In 2015 Oliver Kamm gave us Accidence Will Happen, a guide to English usage. Whether by accident or by design, Kammâs guide traces the steps of Wittgensteinâs conversion from the logical atomism of the Tractatus to the intricately piecemeal non-philosophy of the Investigations.
The parallel isnât perfect, I know. Kamm is more of an empiricist than Wittgenstein, and Wittgenstein more of a pessimist than Kamm. Both authors make appeals to the empirical, but Kammâs appeal to the empirical has a statistical literalism that is alien to the Investigations. For Wittgenstein makes scant reference to any entity so literal as a statistic. The items that instantiate Wittgensteinâs doctrine that meaning is âuseâ include custom, practice, circumstance, history, behaviour, the everyday, the ordinary, âform of lifeâ (Lebensform) and just the various things weâve done. Any such appeal to the evidence as Kamm makes â an appeal that aspires also to be a citation of data â would have struck him as vulgar.
Kamm adopts the dispassionate attitude of a scholar; he shows little enthusiasm for speculative exploits. But â even so â his book reads like an allegorical play on Wittgensteinâs career. Where in Wittgenstein there are philosophical problems, there are, in Kamm, the well-known cruxes of English usage ('shibboleths', for Kamm). And in the place of philosophical practitioners there are the âgrammariansâ* â the authors of the prescriptive surveys of English usage, running from Fowler in 1926 to lesser lights such as Cochrane, Gwynne, Heffer and Humphrys in this century. Kamm calls these people âsticklersâ, taking a disobliging delight in cutting them down to size in a compressed, aridly careful English style.
I remember when Kamm was a banker who blogged. He now works as a journalist, writing leaders for the London Times. But, until he got the Times job, he was a blogger. In 2006 or 2007, his must have been one of the first blogs I read. I remember what a sententious scold he was, and how quick he was to take other writersâ offences against good usage as proof of intellectual inadequacy.**
That was Kammâs Tractarian phase. At that time, he stood for a linguistic essentialism whose equivalent in Wittgenstein is the logical atomism of the first six or so sections of the Tractatus. He confesses to it, too. In his own words he is a âreformedâ (or ârecoveringâ) stickler. Diagnosing his own disorder, he explains that it is characteristic of sticklers to make an appeal to a kind of essential grammar. This, for sticklers, is a grammar founded in logic. Its invariable presence promises in some way to make good the imperfect flux of day-to-day usage. In its light, and only its, our everyday English shall be judged.
Perhaps Kamm himself once thought he could write in so crystalline and clean a style as never to stray from the logic that underlies language. Perhaps the passionate conviction with which he pursues his anti-pedantic campaign derives from an experience of conversion which he only wishes others could experience as passionately as he did. If this is the case, Kammâs project is something like that of the reformed addict who guides initiates to the group through the first steps towards a cure. And it is something like that of the later Wittgenstein, who â in the Investigations â had already fallen victim to the philosophical disorder, and who, in practising a therapy, also pursued a cure. One of the critics cited on the back cover of Kammâs book calls it âdeeply reassuringâ. If so, it is presumably reassuring in the way that therapy is.
I wouldnât like to overstate the analogy. For one thing, Kamm is as much a satirist as he is a therapist. He has great fun with the risible ranters who write about grammar for a conservative readership. (As he nicely observes, their way of popularising discussion of language seems to be to link it to a generalised kind of declinist nostalgia.â ) But, if only pedantry is at issue, there's a surfeit of spite in his satire. In this book, Kamm likes not just to laugh but to lacerate â so that we may wonder whether his anti-pedantic project is in fact a metapedantic one. As far as this âmetapedanticâ project is concerned, only lesser pedants are his targets, but still, and again and again, they provide him with opportunities to reap a paltry glory from the humiliation of the ignorant â just as did the old pedantry he disowns.
For all that, the parallels are clear. Kamm and Wittgenstein are preoccupied by the same question. Both think theyâre in possession of a way to answer it. They both attract the same sorts of antagonists (there are compelling parallels to be drawn between the pathologies of the grammarians and the pathologies of the professional philosophers whom Wittgenstein has in his sights in the Investigations). Both, in the end, succumb to the same problems.
Framed philosophically, the question is this: Are there norms without essences? How can our words mean anything at all without being keyed in some way to a system of reference that lies immutably outside the individual? Of course there arenât just semantic norms: there are norms of all kinds. In epistemology, for instance, there are âepistemicâ norms, and there is this question: What is knowledge without an epistemological guarantee such as Descartesâ God grants in the Meditations? Wittgenstein addresses this question in On Certainty, but Kamm isnât so interested in it.
Take the question of what art is. This is a question in the philosophy of art. Or take the question of how we ought to use the word âartâ. This is a linguistic question, or a grammatical one (if you insist). It is a question on which an authority on usage might conceivably wish to issue a verdict.
Both Wittgenstein and Kamm take it for granted that we cannot have recourse to any essential entity in our answers to these questions. Instead, they ask that we appeal to the way the concept (in Wittgensteinâs case) or the word (in Kammâs) is used. So Wittgenstein asks that we have full confidence in any of his many proxies for âuseâ â in history, in custom, in practice, in the everyday or the ordinary. (The Wittgensteinian question, then, is not âWhat is art, in essence?â but âWhat have we customarily [or historically, etc] considered art to be?â) Kamm, for his part, asks that we have full confidence in something he calls âStandard Englishâ. When we ask him how to gauge, or even to identify, such a Standard, he is able to give a much more precise answer than Wittgenstein is when he is asked how to define âuseâ. In Kammâs case â in the case of the grammar of the word âartâ â we can have recourse to a corpus, or the OED, or, in quick-and-dirty style, to Google. Itâs in terms of bodies of evidence such as linguistic corpora that Standard English is defined. Is there any analogous method we can introduce in philosophy? It would be quite self-defeating to try to appeal to a linguistic corpus to adjudicate the question of what art is. But perhaps there are other literatures or bodies of evidence to which we might turn.
If the question asked by Wittgenstein and Kamm in context of their distinct concerns is the same, their answers are also alike: they would both have us derive definitions from the evidence. Lacking a belief, for instance, in Art Itself, we must derive our definition of it from a study of the conceptâs sublunary career, taking into account its history as well as its present state (so the abstract demand as to âthe evidenceâ can be met by citation from a corpus or a historical dictionary). Therefore neither thinks there is one answer. There is no answer in general. They take their questions one at a time. They give their answers case by case. They move from crux to crux, and from problem to problem, in a piecemeal way.
The essentialists' different projects â either the grammarians' project or that of the philosophers who believe that the definition of art is implicit in its essence â suffer the same setback. The grammarians are always getting caught making mistakes. In the treatises in which they lament the mistakes of others they invariably make many of their own. But even a book unblemished by any hypocrisy, an immaculate book that never fell foul of its own prescriptions â even this work would give the lie to the grammarianâs case. Imagine a grammarian who tried to define a certain class of uses of the word âartâ as ungrammatical. To refer to any sort of stand-up comedy as art, he says, is categorically bad English. Now if we ask why it is he exactly thinks this âbadâ grammar is bad, he typically gives one of two answers that will carry any weight with those of us who canât expect to share his sense of what is and isnât artistic. He may say that it is illogical to say stand-up comedy is art or that it is unintelligible.
If he says it is illogical, we should ask him what he means by logic. If he means merely that it doesnât make sense, he should explain himself (without accounting for the assertion, heâll just have restated his view). But itâs worth challenging him simply on the principle, here, and asking whether language should 'make senseâ in so rigid a way as to preclude a user of English from, for instance, saying âlessâ for âfewerâ with count nouns. How literally logical should natural language be? For what itâs worth, there are already many languages for writing logic (as there are many systems of logic): the pedant, if they want to communicate as logically as possible â and if they are not already Michael Dummett â should learn one. Natural language has to be recast as logic in a laborious process that bears comparison to translation, and â when it is â much of its content is lost with its context. If it has been so recast, this is no guarantee that it will be logically valid. Indeed, part of the point of rewriting things in logic is to check them for validity. If a sentence in its literal, logical form is found valid, it may still, of itself, make less sense and say far less than the natural-language sentence it was in the beginning. To insist that natural language be logical is something close to a category mistake.
But if he claims that itâs unintelligible to say stand-up is art, we need only turn to the treatise in which he offers a taxonomy of all the errors that offend him. In that treatise there will naturally be an entry for âartâ, and in that entry there will be examples of the word âartâ in the erroneous usage. In that entry, too, the grammarian will seem to have understood the instances of art in this, its erroneous usage: he will seem to have understood them so well as to know exactly why they were in error. Is it not legitimate to infer that he must have understood such usages if he has been able criticise them as systematically as he has? In effect, his treatise will show that the solecisms that afflict the sentences he employs as examples are not so severe as to have made them unintelligible.
It therefore proves impossible for the grammarian to flesh out his complaints without making them seem empty. Perhaps the only way would have been to list them anonymously â that is, to make a list of sentences seen to be in error without glossing it as such. The idea of silence as a solution is familiar from the Tractatus. Wittgenstein's project in that book is to put an end to philosophy, reducing all its questions to an atomised array of answers, and this project falls victim to the same sort of problem as the grammarian's. For if the philosopher aspires to seal off philosophy in a sphere of pure sense, as the grammarian might aspire to separate all correct sentences from their incorrect counterparts, he would seem to have to implement the idea without letting us know, in silence and in secret.
The grammarianâs problem is that the sentence or sentences in which he defines the errors that offend him show those errors not to be errors at all, or are otherwise themselves in error; this problem is illustrated ideally by the pedantic intervention that makes a mistake of its own.⥠The philosopherâs problem is that the sentence in which he might have expected to declare the final unravelling of philosophy, and the full separation of sense from nonsense, may not itself speak sense. If the sentence speaks sense, it invalidates itself: it makes nonsense. If it instead speaks nonsense, and is unintelligible, its success remain unknown. But if its success does become known, it must be undone
The task then is to find a way to fall silent without letting on, for silence is the last thing the word âsilentâ achieves. The philosopher should perhaps become a performer, a mime. Failing this, the philosopher must give up the project to put an end to philosophy as misguided, as arising finally from a mistake. The grammarian who cures himself of pedantry, and learns to see language as a linguist does, learns also to see grammarians as dilettantes and cranks, each with his own little collection of cheap complaints, each on his own irrelevant crusade against the corruption of language, each claiming to speak more logically than the last.
The philosopher must likewise remake himself. He must be a non-philosopher (and Kamm is ânon-pedanticâ). As a philosopher, he had suffered from a sort of disorder; it left him incurably confused, and the problems of philosophy gave him no relief. It all seemed as vital as if his very life were at stake. But, as a non-philosopher, he knows that these problems are no problems at all. In our difficulties with them we betray a basic confusion, attesting to an approach that does not discover philosophical problems, as it imagines itself to be doing, but creates them. For they are spurious. They are figments, and merely spells to fall under. No one need try to solve them. They are as spurious as the grammarianâs recommendation to use âdecimateâ only when speaking of divisions by ten.
The philosopher then plays the part of the philosopher against philosophy, as Kamm plays the pedant against pedantry. His philosophy lurches from particular to particular. His solution is not to define, but to do, and do, and do. He has no more disciples: his philosophy is no longer pressed into a package of ideas fit to be digested, disseminated and unpacked again. He has only imitators now. In a sense his method is only to be himself. It does not survive him: it depends on his being there to show us how itâs done
So say he appeals to what people actually do, or to how they actually use the word âartâ, preferring not to prescribe to them what they should ideally do, or how they should use the word âartâ ideally. Thatâs all well and good â if he is prepared never to stop philosophising, as his previous approach promised he might. The thing is, language is adding to itself all the time. Invoking an evidence-base such as that of the linguistâs corpora, he has to watch it remake itself before his eyes. When he asks âHow is the word âartâ standardly used?â, heâs obliged to answer that âartâ must be defined, defined and yet again defined. If he issues a verdict, he's bound to advise that his verdicts are subject to revision from the moment that they are issued. As fresh evidence re-establishes the Standard, and more of our âerrorsâ and lapses become acceptable by its lights, work always remains to be done.
We shouldn't need to be persuaded that Kammâs problem, the pragmatic problem of repetition, is preferable to the philosophical problem of regress. That the non-philosopher swaps the philosopherâs old problems for newer, better problems â this is undeniable. But he would be wise not to claim too much for his reformed method, result as it is of failure â or of the success that is achieved by an ability to admit defeat where others simply keep going hopelessly on. And Kammâs way of being high-handed with his antagonists does suggest that he has misjudged the actual power of his approach.
Take the question of art, again. A variable definition of art derived from the evidence was never what we wanted, any more than we wanted an invariable definition derived from its essence. Inconvenient though it is for philosophy, the one definition enters into the other. Thatâs why the philosopherâs project went wrong: the definition of nonsense is implicated in a dynamic way in the definition of sense, putting into question the project's pivotal sentence, the sentence in which it is to be completed. And itâs not that, when the philosopher underwent his conversion and cured himself of philosophy, his project was saved. No, he gave it up, or at least outsourced it to the ambiguous public whose practice is what âordinaryâ language is. Like the philosopherâs, the pedantâs project failed because correct and incorrect language are interdependent and impossible in isolation. Kamm (as an anti-pedant) does not bring this project back to life, crediting the corpus or the OED with the power to determine what is and what is not correct. Instead he credits them â a little circularly â with the power to define what is Standard.§
In his role as a âreformedâ pedant, Kamm is required to act as a publicist and even as an activist on behalf of his new programme. He is therefore unable to give due weight to the antithetical principle that makes his analysis possible. No meaningful appeal to the empirical could be made without the involvement of this principle, which â though it neednât be an essence or even a constellation of atomic propositions â cannot either be supplied by evidence alone. The evidence cannot of itself furnish criteria for its own evaluation: indeed, it provides no intrinsic guarantee that âartâ should form any kind of category at all, and still less does it contain any intrinsic criterion that would serve to restrict the instances of âartâ as they arise in it to a meaningful minimum. In itself, it lacks the materials to make provision for any degree of limitation such as might be prescribed by an individual set of specifications particular to an individual inquirer. This is why, without any antithetical principle to focus oneâs inquiry, one cannot define its terms, and cannot hope for significant results (without such a principle, one is constrained merely to repeat the evidence as it emerges). In the case of the word âartâ, for example, the evidence does not of itself provide the means for us either to include this quotation in our inquiry or not to include it:
To find out that, good Shepherd, I suppose,
In such a scant allowance of Star-light,
Would over-task the best Land-Pilots art
Without the sure guess of well-practizâd feet.
When we ask what art is, are we asking about such âartâ as a âLand-Pilotâ might practise?
Of course Kamm is no kind of fundamentalist, and he never makes the dogmatic claims I criticise here. He is a columnist and, to that extent, he is concerned to impress his opinions on us. His true error is to take his opponentsâ indulgence in dogmatism as an occasion for an overreach of his own.
Kammâs strategy is to insist that the grammariansâ concerns are groundless, but in this he takes the empiricist approach too far. Itâs easy to see that, though they arenât grounded in evidence, they are grounded in intuitions or âhunchesâ such as most English-speakers share. Had they been purely idiosyncratic, they could not have endured in the unkillable way that they have. The grammariansâ crime isnât to have conjured up complaints out of snobbery or whatever other prejudice takes their fancy. Their crime is to have raised their aesthetic complaints to the status of logical ones. The split infinitive is an example. Are their complaints about it legitimate? The intuition in this instance is that the split infinitive seems too much to mix adverb and verb, making a new verb where none is wanted and none needed. A split infinitive too often primes us to hear an intrusive hyphen between a verb and its adverb (and is it the same to 'go boldly' as it is to 'boldly-go'?§). But Kammâs politicised empiricism sees him entangled in still deeper difficulties.
It is to his credit that he has applied a little scholarship to the issue, putting in time at the library in an attempt to produce a history of the prescriptions against this or that pedantic bugbear. To that extent, his appeal to the evidence pays off. But, by the same token, he might feel compelled to produce statistics showing how much traction was gained by all the particular rules at particular points in this broader story. There are judgements he would need to make in order to produce such statistics, and the evidence cannot make these judgements for him. Regarding split infinitives, he would need to determine: did a particular author decide not to split a particular infinitive in deference to a prescription or not? Should any particular non-split infinitive be considered a product of the kind of pedantry his method calls into doubt â and, if it is, should he then exclude it from the evidence constituting the Standard?
The âŹ60 000 question is this. If the evidence shows some infinitives unsplit and some split, but says nothing about whether any one infinitive might stand unsplit as a result of a piece of pedantry, how can Kamm appeal only to the evidence in judging whether or not split infinitives are Standard? Say the evidence is divided 50/50, or even 60/40. If the empiricist knows that infinitives with adverbs only go unsplit when a writer has bought into some piece of false pedantry, can he then avoid recommending that as Standard? Or would his own method compel him to recommend as Standard a prescription that has risen to a regrettable popularity?â
Above, I made mention of an âantithetical principleâ to which Kamm does not give adequate weight in his arguments. Nothing about Kammâs method works if his appeal to this principle is denied him, and heâs made to rely on evidence alone. It is a principle that could not be derived from any body of evidence; at the same time, it informs all his judgements about the evidence. Once it is admitted, however, there is no knowing how far it will take him back to the arbitrary pedantry he rejects. Insofar as there is a lesson to be learned here, it is that Kamm should not have made his prescription against prescriptions so perversely strict. If his righteous reaction against the pedants leads him never to split, something â somewhere â must be going wrong. He should surely not be seen to be indulging self-consciously in the mistakes he used to scold others for making in ignorance. That is no kind of recovery from pedantry. It is its grisly sequel: Pedantry â The return.
If the appeal to the empirical is to be more than a gesture â that is, if it is to be pursued for more than its effect as a gesture â then it will need to be aided by an antithetical principle, which will complement it as the opposable thumb complements the open hand. But there is a catch here â there is a bug in our code â which is that the moment such a principle is invoked, the empiricist restores to life all the old problems his appeal was supposed to sweep away. So â instead â he holds one hand behind his back. And this is his fate: to proceed serenely, without encountering any pitfall or trap, but to proceed without progress, for all hope of progress has been sacrificed for his serenity, a serenity that the hand behind his back might at any moment sweep aside once again.
This doesnât make Kammâs book worthless. It is not. But its readers should bear in mind that its prescriptions against prescription are made in the light of a method with only a little more intrinsic validity than the philosophy it replaces. Its readers should remember, too, that its anti-pedantic practice doesnât so much arise from an honest desire to offer guidance and advice as from an arbitrary ambition to prove the merits of this method (whose main feature is more that it is not the philosophy that failed us than that it is the method it is). So Kamm recommends his method in the spirit in which a reformed alcoholic might recommend absolute sobriety to an audience of lightweights and social drinkers; perhaps it comes as a shock to him, but they have never had much difficulty keeping their drinking in check.
And the philosopher of art? The aesthetician? Where is he in all this? Dishevelled and spent, is he not now a defeated figure, a figure of fun? Will he not now be the butt of the studentsâ jokes?
Throwing in his lot neither with the invariable nor in the variable definition of art, and unprepared to claim either that there is one thing art is or that there is no thing it is, he finds himself forced to compromise. Art â he accepts â must lie in the difference between its ideal instance, the last artwork, the art of the infinite, which we may deny, but never renounce, and whatever art we call âartâ, the finite art of time. In us, each informs the other, and, each being indispensable, neither is privileged. This give-and-take is what is, and what will be, so long as we live without a norm of norms â a rule of rules â that resolves all regresses, and forever justifies their coming to an end with us, here, at this time.
* In Kamm, the term âgrammarianâ doesnât always appear in the pejorative sense (as if it was a synonym for âpedantâ). In this review it is only used in that sense. When I want to refer to grammarians in the positive sense, I call them âlinguistsâ.
** I remember too that the Iraq War had few more dogmatic or more vigorous advocates in the British blogosphere. I venture no opinion as to whether his loss of faith in linguistic hawkishness coincided with any comparable loss of faith in the neoconservative ideas he espoused in his 2005 book, Antitotalitarianism.
â Kamm is too confident that the pedantâs sense that his language is in some way suffering a decline amounts to a fallacy (âmany of the [language] pundits instinctively confuse change in the language with decline. Thatâs an errorâ). This idea of decline may be expressed more colourfully as decadence, debasement, degeneracy, corruption etc. And we may rightly register our distaste for such language, but if we call it fallacious we invest our distaste with a false rigour that is merely rhetorical in force. It is dishonest of the linguist, taking a generalised perspective on the evidence, to so insistently deny the particular perspective of the pedant â for the fact is that there is a pedant. The pedant is a mortal man or woman born in a particular place at a particular time. If there is that mortal man or woman, there is decline. Unless that man or woman is a fallacy, there is no fallacy.
⥠This, from the Guardian, is interesting: âSo if guns kills people, I guess pencils miss spell words [sic], cars drive drunk and spoons make people fat.â (I'm quoting a quotation). Does the 'sic' apply to both errors ('guns kills', 'miss spell)', or did the Guardian itself introduce the first error? Is the sign's 'miss'-spelling not ironic, and the Guardian's 'sic' not a bit tin-eared, a sort of subeditorâs kludge?
§ An Ngram on this famous split infinitive (credit: Yian Shang). Itâs interesting to see how the rise in popularity of âto boldly goâ is reflected in the rise of âto go boldlyâ in a sort of pedantic afterimage. Iâm not sure how a descriptivist wishing to provide guidance on this point should react to whatâs happened here.Â
â Note that when Kamm or any other commentator makes a self-effacing appeal to the evidence as part of an anti-pedantic project, they also assert their authority as a sort of spokesperson. If the grammarian tries to demote himself in deference to something greater than him, the anti-pedant denies that there is any such entity to defer to, and characterises the grammarian as idiosyncratic, as a crank. But the evidence does not have a voice of its own, either. No linguist is its transparent tribune, any more than the grammarian is a mere instrument of a greater truth. Does the anti-pedant demote himself in deferring to the evidence, or does he promote himself as its spokesperson? In Britain in 2016, were we not reminded that, for every poll presuming to take a neutral reading of the nation's temperature, there is also a pollster?
