GIFTED LINGUISTS.

Mark Liberman has a post at the Log in which he waxes wroth about what he calls a “bizarre meme” by which “every piece of linguistic research is spun as a challenge to ‘universal grammar’.” I wasn’t transfixed by it (my throwaway comment: “I, on the other hand, welcome the new wave of anti-’universal grammar’ spinners”), but the dreaded name of Chomsky came up, and Dominik Lukeš wrote: “Something about what the statement ‘Chomsky was a massively gifted linguist’ by the other Mark P rubbed me the wrong way. I was trying to figure out in writing what the reason for my discomfort might be but it got a bit long and a bit too off topic, so I wrote a separate blog post about it.” I urge anyone interested in modern linguistics to read his post; not only does he explain (as his post title says) “Why Chomsky doesn’t count as a gifted linguist,” but he goes on to ask “So who deserves the label ‘gifted linguist’ defined as somebody who repeatedly elucidates legitimate language phenomena in a way that is relevant across areas of inquiry?” He discusses the work of MAK Halliday, Roman Jakobson, Charles Fillmore, William Labov, his “personal favorite linguist” Michael Hoey, and William Croft, whose Radical Construction Grammar is “probably the most interesting and innovative view of language that has come about since de Saussure.” For anyone who, like me, hasn’t been keeping up with the field for a while, it’s a great source for further exploration.
And our own gifted linguist, marie-lucie, has an excellent comment from which I extract this eye-opening passage:

I recently attended a presentation which centered on some syntactic structures in a language I know quite well, or rather, on translations into that language of complex English sentences, for which the consultant had obviously tried to please the linguist by coming up with sentences that were quite ungrammatical in her language – while she would have been quite capable of describing the (odd) situations presented if she had responded naturally with the (differently structured) resources of her own language.
This type of problem is one that linguists should always be aware of: a “wonderful” consultant who can always be counted on to come up with a translation may be linguistically imaginative rather than displaying models of her native grammatical competence. The modalities of thus adapting to the English structures could be a valid subject of study, if recognized by the linguist, but it is very misleading to describe such adaptations as spontaneous utterances typical of the structure of the language. Sentences thus obtained, which have no parallel in those naturally occurring in spontaneous utterances or in texts, should be very suspicious, especially if some of the features are quite at odds with those independently described in works on the same language.


By the way, there’s been an interesting and irritating discussion going on at the Log about the meaning of most: interesting because it turns out there are two very different understandings of the word (some, like me, think it means ‘more than half’; a lot of people, far more than I would have guessed, think it means ‘way more than half, an overwhelming majority’), and irritating because the “way more than half” group doesn’t seem to want to believe the rest of us are telling the truth (“And, so, while it’s a little sketchy to take 51.4% and call it ‘most’, doing so serves the argumentative ends of the writer”; my response: “It is not ‘sketchy’ if that’s what it means to you!”). I haven’t blogged about it because there doesn’t seem to be much to say except “This is what it means to me!” “Well, this is what it means to me!” But if you want to follow it, here are the posts: Most, Most examples, Most and many.
Update. See now this Log post, in which Mark Liberman quotes from publications that “provide a variety of (mostly perceptual) evidence for the view that most really does mean ‘more than half’, while offering a greater variety of theories about the strategies that (different sorts of) people use to determine whether this is true in particular cases.” If you’re interested in the topic, don’t miss it.

Comments

  1. Athel Cornish-Bowden says:

    marie-lucie’s comment is indeed excellent, but I wonder if she could be persuaded to amplify it a bit for us non-linguists who follow your excellent blog. Is it is possible to illustrate it with invented examples from languages like French or Spanish (or Russian, for that matter, though I’m not sure my Russian will be up to understanding the examples)? If it can only be illustrated with examples from Georgian or Algonquian fear she will lose most of us. I understand, of course, her main point that if universal grammar had been introduced by a speaker of a language very different from English it would probably have been presented very differently. But maybe among the western European languages one can still construct examples.

  2. marie-lucie says:

    LH, thank you for quoting from my comment on LLog. The name of the language and of the presenter have been omitted in order to protect the guilty. When I expressed my opinion on the reliability of the examples used, I was treated to a tirade in which I was accused of “denying the whole premise of generative grammar” (basically, that all languages have the same structure) – precisely.
    In addition to “Chomsky is a massively gifted linguist”, “the other Mark P” wrote that “Chomksy had to face massive hostility” at the beginning of his career, a comment to which I wrote a separate reply giving a reference which demonstrates that the opposite was true: Chomsky was actively supported by some of his professors and given opportunities that other beginning scholars did not have.
    I also second Dominic Lukes’ opinion of William Croft’s Radical construction grammar, which is based on Charles Fillmore’s earlier work. I confess that I had never heard of Michael Hoey but will try to become better acquainted.

  3. I was accused of “denying the whole premise of generative grammar”
    Heh. How dare you!

  4. doesn’t seem to want to believe the rest of us are telling the truth
    I agree with you in this instance, but it does happen that people lie about their grammatical perceptions. For instance, those who, to make some kind of classist/prescriptivist point, claim they literally understand the “double negatives” – constructs they’ve been hearing all their lives from working-class speakers and in popular songs – as meaning the opposite of what the speaker intended. The “So, when you say ‘I can’t get no satisfaction’, you mean you can get some satisfaction?” kind of snark.

  5. Athel Cornish-Bowden says:

    Maybe I should invent an example of my own, but I doubt whether it will stand up to serious scrutiny. About a quarter of a century ago I made a not very energetic attempt to learn some Welsh, and was struck by how many words (very short words, to be sure) were needed to express some simple ideas. To say I love you in English requires three words, as also in French (Je t’aime), but it can be said in two words in Spanish (Te quiero) and only needs one in Hungarian (Szeretlek — I cheated by getting that from Google Translate, but it seems to be what I remember). Google Translate offers the compact Cara ‘ch for the Welsh, but what I remember from my Teach Yourself Welsh book was much longer, a good seven words — something like Y’r wyf yn dy garw dyn. Google Translate tells me that that is indeed Welsh, but says it means I am your man rough indication, so doubtless I don’t remember it correctly.
    But no matter. Suppose that my recollection is more or less correct, and that its literal word-by-word translation into English is what I remember, something like There is a loving of me to you. That is clearly not something a native English speaker would say, but it could conceivably be something a Welsh speaker might produce in trying to translate the Welsh into English.
    Does this make any sense? Is it too far removed from what marie-lucie was saying?
    Trying a different sort of example, the following is a quotation from an English “translation” of a rather important couple of sentences from Karl-Friedrich Gauss (1809): Since this cannot be known a priori, we will, approaching the subject from another point of view, inquire upon what function, tacitly, as it were, assumed as a base, the common principle, the excellence of which is generally acknowledged, depends. It has been customary certainly to regard as an axiom the hypothesis that if any quantity has been determined by several direct observations, made under the same circumstances and with equal care, the arithmetic mean of the observations affords the most probable value, if not rigorously, yet very nearly at least, so that it is always most safe to adhere to it. I think it would be safe to assume that this was not intended to resemble anything that would occur naturally in “spontaneous utterances or texts” in English!

  6. But it seems safe to say that the sentences of Gauss’ being translated here don’t closely resemble “spontaneous utterances” in German either. No? Writing about mathematics and philosophy necessarily involves semantic constructions alien to everyday speech, I should think.

  7. marie-lucie says:

    Athel, about your second comment and the German translation, it is true that long, convoluted sentences can be hard to follow, especially if they refer to abstract notions, but you can find sentences quite as complex as that one in nineteenth-century literature, for instance in Dickens (and Proust is probably the most famous example). Such sentences, while not “conversational”, are not translations from another language. Besides, the information in long, complex sentences can usually be expressed in simpler form, shorter chunks of information being packed in shorter, more basic sentences, regardless of the topic.

  8. I agree with you in this instance, but it does happen that people lie about their grammatical perceptions.
    Oh, absolutely.

  9. marie-lucie says:

    Athel, here is a simple example:
    In English, the adverb (such as often/usually/basically/normally/habitually, etc) comes before the verb, while the Object pronoun comes after it:
    - I often/usually/normally see them at the market.
    In French, the adverb comes after the verb, but the Object pronoun comes before the verb:
    - Je les vois souvent/généralement/normalement au marché.
    But Chomskyan descriptions of French place those components of the sentence in the positions where they occur in English (since those are assumed to be the “universal” rules), and separate, secondary, language-specific rules are needed to “move” them to the positions where they actually occur in French. Why not just say, for instance, that the French adverb is normally placed after the verb, the English adverb before the verb? – Ah, but having two basic word orders would violate the universal rule!
    Awkward, stylistically unusual, barely grammatical sentences are not hard to find in translations, even to and from widely known languages with many millions of speakers, but educated speakers of majority languages are usually clear about what they would say and what sounds right to them. But take a speaker of a small, low-status minority language with little tradition of literacy (eg Acadian French) called upon to translate (or accept the proferred translation of) a complex English sentence for a high-status person such as a university professor, whose attitude seems to imply that if the language cannot replicate an English structure, there must be something wrong with it, it is no wonder that some consultants will do their best to produce something that will conform to the structure of that sentence, or accept the linguist’s translation, no matter that they themselves would never spontaneously produce a sentence that goes against their native intuition.
    I am not saying that there is no merit at all in transformational-generative grammar: it is very good for most of English (not surprisingly since it is based on English structure), and for teachers of English (as first or second language) it is very useful in order to understand and explain the structure of phrases and sentences, and how and why you can or cannot expand short ones and reduce complex ones. But beyond the basics, and applied to very different languages, it just gets more and more complex and convoluted, without adding genuine insights. Looking at an article written within this framework is somewhat like reading math, with sentence diagrams in which individual words are placed in strange positions among alien symbols and lines, so that it is difficult to even recognize what the sentence is that corresponds to the diagram.
    There is a series called “Outstanding dissertations in linguistics”, which does not draw from across the discipline of linguistics but is exclusively devoted to transformationalgenerative grammar. After a few years most of these “outstanding” works are unreadable, because the theory has moved on to more and more complexity, generating new vocabulary and formalism which will soon be useless, even if the examples (usually variations on just a few sentences) are grammatical in the language(s) considered.

  10. marie-lucie says:

    [The Welsh for "i love you" is] something like There is a loving of me to you. That is clearly not something a native English speaker would say, but it could conceivably be something a Welsh speaker might produce in trying to translate the Welsh into English. – Does this make any sense? Is it too far removed from what marie-lucie was saying?
    It is not far removed from what I was saying, but it is the opposite: suppose a Welsh linguist largely ignorant of English was trying “There is a loving from me to you” on an insecure speaker of English, and the English speaker accepted that sentence as perfectly good and natural in English, not daring to contradict the Welsh linguist, who then wrote an article about English syntax based on similar examples (and the article was then quoted in an authoritative reference work!).

  11. Athel, where does that passage occur ? I can guess why you call them a “rather important couple of sentences”, that’s why I want to check. It would surprise me to find Gauss juxtaposing Axiom and Hypothese in German with such insouciance as in “to regard as an axiom the hypothesis that …”. But maybe that’s precisely the interesting bit !?

  12. Latin, not German.

  13. I was taught Latin grammar in college using a generative method, which annoyed me no end, since traditional grammar, which was after all designed for Latin, would have been so much easier, at least for me. The teacher – who was also one of the authors of the method – was excellent, but it was one of the things (along with a greater interest in Greek) that kept me away from Latin for years.
    On the philosophical level, I always thought Chomsky’s tenets were implausible, at least in the particular form he gave them. The whole Cartesian, innate structures thing. But I know there must be tons written about that (there already was decades ago when I last looked).

  14. This guy says (bottom of page 141) that Gauss’s “argument was a logical aberration—it was essentially both circular and non sequitur.”

  15. Thanx, MMcM. He writes: Axiomatis scilicet loco haberi solet hypothesis, si quae quantitas …. I am not familiar with the force of the word axiomatis around that time, but it looks like I got excited for the wrong reasons. In any case, from 177 (containing the passage) to 179 he seems to be showing (as far as I can tell …) that a system of variables can be regarded as assuming its most probable values where the sum-of-squared-deviations is minimal. Or something like that …

  16. Yeah, but read on, Mr. Hat, page 143: “Before seeing Gauss’ book Laplace had not seen any connection between the [central] limit theorem and linear estimation, but almost immediately afterward he could see how it all fit together. … And once Gauss’ choice of curve was given a rational basis, the entire development of least squares fell into place, just as Gauss had showed”.
    A few sentences later, there is talk of “bisecting posteriors”. Have you been reading Sade again ??

  17. > […] the “way more than half” group doesn’t seem to want to believe the rest of us are telling the truth (“And, so, while it’s a little sketchy to take 51.4% and call it ‘most’, doing so serves the argumentative ends of the writer”; my response: “It is not ‘sketchy’ if that’s what it means to you!”).
    Hey now, I’m part of the “way more than half” lot, and I had no difficulty believing the “more than half” lot.
    But don’t take it too hard: in general, commenters at Language Log seem to have difficulty internalizing the fact that not all of their audience shares their dialect/idiolect/whathaveyou, even when a relevant *lectal difference is exactly the topic of discussion. For example, I’ve seen many comments of the form “you pronounce X like Y? how weird; I pronounce it like Z!” from people who evidently haven’t yet wrapped their heads around the fact that X, Y, and Z are all homophones for some speakers.

  18. An anecdote:
    A linguist is working with his consultant who speaks a heretofore undocumented language. They have been diligently working for some time and the linguist has built up his theory of the language all the way to morphology or syntax or something.
    The linguist is feeling confident about his ability in the language at this point, he is providing example sentences already translated into the language and checking them with his consultant.
    “Okay, can I say ‘who ate what yesterday’?”
    “Yeah, you could say that.”
    “Great. Then how about ‘what was eaten by who yesterday’?”
    “Hmm, sure, you could say that.”
    “Okay, how about ‘who saw what yesterday’, can you say that?”
    “No.”
    “Really?”
    “Yeah. I’d never say that.”
    “But you could say ‘who ate what yesterday’?”
    “No.”
    “But you just told me that was okay!”
    “Well sure, it’s okay for you. But I would never say it.”
    “So why is it okay for me to say it?”
    “Because you foreigners can talk however you want. All these people come here to talk with us, they can say things any old way. But us, we only say things one way.”
    The lesson? There is almost always a difference between how a non-native speaker is allowed to talk versus how a native speaker is allowed to talk. Non-native speakers can say all sorts of ungrammatical things that native speakers would never say. And since linguists are rarely native speakers in these situations, they are allowed ungrammaticality. A naïve linguist who doesn’t pay attention to this distinction will end up with a notebook full of ungrammatical nonsense that he or she honestly believes is correct. And there are a lot more naïve linguists out there than there are ones who know better.

  19. I should note that the lesson runs deeper than just the anecdote, for those who have no experience in elicitation and fieldwork. The point is that you should always check your data for naturalness as well as for grammaticality. Unnatural speech must be noted as the exception. Theoretical linguists often overlook naturalness, figuring that anything grammatical is good enough. But the fact is there are a lot of utterances that can be produced by a grammar which are totally abnormal (either ungrammatical or infelicitous or both) and would never be spoken by a native speaker in real life.

  20. I love listening to fieldwork stories. Thanks!

  21. But the fact is there are a lot of utterances that can be produced by a grammar which are totally abnormal (either ungrammatical or infelicitous or both) and would never be spoken by a native speaker in real life.
    But that’s true of “normal” grammars as well. I believe Chomsky gave a famous example …

  22. Oops, I misparsed the N to which “normal” applies. But what criteria might be applied to determine normality in a language you don’t know ? You can’t rely on the consultant – that’s the point of introducing the notion of “normality”, isn’t it ?

  23. You have to rely on the consultant; the problem is getting an honest/correct read of what’s normal (i.e., “what would you, the native speaker, say” as opposed to “what would you let me, the ignorant linguist, get away with saying”).

  24. Er, assuming by “consultant” you mean informant.

  25. For what it’s worth I can see both sides of the ‘most’ argument. In my normal usage ‘most’ means ‘nearly all’, ‘way more than half’, but in more formal contexts, especially those influenced by science, maths or statistics, then I would use/understand it to mean more than 50%.

  26. A late friend of mine (not a linguist) wrote an amusing article called “Chomping on Chomsky” in which he shows what could happen if your “informant” were determined to make sense of any utterance. (can’t link but it’s the 2nd result if you google the title in quotes).

  27. marie-lucie says:

    Er, assuming by “consultant” you mean informant.
    The use of “informant” in a linguistic context is discouraged in favour of “consultant”, which does not have the low status and negative police connotations of “informer” with which some people confuse it.
    “normality”
    Many sentences are normal because they occur in the normal course of living in a culture. Such are the greetings that are presented in the first lesson of most second-language textbooks. For instance, in English many people say “How are you?” without being in the least interested in the details of the other person’s life and health, and unless you know a person well and have not seen them for a long time, it is inappropriate to answer in detail. In some other cultures you are expected to ask not just about the other person’s health, etc but about a dozen relatives. I have read that in one Southeast Asian country (I forgot which) people meeting in the morning ask each other “Have you had your bath yet?”, a routine greeting which would sound extremely rude between English speakers, as it would seem to imply criticism of the hygiene standards of the addressee.
    Linguists also tend to make up lists of sentences of a certain structure, not always realizing the real life contexts under which they could credibly be said. I heard of a linguist asking for translations of sentences such as “This is the —- that I —–ed”: fine, but among sentences such as “This is the seal I shot” there was also “These are the fish I ate”, at which point the consultant got up in disgust and refused to have any more to do with that particular linguist.

  28. Liben Cruber says:

    I learned that the human ability to act linguistically was biologically triggered at the times of the development of human consciousness. Syncretic language (well ‘grammatically’ structured!) served the purpose for few hundreds thousand years of its successful practice. Then, its vocal elements triggered and developed in the new physiological phenomenon: human speech apparatus and the ability to vocalise pre-speech routines on the base of new audio-vocal semiotic structure (modern grammar).
    Do you think, talking about Language evolution and universals, we have to discard very long period of the pre-speech language ability and concentrate scientific effort only on relatively new speech (medium) driven grammatical tendencies?
    It looks for me as a study of a horse anatomy on the merit of the modern traffic rules.
    You can read in Russian: http://scipeople.com/publication/99479/

  29. LH, see what Marie-Lucie writes above:
    I am not saying that there is no merit at all in transformational-generative grammar [TGG]: it is very good for most of English …
    I am pleased to find this, because it provides an example concerning most for a point I want to make here – rather than in the relevant threads at Language Log, a forum in which the blogmagisterial tone has become too rude and high-handed for sound discussion.
    LH, would it be reasonable for Marie-Lucie to write what she did if she thought TGG applied to “51% of English”? Then shift to more standard cases (because it is awkward to speak of percentages of a language!). Suppose we have a slightly biased coin C, which lands head-up in 50.1% of throws. Consider these assertions:
    A1. Most throws of C yield heads.
    A2. (?)Most throws of C yield heads, but only just.
    A3. (*)Nearly most throws of C yield tails.
    A4. C lands heads most of the time.
    A5. Most of the time you’ll win if you choose heads for C.
    Consider also, for a beach 50.1% of which is covered with oil:
    A6. Most of the beach is covered with oil.
    A7. (*?)Not quite most of the beach is free of oil.
    The overwhelming majority (“overwhelmingly most”?) of examples considered at Language Log concern fixed numeric results rather than probabilistic projections. They also concern count terms rather than mass terms (like time, and most problematically English; and with looseness in terminology, cases like the beach, where a beach is considered as a continuously divisible whole).
    Do assertions A1 to A7, in the contexts provided, make it seem more acceptable to restrict the sense of most? I think they might. In my own opinion, the “meaning” of most is an equivocal matter. There is a plain alethic question, for which the meaning “>50%” is quite plausible; and there are questions of assertability and implicature, for which “most” of us assume the vague meaning “contextually, significantly more than 50%”. This latter looks right when we reflect on the oddity of A2, A3, and A7. If the context-free truth value of these three assertions were all that mattered, we surely would not be moved to question them.
    So I am not surprised at either preference for “the” meaning of most. But for the record, I personally would not make assertions like A1 to A7, where 50.1% is the proportion. Would you, LH?

  30. Southeast Asian country
    Indonesia: sudah mandi [belum]?

  31. marie-lucie says:

    Perhaps I made a non-native error in writing “most of English”? I think I should have written “most of English structure”.
    For what it’s worth, I think that what I have internalized about English is what a majority of English Canadians would say (at least outside of phonetics). For me, “most” means “definitely more than 51%”, and I would refer to 51% as “slightly more than half” or “a slight majority”.

  32. Perhaps I made a non-native error in writing “most of English”? I think I should have written “most of English structure”.
    I think your usage was perfectly idiomatic, Marie-Lucie. In any case, my analysis would proceed essentially the same if you had written “most of English structure”.
    Addendum to my post:
    I would feel licensed to make assertions like this, assuming 50.1%:
    A8. It is true that most throws of C yield heads.
    But I’d be careful, because for some audiences A8 might still be as misleading as the original:
    A1. Most throws of C yield heads.

  33. The most gifted linguist I ever read or heard was George Carlin.

  34. Marie-Lucie writes:
    But Chomskyan descriptions of French place those components of the sentence in the positions where they occur in English (since those are assumed to be the “universal” rules), and separate, secondary, language-specific rules are needed to “move” them to the positions where they actually occur in French. Why not just say, for instance, that the French adverb is normally placed after the verb, the English adverb before the verb? – Ah, but having two basic word orders would violate the universal rule!
    I don’t think that’s an entirely fair critique. The reason generative descriptions of French put the pronoun after the verb and then move it to a position before the verb isn’t because the pronoun goes after the verb in English—it’s because other direct objects go after the verb in French. In other words, the movement is posited because of a difference between related structures within French, rather than because of a difference between French and English.
    (The situation of the adverb I don’t feel I have sufficient knowledge to discuss.)

  35. marie-lucie says:

    AJD, I simplified the situation slightly, but it is true that the SVO (Subject-Verb-Object order is taken as basic. Actually the fact that in French the Object pronoun occurs before the verb is not the result of an abstract “movement rule” taking the pronoun from its “underlying” position after the verb and placing it before it, it is a historical relic from Latin, a language which had the basic word order SOV (Subject-Object-Verb), which has been maintained in French, Spanish, Italian, etc as concerns the pronouns, while nouns were attached permanently to the position after the verb (a position which was possible in Latin, with a slightly different emphasis) as a result of the loss of the distinctive suffixes which indicated the S or O roles of the nouns, so that the loss of distinctiveness could cause ambiguity about who was affecting whom.

  36. But Marie-Lucie, that’s not an argument against the existence of an abstract “movement rule”. What you’ve written is a summary of how it historically came to be the case that French puts pronoun objects before the verb but noun objects after the verb, not an account of what’s the best way to analytically describe the current state of affairs. There’s no contradiction or conflict between saying “The fact that modern French puts pronoun objects before the verb is a relic from Latin word order” and saying “French grammar places pronoun objects before the verb via movement from the default position of objects after the verb.”

  37. Roger Depledge says:

    Language Log, a forum in which the blogmagisterial tone has become too rude and high-handed for sound discussion.
    Seconded.

  38. Except that the one saying (“modern French puts …”) is a resumé of historical change, while the other (“French grammar places …”) hypostasizes a mechanism that is unobservable by definition.
    Both expressions bear the marks of pathetic fallacy. But the first one can be cleaned up (transformed !), giving: “The fact that modern French-speaking people put pronoun objects before the verb is a relic from Latin word order”. The second expression can’t be cleaned up: *”French-speaking people place pronoun objects before the verb via movement from the default position of objects after the verb” is not going to encounter the assent of many French speakers, apart from French-speaking Chomskyites.

  39. the existence of an abstract “movement rule”
    AJD, that is plain-vanilla reification, what Whitehead called the fallacy of misplaced concreteness. A “movement rule” is a particular kind of convention for shuttling between discourse and description, but it doesn’t “exist”. I would agree that, as you put it, “there’s no contradiction or conflict” between the two expressions, provided you leave out the “existence” bit.
    It’s such a pity that so many people think that objectivity requires essentialism.

  40. Athel Cornish-Bowden says:

    marie-lucie: thank you for your patient and helpful explanation.
    Grumbly Stu: I found this 25 years ago and noted down the following reference: Gauss, K. F. (1809) Theoria motus corporus coelestium in sectionibus solem ambientum (trans. C. H. Davis), Section 177, pp. 257-259. Dover, New York, 1963 (republication of edition of 1857 by Little, Brown & Co.)
    I can’t easily check it now because there was no photocopier in the library where I found it and I no longer have access to the book. (However, it appears to be unnecessary as other comments seem to indicate that the Gauss’s original has been located: easier now than in 1986, when there was no Internet, at least, not in Chile.) This was a very serendipitous moment for me. I knew that Gauss had witten something along these lines, but I didn’t know where or how to find it. I was in a not very good academic library in Chile looking for something quite different when I saw a book on the shelves by Gauss. I took it down, expecting it to be in German and anyway irrelevant to what I might want to know, but to my great surprise I found that it was not only in English (of a sort) but also that it contained exactly the passage I’d been looking for for years.
    Getting away a bit from the linguistic theme, the importance of this passage is that it indicates that Gauss proved exactly the opposite of what he is widely thought to have proved: far from proving on the basis of a probability distribution that the arithmetic mean was the best measure of an average, he assumed it was best and on that basis defined a world in which it would be the best. I’m not sure if this answers your question about juxtaposing “axiomatic” and “hypothesis”, but I think it may do.
    MMcM: I realized that the title of Gauss’s book was in Latin, but the construction struck me as much more German than Latin, but maybe Gauss wrote Latin in a very Germanic style.

  41. he assumed it was best and on that basis defined a world in which it would be the best.
    By golly, Athel, it looks like my idea (“But maybe that’s precisely the interesting bit”) was going in the right direction after all ! Your take on the passage makes good sense to me.

  42. Athel Cornish-Bowden says:

    There is almost always a difference between how a non-native speaker is allowed to talk versus how a native speaker is allowed to talk.
    Something that is very obvious if you live (as I do) in a place where the everyday language is not your native language. No one in France fusses if I get a high proportion of my genders wrong (about 50% of nouns that are not very common words or ones that are not clearly linked to sex), but if they need to use the same word in their reply they say it correctly. Even if the words have quite different meanings, e.g. if I say “la foi” when I mean “le foie”, no one pretends not to understand what I mean.

  43. Sometimes it doesn’t even make a difference, as in the spiffy expression crise de foie.

  44. Meaning to lose your religion or your cookies after overindulgence in alcohol.

  45. I’m a quite gifted linguist.

  46. Noetica, your beach example is very interesting — in this context I think would say and agree with “Most of the beach is covered with oil” if even a small minority of the beach — say 10% — were covered with oil. It’s just a way of saying there’s a lot of oil on the beach. I think in general I understand “most” to mean “substantially more than half” but it’s highly dependent on context, there are certainly sentences where I would hear “most” and understand it to mean “more than half” or (as in this case) “any”.

  47. The use of “informant” in a linguistic context is discouraged in favour of “consultant”, which does not have the low status and negative police connotations of “informer” with which some people confuse it.
    How bizarre. And when “consultant” acquires the low status inevitable from the nature of the relationship, they’ll switch to “associate.” And then when that gets deprecated, “dearly beloved and honored coadjutor”… Sad that even linguists are prey to the delusion that words determine reality.
    Do assertions A1 to A7, in the contexts provided, make it seem more acceptable to restrict the sense of most?
    I don’t even know what that means. The only thing that restricts the sense of any word is how it’s used. Clearly, different people use most in different ways. They don’t have to justify their usage by any elaborate logic.
    But for the record, I personally would not make assertions like A1 to A7, where 50.1% is the proportion. Would you, LH?
    Yes, I would. Once again, I use most to mean ‘more than half.’ I am not bothered by any of the examples people make up to demonstrate that their usage is somehow the “real” one and that if the other faction would just open their eyes and think clearly, they would realize that they too adhere to it. This attitude is what’s bothering me about the discussion at the Log. Why can’t everyone simply accept that there are two different usages? I mean, I’m surprised by the “vast majority” folks, but I’m not trying to get them to recant, I’m just recalibrating my sense of the variety of English usage.

  48. Finally the voice of reasonableness pipes up again. I’m with you, Hat.

  49. low status
    Somehow “cunninglinguists” springs to mind.

  50. I remember our English master at school wouldn’t let us use the phrase “the majority” (of people in England drink gin, e.g.). He made us write “most”.

  51. PLR ebooks says:

    A linguist for me is a gifted human being, the talent itself surpasses everything I can imagine of.
    PLR
    make money online

  52. Surely you mean “most everything I can imagine of,” Mr. ebooks.

  53. LH:
    First, I point out that I wrote exactly this:

    I personally would not make assertions like A1 to A7, where 50.1% is the proportion.

    I did not say that those assertions fail to express truths.
    Now LH, you wrote:
    Once again, I use most to mean ‘more than half.’ I am not bothered by any of the examples people make up to demonstrate that their usage is somehow the “real” one and that if the other faction would just open their eyes and think clearly, they would realize that they too adhere to it.
    Hmm. I made no claim that most simply has one meaning, and simply does not have another. I don’t see how your second sentence follows from your first, nor why you issue the second sentence at all, in responding to what I write (if that was indeed your intention). My claim concerns pragmatics, and implicatures, though I appear not to have made the case clearly enough. Let me try again.
    I’m with John Cowan, who at the Other Place wrote this:

    Most people are men, but only by about 1%: the sex-ratio at birth doesn’t reflect the sex-ratio in the population as a whole.

    So he agrees with you, LH. And I agree also – as a matter of strict, purely alethic meaning. But John continues:

    And I think “Most people are men” is a very bizarre statement, as is “Most people are women”. For me too, “most” has a defeasible implicature of “much more than a majority”.

    Here I agree with John, and you plainly do not. I have put it a different way, and I pointed out an ambiguity that you do not accept. This respected source examines that ambiguity:

     … the word meaning is deployed ambiguously both in ordinary language and by theoreticians, …

    The author’s brief analysis of the divergences is instructive (and relevantly similar to my own inchoate foray); I commend it to interested blogmates. He then makes a procedural declaration:

    I will reserve the word “meaning” for that notion [a "semantic" notion, equivalent I think to my "alethic" conception of meaning].

    We would have to proceed with the same caution to made progress here. We have not done so; and there are similar deficiencies in the discussion at the other forum.

  54. marie-lucie says:

    (Object pronoun position in French)
    AJD, thank you for your comments. Grumbly, thank you for your replies to AJD.
    As a historically-minded linguist, I tend to prefer an explanation which has a historical basis to one that is an artefact of a theory. The first one is motivated. The second one is not: why, if post-verbal position is “the default position” for Objects, would a language ignore this default position and adopt a different order for a subset of Objects? “Because there is a movement rule” is not an explanation, since the “rule” is only formulated in order to deal with this particular exception to the “default position”. This rule does not even have a pragmatic justification such as the WH-movement rule which places question words in first position in a sentence, a position of prominence (not that this argument would matter for a UG explanation).
    Grumbly: the spiffy expression crise de foie.
    What’s so spiffy about it? It is a very ordinary expression.
    MMcM: Thank you for the Indonesian quote.

  55. marie-lucie says:

    LH: “informant” and “consultant”
    I understand what you are getting at, but the point was not to replace a word which had become negative with use among linguists (just the opposite: not everyone can be a good informant, so good ones are prized, and with severely endangered languages, any speaker is an extremely valuable person), but that many people, including speakers, were interpreting “informant” as “informer”, as in “police informer”. If a person is suspicious about a linguist’s motives for wanting to study their language, asking them to be an inform-something-or-other will not be the best way to gain their confidence and cooperation. And working with a linguist is not considered “low status” work.
    As far as I know, “consultant” does not have an unfortunate resemblance to a word with very negative connotations: it is justs the opposite, since “consultants” are experts in their field.

  56. What’s so spiffy about [crise de foie]? It is a very ordinary expression.
    marie-lucie, I would describe only a normal person as “spiffy” when he is dressed-up. Spiff is just polish on the ordinary, indeed mere spit-polish. An unusual person elegantly clad is not “spiffy”. But this is just my vocab.
    So I called crise de foie “spiffy” precisely because it is ordinary, and also because it is 1) a silly expression/notion, and 2) homophonous with the non-ludicrous expression/notion crise de foi. Crise de foie is silly at least when it is taken seriously, as some French people apparently do – as seriously as a crise de foi. When I have encountered it in this context of judicious second-guessing-the-doctor, crise de foie has usually appeared to me to be a high-falutin’ and evasive description of ordinary symptoms of hangover or gluttony. When it is used jocularly, my sense has been that the resemblance to crise de foi is part of the intended humor.
    I may be wrong about the connotations of the French. It’s just the way it has come across to me until now. I fully expect to be set straight on this – my hand is extended …

  57. I pointed out an ambiguity that you do not accept.
    I don’t accept it for myself, or for the many (most?) English speakers who use it as I do. I have no problem with anyone exploring a usage different than mine; in fact, I find it quite interesting. What I have a problem with is people talking as if their own usage, and the ambiguities and other issues that may flow therefrom, must be universal. I’m sorry if it sounds like I’m not being cautious enough, but I am, after all, a blogger rather than a logician.

  58. marie-lucie says:

    OK, Grumbly, let’s shake hands.
    I quite understand the word “spiffy” and use it sometimes: “My, you look spiffy today! What’s the occasion?” – that type of thing. But I don’t see anything “spiffy” about une crise de foie. I know it is a misnomer, but it is an everyday expression, no more spiffy than “heart attack” or “sick to my stomach” in English.
    I don’t read French works very often – mostly newspapers online – but I have never encountered crise de foi which looks suspiciously like a literal translation of “crisis of faith” which is quite common in American papers. I am sure there is a way of expressing the concept (not a direct translation), but I can’t remember it at the moment. Even translating literally, it seems to me that une crise de la foi would be better than une crise de foi. But in more classical style there would be a different phrasing, such as: Sa foi a souvent été mise à l’épreuve, surtout … “His/her faith has often been tested, especially …” or A l’époque j’ai eu une crise religieuse et j’ai failli perdre la foi “At that time I went through a religious crisis and I came close to losing my faith”, that sort of thing.
    Crise de foi is just too close to crise de foie, so even if you were talking about the Pope having gone through a crisis of faith, the first thing that would come to the mind of the person hearing une crise de foi would be that the Pope had had a digestive problem.

  59. Thanks, marie-lucie. The possibly imported crise de foi was messing me up – although on further consideration I’m not so sure I’ve encountered it that often. It seems to have become an idée fixe in me that it was genuine, and that it was related to the “foie” business.
    So everything crashes to the ground at once. I like that kind of demolition work – no reinforced walls still standing that cost even more to tear down.
    crise de la foi makes good sense.

  60. Weird. Like Grumbly Stu, crise de foi is an expression I run into from time to time (in metaphorical, not religious contexts) and never gave much thought. From Google it seems that while it’s by no means unknown in French, it’s about as common in English.
    But, strangely, it’s in Collins Dictionary.
    More strangely, the dictionary we used in high school has, s.v. crise, “crise de foi shaken faith,” and “crise du foie liver upset,” which latter seems suspect on two counts.

  61. marie-lucie says:

    The TLFI has numerous phrases for crise de, but foi is not there (foie is of course). There are also numerous phrases including foi, but not crise de foi.
    I am pretty sure that crise de foi must be a fairly recent translation of crisis of faith, which means (I think) that a believer is led by some intense experience to question their religious faith, or by extension their faith in some aspect of their belief system. But crise de in French means (for a person) an episode of personal upheaval during which the person is powerfully affected by something, such as bodily pain or an intense psychological experience. So une crise de foi (if genuinely French) would mean that the person went through an intense religious experience of believing, rather than questioning – eg for a Christian, having a vision of Christ, or something similarly powerful which would upset the person’s normal life for some time. But this type of experience would be called crise mystique, not crise de foi (which, according to MMcM, is not used in a religious context).

  62. marie-lucie says:

    “crise du foie liver upset,” which latter seems suspect on two counts.
    Indeed. crise de foie “digestive upset”.

  63. re Indonesian mandi (bath or bathe): see also here.

  64. fairly recent translation of crisis of faith
    There are enough occurrences in Google Books of crise de foi in French from the nineteenth Century that I’m inclined to believe that it is a possible spontaneous phrase. For instance, in a biography of Pascal. And, more interestingly, in some lectures entitled Crise de la foi: trois conférences philosophiques. The author, A-J. Gratry, was an immortel (in Voltaire’s seat), so it’s probably illegal for us to question the purity of his French. Likewise Zola.
    But these don’t give the appearance of fixed phrases. So you may well be right about its having picked up some steam later because of English.
    It’s also in there in Catalan, for what that’s worth.
    is not used in a religious context
    That might be overstating it. It’s not impossible there, it’s just that crisis of faith seems perfectly suitable for matters of faith, sensu stricto. For the set phrase crise de foi in English, Google Books results matched with socialism or communism seem to fit well with my intuition. Though I’m eager to hear what Grumbly Stu or anyone else who thinks it’s English supposes it’s used for.

  65. Feuerbach’s The Essence of Christianity has Krisis des Glaubens a few years earlier than crisis of faith in English, though I’m not sure there’s any direct connection. But it does seem to be a solidly mid-19th Century idea, which makes sense.

  66. marie-lucie says:

    Well, MMcM, so it is older than I thought! but I don’t remember ever running into it. And I wonder why the TLFI does not mention it, while it mentions tens of other examples of crise de and of foi separately.

  67. Grumbly Stu or anyone else who thinks it’s English
    I am not of the opinion that crise de foi is English, in the sense that it occurs often in English texts. I’m not saying it ain’t so, but I would not hazard such an opinion for a simple reason: I read relatively little English, primarily novels. My impression is that I have encountered the expression in French texts.
    However, the few non-novels in English that I have read over the last year, say, are precisely of a type where the author might adorn his prose with a phraselet in French: The Metaphysical Club and The Pound Era, for instance.
    In the net I also find crise de croyance (which by now sounds more natural than foi, but what do I know about French usage ?) and crise du péché:

    pour « l’honnête homme » du XVIIIe siècle selon Groethuysen comme pour « l’agréable Renan » selon Claudel, le péché n’existe plus, le péché n’existe pas. « Il n’y aura plus de péché » ; « le pécheur se meurt » ; « le péché se meurt », répète avec insistance Groethuysen dans les Origines. Il est advenu dans le cours du XVIIIe siècle, soutient-il textes à l’appui, une crise du péché : ce sera l’objet même de ce propos.

    As MMcM remarks, “crisis of faith” seems to be a solidly mid-19th century idea. Maybe we can take this claim a step further: the very notion of “crisis” has been inescapable since that time. Think of Charcot and crises hystériques, also crises motrices: où l’on décrivait soigneusement 4 stades: la période épileptoïde, la période du clownisme, la période des attitudes passionnelles et la période terminale avec souvent des hallucinations, où l’on recherchait des stigmates, où l’hypnose apportait son relent de magie. Later we have Husserl and the Krisis der Wissenschaften und die Transzendentale Phenomenologie, Sartre and the crise existentielle induced by trees (La nausée).
    Somehow it’s all rather histrionic: “crisis” and “climax” are not that far from each other in meaning. Not surprising, really: when you lose the plot, everybody tries to take center stage.

  68. Let me say that I belong to the camp that understands ‘most’ as meaning a large minority, not merely anything over 50%. I could, however, imagine a marketing research report in which ‘most’ was used in the meaning ‘a majority’, i.e., ‘more than 50%, and wonder whether the Hattian usage could have its origins in this…
    I also wonder whether this difference might not have caused some major misunderstandings in the past. If you said that most Americans shower in the mornings, I would assume that a large majority of Americans shower in the morning and treat those who shower at night as an interesting minority. But your meaning would be that a mathematical majority (perhaps a bare majority) shower in the mornings, and those who shower at night are just as much the ‘norm’ (for want of a better word) as those who shower in the morning. I am sure that this could lead to some heated arguments until it was realised that the leftover of Hat’s ‘most’ is not necessarily ‘some people’ but could conceivably be ‘almost as many people’.
    I also wonder whether Hattian usage is totally consistent, or whether the nuance depends on the context. For example, Billy Thorpe sang that ‘Most people I know think that I’m crazy’. For me it’s clear that he was generally thought of as crazy. Would Hattian usage include the interpretation that the community was pretty cleanly divided about his sanity?

  69. Phänomenologie

  70. marie-lucie says:

    Grumbly, the word crise is used more widely in French than “crisis”in English, for instance crise cardiaque “heart attack”, crise d’asthme “asthma attack”, crise de nerfs “hysterics”, crise de larmes (I am not sure how to translate this – it means suddenly bursting into tears and crying hard for a while – but not quite what you would call “a good cry” – “collapsing in tears”?), and dozens of others. As with English “attack” it is something that occurs suddenly and seems to be triggered by something from without (I mean outside) the person.

  71. crise de larmes
    How apposite ! That’s the kind of histrionics I was talking about. A fit of sobbing, a crying jag. Who can be the lachrymost.

  72. marie-lucie, why do you say “seems to be triggered by something from without (I mean outside) the person” ? All of your examples are of things that often have no apparent external cause.
    I don’t think of “heart attacks” as necessarily being “triggered” by anything particular. Sometimes they are, sometimes they aren’t. An attack of asthma can come on without apparent external cause, but rather as the result of emotional turmoil.
    Of course you may hold the view that every bodily dysfunction is always triggered by something external. But then I would wonder why you seem to want to distinguish the word crise in this connection, as if there were internally caused kinds of dysfunction after all.

  73. marie-lucie says:

    Grumbly, of course my examples are of sudden overwhelming episodes which don’t seem to have an outside cause, but to the person in the grip of one it seems to come out of the blue and to be beyond their control. Une crise d’épilepsie “an epilepsy attack” is another one. “Attack” in English certain suggests that whatever is happening is coming from somewhere else – it seems that way to the sufferer.
    Notice the number of times I used the word “seem” in these two paragraphs. I am talking about the state of mind of the sufferer, not about objectively verifiable things like their blood pressure or the pollen count or other external conditions and events.

  74. marie-lucie: I though you were saying it seemed to you (and anyone else as an observer of such crises, including the sufferer himself) that they had no internal cause and came “out of the blue”.
    Are you instead saying that crise de .. suggests the phenomenon as seen subjectively by the sufferer ? That’s very interesting. But your claims about the connotations of the word “attack” don’t fit with my use of it. Heart attacks often occur suddenly – but not always – yet “attack” for me has no connotations of external or internal cause. An “attack” here is just something sudden.
    Also, I don’t think of my body as being “under my control”. It regulates itself for the most part. “I” is just the cup-bearer.

  75. marie-lucie says:

    Grumbly, I think we are approaching the topic from different angles and therefore seeing different things. But I am too tired to argue further tonight! Sleep calls. See you tomorrow.

  76. Grumbly: The literal sense of “attack” is surely that of something coming from outside. The sensation of a major heart attack has been (aptly, I’m told) described as “an elephant stepping on your chest”, which if meant literally would certainly count as an attack!
    I have heard EFL speakers say “heartstroke” instead of “heart attack”, presumably a calque of something or other. “Stroke” = “blow”, again something from outside.

  77. the Hattian usage
    Perhaps a more objective term would be “the dictionary sense.” While I am perfectly happy to accept the “large majority” crowd (and I presume when you wrote “a large minority,” you meant “large majority”) as fully valid and paid-up members of the English-speaking community, I am not willing to have them redefine the situation so that their usage is normative and the traditional (and still, apparently, most common) use a personal outlier.

  78. And on Chomsky, I think it’s true that he got a lot of support and encouragement in the earlier part of his career. Syntax, after all, was not something that the structuralist tradition had a good story for, and if Chomsky did, so much the better — he was just doing normal science, extending the existing paradigm.
    It was generative phonology that really set the cat among the pigeons. Now the powers that be had to face really radical dislocations, the worst perhaps being the death of the phoneme. (Really, the anomalies in Russian phonology could have been accounted for perfectly well in traditional phonology: they were not such an epicycle as all that.) And it was quite clear from the beginning that Chomsky’s real argument was “We will bury you”, à la Khrushchev.
    Well, that’s what happened, and Chomsky has quite prudently ensured that no one will bury him by instead burying himself every decade or so. Admirable as a matter of tactics, and indeed of science (“Stand!” “So I do, against my will”). Chomsky is not to be blamed for the fanatical adherence of the Chomskyites, any more than Freud for the excesses of the Freudians; both have enough blame to carry for themselves.

  79. Chomsky is not to be blamed for the fanatical adherence of the Chomskyites, any more than Freud for the excesses of the Freudians
    Oh, I disagree; I think they both are to blame. It is their personalities and styles of behavior that encouraged, almost mandated, the presence of fanatical followers.

  80. And, I would add, in both cases the presence of fanatical followers was necessary to make up for the inherent deficiencies of the theories, which could not have triumphed on their own merits.

  81. John Cowan: The literal sense of “attack” is surely that of something coming from outside. The sensation of a major heart attack has been (aptly, I’m told) described as “an elephant stepping on your chest”, which if meant literally would certainly count as an attack!
    John, you’re right, in a way that is the “literal sense”, and one can even imagine that’s the explanation for the word being used to describe the experiences we’re discussing. However, it’s not a sense that I experience – for instance in January of this year, when a kidney stone suddenly started banging on the door, demanding to be let out.
    That is why I mentioned that I don’t think of my body as being “under my control”. I have as little control over so-called internal matters as so-called external ones (conventionally using the skin as boundary reference). All I’m saying is that when I hear someone has had a “heart attack”, that means for me merely that it was sudden – though not necessarily unexpected, say when the doctor had warned them to stop participating in those olympic bordello orgies. I’m not saying that other people should experience things and use words as I do, or face the consequences.
    If I felt an elephant stepping on my chest, I would first check to see if an elephant is indeed trying to step on my chest (am I on a safari, and dropped into a snooze at an inopportune moment ?). If so, I would try to roll to one side. If not, then I would try to get to a doctor before the Elephant Within got to me.
    What people mean by “having things under control” is probably more accurately described as “experiencing things as being familiar”. I do see advantages to this way I have of experiencing physiological crises – which is of course heavily conditioned by what I have heard about (“learned”). The advantage is that I can look about impartially for the culprit or culprits who are rocking the boat of familiarity. I am not hindered by a prior belief that it’s those fairies in the cauliflower again.

  82. marie-lucie says:

    I am old enough to have studied linguistics in the US and Canada on the cusp of the transition between the old and the new: Chomsky was well-known by then, but did not yet dominate the field to the point of driving out other points or view.
    Whether the advent of Chomksy was a revolution in linguistics or not, it is true that the discipline attracted a very different crowd once he became known. Before that, linguistics in the US had developed from the training and experience of anthropologists needing to describe undocumented or poorly documented languages that they would encounter in their natural environment. Linguists were people intrigued by other languages (and willing to learn them), other cultures, people in general.
    With Chomsky’s emphasis on “native speaker intuition” and on syntax (a good thing at first, since that aspect had been somewhat neglected by linguists, many of whom did not spend enough time with a language to really learn the syntactic subtleties), a different kind of student went into linguistics: a more “nerdy” kind who was glad to operate in a monolingual environment, rely on introspection rather than communication, and use a math-like formalism. Examples were no longer taken from actual utterances or texts displaying a range of syntactic structures in a different language, but created in order to “push the envelope” and test the limits of what could be said (hence the often weird examples in most TGG works – perhaps grammatically correct but pragmatically unusual or inappropriate). This new brand of linguistics student was also encouraged to take a very aggressive position in order to champion the new theory, going to linguistic meetings in which they not only disputed the presentations of the older scholars but did so in an insulting manner. Things have settled down somewhat but the more creative among Chomsky’s disciples ended up disagreeing with some aspects of theory and setting up their own brands, often less narrowly focused. And there have been a few people who were strong enough not to get caught up in the “revolution” and to pursue their own path. It is no coincidence that Fillmore (mentioned above) was at Berkeley, the site of a very strong centre for the study of Amerindian languages.
    The exclusiveness persists in some places: I have seen job ads (the few that are still open) in linguistics departments saying things like “the department subscribes to theory X and expects new members to pursue research in that framework”. This attitude will not expose their students to a wide range of possibilities, not encourage them to think outside the box, the usual exhortation to creativity. It is also very damaging to actual language study, leading to the type of thing I deplored in the comment that LH quoted above. As for the historical dimension of language, forget it in that context.

  83. An attack of asthma can come on without apparent external cause, but rather as the result of emotional turmoil.
    A girlfriend of mine some centuries ago used to suffer asthma attacks when things were getting a bit heated. Now I understand why…though there was both an external cause and emotional turmoil …

  84. In the presence of what one may be inclined to call “emotional turmoil”, it is moot whether some X might be plausibly identified as the “external cause”. I think that entertaining the very notion of “emotional turmoil” (and similar expressions) implies a tendency to play down external causes. External Xs are seen as mere candidates for triggering turmoil surge. In other words, external Xs that would not otherwise be regarded as sufficient causes are admitted to the status of triggers.
    So in such a situation, it may not have been the boyfriend’s fault after all … in the sense that pretty much any boyfriend would have sufficed to set off the girlfriend.

  85. marie-lucie says:

    Grumbly: If I felt an elephant stepping on my chest, I would first …
    Perhaps you have more “control” than most people (I mean the vast majority) and keep your cool in every situation, but there are many cases where people are totally overwhelmed by some experience and have reactions that surprise them, even if they have been saying to themselves “If such and such happened, I would …”. Some things are unprecendented in a person’s experience, and what they might have heard or read or thought about in advance often turns out to be irrelevant to how they actually feel and react when the event actually happens.

  86. If you require native speaker intuition, then there is a statistical problem. It’s not so much that monolingualism is encouraged, as that true bilingualism is required. And there just aren’t that many bilingual students of minority languages entering graduate studies in linguistics.
    But still, it’s not absolute. Chomsky’s early work was on Hebrew. The first full TGG was on Hidatsa; the first PhD thesis on Turkish.
    MIT was a center of Australian language studies under Ken Hale. Some of Wayne O’Neals’s students have been active in East Coast Native American language revivals.
    As far as I can tell as an outsider, Chomsky and his students welcomed Höskuldur Thráinsson’s stay at MIT and Harvard as they wrestled with some of the weirdness (from their point-of-view) of Modern Icelandic.
    Which is not to say that the premises are above suspicion or that the net effect isn’t as bleak as marie-lucie portrays.

  87. I agree, marie-lucie. I have conceded several of your points already, repeatedly.
    But I still don’t understand why it is being argued that it is unlikely that someone should experience these crises as a disturbance of equilibrium rather than an “external attack”. It’s rather peculiar that you write: “Perhaps you have more “control” than most people (I mean the vast majority) and keep your cool in every situation”, precisely when I have just several times written that I do not imagine I “have control” over things. This is not about keeping cool, but about persisting in looking sharply about you even when panicky and/or in great pain.
    I have given my example of January’s Shoot-out at the Renal Corral. A pity I didn’t have a video camera to hand, to record the simultaneous presence of pain, panic and goal-directed counteraction. This is not an unusual reponse in men, I dare say. Nor in women, provided there are no men around to tempt them into screaming helplessly.
    This is also not about never having unexpected reactions in situations which you might have thought about previously. I have had my full share of reacting unexpectedly to things that I had thought about a lot – sometimes reactions that persisted over years. It’s just that I don’t see this as losing any control that I previously imagined I had. I am not an autocontrol freak, but a leaf attempting a few aerodynamic improvements in the wind.
    It appears that I am unexpectedly learning quite a lot about how other people experience physiological crises, and why they think it is only right and proper to experience them in this way, and describe them in the way they do.
    Whatever I say is taken to be an attempt to change their minds about the subject, no matter how often I repeat that this is merely my way of experiencing things. Very interesting … though I haven’t yet figured out the possible cash value of it. There’s a kind of normative animus involved here, as in many of the heated debates about the proper uses of words and notions. of course I myself have been normatively animated on many an occasion.

  88. For what it’s worth, I second the comments by LH and m-l about the state of linguistics as a discipline since the 1970s, when I was a grad student at U. Hawai‘i, where documentation and fieldwork had not yet been eclipsed–and have in fact lately made a resounding comeback!
    On the subject of translationese, here’s a pertinent quote (with my apologies for the linguistic jargon) from a just-published review by Thomas E. Payne (U. Oregon) of a posthumously published book on field linguistics by Terry Crowley (who was truly a gifted linguist):
    This section reminded me of a situation I encountered in my own early fieldwork in Yagua. If one deals strictly with material elicited via Spanish, one could come to the conclusion that Yagua is an SVO, and strictly nominative-accusative language. However, after working with an extended body of texts, we realized that SV/SVO order hardly ever occurred in natural materials. Furthermore, there was a particular intransitive construction in which the S argument was expressed grammatically as an O. This construction would never have arisen in elicitation, as it occurred only at particular discourse junctures in narrative texts.

  89. Sorry. Forgot to paste the URL of the just-published review.

  90. One point that seems to be missed by a lot of generativists studying minority and endangered languages is that the corner cases should be studied last. In school they spend quite a bit of time going over the weird subtleties in English syntax (I did too). But when it comes to a brand new language with no prior syntactic analysis, they often head straight for the same sorts of things: do-support or double wh-movement or whatnot. (If they knew that double wh-questions aren’t permitted – and why? – then they’d know that there’s no movement possible.) Since they generally understand so little of the rest of the language’s syntactic system, it’s very easy to come to ridiculous conclusions due to the lack of perspective. In addition, sometimes the most fascinating things about the language may be in the obvious things (basic phrase order, agreement, and simple questions come to mind). That’s why it’s crucial for a good syntactician working on minority and endangered languages to have some comparative and typological background, as well as a solid understanding of the language’s basic syntax: to be able to see the forest for the trees. Unfortunately, too much work on the syntax of lesser known languages is not done by people who can see the forest for the trees, but instead is done by linguists who couldn’t construct a simple transitive sentence in the language themselves.

  91. Jonathan Mayhew says:

    I’m still thinking about the “most” discussion. I am in the supermajority-moster category myself, but found myself questioning whether a native speaker of English, such as myself, could be wrong about a very simple word with a seemingly clear meaning. That seems implausible. Yet it is equally implausible that the other camp (more =, quite simply, > 50%) could be wrong. Clearly they are not mistaken either. They might even be *most* people, with my own usage the exception.

  92. “Examples were no longer taken from actual utterances or texts displaying a range of syntactic structures in a different language,”
    The data-free approach to science. That has always really expanded the frontiers.
    “I have seen job ads (the few that are still open) in linguistics departments saying things like “the department subscribes to theory X and expects new members to pursue research in that framework”.
    Thay sounds more like a seminary than a university.
    “It’s not so much that monolingualism is encouraged, as that true bilingualism is required. ”
    I can see that if you are relying on the “consultant” to do all the work. But wouldn’t the linguist eventually start to develop some competence, maybe even intuition, of her/his own in the language under study? J. P. Harrington used to argue grammar with a womnan in Santa Barbara, a consultant on one of the Chumash languages, and she conceded now and then that he was right and she was wrong.

  93. Trond Engen says:

    most and moster
    Isn’t this a question of pragmatics rather than lexical meaning? If so, the discussion of introspective interpretations of example sentences will be a dead end, since different readers are envisioning different types of situations, interactions, moods, tones of voice … you name it.
    It could be researched through study of spoken corpus. If there are real differences out there, one would expect frequent misunderstandings, and the type and context of those could tell us a lot.
    (Disclaimer: I haven’t read the Log discussion ro the end.)

  94. Bathrobe says:

    Perhaps a more objective term would be “the dictionary sense.”
    I used Hattian in order to distinguish it from my own. Apologies for suggesting it is ‘aberrant’.
    I was more interested in hearing your interpretation of expressions like ‘Most people think that I’m crazy’. Or ‘most of the eggs were broken’ (still almost half quite usable). Or ‘most of my spare time I spend watching TV’ (but actually almost half of my spare time is spent going to discos and parties). I’m just curious whether a person who goes to discos three nights a week could really say ‘most of the time I just stay at home’. I suspect there are nuances here that form a bridge between the ‘dictionary meaning’ and the Bathrobian meaning of ‘a large majority’ (and yes, that’s what I meant to type).

  95. I’m not sure what I think with the most and more business, but I have only recently recovered from discovering that different people have different expectations for what “a couple” and “a few” means.
    I tend to be strict on a couple, thinking it means 2 on nearly all (most?) occasions (“in a couple of days”, for instance, I take to mean in 2 days, possibly 3 if things don’t pan out according to plan).
    A few I take to mean 3 to 5 or 6 at a pinch, never 2.
    This, however, is not what everyone thinks.

  96. Jonathan Mayhew says:

    Yes. I was surprised a few years ago to find someone who literally-minded insisted “a couple of” meant exactly two, when for me it can mean any number of numbers. (2-7 depending on context).

  97. I wouldn’t go so far as seven, but “a couple” can definitely include three for me, and at a pinch four.

  98. Jonathan Mayhew says:

    For me, a couple of things at the supermarket could even be 8 or 9. 10, though, is definitely never “a couple.”

  99. I remember coming back from Japan after high school in 1967, being largely ignorant of American car culture, being put to work in my uncle’s filling station in rural Virginia, and being asked by one driver to put “a couple of dollars worth” in his gas tank. Well, it turned out he meant precisely $2 worth (about 6 gals. at the time!), but that wasn’t how I would have interpreted it in most other contexts.
    I also had a hell of time finding the gas tank openings in “a couple or three” of those cars, to use an expression I learned at the time. Most gas caps were behind the license plate, but I remember I had to get help to find one behind the rear light. (Was it a 56 Chevy?)

  100. gas caps … behind the rear light. (Was it a 56 Chevy?)
    Cadillacs with fins, 48-56.

  101. Here’s a picture of it flipped up. The caption says they were there from the ’41, though a ’48 is shown. I’m sure they know more than I do, but I associate them with Franklin Q. Hershey (later of the T-Bird) and fins.

  102. Thanks, MMcM, but it might have been a 56 Chevy Belair (scroll down or search for “gas cap”), like the one my dad drove from VA to CA with five of us kids in back (me the eldest), then drove around Kyoto for the rest of the 50s. He was the first foreign victim of those newfangled police radar guns, and had to take remedial driver’s ed for speeding down Shirakawa-dori.

  103. I favour a narrow interpretation of “couple”, but the local English is strongly influenced by the much wider meaning of Dutch “paar” (“pair”) which covers a semantic space up to where my native English would have “handful”.

  104. I asked an Australian friend, a radio engineer so tending towards precision, how he understands “most.” He said:
    I agree with 75%. Three out of four does it, but two out of three would be stretching it. Fowler and New Fowler are only concerned with most vs almost (US usage being Scottish in origin, and still alive in UK dialect at the time of writing), Concise Oxford and Funk & Wagnalls are no help, nor is Brewer, or Authors’ and Printers’ Dictionary ( F. Howard Collins). No mention in Gowers’ Plain Words, and no guidance for Australian usage in Alan Peterson’s Words, Words, Words, usually helpful for local peculiarities.
    My feeling is that most corresponds to the superlative of the comparative more, at least in general effect, and that implies at least four starters. You need three starters before you can use more. On a 51% election result you could say that more people voted for the winning side; but that you could not say most people did, until you got to at least 75%.
    And of course we are talking number, not quantity. Do you draw the line at a different level for quantity, or area? Most of the ice cream was melted, or most of the beach was empty?
    I think most of us would go for at least 75%.

  105. Another Australian view of “most”:
    3 out of 4 is logical. But do we live in a logical world? Words are losing their meaning more rapidly – blame the media!
    So I asked [my wife] what would be “most” if there were 100 people – she instantly said 75%
    So that must be right.

  106. @LH. Do you have recommendations of how to persuade chomskyites to end their fanaticism and lack of rigor? They seem to have endlessly bad arguments which are asserted to support their claims.

  107. I don’t think they can be persuaded, only avoided.

  108. David Marjanović says:

    (“overwhelmingly most”?)

    By far most.

  109. David Marjanović says:

    I have seen job ads (the few that are still open) in linguistics departments saying things like “the department subscribes to theory X and expects new members to pursue research in that framework

    “The closer you get to humans, the worse the science gets.”
    Hi, Grumbly. The competition has arrived. Grmpf.

Speak Your Mind

*