out the principles of UG is subordinate to the theoretical goal of solving Plato's Problem. From this it follows that functional explanations are anathema in generative linguistics ? because they undermine UG, and hence Chomsky's solution for Plato's Problem. Martin P.S. The above argument is presented in some detail in my review article on Fritz Newmeyer's (1998) book, due to appear in the next issue of Lingua. -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From W.Schulze at LRZ.UNI-MUENCHEN.DE Tue Dec 21 12:53:57 1999 From: W.Schulze at LRZ.UNI-MUENCHEN.DE (Wolfgang Schulze) Date: Tue, 21 Dec 1999 13:53:57 +0100 Subject: antonomy Message-ID: Johanna Rubba wrote: > (...) I did some browsing in some foundational works of cognitive and > functional linguistics, and several make explicit mention of autonomy of > syntax being a central tenet of generative linguistics which they > reject, replacing it with a tenet that claims the opposite. These ideas > are laid out in parts of the work which are intended to summarize the > core, central aspects of the cognitive/functional program, so I assume > this means the autonomy thesis is a major difference between the > research programs, at least in the view of some of the founders of these > alternative theories. With respect to the autonomy debate we should perhaps add another view that is related to the cognitive program. This view claims that 'language' is the output of cognitive routines that interprets communicative experience in its interaction with information processing as a systematic entity (called 'language'). 'Language' is regarded as a learned knowledge system that is imposed (by tradition) on its own cognitive prerogatives, a recursive process that is typical for human cognition. This view does not propose that the process of language acquisition starts at say the age 1.0 years _because_ of the 'fact' that the brain has arrived at a certain maturation state that establishes the language faculty. Rather, it is assumed that people have become used to 'interpret' a certain mental stage, namely the maturation of the corresponding senso-motoric domains as a communicative signal that a child is 'ready for language acquisition'. The collective (and very old) experience that it is best to start to 'linguistically' train children at a certain age has established the collective mental hypothesis (or idealized cognitive model) that language is something more or less autonomous. The assumption of a modular linguistic 'substance' (in what shape so ever) represents a collective mental construction and constitutes an important part of folk-psychology. In this sense, this view has to observe two determinatives with respect to the ontology of language: On the one hand, language is constituted by nothing but the emergent activities of the cognition communication interface that in themselves are not 'language' but senso-motoric schemata related to the complex (and polycentric) network of senso-motorics, audiovision, and information storage procedures. In this sense, 'language' does not have a proper substance (not to speak of 'essence'), but represents cognitive 'events' that acquire a 'communicative and linguistic reading' via mental constructions. On the other hand, the experience of these events together with their paradigmatization during language acquisition ends in some kind of systematic knowledge that is _construed_ as a more or less autonomous something. Both aspects are structurally coupled and lead to what we experience as 'language'. The hypothesis that 'language' has acquired an own mental 'substance' (in terms of Universal Grammar, part of the Language of Thought (LOT) or what so ever) could then be regarded as a highly sophisticated 'scientifization' of the popular hypothesis about language: It refers to a modular interpretation of cognitive organization in which modules are treated as the substantial 'resultant' of the evolution of cognition. This is exactly what people in ordinary life normally think about language (at least in a Western tradition). This is understandable if we bear in mind that in Western tradition language is strongly coupled with '(self-)consciousness' etc.: 'Language' constitutes one of the basic parameters of human ontology in Western folk psychology. It is declared to represent some kind of autonomous substance that is correlated with a certain stage in language acquisition. Traditions related to the autonomy hypothesis refer to this kind of mental construction. Naturally, this is not done 'consciously', saying: "Oh, let's make the pop linguistic experience scientific". In fact, the Chomskyan paradigm represents a unwanted (?) reaction to this folk experience, a process that in itself forms part of the scientifization of popular experience (or folk philosophy) which started (at the latest) in the time of enlightenment. It is 'modern' - whereas some cognitive approaches in the sense described above have more in common with 'post-modern' paradigms. In terms of the historiography of sciences (esp. linguistics) the Chomskyan paradigm is a 'function' of the 'functional/cognitive' paradigm: Post-modern linguistics (if you allow this term) depend on the experience of linguistic modernity, but it tries to overcome it in a dialectic discourse that 'explains Chomsky' just as Chomsky 'explained' pre-modern traditions. Modernity can only _react_ on post-modern arguments (out of logical reasons) by disclaiming approaches related to this paradigm as 'Neo-Hocuspocus' (if ever they do). Wolfgang -- ***************************** Prof. Dr. Wolfgang Schulze Institut fuer Allgemeine und Indogermanische Sprachwissenschaft Ludwig-Maximilians-Universitaet München Geschwister-Scholl-Platz 1 D-80539 München Tel.: +89-21805343 / Fax: +89-21805345 Email: W.Schulze at lrz.uni-muenchen.de http://www.lrz-muenchen.de/~wschulze/ ***************************** From edith at CSD.UWM.EDU Tue Dec 21 15:54:55 1999 From: edith at CSD.UWM.EDU (Edith A Moravcsik) Date: Tue, 21 Dec 1999 09:54:55 -0600 Subject: autonomy and related issues Message-ID: This is in response to a number of recent postings on the concept of the autonomy of syntax. Talmy Givo'n argued that the difference between the functionalist and formalist persuasions was rather large because it was one between an adaptive, evolutionary approach to language and a non-adaptive, anti-evolutionary approach. I think this would be the case if functional explanations were entirely excluded in the formalist approach; but I do not think this is so, for two reasons. First, formalists I believe do not claim that ALL grammatical phenomena are derivable from general laws of form; only some - or most - are. This leaves room for functional explanations. Second, there is no reason why general laws of form posited by formalists should not themselves be derivable from functionally-based generalizations. In fact, the very concept of innateness seems to me to be a functional principle (although not well-delimited). That is to say, if a functional explanation is one that makes reference to the goals and means of language use - i.e., the communicative and expressive needs of humans and the psychological and physical constraints under which these goals are to be achieved - than innateness is a functional "explanation" since it makes reference to psychological and physical contraints of the language user. The very interesting quotes that Johanna Bubba listed show that the autonomy of syntax has indeed been considered to be a divisive issue between formalists and functionalists. A further question, however, is whether it has been so by necessity - or, as John Moore put it, whether the two opposing views on this issue are "integral part/s/ of the /two/ paradigm/s/". I would think the answer is no. As I see it, a purely formal description - that is, one without reference to meaning and function - is actually a necessary part of any functionalist's agenda. As Martin Haspelmath pointed out, everybody's grammar, including functionalists', will make reference to purely formal concepts and this is understandable: if functionalist are interested in how form relates to meaning and function, the very question necessitates a characterization of form all by itself - simply because we cannot talk about the relationship between objects unless there is first a characterization of those object constructed independently of their relationship. If this point is correct, it would constitute an answer to Carl Mills' query re why construct autonomous syntactic descriptions. All in all, I am proposing two points: (a) A purely formal description of sentence structure is a necessary part of functional accounts because sentence form is the entity the relationship of which to other things functionalists are interested in. (b) Functional explanations for sentence form may not be necessary parts of formalist accounts but they are possible parts of them first, because not all syntactic phenomena may be explainable in terms of laws about form, and, second, because laws of form may themselves be derivable from functional considerations related to human perception and cognition. The notion that before we can talk about the relationship between things, we need to characterize those things independently of each other may also provide an answer to the question raised by Stephen Straight and also discussed by Wolfgang Schulze - namely, why posit the very concept of grammar to beging with? I think we need a characterization of grammar because what we want to know things about it: how grammar is acquired, how grammar changes (ontogenetically and historically), how it is stored, and how it is used. If these are our questions, then we do need a characterization of grammar that is independent of how it is acquired, changed, and used. In other words, both the concept of a strictly-formal account of sentence structure and also the concept of a grammar are entitites that we need in order for the questions that we are asking to make sense - namely, how does form relate to meaning and function and how is grammar acquired, stored, processed, etc. The only way, it seems to me, not to be obliged to construct autonomous characterizations of the notions "syntactic form" and "grammar" would be by changing our research questions so that they are not about the relationship of these constructs to other things. Is this correct? Edith PS I very much agree with Martin Haspelmath that the/a main difference between the Chomskian and the functionalist agend is one of what the basic question is - one about language structure or one about the acquisition of language. ************************************************************************ Edith A. Moravcsik Department of Foreign Languages and Linguistics University of Wisconsin-Milwaukee Milwaukee, WI 53201-0413 USA E-mail: edith at uwm.edu Telephone: (414) 229-6794 /office/ (414) 332-0141 /home/ Fax: (414) 229-2741 From moorej at UCSD.EDU Tue Dec 21 16:57:14 1999 From: moorej at UCSD.EDU (John Moore) Date: Tue, 21 Dec 1999 08:57:14 -0800 Subject: the pseudo-issue of Plato's problem Message-ID: At 09:06 AM 12/21/99 +0000, Martin Haspelmath wrote: >In my view, what often divides the two camps is that Chomskyans are >primarily interested in solving Plato's Problem ('How can we acquire >language?'), whereas functionalists are primarily interested in >explaining language structure. Unfortunately, this may be true. However, if one looks at what formal linguists actually do, there is very little solving of Plato's problem, and a lot of explaining (or at, least accounting for) language structures (gratuitous references to "The Child" notwithstanding). Occasionally one finds an epistemological introduction in formal papers, but I just skip it and get directly to the linguistics, which is often very good. While perhaps not be typical of formal linguists' beliefs, at least some take a very agnostic stand on what the deep psychological explanation for language structures may be. This was overtly stated in GPSG work (I think Geoff Pullum has a Topic Comment essay on this issue). Joan can speak to the strong competence hypothesis in LFG, but I don't see the same kind of arm-chair psychology in LFG work that one sometimes encounters in some GB and P&P work; one never found it in RG work, as far as I know. Finally, I suspect that it is less common even in GB/P&P work than some of Chomsky's writings might lead us to believe. Again, Plato's problem was never mentioned by any of the formalists at the explanation conference. If we strip away these issues, which I think are rather peripheral to the enterprise, we may find that the differences between formal and functional linguistics are fewer than we thought - perhaps some methodological ones, but even those currently divide functional and cognitive linguistics (and they seem to get along). John Moore From macw at CMU.EDU Tue Dec 21 17:20:11 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Tue, 21 Dec 1999 12:20:11 -0500 Subject: autonomy, Plato's Problem, and deep structure Message-ID: I would like to second Martin Haspelmath's take on the importance of Plato's Problem in the definition of the Chomskyan program. Since 1965, Chomsky has emphasized the role of the Logical Problem of Language Acquisition (LPLA) in generative theory. The formulation of the LPLA was an intellectual move necessitated by the commitment in 1957 in Syntactic Structures to deep structure. The "discovery" of deep structure was, in turn, the achievement that justified the introduction of transformations and the application of automata theory to language. As a result of these linkages, the commitment to deep structure became central to the Chomskyan program. In Plato's cave, true ideas are only seen as reflections. However, we often don't really need to look at the linguistic input at all, according to Chomsky, since the true shape of the deep structure of language is resident in our minds from birth. It is possible to imagine routes to deep structure that do not go through Plato's cave. Competence-performance and introspection regarding degrees of grammaticality provide one such route. I agree with Martin, Edith, and others that the autonomy of syntax is not a crucial support for deep structure. For example, Freud's deep structure is nicely grounded on claims about biological drives and developmental principles. Freud's example shows that there is no reason in principle that a functionalist theory in linguistics could not be grounded on a deep structure that was surrounded by a shell of protective concepts such as competence-performance, LPLA, sentences instead of utterances, the ideal speaker-hearer, degrees of grammaticality, and the rest. However, I doubt that many of us would be attracted to a functionalist theory of this type. For this reason, I think that a willingness to minimize the role of an abstract, underlying deep structure that has no easy match to observed facts about learning, processing, and physiology is the fatal flaw in the Chomskyan program. Having said this, I wonder whether Martin and Edith really want to say that the functionalist needn't care much about language acquisition. I agree that people in language development have not done a good job of showing functionalists why learning is important. But, if functionalism wants to achieve explanation, it cannot ignore learning. --Brian MacWhinney From noonan at csd.uwm.edu Tue Dec 21 23:34:34 1999 From: noonan at csd.uwm.edu (Michael Noonan) Date: Tue, 21 Dec 1999 17:34:34 -0600 Subject: job in Punjabi linguistics Message-ID: Below is a job announcement for a position in South Asian Studies at the University of Wisconsin-Milwaukee. The ad is fairly generic, covering a number of fields, but the position there described could be filled by a linguist whose specialization is the Panjabi language. If you have questions about the position, feel free to contact me. However, do not send application materials to me. The job ad copied below specifies to whom applications should be sent. Mickey Noonan Michael Noonan Professor of Linguistics Chair Office: 414-229-4539 Dept. of English Fax: 414-229-2643 University of Wisconsin Messages: 414-229-4511 Milwaukee, WI 53201 Webpage: http://www.uwm.edu/~noonan USA University of Wisconsin - Milwaukee South Asian Studies The College of letters and Science invites applications for a tenure track assistant professor position in South Asian Studies beginning Fall, 2000. In particular, the College seeks a person whose teaching and research interests cover one or more of the following areas: language, culture, civilization, or economic development. A strong preference will be given to candidates whose research and teaching interests include Punjab, India and Sikhism. The successful candidate will have a broad knowledge of the region, and his/her tenure home will be in one of the following departments: Economics, English, History, Political Science, Sociology, or Foreign Languages and Literature. The candidate will be expected to teach both undergraduate and graduate courses and direct graduate research. The successful candidate should be able to demonstrate evidence of excellent research potentials and teaching ability. Applicants must have a completed the Ph.D. by August 2000. ABD candidates may be eligible for a lecturer position. Candidates should submit a letter of intent, curriculum vita, research papers, teaching evaluations and three letters of recommendation by January 24, 2000. The University of Wisconsin - Milwaukee is an equal opportunity institution committed to diversity. Contact: Swarnjit S. Arora, Chair South Asian Faculty Search Committee, Department of Economics, Bolton Hall, University of Wisconsin - Milwaukee, WI, 53201. From jl.mackenzie at LET.VU.NL Wed Dec 22 14:03:56 1999 From: jl.mackenzie at LET.VU.NL (J.L. Mackenzie) Date: Wed, 22 Dec 1999 15:03:56 +0100 Subject: Bibliography of Functional Grammar Message-ID: I am pleased to announce the publication of "Functional Grammar Publications 1978-1998", compiled by Casper de Groot and Hella Olbertz. This is a complete overview of the work of Simon Dik and his co-workers that appeared between 1978 and 1998. The bibliography, which contains 1141 alphabetically arranged entries, also has a subject index and a language index. The bibliography has appeared as Working Papers in Functional Grammar 72 (ISSN 0924-1205) and is available for NLG 10 (Dutch guilders) from the executive editor: Aletta Smits, IFOTT, Faculty of Humanities, University of Amsterdam, Spuistraat 210, 1012 VT Amsterdam, Netherlands (e-mail: fg at hum.uva.nl; fax: +31-20-5253052). Other WPFGs that appeared this year were: 67 Dik Bakker, "FG expression rules: from templates to constituent structure" 68 Kwee Tjoe Liong, "Questions in the quasi-productive mode of the Functional Grammar model" 69 Ahmed Moutaouakil, "Exclamation in Functional Grammar: sentence type, illocution or modality?" 70 Kwee Tjoe Liong, "Adverbial clauses, FG, and the change from sentence grammar to discourse-text grammar" 71 Dik Bakker & Ewald Hekking, "A functional approach to linguistic change through language contact" All enquiries regarding WPFGs should be addressed to Aletta Smits at the address above. Lachlan Mackenzie Editor WPFG From W.Schulze at LRZ.UNI-MUENCHEN.DE Wed Dec 22 17:39:35 1999 From: W.Schulze at LRZ.UNI-MUENCHEN.DE (Wolfgang Schulze) Date: Wed, 22 Dec 1999 18:39:35 +0100 Subject: Language acquisition and functionalism Message-ID: Dear Brian and Funknetters, the recent comments upon the Chomskyan program have rightly focused on the role that language acquisition plays in this program. The chain of arguments saying that (1) deep structure presupposes universal categories and universal mental 'operations' in terms of Universal Grammar, (2) UG presupposes a language acquisition device (LAD), and (3) LAD presupposes its embedding in something like the evolution of a 'Language of Thought' (LOT) yields at - if I understand it correctly - the 'reconstruction' of UG to establish a 'meta-empiric' (or meta-descriptive) domain of research. In this sense, however, the 'reconstruction' of the LAD represents an _intermediate_ step in a much larger prospectus, namely the 'reconstruction' of LOT. It _may_ play the most important part in the game just because language acquisition procedures seem to be accessible to direct empirical observation, contrary to UG, LOT and LAD itself that can only be indirectly accessed via primary (UG) and secondary (LOT) extrapolation. But why do 'functionalists' have to refer to language acquisition (or, better, 'learning')? Is it just because the Chomskyan paradigm has opened a perspective so far ignored by functionalists? In this case, the Chomskyan paradigm would yet again represent the argumentative 'anchor' for 'deciding' which explanatory and/or descriptive domains of research have to be accessed within the paradigm of functionalism. In other words: Is it that we have to deal with language acquisition just because Chomsky did so (since 1965)? The observable tendency, namely that functional argumentation is sometimes (even often?) based on the 'reflex' to _react_ on Chomskyan hypotheses, to accommodate these hypotheses to the program of functionalism, and - in a second step - to declare the outcome as an intrinsic part of functionalism has represented a salient aspect in the history of (younger) functionalism (say since roughly 1961 (Dobbs Ferry)]. This is nothing new and it is self-evident if we consider the 'private histories' a number of (now) functionalists. It is logical, if we remember the fact that 'Standard Functionalism' (to use a cover term) is a program that is younger than the MIT 'orthodoxy'. It is logical, too, if we bear in mind that for the first time in the history of linguistics, two scientific paradigms coexist (in a more or less friendly manor) for more than 40 years by now. The scientific discourse between these two paradigms, however, is rather unilateral, which means that functionalists tend to carefully watch the activities of the East Pole (to quote R. Hudson) whereas the MIT orthodoxy usually refers to the output of the West Pole in an rather unspoken manor (if ever they do). Naturally, functionalists cannot deny the existence of the MIT paradigm. They cannot start from zero. This would be both uneconomical and ahistorical. However, what they can do is formulating a research program that would allow to tell people _why_ functionalism has to deal with some of the major issues also discussed by the MIT tradition. In other words, functionalism needs arguments that are 'deduced from it own (unspoken) deductions'. Let me again address the question "why functionalists should deal with language acquisition". Does 'functionalism' (what ever this means) have a program from which we can derive an answer to this question saying 'yes!'? Brian is probably right saying that "if functionalism wants to achieve explanation, it cannot ignore learning". Obviously, 'explanation' forms a major point in this program yet far from being fully formulated. But what can 'language acquisition' tell a functionalist with respect to the explanation of linguistic data (or, communication etc.)? I for myself am sure that 'learning' (as well as the accommodation of learning experience and maturation of brain structures) represents a decisive point even in functional typology, just as for language change and formal/functional diachrony (not to speak of language contact: The linguistic interaction of children and parents etc. conform to many aspects of language contact). However, is most cases functionalists seem to refer to language acquisition only in case this explanatory domain is thought to be adequate (and others fail). What we need (in my mind) is a more general (but 'polycentric') frame work for functionalism that explains _why_ domains like language acquisition form a necessary part of the functional paradigm. Merry Christmas and a Happy New Year to all, Wolfgang ***************************** Prof. Dr. Wolfgang Schulze Institut fuer Allgemeine und Indogermanische Sprachwissenschaft Ludwig-Maximilians-Universitaet München Geschwister-Scholl-Platz 1 D-80539 München Tel.: +89-21805343 / Fax: +89-21805345 Email: W.Schulze at lrz.uni-muenchen.de http://www.lrz-muenchen.de/~wschulze/ ***************************** From wcmann at JUNO.COM Wed Dec 22 22:25:15 1999 From: wcmann at JUNO.COM (William Mann) Date: Wed, 22 Dec 1999 17:25:15 -0500 Subject: submission deadline for Soc. for Text and Discourse: now Jan. 10, 2000 Message-ID: Dear Funknet subscribers: At their request, I am forwarding the information below concerning a conference of the Society for Text and Discourse in Lyon, France. (ST&D publishes Discourse Processes.) Bill Mann ====================================================== Dear Colleagues, The deadline for submitting a proposal to the 10th Annual Meeting of the Society for Text and Discourse in Lyon has been extended to January 10th, 2000. All the information on the meeting are available on the WebSite: http://unpc.univ-lyon2.fr/STD/index.html Have a good Christmas vacation Regards, Isabelle Tapiero [SORRY IN CASE OF MULTIPLE RECEPTIONS OF THIS MESSAGE] Dr. Isabelle Tapiero Chairman of the Scientific Committee for the 10th Annual Meeting for the ST&D E.mail:Isabelle.Tapiero at univ-lyon2.fr From nrude at ucinet.com Sun Dec 26 03:44:37 1999 From: nrude at ucinet.com (Noel Rude) Date: Sat, 25 Dec 1999 19:44:37 -0800 Subject: autonomy and related issues Message-ID: Greetings all, Yes to my mind Edith sums it up well. Though I might add that Givon has never repudiated "transformations" per se. In fact (I've heard him say) the Chomskians ought have kept them. Transformations are simply a mathematical way of relating structures to one another. The notion, e.g., has given us more precision in describing voice constructions, though most of us would assume the Chomskians' "autonomous" level of deep structure superfluous. In our tradition transformations mediate between "surface structure" and semantic case roles and discourse/pragmatics. Such "transformations" can be thought to preserve semantic structure and meaning, but as functionalists we will note that they add a discourse/pragmatic perspective. Noel From nrude at ucinet.com Sun Dec 26 04:10:37 1999 From: nrude at ucinet.com (Noel Rude) Date: Sat, 25 Dec 1999 20:10:37 -0800 Subject: autonomy of syntax Message-ID: Prof Straight says: "The Myth of G would have us accept the existence of a Grammar as a 'source' of [interpreting and creating linguistic events] independent of the processes of reception and expression, and would have us ignore the evidence for discrepancies, separateness, dissociability, cognitive multiplexity, and other (often highly 'creative') interactivity of receptive, expressive, and other language processes. Arguments for the existence of G have considerable intuitive appeal, but so do arguments for numerous other common-sensical and supernatural entities that rightly play no role in scientific accounts of reality." Probably I don't understand what Prof Straight is saying. Or maybe we can agree to disagree, perhaps mostly in what Science is. For it seems to me that theories can be very abstract and "supernatural" (if you will). If they are predictive-refutable--as our grammars and Grammar should be--they are "scientific". Also it seems to me that the messiness of linguistic data, rather than refuting an underlying system, actually suggests it, whatever flaws there might be in Saussure's langue et parole and Chomsky's competence-performance models. Is this just an esoteric argument where in practice we come down to the same thing? Will we both draw up verbal and nominal paradigm charts, describe grammatical relations, posit functions, etc., and some of us will call it "grammar" and "rules" and others will call it something else? Aren't we all looking for regularities--whatever we might call them? Noel From john at RESEARCH.HAIFA.AC.IL Sun Dec 26 14:04:53 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Sun, 26 Dec 1999 16:04:53 +0200 Subject: Who can you 'love'? Summary Message-ID: Dear Funknetters, I have finished my little survey of who people can 'love' aside from romantically, through family ties, and hyperbolically ('I just LOVE Frank Sinatra')--in other words, friends. I'm sorry I took so long to put together the results. I foolishly sent out the question when I didn't really have time to deal with it, so it took me a while to get answers I could put into a framework for presentation. Of course, the biggest problem, aside from generally small numbers, is that the respondents were self-selected. Oh well. Anyone else out there who feels like giving me more data with their own usage, I would welcome your contribution very much. Here are the results; thanks very very much to those of who who provided the data: In the end, I received 13 responses. I guess 14, if I count myself. Actually in the end there were 9 men and 5 women (presumably as a result of me specifically encouraging men to respond after my first attempt). Of these, we can make a general division into three categories: (1) For 4 people, all males, they would not use the word 'love' for anyone outside their family (I include in this group a gay male who said he could use the word 'love' for one `gay personal friend, with whom I never had a sexual relationship--the relationship was more like he was a son to me.)' (2) For 2 people (1 male, 1 female), there seemed to be a more or less open set of non-family, non-relatives they 'love.' One was a male who describes himself as a Christian `as a result of commitment to basic truths of Christianity', who approximated the number of people he 'loves' at several dozen. The other is a woman who attributes her understanding of 'love' to the fact that she lives in Southern California (`I am in Southern California and here you can love anything or anyone-from bare acquaintances to broccoli-(similarly, nothing is fine it is either great, wonderful, or fantastic)'); interestingly, when she described her understanding of 'love' (she wrote `I'm not sure it is actually a semantic difference but a pragmatic one; in other words, if the 'Default" is loving someone, what is that attitude of the speaker towards the referent if he/she merely 'likes' that person?'), it sounded remarkably similar to what Israelis have said to me about ''ahav'. (3) For the remaining 8 people (4 females, 4 males), 'love' can be applied to a limited number of very close friends. All of the women can use it towards either males or females, but all use it for more females than males. For the men, on the other hand, there is no pattern-one uses it only for one male friend, one for 2-3 female friends, one for 7 male friends and one female friend, another for mostly female friends. There seems to be considerably more similarity in female usage than male usage. There also seemed to be, initially, considerably more willingness on the part of women to volunteer information about themselves before I made my second posting. On the other hand, once I had gotten men to participate, some were surprisingly forthcoming and much more specific than I had expected in terms of exactly who they love and who they don't love. The impression I had is that the women seemed to know what to answer, and they answered in a generally similar way, whereas the men seemed to be applying some criteria which they had made up for themselves somewhere along the way, which they didn't agree on among themselves at all (after having conducted the survey, I realized that perhaps part of the reason I feel that I only 'love' people in my family is that otherwise I would have to make up some other criteria for non-romantic 'loving', and I have no idea of what these might be). I seem to be saying 'seem' a lot in this report. Thanks again for the input. John From dgolumbi at PANIX.COM Sun Dec 26 23:36:06 1999 From: dgolumbi at PANIX.COM (David Golumbia) Date: Sun, 26 Dec 1999 18:36:06 -0500 Subject: autonomy of syntax In-Reply-To: <38659539.1B65@ucinet.com> from "Noel Rude" at Dec 25, 1999 08:10:37 PM Message-ID: here's a sentence from a recent posting that, for me, captures the general intuition i find most troubling in both Chomskyan and non-Chomskyan generativist approaches: > Also it seems to me that the > messiness of linguistic data, rather than refuting an underlying system, > actually suggests it, whatever flaws there might be in Saussure's langue > et parole and Chomsky's competence-performance models. Not just that the messiness would *accommodate* or *allow* a system, but "suggests" it -- how have we got to the point where this seems a reasonable interpretation of the world's barely-documented, hardly known linguistic diversity? -- dgolumbi at panix.com David Golumbia From kemmer at RUF.RICE.EDU Wed Dec 1 02:15:19 1999 From: kemmer at RUF.RICE.EDU (Suzanne E Kemmer) Date: Tue, 30 Nov 1999 20:15:19 -0600 Subject: Doctoral Fellowships at Rice University Message-ID: DOCTORAL FELLOWSHIPS IN LINGUISTICS AT RICE UNIVERSITY The Department of Linguistics at Rice University announces the opening of competition for its doctoral fellowships for 2000-2001. The Ph.D. program at Rice emphasizes the study of language use, the relation of language and mind, and functional approaches to linguistic theory and description. A strong component of the program is field studies in particular language areas. Areas of intensive research activity in the department include cognitive/functional linguistics, language universals and typology, language change and grammaticalization studies, lexical semantics, corpus linguistics, computational modelling, neurolinguistics, phonetics, sociolinguistics, and second language acquisition. Interdisciplinary opportunities are available with the Ph.D. programs in Cognitive Psychology, Philosophy, Anthropology, the interdisciplinary group in Cognitive Sciences, and the Center for Cultural Studies. The department hosts a distinguished speakers series, whose recent and imminent speakers include Marianne Mithun, Wallace Chafe, Tom Givon, Megan Crowhurst, and Knud Lambrecht. The department also sponsors a biennial Symposium on Language. The upcoming Symposium in April 2000 is "Causation and Interpersonal Manipulation in the Languages of Central and South America." The last two symposia were "The Interface between Comparative Linguistics and Grammaticalization: Languages of the Americas" (1997) and "Usage-Based Models of Language" (1995). Speakers have included Bernd Heine, Alexandra Aikhenvald, Berend Hoff, Ronald Langacker, Joan Bybee, Brian MacWhinney, Janet Pierrehumbert, Douglas Biber, Tom Givon, John Du Bois, Mira Ariel, and Arie Verhagen. FACULTY AND RESEARCH INTERESTS Michel Achard, Ph.D. Linguistics, University of California, San Diego. Cognitive linguistics, French syntax, second language acquisition. Michael Barlow, Ph.D. Linguistics, Stanford University. Grammatical theory, corpus linguistics, computational linguistics, second language acquisition, discourse. James Copeland, Chair, Ph.D. Linguistics, Cornell University. Functional linguistics, Germanic linguistics, grammaticalization, American Indian linguistics (Tarahumara). Philip W. Davis, Ph.D. Linguistics, Cornell University. Semantics and syntax, language and intelligence, Amerindian (Bella Coola; Alabama), Austronesian (Atayal, Ilokano, Yogad inter alia). Spike Gildea, Ph.D. Linguistics, University of Oregon. Diachronic syntax, field methods and ethics, phonology, typological/functional linguistics, Amazonian linguistics, Cariban languages (Akawayo, Arekuna, Tiriyo). Suzanne Kemmer, Ph.D. Linguistics, Stanford University. Cognitive linguistics, typology and universals, lexical semantics, semantics of grammar, syntactic and semantic change, Germanic, Austronesian, Nilo-Saharan (Luo). Sydney Lamb, Ph.D. Linguistics, University of California, Berkeley. Cognitive linguistics, neurolinguistics, neural network modelling, Amerindian (Monachi). E. Douglas Mitchell, Ph.D. Linguistics, University of Texas at Austin. Comparative Indo-European linguistics, historical linguistics, history of linguistics, early Germanic dialects, Sanskrit. Nancy Niedzielski, Ph.D. Linguistics, University of California, Santa Barbara. Phonetics, digital speech processing, language and society, American dialectology. Stephen A. Tyler, Ph.D. Anthropology, Stanford University. Cognitive studies, philosophy of language, anthropological linguistics, languages of India (Koya). FINANCIAL AID Graduate fellowships include tuition and a cash stipend. Fellowships are normally renewable for four years upon satisfactory performance, and students can apply for a fifth year of support. The department has so far been fortunate to be able to support all its graduate students. RICE UNIVERSITY Rice University, founded in 1912, is a private university dedicated to the promotion of arts and letters, science, and engineering. The university is highly selective, and departments are small and focused. The campus is spacious, tree-lined, and architecturally distinctive (a blend of Mediterranean and Renaissance). Rice is a close-knit academic community and the Department of Linguistics in particular offers opportunities for personalized interaction and collaboration with faculty. Current enrollment is ca. 2700 under- graduates and 1,200 graduate students; faculty:student ratio is 1:9. Houston is the America's fourth largest city and offers the full array of urban amenities (fine arts, large city parks, museums...). It is ethnically extremely diverse (affording not only excellent opportunities for working with linguistic consultants, but also a huge number of restaurants representing a wide spectrum of cuisines at all levels of affordability.) The university is 45 minutes from the Gulf Coast (Galveston Island). Rents are affordable on a graduate stipend. The university and department offer a wide range of computing facilities available to students. The library has an outstanding linguistics collection, including a vast array of reference grammars. Doctoral students are eligible for support for travel to conferences and for summer research funds. The department also supports linguistic consultant fees and photocopying accounts for its doctoral students. Both U.S. and international applicants are admitted on the same basis. Current graduate students include not only U.S. students but also students from Australia, Brazil, China, and Germany. Prospective students of diverse linguistic backgrounds are encouraged to apply. Visiting students with their own funding who would like to come to Rice for a limited time to work with an individual faculty member should contact that faculty member directly. APPLICATION DEADLINE: February 1, 2000. Prospective applicants for the Ph.D. program must take the Graduate Record Examination as soon as possible, and have the results sent to the university in time for consideration in February. Non-native speakers of English must also take the TOEFL test. Admission is competitive. For more information about the program and the application process, please contact: Department of Linguistics, MS 23 Rice University P.O. Box 1892 Houston TX 77251-1892 (713) 527-6010 Departmental Coordinator: Ursula Keierleber, ukeie at ruf.rice.edu Graduate Adviser: Philip Davis, pwd at ruf.rice.edu See also the home page at http://www.ruf.rice.edu/~ling From brucerichman at HOTMAIL.COM Wed Dec 8 15:38:29 1999 From: brucerichman at HOTMAIL.COM (bruce richman) Date: Wed, 8 Dec 1999 07:38:29 PST Subject: Announcement: Alternatives to Chomsky meeting Message-ID: ALTERNATIVES TO CHOMSKY A NEW PARADIGM FOR LANGUAGE STUDIES FOR A NEW MILLENIUM A Conference to be held on September 4, 2000 at Rutgers University in New Brunswick, New Jersey, in association with the Language Origins Society year 2000 meeting. The main obstacle that we have today to clearly understanding the nature and origins of language is the overly formalistic, anti-empirical, anti-historical influence of Chomsky's paradigm for doing linguistics. It has become very clear in recent years that Chomsky's generative calculus model of linguistics has no relevance at all to anything about actual language. It has also become clear that the main generative notion of the innateness of language is based on flimsy, non-existent evidence. Despite this, and despite the fact that many people in many different fields are now actually studying real language in realisitic, empirical ways, and despite the fact that many people intuitively understand the emptiness of generative claims and practices, it still remains true, that to the world at large, Chomsky's theories are somehow considered an important "scientific breakthrough." So, the time has come that those of us who want to start a new paradigm for language studies, who want to begin an empirical way of studying real language, should simply explain why the entire Chomskyan method must be thrown away. We are planning to invite 7 or 8 people from a wide range of fields to speak at our one-day seminar which will be held on Monday, September 4, just prior to the regular meetings of the Language Origins Society. (LOS 2000 will continue on through Saturday, September 9.) If you are interested in participating please send a brief abstract of what you would like to present to: Bruce Richman e-mail brucerichman at hotmail.com 3805 Woodridge Rd. Cleveland Hts., Ohio 44121 216-381-7510 ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From moorej at UCSD.EDU Wed Dec 8 23:34:51 1999 From: moorej at UCSD.EDU (John Moore) Date: Wed, 8 Dec 1999 15:34:51 -0800 Subject: Alternatives to Chomsky Message-ID: At 07:38 AM 12/8/99 PST, bruce richman wrote: > It has become very clear in recent years that Chomsky's generative >calculus model of linguistics has no relevance at all to anything about >actual language. It has also become clear that the main generative notion >of the innateness of language is based on flimsy, non-existent evidence. > Despite this, and despite the fact that many people in many different >fields are now actually studying real language in realisitic, empirical >ways, and despite the fact that many people intuitively understand the >emptiness of generative claims and practices, it still remains true, that to >the world at large, Chomsky's theories are somehow considered an important >"scientific breakthrough." > So, the time has come that those of us who want to start a new paradigm >for language studies, who want to begin an empirical way of studying real >language, should simply explain why the entire Chomskyan method must be >thrown away. I agree that there are severe problems with the way much work within the Chomskyan paradigm has been conducted, but I wouldn't agree that this means that it "has no relevance at all to anything about actual language." Furthermore, very good empirical work has come out of the generative tradition. I would stop and consider what would be lost by throwing the entire generative methodology away. Finally, there are a number of generative frameworks. Do the above comments apply equally to P&P, Minimalism, LFG, HPSG, OT., etc.? How much of this is going to be thrown away in this sweeping paradigm shift - do any of these approaches offer any insight that one might want to keep? John Moore http://ling.ucsd.edu/~moore/ From lakoff at COGSCI.BERKELEY.EDU Thu Dec 9 05:05:41 1999 From: lakoff at COGSCI.BERKELEY.EDU (George Lakoff) Date: Wed, 8 Dec 1999 22:05:41 -0700 Subject: Bruce Richman's announcement In-Reply-To: <19991208153829.48152.qmail@hotmail.com> Message-ID: In case anyone is interested, I'm attaching a copy of the overheads from an all-too-brief and oversimplified talk I gave last weekend at the UCSD Conference on Explanation in Linguistics. Jerry Feldman and I are currently writing a book on the emerging Neural Theory of Language that we are working on at Berkeley, with collaborators at the University of Chicago and SRI.This is intended to supplement the already existing field of Cognitive Linguistics. For a selected bibliography, see the references in Philosophy in the Flesh, by myself and Mark Johnson. Also take a look at Mike Tomasello's THE NEW PSYCHOLOGY OF LANGUAGE. At the International Cognitive Linguistics Conference in Stockholm last year, about 750 researchers from 34 countries attended. We don't need to start an alternatives to Chomsky movement. Such a worldwide movement already exists, with a huge body of work, a journal, a number of book series, an international society, regular conferences, and so on. I suspect that most ICLA members would probably agree with Richman's negative assessment of Chomskyan linguistics. But we are interested in positive research, not negative polemics. A huge amount of such research is going on at an extermely high level. It's an exciting field -- at least as exciting as generative linguistics was back in the 60's, but far more profound. Anyone can join in. But there is a bit of literature to catch up on. I'm reading as fast as I can to keep up. I wish Richman well with his conference and hope it takes a positive form. George Lakoff -------------- next part -------------- A non-text attachment was scrubbed... Name: UCSD_Overheads_Revised_2 Type: application/mac-binhex40 Size: 180164 bytes Desc: not available URL: From jlmendi at POSTA.UNIZAR.ES Thu Dec 9 11:38:50 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Thu, 9 Dec 1999 12:38:50 +0100 Subject: Alternatives to Chomsky In-Reply-To: <3.0.3.32.19991208153451.009c5610@ling.ucsd.edu> Message-ID: At 07:38 AM 12/8/99 PST, bruce richman wrote: > > It has become very clear in recent years that Chomsky's generative >calculus model of linguistics has no relevance at all to anything about >actual language. It has also become clear that the main generative notion >of the innateness of language is based on flimsy, non-existent evidence. > Despite this, and despite the fact that many people in many different >fields are now actually studying real language in realisitic, empirical >ways, and despite the fact that many people intuitively understand the >emptiness of generative claims and practices, it still remains true, that to >the world at large, Chomsky's theories are somehow considered an important >"scientific breakthrough." > So, the time has come that those of us who want to start a new paradigm >for language studies, who want to begin an empirical way of studying real >language, should simply explain why the entire Chomskyan method must be >thrown away. > Ok, the world is wrong and you're right! I should be grateful if you could define 'realistic' and 'empirical'. Thanks in advance. ***************************************** Dr. Jose-Luis Mendivil Giro Dept. of General and Hispanic Linguistics Universidad de Zaragoza C/ Pedro Cerbuna, 12 50009 Zaragoza (Spain) Phone: (+34) 976 761 000 Fax: (+34) 976 761 541 E-mail: jlmendi at posta.unizar.es From john at RESEARCH.HAIFA.AC.IL Thu Dec 9 13:42:11 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Thu, 9 Dec 1999 15:42:11 +0200 Subject: Alternatives to Chomsky Message-ID: While I agree fully with Bruce Richman's assessment of the value of Chomskyan linguistics, I disagree that we should spend much effort on `simply explain(ing) why the entire Chomskyan method must be thrown away.' Chomskyism is a religion, not an intellectual program. It is not something which can be rationally argued with. Chomsky himself does not even attempt to give rational arguments for the allegedly sweeping paradigm changes he seems compelled to make every decade or so to give the appearance of progress--he simply says that the new approach seems more attractive, or something like that, and that's enough to change the minds of his apostles. Several years ago I challenged funknetters to identify themselves if they had been convinced for or against a Chomskyan position after graduate school. Although I repeated this challenge several times in the context of a heated debate which many people were involved in, and although many people responded to other points in my messages, only a single person (Ellen Prince, towards Chomskyan linguistics) reported having had such an experience. I concluded that any convincing arguments against Chomsky must be directed at people who have not yet completed graduate school (in any case, Chomskyan devotees have a tendency to stop doing active research and detach from research affairs when they have been away from The Master for a few years, so I would not say 'conversions' like, e.g. Tom Wasow are worth anything). And since almost all graduate schools now are already clearly pro-Chomsky or anti-Chomsky (I include places such as Stanford as 'pro-Chomsky', although some of the faculty members have made names for themselves as professional gadflies), there is no realistic way to compete for the loyalties of graduate students either. Arguing with Chomsky is a waste of time. I'm all for alternative approaches, but Chomskyism is only going to die with Chomsky--and the man is 71 years old. John From bresnan at CSLI.STANFORD.EDU Thu Dec 9 17:46:32 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 9 Dec 1999 09:46:32 -0800 Subject: Alternatives to Chomsky In-Reply-To: Your message of Thu, 09 Dec 1999 15:42:11 +0200. Message-ID: In a message of 9 Dec 1999 John Myhill wrote: > ... Chomskyan devotees have a tendency to stop doing active research and > detach from research affairs when they have been away from The Master for a > few years, so I would not say 'conversions' like, e.g. Tom Wasow are worth > anything). And since almost all graduate schools now are already clearly > pro-Chomsky or anti-Chomsky (I include places such as Stanford as > 'pro-Chomsky', although some of the faculty members have made names for > themselves as professional gadflies), ... My, my. Let's not get personal. John, you may not be in touch any more with what's been going on at Stanford. We have been building intellectual bridges between formal and functional approaches. Elizabeth Traugott and Paul Kiparsky are working together on a historical project that centrally involves grammaticalization. Ivan Sag and some of his students have become deeply involved with the Fillmore-Kay Construction Grammar, which bears close family resemblances to HPSG/LFG. Bresnan (that's me) has a collaborative research project with Judith Aissen at Santa Cruz on "Optimal Typology" which focusses on using Optimality Theory as a formal tool to explore syntactic markedness hierarchies and explain some of their problematic properties (softness, variable expression, recurrence). We are arguing for using typologically motivated and functionally grounded constraints, much as has happened in phonology. Edward Flemming in phonology is one of the leading young exponents of the OT functionalist approach to phonology which argues that perception shapes the structure of language. Tom Wasow does corpus linguistic studies and natural language processing, so you're more likely to find his recent papers in _Cognitive Psychology_ and _Language Variation and Change_ than in _Linguistic Inquiry_. The corpus approach is very big at Stanford. Our new young faculty memember Chris Manning who holds a joint appointment with Computer Science argues that the new surge of statistical approaches to Natural Language can restore some of the balance in the field by strengthening areas such as historical linguistics and sociolinguistics that were somewhat marginalized by the Chomskyan revolution. Come spend a sabbatical year at Stanford, and see for yourself! Joan Bresnan http://www-ot.stanford.edu/bresnan/ From lakoff at COGSCI.BERKELEY.EDU Thu Dec 9 18:26:14 1999 From: lakoff at COGSCI.BERKELEY.EDU (George Lakoff) Date: Thu, 9 Dec 1999 11:26:14 -0700 Subject: Alternatives to Chomsky In-Reply-To: Message-ID: >I should be grateful if you could define 'realistic' and 'empirical'. >Thanks in advance. > > >***************************************** >Dr. Jose-Luis Mendivil Giro >Dept. of General and Hispanic Linguistics > >Universidad de Zaragoza >C/ Pedro Cerbuna, 12 >50009 Zaragoza (Spain) > >Phone: (+34) 976 761 000 >Fax: (+34) 976 761 541 >E-mail: jlmendi at posta.unizar.es That's easy to find out. Check with any member of the Cognitive Linguistic Society of Spain (AESLA). Their e-mail list address is: lingcog at fcu.um.es. Also take a look at the discussion of Chomsky's philosophy in Philosophy in the Flesh (obtainable via amazon.com). From nrude at ucinet.com Thu Dec 9 22:29:16 1999 From: nrude at ucinet.com (Noel Rude) Date: Thu, 9 Dec 1999 14:29:16 -0800 Subject: Announcement: Alternatives to Chomsky meeting Message-ID: Well, folks, Maybe there's even one more approach besides Chomsky's biological innateness vs. our supervenience upon a thousand-and-one physical phenomena. Like it or not, a large number (perhaps a majority) of mathematicians (and lots of other people in the nonbiological "hard sciences") are still Platonists (don't shoot me!) of the old-fashioned stripe. These blokes believe that the laws of physics lie further down in the hierarchy of reality than the primitive axioms from which mathematics (and logic) derive. Modern cosmological theory generally assumes that physical law traces from the Big Bang (however conceived), and today's theoreticians profit (they think!) from studying other possible worlds all the while assuming that the higher-arching principles of math and logic must obtain there too (and not necessarily the laws of physics). Whereas the formalists believe we invent math/logic, the mathematical Platonists believe we discover it. And--lest anyone gets too frightened--many maybe most of these Platonists are also hard core atheists. And they are reductionists. They just believe that "those things which could be no other way" have a reality of their own. So instead of throwing spit-wads at each other from these various camps within a strictly reductionist "physical law" materialism, why not keep our cool and at least acknowledge what the big boys in the hard sciences are thinking. Maybe Chomsky is on to something. Maybe some of the core of Language--maybe rather than being biologically "hard-wired"--maybe we as humans are equipped to "discover" it. No, let's avoid extremism (as Givon always warns against). Let's milk the biological and discourse-pragmatic sides of the coin for all they're worth. But maybe some folks who study the more abstractly logical side of things should hang in there too. Just what is there about Language "that could be no other way"? What does Natural Language share with other information systems? Is there something there that doesn't merely "emerge" from a swirl of atoms? Why is it we can't have both pragmatics AND Plato? Noel the Rude From fjn at U.WASHINGTON.EDU Thu Dec 9 22:55:45 1999 From: fjn at U.WASHINGTON.EDU (Frederick Newmeyer) Date: Thu, 9 Dec 1999 14:55:45 -0800 Subject: Alternatives to Chomsky meeting Message-ID: Joan Bresnan made an excellent point in her recent Funknet posting. The world of linguistic theorizing is a much bigger place than it was in the past. In my paper at the recent UCSD Conference on Explanation in Linguistics (the conference that George Lakoff referred to), I referred positively to work by John Haiman, Talmy Givon, and Jack Hawkins, and did not mention Chomsky's name even once. There are lot of people out there who are actively trying to synthesize the contributions of various linguistic traditions, including the generative. I do hope that the organizers of the 'Alternatives to Chomsky' conference will organize a meeting that is sensitive to that fact. Fritz Newmeyer From jmacfarl at unm.edu Fri Dec 10 01:06:20 1999 From: jmacfarl at unm.edu (jmacfarl at unm.edu) Date: Thu, 9 Dec 1999 18:06:20 -0700 Subject: How Functional is OT? In-Reply-To: <199912091745.JAA25025@hypatia.Stanford.EDU> Message-ID: On 9 Dec 99, at 9:46, Joan Bresnan wrote: > Bresnan (that's me) has a collaborative > research project with Judith Aissen at Santa Cruz on "Optimal > Typology" which focusses on using Optimality Theory as a formal tool > to explore syntactic markedness hierarchies and explain some of their > problematic properties (softness, variable expression, recurrence). We are > arguing for using typologically motivated and functionally grounded > constraints, much as has happened in phonology. Edward Flemming in > phonology is one of the leading young exponents of the OT functionalist > approach to phonology which argues that perception shapes the structure of > language. Dear Funknetters, As I understand it, OT posits innate or a-priori constraints on the well-formedness of linguistic structure. This leaves me confused when people refer to OT as a functional theory. For example, in Bresenan's description of her work.... We are arguing for using typologically motivated and functionally grounded constraints, much as has happened in phonology. My question is how does "function" inform "constraints" if these constraints are said to be innate or a-priori? Thanks, ************************* James MacFarlane University of New Mexico ************************* From bresnan at CSLI.STANFORD.EDU Fri Dec 10 03:05:03 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 9 Dec 1999 19:05:03 -0800 Subject: How Functional is OT? In-Reply-To: Your message of Thu, 09 Dec 1999 18:06:20 MST. <199912100113.TAA00954@listserv.rice.edu> Message-ID: >>>jmacfarl at unm.edu said: > > As I understand it, OT posits innate or a-priori constraints on the > well-formedness of linguistic structure. This leaves me confused > when people refer to OT as a functional theory. > > For example, in Bresenan's description of her work.... > > We are arguing for using typologically motivated and functionally > grounded constraints, much as has happened in phonology. > > My question is how does "function" inform "constraints" if these > constraints are said to be innate or a-priori? Oh, dear. There's so much confusion about this topic. I'm afraid of stepping into a FUNKpit. Well, here goes nothing... 1. Constraints can be innate without being a priori. For example, the possible inventories of phonological segments and the distribution of phonological contrasts within and across languages reflect constraints imposed by the limits of our articulatory and perceptual systems, which are innate in the sense that we are born with them. The work of Paul Boersma (http://fonsg3.let.uva.nl/paul/), Donca Steriade (http://www.humnet.ucla.edu/humnet/linguistics/people/steriade/steriade.htm), and Edward Flemming (http://www.stanford.edu/~flemming/) is representative. For example, Steriade's web page says: Donca Steriade works mainly on the interactions between phonological patterns and speech properties considered to be exclusively phonetic or even non-linguistic, such as perceptibility and avoidance of articulatory effort. The work is based on the hypothesis that these factors play a role in shaping sound patterns, not only in an evolutionary sense--as argued by John Ohala and Bjorn Lindblom--but also by defining the grammatical constraints whose interactions yield the phonologies of individual languages. In an analogous way, markedness hierarchies in syntax and semantics (Silverstein, Givon, Comrie et al.) have been postulated to reflect innate properties of the human perceptual and cognitive systems. --Not necessarily language-specific properties, but part of the perceptual/cognitive apparatus people are born with. For some VERY BEGINNINGS of OT work incorporating constraints of this kind, have a look at the background papers by Aissen and Bresnan on our Optimal Typology web site at Stanford: http://www-ot.stanford.edu/ot/. 2. Constraints can be innate without having language-specificity. (See the discussion on this very list some time ago, in which important distinctions about innateness and language specificity were made by Liz Bates, among others... don't you remember it?) 3. And constraints can be universal without being innate. Examples would be pragmatic constraints on reference, communicative constraints that influence discourse organization, cohesion, etc. -------- OT is really a theory of constraint INTERACTION, not a theory of what the substance of constraints must be, or where they must come from. You can certainly invent as many a priori, autonomous, artificial, mechanical, etc., constraints as you like (and many have...). But you don't have to. Explicit constraints within a precise (I'm afraid to use the word "formal" on this list!) theory of constraint interaction are useful in that they enable you to experiment with prioritizing constraints in order to generate typologies and test language-particular interactions of constraints. Boersma's model adds the nice refinement of probabilistically variable ranking to the constraints. See Boersma and Hayes' recent paper in the Rutgers Optimality Archive. ------- TTFN-- Joan From gentner at ILS.NWU.EDU Fri Dec 10 03:54:03 1999 From: gentner at ILS.NWU.EDU (Dedre Gentner) Date: Thu, 9 Dec 1999 21:54:03 -0600 Subject: language learning Message-ID: If the discussions of new approaches to grammar extend to issues of how it is learned, I'd like to suggest that analogy be reconsidered as an important mechanism in language learning. From haspelmath at EVA.MPG.DE Fri Dec 10 13:34:24 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Fri, 10 Dec 1999 13:34:24 +0000 Subject: How functional is OT? Message-ID: I found Joan Bresnan's remarks very interesting indeed, and I feel that FUNKNET subscribers should pay close attention to the work that is going on at Stanford and Santa Cruz. I completely agree that OT says nothing about the nature of the constraints, and that OT analyses with functionally-motivated constraints are possible. In fact, people in Natural Morphology (W. Wurzel, W. Dressler and colleagues) have been talking about the ranking of functional principles for almost two decades. But although especially in phonology more and more people seem to be interested in functional explanations of OT constraints, it is also true that most OT practitioners probably still think of constraints as abstract entities that are directly innate and need not and cannot be explained further. If the new wave of (post-MIT) functionalism comes from Stanford/Santa Cruz, that's great. And if you people there occasionally acknowledge that some people have said similar things about language function explaining language form before, that would be nice. Martin -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From geoffn at SIU.EDU Fri Dec 10 14:33:09 1999 From: geoffn at SIU.EDU (Geoffrey S. Nathan) Date: Fri, 10 Dec 1999 08:33:09 -0600 Subject: How Functional is OT? In-Reply-To: <199912100113.TAA00954@listserv.rice.edu> Message-ID: Let me chime in here with an endorsement of what Joan Bresnan and Martin Haspelmath have said about the functional nature of current phonological theory. It was interesting to note that at the Milwaukee conference on formalism and functionalism held a few years back there was a radical difference between the kinds of talks that the syntacticians and semanticists and typologists gave and those that the three phonologists gave. I felt that Joan Bresnan, Bruce Hayes and I all had things to say to each other, and listened and learned from each other, while the others seemed to be talking past each other. And it seems to me it was precisely because the 'formalists' in phonology have (in many cases) insisted that phonological structures be grounded (either in a technical sense, as used by Archangeli et al., or in the more general sense that functionalists use the term) in facts about production and perception. This, of course, is what Stampe and the Natural Phonologists proposed in the sixties and seventies. When Stampe used the term 'innate' he meant, explainable by the nature of the physiology, physics and psychophysics involved. Unfortunately, at the time, Chomsky was using 'innate' to mean unexplainable extrinsically, and NP's message got rejected by those rejecting Chomsky's version of innateness. All this is water under the bridge now, but I will second, (third?) the plea that theories founded purely on negative impulses are unlikely to gain us the insights we are looking for. And many valuable insights about the nature of language have arisen from purely formalist investigations, even if, as Karen van Hoek and Paul Deane (not to mention Lakoff, Langacker etc.) have shown, the insights have reasonable non-linguistic explanations after all. Geoff Geoffrey S. Nathan Department of Linguistics Southern Illinois University at Carbondale Carbondale, IL, 62901-4517 Phone: (618) 453-3421 (Office) / FAX (618) 453-6527 (618) 549-0106 (Home) geoffn at siu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From macw at CMU.EDU Fri Dec 10 15:47:06 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Fri, 10 Dec 1999 10:47:06 -0500 Subject: OT Message-ID: Dear Funknet, I think that a discussion on Funknet of the claims underlying OT could be extremely helpful. I agree with Joan Bresnan and Geffrey Nathan that there is nothing afunctional about grounding constraints on the facts of the perceptual and production apparatus. On the contrary, it would seem to me that the natural phonologies of Dressler, Bybee, Vennemann, Ohala, and others make an interesting functional statement when they trace the grounding of constraints on the speech production and perception apparatus. I agree that Hayes and others are doing a nice job of providing this type of grounding to OT. I promise to read the various sources that Joan points too, since I imagine that they further elaborate this important contribution. However, as a psycholinguist, I have been disappointed by five crucial "strategic" decisions in the development of OT that have tended to vitiate its potential for constructing a psychologically plausible linguistic theory of the type that Joan Bresnan and others have often sought. In particular, 1. Early on, OT was supposed to be linked to connectionist modeling. However, after the first few years, this linkage was largely dropped. Dedre Gentner's interest in analogy as an acquisition or production mechanism has pretty much suffered the same fate, I would guess. 2. Early on, OT constraints were supposed to have strength levels. However, later on this feature was eliminated. In our work on the Competition Model, Liz Bates and I learned how important strength levels are for describing and predicting psycholinguistic data. In fact, one has to go beyond strength levels for separate constraints and look at what we can conflict validity, but none of this could possibly fit in with current OT. 3. Early on, learning of a phonology (or grammar) could have involved the strengthening and weakening of constraints. Later on, it required the types of triggers used by G-B and P&P theories. 4. Even from the beginning, OT never questioned the need to provide a single abstract underlying structure for each lexical item. As far as I can tell, this commitment is the one that tends to lead linguistic theories away from being able to develop psychological reality. Trying to preserve this approach in OT models of syntax would be equally problematic. 5. At no point was OT really committed to an account of online processing. If problems 1-4 were not present, I would not consider this a fatal flaw, since the notion of constraint ordering has clear interest for typology and language change, at the very least. I am just a psycholinguist, so I am happy to have linguists explain to me how I have strayed in my judgments. But I would really like to hear some open discussion of these issues. If it has already occurred on some OT bulletin board, perhaps people can simply point me to an archived discussion of these issues. --Brian MacWhinney From Twright at ACCDVM.ACCD.EDU Fri Dec 10 15:53:36 1999 From: Twright at ACCDVM.ACCD.EDU (Tony Wright) Date: Fri, 10 Dec 1999 09:53:36 -0600 Subject: How Functional is OT? In-Reply-To: <4.2.0.58.19991210082140.009f5190@saluki-mail.siu.edu> Message-ID: At 08:33 AM 12/10/99 -0600, Geoffrey S. Nathan wrote: > Let me chime in here with an endorsement of what Joan Bresnan and > Martin Haspelmath have said about the functional nature of current > phonological theory. Indeed, this very trend is lamented by Mark Hale and Charles Reiss in _Substance Abuse and Dysfunctionalism: Current Trends in Phonology_ in which it is argued that phonology, particularly current implementations of OT, are drifting off toward functionalism, and that phonology qua phonology should be "all form, no substance." (This paper was once on the Rutgers Optimality Archive, but I find it has been removed, probably owing to its subsequent publication in Linguistic Inquiry (if memory serves)). --Tony Wright From nrude at ucinet.com Fri Dec 10 18:47:28 1999 From: nrude at ucinet.com (Noel Rude) Date: Fri, 10 Dec 1999 10:47:28 -0800 Subject: Phonology Message-ID: Folks! Have been away from the lit for quite a spell now and can't say I know anything about acronymic linguistics (nor much of anything, for that matter). What bothers me about letting linguistics degenerate into phonology is that this is what happens when we insist on grounding everything in neurology and physiology. Such was the error of the Bloomfeldians. Chomsky was right in so far as he argued for studying Language--syntax and semantics and universals and all--apart from any purely mechanistic theory (behaviorist or otherwise--is that what his "innateness" was meant to do?). We can still study Language and keep it empirical--grounding it in legitimate data (texts, etc.)--even if the Chomskians have tended not to do this. We can operate within a purely communicative theory of language. And yes as far as possible our functional explanations should be grounded in reality (biology, psychology, pragmatics). But we might remember that no one as yet has succeeded in defining information in purely physical terms (grams, centimeters, volts, etc.). If we don't want to deal with the logical/informational side of language then it will have to fall to the philosophers and mathematicians to do so. Of course I know most of us still believe in syntax and semantics. I just thought it would be good to remind ourselves that Chomsky did help in delivering us from the "biological extremism" of mid century America. Rude again From bresnan at CSLI.STANFORD.EDU Fri Dec 10 19:02:40 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Fri, 10 Dec 1999 11:02:40 -0800 Subject: OT, functionalism, etc. Message-ID: This is a very interesting discussion, and I will try to respond to specific questions raised, in this and a following msg. But remember that I am only a student of Optimality Theory and can't speak with the voice of Authority. James MacFarlane asked: >My question is how does "function" inform "constraints" if these >constraints are said to be innate or a-priori? My reply was that OT does not require the constraints to be a-priori, or even innate, but that innateness is quite compatible with functionalism, a point already made on this list by Liz Bates some time ago, when she distinguished the hypothesis of innateness from the hypothesis of specificity to language (or autonomy). I would like to add that OT does make an important claim of (near) universality of constraints that is used to derive certain learnability and typological results. This is not something that functionalists should find alien. Couldn't conflicting constraints such as iconicity and economy be universal, but prioritized differently across different domains and different languages? Sergio Meira asks: > ... I mean, any constraints which are derived >from the biology of our speech aparatus must be a-priori. Or do I >misunderstand the way this term is being used? Well, my epistemology has grown rusty since my time as a student of philosophy. What you say could be true in an Aristotelian sense of the a priori as that from a causal explanation can be logically derived. But I was thinking of the later philosophical uses of the terms "a posteriori" and "a priori" to mean, roughly, what is derived from experience (empirical truths) and what is derived from reason alone (logical truths). Functional OT phonology appeals to experimental work in human speech perception and articulation to support the constraints hypothesized, so I would not call these constraints a priori. Sergio Meira states: >If OT is only a theory of constraint interaction, then the sources >and natures of constraints probably aren't important... Please don't misunderstand me. OT does not claim that the sources and natures of its constraints are unimportant. My point was rather that as a theory of constraint interaction OT does not dictate the substance of the constraints. It does assume that constraints are (near) universal and violable, and it does urge simple and natural constraints (so that the interactions are not concealed inside the content of complex constraints), but these assumptions are compatible with functionalist views, I believe. It is true that a variety of researchers have been attracted to OT, and in the field of syntax these have included those who like to import their favorite constraints from the Minimalist Program (or whatever) into the new theory. But a number of us have argued against this, and see functional/typological work as the best source for well motivated constraints in OT syntax. Martin Haspelmath writes: >But although especially in phonology more and more people seem to be >interested in functional explanations of OT constraints, it is also true >that most OT practitioners probably still think of constraints as >abstract entities that are directly innate and need not and cannot be >explained further. I agree that this is true, especially in syntax. If I may quote from the conclusion of my paper "Optimal Syntax": Because OT per se is a theory of constraint interaction rather than a theory of substantive linguistic constraints, it is compatible with a wide range of substantive theoretical choices. (Some consider this an explanatory weakness of the framework, but it is also the source of its great integrative potential.) In phonology and to a lesser extent morphology, OT has led to a fundamental rethinking of the domain and to the widespread adoption of nonderivational theories. Syntax, in contrast, is still greatly influenced by the derivational frameworks advanced by Chomsky, and much of the initial work applying OT to syntax reflects this way of thinking by simulating derivational analyses. It is instructive to consider the history of architectural design, which shows that earlier designs, for example in bridge-building, persist long after the development of new materials with radically different engineering properties (e.g. steel compared to wood and stone). Martin Haspelmath adds: >If the new wave of (post-MIT) functionalism comes from Stanford/Santa >Cruz, that's great. And if you people there occasionally acknowledge >that some people have said similar things about language function >explaining language form before, that would be nice. I agree completely about this. Many of the most attractive and central ideas of OT about constraint interaction were anticipated by researchers in Natural Morpology, Natural Syntax, and pragmatics. That is one reason why OT seems to be a natural framework for incorporating and further developing and testing these ideas. The OT framework was initially developed by phonologists, who may be forgiven for not being aware of related work in different fields. But it would be unforgivable for us syntacticians to reinvent functional/typological theories in the guise of OT, without any acknowledgement... Geoffrey S. Nathan writes: > ... It was interesting to note that at the Milwaukee >conference on formalism and functionalism held a few years back there was a >radical difference between the kinds of talks that the syntacticians and >semanticists and typologists gave and those that the three phonologists >gave. I felt that Joan Bresnan, Bruce Hayes and I all had things to say to >each other, and listened and learned from each other, while the others >seemed to be talking past each other. I am pleased that we agree. But it must have been my agreeable spirit you were talking to in Milwaukee, because I wasn't there! TTFN-- Joan *---------------------------------------- ______ __o __o Joan Bresnan bresnan at stanford.edu ______ _`\<,_ _`\<,_ *---------------------------------------- ______ (*)/ (*) (*)/ (*) From dkp at U.ARIZONA.EDU Fri Dec 10 19:09:41 1999 From: dkp at U.ARIZONA.EDU (Dianne K. Patterson) Date: Fri, 10 Dec 1999 12:09:41 -0700 Subject: Phonology In-Reply-To: <38514AB9.4EFE@ucinet.com> Message-ID: On Fri, 10 Dec 1999, Noel Rude wrote: > Folks! > > Have been away from the lit for quite a spell now and can't say I know > anything about acronymic linguistics (nor much of anything, for that > matter). > > What bothers me about letting linguistics degenerate into phonology is > that this is what happens when we insist on grounding everything in > neurology and physiology. Perhaps I am missing something...but the standard position is that we do not want to confuse phonetics (which is mechanistic and mechanical) with phonology (which is a rule governed system and usually included as part of the grammar)...now people may wish to say phonology is not on the right track and should in fact be grounded in biology (e.g., John Ohala...and I have my own leanings in this direction)...but such views are outside the standard accepted views. > Such was the error of the Bloomfeldians. > Chomsky was right in so far as he argued for studying Language--syntax > and semantics and universals and all--apart from any purely mechanistic > theory (behaviorist or otherwise--is that what his "innateness" was > meant to do?). We can still study Language and keep it > empirical--grounding it in legitimate data (texts, etc.)--even if the > Chomskians have tended not to do this. We can operate within a purely > communicative theory of language. And yes as far as possible our > functional explanations should be grounded in reality (biology, > psychology, pragmatics). But we might remember that no one as yet has > succeeded in defining information in purely physical terms (grams, > centimeters, volts, etc.). There is a standard information theory. It is grounded in bits and underlies computer technology...but it is also used in standard ethology (the study of the behavior of communicating) which is generally used when looking at nonhuman behavior...but can, of course also be used to look at human behavior > If we don't want to deal with the > logical/informational side of language then it will have to fall to the > philosophers and mathematicians to do so. > > Of course I know most of us still believe in syntax and semantics. I > just thought it would be good to remind ourselves that Chomsky did help > in delivering us from the "biological extremism" I hope this is helpful, Dianne Patterson From geoffn at SIU.EDU Fri Dec 10 20:25:09 1999 From: geoffn at SIU.EDU (Geoffrey S. Nathan) Date: Fri, 10 Dec 1999 14:25:09 -0600 Subject: Phonology In-Reply-To: Message-ID: At 12:09 PM 12/10/1999 -0700, Dianne K. Patterson wrote: >On Fri, 10 Dec 1999, Noel Rude wrote: > > > Folks! > > > > Have been away from the lit for quite a spell now and can't say > I know > > anything about acronymic linguistics (nor much of anything, for that > > matter). > > > > What bothers me about letting linguistics degenerate into > phonology is > > that this is what happens when we insist on grounding everything in > > neurology and physiology. > >Perhaps I am missing something...but the standard position is that we do >not want to confuse phonetics (which is mechanistic and mechanical) with >phonology (which is a rule governed system and usually included as part of >the grammar)...now people may wish to say phonology is not on the right >track and should in fact be grounded in biology (e.g., John Ohala...and I >have my own leanings in this direction)...but such views are outside the >standard accepted views. Alas, the 'standard position' has become greatly confused. Many phonetician/phonologists now believe that there are language-specific (and thus not mechanistic and mechanical) phonetic implementation rules that are distinct from phonological rules, although perhaps both need to be grounded. These rules include the implementation of the feature [+voice] (which is voiceless in onset position in English, but voiced in French) and similar cases. Pat Keating has written on this issue, for example. And it should be made clear that Natural Phonology, for instance (and I think OT would agree with this way of expressing things) argued that while phonological constraints have physiological explanations, they are not mechanical in and of themselves, precisely because they are violable. To put things in a way that Hale and Reiss object to, it's easier to voice obstruents intervocalically, (for physiological reasons) and some languages give in to the temptation while others don't (i.e. suppress the process, as NP would say it, or rank Faithfulness higher, as OT would say it). Stampe always argued that processes are substitutions made on behalf of the vocal tract, not by it, and arguing that certain phonetically-motivated constraints are outranked by faithfulness constraints is saying the same thing. Geoff P.S. Nobody is arguing that all grammar is reduceable to physiology, but only that phonological processes (however interpreted these days) have physical explanations. Grammatical processes, according to Cognitive Grammar, at least, have extragrammatical cognitive explanations. GN Geoffrey S. Nathan Department of Linguistics Southern Illinois University at Carbondale Carbondale, IL, 62901-4517 Phone: (618) 453-3421 (Office) / FAX (618) 453-6527 (618) 549-0106 (Home) geoffn at siu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From bbergen at socrates.berkeley.edu Fri Dec 10 20:10:50 1999 From: bbergen at socrates.berkeley.edu (bbergen at socrates.berkeley.edu) Date: Fri, 10 Dec 1999 12:10:50 -0800 Subject: OT, functionalism, etc. In-Reply-To: <199912101901.LAA28017@hypatia.Stanford.EDU> Message-ID: Since the cents seem to be flying, I'd like to bring up a couple critiques of functionalist OT. I apologize if this commentary is on the long side, but I think that this position has been inadequately expressed in the recent discussion on this list. First, like other "natural" phonological models, it is very difficult for functional OT to explain " unnatural" phonological behavior. Juliette Blevins and Andrew Garret have recently shown for consonant harmony and metathesis that is is essentially impossible to explain why these processes have variant behavior on the basis of synchronic natural models. Instead, synchronic idiosyncracies result from regular, natural diachronic developments. The problem is not that there are no "natural" behaviors in synchronic phonology - the problem is that trying to explain all synchronic phonological properties by appealing to "natural" effects obscures an essential question: what properties of phonology are a direct result of (caused by, explained by) the synchronic grammar (including, perhaps functional constraints), and which are historically contingent (although historically functionally motivated)? Another questionable property of OT relates to issues brought up by Brian MacWhinney. OT has become an extremely powerful device. Such notions as Sympathy, Overlapping Constraints (Hayes), Output-Output correspondences (e.g. Benua), integration of phonological and extraphonological constraints, and levels of OT derivation have rendered OT essentially omnipotent. This makes it difficult to call it an explanatory theory. Of course, there are various attempts to reduce its power, but the real problem, I think, lies in the very notion that this OT grammar is actually explanatory of the language it models, independent of the "semantics" of the constraints. (Although I do recognize that Joan Bresnan disagrees with me here, I simply mean that there is no way in which the extra-constraint properties of functional constraints enter into their behavior within an OT model.). I would bet that most functionalists agree that whatever capacity for language individual speaker-hearers have, it is extremely powerful, and that certain restrictions are levied on what form a phonology can have by factors external to it. Some of these restrictions will come from the typical sources - processing constraints, neural architecture constraints, information processing constraints, physiological constraints, cognitive representation constraints, and so on. It might be a problematic step to reverse these roles as functional OT seems to - the grammar becomes explanatory, and the constraints have very little explanatory role. On the other hand, I couldn't be happier that the interface between phonetics and phonology is being closely scrutinized. Functional OT may very well provide the kind of theoretical framework required for good empirical work to get done. Ben Bergen ------------------------------------------- Benjamin K. Bergen Graduate Student Department of Linguistics U.C. Berkeley www.icsi.berkeley.edu/~bbergen ------------------------------------------- From chafe at HUMANITAS.UCSB.EDU Sat Dec 11 20:54:38 1999 From: chafe at HUMANITAS.UCSB.EDU (Wallace Chafe) Date: Sat, 11 Dec 1999 12:54:38 -0800 Subject: Inputs In-Reply-To: <3548217.3153811626@agate.psy.cmu.edu> Message-ID: I'd like to expand a little on the following point made by Brian MacWhinney: > 4. Even from the beginning, OT never questioned the need to provide a single > abstract underlying structure for each lexical item. As far as I can tell, > this commitment is the one that tends to lead linguistic theories away from > being able to develop psychological reality. I've worked a lot with two polysynthetic and highly "fusional" languages (Seneca and Caddo) that have obviously undergone a lot of phonological change. I'm convinced that the only way to explain/predict/understand the shapes of their words is by reconstructing earlier forms along with the changes that have made these forms what they are today. Now, earlier "derivational" approaches to phonology essentially mirrored what I've just described in "underlying forms" and "rules" (whatever deviations there were from historical shapes and processes were trivial). But it seems pretty silly to think that speakers of these languages "know" such systems in any realistic sense. Just how they do produce these words is a different and fascinating question, but it seems to involve very large memories combined with the ability to analogize. To put it all too briefly, OT has done away with the rules but kept the underlying forms in the guise of "inputs" to the constraints. So I'm left with the question of what those inputs could possibly mean when it comes to what's going on in the minds of speakers of these languages. It would be easy for me to give examples, which I've done a little in a paper called "How a Historical Linguist and a Native Speaker Understand a Complex Morphology" in the volume Historical Linguistics 1997: Selected Papers From the 13th International Conference on Historical Linguistics (Benjamins). I think this is the same problem that Brian was talking about. Wally Chafe From sylvester.osu at wanadoo.fr Sun Dec 12 08:24:50 1999 From: sylvester.osu at wanadoo.fr (Sylvester OSU) Date: Sun, 12 Dec 1999 09:24:50 +0100 Subject: Alternatives to Chomsky. A word to the initial Author Message-ID: Dear Bruce Richman, It's high time everyone realised that "killing" a man does not in any way "kill" his ideas. What does that really mean? Throwing Chomskyan method away? And what next? Alternatives to Chomsky. No. No. No. Chomsky has never been proclaimed the god of linguistic science, so why take him to be one. I thought we were doing some kind of objective work on languages. The truth is that Chomsky has made his contribution towards understanding the Human Language. If today we find a means of going somewhat further, then that's good. We push on. But we can't come back to square one everday 30 or 40 years and do as if nothing ever happened. Besides, I don't see anyway we can continue without making reference to his works (negatively or positively). So please, take it easy. Secondly, I'll like to remind you that other people have been working using other methods and proposing other frameworks. I refuse to believe that you are not aware of that anyway. Otherwise, why talk of proposing Alternatives to Chomsky? Or is it a provocative method of attracting people to your conference? Once again, Chomsky's method as far as I am concerned is just one among many others. Take for instance in Europe, many people have come out to propose some nice approaches to the study of Language and languages. Permit me here to simply name Antoine Culioli, a French linguist, and his research team, with their fantastic theory of predicative and enunciative operations. I bet you, it's something you should know about before going deeper in your conception of linguistic studies. Apart from Culiolian approach, you have Oswald Ducrot, you have Gustave Guillaume, Gross, Hag?ge, all French, who have proposed some different approaches. Think about it. Well, I just wish that over there, you don't think that the world outside is at a standstill. Try and get in touch, and you'll see that things are moving and that one approach has never and will never offer all the answers to all the problems that the study of Language raises. Should you need references from the Linguists I just mentioned, I'll be glad to submit them to you as quickly as possible. Wishing you well. Sylvester OSU. (France) ps: I'll like you to know that I'm not a Chomskyan and that I have only taken courses and read his works as any average linguist would do. Even, my mastery of his theories is just below what's expected of a contemporary linguist. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bresnan at CSLI.STANFORD.EDU Sun Dec 12 19:01:57 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Sun, 12 Dec 1999 11:01:57 -0800 Subject: OT--5 crucial strategic decisions that vitiate... In-Reply-To: Your message of Fri, 10 Dec 1999 10:47:06 EST. <3548217.3153811626@agate.psy.cmu.edu> Message-ID: Sorry I couldn't get to this till now. My real life called me.... This is a very long msg, because Brian MacWhinney has asked some very substantial questions. Brian MacWhinney wrote on Friday, 10 December: > However, as a psycholinguist, I have been disappointed by five crucial > "strategic" decisions in the development of OT that have tended to vitiate > its potential for constructing a psychologically plausible linguistic theory > of the type that Joan Bresnan and others have often sought. In particular, > 1. Early on, OT was supposed to be linked to connectionist modeling. > However, after the first few years, this linkage was largely dropped. Dedre > Gentner's interest in analogy as an acquisition or production mechanism has > pretty much suffered the same fate, I would guess. It is interesting to read what Prince and Smolensky say about this in their article "Optimality: From Neutral Networks to Universal Grammar" in the March 14, 1997 issue of Science Magazine. (available on-line: http://www.sciencemag.org/cgi/reprint/275/5306/1604): The principal empirical questions addressed by optimality theory, as by other theories of universal grammar, concern the characterization of linguistic forms in and across languages. A quite different question is, can we explicate at least some of the properties of optimality theory itself on the basis of more fundamental cognitive principles? A significant first step toward such an explanation, we will argue, derives from the theory of computation in neural networks. Linguistic research employing optimality theory does not, of course, involve explicit neural network modeling of language. The relation we seek to identify between optimality theory and neural computation must be of the type that holds between higher level and lower level systems of analysis in the physical sciences. ... Like thermodynamics, optimality theory is a self-contained higher-level theory; like statistical mechanics, we claim, neural computation ought to explain fundamental principles of the higher level theory by deriving them as large-scale consequences of interactions at a much lower level. They give some interesting examples where OT has been conceptually driven by the hypothesized underlying neural implementation, but point out that it doesn't explain the universal recurrence of linguistic constraints, so there's still a gap between the two approaches. I suspect that Paul and Alan have a deeper view of the relation to connectionist ideas than most generative linguists who have taken up OT, but *some* linguists have been curious enough about the underlying cognitive issues to walk part way across the bridge. I have heard Helen de Hoop, for example, start out a lecture on OT semantics by talking about tensor products... brave woman, intellectually valiant. > 2. Early on, OT constraints were supposed to have strength levels. However, > later on this feature was eliminated. In our work on the Competition Model, > Liz Bates and I learned how important strength levels are for describing and > predicting psycholinguistic data. In fact, one has to go beyond strength > levels for separate constraints and look at what we can conflict validity, > but none of this could possibly fit in with current OT. My colleague Edward Flemming (http://www.stanford.edu/~flemming/) makes a similar argument. He concludes his most recent paper, "Scalar Representations in a Unified Model of Phonetics and Phonology" as follows: The proposed model of phonetics and phonology is similar to Optimality Theoretic phonology in that outputs are selected so as to best satisfy conflicting, violable constraints. However, the constraints considered here (particularly implementations of minimization of effort and maximization of distinctiveness) trade-off against each other in an additive fashion, implying that these interactions are better modeled in a weighted constraint system rather than one which exclusively employs strict constraint dominance, as is the case for Optimality Theory. It is an interesting question whether the scalar-valued functions that play a role in assimilation and coarticulation are just what is needed for studying the typology and structure of, say, pronominal inventories or voice systems in syntax. But even if one doesn't go that far (into continuous modelling), the idea of optimizing symbolic structures using a discrete evaluation function defined over universal, violable constraints (like the markedness constraints of functional/typological linguistics) is a radical shift that seems promising, and creates an intellectual bridge where there wasn't one before. > 3. Early on, learning of a phonology (or grammar) could have involved the > strengthening and weakening of constraints. Later on, it required the types > of triggers used by G-B and P&P theories. ???? I find this a bit hard to understand in view of Tesar and Smolensky's 1998 article "Learnability in Optimality Theory" (Linguistic Inquiry 29, 229--68) (and their long technical report) which explicitly argues against the triggers learning models. (You can find a lot of these papers on Paul Smolensky's web page: http://www.cog.jhu.edu/faculty/smolensky.html.) However, I also like the new model of gradual learning of constraints whose ranking varies probablistically (see Boersma and Hayes, ROA, for references: http://ruccs.rutgers.edu/roa.html), even more. > 4. Even from the beginning, OT never questioned the need to provide a single > abstract underlying structure for each lexical item. As far as I can tell, > this commitment is the one that tends to lead linguistic theories away from > being able to develop psychological reality. Trying to preserve this > approach in OT models of syntax would be equally problematic. If you have a generative, derivational model of underlying structure, that is certainly true. It is hard to free ourselves from the kind of thinking we were originally trained in... and it may be that OT imported a bit too much of the derivational ways of thinking in generative grammar at the beginning. But we know much more now about nonderivational, constraint-based representations than we used to (especially those of us who have explored alternatives to Chomsky's grammatical architectures), and some OT work has a very different model of the nature and role of the input. (I would refer to several of my recent papers here, if I weren't so modest.) This is also true in phonology. I really like Rene Kager's new (1999) book, _Optimality Theory_ in the red textbook series of Cambridge University Press. It does a beautiful job of explaining some of the functionalist ideas and motivations for OT phonology, and showing some of the recent developments that do away with inputs in various ways... I would address this answer to Wally Chafe's message of December 11, as well. > 5. At no point was OT really committed to an account of online processing. > If problems 1-4 were not present, I would not consider this a fatal flaw, > since the notion of constraint ordering has clear interest for typology and > language change, at the very least. > This is a *very* difficult problem: how to do online computations comparing infinite sets of recursive symbolic structures? Here, the mathematically formalized representational theories of syntax (LFG, HPSG, categorial grammar, etc.) have some advantages, I think, over the Chomskyan syntactic approaches that get more press. Some new theoretical work has been done on this problem in OT-LFG: one, using conventional kinds of computation, is by Jonas Kuhn (http://www.ims.uni-stuttgart.de/~jonas/): Jonas Kuhn. 1999. Generation and Parsing in Optimality Theoretic Syntax---Issues in the Formalization of OT-LFG (draft version of November 1999. To appear in a CSLI volume _OT-LFG: Optimality Theory and LFG_, edited by Peter Sells. I can also point you to a remarkable paper on Optimality Theoretic LFG on Neuroidal Nets by Professor Tetsuro Nishino using a dynamic, parallel computational model to evaluate the OT-LFG candidate space: http://www.sw.cas.uec.ac.jp/tnlab/member/nishino.html. These works are specific to LFG, and the first I know of on generation and parsing in OT Syntax. But once a problem has been solved in LFG, it usually doesn't take too long to port it to HPSG and other constraint-based frameworks, ... :-) > I am just a psycholinguist, so I am happy to have linguists explain to me > how I have strayed in my judgments. But I would really like to hear some > open discussion of these issues. ... What irresistable modesty! And so unwarranted. You FUNKfolk are a pleasure to discuss things with. TTFN- Joan From macw at CMU.EDU Sun Dec 12 20:57:31 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Sun, 12 Dec 1999 15:57:31 -0500 Subject: rote, analogy and OT Message-ID: Dear Wally, Many thanks for your thoughtful response to my queries and concerns about OT. I believe I share your basic perspective. In my 1978 account of the acquisition of morphophonology in Hungarian, I relied largely on the mechanisms of rote and analogy which you (and Dedre Gentner) suggest. Of course, one of the big challenges facing connectionist theory is trying to cash out the meaning of an account based on analogy. It seems to me that OT might have some promise in this regard. However, there seem to be two core problems that have to be ironed out. I would guess that you and I would agree that rote and analogy play their largest role in the area of morphophonology. But, when we turn to issues in phonotactics, assimilatory processes, and metrical phonology, then basic OT-type constraints seem to move to the fore. So, how do we link up the mechanisms of rote and analogy with constraints in a single language system? We certainly want to avoid modularizing two systems that, in fact, far from separable. After all, many morphophonological processes are echoes of phonotactic and assimilatory processes. So, a better solution is to try to figure out how physiologically-grounded constraints impact complex word combinations. To do this, we need to have a system of graded (or scalar) constraints and a method for talking about their interaction. Ideally, OT would have provided this, but the jury is still out on whether strengths and interactions will be a part of OT. The second problem is the ongoing reliance on underlying form in OT. In my still not properly informed opinion (see my forthcoming email reply to Joan Bresnan), the basic problem in OT is its commitment to the doctrine of underlying form. If we build a system that is designed to work off of forms that have little relation to anything stored in the brain, we will be lacing our synchronic theories with diachronic commitments. I will read your paper from the Historical Linguistics volume as soon as I can. The question will be how to reshape OT in a way that solves these problems. Thanks again for the helpful comments. I will continue to think more about these issues. --Brian MacWhinney From pesetsk at MIT.EDU Sun Dec 12 21:03:23 1999 From: pesetsk at MIT.EDU (David Pesetsky) Date: Sun, 12 Dec 1999 16:03:23 -0500 Subject: Myhill's message Message-ID: Dear Funknet subscribers, We want to express our surprise and disappointment that no subscribers to this list have disassociated themselves from Myhill's optimistic anticipation of our colleague's death. We refer to the closing lines of his message sent three days ago: > Arguing with Chomsky is a waste of time. I'm all for alternative > approaches, but > Chomskyism is only going to die with Chomsky--and the man is 71 years old. The strongest rebuke that Myhill received was an invitation to visit Stanford on his sabbatical. Surely the readers of this list are not functionalists first and human beings second! Sincerely, Sabine Iatridou Michael Kenstowicz David Pesetsky From macw at CMU.EDU Sun Dec 12 21:37:13 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Sun, 12 Dec 1999 16:37:13 -0500 Subject: OT--5 crucial strategic decisions that vitiate... In-Reply-To: <199912121900.LAA00634@hypatia.Stanford.EDU> Message-ID: Dear Joan (and Funknet), Thank you for the extremely helpful reply. I have downloaded the various papers you mentioned from the OT archive. I had read the Tesar ones earlier, but I will reread them now. I realize that Tesar's learning algorithm for OT doesn't use triggers, but it does use single trial evidence to reorder constraints. The dominance ordering of constraints in his model is absolute, so there is no way in which the slow accumulation of data in favor of a particular pattern in a language can lead to the gradual accumulation in its strength. However, I believe that the real problem with the Tesar model is not the strict dominance assumption, but the commitment to requiring the child to learn an abstract underlying form. If the Boersma and Hayes model addresses this problem, as well as the strict dominance issue, that would be great. I recently finished Bernhardt and Stemberger's OT treatment of phonological development (primarily for English). Although the authors set up a nice framework for discussing phonological acquisition, I think that the mechanics of OT tend to make the actual presentation of constraint patterns a fairly tough job. At this point, I think it would be best for me to go "offline" and do my homework. I have contacted Flemming and Kuhn, asking for the relevant papers. I will read them, the OT files, and Kager (1999). In a sense, what is important is not whether OT has problems, but whether it is recognizing these problems and moving to address them. I found your letter extremely encouraging in this regard. However, it is now time for me to do some more reading. Until then. --Brian MacWhinney From bresnan at CSLI.STANFORD.EDU Sun Dec 12 21:45:50 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Sun, 12 Dec 1999 13:45:50 -0800 Subject: Myhill's message In-Reply-To: Your message of Sun, 12 Dec 1999 16:03:23 EST. Message-ID: Oh, come on, David! This is flame-bait, just like Myhill's post. Joan >>>David Pesetsky said: > Dear Funknet subscribers, > > We want to express our surprise and disappointment that no subscribers to > this list have disassociated themselves from Myhill's optimistic > anticipation of our colleague's death. > > We refer to the closing lines of his message sent three days ago: > > > Arguing with Chomsky is a waste of time. I'm all for alternative > > approaches, but > > Chomskyism is only going to die with Chomsky--and the man is 71 years old. > > The strongest rebuke that Myhill received was an invitation to visit > Stanford on his sabbatical. Surely the readers of this list are not > functionalists first and human beings second! > > Sincerely, > > Sabine Iatridou > Michael Kenstowicz > David Pesetsky > --------------------------------- Joan Bresnan From faucon at COGSCI.UCSD.EDU Sun Dec 12 22:37:01 1999 From: faucon at COGSCI.UCSD.EDU (Gilles Fauconnier) Date: Sun, 12 Dec 1999 14:37:01 -0800 Subject: Myhill's message Message-ID: Flame-bait or not, it's nothing for any of us to be proud of. Same goes for the post right after that from a long-time Chomsky admirer, boasting that he had given an entire talk without mentioning the Master's name. Gilles Fauconnier Center for Advanced Studies in the Behavioral Sciences 75 Alta Road Stanford CA 94305 USA tel. 650 - 321 2052 From bresnan at CSLI.STANFORD.EDU Sun Dec 12 23:19:46 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Sun, 12 Dec 1999 15:19:46 -0800 Subject: Myhill's message In-Reply-To: Your message of Sun, 12 Dec 1999 14:37:01 PST. <9912122243.AA10926@cogsci.UCSD.EDU> Message-ID: Gilles, I think you are wrongly accepting the truth of David's premise, that John Myhill's msg showed "optimistic anticipation of our colleague's death". This is ridiculous. I certainly didn't interpret Myhill's comment about the strength of Chomsky's ideas in this way, and I doubt that Fritz Newmeyer, Martin Maspelmath, Geoff Nathan, Tony Wright, Brian MacWhinney, or any of the other follow-up posters did, either. This is what I meant by "flame bait". It's not only ridiculous but deeply insulting to draw this inference from the substantive discussion about functionalism, OT, etc., that we were engaging in. I thought that Myhill was wrong to publicly disparage by name both a colleague of mine and by implication Stanford's entire linguistics department, and I directed my reply toward those points, trying to be good humored and constructive about it. It appears, though, that our friends at MIT are quite willing to believe that we all harbor death wishes for their colleague. That's a shame, and I find it hard to believe that it could be said in sincerity. If it was, then it shows what a long way we have to go to build some of the intellectual bridges I was talking about. Joan > > Flame-bait or not, it's nothing for any of us to > be proud of. > > Same goes for the post right after that from a > long-time Chomsky admirer, boasting that he had > given an entire talk without mentioning the > Master's name. > > Gilles Fauconnier > Center for Advanced Studies in the Behavioral Sciences > 75 Alta Road > Stanford CA 94305 > USA > > tel. 650 - 321 2052 > From macw at CMU.EDU Mon Dec 13 00:46:37 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Sun, 12 Dec 1999 19:46:37 -0500 Subject: Myhill's message Message-ID: Like others on FunkNet, I did not parse Myhill's message as an "optimistic anticipation of our colleague's death." However, I can see that it could be interpreted in this way. Perhaps we should wait until the end of the weekend and allow John to clarify his position. As my colleague Herb Simon would say, at 71, Chomsky is still a young man. Herb, who received the Nobel Prize in Economics back in his 40s, is now about 85 and is producing as much high quality intellectual output as ever. He comes in to work and teach everyday and seems to have no interest in ever really retiring. He is a great role model and a great thinker. I think we should also recognize the fact that the Chomskyan program will outlive the man. If we believe that there are reasonable alternatives to Chomsky's ideas, let's get on with the work of constructing them. --Brian MacWhinney From john at RESEARCH.HAIFA.AC.IL Mon Dec 13 08:00:28 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Mon, 13 Dec 1999 10:00:28 +0200 Subject: Myhill's message Message-ID: Thank you for defending me, Joan. I certainly did not intend my comment to be interpreted as an `optimistic anticipation of our colleague's death.' I was simply stating that the social phenomenon of Chomskyism--a phenomenon in which the beliefs of a single person, whose basic contribution to linguistics was made about 40 years ago and who, in my opinion, has not done much new of interest since then (with the exception of 'On wh-movement', which I think is a really great article with actual insights, although it is more than 20 years old), to a significant extent direct the course of the field, both in terms of blindly following on the one hand and serving as a rallying cry for 'alternatives to...' on the other hand--this phenomenon is not something which can be dealt with by a conference on 'alternatives'. There is obviously a magnetic appeal which the guy has on some people which can't be argued with rationally and which is going to be around as long as he is around. Let me draw a parallel with physics (SORRY, I know I've criticized other functionalists for doing things like this, but I'm talking to the formalists on this network now, and this is the kind of argument they like). Let's suppose that after 1910, `Einsteinism' developed as a phenomenon comparable to today's Chomskyism, AFTER Einstein had made what is now recognized as his contribution, but 48 years BEFORE he died. Since Einstein didn't care much for quantum mechanics ('God doesn't play dice'), quantum mechanics might not have developed. Everyone would have spent all their time arguing about Einstein's post-1910 theories, which don't seem to be leading to anything, or on organizing conferences on `alternatives to Einstein.' There wouldn't have been much progress. Fortunately that didn't happen. But that IS what's been happening to a large extent in syntax in the last 30 years. After spending my whole career as a linguist listening to arguments going nowhere and convincing no one on the merits or lack of such of the Chomskyan enterprise, I am firmly convinced that they are a waste of time. I did not intend what I wrote to be 'flame-bait'. I was addressing my message to functionalists, suggesting that we not waste time arguing with Chomskyists. Some formalists on the network got set off by this inter-functionalist dialogue and took what I said as provocative in a very different way than I had intended it. I mentioned Stanford because of the presence there of several people who have in one way or another directed their careers towards being 'alternatives to Chomsky' in a way which I have not found to be any better than the original. It was part of a general criticism of building a program around `alternatives to Chomsky.' Brian MacWhinney wrote: > However, as a psycholinguist, I have been disappointed by five crucial > "strategic" decisions in the development of OT that have tended to vitiate > its potential for constructing a psychologically plausible linguistic theory > of the type that Joan Bresnan and others have often sought. Well, gee, let's try to figure out why this may have happened. Could it possibly be that Chomsky isn't interested in such a theory, and his followers aren't interested in thinking for themselves? So when will this change? I guess when Chomsky isn't on the scene, right? Any other ideas when it will change? Any other ideas on why OT has made these decisions? And Joan's comment is quite to the point: It appears, though, that our friends at MIT are quite willing to believe that we all harbor death wishes for their colleague. That's a shame, and I find it hard to believe that it could be said in sincerity. If it was, then it shows what a long way we have to go to build some of the intellectual bridges I was talking about. Again, though, I'm afraid Joan is being overly optimistic. The problem (in this case) is not 'intellectual bridges.' The problem is the paranoid mindset of our friends at MIT and the degree of influence they have over the field. My statement touched a raw nerve, because, I would hypothesize, our friends at MIT fear they will be intellectually lost without Chomsky and can't imagine that they will one day have to think for themselves. Please note that I am by NO means saying that everyone to do with Chomsky should be excised from intellectual history, as other functionalists have suggested. By no means. Chomsky has made a great contribution with some of his articles, although this was a pretty long time ago. But the overall effect of Chomskyism, the personality cult which has built around him, has at the same time been intellectually debilating, if not devastating, to the field as a whole. And with regard to Brian's statement that: `I think we should also recognize the fact that the Chomskyan program will outlive the man'--it depends upon what is understood by 'the Chomskyan program'. Chomsky is not interested in `constructing a psychologically plausible linguistic theory', taking into consideration functional factors, and he never will be. He is resolutely, ideologically, intractably opposed to this. I consider this bias and blindless to be the essence of 'the Chomskyan program', as it has characterized this program from beginning to end. If Brian wants to make up his own version of 'the Chomskyan program' which leaves this bias out, then I would agree with his statement (though this definition of 'the Chomskyan program' seems bizarre to me). But if we understand 'the Chomskyan program' to mean a rejection of psychological plausibility, empirical data on usage, functional constraints, etc., as has been uniformly characteristic of Chomsky's writings, then I am not at all sure that this will survive him by much. I hope not. John From bresnan at CSLI.STANFORD.EDU Mon Dec 13 18:06:46 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 10:06:46 -0800 Subject: Myhill's message In-Reply-To: Your message of Mon, 13 Dec 1999 10:00:28 +0200. Message-ID: John, I apologize for characterizing your message as "flame bait". You are obviously sincere in your beliefs and not merely trying to elicit reactions from others. As I've tried to show, I think you may be mistaken in your judgements about the state of current linguistic research as being completely polarized into binary camps "pro-Chomsky" and "anti-Chomsky", or "formalist" and "functionalist". You seem to advocate simply writing off all work emanating from institutions or people that you characterize as "pro-Chomsky". (And there are those at MIT who have the same attitude toward you and the other "anti-Chomsky" posters to this list.) But your arguments do not refer to the substantive ideas and empirical claims that are made (as Brian MacWhinney's do, for example). Instead you refer to your judgments about the sociology of the field and even about the personalities and worth of individuals. I was pretty shocked when I saw how personal the remarks in your postings were. Neverthless, like it or not, there is a new functionalism attracting (among others) researchers who have been trained in formal methods and models, and there are new mathematical models for language making use of optimization and probability. The widespread availability of on-line corpora and powerful computers is also having its effect on attitudes toward data and methology: Stanford students working in syntax and semantics, for example, simply start doing corpus based research as a matter of course. I personally think that the most interesting work will hybridize the best from both worlds--research on symbolic representation (the work of the "formalists") and research on data driven cognitive modelling (the work of the "functionalists"), to put it very roughly. In other words, I think that substantive and serious discussions are possible between those on this list and those who you write off. There's no need for us to be so sectarian about our work! Joan From dryer at ACSU.BUFFALO.EDU Mon Dec 13 21:10:56 1999 From: dryer at ACSU.BUFFALO.EDU (Matthew S Dryer) Date: Mon, 13 Dec 1999 16:10:56 -0500 Subject: OT and functionalism Message-ID: The recent discussion of OT and functionalism misses what to me is an essential difference between OT and the relation of functional explanation, particularly competing motivations, to grammar. In OT, as I understand it, the constraints are "in" the grammar, either in the sense that the constraints are stipulated for particular languages, or in the sense that the grammar of the language makes reference to universal constraints, for example in specifying which constraints outrank which other constraints. But on my view, grammars of languages do not make reference to functional principles or motivations. For example, if two languages have a difference that one might describe in terms of economic motivation competing with iconic motivation, where economic motivation wins out in one language and iconic motivation in the other, I would not want to say that there are grammatical rules in these languages that refer to economy or to iconicity. To put the point another way, functional principles and motivations apply primarily at the level of the evolution of particular languages. If economic motivation has won out over iconic motivation with respect to some aspect of the grammar of a particular language, that means that over the past few thousand years, the particular grammatical changes that have led to the current state of the language reflect the influence of that functional principle. But once that has happened, the grammatical rules have an existence that is independent of the explanatory principles that have influenced them. OT shares with theories in the GB tradition of trying to build explanation for why languages are the way they are into the grammars themselves. From my perspective, this is deeply misguided. Languages are highly complex systems, whose properties reflect the interaction of diverse explanatory principles, and I believe that they can and should be described in terms that are independent of the explanatory principles that underlie them. While there are a wide variety of approaches which have been labeled "functionalist", OT as I understand it is fundamentally incompatible with the type of functionalism that I and many others espouse. Matthew Dryer From fjn at U.WASHINGTON.EDU Mon Dec 13 22:45:21 1999 From: fjn at U.WASHINGTON.EDU (Frederick Newmeyer) Date: Mon, 13 Dec 1999 14:45:21 -0800 Subject: OT and functionalism Message-ID: I feel that Matthew Dryer makes an excellent point when he writes: > functional principles and motivations apply > primarily at the level of the evolution of particular languages. If > economic motivation has won out over iconic motivation with respect to > some aspect of the grammar of a particular language, that means that over > the past few thousand years, the particular grammatical changes that have > led to the current state of the language reflect the influence of that > functional principle. But once that has happened, the grammatical rules > have an existence that is independent of the explanatory principles that > have influenced them. As I will try to argue now as concisely as possible, the kind of 'functionalist OT model' that Joan Bresnan has been advocating in her postings (if I understand her correctly), in which the constraints themselves are directly functionally motivated, falters when one tries to apply it to a wide variety of disparate phenomena within a particular language. OT analyses, particular in the realm of syntax, have, in general, been rather circumscribed in their domain of application. Typically, they focus on some little corner of the syntax, such as clitic order, auxiliary inversion, and so on. The machinery intrinsic to the OT approach works beautifully in such situations. But grammars are tightly integrated wholes, so decisions about how one process should be handled are likely to have repercussions for another, seemingly unrelated, process. This fact leads ultimately to a much greater disparity between grammatical principles and their functional roots than is posited by 'functionalist OT models'. Let's posit two functionally-motivated OT constraints for English, ICONICITY and HEAVY-LAST: ICONICITY: Syntactic constituents reflect semantic units. HEAVY-LAST: Heavy constituents follow light constituents. ICONICITY can be illustrated by the fact that, in every formal account, adjectives are generated under the same phrasal node as the noun that they modify. HEAVY-LAST can be illustrated by the fact that within the verb phrase, sentential complements are positioned after phrasal complements. The functional roots of these constraints are so obvious that no discussion is necessary. Now, then, which constraint is ranked higher for English, ICONICITY or HEAVY-LAST? Well, it depends. In some cases, we have identical grammatical elements in variant orders with no meaning difference, each option corresponding to a different ranking of the two constraints. For example, both of the following are grammatical English sentences: (1) a. A man who was wearing a silly-looking red hat dropped by today. b. A man dropped by today who was wearing a silly-looking red hat. Sentence (1a) reflects a ranking of ICONICITY over HEAVY-LAST; sentence (1b) the reverse ranking. In some cases, however, only a higher ranking of ICONICITY is possible. Simple adjective phrases cannot be extraposed from the nouns that they modify, no matter how heavy they are: (2) a. An extremely peculiar-looking man dropped by today. b. *A man dropped by today extremely peculiar-looking. And in other cases, HEAVY-LAST seems to be ranked over ICONICITY. When comparatives are used attributively, the adjective is separated from its complement by the head noun, despite the fact that together they serve to modify semantically that head noun (3a-b). Their structural unity can be obtained only in a manner that is consistent with HEAVY-LAST (3c): (3) a. That's a more boring book than any I have ever read. b. *That's a more boring than I have ever read book. c. That's a book more boring than any I have ever read. What is the solution to this ranking paradox? Only, I would say, to abandon ICONICITY and HEAVY-LAST as constraints internal to English grammar. Rather, what we need are something much more like the 'parochial' rules, principles, and constraints of standard models of grammar that interact to yield the grammatical sentences of the language. That move, however, is incompatible with 'functionalist OT models'. The relationship between detailed grammatical statements --- such as those that position the constituents of NP --- and their functional motivations can be extremely indirect. In other words, we are left with the sort of view advocated in Newmeyer (1998), in which, in a global / historical sense, grammars reflect external forces, but without each language-internal grammatical statement being tied to a particular functional motivation. Newmeyer, Frederick J. (1998). Language form and language function. Cambridge, MA: MIT Press. --fritz newmeyer From sstraigh at BINGHAMTON.EDU Tue Dec 14 04:17:32 1999 From: sstraigh at BINGHAMTON.EDU (H Stephen Straight) Date: Mon, 13 Dec 1999 23:17:32 -0500 Subject: OT, Functionalism, and the Myth of G Message-ID: The problem with Chomskyan (and post-Chomskyan, and the preponderance of pre-Chomskyan) theorizing derives not from over-formalizing, under-empiricizing, or most of the other errors identified in the current FUNKNET discussion. What it all boils down to is belief in the Myth of G. Brian Macwhinney points us in the right direction by urging us to reject the idea that each linguistic event derives from a single underlying form. Underlying forms derive in turn from a Grammar, in which each such form serves as the key to an actual or potential linguistic event, whether as a product of language production or an object of language perception. Let me attempt to debunk this pernicious Myth of G by critiquing Fritz Newmeyer's helpful contribution to this awesome FUNKNET thread. Fritz reasonably (and pro-functionally) asks us to "posit two functionally-motivated OT constraints for English [:] ICONICITY: Syntactic constituents reflect semantic units. HEAVY-LAST: Heavy constituents follow light constituents. ICONICITY can be illustrated by the fact that, in every formal account, adjectives are generated under the same phrasal node as the noun that they modify. HEAVY-LAST can be illustrated by the fact that within the verb phrase, sentential complements are positioned after phrasal complements." Fritz then poses the (false, OT-motivated) question, "Now, then, which constraint is ranked higher for English, ICONICITY or HEAVY-LAST?" Because Fritz, like OT, has only G to turn to, he concludes that "In some cases, we have identical grammatical elements in variant orders with no meaning difference, each option corresponding to a different ranking of the two constraints." But Fritz's supposedly conclusive examples mix receptive and expressive parameters: > For example, both of the following are grammatical English > sentences: > > (1) a. A man who was wearing a silly-looking red hat dropped by > today. > b. A man dropped by today who was wearing a silly-looking red hat. > > Sentence (1a) reflects a ranking of ICONICITY over HEAVY-LAST; sentence > (1b) the reverse ranking. In some cases, however, only a higher ranking of > ICONICITY is possible. Simple adjective phrases cannot be extraposed from > the nouns that they modify, no matter how heavy they are: > > (2) a. An extremely peculiar-looking man dropped by today. > b. *A man dropped by today extremely peculiar-looking. Omitted from Fritz's purview is the fact that 2b works just fine with a healthy, listener-warning pause: c. A man dropped by today ... extremely peculiar-looking. Also omitted is the fact that the explicit relativizer present in examples (1) is in fact optional: (1) c. A man wearing a ... d. A man dropped by today wearing ... So what can we make of these facts? Nothing about Grammar, if it is assumed to be neutral with regard to the demands of speakers and listeners. Fritz appeals to G-specific arbitrarinesses specifiable only in non-functional (i.e., formal) terms, though given the counter-examples I don't think this tactic will save him. However, if we instead see these competing demands as resulting in a gradient of preferences weighted differently for the domain of the receiver versus the domain of the producer, and perhaps even competing within each of these domains, we (rightly) conclude that one and the same "sentence" (a term already unfairly biased toward the G side of this issue) can be OK or problematic for the one but bad or differently problematic for the other. Take for example Fritz's frightful "*That's a more boring than I have ever read book." If we consider that this example violates the pragmatic (functional?) OLD-BEFORE-NEW principle because "book" is clearly topic rather than comment, and that it also ignores counterexamples such as "Thank you for the much needed by a teenager present" (attested in my own experience), we can conclude that Fritz has allowed his asterisks to lull him into complacency about the unidimensionality of grammatical judgments. Fritz rightly criticizes "functionalist OT models" for trying to use functionalist principles to predict grammaticality judgments, but what's wrong is not the principles but rather the judgments: They are a mixed and inconclusive source of evidence because language users employ a wide variety of different criteria to arrive at such (often demonstrably ephemeral) judgments. Fritz's conclusion that "grammars reflect external forces, but without each language-internal grammatical statement being tied to a particular functional motivation" sums up the Myth of G in a manner both ironic and poignant: "Language-internal" statements amount to no more than the unidimensional "prescriptive" judgments of popular grammarians of the likes of Kilpatrick, Newman, and Simon, judgments long since debunked by linguists from Sledd to Bolinger to Pinker. Why can't linguists admit once and for all the uncertainty and contradictoriness of their data on "grammaticality"? When will linguists stop living in armchairs and consider seriously the evidence provided by corpora, elicitation, and experiment? We say things no one can (easily) parse, and we (readily) parse things no one (hardly ever) says. Shouldn't these facts (along with a myriad of others) tell us something really basic about the (non-G) nature of construing and saying? Best. 'Bye. Steve ________________________________________________________________ H Stephen Straight - straight at binghamton.edu From bresnan at CSLI.STANFORD.EDU Tue Dec 14 05:22:25 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 21:22:25 -0800 Subject: OT and functionalism In-Reply-To: Your message of Mon, 13 Dec 1999 16:10:56 EST. Message-ID: In his posting of Monday, 13 Dec., Matthew Dryer makes the point that functional (principles, motivations, or) constraints are external to grammars, while OT constraints are internal to (OT) grammars. Hence, he believes: > OT shares with theories in the GB tradition of trying to build explanation > for why languages are the way they are into the grammars themselves. From > my perspective, this is deeply misguided. Languages are highly complex > systems, whose properties reflect the interaction of diverse explanatory > principles, and I believe that they can and should be described in terms > that are independent of the explanatory principles that underlie them. > > While there are a wide variety of approaches which have been labeled > "functionalist", OT as I understand it is fundamentally incompatible with > the type of functionalism that I and many others espouse. This certainly sounds like an unassailable position, but that is because it is true by definition. If you *define* a functional constraint to be something external to a "grammar", and you *define* an OT "grammar" as something that contains internal constraints, then how could his conclusion be other than true? But I think Matthew's point betrays a misconception about what exactly an OT grammar is. He elaborates: > But on my view, grammars of languages do not make reference to functional > principles or motivations. For example, if two languages have a > difference that one might describe in terms of economic motivation > competing with iconic motivation, where economic motivation wins out in > one language and iconic motivation in the other, I would not want to say > that there are grammatical rules in these languages that refer to economy > or to iconicity. OT has no grammatical rules that refer to economy or to iconicity. Language particularity is ultimately simply a harmonic function over the space of possible forms. If you look inside an OT "grammar", you find a representational basis (this is sometimes called "GEN") which specifies the set of possible structures, and an optimizing component (called "EVAL") which optimizes the candidate structures against the conflicting universal constraints in such a way as to minimize violations. The constraint component is "external" to the structure component. He writes: > To put the point another way, functional principles and motivations apply > primarily at the level of the evolution of particular languages. If > economic motivation has won out over iconic motivation with respect to > some aspect of the grammar of a particular language, that means that over > the past few thousand years, the particular grammatical changes that have > led to the current state of the language reflect the influence of that > functional principle. But once that has happened, the grammatical rules > have an existence that is independent of the explanatory principles that > have influenced them. As I indicated in my reply to J. MacFarlane's message, the work of "new functionalist" OT phonologists like Donca Steriade is based on the hypothesis that factors such as perceptibility and avoidance of articulatory effort "play a role in shaping sound patterns, not only in an evolutionary sense--as argued by John Ohala and Bjorn Lindblom--but also by defining the grammatical constraints whose interactions yield the phonologies of individual languages." Judith Aissen and I believe that the same may be true in syntax: The same functionally motivated constraints or "forces" that shape evolutionary properties and appear in crosslinguistic typological asymmetries also emerge within the grammars of particular languages. We have several papers making this point in the domain of syntax in some detail, and these are available from our Optimal Typology Project website (http://www-ot.stanford.edu/ot/), so I won't repeat the evidence and arguments here. This is what we are trying to do in our project. I will only give one very simple example (from Rene Kager's 1999 CUP textbook on Optimality Theory). Across languages, voiced obstruents are typologically marked: there are languages having only voiceless obstruents (e.g. Polynesian), and languages having both voiced and voiceless obstruents (e.g. English), but no (or hardly any) languages having only voiced obstruents. This typological asymmetry might be explained in terms of constraints on articulatory effort and perception that shaped the evolution of languages to explain why voiced obstruents are more restricted crosslinguistically. But within the phonological components of the grammars of particular languages we see the same thing: voiced obstruents are restricted in their *language-internal* distributions. For example, Dutch and German have a voicing contrast in obstruents, which appears in syllable onsets--a very salient position-- but voiced obstruents are devoiced in syllable codas. In contrast, there appear to be few or no languages whose language-internal phonologies have a voicing contrast in syllable codas and no voicing contrast in syllable onsets. OT can explain the positional neutralization of obstruent voicing in syllable codas and relate it to the typological of segment inventories across languages. The reason is simply that the *same* constraints reflecting ease of perception and articulatory effort are present in the grammars of all languages (as the universal constraint set). No other of grammatical theory that I know of does this. In my work--and in Judith's--we are developing analyses of syntactic analogues of these phonological phenomena. (I refer you to the papers at the website above.) I think it's extremely exciting, and has potential for truly integrative collaborations across the formal/functional divide. Of course, at the level of syntax it is much harder to find direct perceptual and cognitive evidence for the markedness constraints, compared to the work on phonetic perception and articulation that Steriade and others appeal to. There is much to be done. I hope that Matthew's statement of what a true functionalist believes isn't the last word on this topic. Joan From bresnan at CSLI.STANFORD.EDU Tue Dec 14 05:24:02 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 21:24:02 -0800 Subject: OT and functionalism In-Reply-To: Your message of Mon, 13 Dec 1999 14:45:21 PST. Message-ID: Isn't this fun? Too bad grades are due tomorrow. I will be brief in my reply to Fritz's message, since it piggybacks on Matthew's anyway, to which I replied at length. Fritz thinks that OT doesn't mesh with his conception of what functionalism is, namely-- "... the sort of view advocated in Newmeyer (1998), in which, in a global / historical sense, grammars reflect external forces, but without each language-internal grammatical statement being tied to a particular functional motivation. Newmeyer, Frederick J. (1998). Language form and language function. Cambridge, MA: MIT Press." My previous reply to Matthew Dryer applies just as well here, so I won't repeat it. But Fritz argues for this point in a peculiar way, by giving a failed OT analysis of his own design, and blaming it on the framework. (Hmmm... We certainly wouldn't accept this reasoning from a student, would we? ...sorry, I have grades on my mind. :-)) He also makes several very general characterizations about OT work, such as: "OT analyses, particular in the realm of syntax, have, in general, been rather circumscribed in their domain of application. Typically, they focus on some little corner of the syntax, such as clitic order, auxiliary inversion, and so on." ---and: "The problem with the framework seems to be that it "falters when one tries to apply it to a wide variety of disparate phenomena within a particular language." But Fritz hasn't really shown us this, has he? It's a bit odd for Fritz to talk about grammatical coverage, since OT syntax is still in its infancy, and the only formal grammatical frameworks that have produced genuine wide coverage grammars of real languages are the constraint-based lexicalist grammars in LFG, HPSG, etc. which are actively used in large-scale computational NLP projects and applications. (I am told that the Minimalist Program has advanced by shrinking its coverage of a wide variety of disparate phenomena.) It might be of interest to those concerned about wide-scale descriptive grammar coverage and OT that Jonas Kuhn (whose work I referred to in previous postings) has implemented the Grimshaw fragment of English syntax (on auxiliary inversion) in the XLE parser at Stuttgart. This is, of course, a toy fragment. But his formal analysis of the generation and recognition problem in OT-LFG now makes it possible to generate, parse, and study real grammars of the sort that have broad descriptive coverage. This is a remarkable advance. Have a look at: http://www.ims.uni-stuttgart.de/~jonas/ This is just the beginning, but the new work coupling the mathematically well defined group of constraint based grammars with the OT model of constraint interaction seems promising to me. TTFN-- Joan P.S. Lest anyone get the impression that I am an advocate for OT, let me repeat my warning that I am only a student of OT, and I'm only doing it in my spare time (of which there isn't much). *---------------------------------------- ______ __o __o Joan Bresnan bresnan at stanford.edu ______ _`\<,_ _`\<,_ *---------------------------------------- ______ (*)/ (*) (*)/ (*) From bresnan at CSLI.STANFORD.EDU Tue Dec 14 06:13:21 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 13 Dec 1999 22:13:21 -0800 Subject: Chomsky bashing Message-ID: Several of my friends at MIT have privately told me that they consider it inappropriate that I spoke up against the public disparagement of Tom Wasow and Stanford by a post-er to this list, and yet said nothing about the disparagement of Noam Chomsky in the same message. It didn't occur to me that I should speak up for MIT as well as Stanford, since they are certainly an articulate bunch there at MIT and could reply even better than I. But they point out that since I am the current President of the LSA, my silence in regard to some of the Chomsky bashing that passes for discussion on this list seems signficant. I would therefore like to urge the members of this list to consider that we are all human beings, even the most famous (and the least famous!) among us. It is good fun to have a lively interplay of ideas and arguments, and we all enjoy witticisms (especially at others' expense)---but isn't it more enlightening to respect each other enough to listen and answer with civility? I have learned a lot from the intelligent postings and queries on this list in reponse to my recent messages, and I very much respect those of you who took the trouble to formulate such thoughtful and interesting questions and criticisms. I also have great respect for Noam Chomsky, who was my teacher. Even John Myhill's very high regard for Noam Chomsky is evident in his postings--comparing Chomsky in linguistics to Einstein in physics, referring to those of us who have explored alternative grammatical architectures as "not as good as the original". Chomsky is the original, and we are all of us his students, directly or indirectly. Joan P.S. That "we are most of us" construction in my last sentence surely shows that the V-movement analysis of Q-float under the internal-VP subject analysis can't be right. *---------------------------------------- ______ __o __o Joan Bresnan bresnan at stanford.edu ______ _`\<,_ _`\<,_ *---------------------------------------- ______ (*)/ (*) (*)/ (*) From beaumar at HOTMAIL.COM Tue Dec 14 06:44:12 1999 From: beaumar at HOTMAIL.COM (A R) Date: Mon, 13 Dec 1999 22:44:12 PST Subject: Chomsky Message-ID: You know, Funknetters, I don't know that I'd call myself either a formalist or a functionalist. But one of the most damning things I know about Chomsky is that he *never* goes to LSA, the one place where he might face genuine criticism. A friend of mine, a formalist, used to argue that he simply never went to academic conferences. This year, I discover that this is clearly untrue. Chomsky is going to talk at the MLA, to those who do literature. It's only his own that he refuses to talk to at conferences. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From john at RESEARCH.HAIFA.AC.IL Tue Dec 14 07:29:30 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Tue, 14 Dec 1999 09:29:30 +0200 Subject: common ground Message-ID: You may be right, Joan. You may be right. For the first 15 years of my linguistics career, I dutifully read formal linguistics articles of all types, in approaches such as your LFG or Simon Dik's functional grammar which claimed to be something new, something integrating formal and functional findings and results, to an extent I find hard to believe any other functionalist could stand (with the exception of Matthew Dryer). I have thought about formal issues enough that I can say that, e.g., I have been more influenced by 'On Wh-movement' than by anything else Chomsky has written, more by Burzio than by Rizzi, etc. I have never refrained from referring to formal articles or even arguments when they were relevant to my research. I even once gave a straight LFG paper on Indonesian at the LSA. But none of it, NONE of it, not LFG, not FG, not Construction Grammar, escaped from the same conceptual trap. So I gave up attempting to keep up; five years ago, after yet another 'Chomskyan revolution' which was more of the same, I deduced that it was a waste of time and decided to devote my research time to other things. I have been listening to you saying for nigh-on 20 years that you-all are doing something 'really different' from Chomsky, I spent a lot of time investigating whether it was true, and I came to the conclusion that it simply wasn't (which was what Chomsky himself maintained as well). But I have fallen behind. You may be right. This time there may really be a wolf. It is quite possible. But it will be hard for me to convince myself to take the time out from other things to check it out for myself now. But maybe some day... John >John, I apologize for characterizing your message as "flame bait". >You are obviously sincere in your beliefs and not merely trying to >elicit reactions from others. > >As I've tried to show, I think you may be mistaken in your judgements >about the state of current linguistic research as being completely >polarized into binary camps "pro-Chomsky" and "anti-Chomsky", or >"formalist" and "functionalist". You seem to advocate simply writing >off all work emanating from institutions or people that you >characterize as "pro-Chomsky". (And there are those at MIT who have >the same attitude toward you and the other "anti-Chomsky" posters to >this list.) But your arguments do not refer to the substantive ideas >and empirical claims that are made (as Brian MacWhinney's do, for >example). Instead you refer to your judgments about the sociology of >the field and even about the personalities and worth of individuals. >I was pretty shocked when I saw how personal the remarks in your >postings were. > >Neverthless, like it or not, there is a new functionalism attracting >(among others) researchers who have been trained in formal methods and >models, and there are new mathematical models for language making use >of optimization and probability. The widespread availability of >on-line corpora and powerful computers is also having its effect on >attitudes toward data and methology: Stanford students working in >syntax and semantics, for example, simply start doing corpus based >research as a matter of course. > >I personally think that the most interesting work will hybridize the >best from both worlds--research on symbolic representation (the work >of the "formalists") and research on data driven cognitive modelling >(the work of the "functionalists"), to put it very roughly. > >In other words, I think that substantive and serious discussions are >possible between those on this list and those who you write off. >There's no need for us to be so sectarian about our work! > >Joan From john at RESEARCH.HAIFA.AC.IL Tue Dec 14 07:42:32 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Tue, 14 Dec 1999 09:42:32 +0200 Subject: Chomsky Message-ID: Relating to the comment from AR (beaumar at HOTMAIL.COM, copied below), my father, a mathematical logician of the same name as me, met and spent a year around Chomsky at Princeton in the late 1950's, when Chomsky was attempting to pass himself off as a mathematician. The real mathematicians there observed with amusement that Chomsky would attempt to act like a mathematician, throwing around terms designed to impress, when speaking to philosophers, political scientists, etc., but when a real mathematician entered the conversation, Chomsky would either making an excuse to leave or change the topic to something like politics. John >You know, Funknetters, I don't know that I'd call myself either >a formalist or a functionalist. But one of the most damning things >I know about Chomsky is that he *never* goes to LSA, the one place where he >might face genuine criticism. A friend of mine, a formalist, >used to argue that he simply never went to academic conferences. > >This year, I discover that this is clearly untrue. Chomsky is going >to talk at the MLA, to those who do literature. It's only his own that he >refuses to talk to at conferences. > > >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com From haspelmath at EVA.MPG.DE Tue Dec 14 11:08:38 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Tue, 14 Dec 1999 11:08:38 +0000 Subject: optimality in synchrony and diachrony Message-ID: Joan, Matthew and Fritz, Thanks a lot for this truly illuminating discussion on the place of functional constraints in grammatical explanation. I think Matthew formulated the most coherent functionalist position very clearly: functional forces operate in language change, and should not be part of grammatical description (functionalist pronouncements have often been less clear, so that Fritz understandably citicizes functionalists for "stating functional forces in the grammar itself", 1998:141). In this debate, my tendency is to side with Matthew and Fritz. In my paper "Optimality and diachronic adaptation" (ROA-302-0399) I argued in particular that the diachronic dimension is necessary for true explanation. (This paper will appear in Zeitschrift fuer Sprachwissenschaft, with peer commentary by Newmeyer, Croft, Traugott, Wurzel, Dahl, Haider, Kirby, and others.) I am not so sure, however, that both sides could not be right, i.e. that functional factors are crucial for shaping grammars in diachronic change, AND that the best synchronic description is one that uses these constraints, perhaps in an OT-like fashion. Matthew, why would you exclude this possibility? (We have seen Fritz's answer, but as Joan pointed out, it is quite incomplete.) Joan, I must confess that I didn't understand your remarks that address Matthew's claim that OT "is trying to build explanation for why languages are the way they are into the grammars themselves": >OT has no grammatical rules that refer to economy or to iconicity. >Language particularity is ultimately simply a harmonic function over >the space of possible forms. If you look inside an OT "grammar", you >find a representational basis (this is sometimes called "GEN") which >specifies the set of possible structures, and an optimizing component >(called "EVAL") which optimizes the candidate structures against the >conflicting universal constraints in such a way as to minimize >violations. The constraint component is "external" to the structure >component. Are you saying that functional OT constraints are neither "built into the grammars themselves" nor are restricted to diachrony? Could you please clarify this? I think I didn't understand the thrust of Joan's example about the devoicing tendency across languages and in particular languages either. This is something that typologists have long been aware of (see e.g. Bill Croft's markedness chapter in his typology textbook). The usual functionalist explanation for both cross-linguistic patterns and language-particular regularities is that they show the effect of diachronic change. Martin -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From jlmendi at POSTA.UNIZAR.ES Tue Dec 14 12:24:57 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Tue, 14 Dec 1999 13:24:57 +0100 Subject: Myhill on Chomsky Message-ID: John Myhill gave us the following information: >Relating to the comment from AR (beaumar at HOTMAIL.COM, copied below), my >father, a mathematical logician of the same name as me, met and spent a >year around Chomsky at Princeton in the late 1950's, when Chomsky was >attempting to pass himself off as a mathematician. The real mathematicians >there observed with >amusement that Chomsky would attempt to act like a mathematician, throwing >around terms designed to impress, when speaking to philosophers, political >scientists, etc., but when a real mathematician entered the conversation, >Chomsky would either making an excuse to leave or change the topic to >something like politics. > >John If this kind of information is considered useful in order to evaluate the validity of the Chomskyan program, I think the following will serve as well: "For years now, Chomsky has been one of the very few scientists and philosophers who is widely read. The number of citations of his work in the Arts and Humanities Citation Index (nearly 4.000 between 1980 and early 1992) makes him the most cited living person and the eighth overall (...), and the citations of his work in the Social Science Citation Index (7.499 between 1972 and early 1992) are likely to make him the most cited living person there as well. Last, but certainly not least, from 1974 to 1992 he was cited 1.619 times in the Science Citation Index" (I quote from the Foreword to _Noam Chomsky. Critical Assessments_, Routledge, London & New York, 1994: p. xxii) It seems quite exaggerated for a false mathematician! Best Regards, Jose-Luis From john at RESEARCH.HAIFA.AC.IL Tue Dec 14 13:57:27 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Tue, 14 Dec 1999 15:57:27 +0200 Subject: corpus based research at Stanford Message-ID: Joan (Bresnan), you wrote: `Stanford students working in syntax and semantics, for example, simply start doing corpus based research as a matter of course.' Is this really true? Who's teaching these classes? Have you committed tenure-track faculty lines to people trained in corpus-based research? Can you give me references to relevant papers written by such people or their professors? This sounds interesting. John From reich at chass.utoronto.ca Tue Dec 14 18:20:36 1999 From: reich at chass.utoronto.ca (Peter A. Reich) Date: Tue, 14 Dec 1999 13:20:36 -0500 Subject: Myhill on Chomsky Message-ID: I'm sure God is cited more than Chomsky (I will not engage in a discussion as to whether God is Living), and probably Bill Gates is mentioned more than Chomsky as well in the popular media. And Clinton. So what? It is perfectly respectable in Canada, at least, to be an Atheist, and to prefer Linux or Macintosh. And, in the US, to be a Republican. Eat shit; 50 billion insects can't be wrong.--Peter Reich "For years now, Chomsky has been one of the very few scientists and philosophers who is widely read. The number of citations of his work in the Arts and Humanities Citation Index (nearly 4.000 between 1980 and early 1992) makes him the most cited living person and the eighth overall (...), and the citations of his work in the Social Science Citation Index (7.499 between 1972 and early 1992) are likely to make him the most cited living person there as well. Last, but certainly not least, from 1974 to 1992 he was cited 1.619 times in the Science Citation Index" (I quote from the Foreword to _Noam Chomsky. Critical Assessments_, Routledge, London & New York, 1994: p. xxii) It seems quite exaggerated for a false mathematician! Best Regards, Jose-Luis Mendivil Giro From bresnan at CSLI.STANFORD.EDU Wed Dec 15 01:58:52 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 17:58:52 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Tue, 14 Dec 1999 15:57:27 +0200. Message-ID: In addition to Tom Wasow's reply concerning his own corpus-based research, I would point you to Chris Manning, a new member of our department (http://www.stanford.edu/~manning/). Manning maintains a very useful and well-regarded web list of resources for statistical and corpus-based natural language processing: http://www.sultry.arts.usyd.edu.au/links/statnlp.html In addition, Anne Copestake, Senior Researcher at CSLI, provides a wealth of technical and practical experience available to our student researchers. Joan Bresnan > > `Stanford students working in syntax and semantics, for example, simply > start doing corpus based research as a matter of course.' > > Is this really true? Who's teaching these classes? Have you committed > tenure-track faculty lines to people trained in corpus-based research? Can > you give me references to relevant papers written by such people or their > professors? > This sounds interesting. > > John > From ellen at CENTRAL.CIS.UPENN.EDU Wed Dec 15 05:20:30 1999 From: ellen at CENTRAL.CIS.UPENN.EDU (Ellen F. Prince) Date: Wed, 15 Dec 1999 00:20:30 EST Subject: corpus based research at Stanford Message-ID: >In addition to Tom Wasow's reply concerning his own corpus-based >research, I would point you to Chris Manning, a new member of our >department (http://www.stanford.edu/~manning/). Manning maintains a >very useful and well-regarded web list of resources for statistical and > corpus-based natural language processing: > >http://www.sultry.arts.usyd.edu.au/links/statnlp.html > >In addition, Anne Copestake, Senior Researcher at CSLI, provides a >wealth of technical and practical experience available to our student >researchers. > >Joan Bresnan In addition, you've got John Rickford... :-) >> >> `Stanford students working in syntax and semantics, for example, simply >> start doing corpus based research as a matter of course.' >> >> Is this really true? Who's teaching these classes? Have you committed >> tenure-track faculty lines to people trained in corpus-based research? Can >> you give me references to relevant papers written by such people or their >> professors? >> This sounds interesting. >> >> John >> - --AAA10908.945235077/central.cis.upenn.edu-- ------- End of Forwarded Message From bresnan at CSLI.STANFORD.EDU Wed Dec 15 06:38:43 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 22:38:43 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Wed, 15 Dec 1999 00:20:30 EST. <199912150520.AAA10973@central.cis.upenn.edu> Message-ID: Yes, thank you. And we've got Penny Eckert, and we've got Beth Levin... I think that more corpus-based research from more perspectives on language is possible at Stanford than almost anywhere. (Penn being an exception to the rule... :-) J. >>>"Ellen F. Prince" said: > >In addition to Tom Wasow's reply concerning his own corpus-based > >research, I would point you to Chris Manning, a new member of our > >department (http://www.stanford.edu/~manning/). Manning maintains a > >very useful and well-regarded web list of resources for statistical and > > corpus-based natural language processing: > > > >http://www.sultry.arts.usyd.edu.au/links/statnlp.html > > > >In addition, Anne Copestake, Senior Researcher at CSLI, provides a > >wealth of technical and practical experience available to our student > >researchers. > > > >Joan Bresnan > > In addition, you've got John Rickford... > > :-) > > > >> > >> `Stanford students working in syntax and semantics, for example, simply > >> start doing corpus based research as a matter of course.' > >> > >> Is this really true? Who's teaching these classes? Have you committed > >> tenure-track faculty lines to people trained in corpus-based research? Can > >> you give me references to relevant papers written by such people or the ir > >> professors? > >> This sounds interesting. > >> > >> John > >> > > - --AAA10908.945235077/central.cis.upenn.edu-- > > > ------- End of Forwarded Message > --------------------------------- Joan Bresnan From bresnan at CSLI.STANFORD.EDU Wed Dec 15 06:57:22 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 22:57:22 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Wed, 15 Dec 1999 00:20:30 EST. <199912150520.AAA10973@central.cis.upenn.edu> Message-ID: Eve Clark! That brings us to: people who do corpus-based research in Stanford linguistics and CSLI: Tom Wasow, Chris Manning, John Rickford, Beth Levin, Penny Eckert, Eve Clark, Ann Copestake. representing syntax, computational linguistics, sociolinguistics, semantics, language acquisiton, and the lexicon... > > >In addition to Tom Wasow's reply concerning his own corpus-based > >research, I would point you to Chris Manning, a new member of our > >department (http://www.stanford.edu/~manning/). Manning maintains a > >very useful and well-regarded web list of resources for statistical and > > corpus-based natural language processing: > > > >http://www.sultry.arts.usyd.edu.au/links/statnlp.html > > > >In addition, Anne Copestake, Senior Researcher at CSLI, provides a > >wealth of technical and practical experience available to our student > >researchers. > > > >Joan Bresnan > > In addition, you've got John Rickford... > > :-) > > > >> > >> `Stanford students working in syntax and semantics, for example, simply > >> start doing corpus based research as a matter of course.' > >> > >> Is this really true? Who's teaching these classes? Have you committed > >> tenure-track faculty lines to people trained in corpus-based research? Can > >> you give me references to relevant papers written by such people or their > >> professors? > >> This sounds interesting. > >> > >> John > >> > > - --AAA10908.945235077/central.cis.upenn.edu-- > > > ------- End of Forwarded Message > From bresnan at CSLI.STANFORD.EDU Wed Dec 15 06:52:27 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Tue, 14 Dec 1999 22:52:27 -0800 Subject: optimality in synchrony and diachrony In-Reply-To: Your message of Tue, 14 Dec 1999 11:08:38 GMT. <38562528.818756A1@eva.mpg.de> Message-ID: (long message, last message, please forgive) Martin Haspelmath asks me to clarify my reply to Matthew Dryer. Matthew argued (in essence) that OT can't be truly functionalist because functional constraints are external to grammar, shaping language in an evolutionary way, while OT constraints are internal to grammar. Matthew and Fritz Newmeyer believe that for this reason OT is fundamentally misguided, and Martin Haspelmath is inclined to agree with them. My reply was that this argument sounds plausible because it is a tautology, true only by definition. It rests on an erroneous definition of "grammar", which is not what an OT "grammar" is. What is a grammar and how exactly can it be "shaped" by evolutionary forces? Matthew's view of grammar may be shared by many on this list, but it one that has been completely abandoned among most phonologists and also among a number of OT syntacticians. It is the view expressed in the following quotations from Matthew's original msg (emphasis added): "For example, if two languages have a difference that one might describe in terms of economic motivation competing with iconic motivation, where economic motivation wins out in one language and iconic motivation in the other, I WOULD NOT WANT TO SAY THAT THERE ARE GRAMMATICAL RULES IN THESE LANGUAGES THAT REFER TO ECONOMY OR TO ICONICITY." and: "... over the past few thousand years, the particular grammatical changes that have led to the current state of the language reflect the influence of that functional principle. But once that has happened, THE GRAMMATICAL RULES HAVE AN EXISTENCE THAT IS INDEPENDENT of the explanatory principles that have influenced them." The conception of grammar that these passages invoke is one in which a grammar consists of RULES and/or PRINCIPLES AND PARAMETERS. Since the structure of a language is determined by the rules and/or parameter settings of its grammar, on this view, it doesn't make sense for there to be "internal" functional forces also shaping the language. [I leave aside the question of how exactly languages are supposed to be shaped by EXTERNAL forces on this view.] But an OT "grammar" is not a grammar in this sense. Nor is it a "grammar" in the Principles and Parameters sense. OT rejects Principles (because they are inviolable) and Parameters (because, among other reasons, they are categorically either on or off and therefore incapable of explaining the EMERGENCE of the UNMARKED) [see also the critique of parameter learnability in Tesar and Smolensky]. So OT is a much more radical departure from the conventional thinking about what grammars are than Matthew and Fritz seem to recognize. An OT "grammar" is a model of how a set of possible structures (the candidate set) is shaped by the prioritization of conflicting universal constraints, into a particular language. The constraints reflect (hypothesized) properties of the human articulatory, perceptual, and cognitive systems. They evaluate the markedness and the faithfulness of the structures as expressions of content. The faithfulness constraints require that features of the input content be preserved in the output expression. They thus serve the communicative function of expressing contrasts in content, protecting content against the eroding effects of markedness constraints on forms. Markedness constraints penalize complex or `difficult' structures, and so tend to erode contrasts. A particular language harmonizes the conflicting constraints by prioritizing (ranking) them in its own way. Note that the constraint component that does the selection is external to the structure-generating component. Unlike the older, conventional conceptions of grammar which it replaces, OT provides both an EXPLICIT way to model how languages can be shaped by functionalist pressures and also an explanation for why language-internal processes or "rules" (which it deconstructs completely in terms of markedness and faithfulness) reflect typological patterns. My example (from Kager's textbook): Final obstruent devoicing is a productive process in the grammar of Dutch among other languages. Presumably it applies to borrowed words, and affects Dutch second-language learners of English, etc. etc. It is not a historical relic, frozen in a few lexical forms, but a living process or "rule" of the sort that phonologists would write in their descriptions of the grammar of Dutch (in the broad sense of grammar used here). The OT analysis of this devoicing "rule" in Dutch derives its effects from ranking three constraints: (1) faithfulness to voicing contrasts, (2) one markedness constraint reflecting the difficulty of perceiving voicing contrasts in certain less-salient positions (syllable codas), and (3) another constraint reflecting the general markedness of voiced obstruents (compared to vowels, for example)--you've got a closure and you're vocalizing, and it's tough. [pardon my phonetics]. Now if you take this analysis and arbitrarily rerank these constraints in all possible ways, you create little obstruent-voicing/devoicing "grammars" for different languages. What you find, is that all these rankings give you only three different possibilities: voicing contrasts in obstruents everywhere (like English); voiceless obstruents everywhere (like Polynesian); and voicing contrasts present in the salient onset position of syllables and absent is the coda position (Dutch). In a nutshell, the analysis of final devoicing in Dutch predicts the existence of the well known typological asymmetry in obstruent voicing across languages. It is because the OT constraints are universal and not language-particular rules or parameter settings, that this deep connection between typology and language-internal "grammar" is possible. You literally cannot do the (real) OT grammar of any particular language without doing typology. Martin Haspelmath wrote: > I think I didn't understand the thrust of Joan's example about the > devoicing tendency across languages and in particular languages either. > This is something that typologists have long been aware of (see e.g. > Bill Croft's markedness chapter in his typology textbook). The usual > functionalist explanation for both cross-linguistic patterns and > language-particular regularities is that they show the effect of > diachronic change. > But precisely how? Joan Bresnan From Zylogy at AOL.COM Wed Dec 15 08:15:21 1999 From: Zylogy at AOL.COM (Jess Tauber) Date: Wed, 15 Dec 1999 03:15:21 EST Subject: optimality in synchrony and diachrony Message-ID: I've been very interested in this particular thread. I'll make my points and then get out. Storage forms of lexical items to be acted on by faithfulness and markedness constraints have to come from somewhere. Historical change shows that simplification of complex structures is the rule once structures have sufficient staying power and are in the lexicon proper. Combinatory processes create new structures of higher complexity and markedness, which can then undergo further simplification, etc., but these process are live, synchronic ones. Mastication metaphorically. Yet ultimately the origins of roots must be in something similar to expressive vocabulary such as ideophones, even if such origin is quite ancient, and the path traveled through the generations and between languages tortuous. And as morphology derives from free forms, it is probably safe to bet that everything boils down to old expressive vocabulary. It is of interest, then, that such vocabulary is often seemingly resistant to historical changes, if only because it isn't really in the lexicon, but created anew (if not de novo) with each use. And such vocabulary is hardly integrated syntactically, and as such is extraclausal and uninflected. My own research shows that expressive vocabulary obeys universal typological principles. But humanity just can't leave well enough alone. Combination is just too valuable a way to be precise, if not concise. As integration does start to occur the size and complexity of combinatory products starts getting in the way of communicative efficiency, and so we have the slippery slope to simplification. You've heard this all before, in other contexts. I bring all this up because I believe we need a theoretical treatment of how synchronic combination and historical change create inputs in the lexicon, as much as we need one for outputs. Something akin to antimarkedness (if ideophones are maximally unmarked already) to allow for the combinatory creation of marked structures, and antifaithfulness, which would allow forms to infringe/converge on structure already occupied by other forms. The system starts from a state of matrix-like crystallinity of form/meaning, and ends up with a fully mixed, seemingly randomly distributed state. It will be interesting to see what formalists and functionalists, and the folks trying to integrate the two viewpoints, have to say about it (assuming of course you buy the main thrusts). Best to all. Sincerely, Jess Tauber zylogy at aol.com From bresnan at CSLI.STANFORD.EDU Wed Dec 15 21:36:41 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Wed, 15 Dec 1999 13:36:41 -0800 Subject: corpus based research at Stanford In-Reply-To: Your message of Tue, 14 Dec 1999 22:57:22 PST. <199912150655.WAA07293@hypatia.Stanford.EDU> Message-ID: For the record, I should have added Elizabeth Traugott, who does corpus work in both research and teaching, in all of her historical courses in her course on Discourse Analysis. The corrected list is now as follows. People who do corpus-based research in Stanford linguistics and CSLI: Elizabeth Traugott, Tom Wasow, Chris Manning, John Rickford, Beth Levin, Penny Eckert, Eve Clark, Ann Copestake. --representing historical linguistics, discourse analysis, syntax, computational linguistics, sociolinguistics, semantics, language acquisition, and the lexicon... I would have thought it was completely unnecessary to point out these obvious facts about linguistics at Stanford University, but apparently there is quite a lot of misunderstanding or misinformation about our program. This point is, data-driven methodologies in linguistics are a long-standing tradition of Stanford Linguistics, and they are being reinforced both by recent additions to our faculty and new research directions among some of our continuing faculty. At the same time, we probably offer a greater variety of the formal modelling and analytic techniques used in contemporary theoretical and computational linguistics than any other linguistics department. From nrude at ucinet.com Wed Dec 15 23:00:55 1999 From: nrude at ucinet.com (Noel Rude) Date: Wed, 15 Dec 1999 15:00:55 -0800 Subject: optimality in synchrony and diachrony Message-ID: Howdy! Don't really know what the folks are trying to say so maybe I should keep quiet, but then again--am I right that most of us believe that structure codes function, and that if function shapes grammaticalization then this structure-function relationship begins synchronically and persists until bleached out and/or changed? I thought we believed 1) in structure which includes segmantal sequences and prosodies and affixation and word order all of which is subject to various psycho-physical constraints, and which codes 2) simple and complex information 3) according to various principles (propositional structure and semantics and discourse/pragmatics and social stuff ...) which may or may not be related to structure iconically and which may be either "internal" or "external" (whatever this might mean) to Grammar or grannars and 4) which in practice is physically/psychically automated (or routinized) and 5) that anywhere along the line any rules (or "principles and parameters") which CAN be broken WILL be broken (we are not automatons and the rules are not the laws of physics). If we mostly agree on this, how far are we from mopping up the details? Noel From mithun at HUMANITAS.UCSB.EDU Wed Dec 15 23:16:17 1999 From: mithun at HUMANITAS.UCSB.EDU (Marianne Mithun) Date: Wed, 15 Dec 1999 15:16:17 -0800 Subject: corpus based research at Stanford In-Reply-To: <199912150637.WAA07099@hypatia.Stanford.EDU> Message-ID: > I think that more corpus-based research from more perspectives on language > is possible at Stanford than almost anywhere. (Penn being an > exception to the rule... :-) > > J. Actually, there may be considerably more active, corpus-based linguistics going on than many realize. There's been a long tradition among most Santa Barbara linguists of corpus-based analysis, in many cases dating from well before the formation of the department over a decade ago. It ranges from close work on phonetics and prosody through analysis of morphology, syntax, discourse, language change, and language acquisition. The corpora vary in size, but most are quite extensive. They generally consist of spontaneous spoken language, and considerable thought and discussion have gone into issues of content, collection, format, etc. Languages represented include Mohawk, Tuscarora, Cayuga, Seneca, Caddo, Central Pomo, Central Alaskan Yup'ik, Kapampangan, Mandarin, Korean (child language), Japanese (both adult and child language), American English, and others. Corpus-based linguistics is something most of our students simply do as a matter of course in much of their work. Marianne Mithun From macw at CMU.EDU Wed Dec 15 23:59:20 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Wed, 15 Dec 1999 18:59:20 -0500 Subject: corpus based research at Stanford In-Reply-To: Message-ID: Marianne, Joan, and other people interested in corpora, Mariane is right in saying that "there may be considerably more active, corpus-based linguistics going on than many realize." But why is this pivotal aspect of linguistics so low-profile? One way of increasing the profile of corpus-based work is for linguists and their allies to begin to develop more effective ways of sharing corpora, including spoken language corpora. One move in this direction is the new TalkBank project, which NSF (KDI/Linguistics/SBE) has recently funded. (see http://talkbank.org) The goal of TalkBank is to provide computational tools that support corpus-based linguistics and related efforts in about a dozen disciplines devoted to the study of spoken communication. On Dec 4-5, we held a first TalkBank workshop that explored the construction of a database for the study of language used in classrooms and tutorial interactions. http://www.talkbank.org/meetings.html The next TalkBank meeting is devoted to Linguistic Exploration (or what some people might call "field linguistics"). If you are attending LSA this year and are interested in sharing data on spoken communications, please take a look at the program for January 6 at http://www.talkbank.org/exploration.html My guess is that there is a wealth of fantastic spoken language data out there from languages such as "Mohawk, Tuscarora, Cayuga, Seneca, Caddo, Central Pomo, Central Alaskan Yup'ik, Kapampangan, Mandarin, Korean, and Japanese" to name just a few that would greatly benefit the progress of empirically-grounded research across linguistics and allied areas. With our new computational tools we can access these data directly over the Internet (while respecting confidentiality as required). Sounds can be directly linked to transcripts and data can be elaborated with collaborative commentary. TalkBank can provide us all with a way of gaining shared access to these data. In this way, we can also gain a better understanding of the actual data our colleagues are looking at. --Brian MacWhinney From haspelmath at EVA.MPG.DE Thu Dec 16 09:31:47 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Thu, 16 Dec 1999 09:31:47 +0000 Subject: diachronic functionalism Message-ID: Joan Bresnan asks an important question, referring to my earlier posting: >> The usual functionalist explanation for both cross-linguistic patterns and >> language-particular regularities is that they show the effect of >> diachronic change. > >But precisely how? The basic idea is that functional factors apply in performance. Take the devoicing example again: Voiced obstruents are harder to pronounce than voiceless ones, especially in (syllable- or word-)final position. Speakers are thus constantly "tempted" to devoice consonants (partially or completely) under these conditions, and sometimes they give in to the temptation. This may then spread throughout the speech community and result in a language change, such as that from Old High German (e.g. tag 'day') to Modern German (Ta[k] 'day'). Since the phonetic factors are universal, we find languages that have completely lost voiced obstruents (if they ever had any), as well as languages that have devoicing only in final position. We find no languages lacking voiceless obstruents, because there is no possible diachronic change that could give rise to such languages. The source of the universals is thus in performance and (hence) in diachrony, and they are reflected in competence only secondarily. The prediction that this view makes is that impossible languages (e.g. with initial devoicing, or with only voiced obstruents) should NOT be unlearnable, but that they should be diachronically unstable. Unfortunately, learnability experiments are impractical (and perhaps unethical), so the issue is so hard to resolve. Let me cite Joan Bresnan again: >In a nutshell, the analysis of final devoicing in Dutch predicts the >existence of the well known typological asymmetry in obstruent voicing >across languages. The same applies to the diachronic-functional view, except that we would talk in terms of "explanation" rather than "analysis". (Of course, in both approaches the prediction is not literal, because one would not have adopted this particular analysis/explanation for Dutch without the knowledge of the typological asymmetry.) >It is because the OT constraints are universal and not >language-particular rules or parameter settings, that this deep >connection between typology and language-internal "grammar" is >possible. You literally cannot do the (real) OT grammar of any >particular language without doing typology. This would be an advantage only if it turned out that that's the right way of "doing grammar". And diachronic functionalists (such as Matthew Dryer, Fritz Newmeyer, Joan Bybee, Joe Greenberg, Bill Croft) would reject the view that anything is gained by subsuming language-particular description and explanation of universal tendencies under a single theory. (Or in Newmeyer's case, he already seems to be convinced that this is impossible.) Martin P.S. Joan Bresnan seems to use the term "evolutionary" where I use "diachronic". I think this terminological usage should be avoided, because "evolutionary" is ambiguous between "phylogenetic (Darwinian) evolution" and "glossogenetic evolution" (i.e. diachronic change"). -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From john at RESEARCH.HAIFA.AC.IL Thu Dec 16 08:38:00 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Thu, 16 Dec 1999 10:38:00 +0200 Subject: Translation databases? Message-ID: This is terrific, Brian, but what I would really like to know is: Are there any TRANSLATION databases? That is, databases which have both originals and translations in other languages? Allowing for word searches (construction searches would be even better, but this is too much to hope for)? In a variety of genetically unrelated and geographically separated languages? If we want to do comparisons of functions of different structures, or meanings of different word, in different languages, translations are really helpful. For those of us who are seriously interested in language universals, translation data, like nothing else, force us to come to grips directly with differences between languages; we cannot, for example, so well blather about the `universal' or `cognitive' functions of voice alternations based exclusively on English data when confronted with translation data clearly showing that other languages use voice alternations in extremely different ways. I have applied for grants to develop such a translation database twice and been rejected both times. Wally Chafe tells me that the Pear Stories have never been rendered into a usable form (and in any case they are quite short). I have done a number of studies using the Bible, because at least there are a lot of texts with interlinear glosses in both languages, and there are concordances of particular words--but there aren't so many languages with such data, and Bible translations tend not to be into the most naturalistic language, if you know what I mean. There are of course many texts with interlinear glosses in, e.g. Native American or Australian languages and English, but each of these is in only two languages, and there's no concordance to help searches for individual words (in addition to difficulty in accessing native speakers for help). So, in order to get comparison between more than just two languages, I have been forced to do things by hand. I am presently doing a study of the comparative meanings of speech act verbs in Hebrew, English, Japanese, and Spanish by using novels and short stories by Gabriel Garcia Marquez, A.B. Yehoshua, and Banana Yoshimoto, with translations of each into each of the other languages, and let me tell you, it is pretty slow going. I have to search for each occurrence of a given word, then search to see how it is translated into each of the other three languages, without the use of concordances or interlinear glosses (it goes without saying that I read some of these languages more quickly that others). If I have only like 5 occurrences of a given word, the translation data often looks kind of chaotic, but if I can get 30 or 40 tokens, very clear patterns always emerge, but unless the word is pretty common it simply takes too long to get this number of tokens. I can do it, and the results are very interesting, but it takes a long time, and I more or less have to study only words which are pretty common (e.g., I would love to do a study of emotion words like 'angry', 'sad', etc. to see how they're translated, but it would take an enormous amount of time to locate enough tokens to use--I've tried). Any ideas or data sources which might speed things up? Hopefully, John From jlmendi at POSTA.UNIZAR.ES Thu Dec 16 09:14:49 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Thu, 16 Dec 1999 10:14:49 +0100 Subject: Myhill on Chomsky In-Reply-To: <38568478.C9732208@chass.utoronto.ca> Message-ID: This is my last message about this subject. Peter A. Reich wrote: >I'm sure God is cited more than Chomsky (I will not engage in a >discussion as to whether God is Living), and probably Bill Gates is >mentioned more than Chomsky as well in the popular media. And Clinton. >So what? It is perfectly respectable in Canada, at least, to be an >Atheist, and to prefer Linux or Macintosh. And, in the US, to be a >Republican. Eat shit; 50 billion insects can't be wrong.--Peter Reich Of course, when I used a 'quantitative' argument I was not trying to demonstrate that Chomsky's ideas about language and grammar are correct. This is an empirical matter, a scientific matter we are not dealing with here. In fact, someone here has said that the use of, say, 'qualitative' arguments would be a waste of time. I was only trying to show Myhill that his vision of Chomsky as a Pope indoctrinating an international herd of hundreds (or thousands) of icapable of thinking by themselves linguists and scientists is simply not solvent. That it is just an insult. Best Regards, Jose-Luis. From nc206 at HERMES.CAM.AC.UK Thu Dec 16 09:55:48 1999 From: nc206 at HERMES.CAM.AC.UK (N. Chipere) Date: Thu, 16 Dec 1999 09:55:48 +0000 Subject: use of corpora based In-Reply-To: <350474.3154273160@agate.psy.cmu.edu> Message-ID: On Wed, 15 Dec 1999, Brian MacWhinney wrote: > Marianne, Joan, and other people interested in corpora, > > Mariane is right in saying that "there may be considerably more active, > corpus-based linguistics going on than many realize." But why is this > pivotal aspect of linguistics so low-profile? > One way of increasing the profile of corpus-based work is for linguists and > their allies to begin to develop more effective ways of sharing corpora, > including spoken language corpora. One move in this direction is the new > TalkBank project, which NSF (KDI/Linguistics/SBE) has recently funded. (see > http://talkbank.org) Yet another way is, as far as possible, to base discussions of grammar on corpora rather than to rely purely on intuition. - Ngoni From david_tuggy at SIL.ORG Thu Dec 16 15:55:51 1999 From: david_tuggy at SIL.ORG (david_tuggy at SIL.ORG) Date: Thu, 16 Dec 1999 10:55:51 -0500 Subject: The rest of us on Chomsky Message-ID: In the interests of Chomsky-discussion rather than Chomsky-bashing: It seems clear that a number of people on this list think at least some of Chomsky's ideas have been problematic for linguistics rather than helpful. I'd enjoy hearing from any who'd like to respond to questions of the following sort. What do you see as *the* most problematic/pernicious/unhelpful of Chomsky's theoretical positions / basic analytical stances / posited grammatical mechanisms / etc. ? What's wrong with it? Why is it so bad? What's right with it? Why has it convinced so many intelligent people? How does your favorite alternative (functionalist or not) avoid what was wrong with it and keep what was right? --David Tuggy From MILLSCR at UCMAIL.UC.EDU Thu Dec 16 17:03:40 1999 From: MILLSCR at UCMAIL.UC.EDU (Mills, Carl (MILLSCR)) Date: Thu, 16 Dec 1999 12:03:40 -0500 Subject: Chomsky Message-ID: Actually, many of us who went to LSA in the late 1970s and '80s discovered that the "genuine criticism" our papers received, most of it from Chomskyans, was so ill-informed and narrow minded that we stopped going. So I don't blame Chomsky for not going. On the other hand, it is not surprising that Chomsky is addressing the MLA. Anyone who reads stuff from the sciences, e.g. biology, chemiostry, anthropology (the scientific, i.e. physical or biological, parts of it) is apt to discern that linguistics since the Chomskyan revolution (pick whichever one of them you want) has becoming less scientific and more humanistic over the years. Personally, I think that pre-Chomskyan structuralism was a sort of natural history, but what we see mostly today is on a par with literary theory (sic). Incidentally, my own work on how the sciences and a few other disciplines display their results indicates that there are a number of disciplines between the sciences and the humanities (psychology, sociology) that are not unlike linguistics in being unable or unwilling to decide whether they are sciences or humanities. Carl -----Original Message----- From: A R You know, Funknetters, I don't know that I'd call myself either a formalist or a functionalist. But one of the most damning things I know about Chomsky is that he *never* goes to LSA, the one place where he might face genuine criticism. A friend of mine, a formalist, used to argue that he simply never went to academic conferences. This year, I discover that this is clearly untrue. Chomsky is going to talk at the MLA, to those who do literature. It's only his own that he refuses to talk to at conferences. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From bresnan at CSLI.STANFORD.EDU Thu Dec 16 18:07:49 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 16 Dec 1999 10:07:49 -0800 Subject: diachronic functionalism In-Reply-To: Your message of Thu, 16 Dec 1999 09:31:47 GMT. <3858B17F.1D87F188@eva.mpg.de> Message-ID: Martin Haspelmath's response to my question ("But precisely how?") makes it clear to me that there might not be any real incompatibility between our views, once they are spelled out more precisely and terminological differences are clarified. >From his initial statement-- > The basic idea is that functional factors apply in performance. --I infer that he thinks that the differences in our views have something to do with "performance" vs. "competence". But I don't recall mentioning that chestnut in any of my postings. [In fact, I believe I once wrote that the competence/performance distinction has served as a convenient way for some linguists to insulate their theories from empirical disconfirmation.] Is Martin embracing the classic generative view of competence grammar as part of his historical explanation or just (incorrectly) presupposing that I own that view or that anyone who does OT must subscribe to that particular view articulated by Chomsky in 1965? It certainly isn't true. Let's simply drop it as a false barrier to communication. To explain the typological and language-internal patterns of voiceless obstruents, Martin supposes that > Speakers are thus constantly "tempted" to devoice consonants (partially > or completely) under these conditions, and sometimes they give in to the > temptation. Is the "constant" presence of this "phonetic temptation" any different the OT hypothesis that the contextual markedness of voiced obstruents is a universal constraint is present in every individual by virtue of the human articulatory and perceptual systems? Martin suggests that once speakers "give in to the temptation" to devoice, devoicing may spread throughout the speech community and result in language change. Yes, but when we try to model precisely what "giving in to the temptation" means, we may come up with the idea that the "phonetic temptation" becomes more dominant compared to the temptation to preserve constrasts. OT provides some very precise and nice ways to model this. One I like very much is the Boersma-Hayes model of probabilistically varying constraint rankings that I referred to recently. (See their recent ROA paper.) Constraints have continuous (not scalar) ranking values which vary according to a normal distribution. When two conflicting constraints very close in their ranking are applied to the same form, either one may dominate, yielding a certain frequency of variation in the forms produced. I said nothing in my postings about how variation of this kind spreads through populations, and that is certainly where diachronic and sociolinguistic explanations come in. We must look at all sorts of "external" issues such as the relative influence and power of social groups and networks of individuals. One way such factors can exert their influence on individual linguistic patterns, in OT terms, is by modifying the rankings of constraints. In some recent work I have done on negative auxiliary inversion in several English dialects including Scots and Hiberno-English, I show how through a sociolinguistically determined ranking of this kind, variations in both the syntactic distribution and the semantic scope of negation can be explained. One striking syntactic difference is this: Scots English: Amn't I your friend? *I amn't your friend? Hiberno English: Amn't I your friend? I amn't your friend? Note that OT does not determine the ranking of constraints, but only an explicit space of possible variation within which changes and their implications and interactions may be studied in precise detail. [Nigel Vincent has begun to apply this kind of model to historical change from Latin to Romance: see his recent posting to the LFG archive: http://www-lfg.stanford.edu/lfg/.] --Unfortunately, I must interrupt this conversation and go now. Best wishes to you functionalists and here's hoping for a productive and interactive new millenium! Joan > > P.S. Joan Bresnan seems to use the term "evolutionary" where I use > "diachronic". I think this terminological usage should be avoided, > because "evolutionary" is ambiguous between "phylogenetic (Darwinian) > evolution" and "glossogenetic evolution" (i.e. diachronic change"). I completely agree. Sorry, I just unthinkingly picked up some words from Matthew's msg, and realized later that they weren't what I really intended. From bresnan at CSLI.STANFORD.EDU Thu Dec 16 18:12:04 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Thu, 16 Dec 1999 10:12:04 -0800 Subject: diachronic functionalism In-Reply-To: Your message of Thu, 16 Dec 1999 10:07:49 PST. <199912161806.KAA12891@hypatia.Stanford.EDU> Message-ID: correction: > Constraints have > continuous (not scalar) ranking values .. I meant "continuous (scalar) ranking values". J.B. From chafe at HUMANITAS.UCSB.EDU Thu Dec 16 18:19:23 1999 From: chafe at HUMANITAS.UCSB.EDU (Wallace Chafe) Date: Thu, 16 Dec 1999 10:19:23 -0800 Subject: Corpora Message-ID: Evidently Joan wanted to make the valid point that Stanford is not just MIT West, but then it degenerated into "my department has more corpus linguists than your department" (unless you're Penn) (and I'm having a little trouble remembering who they are). More to the point might be a discussion of what corpus research involves, what kinds of corpora there are, how they can best be exploited, etc. Brian and others are providing an important service in this regard with the Talkbank project. As I see it, we're talking about an alternative to the popular kinds of data-gathering that have involved inventing sentences in isolation and asking people whether they "get" them, or measuring the reaction times of college sophomores who see them written on computer screens. The alternative is to examine how people actually use language, a process that necessarily involves confronting more than single sentences. In that sense it's part of what has come to be called the study of "discourse", which of course can be conducted in many different ways. It's worth noting that some people have been examining such data for a long time: one thinks, for example, of many language acquisition studies, of the analysis of conversation (from various points of view), and of the recording and analysis of "texts" collected by those who have been studying lesser known languages. This last kind of corpus work has been going on for well over a hundred years. It's worth noting, too, that this distinction between constructed language and what I like to call "real" language ("natural language" has been coopted with a different meaning) is orthogonal to the formalist-functionalist dichotomy, at least in the sense that while many functionalists do work with corpora, many do not. It might be worth discussing the problems that arise from the supposedly accidental nature of corpora, and the lack of the control and replicability that are so dear to the hearts of psychologists. One might actually find some significance, for example, in the fact that people rarely use a construction one might think easy to invent. And of course the problem tends to diminish with very large corpora. But very large corpora may introduce a problem of their own. Some of you may remember Zellig Harris's book Methods in Structural Linguistics, where he suggested we could get around the vexing problem of meaning by examining the distribution of linguistic forms in very large corpora. Machines to do that weren't available at the time (1950), but now they are, and it looks to me as if some people are doing what Harris had in mind, though so far as I know they don't refer to him. It makes me uncomfortable because I think it's more rewarding in the long run to confront semantics head-on, not trying to avoid it with big corpora and machines. Just one last reservation. Corpora make it easy to count things and come up with interesting findings regarding the frequency of this or that. But knowing exactly what you're counting may not be such a simple matter, and it's easy to come up with "operational definitions" that turn out in the end to be spurious. What I'm trying to say is that there's much of importance to learn from examining real language, but it shouldn't seduce us into thinking we can just crank out analyses mechanically. Understanding the nature of language is always going to require the intervention of perceptive human minds. Wally Chafe From MILLSCR at UCMAIL.UC.EDU Thu Dec 16 18:33:47 1999 From: MILLSCR at UCMAIL.UC.EDU (Mills, Carl (MILLSCR)) Date: Thu, 16 Dec 1999 13:33:47 -0500 Subject: Counting things Message-ID: Having spent three decades counting things, I want to second Wallace Chafe's "last reservation." Of course, there is always going to be a need for the intervention of human minds. But those minds tend to work best on inputs that do not originate within themselves. There is a real world, and it is, in some important sense, out there. Corpus linguistics offers the possibility of starting with something other than "Can you say S?" or worse "Can I say S?"and I don't think that is a small achievement. Carl -----Original Message----- From: Wallace Chafe To: FUNKNET at LISTSERV.RICE.EDU Just one last reservation. Corpora make it easy to count things and come up with interesting findings regarding the frequency of this or that. But knowing exactly what you're counting may not be such a simple matter, and it's easy to come up with "operational definitions" that turn out in the end to be spurious. What I'm trying to say is that there's much of importance to learn from examining real language, but it shouldn't seduce us into thinking we can just crank out analyses mechanically. Understanding the nature of language is always going to require the intervention of perceptive human minds. Wally Chafe -------------- next part -------------- An HTML attachment was scrubbed... URL: From wcmann at JUNO.COM Thu Dec 16 18:31:29 1999 From: wcmann at JUNO.COM (William Mann) Date: Thu, 16 Dec 1999 13:31:29 -0500 Subject: Translation databases? Message-ID: It seems timely to point out that there is a very active Corpora email list, and that discussions of multilingual databases appear there regularly. Participation is very international, at perhaps 5 messages per day. The control is through Majordomo at iub.no and a message body sent to it, containing subscribe corpora <> should begin to get you in. It also responds to the message body help with help. I don't know whether they have searchable archives. Happy processes. Bill Mann From jlmendi at POSTA.UNIZAR.ES Thu Dec 16 19:06:49 1999 From: jlmendi at POSTA.UNIZAR.ES (Jose-Luis Mendivil Giro) Date: Thu, 16 Dec 1999 20:06:49 +0100 Subject: The rest of us on Chomsky In-Reply-To: <9912169453.AA945364013@router-8.camnet.com> Message-ID: At 10:55 -0500 16/12/99, david_tuggy at SIL.ORG wrote: > In the interests of Chomsky-discussion rather than Chomsky-bashing: > > It seems clear that a number of people on this list think at least > some of Chomsky's ideas have been problematic for linguistics rather > than helpful. > > I'd enjoy hearing from any who'd like to respond to questions of the > following sort. > > What do you see as *the* most problematic/pernicious/unhelpful of > Chomsky's theoretical positions / basic analytical stances / posited > grammatical mechanisms / etc. ? > > What's wrong with it? Why is it so bad? Please forgive me for using the words of a member of this list, but I could not state better than him: "An undercurrent of hostility to generative grammar arises because many who identify themselves as linguists believe that Chomsky wishes to define them out of the field of linguistics" (Newmeyer, 1983: 138). Of course, this is not the only problematic affair of Chomsky's theories and attitudes at all, but I think it explains quite well some points of view showed here. In other words, it seems to me that the major problem is related, first, to the fact that Chomsky's program has not been properly understaken and, second, to the fact that Chomsky and, specially, Chomskyans tend to ignore what is not Chomskyan. There is mainly a problem of communication. Best regards, Jose-Luis. From dgolumbi at PANIX.COM Thu Dec 16 19:21:54 1999 From: dgolumbi at PANIX.COM (David Golumbia) Date: Thu, 16 Dec 1999 14:21:54 -0500 Subject: the rest of chomsky Message-ID: during this intense and often enlightening discussion, this sentence of recent vintage, by a contributor whose first language appears to me not to be English, strikes me with special force: > Chomsky's program has not been properly understaken rooting out the "problems" in Chomsky's program could (and should) be the object of even more informed comment than it has already (botha, r[andy] harris, r[oy] harris, huck/goldsmith, matthews, mccawley, newmeyer) - but you could take less productive starting points than this fun gem, which to start with, is*? a specimen of English? -- dgolumbi at panix.com David Golumbia From kfeld at CITRUS.UCR.EDU Thu Dec 16 20:48:08 1999 From: kfeld at CITRUS.UCR.EDU (David B. Kronenfeld) Date: Thu, 16 Dec 1999 12:48:08 -0800 Subject: Corpora In-Reply-To: Message-ID: Out of curiosity, how does the corpora vs. invented speech distinction relate to the old distinction between the way people normally talk vs. the way the talk which you ask them to slow down and be careful (or ask them what they meant to say--or some such)--particularly re how grammatical relations are approached ? And-- "amen" on the Z. Harris comment and what follows it. David Kronenfeld At 10:19 AM 12/16/99 -0800, you wrote: >More to the point might be a discussion of what corpus research involves, >what kinds of corpora there are, how they can best be exploited, etc. >Brian and others are providing an important service in this regard with >the Talkbank project. > >As I see it, we're talking about an alternative to the popular kinds of >data-gathering that have involved inventing sentences in isolation and >asking people whether they "get" them, or measuring the reaction times of >college sophomores who see them written on computer screens. The >alternative is to examine how people actually use language, a process that >necessarily involves confronting more than single sentences. In that >sense it's part of what has come to be called the study of "discourse", >which of course can be conducted in many different ways. It's worth >noting that some people have been examining such data for a long time: >one thinks, for example, of many language acquisition studies, of the >analysis of conversation (from various points of view), and of the >recording and analysis of "texts" collected by those who have been >studying lesser known languages. This last kind of corpus work has been >going on for well over a hundred years. > >It's worth noting, too, that this distinction between constructed language >and what I like to call "real" language ("natural language" has been >coopted with a different meaning) is orthogonal to the >formalist-functionalist dichotomy, at least in the sense that while many >functionalists do work with corpora, many do not. > >It might be worth discussing the problems that arise from the supposedly >accidental nature of corpora, and the lack of the control and >replicability that are so dear to the hearts of psychologists. One might >actually find some significance, for example, in the fact that people >rarely use a construction one might think easy to invent. And of course >the problem tends to diminish with very large corpora. > >But very large corpora may introduce a problem of their own. Some of you >may remember Zellig Harris's book Methods in Structural Linguistics, where >he suggested we could get around the vexing problem of meaning by >examining the distribution of linguistic forms in very large corpora. >Machines to do that weren't available at the time (1950), but now they >are, and it looks to me as if some people are doing what Harris had in >mind, though so far as I know they don't refer to him. It makes me >uncomfortable because I think it's more rewarding in the long run to >confront semantics head-on, not trying to avoid it with big corpora and >machines. > >Just one last reservation. Corpora make it easy to count things and come >up with interesting findings regarding the frequency of this or that. But >knowing exactly what you're counting may not be such a simple matter, and >it's easy to come up with "operational definitions" that turn out in the >end to be spurious. What I'm trying to say is that there's much of >importance to learn from examining real language, but it shouldn't seduce >us into thinking we can just crank out analyses mechanically. >Understanding the nature of language is always going to require the >intervention of perceptive human minds. > >Wally Chafe > David B. Kronenfeld Phone Office 909/787-4340 Department of Anthropology Message 909/787-5524 University of California Fax 909/787-5409 Riverside, CA 92521 email kfeld at citrus.ucr.edu http://www.ucr.edu/CHSS/depts/anthro/home.htm http://pweb.netcom.com/~fanti/david.html From fjn at U.WASHINGTON.EDU Thu Dec 16 22:05:38 1999 From: fjn at U.WASHINGTON.EDU (Frederick Newmeyer) Date: Thu, 16 Dec 1999 14:05:38 -0800 Subject: OT and functional explanation Message-ID: Joan Bresnan's clarification of her position with respect to Martin Haspelmath's was very helpful, but there may be differences that remain. I've been able to discern a number of views out there with respect to the 'functionality' of OT constraints. Here are three of them: (A) Constraints are universal, and may or may not be functionally motivated (e.g. Jane Grimshaw's LI paper, where the issue does not come up). (B) Constraints are universal and functionally motivated (Martin Haspelmath's forthcoming Z fuer S paper). (C) Constraints are universal and are the actual functional motivations themselves. That is how I interpret the following remark from Joan's December 10 posting: 'Couldn't conflicting constraints such as iconicity and economy be universal, but prioritized differently across different domains and different languages?' Martin and Joan: does that put the finger on the different ways that you view the constraints? I'm an utter novice at OT, so the issue I will now raise could be based on my own ignorance of that framework. If so, no doubt several of you will tell me so. But I have trouble seeing how either (B) or (C) could be made to work. My primary qualm comes from the fact that the forces (functional or otherwise) that bring a construction into a language are not necessarily the same ones that keep it there. An example: English is a primarily head-initial right-branching language. There are good functional reasons for a language to be 'consistent' in this regard (Dryer, Hawkins, et al.). So a constraint for English like HEAD-LEFT (or BRANCH-RIGHT) most certainly has a functional motivation. Now there are, in fact, left-branching constructions in English: the GEN-N construction is the best-known ('Mary's mother's uncle's lawyer'). A Grimshavian OT analysis could license this by a constraint called 'GEN-N' (or whatever) that would dominate HEAD-LEFT (the details are more complicated since English also has N-GEN, but we can put them aside). Now how might positions (B) or (C) handle this fact about English? I don't know. As I understand the history of English, 1000+ years ago it was largely left-branching, so the GEN-N construction was indeed functionally motivated at one time. For whatever reason (and many have been suggested), English has become largely right-branching. GEN-N survives as a conventionalized relic of the days when it had a real functional motivation. So, what does a functionally-oriented OT analysis do about this? Surely it would not want to say that the constraint licensing this left-branching structure is functionally motivated by parsing pressure, since that is manifestly false. In that respect this construction is counter-functional. Would the program necessitate finding some other functional explanation for its existence? Is 'convention' admissible as a category of functional motivation? (That would raise a host of problems, no?) I'm raising a number of questions, but don't mean to seem dogmatic about the answers. Languages are filled with constructions that arose in the course of history to respond to some functional pressure, but, as the language as a whole changes, cease to be very good responses to that original pressure. Such facts seem challenging to any theory (like versions of OT that have been suggested) in which the sentences of a language are a product of constraints that must be functionally motivated, or are the actual functional motivations themselves. --fritz newmeyer From macw at CMU.EDU Fri Dec 17 00:31:04 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Thu, 16 Dec 1999 19:31:04 -0500 Subject: Chomsky and deep structure Message-ID: David and FunkNet, Not to distract from the really fascinating issues being discussed in the functionalism-and-OT thread and the corpus linguistics thread, I thought I would take a shot at a quick response to David Tuggy's reasonable request for a statement regarding *the* most problematic of Chomsky's positions. As far as I can see, the core problem with Chomsky's generative grammar is the commitment to deep structure. All of the other problems derive from that commitment. "Underlying form" is a reflex of deep structure in that it allows us to map the many onto the few, thereby appearing to achieve "linguistically significant generalizations" along with abstractions that can be used for further abstract theory building. "Competence-performance" and the "ideal native speaker" are constructs that are required to support theories of deep structure. Without deep structure, the notion of universal grammar is hard to imagine. I realize that the importance of deep structure has waxed and waned in the various versions of the theory. But I see no evidence that, the core conceptual importance of deep or abstract structure has ever disappeared. What is wrong with deep structure? The problems have to do with the (1) the nature of empirical evidence and (2) the structure of human thought. People don't speak deep structures and it is essentially impossible to find them in natural interactions or experiments. Of course, one can argue that physicists rely on a particle and force-based deep structure in quantum dynamics. This is true, but the hypothesization of particles has been supported at each point by precise empirical measurements matched with detailed mathematical accounts that are remarkably precise. No such accuracy of measurement supports the constructs of linguistic deep structure. Psychologists like Sigmund Freud have shown us how dangerous it can be to construct theories of deep structure that cannot be clearly pegged to empirical tests. I am certainly not opposed to abstract theories in principle, but I am only willing to work with such theories when the individual components of the theory have demonstrable empirical grounding. Many of the constraints of certain versions of OT seem to fulfill this requirement, for example. My second reservation about deep structure has to do with what we know about other areas of perception and cognition. It is certainly true that the construction of a visual percept relies on low-level feature detection. In this sense, we could think of vision as building up constructs from a deep structure base. But this is not what Chomsky has in mind with the non-perceptually-grounded categories of universal grammar. In principle, one could imagine that a cognitive system could have arisen through evolution that was structured in this way. But the fact that language is a relatively recent event in primate evolution makes me wonder how something so abstract and ungrounded as deep structure could have arisen in such a short evolutionary time span. Calling language a sprandrel (ala Chomsky and Gould) is fine, but sprandrels are simple emergent structures and deep structure is not. So where would deep structure come from? One can argue that deep structure is just a hypothesis that should be pursued with the same tenacity with which we pursue any scientific hypothesis. I agree with that. But, just as I would not place all my funding resources on one approach to fusion research, I don't think that all of linguistics should be dedicated to pursuing theories grounded on deep structure. In reality, linguistics is far more diverse than the general public realizes. Some high profile media events have made it seem monolithic, but this is misleading. I think that Chomskyan approaches will never disappear, since the hypothesis of deep structure will always be tenable and worth exploring. Once we really understand how language is processed and stored by the brain, we may reinvent a empirically-grounded type of deep structure. But it will have only a vague family resemblance to the deep structures of "Cartesian linguistics" or Chomskyan linguistics. What is right about Chomsky? I think Chomsky's great contributions are his emphasis on generativity and mechanism, his treatment of language as a component of human cognition, his ideas about language and creativity, and his stimulating collaborations with George Miller and Eric Lenneberg. So, David, that's how I see it. --Brian MacWhinney From ward at MERLE.ACNS.NWU.EDU Fri Dec 17 01:08:40 1999 From: ward at MERLE.ACNS.NWU.EDU (Gregory L Ward) Date: Thu, 16 Dec 1999 19:08:40 -0600 Subject: LSA (bashing) Message-ID: Carl Mills writes: > Actually, many of us who went to LSA in the late 1970s and '80s > discovered that the "genuine criticism" our papers received, most of > it from Chomskyans, was so ill-informed and narrow minded that we > stopped going. So I don't blame Chomsky for not going. A R writes: > You know, Funknetters, I don't know that I'd call myself either > a formalist or a functionalist. But one of the most damning things > I know about Chomsky is that he *never* goes to LSA, the one place > where he might face genuine criticism. A friend of mine, a formalist, > used to argue that he simply never went to academic conferences. To sum up: Chomsky doesn't go to the LSA because he'd face "genuine criticism", and Mills doesn't go to the LSA because he faced 'genuine criticism' ("in the late 1970s and '80s"). So the LSA is a conference where the work of functionalists and formalists alike is criticized, presupposing the presence of both. Sounds like my kind of conference... ;-) Gregory _____ Gregory Ward Professor and Chair Department of Linguistics Northwestern University 2016 Sheridan Road Evanston IL 60208-4090 e-mail: gw at nwu.edu tel: 847-491-8055 fax: 847-491-3770 www: http://www.ling.nwu.edu/~ward From haspelmath at EVA.MPG.DE Fri Dec 17 10:15:41 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Fri, 17 Dec 1999 10:15:41 +0000 Subject: OT and diachronic functionalism Message-ID: I'm somewhat surprised to see that Joan Bresnan dissociates herself from the competence/performance terminology. OK, it probably has a lot of baggage attached to it that I may not be aware of, but my impression is that we all pretty much agree that there is a necessary conceptual distinction between language use and language structure, or speech and grammar, or more generally between processing and storage, or cognitive events and cognitive patterns. I used "competence/performance" as a convenient shorthand, following a well-known functionalists's (John Hawkins's) usage. True, the competence/performance distinction has often been attacked by functionalists, but I believe that this attack is misguided. The problem with the Chomskyan approach is not that it draws this distinction, but that all the ways in which performance influences competence are systematically ignored. (See also my review article on Fritz Newmeyer's 1998 book, soon to appear in Lingua). So in my view, three tasks of linguistics are: (i) describing competence, (ii) describing performance, and (iii) explaining competence, which must apparently often make reference to (ii). I understand the Chomskyan approach as claiming that (iii) is irrelevant, (ii) is moderately interesting, and (i) is the core task of linguists (especially concerning the innate aspects of competence). Joan Bresnan's OT functionalism draws the lines quite differently, it seems, and I haven't understood yet, how. So far my impression of OT had been that the Chomskyan program is virtually unchanged: The main goal is a maximally elegant description (or "characterization", the euphemism that is often used instead) of the competence grammar, but the archtecture of the formal grammar is different: The machinery makes use of abstract entities called "constraints" which often happen to correspond closely to the actual constraints on language use that speakers grapple with while speaking. (I call these "user constraints" in my ZS paper.) Joan Bresnan asks: >> Speakers are thus constantly "tempted" to devoice consonants (partially >> or completely) under these conditions, and sometimes they give in to the >> temptation. >Is the "constant" presence of this "phonetic temptation" any different from >the OT hypothesis that the contextual markedness of voiced obstruents >is a universal constraint that is present in every individual by virtue of >the human articulatory and perceptual systems? Until recently I thought yes, crucially, in that the OT constraint is part of the competence grammar (and is perhaps even innate as part of UG), whereas the "user constraint" is a constraint on the articulatory system and has nothing to do with the conventional language system (= competence). (But now Joan Bresnan's contributions to this discussion have confused me.) Joan continues: >Martin suggests that once speakers "give in to the temptation" to >devoice, devoicing may spread throughout the speech community and >result in language change. Yes, but when we try to model precisely >what "giving in to the temptation" means, we may come up with the idea >that the "phonetic temptation" becomes more dominant compared to the >temptation to preserve constrasts. Sure, intuitively that's what is going on, and linguists have been talking in such terms for centuries (though not as precisely as in OT, of course). But how can a "phonetic temptation" become more dominant than another temptation? After all, these temptations or constraints are universal ? identical for all speakers and all languages! Clearly, what happens is that the linguistic *conventions* change: It suddenly becomes socially acceptable to "give in to the temptation". Or in other words, the grammar has changed, and the functional motivation has left its mark on the competence system. Since the grammar is so clearly shaped by the performance constraint, it is possible (and elegant) to describe the grammar in terms of a competence constraint that mimics the performance constraint. My point in my ZS paper is that such descriptions are clearly better than arbitrary descriptions (so I welcome OT), but that we still need the link to the performance constraints if we want to explain grammar. Fritz Newmeyer asks: >(A) Constraints are universal, and may or may not be functionally >motivated (e.g. Jane Grimshaw's LI paper, where the issue does not come >up). >(B) Constraints are universal and functionally motivated (Martin >Haspelmath's forthcoming Z fuer S paper). >(C) Constraints are universal and are the actual functional >motivations themselves. That is how I interpret the following remark from >Joan's December 10 posting: 'Couldn't conflicting constraints such as >iconicity and economy be universal, but prioritized differently across >different domains and different languages?' >Martin and Joan: does that put the finger on the different ways that you >view the constraints? I thought so initially, but I haven't understood Joan yet. Fritz mentions some word order facts as difficult for an OT-functionalist position like Joan Bresnan's, and again Joan might reply that her general approach is only in its infancy, so it's too early to judge whether it might not work after all. But I agree with Fritz's general skepticism: Languages are so full of patterns that are there because of speakers' conservatism, not because they allow speakers to give in to some temptation. Fritz is absolutely right in pointing out that if one wanted to make these structures follow from an OT-functionalist grammar, one would have to introduce a functional constraint such as "conventionality". Another consideration that makes me skeptical about OT is that the descriptions rely so heavily on non-functional, language specific constraints such as alignment. HEAD-LEFT might be seen as a constraint in English grammar, but it surely cannot be functionally motivated, because HEAD-RIGHT is just as good from the language user's point of view. As John Hawkins has argued convincingly, the real motivation for the Greenbergian word order correlations is Early Immediate Constituents, and it seems difficult to integrate this into an OT-style grammar. But let's wait and see. Martin -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From david_tuggy at SIL.ORG Thu Dec 16 22:28:33 1999 From: david_tuggy at SIL.ORG (david_tuggy at SIL.ORG) Date: Thu, 16 Dec 1999 17:28:33 -0500 Subject: OT and functional explanation Message-ID: Fritz Newmeyer asks, **** Is 'convention' admissible as a category of functional motivation? (That would raise a host of problems, no?) **** I for one think it has to be, or else we can not begin to think of functionalist linguistics as covering anywhere near all of language. (It also seems prima facie obvious that doing things the same way you have done them before/heard others doing them is functionally advantageous!) --David Tuggy From wilcox at UNM.EDU Fri Dec 17 17:10:45 1999 From: wilcox at UNM.EDU (Sherman Wilcox) Date: Fri, 17 Dec 1999 10:10:45 -0700 Subject: Linguistic motivation Message-ID: A request for suggestions: I'll be teaching a new course this spring on linguistic motivation and I'm putting together the syllabus and reading list. I have a course outline, topics, and several readings in mind. I'd appreciate, though, any suggestions that list members might have regarding articles or topics you would consider important to include. I'll be happy to summarize and post the responses back to this list (I am posting this message on both cogling and funknet). Sherman Wilcox University of New Mexico From W.Schulze at LRZ.UNI-MUENCHEN.DE Fri Dec 17 18:58:15 1999 From: W.Schulze at LRZ.UNI-MUENCHEN.DE (Wolfgang Schulze) Date: Fri, 17 Dec 1999 19:58:15 +0100 Subject: Function and convention Message-ID: > David Tuggy quotes Fritz Newmeyer: > > **** > Is 'convention' admissible as a category of functional motivation? (That > would raise a host of problems, no?) > **** Perhaps we should replace 'convention' by 'routinization' (based on 'ritualization') in order to avoid the intentional connotation, the term 'convention' seems to imply. I think that the grammar of a language has to be explained on the basis of the acquired cognitive and (cognition based) communicative practise of an individual integrated into a collective. This practise is dominated by massive hypotheses about the self-attachment to a collective; it RE-presents a strongly ritualized but construing interaction of the individual with environmental or world stimuli which corresponds to the habitus of a collective and which takes place in form of the tacit (poiematic) and/or articulate (pragmatic) activation of an acquired (and tradi-tional) knowledge system. Linguistic practise can be thought to represent the individual reaction to a collective communicative and cognitive standard which itself is predominantly historical in nature. By this is meant that the linguistic knowledge system of an individual and its instantiation in a 'communicative community' always reflects strategies of linguistic adaptation that have been functionalized long before the individual has acquired a given system. Hence, it can be argued that language as a ?metaphysical? phenomenon owns strong anachronistic features: It hardly ever meets the immediate synchronic needs of information processing and communication. It follows that functional and semantic aspects of language architectures are mainly to be explained in a diachronic perspective, though the potential to adopt newly established communicative and cognitive routines plays an important role in the dynamic potential of language systems (what I call 'Pragmatic Intervention' (PI)). The assumption of an anachronistic ontology of language systems has an important consequence for linguistic explanation: Contrary to some other cognitive approaches, the framework underlying these assumptions ('Grammar of Scenes and Scenarios' (GSS) does not establish a direct synchronic relationship between language systems and cognition. Language systems and cognitive activities are thought to be structurally coupled on the basis of a mainly diachronic relationship. From this it follows that routinization (or if you want 'conventionalization') plays a crucial role not only in the dynamics of linguistic functions and in the functional architecture on language systems, but also in the motivation of these functions emerging from the cognition<>communication interface. Wolfgang ***************************** Prof. Dr. Wolfgang Schulze Institut fuer Allgemeine und Indogermanische Sprachwissenschaft Ludwig-Maximilians-Universitaet M?nchen Geschwister-Scholl-Platz 1 D-80539 M?nchen Tel.: +89-21805343 / Fax: +89-21805345 Email: W.Schulze at lrz.uni-muenchen.de http://www.lrz-muenchen.de/~wschulze/ ***************************** From bresnan at CSLI.STANFORD.EDU Fri Dec 17 19:57:46 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Fri, 17 Dec 1999 11:57:46 -0800 Subject: OT and diachronic functionalism In-Reply-To: Your message of Fri, 17 Dec 1999 10:15:41 GMT. <385A0D42.C59427A2@eva.mpg.de> Message-ID: 1. Fritz Newmeyer asks whether I subscribe to (a), (b), or (c): a. Constraints are universal, and may or may not be functionally motivated b. Constraints are universal and functionally motivated c. Constraints are universal and are functional motivations themselves. ???? I am subscribing to (b) as a hypothesis [not a religious tenet]. 2. Fritz asks how (b) or (c) can be reconciled with the existence of divergent word order patterns in English: predominant right branching coexisting with the left branching genitive N construction. If harmonic consistency in branching direction is functionally motivated (as argued by Dryer, Hawkins, etc.), how can we explain disharmonic patterns except as dysfunctional historical relics that belie (b) above? The mistake here, I think, is to talk about the functional motivation of *constructions* (or sentence types) instead of the functional motivation of *constraints* (which is what I am talking about). Constraints may be functionally motivated and yet conflict with each other, as for example constraints favoring ease of perception conflict with constraints favoring ease of production. There are many dimensions of harmony or markedness all at work at the same time, and a specific syntactic pattern which is optimal may in fact violate many well-motivated constraints. Consistency of branching direction is certainly not the only motivation at work in determining word order. As all good functional/typological linguists know, across languages possessive constructions often have special syntax, particularly when involving inalienable possession (but not exclusively so, if I am not mistaken). Hierarchies of animacy and topicality have often been invoked to account for this. (Cathy O'Connor has a very nice OT analysis of a case of this in Northern Pomo, a completely dependent-marking language everywhere except in the syntax of possession, which is both head-marking and strongly influenced by animacy and discourse topicality factors.) Many formalists would say that such things are merely historical wrinkles or accidents, reflecting nothing systematic about language. But I'm not so sure... Joan From gthomson at GPU.SRV.UALBERTA.CA Fri Dec 17 23:18:08 1999 From: gthomson at GPU.SRV.UALBERTA.CA (Greg Thomson) Date: Sat, 18 Dec 1999 04:48:08 +0530 Subject: competence/performance In-Reply-To: <385A0D42.C59427A2@eva.mpg.de> Message-ID: At 10:15 +0000 12-17-1999, Martin Haspelmath wrote: >I'm somewhat surprised to see that Joan Bresnan dissociates herself from >the competence/performance terminology. OK, it probably has a lot of >baggage attached to it that I may not be aware of, but my impression is >that we all pretty much agree that there is a necessary conceptual >distinction between language use and language structure, or speech and >grammar, or more generally between processing and storage, or cognitive >events and cognitive patterns. I believe Chomsky has used phraseology along the lines of "a disticntion between what one knows and what one does with what one knows". For a long time I couldn't imagine how anyone could take issue with the validity of such a distinction. However, suppose there are comprehension mechanisms that just sit there until they encounter speech, and at that point they process it. Suppose also that there are other mechanisms that just sit there until a to-be-verbalized concept gets kicked into them, and at that point they just verbalize it. In that case, competence is the performance systems. The only distinction to be made is a distinction between when they are running and when they aren't running. If there is a third system in people in addition to the comprehension and production systems--one that specifies or characterizes the sentences of their languages and the structural properties of those sentences--what is it for? In answer to Tuggy's question regarding harmful Chomskyan ideas, I would cite this idea as a damaging one. In my own area of special interest, Second Language Acquisition, it has lots of people thinking about acquiring grammars, rather than thinking about developing comprehension and production mechanisms. Greg Thomson From dubois at humanitas.ucsb.edu Fri Dec 17 22:25:12 1999 From: dubois at humanitas.ucsb.edu (John W. Du Bois) Date: Fri, 17 Dec 1999 14:25:12 -0800 Subject: CSDL 5 Conference Announcement Message-ID: CSDL 5 -- PRELIMINARY ANNOUNCEMENT The 5th conference on: "Conceptual Structure, Discourse, and Language" will be held at the University of California, Santa Barbara, on May 11-14, 2000. The conference is sponsored by UC Santa Barbara's Center for the Study of Discourse, Cognitive Science Program, Department of Linguistics, and Department of Geography. Papers in all areas of cognitive linguistics and related research areas are welcome, including research on conceptual structure, grammar, meaning, cognitive processing, acquisition, language use, discourse function, and other issues. Papers are especially encouraged bearing on, but not limited to, the special conference themes of: Metaphor Analogy Irony Space Grammar and Cognition Discourse and Cognition Learning and Acquisition Interactionally Distributed Cognition Abstracts are due February 14, 2000. An abstract of 500 words should be submitted via email to Patricia Clancy at: pclancy at humanitas.ucsb.edu Additional information about the conference will be made available shortly via the Internet. For additional information contact John Du Bois at dubois at humanitas.ucsb.edu. From jrubba at CALPOLY.EDU Fri Dec 17 22:43:08 1999 From: jrubba at CALPOLY.EDU (Johanna Rubba) Date: Fri, 17 Dec 1999 14:43:08 -0800 Subject: On Chomsky Message-ID: In response to Dave Tuggy's query: Autonomy of syntax certainly seems a most problematic aspect of Chomsky's view of language for those of the cognitive/functional/discourse analysis persuasion (not that I feel that any of these three labels are necessarily mutually exclusive). These lines of investigation seek motivation for syntactic structures in various places -- discourse, typology, the nature of the language-producing organism, history of languages and grammaticization, and synchronic semantics. Seeking motivations other than the 'innate universals in a syntax-specific module' posited by generativism seems to be a defining characteristic of many non-generative approaches to language. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Johanna Rubba Assistant Professor, Linguistics English Department, California Polytechnic State University One Grand Avenue ? San Luis Obispo, CA 93407 Tel. (805)-756-2184 ? Fax: (805)-756-6374 ? Dept. Phone. 756-259 ? E-mail: jrubba at calpoly.edu ? Home page: http://www.calpoly.edu/~jrubba ** "Understanding is a lot like sex; it's got a practical purpose, but that's not why people do it normally" - Frank Oppenheimer ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From chafe at HUMANITAS.UCSB.EDU Sat Dec 18 20:34:09 1999 From: chafe at HUMANITAS.UCSB.EDU (Wallace Chafe) Date: Sat, 18 Dec 1999 12:34:09 -0800 Subject: Problems with Chomsky In-Reply-To: <9912169453.AA945364013@router-8.camnet.com> Message-ID: David Tuggy's questions invite so many different answers from so many points of view that I imagine most subscribers to this list will throw up their hands and not even attempt to answer them. "Let me count the ways." But we shouldn't leave things with the answer that the only thing wrong with Chomsky's program is deep structure. Many of us, I'm sure, could write books on this subject if we wanted to, and the books would be quite different. But let me mention several general points that immediately occur to me, without trying to go into the detail they deserve. (1) The nature of the data. I said a little about this in a recent message to this list, so won't repeat it here. But it's worth adding that an extraordinarily restricted set of phenomena have been involved. (2) The focus on the "sentence" as the preeminent unit of language. (3) The assignment of too many phenomena to innateness. (4) The refusal or inability to explain linguistic phenomena in the context of other human endowments, whether cognitive (including memory, consciousness, imagery, emotions, etc.); social (clearly language is an interactive phenomenon); or historical (one can hardly understand the shape of language without taking account of language change). (5) The disregard for the fundamental importance of meaning or content, which can be traced to Bloomfield's infatuation with behaviorism and logical positivism as subsequently exaggerated by Harris. Language organizes thoughts as much as sounds, and in some ways thoughts are more important. In even more general terms, one might say that this vast, wonderfully beautiful human endowment we call language has been reduced to a few rather mundane phenomena of limited interest, over which a house of cards (or series of them) has been erected. Postscript: It would be easy to question some of the positive contributions listed by Brian, but I'll mention just the notion of "creativity", which has always puzzled me. The ability of language users to produce and understand novel sentences (or whatever) doesn't come from recursion, but from the insertion of a vast lexicon into a relatively small set of patterns. Yesterday my wife said to me "Don't mist the ribbon." It had to do with squirting water on a Christmas wreath. I doubt that she had ever said that before, and I certainly had never heard it before, but there was no problem (and no recursion). Wally Chafe From wilcox at UNM.EDU Sat Dec 18 21:07:12 1999 From: wilcox at UNM.EDU (Sherman Wilcox) Date: Sat, 18 Dec 1999 14:07:12 -0700 Subject: Problems with Chomsky In-Reply-To: Message-ID: On 12/18/99 1:34 PM, Wallace Chafe said: > Postscript: It would be easy to question some of the positive > contributions listed by Brian, but I'll mention just the notion of > "creativity", which has always puzzled me. Hurray! I've always thought that the Chomskyan perspective on linguistic creativity was just plain bizarre. For me, language creativity is more akin to musical creativity than it is to mathematical recursion. So I find the words of a fine jazz musician such as Clark Terry more insightful than Chomsky on the subject. In an interview on a recent reissue of an old jazz album (with Oscar Peterson), Terry commented on his performance of "Mack the Knife" (PP is the interviewer): ***** CT: Well, all I can say is, I didn't make too many obvious mistakes. I got through that one okay. When you give vent to your feelings, playing jazz, some of your better solos spin off from what could have been catastrophes. This is something that we learned in the ghetto, where there's varmints -- we used to say, "Damn rat, I'm going to fix him -- I know where his hole is." So we would chink up the holes with a lid from a tin can, nail it down. But he finds another way. If you're attempting to get your idea through one way and it's closed, you have to get it out, so you take another way. Like a rat going from one rat hole to the other, if you're trapped, you can't stop and say, "Let's do that over again." You have to figure out ways and means, through the medium of your having mastered the instrument. It's not like the classics. It's extemporaneous composition. ... It's avoiding catastrophe. PP: But you can always base what you're playing on the standard. CT: Oh, yes. You're playing the changes, and you do what the old-timers did before they knew anything about theory or harmony or counterpoint. They used the melody as a guy wire to extemporaneously superimpose another melody. That's [still] how cats develop ears, and are able to play by ear. But this was all that the old-timers had to go by. PP: ... so they improvised by necessity, because they didn't have or couldn't read music? CT: Yes. Just finding a way to get through the song helped to develop their creativity. ***** Not only does Clark Terry hint at Wally's "insertion of a vast lexicon into a relatively small set of patterns" and the "real" language that he discussed in a previous message, but Terry provides what I'd like to offer as an alternative definition of linguistic creativity: avoiding catastrophe in extemporaneous composition. Sherman Wilcox Department of Linguistics University of New Mexico From bresnan at CSLI.STANFORD.EDU Mon Dec 20 18:46:24 1999 From: bresnan at CSLI.STANFORD.EDU (Joan Bresnan) Date: Mon, 20 Dec 1999 10:46:24 -0800 Subject: OT and functionalism In-Reply-To: Your message of Mon, 20 Dec 1999 11:22:32 +0100. <3.0.5.32.19991220112232.00797e80@mail.hum.uva.nl> Message-ID: Dear Harry Perridion, Thank you very much for your remarks on Dutch phonology, which are obviously intended for the FUNKNET audience, in view of the third person form of address to me. You make the point that it is not literally true that Dutch devoicing occurs in coda position, because in that position but before b or d voicing assimilation occurs (if there is no pause). You comment: > It is remarkable that OT is capable of explaining non-existent facts. I think you are correct that the generalization about the distribution of voiced obstruents in Dutch is overridden by anticipatory voicing assimilation in the particular context you note. But you are mistaken in concluding that this fact somehow undermines OT phonology or the point I was using the Dutch example to make. Phonological systems actually have more than the three constraints I used in my example. In particular, there are very well motivated constraints that favor voicing assimilation. In Dutch, such a constraint overrides the positional markedness constraint against voicing contrasts in codas. The reference I gave for the Dutch devoicing example in my original message is Rene' Kager's 1999 CUP book _Optimality Theory_. There and in the references you will find an excellent discussion, with many illustrations, of how the interactions of various phonological processes can be explained more insightfully in OT than in previous rule-based terms. Best wishes, Joan Bresnan xxxxxxxxxx > In one of her contributions to the interesting discussion on OT and > functionalism on this list Joan Brennan wrote the following on final > devoicing in Dutch: > > >>For example, Dutch and German have a voicing contrast in obstruents, > which appears in syllable onsets--a very salient position-- but voiced > obstruents are devoiced in syllable codas. << > > This is not correct. The facts are as follows. Voiced obstruents are not > possible before a pause, or a voiceless obstruent. The 'f' in 'leefde' > (lived) e.g. is both syllable- and morpheme-final, but voiced, as is the > 'b' in 'ebde weg' (ebbed away). In fact, syllable-, morpheme- and even > wordfinal obstruents are voiced before /d/ and /b/, if there is no > intervening pause, e.g. 'op' [p] (on) + 'doen' (do) -> 'opdoen' (to put > on) with [-bd-]. > What we have here is a rather simple case of voice-assimilation, that is, > if we assume that a pause functions like a voiceless obstruent. > CC-sequences in Dutch (C is either a stop or a fricative) are either voiced > or voiceless. The assimilation is regressive, except when the second > obstruent is a voiced fricative (/z/ or /v/), in which case it is progressive. > > >>OT can explain the > positional neutralization of obstruent voicing in syllable codas and > relate it to the typological of segment inventories across languages. > The reason is simply that the *same* constraints reflecting ease of > perception and articulatory effort are present in the grammars of all > languages (as the universal constraint set). No other of grammatical > theory that I know of does this.<< > > It is remarkable that OT is capable of explaining non-existent facts. > > >>In contrast, there appear to be few or no languages whose > language-internal phonologies have a voicing contrast in syllable > codas and no voicing contrast in syllable onsets.<< > > To a certain extent, American English is such a language: there is a > contrast in syllable codas between voiced and voiceless stops ('bed' vs. > 'bet') but not in onsets of syllables that follow a syllable ending in a > vowel ('latter' vs. 'ladder'). > Such voicing of previously voiceless stops in intervocalic (but > 'syllable-initial') position is well-attested in the history of a number of > languages. > > Another fact that might disturb the serene picture of voice in the > languages of the world sketched by some proponents of OT and/or > typologists, comes from Scandinavian: > > In Southern Scandinavian (Danish, southern Swedish and southern Norwegian) > there is no opposition between voiced and voiceless stops in final position > after a long vowel, but in S-Swedish and S-Norwegian the merger is a voiced > stop. Danish developed in most cases a voiced fricative (Old Danish 'ut' > (out) -> 'ud' -> 'udh' ('dh' stands for the fricative). > > It is possible that these Germanic languages are all typologically weird, > but it is more probable that the assumption that: > > >>Across languages, voiced obstruents > are typologically marked: there are languages having only voiceless > obstruents (e.g. Polynesian), and languages having both voiced and > voiceless obstruents (e.g. English), but no (or hardly any) languages > having only voiced obstruents. This typological asymmetry might be > explained in terms of constraints on articulatory effort and > perception that shaped the evolution of languages to explain why > voiced obstruents are more restricted crosslinguistically. But within > the phonological components of the grammars of particular languages we > see the same thing: voiced obstruents are restricted in their > *language-internal* distributions. << > > is simply wrong. In fact it surprises me that respected linguists like Joan > Bresnan (see the quotations above) and Martin Haspelmath feel obliged to > make sweeping statements like > > >>Voiced obstruents are harder to pronounce than > voiceless ones, especially in (syllable- or word-)final position.<< > > What evidence is there for such a statement? Is it a fact that voiced > consonants are harder to pronounce?, in what way? and why? > > Let us get the facts straight before arguing about their interpretation. > > > Harry Peeridon > University of Amsterdam > > From edith at CSD.UWM.EDU Mon Dec 20 18:56:54 1999 From: edith at CSD.UWM.EDU (Edith A Moravcsik) Date: Mon, 20 Dec 1999 12:56:54 -0600 Subject: antonomy of syntax Message-ID: Johanna Bubba has suggested that the most problematic aspect of Chomsky's view of language is the notion of autonomy of syntax. But I cannot see how this could be an important issue of divisive force. If the claim that syntax is autonomous means that it can be described without reference to meaning (and function), then the claim is almost true by definition and thus I do not believe even functionalists would take issue with it. The claim is necessarily true in the sense that any object in the world that includes recurrent parts and/or properties can be described in terms of the distribution (selection and arrangement) of these basic elements. Sentences are analyzable into recurrent parts - such as words - and thus sentences can be given a distributionally based description without reference to the meaning of these parts. To claim the opposite - that sentences can be described unless we consider meaning - would be like saying that, while the structure of a string of beads can be described in terms of the choice and order of the beads, as soon as this string serves the purposes of a rosary, the former structural description becomes invalid and a valid description can be constructed only if it is known which bead stands for which prayer. This is, of course, not so. As I think was suggested by several people earlier in this discussion, the difference between functionalism and formalism seems to exist not on the descriptive but on the explanatory level; but even there, it is less than categorical. Functionalists claim that most or all structural features of sentences can be explained in terms of meaning (or function). Formalists in turn claim that most - but not necessarily all - structural characteristics are subject to form-related (i.e., non-semantic and non-functional) explanations - which in turn may or may not be ultimately functionally-based themselves. If this is a correct characterization of the difference between the two persuasions, the differences is not a huge one. Edith M. ************************************************************************ Edith A. Moravcsik Department of Foreign Languages and Linguistics University of Wisconsin-Milwaukee Milwaukee, WI 53201-0413 USA E-mail: edith at uwm.edu Telephone: (414) 229-6794 /office/ (414) 332-0141 /home/ Fax: (414) 229-2741 From jrubba at CALPOLY.EDU Mon Dec 20 20:44:40 1999 From: jrubba at CALPOLY.EDU (Johanna Rubba) Date: Mon, 20 Dec 1999 12:44:40 -0800 Subject: antonomy of syntax Message-ID: To respond to Edith, I was writing about _explanation_ in linguistics, not description. It was clear from early in the program that generative linguistics had explanation as its goal, whether or not the program has ever achieved this. I think it is also clear that it has long been a goal of generativism to prove the existence of an innate, syntax-specific module, _physically_ distinct from other parts of the brain that process language (Chomsky's 'language organ'), to _explain_ the facts of syntactic structure, and thus to minimize to zero if possible the involvement of semantics in the explanation of syntax. The description of this module's workings would then constitute an explanation of why syntax is the way it is. I did some browsing in some foundational works of cognitive and functional linguistics, and several make explicit mention of autonomy of syntax being a central tenet of generative linguistics which they reject, replacing it with a tenet that claims the opposite. These ideas are laid out in parts of the work which are intended to summarize the core, central aspects of the cognitive/functional program, so I assume this means the autonomy thesis is a major difference between the research programs, at least in the view of some of the founders of these alternative theories. Some quotes: "Central to [Cognitive Grammar's] conception of grammatical structure are three closely related claims, which define the focal concern of this book ... 2. Grammar (or syntax) does not constitute an autonomous formal level of representation. Instead, grammar is sysmbolic in nature, consisting in the conventional symbolization of semantic structure." Langacker, _Foundations of Cognitive Grammar_ Vol. 1, p. 2. Givon, in _Syntax: A functional/typological introduction_, lists among 8 core components of generative grammar, the following one: "a. Structure and function: Language -- and syntax -- were conceived of as structure, existing and understandable independently of meaning or function. 'Autonomous syntax' then consituted its own explanation even within linguistics ..." (p. 7). He goes on to say at the end of this list: "The approach to the study of syntax adopted in this book developed gradually as a rejection of all the tenets of the transformational-generative tradition as listed above" (p. 9). George Lakoff, in _Women, Fire and Dangerous Things_, devotes a brief chapter to "The Formalist Enterprise", tracing the origin of formal theories of syntax to the history of mathematics, logic, and philosopy. I quote: "The idea that natural language syntax is independent of semantics derives from the attempt to impose the structure of mathematical logic on the study of human language and human thought in general" (p. 225). He writes a short bit later "The question of whether there is an independent syntax for natural language comes down to the question of whether the metaphorical definition that defines the enterprise of generative grammar is a reasonable way to comprehend natural language. Intuitively the idea that a natural language is made up of ininterpreted symbols is rather strange ... if language is a way of framing and expressing thought so it can be communicated, then one would expect that many (not nedessarily all) aspects of natural language syntax would be dependent in at least some way on the thoughts expressed" (p.228). The fact that both 'camps' have had to compromise a bit due to problematic data (as Lakoff notes in the quote given here) doesn't minimize the importance of autonomy vs. non-autonomy. There may be practitioners of each kind of linguistics who take on this particular 'article of faith' with more or less confidence/enthusiasm, but the salience of discussions of autonomy in these foundational works seems to me to be evidence of major disagreement on this principle. It also nearly always comes up in cross-theory discussions that I have with generative theorists. A favorite challenge of generativists is to name some favorite point of syntax (co-reference, for example) and ask 'How are you going to motivate _that_ semantically?' ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Johanna Rubba Assistant Professor, Linguistics English Department, California Polytechnic State University One Grand Avenue ? San Luis Obispo, CA 93407 Tel. (805)-756-2184 ? Fax: (805)-756-6374 ? Dept. Phone. 756-259 ? E-mail: jrubba at calpoly.edu ? Home page: http://www.calpoly.edu/~jrubba ** "Understanding is a lot like sex; it's got a practical purpose, but that's not why people do it normally" - Frank Oppenheimer ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From moorej at UCSD.EDU Mon Dec 20 22:02:27 1999 From: moorej at UCSD.EDU (John Moore) Date: Mon, 20 Dec 1999 14:02:27 -0800 Subject: defining generative grammar Message-ID: Jo Rubba writes: >I think it is also clear that it has long been a >goal of generativism to prove the existence of an innate, >syntax-specific module, _physically_ distinct from other parts of the >brain that process language (Chomsky's 'language organ'), to _explain_ >the facts of syntactic structure, and thus to minimize to zero if >possible the involvement of semantics in the explanation of syntax. The >description of this module's workings would then constitute an >explanation of why syntax is the way it is. This has been a long-standing point of contention here at UCSD, but it does not take the form one might suspect. Rather, those of us who are identified as formalists are constantly hit over the head with this autonomy thing from the members of the cognitive science community, who tend to favor cognitive/functional approaches. We, putative formalists, on the other hand answer: "but we don't necessarily believe in the autonomy of syntax - perhaps some do, but we don't." Jo was a student here many years ago, and must remember these exchanges. I find it interesting that Jo uses quotes from cognitive/functional linguists to bolster the idea that autonomy is a major tenet of formal grammar. I suspect that this is a position that is often attributed to formal linguistics by others, and not an integral part of the paradigm. During my graduate training in formal syntax at UCSC in the late 80s I never heard anyone make the above claims. Of course, it is true that perhaps a majority of formal syntacticians in the GB/P&P tradition believe something along the lines of the autonomy hypothesis. It is also true that work in formal syntax tends to not make reference to anything other than formal devices to account for syntactic data. However, this is clearly changing, and has been for some time (I'm struck by a number of functionalist works that argue, in the 1980s and 1990s, against the generative tradition, and cite only _Aspects_). We recently had a seminar on explanation here - it was interesting that the psychological reality/autonomy of a generative grammar was never mentioned. As far as I remember, the closest any of the formalist participants came to Jo's characterization of generative grammar were: (i) the proposal that formal principles should be maintained as constants and pushed to their limits in a deductive research strategy (papers by Baker and Borer). (ii) the claim that syntactic accounts are preferable because they are cast in a well-understood algebra, whereas the nature of functional principles invoke a less well-understood domain. (iii) a similar point that generative (as some functional) accounts are of particular value when they are explicit and make clear predictions. I realize that these three points are controversial (and I don't fully subscribe to i-ii); however, they are a far cry from the above characterization of generative grammar. >A favorite challenge of generativists is to name >some favorite point of syntax (co-reference, for example) and ask 'How >are you going to motivate _that_ semantically?' I've heard (and expressed this) somewhat differently - generative research has uncovered a lot of generalizations about language - the challenge for a cognitive/functional approach is to either capture similar generalizations or show that the generalizations are epiphenomenal. Conversely, of course, formal linguists need to take generalizations discovered by functionalists seriously - this is what I understand Joan and Judith to be doing. John Moore P.S. Bach and Partee sketched a semantic account of some binding facts in a CLS paper almost 20 years ago. Aspects of their proposal have been very influential in subsequent RG, LFG, HPSG works, as well as work by Will imams and Reinhart and Reuland. http://ling.ucsd.edu/~moore/ From MILLSCR at UCMAIL.UC.EDU Mon Dec 20 23:06:14 1999 From: MILLSCR at UCMAIL.UC.EDU (Mills, Carl (MILLSCR)) Date: Mon, 20 Dec 1999 18:06:14 -0500 Subject: antonomy of syntax Message-ID: There are a few of us here on the fringes for whom the exchange between Edith Moravcsik and Johanna Rubba, with an interesting sidebar by John Moore, on the autonomy of syntax sounds like a debate on how many agels can dance on the head of a pin. Not that this thread is not fascinating, but at a recent conference another internationally known linguist said something like, "Nobody is trying to do away with rules of grammar." To which I replied, "Some of us are." Syd Lamb has published several papers on the "Unreality of Syntax." Vic Yngve has argued that the notion of language itself is a historical inheritance that impedes understanding what we are talking about. Steve Straight has posted his views on "The Myth of G" to this list recently. Paul Hopper continues to publish and present work on emergent grammars. I have no doubt that syntax can be described autonomously. The question is "Why bother?" or rather, "What is gained and what is lost in linguistic theories that seem to require an autonomous syntactic module?" It would appear that for most linguists today the gains associated with an autonomous syntax outweigh the losses. But some of us have concluded the opposite. Carl Mills University of Cincinnati - -------------- next part -------------- An HTML attachment was scrubbed... URL: From nrude at ucinet.com Tue Dec 21 00:12:35 1999 From: nrude at ucinet.com (Noel Rude) Date: Mon, 20 Dec 1999 16:12:35 -0800 Subject: autonomy of syntax Message-ID: Hmm, To me, as in all the mind sciences, "emergence" and "supervenience" seem like covers for ignorance, merely reductionist "promisory notes": When we know enough about neurons and whatnot, then language will be accounted for bottom-up. But until we do, how can we be so sure that there is no top-down phenomena like grammar? Noel From sstraigh at BINGHAMTON.EDU Tue Dec 21 04:03:43 1999 From: sstraigh at BINGHAMTON.EDU (H Stephen Straight) Date: Mon, 20 Dec 1999 23:03:43 -0500 Subject: autonomy of syntax In-Reply-To: <385EC5E1.6B1@ucinet.com> Message-ID: Noel Rude ruminates: > To me, as in all the mind sciences, "emergence" and "supervenience" > seem like covers for ignorance, merely reductionist "promisory notes": > When we know enough about neurons and whatnot, then language will be > accounted for bottom-up. But until we do, how can we be so sure that > there is no top-down phenomena like grammar? Granted, we can't be sure that there is no overarching set of linguistic "rules" (Grammar), but we can be sure that the positing of such a "supervenient" entity violates the rule of parsimony. To pursue the study of language scientifically we must avoid positing any theoretical entities beyond those necessary for the pursuit of our inquiry. In this case, we know we need separate accountings of how people interpret the linguistic events they perceive and of how they create linguistic events. The Myth of G would have us accept the existence of a Grammar as a "source" of such events independent of the processes of reception and expression, and would have us ignore the evidence for discrepancies, separateness, dissociability, cognitive multiplexity, and other (often highly "creative") interactivity of receptive, expressive, and other language processes. Arguments for the existence of G have considerable intuitive appeal, but so do arguments for numerous other common-sensical and supernatural entities that rightly play no role in scientific accounts of reality. H Stephen Straight From haspelmath at EVA.MPG.DE Tue Dec 21 09:06:40 1999 From: haspelmath at EVA.MPG.DE (Martin Haspelmath) Date: Tue, 21 Dec 1999 09:06:40 +0000 Subject: the pseudo-issue of autonomy Message-ID: I agree with Edith Moravcsik and John Moore that autonomy is not the central issue that divides Chomskyans and functionalists. Just look at any descriptive grammar (quite a few functionalists have published such works) and examine the morphology section: Inflectional classes, various stem forms, morphophonemic alternations ? all this is systematic arbirariness, i.e. autonomy (in Newmeyer's 1998 definition). In my view, what often divides the two camps is that Chomskyans are primarily interested in solving Plato's Problem ('How can we acquire language?'), whereas functionalists are primarily interested in explaining language structure. Jo Rubba is right that "it was clear from early in the program that generative linguistics had explanation as its goal", but not explanation of language structure. Chomsky's "explanatory adequacy" is only about explaining language acquisition. The practical goal of spelling out the principles of UG is subordinate to the theoretical goal of solving Plato's Problem. From this it follows that functional explanations are anathema in generative linguistics ? because they undermine UG, and hence Chomsky's solution for Plato's Problem. Martin P.S. The above argument is presented in some detail in my review article on Fritz Newmeyer's (1998) book, due to appear in the next issue of Lingua. -- Martin Haspelmath (haspelmath at eva.mpg.de) Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22 D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616) From W.Schulze at LRZ.UNI-MUENCHEN.DE Tue Dec 21 12:53:57 1999 From: W.Schulze at LRZ.UNI-MUENCHEN.DE (Wolfgang Schulze) Date: Tue, 21 Dec 1999 13:53:57 +0100 Subject: antonomy Message-ID: Johanna Rubba wrote: > (...) I did some browsing in some foundational works of cognitive and > functional linguistics, and several make explicit mention of autonomy of > syntax being a central tenet of generative linguistics which they > reject, replacing it with a tenet that claims the opposite. These ideas > are laid out in parts of the work which are intended to summarize the > core, central aspects of the cognitive/functional program, so I assume > this means the autonomy thesis is a major difference between the > research programs, at least in the view of some of the founders of these > alternative theories. With respect to the autonomy debate we should perhaps add another view that is related to the cognitive program. This view claims that 'language' is the output of cognitive routines that interprets communicative experience in its interaction with information processing as a systematic entity (called 'language'). 'Language' is regarded as a learned knowledge system that is imposed (by tradition) on its own cognitive prerogatives, a recursive process that is typical for human cognition. This view does not propose that the process of language acquisition starts at say the age 1.0 years _because_ of the 'fact' that the brain has arrived at a certain maturation state that establishes the language faculty. Rather, it is assumed that people have become used to 'interpret' a certain mental stage, namely the maturation of the corresponding senso-motoric domains as a communicative signal that a child is 'ready for language acquisition'. The collective (and very old) experience that it is best to start to 'linguistically' train children at a certain age has established the collective mental hypothesis (or idealized cognitive model) that language is something more or less autonomous. The assumption of a modular linguistic 'substance' (in what shape so ever) represents a collective mental construction and constitutes an important part of folk-psychology. In this sense, this view has to observe two determinatives with respect to the ontology of language: On the one hand, language is constituted by nothing but the emergent activities of the cognition communication interface that in themselves are not 'language' but senso-motoric schemata related to the complex (and polycentric) network of senso-motorics, audiovision, and information storage procedures. In this sense, 'language' does not have a proper substance (not to speak of 'essence'), but represents cognitive 'events' that acquire a 'communicative and linguistic reading' via mental constructions. On the other hand, the experience of these events together with their paradigmatization during language acquisition ends in some kind of systematic knowledge that is _construed_ as a more or less autonomous something. Both aspects are structurally coupled and lead to what we experience as 'language'. The hypothesis that 'language' has acquired an own mental 'substance' (in terms of Universal Grammar, part of the Language of Thought (LOT) or what so ever) could then be regarded as a highly sophisticated 'scientifization' of the popular hypothesis about language: It refers to a modular interpretation of cognitive organization in which modules are treated as the substantial 'resultant' of the evolution of cognition. This is exactly what people in ordinary life normally think about language (at least in a Western tradition). This is understandable if we bear in mind that in Western tradition language is strongly coupled with '(self-)consciousness' etc.: 'Language' constitutes one of the basic parameters of human ontology in Western folk psychology. It is declared to represent some kind of autonomous substance that is correlated with a certain stage in language acquisition. Traditions related to the autonomy hypothesis refer to this kind of mental construction. Naturally, this is not done 'consciously', saying: "Oh, let's make the pop linguistic experience scientific". In fact, the Chomskyan paradigm represents a unwanted (?) reaction to this folk experience, a process that in itself forms part of the scientifization of popular experience (or folk philosophy) which started (at the latest) in the time of enlightenment. It is 'modern' - whereas some cognitive approaches in the sense described above have more in common with 'post-modern' paradigms. In terms of the historiography of sciences (esp. linguistics) the Chomskyan paradigm is a 'function' of the 'functional/cognitive' paradigm: Post-modern linguistics (if you allow this term) depend on the experience of linguistic modernity, but it tries to overcome it in a dialectic discourse that 'explains Chomsky' just as Chomsky 'explained' pre-modern traditions. Modernity can only _react_ on post-modern arguments (out of logical reasons) by disclaiming approaches related to this paradigm as 'Neo-Hocuspocus' (if ever they do). Wolfgang -- ***************************** Prof. Dr. Wolfgang Schulze Institut fuer Allgemeine und Indogermanische Sprachwissenschaft Ludwig-Maximilians-Universitaet M?nchen Geschwister-Scholl-Platz 1 D-80539 M?nchen Tel.: +89-21805343 / Fax: +89-21805345 Email: W.Schulze at lrz.uni-muenchen.de http://www.lrz-muenchen.de/~wschulze/ ***************************** From edith at CSD.UWM.EDU Tue Dec 21 15:54:55 1999 From: edith at CSD.UWM.EDU (Edith A Moravcsik) Date: Tue, 21 Dec 1999 09:54:55 -0600 Subject: autonomy and related issues Message-ID: This is in response to a number of recent postings on the concept of the autonomy of syntax. Talmy Givo'n argued that the difference between the functionalist and formalist persuasions was rather large because it was one between an adaptive, evolutionary approach to language and a non-adaptive, anti-evolutionary approach. I think this would be the case if functional explanations were entirely excluded in the formalist approach; but I do not think this is so, for two reasons. First, formalists I believe do not claim that ALL grammatical phenomena are derivable from general laws of form; only some - or most - are. This leaves room for functional explanations. Second, there is no reason why general laws of form posited by formalists should not themselves be derivable from functionally-based generalizations. In fact, the very concept of innateness seems to me to be a functional principle (although not well-delimited). That is to say, if a functional explanation is one that makes reference to the goals and means of language use - i.e., the communicative and expressive needs of humans and the psychological and physical constraints under which these goals are to be achieved - than innateness is a functional "explanation" since it makes reference to psychological and physical contraints of the language user. The very interesting quotes that Johanna Bubba listed show that the autonomy of syntax has indeed been considered to be a divisive issue between formalists and functionalists. A further question, however, is whether it has been so by necessity - or, as John Moore put it, whether the two opposing views on this issue are "integral part/s/ of the /two/ paradigm/s/". I would think the answer is no. As I see it, a purely formal description - that is, one without reference to meaning and function - is actually a necessary part of any functionalist's agenda. As Martin Haspelmath pointed out, everybody's grammar, including functionalists', will make reference to purely formal concepts and this is understandable: if functionalist are interested in how form relates to meaning and function, the very question necessitates a characterization of form all by itself - simply because we cannot talk about the relationship between objects unless there is first a characterization of those object constructed independently of their relationship. If this point is correct, it would constitute an answer to Carl Mills' query re why construct autonomous syntactic descriptions. All in all, I am proposing two points: (a) A purely formal description of sentence structure is a necessary part of functional accounts because sentence form is the entity the relationship of which to other things functionalists are interested in. (b) Functional explanations for sentence form may not be necessary parts of formalist accounts but they are possible parts of them first, because not all syntactic phenomena may be explainable in terms of laws about form, and, second, because laws of form may themselves be derivable from functional considerations related to human perception and cognition. The notion that before we can talk about the relationship between things, we need to characterize those things independently of each other may also provide an answer to the question raised by Stephen Straight and also discussed by Wolfgang Schulze - namely, why posit the very concept of grammar to beging with? I think we need a characterization of grammar because what we want to know things about it: how grammar is acquired, how grammar changes (ontogenetically and historically), how it is stored, and how it is used. If these are our questions, then we do need a characterization of grammar that is independent of how it is acquired, changed, and used. In other words, both the concept of a strictly-formal account of sentence structure and also the concept of a grammar are entitites that we need in order for the questions that we are asking to make sense - namely, how does form relate to meaning and function and how is grammar acquired, stored, processed, etc. The only way, it seems to me, not to be obliged to construct autonomous characterizations of the notions "syntactic form" and "grammar" would be by changing our research questions so that they are not about the relationship of these constructs to other things. Is this correct? Edith PS I very much agree with Martin Haspelmath that the/a main difference between the Chomskian and the functionalist agend is one of what the basic question is - one about language structure or one about the acquisition of language. ************************************************************************ Edith A. Moravcsik Department of Foreign Languages and Linguistics University of Wisconsin-Milwaukee Milwaukee, WI 53201-0413 USA E-mail: edith at uwm.edu Telephone: (414) 229-6794 /office/ (414) 332-0141 /home/ Fax: (414) 229-2741 From moorej at UCSD.EDU Tue Dec 21 16:57:14 1999 From: moorej at UCSD.EDU (John Moore) Date: Tue, 21 Dec 1999 08:57:14 -0800 Subject: the pseudo-issue of Plato's problem Message-ID: At 09:06 AM 12/21/99 +0000, Martin Haspelmath wrote: >In my view, what often divides the two camps is that Chomskyans are >primarily interested in solving Plato's Problem ('How can we acquire >language?'), whereas functionalists are primarily interested in >explaining language structure. Unfortunately, this may be true. However, if one looks at what formal linguists actually do, there is very little solving of Plato's problem, and a lot of explaining (or at, least accounting for) language structures (gratuitous references to "The Child" notwithstanding). Occasionally one finds an epistemological introduction in formal papers, but I just skip it and get directly to the linguistics, which is often very good. While perhaps not be typical of formal linguists' beliefs, at least some take a very agnostic stand on what the deep psychological explanation for language structures may be. This was overtly stated in GPSG work (I think Geoff Pullum has a Topic Comment essay on this issue). Joan can speak to the strong competence hypothesis in LFG, but I don't see the same kind of arm-chair psychology in LFG work that one sometimes encounters in some GB and P&P work; one never found it in RG work, as far as I know. Finally, I suspect that it is less common even in GB/P&P work than some of Chomsky's writings might lead us to believe. Again, Plato's problem was never mentioned by any of the formalists at the explanation conference. If we strip away these issues, which I think are rather peripheral to the enterprise, we may find that the differences between formal and functional linguistics are fewer than we thought - perhaps some methodological ones, but even those currently divide functional and cognitive linguistics (and they seem to get along). John Moore From macw at CMU.EDU Tue Dec 21 17:20:11 1999 From: macw at CMU.EDU (Brian MacWhinney) Date: Tue, 21 Dec 1999 12:20:11 -0500 Subject: autonomy, Plato's Problem, and deep structure Message-ID: I would like to second Martin Haspelmath's take on the importance of Plato's Problem in the definition of the Chomskyan program. Since 1965, Chomsky has emphasized the role of the Logical Problem of Language Acquisition (LPLA) in generative theory. The formulation of the LPLA was an intellectual move necessitated by the commitment in 1957 in Syntactic Structures to deep structure. The "discovery" of deep structure was, in turn, the achievement that justified the introduction of transformations and the application of automata theory to language. As a result of these linkages, the commitment to deep structure became central to the Chomskyan program. In Plato's cave, true ideas are only seen as reflections. However, we often don't really need to look at the linguistic input at all, according to Chomsky, since the true shape of the deep structure of language is resident in our minds from birth. It is possible to imagine routes to deep structure that do not go through Plato's cave. Competence-performance and introspection regarding degrees of grammaticality provide one such route. I agree with Martin, Edith, and others that the autonomy of syntax is not a crucial support for deep structure. For example, Freud's deep structure is nicely grounded on claims about biological drives and developmental principles. Freud's example shows that there is no reason in principle that a functionalist theory in linguistics could not be grounded on a deep structure that was surrounded by a shell of protective concepts such as competence-performance, LPLA, sentences instead of utterances, the ideal speaker-hearer, degrees of grammaticality, and the rest. However, I doubt that many of us would be attracted to a functionalist theory of this type. For this reason, I think that a willingness to minimize the role of an abstract, underlying deep structure that has no easy match to observed facts about learning, processing, and physiology is the fatal flaw in the Chomskyan program. Having said this, I wonder whether Martin and Edith really want to say that the functionalist needn't care much about language acquisition. I agree that people in language development have not done a good job of showing functionalists why learning is important. But, if functionalism wants to achieve explanation, it cannot ignore learning. --Brian MacWhinney From noonan at csd.uwm.edu Tue Dec 21 23:34:34 1999 From: noonan at csd.uwm.edu (Michael Noonan) Date: Tue, 21 Dec 1999 17:34:34 -0600 Subject: job in Punjabi linguistics Message-ID: Below is a job announcement for a position in South Asian Studies at the University of Wisconsin-Milwaukee. The ad is fairly generic, covering a number of fields, but the position there described could be filled by a linguist whose specialization is the Panjabi language. If you have questions about the position, feel free to contact me. However, do not send application materials to me. The job ad copied below specifies to whom applications should be sent. Mickey Noonan Michael Noonan Professor of Linguistics Chair Office: 414-229-4539 Dept. of English Fax: 414-229-2643 University of Wisconsin Messages: 414-229-4511 Milwaukee, WI 53201 Webpage: http://www.uwm.edu/~noonan USA University of Wisconsin - Milwaukee South Asian Studies The College of letters and Science invites applications for a tenure track assistant professor position in South Asian Studies beginning Fall, 2000. In particular, the College seeks a person whose teaching and research interests cover one or more of the following areas: language, culture, civilization, or economic development. A strong preference will be given to candidates whose research and teaching interests include Punjab, India and Sikhism. The successful candidate will have a broad knowledge of the region, and his/her tenure home will be in one of the following departments: Economics, English, History, Political Science, Sociology, or Foreign Languages and Literature. The candidate will be expected to teach both undergraduate and graduate courses and direct graduate research. The successful candidate should be able to demonstrate evidence of excellent research potentials and teaching ability. Applicants must have a completed the Ph.D. by August 2000. ABD candidates may be eligible for a lecturer position. Candidates should submit a letter of intent, curriculum vita, research papers, teaching evaluations and three letters of recommendation by January 24, 2000. The University of Wisconsin - Milwaukee is an equal opportunity institution committed to diversity. Contact: Swarnjit S. Arora, Chair South Asian Faculty Search Committee, Department of Economics, Bolton Hall, University of Wisconsin - Milwaukee, WI, 53201. From jl.mackenzie at LET.VU.NL Wed Dec 22 14:03:56 1999 From: jl.mackenzie at LET.VU.NL (J.L. Mackenzie) Date: Wed, 22 Dec 1999 15:03:56 +0100 Subject: Bibliography of Functional Grammar Message-ID: I am pleased to announce the publication of "Functional Grammar Publications 1978-1998", compiled by Casper de Groot and Hella Olbertz. This is a complete overview of the work of Simon Dik and his co-workers that appeared between 1978 and 1998. The bibliography, which contains 1141 alphabetically arranged entries, also has a subject index and a language index. The bibliography has appeared as Working Papers in Functional Grammar 72 (ISSN 0924-1205) and is available for NLG 10 (Dutch guilders) from the executive editor: Aletta Smits, IFOTT, Faculty of Humanities, University of Amsterdam, Spuistraat 210, 1012 VT Amsterdam, Netherlands (e-mail: fg at hum.uva.nl; fax: +31-20-5253052). Other WPFGs that appeared this year were: 67 Dik Bakker, "FG expression rules: from templates to constituent structure" 68 Kwee Tjoe Liong, "Questions in the quasi-productive mode of the Functional Grammar model" 69 Ahmed Moutaouakil, "Exclamation in Functional Grammar: sentence type, illocution or modality?" 70 Kwee Tjoe Liong, "Adverbial clauses, FG, and the change from sentence grammar to discourse-text grammar" 71 Dik Bakker & Ewald Hekking, "A functional approach to linguistic change through language contact" All enquiries regarding WPFGs should be addressed to Aletta Smits at the address above. Lachlan Mackenzie Editor WPFG From W.Schulze at LRZ.UNI-MUENCHEN.DE Wed Dec 22 17:39:35 1999 From: W.Schulze at LRZ.UNI-MUENCHEN.DE (Wolfgang Schulze) Date: Wed, 22 Dec 1999 18:39:35 +0100 Subject: Language acquisition and functionalism Message-ID: Dear Brian and Funknetters, the recent comments upon the Chomskyan program have rightly focused on the role that language acquisition plays in this program. The chain of arguments saying that (1) deep structure presupposes universal categories and universal mental 'operations' in terms of Universal Grammar, (2) UG presupposes a language acquisition device (LAD), and (3) LAD presupposes its embedding in something like the evolution of a 'Language of Thought' (LOT) yields at - if I understand it correctly - the 'reconstruction' of UG to establish a 'meta-empiric' (or meta-descriptive) domain of research. In this sense, however, the 'reconstruction' of the LAD represents an _intermediate_ step in a much larger prospectus, namely the 'reconstruction' of LOT. It _may_ play the most important part in the game just because language acquisition procedures seem to be accessible to direct empirical observation, contrary to UG, LOT and LAD itself that can only be indirectly accessed via primary (UG) and secondary (LOT) extrapolation. But why do 'functionalists' have to refer to language acquisition (or, better, 'learning')? Is it just because the Chomskyan paradigm has opened a perspective so far ignored by functionalists? In this case, the Chomskyan paradigm would yet again represent the argumentative 'anchor' for 'deciding' which explanatory and/or descriptive domains of research have to be accessed within the paradigm of functionalism. In other words: Is it that we have to deal with language acquisition just because Chomsky did so (since 1965)? The observable tendency, namely that functional argumentation is sometimes (even often?) based on the 'reflex' to _react_ on Chomskyan hypotheses, to accommodate these hypotheses to the program of functionalism, and - in a second step - to declare the outcome as an intrinsic part of functionalism has represented a salient aspect in the history of (younger) functionalism (say since roughly 1961 (Dobbs Ferry)]. This is nothing new and it is self-evident if we consider the 'private histories' a number of (now) functionalists. It is logical, if we remember the fact that 'Standard Functionalism' (to use a cover term) is a program that is younger than the MIT 'orthodoxy'. It is logical, too, if we bear in mind that for the first time in the history of linguistics, two scientific paradigms coexist (in a more or less friendly manor) for more than 40 years by now. The scientific discourse between these two paradigms, however, is rather unilateral, which means that functionalists tend to carefully watch the activities of the East Pole (to quote R. Hudson) whereas the MIT orthodoxy usually refers to the output of the West Pole in an rather unspoken manor (if ever they do). Naturally, functionalists cannot deny the existence of the MIT paradigm. They cannot start from zero. This would be both uneconomical and ahistorical. However, what they can do is formulating a research program that would allow to tell people _why_ functionalism has to deal with some of the major issues also discussed by the MIT tradition. In other words, functionalism needs arguments that are 'deduced from it own (unspoken) deductions'. Let me again address the question "why functionalists should deal with language acquisition". Does 'functionalism' (what ever this means) have a program from which we can derive an answer to this question saying 'yes!'? Brian is probably right saying that "if functionalism wants to achieve explanation, it cannot ignore learning". Obviously, 'explanation' forms a major point in this program yet far from being fully formulated. But what can 'language acquisition' tell a functionalist with respect to the explanation of linguistic data (or, communication etc.)? I for myself am sure that 'learning' (as well as the accommodation of learning experience and maturation of brain structures) represents a decisive point even in functional typology, just as for language change and formal/functional diachrony (not to speak of language contact: The linguistic interaction of children and parents etc. conform to many aspects of language contact). However, is most cases functionalists seem to refer to language acquisition only in case this explanatory domain is thought to be adequate (and others fail). What we need (in my mind) is a more general (but 'polycentric') frame work for functionalism that explains _why_ domains like language acquisition form a necessary part of the functional paradigm. Merry Christmas and a Happy New Year to all, Wolfgang ***************************** Prof. Dr. Wolfgang Schulze Institut fuer Allgemeine und Indogermanische Sprachwissenschaft Ludwig-Maximilians-Universitaet M?nchen Geschwister-Scholl-Platz 1 D-80539 M?nchen Tel.: +89-21805343 / Fax: +89-21805345 Email: W.Schulze at lrz.uni-muenchen.de http://www.lrz-muenchen.de/~wschulze/ ***************************** From wcmann at JUNO.COM Wed Dec 22 22:25:15 1999 From: wcmann at JUNO.COM (William Mann) Date: Wed, 22 Dec 1999 17:25:15 -0500 Subject: submission deadline for Soc. for Text and Discourse: now Jan. 10, 2000 Message-ID: Dear Funknet subscribers: At their request, I am forwarding the information below concerning a conference of the Society for Text and Discourse in Lyon, France. (ST&D publishes Discourse Processes.) Bill Mann ====================================================== Dear Colleagues, The deadline for submitting a proposal to the 10th Annual Meeting of the Society for Text and Discourse in Lyon has been extended to January 10th, 2000. All the information on the meeting are available on the WebSite: http://unpc.univ-lyon2.fr/STD/index.html Have a good Christmas vacation Regards, Isabelle Tapiero [SORRY IN CASE OF MULTIPLE RECEPTIONS OF THIS MESSAGE] Dr. Isabelle Tapiero Chairman of the Scientific Committee for the 10th Annual Meeting for the ST&D E.mail:Isabelle.Tapiero at univ-lyon2.fr From nrude at ucinet.com Sun Dec 26 03:44:37 1999 From: nrude at ucinet.com (Noel Rude) Date: Sat, 25 Dec 1999 19:44:37 -0800 Subject: autonomy and related issues Message-ID: Greetings all, Yes to my mind Edith sums it up well. Though I might add that Givon has never repudiated "transformations" per se. In fact (I've heard him say) the Chomskians ought have kept them. Transformations are simply a mathematical way of relating structures to one another. The notion, e.g., has given us more precision in describing voice constructions, though most of us would assume the Chomskians' "autonomous" level of deep structure superfluous. In our tradition transformations mediate between "surface structure" and semantic case roles and discourse/pragmatics. Such "transformations" can be thought to preserve semantic structure and meaning, but as functionalists we will note that they add a discourse/pragmatic perspective. Noel From nrude at ucinet.com Sun Dec 26 04:10:37 1999 From: nrude at ucinet.com (Noel Rude) Date: Sat, 25 Dec 1999 20:10:37 -0800 Subject: autonomy of syntax Message-ID: Prof Straight says: "The Myth of G would have us accept the existence of a Grammar as a 'source' of [interpreting and creating linguistic events] independent of the processes of reception and expression, and would have us ignore the evidence for discrepancies, separateness, dissociability, cognitive multiplexity, and other (often highly 'creative') interactivity of receptive, expressive, and other language processes. Arguments for the existence of G have considerable intuitive appeal, but so do arguments for numerous other common-sensical and supernatural entities that rightly play no role in scientific accounts of reality." Probably I don't understand what Prof Straight is saying. Or maybe we can agree to disagree, perhaps mostly in what Science is. For it seems to me that theories can be very abstract and "supernatural" (if you will). If they are predictive-refutable--as our grammars and Grammar should be--they are "scientific". Also it seems to me that the messiness of linguistic data, rather than refuting an underlying system, actually suggests it, whatever flaws there might be in Saussure's langue et parole and Chomsky's competence-performance models. Is this just an esoteric argument where in practice we come down to the same thing? Will we both draw up verbal and nominal paradigm charts, describe grammatical relations, posit functions, etc., and some of us will call it "grammar" and "rules" and others will call it something else? Aren't we all looking for regularities--whatever we might call them? Noel From john at RESEARCH.HAIFA.AC.IL Sun Dec 26 14:04:53 1999 From: john at RESEARCH.HAIFA.AC.IL (John Myhill) Date: Sun, 26 Dec 1999 16:04:53 +0200 Subject: Who can you 'love'? Summary Message-ID: Dear Funknetters, I have finished my little survey of who people can 'love' aside from romantically, through family ties, and hyperbolically ('I just LOVE Frank Sinatra')--in other words, friends. I'm sorry I took so long to put together the results. I foolishly sent out the question when I didn't really have time to deal with it, so it took me a while to get answers I could put into a framework for presentation. Of course, the biggest problem, aside from generally small numbers, is that the respondents were self-selected. Oh well. Anyone else out there who feels like giving me more data with their own usage, I would welcome your contribution very much. Here are the results; thanks very very much to those of who who provided the data: In the end, I received 13 responses. I guess 14, if I count myself. Actually in the end there were 9 men and 5 women (presumably as a result of me specifically encouraging men to respond after my first attempt). Of these, we can make a general division into three categories: (1) For 4 people, all males, they would not use the word 'love' for anyone outside their family (I include in this group a gay male who said he could use the word 'love' for one `gay personal friend, with whom I never had a sexual relationship--the relationship was more like he was a son to me.)' (2) For 2 people (1 male, 1 female), there seemed to be a more or less open set of non-family, non-relatives they 'love.' One was a male who describes himself as a Christian `as a result of commitment to basic truths of Christianity', who approximated the number of people he 'loves' at several dozen. The other is a woman who attributes her understanding of 'love' to the fact that she lives in Southern California (`I am in Southern California and here you can love anything or anyone-from bare acquaintances to broccoli-(similarly, nothing is fine it is either great, wonderful, or fantastic)'); interestingly, when she described her understanding of 'love' (she wrote `I'm not sure it is actually a semantic difference but a pragmatic one; in other words, if the 'Default" is loving someone, what is that attitude of the speaker towards the referent if he/she merely 'likes' that person?'), it sounded remarkably similar to what Israelis have said to me about ''ahav'. (3) For the remaining 8 people (4 females, 4 males), 'love' can be applied to a limited number of very close friends. All of the women can use it towards either males or females, but all use it for more females than males. For the men, on the other hand, there is no pattern-one uses it only for one male friend, one for 2-3 female friends, one for 7 male friends and one female friend, another for mostly female friends. There seems to be considerably more similarity in female usage than male usage. There also seemed to be, initially, considerably more willingness on the part of women to volunteer information about themselves before I made my second posting. On the other hand, once I had gotten men to participate, some were surprisingly forthcoming and much more specific than I had expected in terms of exactly who they love and who they don't love. The impression I had is that the women seemed to know what to answer, and they answered in a generally similar way, whereas the men seemed to be applying some criteria which they had made up for themselves somewhere along the way, which they didn't agree on among themselves at all (after having conducted the survey, I realized that perhaps part of the reason I feel that I only 'love' people in my family is that otherwise I would have to make up some other criteria for non-romantic 'loving', and I have no idea of what these might be). I seem to be saying 'seem' a lot in this report. Thanks again for the input. John From dgolumbi at PANIX.COM Sun Dec 26 23:36:06 1999 From: dgolumbi at PANIX.COM (David Golumbia) Date: Sun, 26 Dec 1999 18:36:06 -0500 Subject: autonomy of syntax In-Reply-To: <38659539.1B65@ucinet.com> from "Noel Rude" at Dec 25, 1999 08:10:37 PM Message-ID: here's a sentence from a recent posting that, for me, captures the general intuition i find most troubling in both Chomskyan and non-Chomskyan generativist approaches: > Also it seems to me that the > messiness of linguistic data, rather than refuting an underlying system, > actually suggests it, whatever flaws there might be in Saussure's langue > et parole and Chomsky's competence-performance models. Not just that the messiness would *accommodate* or *allow* a system, but "suggests" it -- how have we got to the point where this seems a reasonable interpretation of the world's barely-documented, hardly known linguistic diversity? -- dgolumbi at panix.com David Golumbia