Leaving the Myth Behind.

My apologies to those of you who either subscribe to Chomskyan linguistics or aren’t interested in it, but I still bear the scars of attempted brainwashing from my time in grad school four decades ago, and I can never get enough of attacks on the Great Man and his Theory (or, more accurately, Theories). Herewith, for those who are interested, Christina Behme and Vyvyan Evans, “Leaving the myth behind: A reply to Adger (2015)” (pdf), a satisfying response to Adger’s defense of Chomsky against Evans’s The Language Myth (a book I’ll have to get hold of some time) and article “There Is No Language Instinct.” Here’s the concluding paragraph; click through for the detailed discussion:

Minimalists have directed harsh criticism at The Language Myth and There is no language instinct, but little of this criticism seems to concern substantial issues. Alleged misunderstandings or misinterpretations of the Chomskyan commitment and the Chomskyan framework are in large degree due to the imprecise and, at times, inconsistent formulation of its basic assumptions. Chomskyans refer informally to language as an instinct, do not use key terms (e.g. I-language) consistently, do not provide precise definitions of important concepts (e.g. ‘innate’, ’language organ’, Universal Grammar), and they regularly conflate the meanings of ‘recursion’ Adger wishes to keep separate. Given that most of The Language Myth has been ignored by Adger, he is no position to judge whether it makes a valuable contribution or should be dismissed. And he has given little reason to think that the minimalist research program can shed light on there being “new exciting challenges to be addressed about how language is implemented in the brain, how what we know about language structure can improve statistical translation techniques, how language interacts with other systems in our minds and how it’s put to use in situations of social complexity” (Adger, 2015: 80). Adger seems to believe that generative grammarians continue to play the central role in syntactic research, and that they ought to shape the agenda of a larger, multidisciplinary research community. Yet, as Chomsky pointed out decades ago: “this framework is only taken seriously by a tiny minority in the field … it does not represent a major tendency within the field in statistical terms” (Chomsky, 1982: 41). It is arguable whether this evaluation was accurate in 1982. But, Chomsky could have hardly offered a better prediction for that state of the field in 2015. Anyone who wishes to defend the Chomskyan framework ought to move beyond the fruitless quarrelling that has distracted so much attention from the real issues, and address the following questions: [i] what are the specific theories Chomskyans are currently committed to, [ii] which concrete findings from developmental psychology and neurobiology support the Chomskyan framework, and [iii] how can the Chomskyan paradigm overcome the familiar, long standing challenges stated in the technical literature, including those by other
generativists (e.g. Culicover and Jackendoff, 2005; Jackendoff, 2011; Seuren, 2004).


  1. marie-lucie says

    I was glad to read the joint reply to Adger’s review, an attempted hatchet job on Evans’ work which was particularly virulent and therefore cast doubt on the credibility of the review’s author even without considering specific points of disagreement. One has to agree with Evans & Behme’s comment: [Adger] has given little reason to think that the minimalist research program can shed light on there being “new exciting challenges to be addressed about how language is implemented in the brain, how what we know about language structure can improve statistical translation techniques, how language interacts with other systems in our minds and how it’s put to use in situations of social complexity”.

  2. (ii) narrowly focuses on alleged misunderstandings but provides no thorough clarification of the Chomskyan framework

    Fish gotta swim, birds gotta fly…

  3. From the Jargon File:

    adger, v. [UCLA mutant of nadger, poss. also from the middle name of an infamous tenured graduate student] To make a bonehead move with consequences that could have been foreseen with even slight mental effort. E.g., “He started removing files and promptly adgered the whole project” . Compare dumbass attack.

    The grad student in question was Ellison Adger Williams III. Tenured is a pun on ten-yeared.

  4. David Eddyshaw says

    Chomskyism is to linguistics much as psychoanalysis is to psychology, except that it has shown much less of the mythopoeic fruitfulness of psychoanalysis, inspiring remarkably little in the way of actual achievement in increased understanding; claimed achievements, over and over, are recognisable as such only to existing cultists, and celebrated as achievements precisely because they are thought to validate the preexisting theory rather than shed any new light on the real data.

    The (considerable) scientific value of Chomskyism is as study material for sociologists investigating pathologies of academic life.

  5. @John Cowan: I have always wondered whether “tenured grad student” really started out as a pun at all, or whether that was added as an additional layer of explanation after the joke was already in circulation.

  6. I wouldn’t bet either way.

  7. “Alleged misunderstandings or misinterpretations of the Chomskyan commitment and the Chomskyan framework are in large degree due to the imprecise and, at times, inconsistent formulation of its basic assumptions.”

    I have noticed a tendency in Chomsky to equivocate quite naturally and probably unintentionally, as if words were realities unto themselves rather than markers for fairly arbitrary categories, so that there is no need to attend to any possible shades of meaning or semantic discontinuities of the sort that equivocation exploits.

    “The (considerable) scientific value of Chomskyism is as study material for sociologists investigating pathologies of academic life.”

    Here’s a start on that: http://www.amazon.com/Linguistics-Wars-Randy-Allen-Harris/dp/019509834X

  8. David Eddyshaw says

    Genuine scientific theories and research programs are not routinely referred to by the names of charismatic founders. There is no General Theory of Einsteinism.

  9. J.W. Brewer says

    I note with interest the unselfconscious use of “minimalists” as pejorative shorthand for the Bad Guys. Of interest I suppose since as I understand it the whole “minimalist program” iteration of Chomskyanism didn’t really arise until approx. the last two decades. It must somehow or other be materially different from the prior iteration I encountered as an undergraduate in the ’80’s which in turn might have been different from the iteration hat encountered as a grad student in the ’70’s, and presumably is not wrong (or incoherent/non-falsifiable) for *exactly* the same reasons. But I don’t feel motivated to understand it well enough to know how the critiques of the earlier versions would need to be adjusted to fit.

  10. David Eddyshaw says

    I suppose one must in fairness admit that Chomskyism in its various iterations has in fact dallied with falsifiability and made some predictions about how language ought to work; in cases where this has happened, the predictions have been fairly consistently shown to be wrong. Minimalism seems be the response, retreating to an area where statements about language are essentially vacuous. The response to the Piraha brouhaha seems to boil down to saying that even if Everett (per impossibile) has described the language correctly, Minimalism would not be falsified. This may very well be true …

    I should perhaps own up to not being altogether persuaded that falsifibility a la Popper is quite the acid test for rigour that some think. There have been numerous valuable systems which at the very least have inspired productive research despite falling on the wrong side of that line. But Chomskyism seems rather to have spent decades diverting many very clever people away from actually addressing messy and beautiful linguistic reality.

  11. That’s what gets me about Chomskyism — what drew me to the field was exactly the messy and beautiful linguistic reality that so many brilliant people had managed to figure out how to describe in concise and helpful ways, and it boggles my mind that someone could come along and say “Nah, that doesn’t interest me, I’d rather play with logical structure and arrows and stuff.” I mean, if you’re not interested in language as it is (rather than the imaginary purified version you make up in your head), why not go into math or computers or something?

  12. Hat: Or economics.

    David E: Darwinism seems to be a counterexample.

  13. marie-lucie says

    Isn’t Darwinism a relatively recent coinage intended to be a derogatory pendant to Creationism? Neither of these two terms seems to be used seriously outside of the “Biblical Science” circles.

  14. Trond Engen says

    John C.: Hat: Or economics.

    Arguably the world would have been better off today if all post-Friedman Chicago economists had gone into linguistics (and that includes, at the very least, the post-Friedman Friedman). Given their relationship to Chomsky, it’s likely that even linguistics would have gained on the trade.

  15. J. W. Brewer says

    “Platonism” isn’t necessarily derogatory, is it? Maybe it doesn’t purport to be “scientific” in some narrow sense of the word . . .

  16. Trond Engen says

    marie-lucie: Isn’t Darwinism a relatively recent coinage intended to be a derogatory pendant to Creationism?

    I think it was a pendant to Lamarckianism. It must be at least as old as Social Darwinism, and that was coined in 1877, says Wikipedia.

  17. marie-lucie says

    Trond, you must be right. But from a pendanst to Lamarckianism, which seems to be mostly of historical interest, it has apparently become a pendant to Creationism.

  18. Newtonian mechanics is a thing, although I don’t know if anyone referred to it as such until non-newtonian mechanics was discovered…

  19. David Eddyshaw says

    Philosophical schools certainly do get named for their charismatic creators. Although they are not scientific theories, of course, I am very far from thinking that they are valueless. That was part of what I had in mind in saying that falsifiability is not coterminous with rigour.

    Economics seems to be one of those amphibious disciplines with characteristics partaking of both science and philosophy, so I can save my rash overgeneralisation in cases like Marxism and Keynesianism.

    I thought about “Darwinism” as an exception. I share marie-lucie’s feeling that the use of the term tends to suggest hostile polemic; though it’s also interesting that Darwinism itself doesn’t fit the Popperian falsifiability criterion without a bit of fancy footwork, a fact made much of by creationists, which annoyed Popper himself considerably. He described it as a “metaphysical research program” but certainly did not himself intend to impugn its scientific bona fides.

    “Newtonian mechanics” I concede. I now claim that my original assertion is still absolutely true, but at a more abstract level than in my original formulation.

  20. marie-lucie says

    I don’t think that “Newtonian mechanics” is ever referred to as “Newtonism”.

  21. David Eddyshaw says

    Ah! Precisely! proved right again.

  22. J. W. Brewer says

    “Gaussianism” is Out There if you do some googling, at least sometimes in what appears to be a pejorative context.

  23. David Eddyshaw says

    So it is. Not in the sense of “Disciples of Gauss” though. Now there’s a thought for a movie script. Steampunk conspiracy story? Eat your heart out, Dan Brown. I thought of it first!

  24. The early quotations forDarwinism in the OED3 are pretty much all from the writings of evolutionary biologists: Thomas Henry Huxley 1860, E. Ray Lankester 1876, A. R. Wallace 1889 (the title of his book), T. H. Morgan 1909. The word had an earlier sense, ‘the literary style or scientific views of Erasmus Darwin’, Charles’s grandfather. Erasmus was an evolutionist, like a minority of pre-Charles biologists, but did not propose a mechanism like natural selection.

  25. Googling for “newtonism” reveals a few gems bordering on examples of Poe’s law (poesism?), but mostly it is either the philosophical views of or ascribed to Newton or parodies of creationist arguments.

  26. In the old bad days of Lysenkoism the evolutionary theory (Darwinism) was not decried, but Weissmanism-Morganism (tah is genetics) was. Particularly mind boggling for me is the fact that W-M provided materialistic basis for rather abstract Darwin’s principles. But, I guess, if politics is at stake all reason goes through the window. There should be an -ism about that.

  27. Whether or not Popper was right about the key role of falsifiability, he was notoriously bad at understanding what theories are falsifiable. In particular, he initially did understand the use of statistical evidence in fields such as quantum physics.

  28. Brett, you probably meant that he did NOT understand etc. And, honestly, it is not particularly damning not to understand a thing or two about quantum mechanics, especially such a long time ago.

  29. George Gibbard says

    One thing Chomsky has going for him is that he admonished working linguists (a long time ago) to include negative data: so you should no longer just report ‘the Roman general destroying the city (was horrendous)’, ‘the Roman general’s destroying the city’ and ‘the Roman general’s destroying of the city’ are all correct nominalizations of clauses, you must also look at your data and go back and ask whether ‘The Roman general destroying of the city’ is also correct as a noun phrase (/determiner phrase) (it is not). Although, Chomsky has pissed me off (he even once told an interviewer that reports that he was looking for ‘Universal Grammar’ was ‘a common misconception’, unless the interviewer mischaracterized the exchange). And in The Minimalist Program there is little data, and frequent resorts to ‘it seems most parsimonious to assume…’

    I guess linguists have let the baby in with the bathwater.

  30. David Eddyshaw says

    Thinking about some of these examples, there’s an evident tendency for proper name +ism forms to be coined by opponents in debate. Perhaps this is actually a rhetorical attempt to *imply* that one’s opponents are really more of a personality cult than a respectable body, and trades on a preexisting sense that these forms are pejorative in a scientific context, with their implication that opponents have really nothing more to offer than the argument from authority.

    It’s probably also true that the actual term “Chomskyism” is used chiefly by those not altogether happy with the great man’s influence on linguistics. So while my original comment is probably very broadly valid, it most likely inverts cause and effect.

    So far as there is anything true in the idea that real scientific theories aren’t named after their inspired originators, this probably reflects the very nature of science: valid and fruitful ideas develop a life of their own as other investigators see new implications, devise corrections and improvements and so forth until it would be plainly ridiculous to label the entire field with one person’s name, however gifted. A real advance will be so fruitful that there is no way it will remain forever in the shadow of a guru.

  31. David Eddyshaw says

    And on the other hand, in those areas of intellectual endeavour where adherents of a particular school of thought proudly self-designate after the Founder, it probably means that they essentially subscribe to the view that intellectual progress is overwhelmingly the work of exceptional individual prophetic figures. I think this is actually the default view, and moreover quite often true; modern science is really quite distinctive in that in reality (as opposed to the minds of journalists) that’s not how it works.

  32. I remember years ago someone saying that she objected to being called a Chomskyite, “as if I were a member of some absurd religion”. So if -ism may or may not be insulting, -ite is probably more so.

  33. As for Newtonian, the OED traces it right back to 1676, the year after Newton’s first book on optics, Hypothesis of Light:

    Philos. Trans. 1675 (Royal Soc.) 10 491 A Reply to a Letter formerly printed by way of Answer to another of the said Mr. Linus, which relates to the Newtonian Theory of Light and Colours.

    As the journal title shows, however, science and philosophy were hardly separated at the time. We no longer speak of Newtonian optics except in connection with Newtonian telescopes, but we do have Newtonian mechanics and Newtonian fluids, sometimes in lower case, which is the ultimate accolade for a scientist or mathematician.

  34. And of course there’s platonic, though I’m not sure Plato would be happy with how it’s used now.

  35. David Eddyshaw says

    Just as no serious self-respecting heresiarch would ever describe his own views as anything other than “Orthodoxy”, so presumably any philosopher would hope to avoid the fate of his own views being pigeonholed by being simply labelled after himself. I would imagine Plato’s preferred term for his own views would be “Truth.”

    As for Platonic Love, the way the term gets used popularly now is a bit like calling the Thirty Years War “an extended period of lively religious discussion occasionally leaving some resentment in its wake.”

  36. But there a number of philosophical schools not named after their founders (positivism, pragmatism, existentialism, analytical philosophy, phenomenology, deconstructionism) and somehow there is not much difference in their outward existence with “named” schools like Marxist or Hegelian or Rawlsian.

  37. David Eddyshaw says


    True enough. Though several of your examples don’t immediately suggest a single name as the founder of the movement (many parents could be adduced for existentialism or analytical philosophy for example.) Some of them too, you might argue, have proved stimulating enough to other thinkers that they have been taken forward by them to a degree where to attach a single name to the school no longer seems particularly sensible (whatever unprintable things one might actually think about deconstructionism.)

    However, even if my (increasingly strained) thesis were correct (that movements named for their founders were more likely to have cultlike qualities and less likely to resemble scientific theories) it wouldn’t necessarily follow that the reverse was also true. After all it’s not hard to think of outright cults that *aren’t* usually named after their founders.

    Basically I revert, however, to the position that my thesis remains entirely valid at a sufficiently high level of abstraction …

  38. Counterpoint: In retrospect, “Mesmerism” was a much better name than “Animal Magnetism.”

  39. Greg Pandatshang says

    Just who are/were these people who have been promoting Chomskyan linguistics? I only ever seem to hear people criticising it. Are there really a lot them? Or there used to be? I must just be really good at assortatively exposing myself only to the side that I expect to be more interesting. It helps that Chomsky-influenced papers are largely unintelligible to me, whereas with normal linguistics papers I was able to find interesting stuff written pretty plainly and then gradually bootstrap up to understanding some more complex concepts. (I do see papers with Chomsky-esque notation sometimes, but these papers seem to have it as a background assumption; what I don’t see is people arguing for Chomskery).

    P.S. I noticed recently that, as I’ve gotten older, I’ve started to appreciate Chomsky’s political thought to an extent, which certainly didn’t used to be the case. But no movement on his linguistic approach starting to seem worthwhile (maybe it will once I get even older). Hat put it beautifully: “I mean, if you’re not interested in language as it is … why not go into math or computers or something?”

  40. David Marjanović says

    it’s also interesting that Darwinism itself doesn’t fit the Popperian falsifiability criterion without a bit of fancy footwork

    Insert here the obligatory discussion of whether the famous “rabbits in the Precambrian” would falsify the theory of evolution or rather constitute stunning evidence for time travel…

    No, seriously this time: the theory of evolution by mutation, selection and drift is being tested all the time. I have seen evolution in a petri dish full of bacteria happen overnight, as a routine part of an obligatory introductory course in molecular biology. All the way to Lenski’s large-scale experiment, the observations fit the predictions so routinely that we don’t even notice when a test has been performed anymore.

    The term “Creationism” is used all the time, at least by everyone who isn’t a creationist. “Darwinism” used to be, as mentioned above, used in contrast to Lamarckism; the term survived the demise of Lamarckism* and was still used by Richard Dawkins not long ago, but is now almost exclusively used by creationists, and even found insulting by many who are not because -ism is understood to indicate a political ideology.

    * Recent tales of Lamarckism being vindicated by epigenetics are greatly exaggerated.

Speak Your Mind