[This was in the old Vocabula Review in March of 2016. I’ve tinkered only a little. – JK]
anguage Log, a fascinating and important website, the blogging home of some very famous linguists and scholars, has a standing feud with what it calls “peeves”: people’s highly particular annoyances with one usage or another, and (especially) the habit of justifying such aversions with hazy ad hoc rules that have not really been tested or thought through. Writers at Language Log are peeved by peeves. The theme runs through post after post, and the page explicitly entitled “peeving” (or assembled when you click that tag) includes discussions of singular data, of irregardless, of where we’re at, of hopefully, of nucular, of tremendous and other empty superlatives, and so on. The category stretches all the way from prescriptive rules that have been put forward formally at some point, to small gripes in eager letters to the editor. Nearly always, LL defends the questioned usage or, at the least, holds it to be a small problem only, not nearly worth the peever’s fustian.
Language Log’s idea about peeves, the descriptivist idea, is that they are accidental, pointless, artificial. The irritations that peeves report would not arise spontaneously, but are dreamed up and deliberately instilled at some point by bad teachers or snide hobbyists (whose categorical inferiority to real linguists is a favorite theme of many posters at LL). Afterwards they are retained as self-conscious shibboleths, not because they assist communication but only because people get a thrill of mean superiority from enforcing them (or, on the other side, feel intimidated). In the worst case, peeves lead to counterintuitive, impossible-to-follow pseudo-rules that overwhelm any natural feeling for language, producing the trademark unhappy stiffness of overly self-conscious usage.
This sense that peeving, and with it too much of traditional grammar, adds nothing productive to natural usage, that it is only what Arnold Zwicky calls “a pointless game of grammar Gotcha,” is a central passion at Language Log. The contributors feel that they have a mission: to tear down false dogma and false authority. And in this cause they are often quite impressive. The typical descriptivist take-down of a peever goes like this:
1) The peever is quoted, often at length, in an apparent show of fair-mindedness, but actually in sly confidence that his histrionic language will play quite badly.
2) The reviewer undertakes the light research — in the OED and MWDEU and so on — which the peever has not bothered to do or known how to do, an oversight which is held to be shocking and inexplicable.
3) Often quite usefully and instructively, for my money at least, the reviewer dispels what Mark Liberman calls “the recency illusion” and “the frequency illusion”: the peever’s usual conviction that his pet solecism was invented just recently but has been multiplying like the Ebola virus.
4) It turns out that the questioned usage has been around, off and on, since 1604, and was denounced in 1659 by a writer who considered it a sure sign of cultural collapse. Bonus points at this stage when the item can be found anywhere in the works of literary greats, as double negatives can be in Shakespeare and nonliteral “literally” in Twain.
5) Googling the phrase produces several million hits, quite close to the number for the supposedly preferred alternative.
6) The grammatical logic of the usage is explored with a rigor and subtlety that are, plainly, beyond the reach of the peever himself. It is gently pointed out that what he calls a gerund is really a participle, and that he has mistaken the conditional tense for the subjunctive. All too often, he has lumped together very different kinds of problems, griping about (let’s say) between you and I and comma faults and quotative like in the same article, simply because all three annoy him.
7) Usages analogous to the peeve are introduced, by way of showing that the offending whatnot is really consistent with the normal operations of the language. If you accept singular agenda for instance, you have no right to reject singular data; if really as an empty intensifier does not bother you, you must in simple fairness accept literally in the same role.
8) Sometimes, if the argument has been going especially well, the practitioner may dare a move that is comparable to a bullfighter’s turning his back on the bull: He confesses to being a little bugged by the usage in question, or perhaps by a related one. But he hastens to reaffirm that there is no real scholarly basis for such feelings, much less for the peever’s excessive wrath and abusive hyperbole.
In short, the linguist runs rings around the peever, making him seem at best ill-informed, at worst a snob, rube, or bigot. Such discussions are addictively interesting, impressively learned. You come away from them enlightened and more than a little abashed, resolved to take a more charitable view of ain’t, or could care less, or between you and I, or the misspelling of led (with an a after the e), or annoy me to no end or omitting the “Oxford comma” as I just did. We are all native speakers, after all, so anything that comes out of our mouths and pens is English ipso facto, emanating from roughly the same majestic body of precedent and unconscious principle. People are nobler and better than you thought, and you will quit finding fault with them.
This euphoria lasts until the next time you hear a human voice, actual or virtual, speaking for a minute or two. Within that interval, nearly always, something gets said that your onboard speech-processor greets with unreflective annoyance. There is simply no way not to notice hone in on or step foot in or based off of or relishing in his vacation, or the opposite choices if in fact these former are your preferences. So then you go back to your normal human state of wondering what the hell is wrong with everybody.
In his recent The Language Wars (a wonderful book, notwithstanding several rote, straw-man denunciations of prescriptivism), Henry Hitchings remarks, “I have yet to come across anyone who is never irritated by other people’s use of language.” Of course not, for no such specimen exists. Peeving, at least to the extent of having strong opinions about the right and wrong ways to say a thing, is roughly as spontaneous as speech itself, and can be readily observed in children. LL posters themselves often note (and lament) the speed with which a posted query or peeve will light up a chat room with (they insist) mostly uninformed, knee-jerk opinions. Commenting on such gripe sessions, Geoffrey Pullum (co-author of The Cambridge Grammar of the English Language, no less) half-jokingly complains, “And remember, when it's language, people never check. They never call a linguist. They just make stuff up.” Often, elsewhere, Pullum makes what seems to me an excellent general point: that those who pronounce on usage from positions of authority, in books or at editor’s desks or in classrooms, should be better guided by actual evidence of what people really do say and write. Amen to that! Here, though, he seems ready to ban amateur language punditry altogether: let the trained linguists pronounce their judgments, while the rest of us meekly suspend ours and suppress the underlying intuitions.
But language is just more democratic than that, and more social and more personal. Peeving turns out to be like voting: something you can’t delegate but must do for yourself. Where language is concerned, we treeless apes are all stakeholders and thus keenly curious, alert to any sign that the ground rules may be different from what we supposed. Even at the unconscious level, our listening is accompanied by metalistening: we are about equally interested in what is getting said and how, on the excellent principle that the language itself is more important, long term, than the French fries Bubba just now requested. We pass the fries, but also notice his usage, filing a quick mental note that, later, who knows, may be a template we either follow or deliberately avoid as we fashion phrases of our own. (If all this sounds improbably complex, well, think about it: how else do we ever learn to speak at all, except by constantly noticing language itself as well as the meanings it carries? How do we keep our sense of grammar and idiom fine-tuned over the years? Not by any system of formal instruction, as any linguist will point out.)
o peeving of at least a rudimentary sort, I want to say, is no dreamed-up mischief but just normal business, a natural by-product of our talent for language. If a peeve is not purely learned and artificial (in which case I grant its obnoxiousness), it begins as a nearly unreflective outcry, a simple alarm bell that goes off when a usage challenges our downstairs grammar. Something goes wrong at the level of easy, unconscious language function, and now, like it or not, the matter must be referred to consciousness for further scrutiny. But the problem is not maliciously made-up.
Of course, peeves tend to lose their primal innocence very quickly, growing shrill and doctrinaire almost as soon as they are expressed. Geoffrey Nunberg, another celebrated name at Language Log, usefully distinguishes between the initial, usually valid perception of “something not quite right” in a phrase and the later process of “fetishization” that “makes us stupid” as the feeling hardens into dogma. When my feeling becomes your taboo, we have a problem. When some tiny particular fetish (rear vs. raise, uninterested vs. disinterested) comes to be regarded as the essence of right thinking and cultivated behavior, we have a big problem. But how many peevers really grind their axes so hard? What most of them are still mostly doing (so I would argue), and what as speakers and stakeholders they have every right to do, is reporting their feelings about this or that usage. Simply that. What matters is not the reasons given, which too often are pure hash (as the bloggers of LL love to show), but the feeling itself, which as far as it goes is neither right nor wrong but simply a social and linguistic fact that needs to be dealt with.
Descriptivists often profess dismay at what they see — fairly enough, up to a point — as prescriptivists’ cavalier attitude toward rational argument: the inchoate edicts and missing evidence, the sitting-duck fallacies, the magisterial because-I-say-so argument. But there is a reason that even knowledgeable, high-end prescriptivists tend to affect this disagreeable persona. In three words, language is arbitrary (though “coherently arbitrary”). It is not finally a natural object governed by inexorable laws, but a body of far more tenuous and shifting social conventions: not what it rationally must be, but what we collectively decide it is going to be.
So the grammatical logic is there, of course, an endless and fascinating study, but in the end frustrating and nugatory. It never decides anything. This, again, is something linguists love to tell you: no language is an integrated logical system. The past tense of work tells you nothing about how to form the past tense of sing. Could care less, for those who accept it, says the opposite of what it logically should, as do inflammable and invaluable. Oversight is its own opposite, and neither doughnut hole nor self-addressed stamped envelope has any right to say what it does. If writers at the New York Times and everywhere else abruptly begin to say good-paying jobs, defying everything I thought I knew about adjectives and adverbs, it is a change no one could have predicted, and I have no choice but to tag the phrase as an allowed exception and soldier on, with a tight smile. And the examples of singular agendaand figurative really do not make a compelling case for singular data and figurative literally: English might simply be inconsistent on these points, as it is on countless others.
Ages ago, in my first year French class, I mustered enough vocabulary to ask why something was done as it was. The instructor laughed and told me that in language there was no “pourquoi,” or rather that the pourquoi of anything was just C’est comme ca. That little axiom states the case quite well I think. Language is the province of Rube Goldberg, not William of Occam. Its republic is ruled not by a sane central government but by innumerable warlords — that is to say, by competing paradigms that all take you a certain distance but no further, rules that invariably peter out at exactly that point where you wish to resolve a question of usage; for such questions occur in the first place only where different paradigms collide.
The upshot is that, in discussions of usage, the appeal to reason tends to be little more than window-dressing, or even a bit of a con, aimed at producing a consensus by pretending one already exists. At the heart of most prescription is a shockingly short argument: here’s how we do it, here’s how I feel about it, period, amen, take it or leave it. Even as thoroughly reconstructed a prescriptivist as Bryan Garner, who by his own account is really a descriptivist too (much as Pullum and Nunberg, whatever they say, are really also prescriptivists), who in his estimable Guide to Modern American Usage always analyzes terabytes of evidence before pronouncing judgment — even Garner in the end is mainly a cocky sonofagun proceeding a priori. For in every case that is even slightly interesting, the evidence really cuts both ways, “usage varies,” and his job is to make a tough call finally based on his own intuitions and sensibilities. Sometimes a brief rationale for the judgment is offered, but what really counts, in each entry, is the fateful asterisk that pronounces a form “invariably inferior” or not. Even the piles of evidence are just the same sort of argument-by-fiat, at one remove: an appeal to the intuitions and sensibilities of others rather than to his own. They endlessly record what choices people make, but tell us nothing about why. Another key point is that such evidence tells us only how the language has been used in the past, not necessarily how it will or should be used in the future, which is the only burning question anyone had in the first place.
But back to peeving. How likely is it that something we do so instinctively and incorrigibly, with such dark glee, is really bad for the language and us and the cosmos? How much sense does it make to condemn the earnest mutterers and eyebrow-raisers and letter-writers, the folks who know that something sounds funny even if they can’t put their finger on it?
I propose instead, (with a prescriptivist’s usual insouciance over the lack of evidence) that such grousing is useful and salutary. Elaborated as doctrine it takes us nowhere, and the crew at LL are dead right to keep challenging such developments. But accepted for what it is — venting, basically — it gets into the mix and is half-consciously assimilated and helps the language decide where to go next. Any language requires an astonishing degree of unquestioning agreement among speakers: convention after convention, governing everything from sound formation to word definitions to inflections and syntax, has to be established at the level of unthinking reflex for even the simplest conversation to proceed. Yet we know ourselves to be a contentious species. How could such a mind-bogglingly huge web of agreements — language itself — ever come into being? And what keeps the compact from unraveling?
Part of the answer, surely, is that our tetchiness about language serves us better than we know. Peeving looks like the ancient, not very efficient means whereby we maintain and repair our collective creation, keeping it more or less shipshape and sending it sailing on down through the ages.
Maybe someone will write a grant, fund a study, and prove that I am right or wrong. Till then, peeve on, my droogies, peeve on.
