to the logical status of the question,rnwhich is to say that it matters enormously.rnSuppose, for example, a demographerrnwere to approach the Chinese problemrnon the subjectivist assumption thatrnany view of the total population of Chinarnis as good as any other; that it all dependsrnon who, where, or what you are; that it isrnall ideological. Surely the effect of suchrnan assumption on the science of demographyrnwould be shattering. No estimaternwould be false; no need could ever arisernto refine or improve estimates alreadyrnmade. In other words, a scientific inquiryrnwould have lost its status as a formrnof knowledge and entered into the shadowyrnworld that critical theory has inhabitedrnnow for a generation or more, wherernno view is better than any other except inrnits power to stimulate or amuse.rnIf the case for critical subjectivism isrnnow lost, where does theory go fromrnhere?rnObjectivism has already scored arnnumber of hard-fought gains in recentrnyears, and it might be useful to summarizernthem. Knowledge is not the same asrnaccount-giving, as the case of the speechlessrninfant shows; so an inability to answerrndoes not, always and necessarily,rnsignify ignorance. Accuracy is not precision,rnso the imprecision of much criticalrnlanguage is not a charge against its accuracy.rnNot all beliefs can be ideologicallyrndistorted: if they were, after all, thatrnwould be an objection to the very beliefrnthat all beliefs are ideologically distorted.rnAnd the ignorance of literary critics, likerntheir failure to agree, makes no casernagainst critical objectivism, since physicistsrnand chemists too are often ignorantrnor divided. Knowledge thrives on conflict,rnin any case, and there is nothing tornbe concerned about wheir critics disagreernor confess to uncertainty. Norndoubt any natural scientist would happilyrnaccept that most of the facts of the presentrncosmos, which number millions ofrnbodies, are still unknown and likely to remainrnso. Add the possibility of anotherrncosmos beyond this one, and the claimrnlooks indisputable and in no way dispiriting.rnAny intellectual inquiry, whether inrnthe arts or the sciences, has reason to rejoicernthat there is work to be done.rnSo what work is there in critical theoryrnto be done?rnOne large task is to describe what peoplerndo, and at some risk to clarity, sincernthe word has other meanings, I shall callrnthat task descriptivist. A descriptivist, inrnthis context, is one who describes howrnjudgments are made rather than dogmatizingrnabout how they ought to be made.rnA clearer way of putting this might be tornsay that, when asked how he knowsrnsomething, he takes the question literally.rnThat, one should be warned, is highlyrnannoying behavior, and no descriptivistrnshould look for an easy ride. If Irnwere asked, for example, how I know therntwo-times table, I might artlessly replyrnwith the name and address of the primaryrnschool where I learned it and even, if Irncan remember, with the name of thernteacher who taught it to me. That answerrnis wholly accurate. It is indeed howrnI know. It is also, like other accurate answers,rnwholly unhelpful, since I mightrnhave been taught the same truths anywhere,rnand for just that reason it is deeplyrnenraging. But then if the skeptic hadrnwanted to know whether my reasons forrnbelieving in simple arithmetic were sufficientrnreasons, he should have had therncourage to ask that question. At least arnliteral answer, enraging as it may be, hasrntaught him one highly important truth:rnthat a truth can be believed for reasonsrnwholly inadequate and irrelevant, andrnthat it is none the less true for that.rnSuppose, similarly, someone were tornask how you know that a work of art isrnbeautiful or that murder is wrong. Suchrnquestions, in practice, are seldom questions.rnThey are more often challenges,rnimplying that beliefs must be justifiablernif they are to be allowed to count asrnknowledge. But the implication is absurd.rnPlenty of certain knowledge cannotrnbe justified. I do not know how anyonernwould justify the belief that murderrnis wrong, and yet, like most people, Irnhold it to be certain. Those who seek torndeny or qualify’ such certainties with ingeniousrncounter-instances like thernbomb-plotters who tried to kill Hider inrn1944 forget that not all killing is murder,rnand that an elaborate body of argumentrnknown as casuistry sorts out suchrnmarginal cases as fyrannicide, justifiablernmanslaughter, or self-defense.rnOnce again, argument can go forwardrnby going back. Coleridge offered thernoutline of a descriptivist answer when, inrnan essay called “On the Principles OfrnGenial Criticism” (1814), he remarkedrnof an ancient marble in the Vatican Museumrnknown as the Apollo Belvederernthat “it is not beautiful because it pleases”rnbut “pleases us because it is beautiful.”rnIrr recent years theorists have toorneasily dismissed naturalistic argumentsrnlike these. Naturalism means acceptingrnthat goodness and beaufy are constituentrnaspects of the world, and that if we fail tornrecognize them the failure is squarelyrnours, much as the color-blind cannotrnrecognize colors even though they arernthere. It ofl^ers a hope for critical theory;rnand if theory has a future, it is there.rnFor theory to turn naturalistic wouldrnrepresent a startling break with the traditionrnof Sartre and Barthes. But so itrnshould. In recent years the skeptical traditionrnhas turned into an unhappy mixturernof the cynical and the vain: cynical,rnbecause the effort to justify grants andrnsalaries has forced the skeptic into pretendingrnthat he is doing something significantrnwhen he knows he is not; vain,rnbecause it has cast the theorist back onrnhimself, claiming an interest in his opinionsrnon the sole ground that they definernhis own nature and personality. “Thernway I see it is . . . ” can easily mean nornmore than “The sort of person I amrni s . . . ” Narcissism is all that is left.rnAs that mood runs its course, therernwill be a mounting demand to say something,rnand something that is not aboutrnthe critic. A cynic is never more than arnspectator, usually a bored one; and thernvain easily forget that other people are interestingrntoo. That is the mood that leadsrnback to naturalism. What if, after all, literaturerncan be known, understood, andrneven (on occasion) described? Thernthought is unnerving. It can also excite.rnUntil it does, literature is condemnedrnto be a playground rather than a subject,rnand that is bad news. True, people willrnsometimes pay good money to watch arngame. But I suspect the literary critic isrnill-placed in any competition with thernMatch of the Day or the World Series.rnThis is a competition he simply cannotrnwin. If literature is only a game, thenrnthere are better games; if it is a kind ofrnknowledge, on the other hand, then it isrnnot always clear there are better kinds ofrnknowledge. That is where its greatrnchance lies. Rilke once remarked admiringlyrnof Cezanne that he did notrnpaint “I like it” but “There it is.” Theoryrnnow needs to do that too—to stop lookingrnat itself, its own likes and dislikes, andrnstart looking at what is there. For there itrnGeorge Watson, a Fellow of St. John’srnCollege, Cambridge, is the author ofrnWriting a Thesis (Longman),rnThe Certainty of Literaturern(St. Martin’s) and British Literaturernsince 1945 (St. Martin’s).rn46/CHRONICLESrnrnrn