Chomsky’s Forever War.

Geoff Pullum has a typically invigorating review of the new second edition of Randy Allen Harris’s The Linguistics Wars: Chomsky, Lakoff, and the Battle over Deep Structure; here are a few excerpts:

It is quite difficult to explain in nontechnical terms what triggered the linguistics wars, but let me try. Chomsky in the mid 1960s maintained that the structures of sentences were nowhere near as simple as the sentence diagrams of yesteryear; they were more like se­quences or layers of such diagrams. The most concrete layer, called the surface structure, captured facts about the overt shape of the sentence (word order, inflec­tion, and pronunciation). More abstract layers accounted for things like con­nections between sentences — relations be­tween active and passive clauses, for example. […]

However, from about 1967 some of Chomsky’s earliest defenders and most talented students began to develop arguments undercutting his case for “Deep Structure.” They claimed it had no theoretical necessity or significance. Instead, they posited much more abstract syntactic layers, and suggested that the most abstract layer of all looked much like what philosophers would call the “logical form” of a sentence. Their hypothesis was dubbed “Generative Semantics” (GS). […] GS emerged directly out of Chomsky’s work, using the kind of arguments he used, but it diverged from his own views, and displeased him. In his view this meant that the promoters of GS had to be defeated and punished.

The punishments Chomsky imposes on publicly announced enemies of his views are of two types: the dungeon and the fire.

Consigning an opponent to the dungeon of ostracism and obscurity is easier than you’d think for someone of Chomsky’s extraordinary prestige. His close followers, who dominate the mainstream of syntactic theory in America, tend to cite only what he cites, so if he resolutely avoids mention of someone’s name, within the mainstream it will soon look as if neither that person nor their publications even exist. (Today, when citation counts affect hiring and pro­motion, that can hurt more than the victim’s pride.)

The citational oubliette worked well for lesser-known or younger GS proponents, especially from low-prestige or foreign institutions. For example, in half a century the extensive and highly relevant works of brilliant European theorists such as Pieter Seuren and Esa Itkonen were never referred to in Chomsky’s vo­luminous writings, be­cause he had identified them early on as enemies.

But more visible proponents, especially those close to Chomsky’s home institution, MIT, faced something more like burning at the stake. Chomsky would immolate their ideas in classes (scores of visitors would travel long distances to absorb his wisdom at MIT each Thursday afternoon), in conference presentations and invited public lectures around the world, and in the many articles and books he produced. He excoriated, misrepresented, and ridiculed his enemies, sometimes in their presence, calling their positions incoherent, irrelevant, or possibly just muddled equivalents (“notational variants”) of his own. He would attribute to his opponents views they rejected, even when they were present and tried to demur. […]

By around 1976, GS proponents could see that they had lost. GS had been crushed in an ideological struggle be­tween an intellectual superstar whose followers saw him as infallible and a foolhardy clique that had dared to depart from the true faith.

Some on the losing side, worn down by a decade of struggle, looked like has-beens, which meant they could be transferred to the other punishment mode: total denial of citation. George Lakoff was tough enough to take Chomsky’s attacks, shake them off, move to Berkeley, and start a new life writing about metaphor and politics instead. But Chomsky’s former star student John Robert Ross was cowed and sidelined, and largely ceased publishing on syntax (a pretext was found for forcing him out of his tenured faculty position at MIT). James McCawley continued writing GS-influenced works from his indepen­dent base at the University of Chicago but was increasingly ignored by the main­stream. […]

Harris quotes Shakespeare’s amusing vignette of Hamlet persuading the sycophantic Polonius that a certain cloud is shaped like a camel, then getting him to agree it is like a weasel instead, and then that it resembles a whale. True Chomsky followers show their Polonius-like loyalty through unquestioning acceptance of each new image Chomsky sees in the theoretical cloud. They adopt the buzzwords, modify the direction of their own work, and pursue the new line — only to be left high and dry eight to ten years later when once again he demolishes the central pillars of the previous conceptual edifice.

Another practice of Chomsky’s is to plunder the work of his opponents for attractive ideas and useful terms to be adopted without credit a decade or two after he first attacked them as misguided. In recent years he has mined the program of GS for insights without crediting the original developers. Around 1990 he completely abandoned his “Deep Structure” level without a word about this being the central proposal of GS in 1967. His “minimalist program” posits a unified rule system relating logical forms directly to surface forms of sentences, a position GS proponent Paul Postal explicitly advocated in 1969, when Chomsky hotly rejected it. Chomsky never mentions that adopting the phi­loso­phers’ term “logical form” for the most abstract level of sentence representation was George Lakoff’s idea.

It’s hard to stop quoting, but if you want more just click on through; I know I’ve bashed Chomsky a lot and quoted a lot of bashing by others, but I was traumatized by him and his minions at a formative age and can’t get enough. I got the link via Mark Liberman’s Log post, where David Marjanović left this comment:

The latest shot in the linguistics wars: an open-access paper on how language can indeed be learned from input alone.

Venceremos!

Comments

  1. jack morava says

    Barely passing Calc II (in particular, essentially flunking integration by substitution, x –> f(t)) traumatized young Noam, making transformations a recurring theoretical obsession. It’s sad that he never got over it.

  2. David Marjanović says

    I posted that paper in the latest Chomsky discussion quite recently, too… OK, probably in December…

    It’s sad that he never got over it.

    He passed, and he was still traumatized? Did he, like, need straight As for a scholarship or something?

    Where I come from, the first thing you… well, the last thing you learn in your first semester is to completely stop caring about all non-failing grades. The professors are allowed to grade as they damn well please, so if you pass, you can simply move on.

  3. I barely passed organic chemistry in my first undergraduate year, and while I wouldn’t say I was traumatized for life, the bitter memory lingers. My recourse is to assert that organic chemistry is fundamentally incomprehensible.

  4. It’s remarkable that the WP article on Chomsky barely hints that his linguistic work has been so polarizing: i.e. that significant numbers think him godhead, and that significant numbers think him worthless.

  5. David Marjanović says

    My recourse is to assert that organic chemistry is fundamentally incomprehensible.

    I’m afraid you just had a bad teacher. Did you have to learn lots of name reactions?

  6. David Eddyshaw says

    I’ve missed Geoffrey Pullum …

  7. David Eddyshaw says

    The only public examination I have ever failed was the Primary examination of the Royal College of Surgeons (in those days, something under 15% of candidates passed in each session.)

    I knew a Professor of Surgery who remarked that this was quite typical: most UK candidates for the Primary have similarly never actually failed a public examination in their entire lives until they sat the Primary, which they then duly failed because everybody does, the first time. The psychological effect can be quite profound.

    He (the Professor of Surgery) had, however, failed every single annual examination as an undergraduate at least once (this was more recoverable-from, academically, in those days than it now is.) When he duly failed the Primary, he simply took it in his stride, and buckled down to revise for the next session …

  8. Did you have to learn lots of name reactions?

    There was a good deal of that. What I mainly remember was that the subject seemed to consist of trying to remember long lists of what molecular group liked to react with some other molecular group, and how they preferred to rearrange themselves under specified conditions, and why they would shun other groups, and so on.

    I have to admit that I had a friend who was good at organic chemistry, and could figure out what would happen when two complicated molecules were introduced to each other. But I could never get the hang of it.

  9. invited public lectures — enviably fluent in style,

    I’ve not attended any Chomsky lecture; I have watched some on Youtube. ‘Fluent’ is not a word I would use. His style has become increasingly impenetrable.

    Like he’s hearing himself getting dangerously close to making a testable claim; so smothers it in caveats and qualifications to the point he fails to convey any sense at all. (Or perhaps Pullum’s ‘fluent’ applies to the production of subjunctives and counterfactuals?)

    His political lectures and interviews are worse: a reciting of conspiracy theory hashtags with no concrete evidence, aimed to evince a knee-jerk reaction that all governments are evil. Really no different in nature/’style’ to Breitbart News.

    precise description of languages (which once figured in linguists’ job descriptions)

    I’m unconvinced Chomsky would recognise a human language, let alone be able to describe one. I’d be interested to know what sort of linguistic input he’s providing to his granddaughter, or her kitten. Probably not something from which anyone could learn a language.

  10. David Eddyshaw says

    His style has become increasingly impenetrable

    When you’re selling snake-oil, clarity is not your friend …

    I suspect that in years to come, the value of Chomsky will be largely as an Awful Warning of how an entire discipline can be seriously damaged for years by a personality cult.* It should provide plenty of material for sociologists to write theses on, anyhow. (I like to look on the bright side,)

    * Thanks to Geoffrey Pullum for the precise formulation. Extraordinarily (in retrospect) I’ve never quite thought of it as that, exactly. But it’s the mot juste. Spot on.

  11. how an entire discipline can be seriously damaged for years by a personality cult

    There’s Freud, too. There are a lot of parallels: both opened new directions in their fields; both had a lasting influence because their theories were unfalsifiable. Freud, however, was not so full of himself as Chomsky, from what I can tell.

    I won’t say anything about economic theories because I don’t know much about the field, but I suspect you could find parallels there, too.

  12. AntC : “I’m unconvinced Chomsky would recognise a human language, let alone be able to describe one. I’d be interested to know what sort of linguistic input he’s providing to his granddaughter, or her kitten. Probably not something from which anyone could learn a language.”

    Do you think Steven Pinker would, or Daniel Everett? Or Dennet?

  13. David Eddyshaw says

    Freud is actually both entertaining and accessible as a writer (wrong, too, but you can’t have everything.) On the other hand, attempting to actually read Chomsky is a considerable trial, aesthetically; I imagine the groupies find this to be the case as well, but feel that they are purified and ennobled by the suffering. A sort of rite of passage, maybe, like the Fulɓe soro or the Comanche Sun Dance (I’m pointing the way for future sociologists here! Give me a credit in your references, guys!)

    Pinker is a capable writer. The irritation his works produce is of a quite different kind.

    I’ve not attempted to read Everett at any length, and have no worthwhile opinion on his abilities as a descriptive linguist; however, at least he has actually done significant language fieldwork, of a kind not possible from an office at MIT.

    Dennett has an excellent beard. These things matter. I think I speak for many Hatters in this.

  14. Daniel Everett certainly would.

  15. Pinker is a very capable writer. Freud, not so much. Chomsky is an _interesting_ writer.

    EDIT : David Eddyshaw : As in “May you live in interesting times”?

    I Guess?

  16. David Eddyshaw says

    Chomsky is an _interesting_ writer

    As in “May you live in interesting times”?

  17. I barely passed organic chemistry in my first undergraduate year, and while I wouldn’t say I was traumatized for life, the bitter memory lingers. My recourse is to assert that organic chemistry is fundamentally incomprehensible.

    Substitute physics for organic chemistry and the same is true for me.

    I have to admit that I had a friend who was good at organic chemistry, and could figure out what would happen when two complicated molecules were introduced to each other. But I could never get the hang of it.

    Again the same is true for me (with the same substitution); in fact, he was the only reason I (and other hapless freshmen as ignorant as myself) were able to pass — every Thursday night we assembled in his dorm room and explained to us the homework assignments due Friday (which were a large enough percentage of the grade to allow us to pass the course). He was large, genial, and Mormon, and after that year he had to go off and do his years of missionary work; I’ve always wondered what became of him. I hope life has been as good to him as he was to us.

  18. David Eddyshaw: “Freud is actually both entertaining and accessible as a writer (wrong, too, but you can’t have everything.)” Maybe like H. P. Lovecraft? It reminds me of him.

  19. David Eddyshaw says

    It’s not an analogy which has ever occurred to me … I shall have to get back to you on that one …

    (Possibilities for crossover fanfic suggest themselves surprisingly readily, however …)

  20. The eldritch horrors of the Id…

  21. (…the Deepest Structure of all…)

  22. I’m very happy I got your muse going 😀

  23. I won’t say anything about economic theories because I don’t know much about the field, but I suspect you could find parallels there, too.

    I think the parallel between Chomskyan linguistics and neoclassical economics is the extreme abstraction from real observable reality in favor of grandiose theorizing. As well, both probably the textbook case of physics envy, using mathematics or mathematical-sounding terms to give the appearance of serious rigor. Although the economists are far more serious about the mathematics; so your typical economics undergraduate needs several math classes, while the linguistic major need none.

  24. @V Do you think Steven Pinker would, or Daniel Everett? Or Dennet?

    Curious rag-bag of thinkers you list. “… would recognise a human language” I presume you mean. Then only one of those claims to have applicable skills.

    (Thanks @Y for that demo, which I’ve seen before.) Then yes I’ve no doubt Everett can and does recognise and describe human languages. (His writing style is maybe rather pedestrian. This adds credibility in my mind.)

    Dennett can and does recognise, describe and critique systems/structures of thought. I don’t see him making claims about language-specific genes, or that linguistic abilities are qualitatively different from general human cognitive abilities.

    Pinker I haven’t paid attention to.

  25. I love me a good Chomsky bashing, but I’d be weary of lending credence to the Yang & Piantadosi paper, which is mostly attacks of straw-men (read it; or if you’re not up to it, check the quote-tweets off P’s announcement of the paper. The counterarguments are very convincing and not necessarily Chomskian).

  26. Dennett can and does recognise, describe and critique systems/structures of thought. I don’t see him making claims about language-specific genes, — you mean memes?

    “or that linguistic abilities are qualitatively different from general human cognitive abilities.” — that’s what the word “meme” means? Cognitive patterns in the brain that reduplicate in the manner that genes do? At least that’s what it meant twenty years ago.

  27. I mean genes, contra “Our language capacity must be due to some unidentified universal genetic inheritance. … He believes that a “mu­tation” in a single human a few tens of thousands of years ago sparked an un­precedented new cognitive ability.” [Pullum summarising Chomsky].

  28. Of the names mentioned in that extract, John Robert Ross is one to conjure with. The most interesting and counterintuitive insight of generative grammar – the discovery of “islands” – is his work. I’d like to learn more about his post-generative career.

    Everett shares with Chomsky an ability to make maximally grand, provocative claims and then dial them down as needed long afterwards. His paper that originally sparked the whole Piraha brouhaha is a fascinating read, if approached with a critical eye.

  29. David Marjanović says

    Speaking of which, here’s the open-access paper from 2016 with Everett as the 3rd of 5 authors, concluding that it’s possible but still quite unclear whether Pirahã has recursion (in a more meaningful sense than “sentences with more than 2 words”).

    Abstract:

    The Pirahã language has been at the center of recent debates in linguistics, in large part because it is claimed not to exhibit recursion, a purported universal of human language. Here, we present an analysis of a novel corpus of natural Pirahã speech that was originally collected by Dan Everett and Steve Sheldon. We make the corpus freely available for further research. In the corpus, Pirahã sentences have been shallowly parsed and given morpheme-aligned English translations. We use the corpus to investigate the formal complexity of Pirahã syntax by searching for evidence of syntactic embedding. In particular, we search for sentences which could be analyzed as containing center-embedding, sentential complements, adverbials, complementizers, embedded possessors, conjunction or disjunction. We do not find unambiguous evidence for recursive embedding of sentences or noun phrases in the corpus. We find that the corpus is plausibly consistent with an analysis of Pirahã as a regular language, although this is not the only plausible analysis.

  30. Athel Cornish-Bowden says

    I disliked organic chemistry as a undergraduate, but I had an excellent tutor who attached more importance to mechanisms than to lists of reactions, and I ended up doing my DPhil in an organic department!

  31. I also barely passed organic chemistry, partly due to a vagueness between e.g. Friedel-Crafts and Diels-Alder. My response was to become an English major, a decision I’ve never regretted.

  32. Somewhere between loving Carlos Quícoli’s introductory linguistics course when I was a freshman and being stymied into two Cs in a row in Maitland Jones’s organic chemistry courses as a sophomore, I probably should have figured out what my passion was.

    But I was hardheaded and afraid of winding up in a field with no career prospects, so I squeaked out a chemical engineering degree (two, actually) by the skin of my teeth and have spent over twenty years as a programmer instead.

  33. I loved learning languages at school (I mean high-school equivalent) and I actually got better reports from my teachers than I did in sciences. But I had no idea what an undergraduate degree in languages would entail, and I was only vaguely aware that a subject known as linguistics existed. And I certainly had no idea what sort of career such a degree would lead to.

    So I studied physics, which I enjoyed and which led to an interesting life. But now that I am retired and occasionally given to fits of nostalgia, I sometimes wonder where I would have ended up if I’d studied languages instead.

    Another dream, of course, was playing cricket for England, but I realized quickly I wasn’t good enough.

  34. David Eddyshaw says

    @Lameen:

    Yes, the “islands” thing has always struck me as about the only real linguistic discovery that the Chomsky school has ever contributed. I hadn’t realised that its architect had been drummed out of the Brownies.

  35. Looking up Quícoli and Jones, this is the point when I also realize my professors are octogenarians, and my high school teachers are largely deceased. Maybe time to have a midlife crisis (assuming I live to 106 or so).

  36. Lars Mathiesen says

    @David L, I was also relatively good at languages, and fascinated by historical linguistics. Only computers more so. I might have been happier in a linguistics department than trying to wedge Configuration as Code into 1990s mindsets, but the money is better. (And I get to do enough of the fun stuff).

  37. Stu Clayton says

    recursion (in a more meaningful sense than “sentences with more than 2 words”)

    Yeah, otherwise what you say right there would be recursive. Non-mathematicians/-programmers who love to talk about recursion seen to have no notion of infinite recursion. Perhaps they would imagine that’s what someone gets into when he can’t stop talking. That’s not primarily a linguistic issue, though, so I suppose that’s why they don’t think about it.

  38. Non-mathematicians/-programmers who love to talk about recursion seen to have no notion of infinite recursion.

    Paul Postal wrote a whole book – I stupidly got rid of my copy – arguing for the grammaticality of transfinitely long sentences. Merely countable infinity was apparently not enough.

  39. Stu Clayton says

    Long ago I read about someone’s work on transfinite sentences/formulas (wff) in logic. Maybe that was Postal. I bet there aren’t *that* many people hooked on the topic. As they might say over here about results in such a field, dafür kann ich mir nichts kaufen.

  40. jack morava says

    re cult of personalities, cf Trofim Lysenko?

    May I say that co-evolution of language, considered as a virus, and the brain, following eg Terrence Deacon, seems like a good idea to me…

  41. But Lysenko was imposed from above; Chomsky propagated himself.

  42. @jack morava: I don’t think there was really much cult around Lysenko personally. He was just the most forceful and successful of those who put forward “theories” of biology that were designed to make Marxism-Leninism look “scientific.” As evidence that Lysenko was not really venerated so much as an individual, there are writings from the 1940s that say, more or less: Lysenko is right, and we know that because Stalin says so. By 1954, real biologists were quietly moving back in to take over Soviet agricultural development, although Lysenko was not publicly repudiated. This led, unfortunately, to Lysenkoism, while it was quietly being abandoned in the Soviet Union, being exported to China, where it played a role in the Great Leap Forward. (In fact, there is, I think, a clear parallel between Lysenkoism in the Soviet Union and the later Great Leap Forward. The butchers really seemed to believe that they could deduce scientific facts and policies from a basis of Marxist tenets. Moreover, when things did not go as planned, the erroneous policies were quietly and slowly tailed off, because that was easier to do without admitting that the underlying ideas had been totally erroneous. How many people died in the last couple years of the Great Leap Forward, just so that the Great Helmsman could save face?)

  43. @David Eddyshaw : What I read of Freud as a teenager was so hilariously wrong I thought it was an elaborate joke, initially. Sort of like Lysenkoism, come to think of it, but wronger. But it was, at the time, puzzling/entertaining/offputting. Mostly puzzling. Not so entertaining if you were a trans person, at that age, reading him, I would imagine — and I don’t need to imagine, trans people who read him also as teenagers have told me it was in no way entertaining, and I don’t find it entertaining now either. But even then, I found it puzzling why he would think those things about gender.

  44. David Eddyshaw says

    Sure. I hold no brief whatsoever for Freud’s actual views on anything. I also share the Popper/Lakatos* etc conception of Freudianism as a poster-child for not-science. It doubtless also helps appreciation of his literary virtues quite a bit if you are a cis male heterosexual. Still, I persist in thinking he’s much better qua stylist than Chomsky. Not a high mark to aim at …

    * Sadly, these gentlemen’s floruit was too early for full-blown Chomskyism to take its place in their gallery.

  45. I persist in thinking he’s much better qua stylist than Chomsky.

    I do too, though I confess I’ve never tried to read him in German. He’s no Nietzsche, but his idiotic and offensive ideas are expressed in perfectly digestible fashion.

  46. jack morava says

    I agree that Chomsky never sent anybody to the camps, but [wikipedia]

    …In 1964, physicist Andrei Sakharov spoke out against Lysenko in the General Assembly of the Academy of Sciences of the USSR:

    He is responsible for the shameful backwardness of Soviet biology and of genetics in particular, for the dissemination of pseudo-scientific views, for adventurism, for the degradation of learning, and for the defamation, firing, arrest, even death, of many genuine scientists…

  47. Nobody’s arguing with that.

  48. I am now curious about whatever Russian word Sakharov used that’s translated as “adventurism.” Is it being used in approximately the same sense as in sentences such as (from a 1976 article in New Left Review) “The adventurism of the KPD in 1921 was condemned by the Third World Congress of the Comintern”? Interesting to see the phase when Sakharov has not yet officially quite become a “dissident” so he is using Communist rhetoric to attack Lysenko as a Bad Communist who made it more difficult for the Party to pursue the correct line in these matters.

  49. Lysenko

    Лысенко был непрост и неоднозначен. Он — агроном «от Бога», чувствовавший землю и болевший за нее. Его агрономические рекомендации, пока он не выходил за рамки «смешивания суперфосфата с навозом» или яровизации, работали, и крестьяне им охотно следовали. С другой стороны, он настойчиво добивался запрета атомных, да и всяких иных, взрывов, так как, по его мнению, живая природа отторгает радиоактивность как чуждое всему живому, естественной природе. Он верил, что «Земля — живая, от взрывов она потеряет способность родить, испугается навечно, и все живое погибнет».

    https://biography.wikireading.ru/53015

  50. juha, would you please translate into English? most people who post here probably know who Lysenko is but future readers might not.

  51. Lysenko was complicated and ambiguous. He was an inspired agronomist, who felt the land and cared about it. His agronomic recommendations, as long as he did not go beyond “mixing superphosphate with manure” or crop rotation, worked, and peasants willingly followed them. On the other hand, he insisted on banning atomic and other explosions because, in his opinion, living nature rejected radioactivity as alien to all living things, to nature. He believed that “the Earth is alive, but it will lose its ability to give birth from explosions, it will be frightened forever, and all living things will perish.

    future readers might not

    Then it might well be another round of history repeating itself, just with different actors.

  52. That appears to be the DeepL translation. The Google one is scarcely comprehensible, ditto Apple’s brand new service.

  53. PlasticPaddy says

    Public criticism of Lysenko was bad for one’s physical and mental wellbeing:
    В пятницу 29 мая 1970 года в доме Медведева в Обнинске появился доктор и несколько
    сотрудников милиции в штатском и забрали ученого в психиатрическую лечебницу в Калуге. Главный врач-психиатр Александр Лифшиц заявил, что одержимость Медведева фигурой Лысенко выявила психический дисбаланс.
    Source:
    http://madan.org.il/ru/news/zhores-medvedev-i-bitva-za-pravdu-v-sovetskoy-nauke

  54. непрост и неоднозначен

    Useful words to know !

  55. David Eddyshaw says

    They may or may not be. It’s a difficult question.

  56. Even in My Day, Freud was mainly touted by people who supposed he was the first person in history to realize that we sometimes act from motivations that aren’t perfectly transparent to our narrative consciousness. And Jung was mainly touted by people who supposed that the only alternative was Freud.

  57. 1. I’ve never tried to read Freud in German but am told he’s a quite good prose stylist and his English translators generally made him sound somewhat more ponderous and turgid than he comes across in the original because of their conception of the right English register for a Serious Scientist or what have you.

    2. Just re marginal intellectual movements and “personality cults” I was struck by an interesting feature of the history of the Objectivist (i.e. the theories of Ayn Rand and her followers) movement. Six or seven years after Rand’s own death, there was a major schism, in which one faction (the Kelley side) tried to separate the intellectual content of the movement from its personality-cult aspects. He and his associates were duly purged and cast into outer darkness by the official guardians of the legacy, but were numerous and well-resourced enough to create a rival institutional structure that endures to this day. No doubt there are plenty of individual linguisticians who publish on syntax in a Chomskyesque framework but try to avoid getting sucked into the personality-cult dynamic, but I don’t know if they ever (at least following the demise of the original GS folks) formed any sort of organized movement or faction with the overt slogan “here’s Chomskyanism without the personality-cult baggage.”

  58. David Eddyshaw says

    I think the problem is that without the cult of Chomsky himself, there isn’t really enough there intellectually to keep up the momentum. It seems unlikely that very much of it will actually survive its founder for long.

    The Objectivist thing is a bit different, as there is always a market for “respectable” reasons to say that shameless greed and an inhuman lack of compassion are Good. It generates its own momentum.

  59. J.W. Brewer says

    David E.: you may be correct, but unlike Objectivism there are still some fair number of doctrinaire Chomskyists embedded in academic institutions and tenured who are considerably younger than their nonagenarian guru. What are they going to say/teach/publish in the years/decades between the guru’s occultation and their own retirements? Who are they going to support or oppose as fresh-out-of-grad-school hires in their departments?

  60. David Eddyshaw says

    Good points …

  61. David Marjanović says

    I’ve never tried to read Freud in German but am told he’s a quite good prose stylist and his English translators generally made him sound somewhat more ponderous and turgid than he comes across in the original because of their conception of the right English register for a Serious Scientist or what have you.

    I’ve never tried to read any serious amount of Freud in any language, but the most famous example of this phenomenon are the terms that got translated as Id, Ego, Superego: that’s exactly what Freud opposed. His goal in terminology, AFAIK, was that it should all be self-explaining rather than sounding scientific, and so he went with the entirely plain Es, Ich, Über-Ich.

    (Naturally, I don’t think that works terribly well either; repurposing everyday words as carefully defined technical terms will always run the risk of being confusing or misleading. But that’s another story.)

  62. Discussed here:

    It’s the same mentality that chose to render Freud’s Besetzung by “cathexis,” Fehlleistung by “parapraxis,” and Ich by “ego.”

  63. David Eddyshaw : I sort of share your lamentation that Imre Lakatos and Karl Popper did no live long enough to witness Chomsky’s star career, but, at the same time I am glad they did not.

    Stu Clayton: неоднозначен means just ambiguous.

    All of that reminded of Tiny Carl Jung from Dresden Codac.

  64. One thing I got from the Pullum piece:

    Chomsky posited an analysis of English in which a maximally abstract structure, for which he used Charles Hockett’s term “Deep Structure,”

    I didn’t know he’d filched that.

    If you google “Chomsky and Hockett”, a few interesting looking results come up. The first one is Charles Hockett Critiques Noam Chomsky, notes on Hockett’s The State of the Art (1968). It goes into things like:
    Chess can be characterized as a formal system.
    Baseball can be characterized as a formal system.
    American football can not be characterized as a formal system.
    American table manners can not be characterized as a formal system.

    Hockett provides an example of a physical system (not well defined) and the well-defined notational system which is used to describe it: the relationship between hydrocarbon molecules of the methane series, and the structural formulae which describe the hydrocarbons. (The fact that you can write down the formula for the case of n=1,000,000 or more does not actually mean that these hydrocarbons exist.)

    Pullum’s article is quite personal in a number of ways. Much is devoted to Chomsky’s personality and behaviour, including his various nefarious tactics (including bare-faced lying) and the fact (which others have pointed out) that Chomsky is ruthless in his determination to always be the winner in arguments.

    Pullum also claims that Chomsky has “few if any widely accepted empirical discoveries about human language to his name”. I agree with David Eddyshaw that “islands” are one of the only true discoveries that generativists have made — Haj Ross, of course, not Chomsky.

    So Pullum is definitely there to bury Chomsky, not to praise him.

    My own acquaintance with Generative Semantics in the 1970s was fleeting but distinctly negative. The claim that “kill” should be derived from “cause to die” struck me as ridiculous. Didn’t they know the difference between 殺す korosu ‘kill’ and 死なせる shinaseru ‘allow or cause to die’ in Japanese? This basically put me off generative grammar. Ironically, then, I was emotionally a follower of Chomsky and his deep structure and an opponent of Generative Semantics.

    The Language Log article was interesting for Liberman’s statement that:

    “But in fact Noam’s fundamental allegiance is to rationalism — he believes that the aspects of language that interest him don’t have to be learned, not because they’re somehow coded in the genome, but because they’re a necessary consequence of the logic of the situation (maybe along with one little mutation giving rise to “recursion/merge” or whatever).”

    This has always seemed like the crux to me: In what meaningful way is Chomsky claiming that his formal analysis of language has any relevance to how people use language.

    Our own DM responded with a link to One model for the learning of language, to which one DJL responded: “Worth noting that that paper is about extrapolating formal grammars from strings of elements, and thus, very possibly of rather limited relevance to the study of the acquisition of natural language, as has been noted ad nauseam over the years (it constitutes a rather narrow and technical point, and perhaps an interesting one, but not one that directly relates to the study of natural language).”

    I think I agree with DJL in this (although by saying this I don’t mean that I disagree with DM). When Chomsky uses examples like “Is the man who is tall happy?”, I always find it faintly ridiculous that Chomsky resorts to positing other completely mechanical possibilities of forming questions (e.g., move the first encountered verb to the head of the sentence) that are not used by human languages. This is supposed to show that there is “structure” in sentences — but didn’t we already know that?

  65. David Eddyshaw says

    The claim that “kill” should be derived from “cause to die” struck me as ridiculous

    True.

    It kinda-works in Kusaal, not in that “kill” can be synchronically derived morphologically from kpi “die” (though it probably was, in some distant protolanguage before Proto-Oti-Volta); however, it’s one of a relatively small set of verbs which has to have a direct object*, the majority of which are morphologically-derived causatives or applicatives (along with stative verbs expressing relationships, like mɔr “have.”)

    * In the sense that if one isn’t actually present, the sense is necessarily anaphoric (“kill him/her.”) “Thou shalt not kill” has to be expressed as “Thou shalt not kill anybody.”

  66. Just a quick shoutout for Charles Hockett:

    … Those who said that Boas was the first, and Kroeber the last, general anthropologist will now have to take it back and make room for Hockett on that pedestal…

    Man’s Place in Nature by C. F. Hockett, review by: D. L. Olmsted, Language, Vol. 52 (1976) 524 – 526, Linguistic Society of America, Stable URL: https://www.jstor.org/stable/412582

  67. The claim that “kill” should be derived from “cause to die” struck me as ridiculous

    Depends on the language, doesn’t it?

    ölmek

    Turkish
    Etymology

    From Ottoman Turkish اولمك‎ (ölmek, “to die”), from Proto-Turkic *öl- (“to die”). Cognate with Old Turkic 𐰇𐰠‎ (öl-, “to die”).
    Pronunciation

    IPA(key): [œlˈmɛc]
    Hyphenation: öl‧mek

    Verb

    ölmek (third-person singular simple present ölür)

    1. to die

    öldürmek

    Turkish
    Etymology

    Causative form of ölmek (“to die”).
    Verb

    öldürmek (third-person singular simple present öldürür)

    1. to kill, to murder (with accusative case)

    Onu öldürdüler.

    They killed [him/her/it].

  68. I’ve modified my views since the 1970s. However, I still wonder about the Generative Semanticists’ attempt to directly link “die” and “kill” together, given the many nuances involved. Causativity is not a simple area.

    My rather fond views of Transformational Grammar took a severe beating when I discovered Principles and Parameters in more recent times. Chomsky pulled the rug out from under TG in favour of theoretical simplicity and purity, but I personally find P&P and its mechanics clumsy and ugly.

  69. I didn’t know he’d filched that.

    The idea that most appealed to me in Syntactic Structures — after which everything went downhill — was ‘transformations’. Which it turns out he filched from Hjelmslev.

    So Pullum is definitely there to bury Chomsky, not to praise him.

    “few if any empirical discoveries” seems to me just the facts. Then this is more of a sky-burial.

  70. I was surprised at your statement that Chomsky filched transformations from Hjelmslev, so I checked with Wikipedia (“Transformational grammar”), and sure enough, it says “Transformational algebra was first introduced to general linguistics by the structural linguist Louis Hjelmslev”. The source cited for this is Pieter Seuren’s Western linguistics: An historical introduction but no page number is given.

    I have a digital copy of Seuren’s book (unfortunately not searchable) and was unable to find anything there to back up the Wikipedia assertion. I’ve asked the editor in question to clarify where in Seuren’s book that particular statement comes from. (Incidentally, Hjelmslev’s father was a mathematician famous for discovering the Hjelmslev transformation, which has nothing to do with linguistics.)

    Do you have a source that backs the Wikipedia statement up? Seuren’s book isn’t dismissive of Hjelmslev, but it certainly makes some play of the fact that his Glossematics never really became a theory of language, being stuck in the preparatory and programmatic stage.

  71. Do you have a source that backs the Wikipedia statement up?

    Ha! Good on you for questioning sources. My source is … Wikipedia on ‘Syntactic Structures’, which cites Seuren’s book again. Hmmm

    Hjelmslev’s Prolegomena is online (in the Whitfield translation). I’ve scanned it quickly. Neither ‘grammar’ nor ‘syntax’ appears in the ‘Register of Defined Terms’. (Neither is the pdf searchable, nor copy-pastable.) But ‘syntax’ does appear on p73, just after:

    We shall ordinarily, even at the stage of analyzing sentences, be able to reach, not merely a resolution of complex sentences into simple clauses, but also a reduction through the whole inventory of a given primary and a given secondary clause to one clause with both functional possibilities. … a specific word order in certain sorts of secondary clauses may be registered as a /signal/ for those clause-variants and thus does not prevent the reduction from being carried out.

    There would then follow an example analysing the primary and secondary clauses of ‘John is easy/eager to please’. Except Hjelsmslev doesn’t do examples nor phrase structures nor sentence diagrams. There’s a few examples for phonemics; fewer for lexicon; hardly any for morphology; none I can find for whole sentences/clauses.

    In that quote I can’t see anything I’d call a ‘transformation’. Rather more like ‘deep structure’. Perhaps there’s other papers showing ‘reduction’ in practice?

    So Chomsky has a hot contender for impenetrability. Hjelmslev is full of formalistic category-building, leaving me unconvinced he knows how to describe an actual human language.

  72. Depends on the language, doesn’t it?

    That may be the most un-Chomskyan thing I’ve ever read.

  73. It sounds like Hjelmslev could be characterised as “generative”, but “generative” is not the same as “transformational”. From what you quote it’s certainly hard to say. I’ll wait to hear what the Wikipedia editor has to say.

  74. I have no commission in this war. (I did hear NC speak in my Brandeis years 1968-72.) Geoffrey Pullum is a lively and skilled writer, one I wish wrote more often. I like his Linguistics: Why it Matters. (That Lingua Franca, Chronicle HE, ended his column was unfortunate, to put it mildly.)
    It’s been years ago, so my memory might be faulty, but Chomsky criticizing BF Skinner’s behaviorism means, to me, that he is, at a minimum, if I may say so here, not all bad.

  75. That’s the lowest imaginable bar. And nobody is “all bad” (no, not even him, he liked his dog); the point is that he was a shitty linguist and made linguistics a shitty field for decades.

  76. The focus on cultist-Chomskyite persecution of deviant/heretic Chomskyites is a bit of a distraction from the perhaps more important task of lamenting how non-“generative” perspectives got sidelined/marginalized in the discipline. Both “sides” in the headline-grabbing “war” were wrong, and both deserved to lose. That the GS crowd may have been nicer people and/or more willing to give up on GS when it became clear that it did not actually work very well as an empirical matter speaks well of them in a comparative sense, but …

    Sometimes I feel bad for being such a distracted and inattentive and all-around lousy student in the “Syntax I” class I took in 1985, which probably had a higher quotient of explicitly-Chomskyan content than any other class I took as an undergrad. But then I reflect that the majority of what poor Larry Horn was unsuccessfully trying to teach me was less than a decade away from being jettisoned from Chomskyite Central with the roll-out of the Minimalist thing, so it’s like I got a (deservedly!) bad grade for not mastering the then-current teaching about the properties of phlogiston.

  77. Yes, “teaching about the properties of phlogiston” is exactly how I regarded (and described) my enforced class on the subject. That and the similar requirement for a class in dialectical materialism over yonder in the USSR.

  78. David Eddyshaw says

    Surely the intellectual discipline of learning about the properties of ph(ono)logiston is valuable in itself? People have such a utilitarian view of education these days.

  79. @David: I was already dreadfully utilitarian at that age in the sense that I had the year before abandoned philosophy for linguistics because of increasing suspicion that the things Hegel was banging on about did not actually exist in even the semi-abstract way that grammatical categories and patterns exist and that non-existence of the object of the study was actually a drawback in an academic discipline.

  80. David Eddyshaw says

    @JWB:

    I may be underestimating the damage potential of phlogiston theory in my reflex condemnation of utilitarianism.

    I once marked an undergraduate multiple-choice paper (this was at such a primeval epoch that such marking was still done by actual people) on which the candidate would have received a higher score if he had answered completely at random. He demonstrated actual negative knowledge. It seems possible that Aspects of Chomsky may fall into this category.

  81. David Eddyshaw says

    I was reading somewhere (unfortunately can’t remember just where) an account of two colleagues talking about how Charles Hockett had been off for a few months over the summer creating the first adequate descriptions of a couple of indigenous North American languages. Said one: “Apparently, it’s just the sort of thing he does …”

  82. @languagehat : not quote comparable, but sort of similar, “history of the CPSU”. At my first history lesson in school, the teacher wrote out “ВОСР” at the blackboard, with great gravitas, as if it was the most important even in history, which left me extremely puzzled. Huge letters that covered most of the 1 by 2 meter blackboard. That’s how indoctrinated some history teachers were.

  83. For non-Russophones, that’s Великая октябрьская социалистическая революция (Great October Socialist Revolution).

  84. It sounds like Hjelmslev could be characterised as “generative”,

    From that quote, all I see is “analyzing … reduction”. So not ‘generating’ sentences from a finite set of rules, only reverse-engineering(?) from finished sentences, by recognising a finite set of patterns. Even in that characterisation, I’m guessing beyond Hjelmslev’s (translated) words.

    I lament that the empirical matter not getting deserved attention is understanding/parsing. (It’s not as easy as supposing listeners just replay the generation steps, to figure out how the speaker got there.) The structure of a language must be such that most listeners most of the time for most sentences can get to an unambiguous meaning. ‘Transformations’ or structure-shuffling from ‘deep’ to ‘surface’ royally screw up parsing strategies. Pullum’s own (with Gazdar) GPSG at least tries to generate in left-to-right surface order.

  85. In Western linguistics: An historical introduction, Seuren claims among other things that Hjelmslev realised that an algorithmic description of a language should generate an infinite number of products from a finite number of primitive elements. In support of this he quotes from page 26 of the Prolegomena. (I won’t quote this here; if you have it you can check it. It is the paragraph that starts: “When we compare the inventories yielded at the various stages of the deduction, their size will usually turn out to decrease as the procedure goes on…. “)

  86. Hmm page 26 are you sure? Is this of the Whitfield translation, or of the original Danish? Or is it Seuren’s own translation? The Whitfield translation is annotated with the original page numbers of the 1943 Danish publication. It was first published 1953 (a little before Syntactic Structures) “as Memoir 7 of the International Journal of American Linguistics (with what pagination?) — which presumably Chomsky would have seen.

    I’m looking at camera copy of the 1969 reprint of the 1961 book-form publication. Curiously, there’s what look like library stamps in Hebrew, plus a whole load of mostly illegible scrawlings.

    … Ah ok found it: page 42 = 38/39 of the Danish.

    The “inventoried entities” at different levels of a chain or hierarchy include phonemes, syllables, words, clauses (primary and secondary), … I see nothing I can interpret as a “algorithmic description of a language”. I’d call it more a slot-and-filler approach. Where else is Seuren getting this from? Is he mixing up the references? Page 26 of the English or Danish has no algorithmic description. Page 28 = Danish 27 has

    Form of the analysis
    The analysis thus consists actually in registering certain dependencies between certain terminals, which we may call, in accordance with established usage, the parts of the text, and which have existence precisely by virtue of these dependencies and only by virtue of them.

    If I put that together with other comments, this seems an admirable caution against attributing psychological existence to artefacts concocted by the theorist. No Phlogiston here!

    From an exegetical point of view, none of these ideas seem specific enough to be attributable to any particular theorist. But how novel/revolutionary were they in 1940’s/1950’s? Is SS a re-interpretation of Zellig Harris’s interpretation of Hjelmslev?

    A pity Chomsky didn’t pay closer attention to Hjelmslev’s metatheoretical admonitions.

  87. The reference is to the Whitfield translation (1943). The abbreviated reference is PTL: 26. PTL being Hjelmslev’s Prolegomena.

    I will type the first part of the quote:

    When we compare the inventories yielded at the various stages of the deduction, their size will usually turn out to decrease as the procedure goes on. If the text is unrestricted, i.e., capable of being prolonged through constant addition of further parts, … it will be possible to register an unrestricted number of words. Sooner or later in the course of the deduction, however, there comes a point at which the number of the inventoried entities becomes restricted, and after which it usually falls steadily. Thus it seems certain that a language will have a restricted number of syllables, although that number will be relatively high. In the case of syllables permitting a division into central and marginal parts, the number of members in these classes will be lower than the number of syllables in the language. When the parts of the syllables are further partitioned, we reach the entities conventionally called phonemes; their number is probably so small in any language that it can be written with two digits, and, in a good many languages, is very low (somewhere about twenty).

    (But this is a bit of truism — the ability to produce an unlimited number of sentences from a limited set of sounds — so I’m not sure that this really the same as Chomskyan ideas of grammar producing an infinite number of sentences.)

    Seuren also covers other principles germane to Chomsky, such as simplicity.

  88. Thanks @Bathrobe.

    Whitfield translation (1943)

    No, 1943 is when Hjelmslev published in Danish. Whitfield’s translation appeared variously 1953 (in a Journal), 1961 (book form), 1969 (reprint of book) (see my post).

    The abbreviated reference is PTL: 26

    I don’t know what the ’26’ relates to: it’s not that page in the Whitfield 1961/1969; it’s not that page in the Danish. It’s possible it’s page 26 in the Journal, but the text you quote appears on page 42 in the 1961/1969 book, so the Journal would have to be using tiny font.

    I didn’t bother quoting the text you give/rather I summarised because — as you say — there’s nothing exciting there.

    So what Seuren’s citing doesn’t support his attributions to Hjelmslev. Was H producing work-papers that got circulated privately rather than formally published? Were there lectures in which he elaborated the ideas, and which students noted down/published? (á la Wittgenstein Blue & Brown Books.)

  89. transfinite

    Langendoen and Postal’s The vastness of natural languages gets there by assuming that natural languages are best modelled within a very rich set theory and that they can create new sentences by coordinating arbitrary sets of sentences. A more interesting book might have studied those objects, instead of trying to argue they have much to do with natural language (except as generalisations of some model of them).

    The above is not meant as implying any comment on Postal’s other work or on his other criticism of Chomsky or Chomskyanism.

    Julian Bradfield has written an informal review that’s clearer than some older ones.

  90. David Eddyshaw says

    Bradfield’s review is excellent (as is JL’s executive summary.) The review is also very funny.
    (And very quotable, but it’s best just to read it.)

  91. @ Jichang Lulu, many thanks for this reference, it’s wonderful!

    Can anyone interested in such things suggest a cogent account of the state of the art of quantification over possible worlds? See for example

    https://en.wikipedia.org/wiki/Meinong%27s_jungle

    and/or the [claimed proofs] section of

    https://en.wikipedia.org/wiki/Abc_conjecture ?

    I think we should be told!

  92. David Eddyshaw says

    Reminds me of my favourite lines from David Lewis’ Counterfactuals (section 4.1):

    What arguments can be given against realism about possible worlds? I have met with few arguments – incredulous stares are more common. But I shall try to answer those that I have heard …

  93. @ David E, thanks, I wasn’t aware of Lewis.

  94. Stu Clayton says

    The Girl Who Never Read Noam Chomsky

    # An ambitious debut, at once timely and timeless, that captures the complexity and joys of modern womanhood. This novel is gem like—in its precision, its many facets, and its containing multitudes. Following in the footsteps of Virginia Woolf, Rona Jaffe, Maggie Shipstead, and Sheila Heti, Jana Casale writes with bold assurance about the female experience. #

    Another example of how smart women are !

    Although gems don’t have containing multitudes, or am I missing something ?

  95. David Eddyshaw says

    Although gems don’t have containing multitudes

    As this gem is both timely and timeless, stale patriarchal antinomies do not apply.

    Similarly, the refusal (playfully presented as a “failure”) to read Chomsky represents a rejection of the distinction between I-language and E-language, so obviously reflective of the male obsession with “performance.”

  96. I am struggling to embrace that side of my masculinity.
    But I compensate by growing a beard (actually I am just too lazy to shave anything).

  97. Its containership of multitudes?

  98. Multitudes of containers?

  99. Stu Clayton says

    Das Vaginarrenschiff, an allegory of patriarchal perplexity.

  100. I read it as “multitudes [Mengen] that contain this novel”.

  101. PlasticPaddy says

    @drasvi
    Do I contradict myself?
    Very well then I contradict myself,
    (I am large, I contain multitudes.)
    https://poets.org/poem/song-myself-51

  102. Whitman clearly refers to paradoxes similar to Russell’s.
    Was “a multitude” used in English in the sense “a set”?

  103. David Eddyshaw says

    Whitman does not, however, claim to contain himself. Or deny it …

  104. J.W. Brewer says

    @drasvi: The current reasonably rigorous mathematical sense of “set” did not arise (in English or more importantly AFAIK in mathematics conducted in any language) until toward the very end of Whitman’s lifetime. Whitman himself had very little formal schooling and I would not be surprised if he had never gotten through Euclid’s Elements, which was about as far as you might expect an educated American of his generation with no particular professional reason to explore trigonometry etc to have gotten in mathematics.

  105. @JWB: I assume drasvi asked because the Russian word for “set”, множество, also means “multitude” (that’s similar to German Menge, which besides “set” in “set theory” (Mengenlehre) also means “multitude” (plus “amount, quantity, portion, crowd…”)

  106. yes.
    In Russian a “set” is a long m-word mnozhestvo which means “multitude”.
    It is calqued from German Menge, I think, which means “multitude” (and Cantor was German).

    Cf. also French Wikipedia:

    En mathématiques, un ensemble désigne intuitivement une collection d’objets (les éléments de l’ensemble), « une multitude qui peut être comprise comme un tout » (au sens d’omnis).

  107. And of course

    -did you ever see such a thing as a drawing of a muchness?’
    ‘Really, now you ask me,’ said Alice, very much confused, ‘I don’t think–‘
    ‘Then you shouldn’t talk,’ said the Hatter.

    in Russian becomes

    ‘–that begins with an M, such as mouse-traps, moon, mathematics, set [multitude] –did you ever see such a thing as a drawing of a set[multitude]?’
    ‘A set of what?’ – asked Alice.
    ‘Nothing,’ – responded the Dormouse, ‘just a set’
    ‘I do not know,’ began Alice…

  108. J.W. Brewer says

    I regret to say that Whitman was probably unaware of the wordplay possibilities of Russian in this regard.

  109. @JWB, Cantor was a German as I said.

    English literature about sets was likely translated (just like Russian) and English mathematicians themselves wrote in many other langauges. If “multitude” was a word that appeared in discussions of sets (as you can see in the French example), Whitman could have heard it.

    Naïve set theory was an inspiration for mathematicians and affected their language. But it was simple. I guess it also inspired philosophers. Today we discuss discoveries in biology without knowing anythign about biology and I am not sure that “things that affect mathematics” have always been esoteric for everyone.

    Do I contradict myself? Very well, then I contradict myself” sounds similar to the style of reasoning found in logic.

  110. compare

    Идите все, идите на Урал!
    Мы очищаем место бою
    Стальных машин, где дышит интеграл,
    С монгольской дикою ордою!

    literally:
    Go all, go to the Ural / We are clearing space for the battle / Of steel machines where the integral breathes/ with wild mongol horde

  111. Идите все, идите на

    I finished this another way in my head, after all the русский военный корабль stories.

  112. J.W. Brewer says

    Poking around Google books, I find a 1905 “Inaugural-Dissertation zur Erlangung der Doktorwuerde” usw usw written in English by the splendidly-named Harold A.P. Pittard-Bullock (aus Guernesey), which uses “theory of multitudes” in a way that may be synonymous with what is now standardly called in English “set theory.” That phrasing does not seem to have caught on all that much although there’s another instance around the same time and a scattering later in the 20th century in texts that may perhaps have a “not-written-by-an-L1-Anglophone” vibe. And there’s a guy claiming that C.S. Peirce was interested in “postnumeral [=transfinite?] multitudes” although he may not have been aware of Cantor and may have meant something related-but-different.

  113. Intergral-Ural is an unexpected rhyme.

  114. @JWB, well, apart of using “a multitude” in the sense of “a set”, it could also commonly occur in logical discourse involving sets and stuff. One favorite word in a logician’s vocabulary. Like basically in the French article (“ensemble”) quoted above. Then as English sailed further away from the continent (English scholars ceased to write and then read in German and French) it could become forgotten.

    Note also that Whitman does not say multitudes of what. Just multitudes. It is reminscent of mathematical usage too (simiarly from Arabic learners I can hear “MSA or dialect” or “MSA or a dialect” – without specifying dialect of what. What a learner means is “vernacular as opposed to the literary langauge, whatever local variant of it among many mutually partly unitelligible variants the person in question may need”).

  115. Dr. Internet: “Harold married Sofia Teresa De Sadow in 1908 and changed his name to Harold Arthur Penrhyn De Sadow–Pittard.”

    (I changed the hyphen to an en space. Is that correct, Hat?)

  116. It is (well, technically an en dash), and you get the Gold Star of Copyediting for today!

  117. Stu Clayton says

    Note also that Whitman does not say multitudes of what. Just multitudes.

    Nor did he say what he didn’t contain. That I find more exciting than the unwashed multitudes he harbored. See my forthcoming monograph with the provisional title Exclusion: The Outside Of Whitman’s Container.

    From the available evidence it seems that he meant to exclude multitudes that contain themselves.

  118. jack morava says

    In French mathematical parlance the category of sets is often denoted (Ens) for (Ensemble) which I have always supposed to be a pun on (being) in scholastic logic but maybe I just made that up.

  119. Stu Clayton says

    Well, that’s what I had always supposed as well. It’s an amusing story at any rate. “Ensemble” never occurred to me, not being Frenchy. “Ens” for the category of sets is not a proprietary hexagonal term.

    Next you’ll be suggesting that V might not refer to the vniverse.

  120. The usual Russian rhyme on ensemble is сам, бля, без ансам-, -бля…

  121. Anyway, I misread it because this agent possesseur thing (“its containing”) is what I can’t wrap my mind around in English.

  122. Stu Clayton says

    That’s because “its containing multitudes” doesn’t make any sense in English. You can unwrap and relax.

    The preceeding expression “timely and timeless” doesn’t make any sense either, but I’m sure whoever wrote it was so pleased with themselves that they got feuchtes Höschen from it.

  123. But then I won’t find the key to joys of womanhood embracing the chomsky-disjointed side of my masculinity.

  124. Though…

    The initial goal of the research reported in this chapter was to develop an extension of the set theory able to handle the meanings of concepts” says an article Theory of Multitudes as an Alternative to the Set Theory.

  125. @DE: i feel confident that whitman refused to contain himself, on principle.

    and i suspect his “multitudes” had “the mob[ile classes]” as a piece of its intertextuality.
    on the other hand, maybe it was just amy.

  126. “… Harold Arthur Penrhyn De Sadow–Pittard.”
    (I changed the hyphen to an en space. Is that correct, Hat?)
    It is (well, technically an en dash), and you get the Gold Star of Copyediting for today!

    Yes, that’s an en dash. But the award is suspended and referred to the committee for a ruling.

    Points:

    1.

    Only in US English does this issue of an en dash come up at all. Chicago (CMOS) style calls generally for an en dash to replace the usual hyphen when either of the connected elements itself includes a hyphen or a space: “the pre–Vietnam War peace movement”; “the car-dependency of Los Angeles–like cities”. That is done infrequently in the UK or Australia, and it is only done under US influence.

    2.

    The CMOS 17 rules (at §§6.78–84) are poorly regimented and not all clear. §6.80 deals with compound adjectives, but does not give any confidence for cases like “Grand Rapids-Wyoming Metropolitan Area”, which are nearly always seen with a hyphen though “Grand Rapids-Wyoming” here is compound and adjectival. §6.80 ends by prescribing a hyphen in “US-Canadian relations”; it glosses this in parentheses with “Chicago’s sense of the en dash does not extend to between“; but we needed to see that as an important separate rule concerning a case like “Australian-Canadian relations”, and a case like “United States–Canadian relations” (or perhaps “United States-Canadian relations”?).

    3.

    Examples above, including the CMOS one, are adjectival and also would mostly occur as nonce formations (contrast cases like “The Diocese of Great Falls-Billings”, online). Things are less clear with fixed and generally non-adjectival forms such as compounded surnames. CMOS does not rule on substituting en dashes in these, but an opinion is found in their Q&As:
    Q. If someone has a compound surname like “De Chicago-Smith,” do we use an en dash? I understand the rationale, but I think it looks weird (but who cares what I think?). What about “De Chicago-Von Suedkurve Auf Der CSS&SBRR,” for example?
    A. Although a simple hyphenated name normally takes (no surprise) a hyphen, a name with multiple appendages might be able to pull off the slightly longer en dash. Anyone with such a dazzling name as “De Chicago–Von Suedkurve Auf Der CSS&SBRR” deserves all the dashes and doodads they want. (And we care what you think.)
    This advice is incomplete, arbitrary, and off the cuff. And it neglects point 4:

    4.

    Who has the final say about such usages in a surname? (I almost wrote “such en-dash-hyphen questions”, but went for discretion instead.) The bearer of the surname? Has anyone anywhere stipulated a CMOS-like en dash for their surname? A Matt van den Bosch-Finkelstein or an Eliza Terranova-di Santo is unlikely to acquiesce in a CMOS en dash being imposed. The Rest of the World is used to US publishers tucking a comma inside a closing quotation mark within a larger quote from a UK source. But CMOS changes in surnames? This time it’s personal.

  127. I wasn’t sure if the en-dash applied for names of this sort as well (in my humble circle there are but few hyphened noblemen.) I figured it was rude to look it up when a professional is present.

    I wrote “en space” because I was dashed out.

    I generally use en dashes as necessary, not because I’m trying to be correct or something, but because I genuinely find that they makes the text clearer. I’m glad I learned about them and that my computer makes it easy for me to type them.

  128. “because I genuinely find that they makes the text clearer”

    The sentiment is familiar, the outcome xxxxxxx-yyyyyyy–zzzzzz is still not pretty:(
    And the cost is high.

  129. Baron du Vallon de Bracieux de Pierrefonds does not contain multitudes of hyphens.

  130. @jack morava: I think it’s interesting that French mathematicians are using the same standard for naming categories as is used in English (three-letter abbreviations with an initial majuscule). They just abbreviate French terms instead of English (or they do so sometimes, at any rate).

  131. Brett: OTAN. The only capable Prime Minister of France during the cold war was Georges Pompidou, and all they remember him for is an, admittedly, nice building.

  132. Stu Clayton: “Next you’ll be suggesting that V might not refer to the vniverse.”

    I don’t even remember why it was the letter “V” that I chose.

  133. Lars Mathiesen says

    But Nous sommes l’ISO, l’Organisation internationale de normalisation. — from their home page. It’s a trademark and all. OSI is Open Systems Interconnect who did the 7-layer model of network protocols.

    So how long before they drop Russian as an official language? Мы являемся Международной организацией по стандартизации.

  134. Lars Mathiesen says

    Danish (you knew it was coming) actually distinguishes mængde (set, quantity, crowd, many interchangable items) from mangfoldighed (diversity, many saliently different items). (Also Sw mängd, mångfald). Whitman’s multitudes would be en mangfoldighed.

    (The word is transparently derived in the same way as E manifold but the adjective mangefold is not in current use, only the abstract noun. If it was, you could have en mangefold mængde).

    Also duplicate really means “fold in two!” Who is with me?

  135. mangfoldighed
    German Mannigfaltigkeit. But that’s a bookish word; normally Vielfalt / Vielfältigkeit are used instead.

  136. David Marjanović says

    So how long before they drop Russian as an official language?

    I don’t think they will. Plenty of native speakers outside Russia and Belarus – in Ukraine and the EU members Estonia and Latvia for example.

    Also, consider the possibilities the great & mighty Russian language offers. These people sound like they could just keep going forever – and if they do, they’ll never get to fight. Win-win-win.

  137. Yeah, I really hope the animus against the Russian state (which at this point basically means Putin) doesn’t spill over onto the language and literature. It would be stupid, pointless, and anti-human.

  138. Lars Mathiesen says

    I’m not sure the non-Russian side will limit itself to sensible, efficacious and humane measures once it gets into the flow of it.

  139. I’m sure it won’t. That doesn’t mean we can’t deplore it. The abuse of people who have nothing to do with their government’s politics (or, worse, with the politics of a government they have nothing to do with, like the Japanese-Americans in WWII) is an ancient and dishonorable tradition. Fortunately, the US is not at war with Russia; if it were, I’m sure people would be attacked for speaking Russian in Brooklyn.

  140. I personally know at least one person whose income is going to suffer because it depends on grants for translating Russian literature.

  141. J.W. Brewer says

    The glass-half-full view of this news story is that the Italian university that cancelled a class on Dostoevsky quickly backtracked and reinstated it after receiving well-deserved criticism. The glass-half-empty view is that some administrator there had thought that a sensible symbolic gesture in the first place. https://www.ansa.it/english/news/world/2022/03/02/ukraine-dostoevsky-course-reinstated-after-cancellation_d1a69d23-d37f-4f05-a255-9e3041e63521.html

  142. I’m glad to hear it!

  143. @V: I was actually thinking yesterday, as I have done a few times before, about the reversed acronyms for NATO in English and French. (Yesterday, my thinking was occasioned by a rereading of the North Atlantic Treaty.) I wonder whether the name chosen for the alliance (there seems to have been some who preferred the name “North Atlantic Alliance” at the time; the 1949 North Atlantic Treaty itself does not name the organization that the member states form), which is bracketed in English as [[[North Altantic] Treaty] Organization with three nested adjectivals, was selected so that the French version would have the letters in the reverse order, thanks to French’s postposed modifiers. The fact that “OTAN” is almost “NATO” under a reflection may make it convenient to print both acronyms (French and English always having been the principal official languages of the alliance*) on a flag, which can almost be read from either side. Some NATO banners indeed include both, and you only need to make a minor mental adjustment for the inverted “N” if you are viewing them from the backs. However, the main NATO flag that was settled upon in the 1950s ended up with no writing.

    * Per the final article of the treaty:

    Article 14

    This Treaty, of which the English and French texts are equally authentic,….

  144. “Furthermoe, the university’s dean will meet Paolo Nori next week for a moment of reflection”.

    Reminded me about the dean of the mathematical faculty of MSU. MSU is a tower, and students were pressing themselves into the elevator and when it was full someone said “move back: we see there is still empty place/space there” and the voice from the empty space said “I am not an empty place, I am the dean of the mathematical faculty!”. He’s short….

    An empty place is an insult here.

  145. Means: “insignificant”. And the faculty (in Russia: a major university department almost independent from other departments*) is mechanico-mathematical, “mech-math”.

    It is cute, two Italians meeting for a moment of reflection but also it is weird. Is he (dean or Nori…) such a queen of Sheba?


    * from student’s perspective you graduate not as much from MSU as from mech-math.

  146. Brett: a petition for Finland to join NATO/OTAN reached 50,000 signatures, required to be discussed in Parliament. Good. If shit goes down, we could use their cocktails.

  147. Brett:

    > I wonder whether the name chosen for the alliance (there seems to have been some who preferred the name “North Atlantic Alliance” at the time; the 1949 North Atlantic Treaty itself does not name the organization that the member states form), which is bracketed in English as [[[North Altantic] Treaty] Organization with three nested adjectivals, was selected so that the French version would have the letters in the reverse order, thanks to French’s postposed modifiers.

    I thought it was obvious that it was deliberate? I think even the people that came up with the acronyms said they made them that way deliberately.

  148. David Eddyshaw says

    Ah me! That’s when we had decent human beings as Prime Ministers …

  149. Bathrobe says

    “Empty space” is an insult? In English the roughly corresponding term is equally devastating: “Waste of space”.

  150. Or “waste of air” (meaning breathed air).

  151. @V: It’s entirely possible that the choice of names and acronyms is well documented and confirms that hypothesis. I just don’t think I have ever come across that information. All I know is, as I said, there was initially some discussion about what the formal name of the alliance should be.

  152. “an insult” – Bathrobe, I was not quite accurate here. The link is to “he is an empty place” in Google and it shows what sort of context it is used in (many texts in Google are translations from Russian).
    Someone insignificant, someone you can or do fully ignore and not count with. Some texts in google speak about children who are made feel so by their parents. Other time you say about an official. It does not necessarily means that the person deserves such treatment.
    But can be an insult too: “how you dare bla-bla-bla, who are you to say it bla-bla, I am the hero, you are an empty place bla-bla-blah”.

  153. David Marjanović says

    Or “waste of air” (meaning breathed air).

    I’ve encountered “waste of oxygen” pretty often.

  154. “Waste of air,” and “waste of oxygen” are both pretty common, with the “oxygen” version probably being a bit more common. I once called somebody a “waste of glucose,” although that misses part of the joke. The real bite of “waste of air” comes from the fact that air is already free to breathe.

  155. J.W. Brewer says

    To quote the “antique frame” guy, writing more than a century ago:

    I balanced all, brought all to mind,
    The years to come seemed waste of breath,
    A waste of breath the years behind
    In balance with this life, this death.

    Now, one could say that “breath” is a lazy rhyme with “death,” but I think it maybe actually works better than “air” or “oxygen” would have regardless of that?

  156. I, for one, am not going to say, “Here, Mr Yeats, let me fix this poem for you…”
    And why are breath/death a lazy rhyme? -/eθ/ is not that common.

  157. The claim that “kill” should be derived from “cause to die” struck me as ridiculous.

    On the contrary, it always struck me as self-evidently correct.

    I feel the GS crowd might lose me instead in that this equally clearly should be understood as an analysis of the meaning ᴋɪʟʟ, and not of any real lexeme ‘kill’, which easily can and usually do accumulate additional semantic nuances, connotations and metaphorical uses. Same for derived forms too. Finnish has a transparent causative of ‘to die’ kuolettaa, indeed attested in dialects as ‘to kill’, but by now it primarily means the economic term ‘to amortize’. Approaching lexemes just by their dictionary glosses is almost always a gross simplification; approaching them just by their morphological makeup is much of the time just flat-out wrong.

    Is the Natural Semantic Metalanguage project considered descended from or an offshoot of GS? That seems like a more productive way to proceed from observations like this, not anything that necessarily drags in syntax.

  158. John Cowan says

    He passed, and he was still traumatized? Did he, like, need straight As for a scholarship or something?

    For a certain kind of U.S. person (including me in my larval stage), any grade other than A is psychologically a failing grade, and in I think it is still true that in U.S. graduate schools a non-A is a warning that you are in the wrong field and should depart forthwith.

    [from Pullum’s piece] Hardly anyone will stand up and call him a liar in front of a lecture theater full of his devotees.

    That’s because none of them were students of either romantic literature or the Edinburgh Review. Had they been, they would have been armed with such statements as:

    You are a liar, sir, a liar, and if I express myself in harsh terms commonly considered unfit for polite society, it is to show in unmistakable terms not merely my utter detestation of what you say but the lengths to which I will go to enforce my views.

    (The lie direct was a challenge to mortal combat.)

    And Jung was mainly touted by people who supposed that the only alternative was Freud.

    As someone who might not have been born if Jung had never lived, I can falsify that remark.

    On the contrary, it always struck me as self-evidently correct.

    That’s probably because you think in a causative language, whereas we anticausitive-speakers find such things hard to swallow.

    Is the Natural Semantic Metalanguage project considered descended from or an offshoot of GS?

    I very much doubt it. Anna Weetabix went straight from Poland to Australia without a stopover in North America worth mentioning.

    waste of space/air/oxygen/etc…

    WOMBAT ‘unfortunate and doomed undertaking (lit. ‘waste of money, brains, and time’)

  159. David Marjanović says

    I think it is still true that in U.S. graduate schools a non-A is a warning that you are in the wrong field and should depart forthwith.

    I knew about grade inflation, but I didn’t know it was quite that bad.

    As someone who might not have been born if Jung had never lived, I can falsify that remark.

    I’m intrigued.

    That’s probably because you think in a causative language, whereas we anticausitive-speakers find such things hard to swallow.

    Causatives are dead in German basically as in English, but töten remains a rather obvious causative, factitive or other verbing of… tot, the adjective “dead”. The verb cognate with die hasn’t survived.

  160. Lars Mathiesen says

    Swedish has döda as the default verb for ‘kill’, but it’s gone from Danish (slå ihjel). SAOB does list a Danish døde but just says “corresponding to,” also to Gothic dauþjan; and it’s not in Hellqvist, so I can’t tell you if it’s native or from MLG.

    TIL that the /g/ in the preterite dog of may be a fortition of an intrusive /w/ in the Old Swedish strong plural preterite döo — otherwise it’s scurrilously claimed to be a Stockholm corruption based on slog to slå where it has impeccable Germanic pedigree. (Danish, again, opts for the simple life and the verb is weak, with død = ‘dead’ as a frozen strong participle; the weak participle døet = ‘died’ is not usable, though. ‘All my houseplants have died’ can only be Alle mine potteplanter er døde — Germans note number accord in the predicate adjective).

  161. David Eddyshaw says

    I presume that the generative idea that “kill” is in some way “derived” from “cause to die” has nothing in particular to do with morphological derivation; the assertion would stand or fall on whether “kill” had significant syntactic properties in common with causative constructions (or derivatives, in languages which actually have productive morphological ways of deriving causatives.)

    As I mentioned above, Kusaal “kill” actually does share at least one syntactic property with causatives and applicatives, in that it’s strictly transitive (as opposed to agentive ambitransitive, which is much commoner in Kusaal.) It’s not a criterion that you could use to demarcate “causatives” from “the rest”, though, unless you could devise some ingenious Chomskyan way to make e.g. “be”, “have”, “resemble” and “love” causative as well.

  162. @David Marjanović: No doubt there has been some grade inflation over time in graduate programs, but not as much as John Cowan suggests. In North American doctoral programs (professional programs can be different) A has been the most common grade for a long time. Grading is supposed to be based on absolute mastery, not on populating some arbitrary curve, and graduate students are supposed to be exceptionally strong academically. A grade of B is a sign of weakness, but isolated B grades are not typically reason to worry unduly. Getting a C, however, is a sign that one is in trouble, although it can sometimes still be overcome.

  163. @DE, ah, love means “to make beloved”:)

    There was a Soviet cartoo about this. A Goatkid Who Counted to Ten. He learned to count and was running accross the forest warning animals that he is going to count them. And was beaten up for that.

  164. https://ru.wikipedia.org/wiki/Козлёнок,_который_считал_до_десяти

    Based on «Geitekillingen som kunne telle til ti» by Alf Prøysen it says…

  165. J.W. Brewer says

    Re “professional programs can be different,” I don’t know whether it’s still the case but When I Were a Lad (or more to the point a law student) many U.S. law schools had rigid “curves” saying that the professor absolutely positively could give out A’s to no more than X% of students (maybe 20 or 25% was a common cap, although my memory is fuzzy about the details) in the big first-year required classes. Best as I can tell, this was because the grading system was in large part intended to communicate relative standing rather than absolute mastery to potential employers. Practices could be looser and more lenient in advanced electives, but not very far above a B+ average over 3 years would (at my school at least) still sufficed to get you into the top 20 or 25% of the graduating class. To be sure that may have been because the school was at the time using a wacky grading scale where you could get an A+ but not an A-, a B+ but not a B- etc. I don’t know the backstory but it seems like the sort of thing someone must have thought was a good idea in the Seventies but by the early Nineties was hanging on via sheer inertia.

  166. That’s probably because you think in a causative language, whereas we anticausitive-speakers find such things hard to swallow.

    I’m not sure of the logic here. Japanese has a clear causative marker, which means that causative can be distinguished from transitives. But the line can of course be fuzzy. Causativity can also involve nuances of causing something to happen and allowing something to happen.

    My main concern is the fact that language is likely to need a distinction between the direct act of killing and that of causing someone or something’s death. “Kill” = “cause to die” just seems too blunt an instrument to me.

  167. running accross the forest warning animals that he is going to count them

    is this the kind of implicit threat of exposure to supernatural attack that it could be in a yiddish context?

    the safe way to count precious beings (children especially – and a kid is often a figure for a child), of course, being “not one, not two, not three…” – because though the supernatural world may be threatening, it runs on very transparent semantic principles.

  168. Stu Clayton says

    Causativity can also involve nuances of causing something to happen and allowing something to happen.

    As in “Whoops ! How careless of me !” Also Hoppla! Das wollte ich nicht! – or, equivalently impertinent: Das war nicht Absicht! Why ist deliberateness denied before a charge of deliberateness is brought ? They need to get better lawyers, or else pay more attention to what they do.

    There is a whole pragmatics of flakey paranoia-inducing nuances. “I meant well.”

  169. Or in Russian “Я нечаянно!” ‘I didn’t mean to! It was an accident!’ (to which the traditional response is “За нечаянно бьют отчаянно” ‘For not meaning to they’re mighty mean to you’ [literally ‘they beat you unmercifully’]).

  170. John Cowan says

    I’m intrigued.

    Both my parents were Jungians, and remember that I am a producer-product Peircean (which is why I have no patience with people saying “Ukraine’s plight is the result of X”, as if things had causes).

    nothing in particular to do with morphological derivation

    Well, from a sensible point of view, of course not. But the GS People were Chomskyites, or thought they were, and so adhered to “We don’t need no steenking morphology!”

    I’m not sure of the logic here.

    I was implying humoris causa that J. Pystynen was a sad victim of Whorfian Mind-Lock, that’s all.

  171. Stu Clayton says

    Stressed thus: не’чаянно ?

  172. @J.W. Brewer: With the system we use, I can give plus grades but not minus grades. However, we don’t give A+ (or D+) either.

    Unofficially, I got an A– in Algebraic Topology I. (Plus and minus modifiers at MIT are recorded but are not actually part of the official grades on one’s transcript.) That A– was one reason, although not the most important one, that I decided to drop Algebraic Topology II.

  173. Well, from a sensible point of view, of course not. But the GS People were Chomskyites, or thought they were, and so adhered to “We don’t need no steenking morphology!”

    The presentations of the history of morphology that I read in grad school implied that the rehabilitation of morphology in Chomskyan circles was essentially a byproduct of Chomsky’s main argument against GS (the Lexicalist Hypothesis).

  174. A dialogue in elementary school: “do you prefer 2+ or 3-?” “2+, of course, a plus is better than minus”.

    My sense of humour was nor very well developed. When the teacher asked her question I thought: “oh, still not 2.” because pluses and minuses do not really matter, and then I was surprised by the boy’s answer.

  175. Yes, that’s my problem with GS. They wanted to derive surface structure direct from the semantic structure, if I remember rightly. “Kill” as a lexical unit meaning “cause to die” (or cause not to be alive” even) makes sense. But deriving the morpheme “kill” from a semantic structure “cause to die” just seems too simplistic.

  176. I am entirely ignorant of what semanticists have been saying about “kill = cause to die” for the last fifty years. I imagine that one side, enamored with truth values and symbolic logic notations, has no use for pragmatic fine points, and sweeps them aside just as generativists sweep aside “performance”. I also imagine someone said, based on handwaving and no evidence, that “kill” means “cause to die” underlyingly, deep in one’s invisible mind, and this unprovable, unfalsifiable idea becomes fodder for generations of linguists trying to prove it or falsify it.

    BTW, Hebrew uses both the dedicated verb לַהֲרֹג lahărog and the causative of ‘to die’, לְהָמִית lǝhāmīt. The former is more blunt, the latter more euphemistic. Both have been used in the OT, in the Mishna, and in the modern language.

  177. J.W. Brewer says

    The etymology of English “kill” is uncertain and debated and until not too many generations ago it had to share semantic space with the not-yet-archaic “slay.” Indeed, the Old English verb usually glossed “kill” evolved into the not-necessarily-lethal modern verb “quell,” which is similar enough to suggest some sort of doublet, yet the uncertainty remains.

    GS probably worked much better on conlangs where such messiness and weirdness would not have been tolerated by the Intelligent Designers.

  178. David Eddyshaw says

    Both my parents were Jungians

    My grandfather was a Jungian too, so we are cousins. Or, possibly, you are my uncle.

  179. Stressed thus: не’чаянно ?

    Yes, hence it rhymes with отчаянно.

  180. David Eddyshaw says

    Kusaal kpi “die” actually has (what looks like) a morphologically-derived causative*, kpiis, but it means “quench” (as in fire), not “kill.” (The corresponding intransitive “go out (of fire)” is kpiig, with a common inchoative derivational suffix.)

    * Many verb-deriving suffixes in Oti-Volta are quite hard to assign any consistent meaning to, but WOV -s- is comparatively well behaved: it’s nearly always either valency-increasing (causative/applicative) or pluractional.

  181. David Eddyshaw says

    Lexicalist hypothesis

    Per WP: phrasal syntax has no access to subword units.

    A schoolboy error, which could have been avoided by even a passing familiarity with Greenlandic.
    Presumably the Great Man himself has long since abandoned it, as he has abandoned all hypotheses actually susceptible of refutation. (There never was any Lexicalist Hypothesis, comrade.*)

    There are languages out there with more than one morphological causative, differing in the degree of control or power of coercion that the causer has over the causee (“let” versus “make”, and so on.)

    With “kill”, all options are covered by Asimov’s First Law, fortunately.

    * How many Political Commissars does it take to change a lightbulb?

  182. yes.
    чаять, “to wait, expect, hope” (intersection rather than union of suitable contexts)
    Wiktionary: *čajati, one of those words where supposed Sanskrit cognate is ideantical: cā́yati, “to perceive, to observe”.
    нечаянно “inadvertently”, as in Wiktionary’s ‘Dear me! I may have inadvertently touched the handle,’ said he. (1929, Sir Arthur Conan Doyle, The Disintegration Machine‎)
    я нечаянно – children say after having destroyed something.

    Adults may say something longer, impersonal and sometimes with a word “accidentally”:
    это получилось случайно, это нечаянно вышло etc. (they are better liars?).
    —-
    I can’t remember an English word expressing this concept which is suitable for little children:/
    —-
    отчаянно is “desperately”. It can mean violent, panic action, but does not mean нещадно, unmercifully. Its use in this response is unexpected/funny.
    But I did not know what “desperately” normally means when I learned the response 🙂

  183. David Eddyshaw says

    The Lexicalist Hypothesis doesn’t even work in English:

    https://www.linguisticsociety.org/sites/default/files/03_94.1Bruening.pdf

    Or Kusaal, of course: salima nɛ anzurifa la’amaan “maker of gold and silver items” (gold with silver item-maker.)

    Footnote 2 of the Bruening paper actually seems to be the most important point:

    A necessary assumption of this view is that it is possible to define the word as a linguistic unit and to unambiguously identify words. As is well known, this assumption is false.

    This is actually extremely apropos to my Kusaal example, in which my very own analysis treats la’amaan as two words (despite the standard orthography), which actually neatly abolishes the whole problem at a stroke. But that makes sense in the light of a whole lot of quite other evidence for the nature of nominal “compounds” in Oti-Volta languages, and is not something I dreamt up just so that I could maintain the Lexicalist Hypothesis in all its inerrant purity; and you can’t pull the same trick with West Greenlandic.

    The Lexicalist Hypothesis seems to be well embedded in some theories still currently regarded by their devotees as undebunked, though.

  184. Haspelmath has recently decided that words can be allowed to exist after all:

    A word is a free morph, or a bound morph that is not an affix, or a combination of a root or a compound with its affixes.

    Yeah, the Lexicalist Hypothesis is a bit of an ad hoc weapon that doesn’t really work. It allowed a handful of linguists to carve out a niche as morphologists at the very height of Chomsky’s heyday without having to pretend to be syntacticians, but since the 1980s and especially 1990s – with GS safely extinct – a lot of the more interesting generative work (notably Distributed Morphology) has focused on trying to pull morphology back under the umbrella of syntax, against the morphologists’ protests. This approach, however, seems to appeal a lot more to linguists looking at nice nearly-agglutinating languages than to ones full of inflectional classes like Russian or Dinka, let alone nonconcatenative morphology like Arabic (though efforts have certainly been made). As morphologists like to put it, item-and-arrangement models of morphology just don’t work very well.

  185. Incidentally, if my vague recollections of The Linguistic Wars are correct, this move, effective though it was in preserving Chomskyan hegemony within linguistics, was also a big part of why Chomsky lost his 1960s cachet in adjacent disciplines like literary criticism. The Lexicalist Hypothesis effectively let syntacticians draw a line between syntactically relevant meaning and all the rest of meaning, and banish the rest to the outer darkness of the lexicon or pragmatics; this came as a great disappointment to all the non-linguists for whom the most appealing thing about Chomsky was the implicit promise of a formal theory of meaning.

  186. Lars Mathiesen says

    Anecdata: Döda/slå ihjel are no more mandatory transitives in Swedish/Danish than kill is in English.

    Danish has kval = ‘tornent’ (noun) and kvæle = ‘strangle’ (verb). I never connected them with kill before so that was today’s learning moment and it’s not even lunchtime. (And kvæle does need its object, it has to be double-plus obvious from context before it can be omitted).

    (Kvæle [and Sw kväva] are widely used as metaphors for ‘suppress’ [undertrykke/undertrycka], with anything from sighs to rebellions, cf E quell. But also kvælstof/kväve = ‘nitrogen’ [it kills by suffocation, or suppresses fire, I don’t know which of those was the original idea]).

  187. PlasticPaddy says

    @lars
    Calque of Stickstoff?

  188. Lars Mathiesen says

    Or both are calques of Lavoisier’s azôte. Some historian has probably teased out who stole what from whom, but this was 250 years ago. (Danish and Swedish have kept a lot of the “domestic” names of the elements that were introduced back then, nice and short. Cp. kultveilte and carbon dioxide, though people mostly say kuldioksid now).

  189. David Eddyshaw says

    Haspelmath has recently decided

    A compound is a combination of two roots which occur next to each other and which cannot be expanded by nominal or adjectival modifiers.

    (“Compound” being a kind of “word”, apud Haspelmath)

    This actually means that Kusaal (and much of Oti-Volta generally) have no compounds, which will come as a surprise to most people who’ve studied the languages, and, indeed, to literate native speakers. As a matter of fact, I agree with the conclusion*, though not the means of getting there. The Oti-Volta languages, with their entirely regular systematic compounding (sic) not only of nouns with adjectives but also of nouns with demonstratives, were evidently created at Babel specifically to rattle people who imagine that they’ve come up with a watertight cross-linguistic definition of “compound.” Kusaal even more so, at least as regards “word”, as it has several perfectly real and distinct enclitics realised as segmental zero.

    * I don’t think this actually has any great cosmic significance, though. All that’s really needed is to define near the beginning of your description what you mean by “word”, and then be consistent about it. (In other words, I agree with pre-recantation Haspelmath.)

  190. January First-of-May says

    BTW, Hebrew uses both the dedicated verb לַהֲרֹג lahărog and the causative of ‘to die’, לְהָמִית lǝhāmīt. The former is more blunt, the latter more euphemistic.

    Russian has убить (lit. ~= “to beat up”*), the usual term for “kill”, as well as умертвить, the (apparent) causative of “to die”; the latter does sound (slightly) more euphemistic, as well as somewhat more archaic.

    OTOH, the Russian legal term for what US law calls “manslaughter” is причинение смерти по неосторожности – literally something like “causation of death by carelessness”.

     
    *) in a ridiculous example of false cognacy, Russian бить and its English translation (to) beat are unrelated at least at the PIE level

  191. in a ridiculous example of false cognacy, Russian бить and its English translation (to) beat are unrelated at least at the PIE level

    Somewhere there’s an LH thread about false cognates/faux ami, and if I find it I’ll add that to it.

  192. David Eddyshaw says

    a ridiculous example of false cognacy

    Kusaal bʋd “sow seeds”, sɛɛd “transplants.” However, these words are not cognate* with the obviously connected English words, which were, rather, borrowed by the Anglo-Saxons when they gave up their hunter-gatherer lifestyle and adopted agriculture from the Kusaasi.

    * English descends from Proto-Welsh and is not a Scandi-Congo language.

  193. I can’t remember an English word expressing this concept which is suitable for little children:/

    My daughter, who to be sure became an English major, said “accidentally” from at least the age of six or so, with no stressed syllable, i.e. as a sort of prefix.

  194. @Rodger C, aha, thank you!
    “Inadvertently” or “accidentally” confuse me, because Russian нечаянно and случайно are shorter – and that’s when Russian words are generally longer than English. Even “….-lly” in English sounds bookish.
    Meanwhile I want something that sounds casual (нечаянно even feels chidlish!). I guess, I must use “accidentally” then (when I need casual tone)….

  195. That’s probably because you think in a causative language, whereas we anticausitive-speakers find such things hard to swallow.

    Distinction without a difference for this purpose. I would not want to insist on a derivation in any sense of chronology or etymology, just on that the basic senses ᴋɪʟʟ and ᴅɪᴇ are related as a causative / noncausative pair. The disassembled English glossing ’cause to die’ is a pointer at this fact, not the actual fact. Something like ᴅɪᴇ = ‘to be killed due to something’ would serve about equally well at this. (2+3 equals 5, but this does not mean that 5 is “derived from” 2+3 specifically and we could well shuffle terms around to gain other equivalencies like 5-2 = 3.)

    The notion of “cause” is, yes, probably also somewhat blunt concept for this and in a formal analysis would have to be unpacked to include something something intentionality.

    the assertion would stand or fall on whether “kill” had significant syntactic properties in common with causative constructions

    Has no bearing whatsoever on my conviction, on the other hand. The core point I get from this particular adage is precisely that semantics has a structure to it independently of its formal realization as words, expressions, etc.

    Much is also dependent of course on the assumption of there existing any shared cognitively deep-set structure of semantics at all. Which is not a priori given and I would be open to a kind of a weak Sapir-Whorf argument that perhaps most or even all semantic categories are at some level built idiosyncratically per each speaker.

    On the Lexicalist Hypothesis, I could probably give a conviction on both why it’s obviously correct and why it’s obviously false, if given enough wiggle space on the definition of “word”… the synthesis from these, though, does also end up at “no, we do not need morphology”, just with morphology divided between other subfields rather than ceded in whole to any one. Or maybe, for the sake of livening things up, we could swap a few labels here and equivalently argue that syntax is a subtype of morphology?

  196. David Marjanović says

    I can’t remember an English word expressing this concept which is suitable for little children:/

    “I didn’t mean to”?

    This actually means that Kusaal (and much of Oti-Volta generally) have no compounds, which will come as a surprise to most people who’ve studied the languages, and, indeed, to literate native speakers.

    Leave a comment there. Mine got answered.

    OTOH, the Russian legal term for what US law calls “manslaughter” is причинение смерти по неосторожности – literally something like “causation of death by carelessness”.

    Is that “manslaughter” or what seems to be called “reckless killing”?

    (Totschlag vs. fahrlässige Tötung in German… but not in Germany, which has tried and amazingly failed several times to rid itself of the Nazi definition of the distinction of murder vs. manslaughter… I digress.)

    Anyway, erschlagen, which looks like “beat to successful completion”, means “beat to death” nowadays, but in a more Biblical register you can do it with swords, too. Schlagen alone is “beat” with about the same semantic range as in English (though “hit” is often more idiomatic there), including to successful completion in sports or in a battle.

    (Death as prefix-marked successful completion is more widespread, though: erschießen “shoot to death”, erstechen “stab to death”, erhängen “hang to death” – though that one is rare –, ertränken “drown [transitive]”, ertrinken “drown [intransitive]”, ersticken “suffocate” [transitive or intransitive]”, erfrieren “freeze to death” [intransitive, or at least no transitive sightings are known].* However, verhungern “starve to death”, verdursten “die of thirst”. That last one may be directly denominal, BTW, because “to thirst”, about as obsolete as in English, was dürsten. But then, that one went with the accusative like “methinks”.)

    * And then there’s the hapax legomenon that occurs in a plea to close the windows: erfroren sind schon viele, erstunken ist noch keiner “many have frozen to death, nobody has died of stench”.

  197. “I didn’t mean to”?

    Yes, that’s what kids actually say (unless they say “It wasn’t me!”).

  198. Stu Clayton says

    There’s a dis legomenon as well: erstunken und erlogen. “A lie that stinks to hgh heaven”.

  199. David Marjanović says

    Oh! Yes. So the transitive and the intransitive one are both hapax legomena. 🙂

  200. the basic senses ᴋɪʟʟ and ᴅɪᴇ are related as a causative / noncausative pair

    What is the difference from a transitive / nontransitive pair? One is a semantic distinction, the other is a lexical distinction? So what kind of mapping are you going to have from semantic to lexical? What kind of semantic-cum-lexical-cum-syntactic theory is going to explain the derivations of “he killed him”, “he caused his death”, “he sentenced him to death”, “he had him put to death”, etc. They were not talking about a theory of pure semantics; they wanted to do away with any kind of Deep (syntactic) Structure and derive surface structures from semantic structures.

    Chomsky was always weak on morphology but I can see why he wanted to introduce a “Lexicalist” hypothesis. It’s a pity, really, that Generative Semantics was nipped in the bud because we might have seen a clearer working out of these ideas. They might have been interesting; they might also have led to a profusion of competing proposals that would have put Chomsky’s later versions of generative grammar, not to mention those of competing schools of generative grammar, to shame. Perhaps I suffer from lack of imagination but the existence of a very large number of competing approaches to explain “Language” has always suggested to me that we can only explain parts of the system. A grand theory to unite them all always seems to stumble over basic details.

  201. The notion of “cause” is, yes, probably also somewhat blunt concept for this and in a formal analysis would have to be unpacked to include something something intentionality
    I don’t think that intentionality is necessary – one can say e.g. “Smoking kills X number of people every year” without assuming a concept like smoking to have intentions.
    The German version of “unintentionally” for children is (or at least was when I was a child) das hab ich nicht extra gemacht / das war nicht extra

  202. That’s odd. Is “extra” short for something?

  203. Is “extra” short for something?

    It’s long for X

  204. Stu Clayton says

    That’s odd. Is “extra” short for something?

    Here it’s just colloquial for “on purpose”. When you do something on purpose it’s “in addition” to just being alive. Das hat er extra gemacht! = “he did that deliberately!”.

    The idea may go back to Schopenhauer: Die Welt als Wille und Couchkartoffel.

  205. Thanks!

  206. …And now that I bother to check, I see that sense is in my big German dictionary.

  207. David Marjanović says

    das hab ich nicht extra gemacht / das war nicht extra

    While that’s intelligible, it’s not something I’ve heard without more context. Rather (das) wollte ich nicht or maybe das war nicht (mit) Absicht.

    (Now I’m wondering if that’s why my dialect keeps simple-past wollte around.)

  208. Stu Clayton says

    (Now I’m wondering if that’s why my dialect keeps simple-past wollte around.)

    Das wollte ich nicht! is something you often hear in Germany. It is in no way dialectic.

    Edit: nor dielectric.

  209. David Marjanović says

    Yes, it isn’t dialectal; my point is that the tense that looks like the English simple past is extinct in the Upper German dialects, with occasional exceptions for a few individual verbs, and in my case (probably throughout most of Bavarian) those are “be” and “want”. One more irregularity in “be” is not surprising, but why “want”? Its simple past isn’t even irregular to begin with.

  210. David Eddyshaw says

    Afrikaans, which has otherwise ditched the simple past, preserves it only in was “was”, had “had”, unsurprisingly, and (more surprisingly) in dag/dog “thought” and wis “knew.” Apparently had and wis are moribund, but dag/dog remains very common. (Purloined from Bruce Donaldson’s MGL grammar.)

  211. Lars Mathiesen says

    According to Ringe, the present of “will” in sufficiently old Germanic looks like the sole survival of the IE optative in that branch. He supposes that people mostly used the optative out of politeness, so that “became” the unmarked present.

    I know this has absolutely nothing to do with the current thread.

  212. I know this has absolutely nothing to do with the current thread.

    Sockpuppet alert!

    If this were the real Lars Mathiesen he would know that this is a meaningless disclaimer on LH.

  213. никопёр (a term from early days of Russian Internet)….

  214. That is, somene who прёт (imp, perfective спёр, упёр) чужие ники (nick[name]s).

  215. I suspect that because a German modal like wollen has a irregular present inflection—in particular, without a –t ending in the third person singular—the simple past may be more likely to survive. This may be due to there being less phonological interference between wollte and will that there would be between wolte and the regular *wollt. It may also be that a with verb that is already highly irregular in unavoidable present forms, it just feels more natural to speakers to preserve other unusual forms than it would be for a completely regular verb; the verb is already a special case, so it may seem to make sense for it to be atypical in other ways. (This is probably part of why the simple past of sein also sticks around.) Finally, some of the modal pasts have specialized senses of their own (a feature shared in general shape, although not in detail, with English). You aren’t going to get very far without the imperfect subjunctive möchte, for example.

    Interestingly, these considerations seem to bleed over into the rules for other verbs that are similar phonetically but not in meaning. There is evidence on the Web that the existence of the simple past wollte for wollen licenses the simple past wöllte for the unrelated verb wöllen, which means “cough up a hairball” for some speakers.

  216. Lars Mathiesen says

    Actually, on LH that is tantamount to trolling — but only the regulars know that. So who did it, if it wasn’t me?

  217. PlasticPaddy says

    @lars
    Unless the troll spoofed or hijacked your email address, admin could find out. Perhaps AJP is making a St. Patrick’s day joke from the other side.

  218. Oh, it was Lars Mathiesen, all right. But (drumroll)… was it Lars (the original one)??

  219. Lars Mathiesen says

    What exactly do you mean by “was” here? Akismet thought the original Lars was a spammer and you see this post innit.

  220. David Eddyshaw says

    The Original One Returns!

  221. Lars Mathiesen says

    Now Akismet just needs to get a Centaurian filter. (Or is that Alphan?)

  222. David Eddyshaw says

    Pah! We scorn your puny Earthling filters!

    I mean, they scorn your puny Earthling filters.

  223. David Marjanović says

    This may be due to there being less phonological interference between wollte and will that there would be between wollte and the regular *wollt.

    Yes, and I should have thought of that, but all the other preterito-presents have lost their simple past nonetheless.

    You aren’t going to get very far without the imperfect subjunctive möchte, for example.

    Sure, but the use of the imperfect subjunctive is even expanded in the same dialects, precisely because there’s nothing to confuse them with anymore.

    licenses the simple past wöllte for the unrelated verb wöllen, which means “cough up a hairball” for some speakers

    Oh! So that’s where Gewölle “(owl’s) pellet” comes from! I’d been wondering if it was related to Wolle “wool”, closer to “hairball” in other words…

    Sure, if wöllen is regular, it must go on wöllte, gewöllt. Is it woll, gewollen? …No, Wiktionary and its sources say it’s regular.

    (I suppose it’s the former causative of wallen “well”; the MHG and OHG attestations of the root vowel are chaotic enough to probably allow that.)

  224. @DM: Do you have some examples for the use of the simple past of wollen in your dialect? Can they be distinguished from subjunctive uses like in (Standard German) Ich wollt ich wär ein Huhn?

  225. David Marjanović says

    Do you have some examples

    The basic way to say “oops, sorry” is /ˈvoɪ̯tɪnɛd/ – [das] wollte ich nicht.

    The subjunctive is either done analytically or gets double-marked, 1/3sg /ˈvoɪ̯tɐd/ *wollt-te.

    Incidentally, many people avoid the 2pl of the simple past: of the two logically possible forms, /ˈvoɪ̯t͡s/ is identical to the 2pl present, /ˈvoɪ̯tɐt͡s/ is identical to the subjunctive. I heard the latter used last summer, though.

    Ich wollt ich wär ein Huhn

    I wonder how widespread this construction is. I haven’t encountered it spoken, and what sense does “I would wish I were a chicken” make? It must have been widespread somewhere at some point, ich wünschte is common in literature, but…

  226. Lars Mathiesen says

    Swedes can pile three politenesses in a heap: Jag skulle gärna vilja ha en glass is what the kids are taught to say.

  227. Stu Clayton says

    Ich wollt ich wär ein Huhn … I wonder how widespread this construction is

    It’s a hapax aeidomenon.

  228. Stu Clayton says

    Or aeidon ? aeison ?? Sung at any rate.

    The fabulous lyrics are chock-a-block with subjunctives. No better way to tart up your German:

    # Ich wollt, ich wär ein Huhn
    Ich hätt nicht viel zu tun
    Ich legte vormittags ein Ei
    Und abends wär ich frei

    Mich lockte auf der Welt
    Kein Ruhm mehr und kein Geld
    Und fände ich das große Los
    Dann fräße ich es bloß

    … #

    I find it’s from the 1936 film Glückskinder. Thanks to Hans for this trip down memory lane at one remove !

  229. Gern geschehen 🙂
    @DM: Thanks for the example!

  230. John Cowan says

    /ˈvoɪ̯tɪnɛd/ – [das] wollte ich nicht

    Voilà une langue admirable, que ce turc!

  231. Plus admirable qu’on ne peut croire. Savez-vous bien ce que veut dire cacaracamouchen?

  232. John Cowan says

    No indeed.

  233. J.W. Brewer says

    I’m thinking about buying a copy of this and would be interested in hearing from anyone who has read it about the following concern: as I take it the principal “victims” of the Chomsky faction are depicted as the GS folks who were themselves generativists and thus (although I guess some of them subsequently repented) also Part of the Problem. To use a potentially insensitive analogy, I am not particularly fond of histories of Stalinism that focus primarily on the Trotskyites and Old Bolsheviks who fell afoul of him as opposed to the more innocent victims who had had the good sense never to be Commies to start with. How much discussion is there of the marginalization within the profession (in the U.S. at least) of overtly non-generativist perspectives as opposed to the negative experience of saboteurs, wreckers, and right deviationists within a fundamentally generativist paradigm?

  234. An excellent question and comparison!

  235. Indeed an interesting (and generlizable!) question. I guess, there is a bit of a “human interest story” in this. Suppose, people couldn’t have expected to study syntax or some part of it as a job outside GS. As long as it was well-known only brave (or foolhardy) people would try and the rest would find another interesting topic to build a career in. And the loss would be to science, not to people. On the other hand, people who engaged with the system would suffer a personal (for scientists, professional is personal, duh) loss, which is inherently more heartstringspullingworthy.

  236. John Robert Ross is one to conjure with. The most interesting and counterintuitive insight of generative grammar – the discovery of “islands” – is his work. I’d like to learn more about his post-generative career.

    The first thing to say, I think, is that Wikipedia describes him as a “poet and linguist”, in that order. The Wikipedians don’t actually call him a terminologist, but give evidence in their second paragraph:

    Ross is also well known for his onomastic fecundity; he has coined many new terms describing syntactic phenomena that are well known to this day, including copula switch, Do-Gobbling, freeze(s), gapping, heavy NP shift, (inner) islands, myopia, the penthouse principle, pied piping, pruning, scrambling, siamese sentences, sluicing, slifting, sloppy identity, squib, squishes and syntactic islands.

    About half of these terms are linked. After being banished from MIT, he seems to have spent the rest of his career at North Texas University (formally the University of North Texas), which is in Denton and is almost as far north as you can go in the non-Panhandle (eastern) parts of the state. There he taught courses in, inter alia, Linguistics and Literature, Syntax, Field Methods, History of English, and Semantics and Pragmatics until retiring in 2021; he also oversaw the doctoral program in Poetics. I am glad to see Field Methods in this list, and wonder how much that class overlapped with Syntax, as I can hardly think of a better way to teach syntax.

Speak Your Mind

*