Contractual Grammar

« previous post | next post »

Today's Grand Avenue:

Like a recent LL commenter, many people think that societies should explicitly legislate grammar, and punish those who disobey:

Trying to refute a prescriptivist by showing that people have always done it the other way is as absurd as saying that there should be no law against murder because millions of people have commited [sic] it.

Rather than calling this "prescriptivist grammar", we should probably call it "statist grammar", or something of the sort. There's a presupposition that leaving people to work such things out for themselves is intolerable, either morally or practically, so that the principles must be codified and enforced by some authority.

The alternative — which most modern linguists prefer — is not to treat language as unsystematic or unconstrained, but rather to view it preferentially as a natural rather than an artificial phenomenon, what Hayek called a "grown" or "endogenous" or "spontaneous" order, rather than a "made" or "exogenous" or "artificial" order. (See "Oblivious to usage advice?", 8/12/2008, for some discussion and links.)

In this respect, descriptive linguists are libertarians, or perhaps classical liberals, while prescriptivists are authoritarian rationalists.

But there's a curious thing. At least in the U.S., the very people who are authoritarian rationalists in linguistic matters, and thus philosophically aligned (for example) with other authoritarian rationalist movements from the French Revolution to Mao's China, tend to think of themselves as political "conservatives". Thus the commenter quoted above complains on his own blog about "that kind of woolly, leftist thinking that particularly inflames me".

[hat tip: David Giacalone, who suggests that the cartoon should count as "cross-disciplinary lawnguistics"]



50 Comments

  1. Sili said,

    February 4, 2009 @ 1:46 pm

    I didn't think it was possible to Godwin (almost) a discussion about grammar.

    Let me just take this opportunity to thank you for opening my eyes to the follies of prescriptivism. When I first got a phone I spelt and punctuated as correctly as I could. Now I'm quite comfortable dropping silent letters without as much as an apostrophe to mark their absense.

    Viva la Revolución!

  2. greg said,

    February 4, 2009 @ 1:58 pm

    I'm trying to decide if the idea that a person who ascribes to a political philosophy must logically ascribe to the similarly aligned linguistic (or any other) philosophy is a product of "wooly, leftist thinking" or of "close-minded, rightwing thinking." Or if it's just a cleverly sarcastic jab. Or both?

  3. vanya said,

    February 4, 2009 @ 2:02 pm

    This is pretty wooly headed argument as well. The Conservative movement in the US has generally favored order, rules and preservation of social norms. It seems quite obvious to me why in every country I can think of conservatives have a vested interest in prescriptivism and left-wingers apparently have a vested interested in overthrowing those rules. The "authoritarian rationalists" you point to – the French revolutionaries, the Russian Bolsheviks and the Maoists – were all initially anti-prescriptivist, and only became prescriptivists after they had created imperial states they were trying to preserve. One could argue that the legitimacy of the nation-state, a concept most conservatives believe in fervently, is deeply entwined with a prescriptivist approach to grammar. If there is an "American people" (or a "French people" or "Han people" or whatever) than it stands to reason that this group of humans, that is somehow clearly distinguishable from those outside the group, must all speak (or at least read, if you are Han) the same mutually intelligible language. And history shows that language change is a slippery slope – without codified rules and a central authority enforcing those rules, spoken and written languages do tend to diverge faster, undermining the eternal unchangeable nature of "the nation." As a practical matter prescriptivism makes a great deal of sense if you are pursuing certain political ends. Which is why the condescending attitude towards prescriptivists on offer on this blog always strikes me as naive.

  4. Thomas Thurman said,

    February 4, 2009 @ 2:13 pm

    But there are libertarians on the left as well as the right. (The ones on the left often call themselves anarchists or something similar.) It seems to me that left-wing anarchism, with no centralised authority but with plenty of rules decided by consensus among society, is a fair analogy to the way linguists perceive language.

  5. Benjamin R. George said,

    February 4, 2009 @ 2:27 pm

    I think that, if the analogy is interpreted at the right level of abstraction, the point the quote makes is formally valid. The two questions of ‘is x wrong?’ (for any chosen sense of ‘wrong’) and ‘is x a common and longstanding practice?’ are logically independent. A lot of anti-prescriptivist arguments found in popular and introductory writing on linguistics offer the longstandingness of the practice as evidence against it's wrongness, and others are quite right to point out the invalidity of this argument, even though the premise of longstandingness and the conclusion of non-wrongness are both quite true in the cases that are usually discussed.

  6. Mark Liberman said,

    February 4, 2009 @ 2:30 pm

    For the irony-impaired: I do recognize that "conservatives" often combine a commitment to keep government out of the marketplace with a desire to insert it forcefully into (say) the bedroom. There are several dimensions of opinion here, and they're at least empirically somewhat independent. But I feel that everyone involved ought to spend a few minutes thinking though what they believe about the nature and source of the rules or norms of behavior in various areas, and the mechanisms for dealing with violations of those rules or norms.

    If you're like most people, you'll find that you give rather inconsistent-seeming answers to apparently similar questions of this kind about different areas of human activity. The questions may not be as similar as they seem, and your answers may turn out to be deeply consistent, when viewed from the right perspective. But then again, maybe not.

  7. Andy Hollandbeck said,

    February 4, 2009 @ 2:35 pm

    @Vanya: "As a practical matter prescriptivism makes a great deal of sense if you are pursuing certain political ends. Which is why the condescending attitude towards prescriptivists on offer on this blog always strikes me as naive."

    I agree with the first sentence here, but it isn't always political ends that prescriptivists aim toward. I posit that violent prescriptivists always have some non-linguistic reason for wanting to solidify and codify the English language. It isn't for the good of language, nor for the clarification of communication, nor "for the children!" that prescriptivists thump their style guides on intellectual streetcorners. There's always some other reason.

    Which is why it's so easy to have such a "condescending attitude toward prescriptivists." Linguists (if I may be so bold to generalize) understand that language is just a tool for communication — a tool that grows and changes on its own. Prescriptivists use language as a tool for something entirely different, that has little to do with communication.

    Giving prescriptivists the total authority to decide proper grammar is a bit like giving creationists sole authority to interpret archaeological discoveries. Neither will wield that authority based solely on the evidence.

  8. kip said,

    February 4, 2009 @ 2:48 pm

    "In this respect … prescriptivists are authoritarian rationalists.

    But there's a curious thing. At least in the U.S., the very people who are authoritarian rationalists in linguistic matters … tend to think of themselves as political "conservatives"."

    So you're saying language/grammar prescriptivists are more often political conservatives in the United States? Or did I misunderstand you? In my experience, conservatives are generally less prescriptive. I mean, they supported Sarah Palin, and they aren't the ones responsible for the "Bushism" industry (except for the fact that they voted him into office, of course).

    [(myl) There are certainly plenty of prescriptivists who vote Democratic. And the New Yorker is a special case, maintaining and even extending Strunkish prescriptivism along with a liberal stance in politics.

    But as a rule, the people who make fun of W for how he talks don't get after him for using which in restrictive relative clauses, or for splitting infinitives, or for stranding prepositions, or for other violations of unnatural made-up rules. Instead, they make fun of him for regionalisms, for speech errors, and for alleged conceptual confusions. Take a look at Slate's collected Bushisms — I don't think you'll find any of the standard prescriptivist bugbears (though I haven't checked, and I might be wrong about this).

    But when (for example) Ann Coulter decides to go after someone for alleged "grammatical errors", she's all over which v. that. It's an interesting difference, don't you think? ]

  9. Mark Liberman said,

    February 4, 2009 @ 2:49 pm

    @vanya: When we adopt a "condescending attitude" towards "prescriptivists", it's because they say things that are historically false, linguistically incoherent, or pragmatically empty.

    Can you cite a contrary example?

    In fact, I've objected that

    genuine scholars of English usage find themselves forced to spend as much time marshaling evidence against the cranks who promote non-existent "rules" as they do correcting the barbarians whose prose is genuinely non-standard, confusing, or mistaken. As a result, the word "prescriptivist" is generally taken to refer to the crazies rather than to the scholars, and this seems unfair to me. The scholars also prescribe, after all, it's just that their recommendations are based on a rational analysis of the facts. It's as if we called witch-doctors "prescriptivists" because they prescribe on the basis of magical thinking about imaginary spirits, while calling practitioners of evidence-based medicine "descriptivists" because their recommendations are based on the factual relationship between remedies and their consequences.

    In addition to the many respected scholars who try to give useful and well-founded usage advice, there's a respectable discipline of "language planning" whose fruits can be seen in (for example) contemporary Indonesian or modern Hebrew or (at least to some extent) Putonghua.

  10. S Hawkins said,

    February 4, 2009 @ 3:13 pm

    At the risk of invoking anecdotal evidence, I have several colleagues who would be described by others and themselves as politically to the left, but who also hold extremely prescriptivist views. Indeed, I have found this to be a fairly common pattern in academia. Granted that this is a subset of the larger population, but it is not trivial, as it the teachers and professors who pass on the prescriptivist attitudes.

    On a separate note, language is about far more than communicating, or rather, more than communicating the content of the words (the referential meaning). The methods by which language is used play a key role in defining social identity, context of interactions, the nature of relationships, and so on. This may account for some of the emotional nature of the prescriptivist stance. Challenging their vision of the standard challenges a component of their identity.

  11. Blake Stacey said,

    February 4, 2009 @ 3:29 pm

    Mark Liberman wrote:

    If you're like most people, you'll find that you give rather inconsistent-seeming answers to apparently similar questions of this kind about different areas of human activity.

    Scott Aaronson suggested automating this process via the Web; some interesting discussion followed.

  12. A Reader said,

    February 4, 2009 @ 3:29 pm

    "Linguists (if I may be so bold to generalize) understand that language is just a tool for communication — a tool that grows and changes on its own."

    Linguistic Anthropology is a major exception to this- a major focus in current research is on language ideologies, and close investigation of language use in almost any context will turn up issues of power and inclusion/exclusion). Prescriptivism (in the usual sense) is essentially a demonstration of this in action, and it is (I think) this ideological element that many linguists object to. However, this also makes the focus on 'prescriptivism' a bit misleading- as Mr. Lieberman has just pointed out, prescribing language usage need not be an ideological exercise*, and prescriptivism hardly begins to encompass the range of language ideologies.

    My main recommendation to the Log in this respect would be to move beyond simply saying that the 'rules' so prescribed do not match the empirical rules of grammar, and get at why the prescriptivisms are supposedly better, the prescriptivists' motivation, and why language has remained a place where wholly arbitrary ideas of correctness are still widely accepted as an excuse to tell other people how they should act. Answering these questions helps build the case for linguistic tolerance in general (tying in closely with the English Only movement, among other things).

    *Except, perhaps, to an extreme post-modernist.

  13. A Reader said,

    February 4, 2009 @ 3:31 pm

    "Challenging their vision of the standard challenges a component of their identity."

    Even sooner done than said.

  14. Boris Blagojević said,

    February 4, 2009 @ 3:56 pm

    I think that the quoted comment illustrates the biggest problem of prescriptivism – mixing grammar with ethics.

    The thing is… due to a large number of differences between splitting an infinitive and committing a murder, you can't really compare them.

  15. kip said,

    February 4, 2009 @ 4:13 pm

    "due to a large number of differences between splitting an infinitive and committing a murder, you can't really compare them."

    Try telling that to the infinitive's children! :)

  16. Jonathan Badger said,

    February 4, 2009 @ 4:14 pm

    I guess I don't understand why there is so much acrimony between prescriptivism and descriptivism in linguistics. Yes, as in any science, the first thing to do in linguistics is to understand the system under consideration as it currently exists (descriptive analysis). But in other sciences, such as biology, it is taken as a given that the point behind learning the current state of a system is to use this knowledge to improve matters (prescriptive applications); molecular biology would be just that much trivia if it had no applications to medicine, agriculture, the environment, and so forth. I'm a basic biologist not an applied one, but in my grant applications I still have to justify why my research will lead to applications.

    [(myl) There is no acrimony between descriptive linguists and people interested in clinical applications (speech pathology, aphasiology and so on). Nor is there any acrimony between descriptive linguists and people interested in improving language teaching, or speech technology, or web search. Nor is there even any acrimony between descriptive linguists and those (some of them linguists) interested in giving usage advice (our own Geoff Nunberg was for many years the head of the AHD Usage Panel, for example), or doing evidence-based research on rhetoric or composition.

    In order to translate "prescriptivist" into the biological context, you have to think about categories like homeopathy, or diagnosis by aura color, or therapeutic use of magnets and crystals. In fact, "prescriptivism" is worse, since the adherents of biomedical woo are usually committed to the view that their cures actually do work, and can be shown to do so if you'll only look at the evidence in the right way. The weird thing about the "prescriptivist" strictures that we sometimes complain about is that their adherents typically insist that these are simply the rules, dammit, and never mind that there's no logical argument in their favor, and that the best writers in the standard language, over the years, haven't obeyed them. (For a half-serious analysis of this form of "prescriptivism" as a sort of cultural obsessive-compulsive disorder, see here.)

    So I continue to feel that "prescriptivist" and "descriptivist" are misleading terms. ]

  17. bulbul said,

    February 4, 2009 @ 4:29 pm

    Jonathan,
    if as Boris pointed out linguistics cannot be compared with law, how do you think biology will do in that scenario? For one, as Mark keeps pointed out over and over and over again, linguistics prescriptivism is not based on any sound understanding of the system. How do you think would a grant committee look upon my application if the ideas behind it were based on theories of Mssrs. Lysenko and Michurin?
    Also, there is such thing as applied linguistics. Could you be perhaps confusing the two? Applied linguistics is just what the name says. Prescriptivism is an ideology.

    As for political affiliation and prescriptivism, I can add a few more anecdotes. In the last few weeks, I noticed the aversion of some of my comrades on the left to the phrase "could care less" (recent example). And boy they sure can prescribe with the best of them.

  18. Bob Ladd said,

    February 4, 2009 @ 4:55 pm

    Anybody interested enough in these issues to have read down this far is urged to have a look at Deborah Cameron's book Verbal Hygiene (1995 or thereabouts). She's just about guaranteed to change your mind about something in this general debate (the title alone is enough to get you thinking), and her mix of political and linguistic views sure defies classification in the terms under discussion in this thread.

  19. J. W. Brewer said,

    February 4, 2009 @ 5:05 pm

    I have long lamented the inability of many of my fellow political conservatives to assimilate the Hayekian understanding of the grammar of a natural language, but I'm dubious about the claim that Democrats/liberals are not equally part of the problem in supporting weird bogus prescriptivism in American language culture. The institutional settings in which bogus prescriptivist rules are typically expounded and enforced (public high schools, newspapers, publishing houses, the legal system) are generally dominated by liberals, whereas more conservative-trending institutional settings like military and business culture tend to generate the sort of neologisms and innovative syntax that arouse prescriptivist ire. ("Impact" as a verb, "impactful" as an adjective . . .) Perhaps liberals are more likely to recognize a "diversity exception" for speakers of AAVE, Spanglish, etc. exempting them from being hassled about split infinitives, but that's a considerably different issue, and it doesn't seem to apply to e.g. anyone with the sort of regional accent who says "nukular" or "might could."

  20. Jonathan Badger said,

    February 4, 2009 @ 5:10 pm

    Given this, I continue to feel that "prescriptivist" and "descriptivist" are misleading terms

    They certainly are misleading if "prescriptivists" means to you *only* the people that you consider to be cranks such as biologists regard homeopaths (although I suspect even the type of cranky prescriptivists you are referring to, such as William Safire etc., know rather more about language than homeopaths know biology.)

    [(myl) This is a bit unfair to William Safire, whom I've often defended. Though whether he knows more about language than homeopaths know about biology might be subject to discussion — I have the impression that some homeopaths, whatever the quality of their arguments, are actually quite extensively informed about many biological issues. ]

    However, if you agree that it is *possible* to argue for language standards or change on the basis of knowledge rather than ignorance, then I agree there isn't any acrinomy. It's just that the standard argument against "prescriptivism" often comes off as supporting some sort of Star-Trekkian "Prime Directive" where the very idea of interfering with the natural evolution of language use is taboo.

    [(myl) There are some people who feel that way. But I don't know any case where a coherent and well-founded intervention has been systematically opposed on such ideological grounds. The commonest arguments are of two types. (To avoid connotational confusion, let's call the two sides of such arguments "P" and "D".)

    (1) Someone on the P side invents a "rule", which is not dictated by any well-supported theory, nor shown to be clearer or better by any experiment, nor found to be characteristic of the practice of any well-respected community of speakers or writers. People on the D side protest that this "rule" is made up out of nowhere, has no logical or empirical support, etc. Iterate to taste. (You may think that this is an exaggeration, but it isn't. See e.g. "The split verbs mystery".)

    (2) In a situation where the standard variants of English do X, a regional or social variant does Y. The P side characterizes this as the non-standard varieties "breaking the rule", due to being "sloppy", "lazy", "careless", "undisciplined" or "ignorant". Even more highly evaluative terms may be used, like "disgusting"; and predictions of cultural dissolution are often added. The D side points out that Y is just as rule-governed as X, it's just that the rule is a different one. (And sometimes Y is the older usage as well.)

    In cases of type (2), there's a legitimate argument about whether the Y-speakers ought to be urged to learn to pretend to be standard X-speakers when the occasion warrants, or whether the X-speakers ought to be urged to tolerate the Y-speakers' different linguistic patterns. But this argument is badly muddled by the P-rhetoric identifying Y-speakers as lazy, ignorant law-breakers.

    ]

  21. Nathan Myers said,

    February 4, 2009 @ 5:46 pm

    Here is an amusing thread responding to apoplexy over changing definitions of the word "dinosaur". To paleontologists, "dinosaur" means exactly "an ornithischian, saurischian, or allied genus". To the rest of us, it could also be any of the lumbering beasts that died out before 65 million years ago, including pterosaurs, ichthyosaurs, plesiosaurs, mosasaurs, or even early Triassic crurotarsans, but not including modern crocs, turtles, and (significantly) birds; Merriam-Webster agrees. It's amusing to find paleontologists insisting that the meaning assigned to the word "dinosaur" is a "simple fact" akin to the simple fact of their (avian-excepted) extinction.

    http://dinobase.gly.bris.ac.uk/forum/viewtopic.php?id=575

    By contrast, botanists seem to have no difficulty with people resisting cucumbers as "berries". Entomologists seem reconciled to all sorts of arthropods being called "bugs". King Kong is a well-known fictional monkey.

  22. Adam said,

    February 4, 2009 @ 5:48 pm

    Loldemort (the commenter quoted in the post) doesn't seem to be advocating explicit legislation of grammar, much less punishment of deviance. Rather, the point is that some normative claim (that murder is wrong, that splitting infinitives is wrong) is not undercut merely by pointing out that the norm is in fact sometimes violated.

    So what's needed is discussion of the epistemology of norms, and some sensible claims that underwrite our normative judgments. This is another way in which the labels 'prescriptivism' and 'descriptivism' are bad—it's not that so-called descriptivists think that there aren't any norms of language, but rather that those norms have to be discovered by empirical investigation into actual language use. And the standard LL complaint about prescriptivists is (again) not that they advocate usage norms, but that those norms are not justified by appropriate methodological principles.

    Sorry if this seems to merely restate Mark Liberman's point @2:49pm. But I think the original commenter was misinterpreted, and that his/her point as it stands is a correct one. In order to get real traction against the "prescriptivists," we've got to take on a different question: what are the right ways to extract norms from the empirical investigation of language?

  23. Geoff Nathan said,

    February 4, 2009 @ 6:05 pm

    And, as a very long-term libertarian I note with exasperation that many well-known libertarians (Hayek being an exception) are rabid prescriptivists on language. Stephen Cox, an English professor and current editor of one of the two major libertarian monthlies (Liberty) has a monthly column called 'Wordwatch' which often grumbles about various shibboleths. On the other hand, at least some of the folks who write for the other mag, Reason, read this blog, and everybody seems to meet at the Volokh Conspiracy.

  24. Nathan Myers said,

    February 4, 2009 @ 6:29 pm

    For those of us who perceive "libertarian" as a specialization of "fetishist", language prescriptivism in libertarians is entirely unsurprising. Language-prescriptivist behavior among fetishists, in general, seems a legitimate topic for academic study.

    I wonder if there's a standard diagnostic for fetishism as there is, for example, for narcissism. All we would need then is a standard diagnostic for language prescriptivism. If the correlation were high enough one diagnostic might stand in for the other.

  25. Mark F. said,

    February 4, 2009 @ 7:52 pm

    I don't think most prescriptivists are really statist about language. I think they're Platonists. They think that using 'which' in a restrictive clause is wrong in an objective, ideal sense, independent of whether people do it or not. The OP, I think, was not implicitly advocating an Academy of English. They were saying that murder is just wrong, no matter how widely it's practiced, and that grammatical 'errors' are also errors, no matter how widely they have been practiced. They weren't saying they are both morally wrong, or that they both should be made illegal; only that they are both objectively wrong. (I don't think they even have that in common myself, but that's another story.)

    I know people do routinely speak of grammatical peeves as if they are morally outraged, but they really aren't. If they were, then we would actually see people trying to build a case for legislative bans on split infinitives, and we don't. So, you may think I am being too generous in my interpretation of the OP's comments, but people say things like that all the time, and if they really meant it in the strong sense, at least some of them would act on it.

  26. joseph palmer said,

    February 4, 2009 @ 8:14 pm

    Adam is quite right, and I tried to say the same thing much less eloquently following the post singled out by the blog.

    Once "descriptive" linguists have formulated some "rules" then those rules become norms used by teachers, and linguists, to judge what standard English is. So called "prescriptivists" arrive at these rules very clumsily. However stating rules for standard English is not in any instance easy. For example, where do the boundaries lie between standard and non-standard?

    One thing is for sure – nearly all linguists are involved, or have been involved, in a process of harshly marking and damaging the careers of those who do not use standard English as prescribed/described by themselves.

  27. Mark Liberman said,

    February 4, 2009 @ 8:25 pm

    Mark F.: I don't think most prescriptivists are really statist about language. I think they're Platonists. They think that using 'which' in a restrictive clause is wrong in an objective, ideal sense, independent of whether people do it or not.

    I don't have any evidence about relative numbers. But there is no question that many important prescriptivist writers do believe that the rules they care about are contingent human creations, not aspects of the eternal structure of reality. For example, Mark Halpern wrote a reply to Geoff Nunberg, in the Atlantic in 1997, which included this imaginary rejoinder from P to D:

    "Yes, we know this; we do not contend that the rules we propose for the sake of clarity and richness of communication were handed down from on high. They are ordinary man-made rules, not divine commandments or scientific laws (although many have support from historical scholarship), and we agree that they, like all man-made things, will need continual review and revision. But these facts are no more arguments against laws governing language usage than they are against laws governing vehicular traffic. Arbitrary laws — conventions — are just the ones that need enforcement, not the natural laws. The law of gravity can take care of itself; the law that you go on green and stop on red needs all the help it can get."

    This not only specifies that the principles in question are social conventions, not platonic ideals, but also calls them "laws" and makes an explicit analogy to laws that are written down and enforced by state power. (Though the analogy of traffic lights is obviously better than the analogy of homicide.)

    Many people seem to find Halpern's reasoning to be convincing. For example, someone who recently commented on a discussion from last summer added this post scriptum: "Read the writings of Mark Halpern, if you dare. He guts descriptive linguists the way a fisherman guts a perch! "

    (I should note that I not only dared to read Mark Halpern's essay, but also assigned it to students in the LING 001 course that I teach, starting in 1997, the very year it was published. The lecture notes for the session of {pre/de}scriptivism for the 1997 edition are preserved here, though of course many of the links are dead.)

  28. James said,

    February 4, 2009 @ 8:36 pm

    Nobody could really be a Platonist about the rules of English the way some (Plato, say) are Platonists about ethics, since everybody, even the prescriptivist you most want to make fun of, knows that people who speak Korean aren't thereby speaking incorrectly, whereas Platonists do think that people who follow a completely different moral code are behaving wrongly.

    But I'm sure Mark F. is onto something nevertheless. It's as if there were a bunch of languages in Platonic heaven, and any of them is fine to speak or write, but each has its own eternal rules. When we speakers get in the neighborhood of one of them, we are subject to the normative gravitational of the nearest, and guilty of misspeaking when we depart… and that's true no matter how many of us do it.

    Think about this. We all have some kind of prescriptivist experience, because we all know what it's like to realize, "Oops, that was wrong, I just said something ungrammatical." We break our own idiolectic rules, and then we notice, and we can experience it as a mistake, wrong, incorrect, bad. Doesn't this feel a bit like Platonism?

  29. The other Mark P said,

    February 4, 2009 @ 10:33 pm

    King Kong is a well-known fictional monkey.

    He is a fictional APE, not a monkey.

    This is precisely the sort of confusion that drives experts crazy.

  30. Nathan Myers said,

    February 4, 2009 @ 11:56 pm

    Google finds 1,440,000 hits for '"+"king kong" +monkey' and 665,000 for '+"king kong" +ape'.

    As noted in the thread linked above,

    The hotel guest demanding that you get the monkey out of his room is fully justified, when you explain that the creature in his room is no monkey, but a fine specimen of Pan troglodytes, in flushing your toupee — even if that guest is in fact a union steward, and fully cognizant of finer distinctions than you could ever make.

  31. Mark Liberman said,

    February 5, 2009 @ 5:03 am

    Adam: Loldemort (the commenter quoted in the post) doesn't seem to be advocating explicit legislation of grammar, much less punishment of deviance.

    Perhaps Loldemort would agree with you, if asked. But why? If we're talking about things that are offensive to morals and public order, why wouldn't we want at least to consider legislation and penalties?

    In his 2008 book Language and Human Nature, Mark Halpern takes care to say that

    The last thing I would propose is the establishment of any kind of central authority over language, and for the same reason that I reject the idea of a central authority over the economy.

    But he also asserts, at length, that "faults of language usage" are "more than just irritants to pedants and esthetes, but offenses that caused or contributed to serious social and political problems". Most people these days, even conservatives, believe that some sort of governmental regulation of economic affairs is necessary; and Halpern repeats in this book his arguments about traffic laws, without any indication that he favors replacing such laws by a system of do-it-yourself vigilante justice.

    If you believe that violations of linguistic "rules" (even artificial ones invented by self-appointed athorities) are actions with "serious social and political consequences", it seems inconsistent (though perhaps realistic) to avoid calling for a governmental role in avoiding or ameliorating these consequences.

  32. Chris said,

    February 5, 2009 @ 9:13 am

    If you believe that violations of linguistic "rules" (even artificial ones invented by self-appointed athorities) are actions with "serious social and political consequences", it seems inconsistent (though perhaps realistic) to avoid calling for a governmental role in avoiding or ameliorating these consequences.

    At the very least, people with an influence on public behavior, such as media figures, ought to be fined for linguistic slips the same way they are fined for wardrobe slips, and if inadvertence is no defense in one case, why should it be in the other?

    (I think this is rather silly, but then, I don't think that linguistic missteps do have serious social and political consequences. For that matter, I don't think nudity does either.)

    Your point about conservatives seems oddly out of touch with the fact that U.S. conservatives *are* authoritarians – and many libertarians are regretting their devil's bargain with conservatives to achieve economic deregulation at the price of torture, warrantless wiretapping, etc. (especially when the unregulated economy led to disaster anyway). I don't want to devote too much space to a political point, but in this case it's also linguistic: "conservative", in contemporary US usage, no longer means what it meant a generation ago.

  33. language hat said,

    February 5, 2009 @ 9:18 am

    And history shows that language change is a slippery slope – without codified rules and a central authority enforcing those rules, spoken and written languages do tend to diverge faster, undermining the eternal unchangeable nature of "the nation."

    Nonsense. Languages change at the same stately pace regardless of the flailings of would-be linguistic legislators.

    As a practical matter prescriptivism makes a great deal of sense if you are pursuing certain political ends.

    Sure! Not scientific sense, of course, but forcing people to do things One Prescribed Way is always useful to authoritarians.

  34. vanya said,

    February 5, 2009 @ 11:29 am

    Nonsense. Languages change at the same stately pace regardless of the flailings of would-be linguistic legislators.

    I'm not talking about linguistic legislators of the Safire variety. I think there is ample evidence that a centralized state promulgating an official language can slow language change – spoken Latin changed more rapidly from 500 AD to 900 AD that it had in the 700 years previous. Sure, this can result in diglossia as the actual spoken language changes more rapidly than the fossilized "official" language but the authoritarian rationalists can deal with that.

  35. Mark F. said,

    February 5, 2009 @ 12:52 pm

    Adam: Loldemort (the commenter quoted in the post) doesn't seem to be advocating explicit legislation of grammar, much less punishment of deviance.

    Mark Liberman: Perhaps Loldemort would agree with you, if asked. But why? If we're talking about things that are offensive to morals and public order, why wouldn't we want at least to consider legislation and penalties?

    OK, that kind of gets at one of my points. People don't at least want to consider legislation and penalties. I guess I can see that as a reductio of some of their claims, whereas before I was introducing it as evidence they were Platonists.

    It's been said of mathematicians that they're formalists on the Sabbath and Platonists the rest of the week (or some such). I think a lot of prescriptivists may be somewhat that way, if you replace "formalist" with "social constructivist". I don't see how else people can argue, as they often do, that a construction in universal use for centuries is an error.

  36. Mark Liberman said,

    February 5, 2009 @ 1:06 pm

    Mark F: I don't see how else [if they're not Platonists on some days of the week] people can argue, as they often do, that a construction in universal use for centuries is an error.

    One coherent and non-Platonistic argument of this general type is based on internal consistency of the system, or perhaps the ranking of one sort of consideration above another when both are valid but they disagree. I've sometimes made arguments of this kind about under- and over-negation, e.g. here; and there's an explicit prescriptivist argument from Robert Beard of this form here; and what may be an argument of that form from James Kilpatrick here.

    This sort of argument might be right or wrong in particular instances — the presumption is against it, I think, in cases where speakers and writers don't see what they produced as a mistake — but it's not false in principle, and I'm pretty sure that it's sometimes right.

    Another argument, which I don't recall having seen in recent prescriptivist work, was advanced by John Dryden with respect to "dangling prepositions": he argued that the rule, though an innovation, was a sign of progress, making the language better (for reasons that he didn't specify, but still…). This was part of an argument that he (as a representative of his age) was a better poet than Shakespeare and other Elizabethans, because

    … these absurdities, which those poets committed, may more properly be called the age's fault than theirs. For, besides the want of education and learning, (which was their particular unhappiness,) they wanted the benefit of converse […] Their audiences knew no better; and therefore were satisfied with what they brought. Those who call theirs the Golden Age of Poetry, have only this reason for it, that they were then content with acorns, before they knew the use of bread …

    Some of the which-hunting arguments are of this form, as I recall.

  37. James Wimberley said,

    February 5, 2009 @ 8:09 pm

    There's a sensible Burkean conservative position that hasn't SFIK been argued here. (I'm in two minds myself). Viz: for any language that has a significant body of literature and other culturally interesting texts, it is desirable to preserve their intelligibility for as long as possible – centuries not decades – otherwise the language community loses touch with its past. Example: Chaucer is nearly unintelligible to modern English readers without a lot of apparatus, which is a shame, as he wrote good stuff. If we'd kept say swive we'd be the richer for it.

    The inference is that so far as these things can be influenced by education and punditry, we should try to slow the pace of change in grammar and vocabulary. In particular, we should resist changes that convey no appreciable advantage: the linguistic equivalent of genetic drift. Whether innovations are worthwhile is a mater of opinion; but the linguistic conservative plays a useful cultural role in asking the question.

    The counterarguments would lie over the necessity and possibility of any such programme. Latin stayed pretty fixed for a thousand years from Livy to Bacon as the written language of the Western European élite, without any self-conscious policing before the Renaissance revival of Ciceronian Latin. Milton's English is entirely comprehensible today. This suggests that the rate of drift in the common core of English is too slow to worry about.

    The other aim of Burkean conservatism would be to slow the splitting of a language, as happened with Vulgar Latin, and the concomitant division of a single language community into several, with impaired mutual communication (as between the Romance languages.) The merits here are unclear. The diversity has its own value. I suspect professional linguists prize it over communication. (Thought experiment: if the price of world peace was a single language, would you pay it?) It's also very doubtful than anything can be done about language speciation, and the current problem is extinction instead.

  38. joseph palmer said,

    February 5, 2009 @ 9:53 pm

    I would say that was a vaild reason for at least sometimes resisting some kinds of language in some contexts, James. Codification of a language initially requires prescriptivism of some sort, and surely without people feeling they must conform to certain codes you get diversity and reduced communicative ability between groups. However the most importanat reason that many people try to stick to the norms is not because of a pragmatic conservatism. It is because they wish to demonstrate how clever they are, having mastered such norms, and their show their allegiance to the community that values them. Nobody sticks to them more rigidly than the professional language person, for of course demonstration of that mastery, for such a person, is vital. Their peers, despite any proclaimed liberalism, will in fact judge them very harshly if they seem to speak non-standard English.

  39. Ken Brown said,

    February 6, 2009 @ 8:18 am

    Nathan Myers said "To paleontologists, "dinosaur" means exactly "an ornithischian, saurischian, or allied genus".

    "Allied genus"? That's fighting talk! As well as being pre-cladistic, if not pre-Darwinian, in its vagueness and essentialism!

    To palaeontologists "dinosaurs" are all the descendents of whatever the original dinosaurs were, and nothing else.

    When I did my palaeontology course not so long ago, the party line was that mammals did not replace dinosaurs – (actually we are an older group than them so they partially replaced us) dinosaurs are not extinct, birds simply *are* dinosaurs. And baby dinosaurs were called "chicks". :)

    "By contrast, botanists seem to have no difficulty with people resisting cucumbers as 'berries'. Entomologists seem reconciled to all sorts of arthropods being called 'bugs'"

    I think we biologists have to common sense to realise that we borrowed the words "bug" and "berry" from ordinary language and we use them in specialist ways & so can't expect everyone to use them our way. So we talk about "true bugs" and so on.

    But we invented the word "dinosaur" and you nicked it!

    Also in a world where all sorts of people are trying to suppress the study of evolutionary biology and palaeontology, spreading knowledge can come to seem like a public duty.

    And its fun. For one thing, those fluffy tyrannosaur chicks might have been cute. For another, those great big pre-dinosaur thingies with the giant sails on their backs are much more closely related to us than they are to dinosaurs – they (or something like them) might even have been our own direct ancestors. Which is sort of interesting.

  40. Chris said,

    February 6, 2009 @ 9:48 am

    If we'd kept say swive we'd be the richer for it.

    On the other hand, if enough people had found swive useful to actually use it, we would, ipso facto, have kept it, which seems to argue against your conclusion. Words are created and abandoned based on the need for them – the reason we don't have a commonly used word for thill anymore is that we don't have much need to talk about thills. Eggcorns like "reign in" exist (and threaten to send "rein" the way of "swive") because we don't have much call to talk about reins in our motor-vehicle-driven culture. On the other hand we do need words for e-mail and spam, and so we have them.

    I suppose this could be seen as validating (partially) the "no word for X" pseudo-linguistic pseudo-anthropology that gets occasionally bashed in this space, but that's not really what I'm getting at. The regularly used vocabulary of a language reflects the subjects that are of interest to its speakers – whether they are part of daily life (car), rare but culturally significant (terrorist), or even entirely fictional but still culturally significant (angel). (And of course we *do* have words for swiving – swive just isn't one of them anymore – which makes it hard to see how we have lost anything.)

  41. Nathan Myers said,

    February 6, 2009 @ 4:47 pm

    Ken: Exactly my point… fluffy T. rex chicks snatched up by Maastrichtian ninja azhdarchids the size of a giraffe, that plunge from the sky and then explode back aloft before Mrs. Rex has time to put down her coffee cup, are intrinsically interesting. Quarreling over who gets to define the lexeme "dinosaur" isn't interesting.

  42. Nathan Myers said,

    February 6, 2009 @ 4:52 pm

    Ken: Oh, and I didn't nick "dinosaur". It was nicked long before you or I were born. The horse hasn't just left the barn, the barn rotted and fell over, and the horse's descendants stampede the plains, tossing their manes in the wind.

  43. James Wimberley said,

    February 6, 2009 @ 5:00 pm

    Chris: "Words are created and abandoned based on the need for them."
    Not entirely; what need other than the avant-garde search for novelty justified the recent shift from symbolic to emblematic in high-maintenance culturespeak?

    Another example (personal peeve):
    The hijacked bus careened through the rush-hour traffic pursued by a dozen police cars.
    Buccaneers used to careen ships by beaching them on their sides to scrape off the barnacles, Not exactly an everyday contingency, but the word was there if Disney ever needed it for Pirates of the Caribbean IV. I assume some newspaper replaced an r by an n in the rushing-about context – an easy thing to do, and it's stuck; but careen in this sense adds nothing to career. It's useless linguistic drift and (loses control, spittle flies in all directions) ILLITERATE VANDALS AND MISCREANTS OF THIS CONTEMPTIBLE SORT SHOULD BE MADE TO WALK THE PLANK.

  44. Mark F. said,

    February 6, 2009 @ 10:25 pm

    Mark Liberman: One coherent and non-Platonistic argument of this general type is based on internal consistency of the system, or perhaps the ranking of one sort of consideration above another when both are valid but they disagree.

    Good point.

    Another argument, which I don't recall having seen in recent prescriptivist work, was advanced by John Dryden with respect to "dangling prepositions": he argued that the rule, though an innovation, was a sign of progress, making the language better (for reasons that he didn't specify, but still…).

    Now this line of reasoning, on the other hand, is exactly what I was calling Platonic. Perhaps I'm using the word over-broadly, but it seems as if Dryden thought there was an ideal version of English that the dialect of his era approximated more closely than that of Shakespeare.

    And even when rational arguments are available, people often don't seem to be relying on them. Even though you can make consistency arguments against Singular They, I have the impression that most people who don't like it think is is Just Wrong. If they were treating it as a policy question they would make more of a show of rebutting arguments for why singular they might be worth using.

  45. Mark F. said,

    February 6, 2009 @ 10:47 pm

    A general comment I've been wanting to add to an LL thread on prescriptivism:

    Has their been any analysis of the relationship between obedience to prescriptive rules and formality of writing? Are infinitives split more or less often in scientific journal articles than Time Magazine? Do the sciences differ from the humanities in non-restrictive which usage? Are people more careful to follow purported rules like these in resume cover letters than in invited addresses?

    If you could show, for instance, that, even before the rule was propounded, people split infinitives less in more formal registers of English than less formal registers, then it would suggest that perhaps there was some organic basis for the rule beyond mere emulation of Latin. On the other hand, if people split them even more often in formal contexts, that would be very interesting. Perhaps these kinds of questions have already been answered.

  46. Loldemort said,

    February 7, 2009 @ 3:45 pm

    Well that's perhaps a reading of what I wrote — if the most violently strained one imaginable. The idea that I think that states (or anyone else for that matter) should legislate grammar and punish infractions is, I am afraid, quite nonsensical. I do hope that all your conclusions are not as far removed from the actual facts as that one.

    No, the point I was making was simply that there is a difference between does and should, that's all. They are logically separate things.

    The problem I think stems from an overly naive characterisation of linguistics as a scientific discipline. In this view, I suppose, speakers are like atoms and they emit speech rather in the way that atoms emit photons. Alongside them sits, eminent and grave, The Linguist, ready to catch their utterances in his detector. The Linguist has, quite properly, no preference as to the forms of the utterances he records; like a true scientist, he wishes only to observe, to analyse, and to construct a theoretical explanation for his observations.

    That's fair enough as far as it goes, and perhaps I shouldn't laugh at it (irresistible though it is) but it only goes half way. People — speakers — are not simple, natural objects, behaving according to immutable laws. The experiment has been done: children unfortunate enough to have been brought up by wild animals do not grow up speaking "excellent Hebrew"!

    So, alongside whatever general language mechanisms may exist in our brains on the one side, we have the essential arbitrariness of any particular language on the other. Part of learning any language therefore is the internalisation of a large set of rules. Now these rules are, again, for the most part perhaps, quite arbitrary, and they need not even be particularly consistent. They will differ slightly from person to person, and from decade to decade within the same person.

    Now here's where I find your — since were in the business, in this post, of calling each other rude names — "primitive descriptivism" particularly offensive. Since we're not simply dried beans sliding down a ramp, or metal balls being shot through the air, or atoms being excited until they give up a photon, it's quite allowable for us to develop opinions about what we do, and indeed about what others do. What's the point in being the crowning glory of the (known) universe if we can't do that?

    So I view prescriptivists as simply people who have an opinion about how we should speak (or write, or text, or whatever). Not, as I said at the very beginning, to be refuted by pointing out how we do speak. Whether they justify their suggestions by reference to alleged historical usage is, I think, beside the point. The suggestions can be evaluated, calmly, on their own merits. Nothing there to get particularly excited about. Nothing to merit the amazing amount of venom that they receive regularly in this blog.

    Finally, I think it odd that you should identify prescriptivists with the right wing. Of the three most famous totalitarian states of the last century, Nazi Germany and the communist USSR and PRC, two were left wing, and they both lasted the better part of a century whereas Nazi Germany only lasted the better part of a decade. It seems to me that the left can be just as authoritarian as the right if not more so, it just tends to be less honest about it, that's all. Let a thousand prescriptivists bloom!

  47. Nathan Myers said,

    February 7, 2009 @ 7:34 pm

    Loldemort: I have just been reflecting on the relationship between "should" and "does". In a careful study of their usage, I find that in fact they are different. "Should", in fact, appears to substitute almost perfectly for "doesn't", and "shouldn't" for "does", except where the participants have not yet had occasion to "do".

    Incidentally, your interpretation of the xkcd strip involving a Cherokee speaker made me laugh, albeit not especially loudly. It must give Randall hives, finding otherwise articulate people so unable to catch the gist of simple humor.

  48. Mark F. said,

    February 8, 2009 @ 1:02 am

    Loldemort: So I view prescriptivists as simply people who have an opinion about how we should speak (or write, or text, or whatever). Not, as I said at the very beginning, to be refuted by pointing out how we do speak. Whether they justify their suggestions by reference to alleged historical usage is, I think, beside the point. The suggestions can be evaluated, calmly, on their own merits. Nothing there to get particularly excited about. Nothing to merit the amazing amount of venom that they receive regularly in this blog.

    What about people who say that some usage is "incorrect"? Usually, when you use that kind of language, you are talking about concordance with some agreed-upon standard (something lacking for English). That isn't the language of preference or opinion, but it is the phrasing often used in usage advice.

    Secondly, why is it (automatically) beside the point to look at what people actually do when trying to judge what people should do? If you choose well the group of people to look at, it seems very much to the point. If the very best writers, people who write better than you do, use a construction, then shouldn't you at least give their judgment some weight?

  49. Loldemort said,

    February 8, 2009 @ 2:16 am

    @Mark F: Yes, I suppose if I read more prescriptivists, or took them more seriously, I might be more annoyed by them. Still, it seems to me that maintaining that people mustn't have an opinion as to what's correct and incorrect in speech is as prescriptivist as anything else. Languages change for all sorts of reasons, and I get the feeling that what might be called bottom-up change is regarded here as somehow more legitimate than top down. Both those phrases might be emotionally laden to some people of course, I'm simply trying to contrast mass/anonymous sources of change with individual/articulate sources.

    As a former programmer, I am a firm believer in the value of correctness in the use of language. Most uses of language have an aim beyond mere utterance. Their purpose is to convey meaning about objects and happenings in the world. Say the wrong thing to a computer and it will do the wrong thing. At the same time, I like my rules to be as simple and clear as possible, without unnecessary exceptions. Which brings me to my favourite dislike: double negatives. If I'm programming a robot waiter to offer people drinks, I want the same simple rule on negation, p and not-p, to work where p already contains a negative. It turns out that people who want to avoid double negatives (except, of course, where by not-not-p you really do mean p) actually want to remove a rule, not to add one.

    I do take your point about the usefulness of looking at existing and archived usage. I just don't regard it as a final authority.

    Finally I would ask how do people here feel about Peking/Beijing? I confess that I would expect most people here to feel an obligation to use "Beijing", perhaps glossing it as all about being civilised when someone asks you to respect their name for their capital city. I don't feel that way, of course. If languages are indeed to be arbitrary, I don't see why a change in the romanisation of a word in one language should have any effect on the word used to denote the same entity in a completely different language. We are after all talking about a change in a non-official mode of writing in the country concerned: the actual spoken sound hasn't changed in Chinese so why should it in English?

    Finally finally, it seems to me that the dangerous lure of prescriptivism can corrupt even the most descriptivist of angels. I refer, of course, to the "[sic]" in the quote from my comment at the start of this post. Now it just so happened that I didn't check my spelling carefully enough before submitting my comment. I don't mean to suggest in any way that Mr Liberman wanted to make some kind of cheap point about the irony of a prescriptivist not being able to spell, or that he wanted to characterise me as uneducated because I disagreed with him. Perish the thought! No, I would just like to point out that Google has "about 111,000,000" results for "commited" as opposed to a mere "about 108,000,000" for "committed".

  50. Mark F. said,

    February 10, 2009 @ 12:00 am

    Loldemort — several comments.

    1. Before, you were talking about it being reasonable to have opinions about language use, and now you are talking about correct language use. They are different concepts and I think people conflate them a lot.

    2. Sometimes the prescriptivists do win. Take double negatives. They really are incorrect in standard English, even by the standard of actual use that you don't like. And they are wrong because prescriptivist grammarians basically convinced educated writers not to use them, as I understand it.

    3. I think Geoff Pullum's periodic railing against linguification is an attempt to get a prescriptivist shibboleth started just to show how it happens. And don't get me started on "Line begins here to pick-up your forms". So, as you say, no one is innocent.

    4. I think people started using Beijing because China started using it in their English language documents. They were not merely changing the Romanization of the word in Mandarin; they were effectively saying "This is the name we would like you to use for this city." The German government, on the other hand, doesn't refer to itself as "Deutschland" in its English documents. I think it's reasonable to call the Chinese capital what they would like me to call it, just as I think it would be rude to insist on calling Pope Benedict "Pope Ratzinger."

RSS feed for comments on this post