France. When ideas die, a wit onee remarked, they go to America.rnWith the new age of international conferences on criticalrntheory, nobody thought Sir Philip Sidney had anything muchrnto do with the matter, or even Coleridge or I.A. Richards. I suggestrnit is time we did.rnAs Roland Barthes said, one never knows what is not to bernfound in the writings of the Anglo-Saxons. Consider, for example,rna book by a learned Dutch theorist, Douwe Fokkema, Theoriesrnof Literature in the Twentieth Century (1977), where I amrnpolitely attacked for having suggested that one can know whatrnterms like romanticism and tragedy mean without being able torndefine them. That is a claim I made in The Study of Literaturern(1969), a book that regrettably appeared in the same year asrnWittgenstein’s posthumous masterpiece On Certainty, which Irnwas consequently unable to use. Wittgenstein, being ViennesernJewish, hardly counts as an Anglo-Saxon, even if he spent mostrnof his adult life in England. But his point, in the notes on certaintyrnhe scribbled in Cambridge shortly before his death inrnApril 1951, was that all thinking beings possess certain knowledgernand need it, since even their uncertainties depend on certainties;rnthat one understands words less by verbal definitionrnthan by use; and that in any contested case, it is the instancernthat takes priority over definitions. “What would get judged byrnwhat here?” as his famous challenge in On Certainty put it. Torngive a blindingly simple literary instance; if a prof erred definitionrnof tragedy were to exclude King Lear or admit As You LikernIt, one would not reallocate the plays but scrap the definition.rnLear is known with certainty to be a tragedy without any definition;rnyou judge definitions against instances, not the otherrnway around.rnThe point is already in Coleridge, an author Wittgenstein isrnnot known to have admired. “On the immediate which dwellsrnin every man,” he wrote in the Biographia, “all the certainty ofrnour knowledge depends.” That is a very Wittgensteinian point,rnusing his word “certainty” to similar purpose. You cannot evenrnbegin to think without being certain of something. When yournlearn the two-times table at school, for example, or are told inrninfancy it is wrong to tell lies, you see with instant certainty thatrnit is so. It is less an act of discovery than of recognition, likernlearning to walk. Science is no exception. Every experimentalrnscientist, as a British scientist remarked some years ago, knowsrnthat he can use color-terms without first proving them to bernright. That was Harold Jeffreys in Scientific Inference (1973). Ifrnthe analytical chemist or botanist were to reject color-terms,rnJeffreys argued, he could no longer report or use his own sensations.rn”But he knows quite well that color seirsation is goodrnenough for his purpose.” In other words, scientists as well asrncritics assert, and rightly and with certainty assert, what theyrnknow they cannot prove: for example, that grass is green. Sorncritics, and for that matter moral philosophers, are not differentrnfrom scientists when they assert what they cannot prove. If wernhad to prove and agree on the foundations of knowledge of literaturernor morality, or of the physical sciences, we could notrneven begin to study such matters, still less progress in such studies.rnWe are right, then, on occasion, to believe what we cannotrndefine or prove.rnDr. Fokkema in his Theories of Literature objects to thisrnunashamedly antidefinitional view, quoting from myrnStudy of Literature, on the ground that it would “make usrnspeechless, and the results of our investigations impervious torncriticism.” But surely it is the other way around. It is the universalrndemand for definition, or the denial of what Coleridgerncalled “the immediate which dwells in every man,” that leavesrnthe critic speechless in the sense of being incapable of assertion.rnUnfortunately, it leaves him capable of talking without asserting;rnand that has become all too possible in an age of radicalrnskepticism recently erected by critical theory, as anyone whornhas ever attended a deconstructionist lecture must know. Onerncan say, and at enormous length, that there is nothing to bernsaid. That was the desperate point of Jacques Dcrrida’s Clasrn(1974), or The Knell, meaning the death-knell of Western civilization;rnand those who deny Coleridge’s point, or Wittgenstein’s,rnabout the immediacy of knowledge are less often renderedrnspeechless than vacuous; it is those who accept it, byrncontrast, who do real work. “Travaillons sans raisonner,” saysrnVoltaire at the close of Candide (1759), showing that you dornnot need to be an Anglo-Saxon to take the point: let us workrnwithout asking why. The critic, like the scientist and the moralist,rncan only work at all because he begins with certainties whichrnhe does not need to prove.rnColeridge returned to the point often. In the first chapter ofrnthe Biographia, for example, he praised his old schoolmaster forrnhaving shown him as a boy that poetry (no less than science) isrna cognitive activity:rnI learnt from him that poetry, even that of the loftiestrnand, seemingly, that of the wildest odes, had a logic of itsrnown as severe as that of science; and more difficult becausernmore subtle, more complex, and dependent onrnmore, and more fugitive, causes.rnThis is as profound a remark as any ever made about poetry, butrnit is regrettably concise and needs to be expanded. If the logicrnof a poem is as severe as that of a scientific proposition, it isrnnonetheless severe in a wholly distinct sense—”a logic of itsrnown”; distinct, in that it is more subtle, complex, and “fugitive,”rnas Coleridge puts it, meaning fleeting and indefinable; partly,rnno doubt, because it does not characteristically deal in technicalrnterms that are subject to verbal definition. It is not an objcctiorirnto the truth of a poem, accordingly, to conclude thatrnone cannot say what it is, just as it is not an objection to ourrnknowledge of color or of palatal taste—knowing how to tell coffeernfrom tea, for example, and blindfolded—that one cannotrngive a defining account of what one knows. Truth is wider thanrnaccount-giving, in fact, and we all know more than we can say.rnTo be unable to answer questions like “How do you know thernpoem means that?” or “How do you know it is good?” is notrnconclusive to the charge that one does not know. Dame MaggiernSmith, when asked (as she often is) about the art of acting,rndeclines to answer, replying “It is just something I do.” But nobodyrnthinks it follows from her refusal or inability to answer thatrnshe does not know how to act.rnColeridge’s profound recognition of silent knowledge and itsrnsignificance could be pressed further. Prunella Scales, to quoternanother actress, has told how she occasionally allows herself tornbe persuaded to give an address on the art of acting, and findsrnfor several evenings that she cannot do it. Or consider thernfollowing instance. I know English better than French, whichrnI speak only as a foreigner. But because I learned French late, Irnwould be better at giving air account of its grammatical rulesrnthan those of English. On the other hand, my English is right,rnand reliably right, and my French is not; if asked whether a sentencernis possible in English, for example, I can reply with cer-rn14/CHRONICLESrnrnrn