Analytic and Continental Philosophy [Explaining the Differences]

ANALYTIC AND CONTINENTAL PHILOSOPHY:

EXPLAINING THE DIFFERENCES

NEIL LEVY

ABSTRACT: A number of writers have tackled the task of characterizing thedifferences between analytic and Continental philosophy. I suggest that theseattempts have indeed captured the most important divergences between the twostyles but have left the explanation of the differences mysterious. I argue thatanalytic philosophy is usefully seen as philosophy conducted within a paradigm,in Kuhn’s sense of the word, whereas Continental philosophy assumes much lessin the way of shared presuppositions, problems, methods and approaches. Thisimportant opposition accounts for all those features that have rightly been held toconstitute the difference between the two traditions. I finish with some reflectionson the relative superiority of each tradition and by highlighting the characteristicdeficiencies of each.

Keywords: analytic; Continental; Kuhn; paradigm.

Since the early twentieth century, Western philosophy has been split intotwo apparently irreconcilable camps: the ‘‘analytic’’ and the ‘‘Continen-tal.’’ Philosophers who belong to each camp read and respond to theirfellows almost exclusively; thus, each stream develops separately, and thedifferences become more entrenched. Relations between the camps arecharacterized largely by mutual incomprehension and not a littlehostility. But because few philosophers are well acquainted with bothsides, the nature of the split is not well understood. This essay is intendedto contribute to understanding this split and perhaps, in a small way,even to overcoming it. Now, at the dawn of a new century, it is time toput the divisions within philosophy behind us. Although, as I shallsuggest toward the end of the paper, we have cause to temper ouroptimism at the prospects for imminent reconciliation, understanding thedifferences is a first and indispensable step toward overcoming them.

A few caveats about the ambitions and limitations of the essay are inorder before we turn to an examination of the differences. First, I onlyintend to characterize general trends and tendencies. Thus a singlecounterexample will not serve to falsify my view; only a sufficient weightof counterexamples could accomplish this. Second, I do not claim that allphilosophers working in contemporary academia, not even all of those

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2003.Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and350 Main Street, Malden, MA 02148, USAMETAPHILOSOPHYVol. 34, No. 3, April 20030026-1068

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2003

working in what is clearly the Western tradition, fit neatly into thesecategories. Some philosophers work between the traditions, as it were, inways influenced by both: Alasdair MacIntyre and Charles Taylor come tomind here. Others work wholly or partially in traditions never entirelyassimilated into one stream or the other: this is true of Americanpragmatism to some extent, and of philosophers whose work developedprimarily through the reception of ancient philosophy. Apart from theintrinsic interest of their work, these philosophers serve a salutaryfunction: they remind us that valuable philosophy can be done by thosewho do not owe allegiance to either side of the divide. However, theirexistence does not alter the fact that our philosophical landscape isdominated by the split.

The essay will fall into two parts. In the first, I shall examine somerecent attempts to characterize analytic and Continental philosophy(hereafter, AP and CP, respectively). I shall suggest that all fail to statenecessary and sufficient conditions that could function as criteria bywhich to classify particular philosophers or their works. Indeed, there areno such criteria, or so I shall contend.

Ought we to conclude, then, that AP and CP are Wittgensteiniancluster concepts – that is, concepts that lump together many differentelements, elements that share no common core? I suggest that thisapproach would be more fruitful. But I do not want to stop there. I wantto see if something more can be said concerning AP and CP, somethingover and above ‘‘characteristically, each tends to possess such and suchfeatures.’’ I want to see, in fact, if it isn’t possible to account for thesecharacteristic features. If my suggestion is on the right track, I shallaccount not merely for the positive features of AP and CP but also for thedifficulty I have experienced when I have attempted to bring them intodialogue with each other, and the unsatisfactory nature of most othersuch attempts at such ecumenicalism.

Characterizing the Traditions

I turn, then, to characterizing the differences. One way of approachingthe topic immediately suggests itself: taking our cue from the wordContinental, we might attempt to classify each camp on the basis ofgeography. CP will thus be that kind of philosophy practiced in mainlandEurope, or taught in philosophy departments there. But it is evident thatthis approach will fail. For one thing, AP itself has its roots on theContinent as much as CP does (Dummett argues that it would moreappropriately be called Anglo-Austrian than Anglo-American [Dummett1993, 1–2]). For another, there are currently plenty of analyticphilosophers on the Continent – Jacques Bouveresse and Pascal Engelin France, for example – and even more Continental philosophers inEnglish-speaking countries. ‘‘Continental’’ cannot, therefore, refer primarily


ANALYTIC AND CONTINENTAL PHILOSOPHY 285

to a place, anymore than ‘‘analytic’’ can be entirely synonymous with‘‘Anglo-American.’’1

If not geography, then perhaps style will provide us with the criterionwe seek. So Bernard Williams suggests, in the preface to Ethics and theLimits of Philosophy:

What distinguishes analytical philosophy from other contemporary philoso-phy (though not from much philosophy of other times) is a certain way ofgoing on, which involves argument, distinctions, and, so far as it remembers totry to achieve it and succeeds, moderately plain speech . . . it distinguishessharply between obscurity and technicality. (Williams 1985, vi)

There are, I think, two claims at issue here. The first concerns the place ofargument in the two traditions. It is often said that what distinguishesanalytic from Continental philosophy is the greater place and respect forargument in the former. Dagfinn F�llesdal, for example, characterizes thedifference between analytic and nonanalytic philosophy as essentially adifference in the place given to arguments, rather than rhetoric (F�llesdal1997). The claim here is that CP is not rigorous. Perhaps the best-knownexample of this claim is the letter sent to the Times in 1992 to opposeCambridge’s proposal to grant an honorary doctorate to Derrida:

In the eyes of philosophers, and certainly among those working in leadingdepartments of philosophy throughout the world, M. Derrida’s work does notmeet accepted standards of clarity and rigor . . . his writings . . . seem to consistin no small part of elaborate jokes and puns (‘‘logical phallusies’’ and the like)and M. Derrida seems to us to have come close to making a career out of whatwe regard as translating into the academic sphere tricks and gimmicks similarto those of the Dadaists or of the concrete poets. (Derrida 1995, 420)

The letter is signed by, among other people, David Armstrong, RuthBarcan Marcus, Keith Campbell, Kevin Mulligan, Barry Smith, and noless a figure than Quine.

I think it is not simply prejudice – though it is also prejudice – that canlead thinkers like Quine to this conclusion. There are really positivefeatures of CP that give the impression to those trained in the analyticschool that it is argument free. The proof of this is that the difference inthe role of argument in the two traditions is recognized by theContinentals too. This recognition comes out in the counteraccusationfrequently heard: that AP is a new scholasticism, where the concern fortechnique overwhelms the very problems that the techniques hadoriginally been designed to solve. An adequate characterization of the

1 As David Cooper says, ‘‘The continent, for our purposes, is not a place, but atendency’’ (Cooper 1994, 2). Bernard Williams has suggested that the very terms in whichthe distinction is drawn are ‘‘absurd.’’ Because one term refers to geography and the other tomethod, they involve ‘‘a strange cross-classification – rather as though one divided cars intofront-wheel drives and Japanese’’ (Williams 1996, 25).


NEIL LEVY286

differences between the two traditions will allow us to account for boththese accusations.

I said there were two issues at work in my original quotation fromWilliams, and for that matter in the letter to the Times as well. The firstissue concerned the place of argument. The second, more promising,places the burden of differentiating the two traditions on style in abroader sense. It has often been noted that CP is more ‘‘literary’’ than isAP; perhaps that is all the difference consists in. This approach wouldhave the advantage of supplying us with the means to account for theapparent lack of arguments in CP. Sometimes, we might conclude, thisschool lets its concern for style override its concern for ideas, allowing thecoloring of sentences to take precedence over their clarity.

Of course, there are important stylistic differences between the twotendencies, but if that were all the difference amounted to, the difficultyin bringing the two schools into dialogue with each other would beinexplicable. Merely stylistic differences are superficial, and such surfacedifferences ought to yield relatively easily.

Perhaps we might try to account for the difference simply in terms ofhistorical origins and reference points. Indeed, there is no doubt that animmediately striking difference between the two involves the standardthinkers that each refers to. If an article cites, on the one hand, Frege,Russell, Quine or Davidson and, or on the other, Husserl Heidegger,Derrida or Gadamer, it is usually clear which tradition the work is in. Butmore than this needs to be said. Noting this fact simply pushes thequestion back one step: it is now in order to ask in what the differencesbetween these thinkers consist.

If the differences cannot be characterized solely in terms of the place ofargument or of style, what of content? David Cooper has made apersuasive case for the difference lying on this level. Cooper claims thatthree themes ‘‘run through the writings of the most influential continentalthinkers . . . which have no similar prominence in the analytical tradition’’;they are ‘‘cultural critique, concern with the background conditions ofenquiry and . . . ‘the fall of the self’’’ (Cooper 1994, 4).

I agree with Cooper that these themes are of special concern for theContinental tradition, but I doubt they can serve as criteria to distinguishanalytic from Continental philosophy. Cultural critique is a necessaryconcern of all political and social philosophers; to say that it ischaracteristic of CP is just to say (rightly) that political and socialphilosophy are more important in the Continental tradition than in theanalytic. This, in turn, is at least in part the result of the relative lack ofspecialization by Continental philosophers, among whom the myriadsubdisciplines into which AP divides itself (ethics and metaethics,philosophy of mind, of language, metaphysics, epistemology, and soon) are relatively unknown. Because Continental philosophers are notengaged in specialized subdisciplines, they do not partition the potential



ethical and political implications of their work off from its other aspects.They thus engage in cultural critique at the same time as they developtheir philosophy of language, or whatever else they happen to be workingon.2 To explain this phenomenon, then, requires the prior explanation of,on the one hand, the presence and, on the other, the relative lack of suchsubdisciplines.

The relative lack of specialization also goes some way to explainingCooper’s second theme, the ‘‘concern with the background conditions ofenquiry.’’ Because Continental philosophers typically tend to bepolitically engaged, they are more interested in the political stakes andconditions of knowledge, and thus in laying bare the nonrational factorsthat condition knowledge. This feature of CP is one with which manyanalytic philosophers are especially impatient, since they see in it aconfusion of the context of discovery with the context of justification, ora commission of the genetic fallacy. Nevertheless, it is not an approachshared by all Continental thinkers, or only by Continental thinkers. Ithink much the same could be said of Cooper’s third theme – Parfit showsas great a delight in dismantling our common-sense picture of the subjectas does any poststructuralist.

More fruitful, I suspect, is another suggestion of Cooper’s: that ‘‘anti-scientism’’ characterizes Continental thought (Cooper 1994, 10). Con-tinental thinkers have often objected to the hegemony of science inmodern culture, insisting that it represents neither the only kind ofknowledge nor even the most basic kind. Instead, they have tended tohold that scientific knowledge is secondary or derivative: derived, that is,from our more primordial existence in the Lebenswelt. This has been atheme common to Husserl, Heidegger, Merleau-Ponty, and Levinas, andit reappears – though much transformed – in Lyotard and Foucault. Incontrast, and as Cooper notes, ‘‘analytical philosophy has generallyproved more friendly and sympathetic to science’’ (10).

I said that I thought this suggestion was a fruitful one. I do not believe,however, that it is sufficient to serve as a criterion to distinguish betweenanalytic and Continental philosophy – not unless we classify JohnMcDowell as a Continental. Moreover, I find Cooper’s ‘‘explanation’’ forthe contrast unhelpful. Cooper argues that, although both traditions tooklinguistic turns, AP turned toward a systematic explanation of language,which is conducive to a scientific approach, whereas CP turned insteadtoward a conception of language that cannot be made systematic, since itholds that language exists only as embodied in linguistic practices(Cooper 1994, 13–15). But if these turns are indeed characteristic of ourtwo traditions, they are part of the data to be explained, not the

2 As Vincent Descombes says, the tracing of its political implications is regarded as the‘‘decisive test, disclosing . . . the definitive meaning of a mode of thought’’ – no matter howapolitical the content of that thought might seem to be (Descombes 1980, 7).


NEIL LEVY288

explanation. This line of thought is promising, but it awaits furtherdevelopment.

Much the same can be said for Simon Critchley’s position. Critchleycontends that the essential difference lies in the two traditions’ attitudesto history. CP, he argues, approaches its problems ‘‘textually andcontextually’’; it holds that ‘‘philosophical problems do not fall from thesky ready-made and cannot be treated as elements in some ahistoricalfantasy of philosophia perennis’’ (Critchley 1997, 353–54). Because thereare no such things as the eternal problems of philosophy – no problem ofuniversals, for example, which might be traced from Aristotle toArmstrong – problems can only be approached in their historicalcontext. Thus, the history of philosophy is almost isomorphic withphilosophy sans phrase for the Continental philosopher. For her, thenotion that there are such perennial problems is hopelessly naıve.

This, too, is a promising suggestion, one that captures much that ischaracteristic of the Continental approach. I suspect, however, that it toocannot serve as a necessary and sufficient condition to define theContinental (John McDowell would once again serve as a counter-example). More important, it leaves the differing attitudes of the twotraditions toward history unexplained. This too is an explanandum, notan explanation.

Let me now consider, finally, one of the best-known attempts tocharacterize analytic philosophy, that of Michael Dummett. Dummettargues that analytic philosophy is defined by its concern with language:

What distinguishes analytical philosophy, in its diverse manifestations, fromother schools is the belief, first, that a philosophical account of thought can beattained through a philosophical account of language, and, secondly, that acomprehensive account can only be so attained. (Dummett 1993, 4)

Now, I doubt neither that Dummett’s characterization accurately reflectscertain tendencies in AP nor that this tendency is central to the self-imageof the school. But this concern with language that he finds definitive ofthe school is in fact neither necessary nor sufficient. It is not sufficientbecause the concern for language is shared with many, perhaps most,Continental philosophers, at least some of whom share the convictionthat thought is reached only via language. Derrida’s obsession with thelimits of language, for example, is a consequence of his conviction thatthey are the limits of thought. And the concern with language is notnecessary because a number of philosophers who are indisputablyanalytic are little concerned with language: Rawls, for example, and mostof political philosophy – or, if political philosophy is felt to be tooperipheral, we could point to the relatively small role that linguisticanalysis plays in contemporary metaethics, where its importance hasdecreased over the past decades.



Another aspect of my disagreement with Dummett’s views isconcerned with his implicit characterization of CP. Throughout OriginsDummett talks of ‘‘the phenomenological school’’ (1993, ix, 26). Now, Isuspect that there is no such school. Or at least if there is such a school, itis very far from being isomorphic with CP. Husserl was and is animportant figure in CP, and a training in phenomenology is consideredimportant in Europe. But it is hard to find even a trace of Husserl’sinfluence in Foucault or the mature Lyotard (despite Lyotard’s havingwritten an introduction to phenomenology). Even in the mature work ofmany figures who stem from the phenomenological tradition – Levinascomes to mind – few of Husserl’s methods and concepts find a place.

I think, in fact, that CP does not form a school, and that this fact is ofmore than marginal importance. CP is not an ongoing research program,in the sense in which AP takes itself to be. Account for this difference,and I think we have the explanation we seek.

Accounting for the Differences

It would be possible to go on multiplying the cliches concerning AP andCP. I suggest, though, that we would continue to fail to locate anynecessary and sufficient conditions definitive of the two traditions. To besure, we might be able to isolate sufficient conditions. We might, that is,think that our list of cliches, perhaps supplemented by others, wouldserve as a set of characteristics, the possession of a sufficient number ofwhich would allow us to classify a philosopher into her correct category.Though this may well be the case, it nevertheless falls far short of anexplanation of the differences.

What, then, could explain the division? I suggested earlier that theattitude of each tradition toward science, though not by itself definitive ofit, would be a promising starting point. For AP, we noted, scienceoccupies a central position. This is true with regard both to its subjectmatter – AP is more often realist, even reductively materialist3 – and to itsstyle. AP, in Pascal Engel’s words, ‘‘mimics the scientific style of inquiry,which proposes hypotheses and theories, tests them in the light of data,and aims at widespread discussion and control by the peers’’ (Engel 1999,222). In contrast, CP is closer to the humanistic disciplines and toliterature and art. Once again, this is the case with regard both to contentand to form. Derrida’s infuriatingly ‘‘literary’’ style has already beenremarked upon; as regards content, a list of books by major Continentalphilosophers that focus on art and artists would include:

3 This is not to say that all work in AP is realist. This is far from being the case.Nevertheless, AP has a realist orientation. We might best express this by saying that theconsensus in AP is that the burden of proof is on those who would deny realism. If anything,the burden of proof in CP is on those who would uphold it.


NEIL LEVY290

What Is Literature? (Sartre)Kafka: Towards a Minor Literature (Deleuze and Guattari)This Is Not a Pipe (Foucault)Signe Malraux (Lyotard)The Truth in Painting (Derrida)

The list could very easily be extended. It is, to be sure, a very French list,and I suspect that this emphasis on art and literature is more pronouncedon the French side of the Rhine. Nevertheless, important strands ofGerman philosophy have been similarly concerned with art – witnessHeidegger’s preoccupation with Holderlin, Trakl, and George, as well ashis famous ‘‘On the Origin of the Work of Art,’’ or the idea propagatedby the Frankfurt School that only the avant-garde artwork can resistcommodification.

I suspect that this contrasting emphasis runs deeper than mostphilosophers have realized, and that the place of science in the twotraditions is the most important element in any explanation of theirdifferences. I think, though, it is not the contrasting objects of AP and CPthat are central here but the formal analogy it is possible to constructbetween, on the one hand, AP and the physical sciences and, on the other,CP and the arts. Analytic aesthetics is, after all, still analytic. It istherefore to the formal analogy that I now turn.

I propose to develop this analogy by comparing AP, as a self-reproducing discipline, to the image of science we find in Kuhn’sStructure of Scientific Revolutions. In Kuhn’s text we shall find not onlymany of the characteristic features of AP repeated in his description ofscience but also the tools we need to explain those characteristics.

My suggestion is this: AP has successfully modeled itself on thephysical sciences. Work in it is thus guided by paradigms that function inthe way Kuhn sketches, and the discipline is reproduced in somethingakin to the way in which the sciences are reproduced. CP has a quitedifferent approach to its subject matter, a quite different model of whatphilosophy is, which guides its characteristic concerns and shapes itsmethods.

Analytic Philosophy as Normal Science

The evidence for this claim rests upon the number of features of AP thatKuhn’s model explains. I shall give a brief account of ‘‘normal science,’’following Kuhn, and then turn to the parallels between it and AP.

Normal science is a research program that is guided by a paradigm.Paradigms are ‘‘universally recognized scientific achievements that for atime provide model problems and solutions to a community ofpractitioners’’ (Kuhn 1970, viii). It is the common acceptance of aparadigm, Kuhn argues, that accounts for the rapid progress made by



mature sciences. Two features of paradigms explain this rapidity.Acceptance of a paradigm is simultaneously the acceptance of anontology and a methodology. It thus puts an end to debate about suchfundamentals. This has the result of allowing scientists to concentratetheir attention on problems rather than methodology. Just as important,it changes the character of that attention itself:

In the absence of a paradigm or some candidate for a paradigm, all the factsthat could possibly pertain to the development of a given science are likely toseem equally relevant. As a result, early fact-gathering is a far more nearlyrandom activity than the one that subsequent scientific development makesfamiliar. Furthermore, in the absence of a reason for seeking some particularform of more recondite information, early fact-gathering is usually restrictedto the wealth of data that lie ready to hand. (Kuhn 1970, 15)

The second important feature of paradigms concerns the kind ofproblems they delineate. A paradigm dramatically narrows the field ofpossible problems. The postparadigm scientist is concerned only withthose problems that are sufficiently similar to those the paradigm hasalready successfully solved. As a result, science is transformed into apuzzle-solving activity. These two factors – the one focusing the attentionof the scientist upon problems and away from fundamentals, the otherrestricting the scope of problems and transforming them into puzzles –explain the appearance of swift progress that characterizes postparadigmscience. This progress is exactly what one should expect from an activitywhose practitioners ‘‘concentrate on problems that only their own lack ofingenuity should keep them from solving’’ (Kuhn 1970, 37).

Once scientists have a relatively clear standard to which to refer injudging relevance, once their field of vision is narrowed by a sharedparadigm, they are enabled to achieve in-depth, rather than broad,knowledge of nature. Prior to the establishment of a paradigm, scientistswork upon whatever features of the natural world seem most interesting,or upon whatever seems related to practical matters, whereas acceptanceof a paradigm enables scientists to perform painstaking research uponarcane and obscure features of reality, features that would not even havecome to light without the work the paradigm permitted: ‘‘The confidencethat they were on the right track encouraged scientists to undertake moreprecise, esoteric, and consuming sorts of work’’ (Kuhn 1970, 18).

The acceptance of a paradigm has one significant drawback, however– or at least what seems at first sight to be a drawback. That is, theparadigm encourages the suppression of novelty. Novelties aresuppressed for two convergent reasons. On the one hand, they ‘‘arenecessarily subversive of its basic commitments’’ (Kuhn 1970, 5). Sinceno place can be found for them within the paradigm, acceptance ofnovelties is rejection of the paradigm. But the scientist cannot reject theparadigm, unless it is to move to an alternative paradigm, on pain of


NEIL LEVY292

ceasing to be a scientist. She is, therefore, bound to suppress the isolatednovelty. In fact, very often she will simply fail to notice it altogether:‘‘Novelty emerges only with difficulty, manifested by resistance, against abackground provided by expectation. Initially, only the anticipated andusual are experienced even under circumstances where anomaly is later tobe observed’’ (Kuhn 1970, 64). Moreover – and here we move to thesecond reason for the suppression of novelty – when anomaly is observed,the scientist is committed to believing that further research within theparadigm will succeed in explaining it. That, ultimately, is whatacceptance of a paradigm amounts to: commitment to the belief thatthe paradigm will be able to account for all the features of the realm towhich it applies.

As is well known, this last feature of paradigms is the mostcontroversial, and it leads to Kuhn’s difficulties with explaining paradigmchange. It is not this feature that concerns me here, however, but theparallels between the practice of AP and Kuhn’s picture of normalscience.

AP, I suggest, is the philosophy built upon acceptance of the work ofFrege and Russell as a paradigm. Reassess the features of AP in the lightof this suggestion, and all those features we noted suddenly fall intoplace. AP is, as we have seen, essentially a problem-solving activity:indeed, it is this feature of their work that its defenders are most proud of:

Continental philosophy . . . is problemarm. ‘‘Ask me what I’m working on, andI’ll reply with the name of a problem,’’ the Analytical Philosopher will proudlysay, ‘‘ask them, and they’ll reply with a proper name.’’ (Mulligan 1991, 115)

[Analytic philosophy] aims to solve particular problems, puzzles andparadoxes, and to build theories in answer to them. It prefers to work upondetails and particular analyses, rather than to produce general syntheses.(Engel 1999, 222)

I am suggesting that the difference noted here is genuine, and that it stemsfrom AP’s being (something akin to) a normal science.

If I am right, and AP is a problem-solving activity, we should expectprecisely that proliferation of subdisciplines which characterizes thediscipline. Normal scientists need precisely delineated puzzles upon whichto exercise their skills. Accordingly, the analytic philosopher cannotaddress herself to the meaning of life, or to discovering the good life.Instead, she focuses on cognitivism versus noncognitivism, or refining theutilitarian calculus, or the mind-brain identity question, and so on.

Of course, to the extent that these are her problems, the work ofRussell and Frege will be relatively unhelpful to her. That paradigmcannot inform her work as directly as it does that of a logician, or aphilosopher of language. Instead, she will be guided in her subdisciplineby what Kuhn calls an exemplar. Exemplars are ‘‘concrete problem-



solutions’’ (Kuhn 1970, 187). Unlike paradigms, however, exemplarsneed not be shared by the scientific community as a whole. Instead, eachsubdiscipline may possess its own – albeit partially overlapping – set. It isthe differences between sets of exemplars that, more than anything else,‘‘provide the community fine-structure of science’’ (187). All physicists,for example, share a large number of symbolic generalizations, as well asa number of exemplars. As their training develops and the youngphysicist specializes in one or other branch of the science, ‘‘the symbolicgeneralizations they share are increasingly illustrated by differentexemplars’’ (187).

In the same way, I suggest, the metaethicist and the logician, thephilosopher of language and the philosopher of mind, possess a set ofshared and a number of divergent exemplars. All may have had thedistinction between sense and reference impressed upon them, butKripke’s and Putnam’s extensions of this work will matter much more tosome of them than to others. These others might find their majorexemplar in the new riddle of induction, for example, or in A Theory ofJustice.4

When a body of knowledge makes the transition to becoming ascience, we can predict on the basis of Kuhn’s work that not only thesubstance of its research but even the manner of its presentation willundergo important changes. As a result of the acceptance of a normalscientific paradigm, the scientist’s ‘‘research communiques’’ begin tochange:

No longer will his researches usually be embodied in books addressed . . . toanyone who might be interested in the subject matter of the field. Instead theywill usually appear as brief articles addressed only to professional colleagues,the men whose knowledge of a shared paradigm can be assumed and whoprove to be the only ones able to read the papers addressed to them. (Kuhn1970, 20)

Once a paradigm is accepted, the scientist can simply assume it: she doesnot have to rehearse previous findings, nor need she defend much of hermethodology. Moreover, the kind of puzzle she will typically beaddressing will be relatively discrete. It will lend itself to treatment in afew – albeit dense – pages.

As these remarks would lead us to expect, AP and CP present theirresearch in differing forms. As Kevin Mulligan notes, ‘‘The book, not thepaper, is the preferred philosophical genre on the continent’’ (Mulligan1991, 116). It is easy to think of important philosophers in the analytictradition whose reputation rests on journal articles alone, or whose books

4 The way in which A Theory of Justice revitalized political philosophy is an especiallystriking instance of the way in which AP requires an exemplar to stimulate research in itssubdisciplines. Prior to its publication, political philosophy had been moribund, a result ofthe difficulty in transferring the analytic paradigm to the political and social arenas.


NEIL LEVY294

tend to consist of collections of previously published articles – FrankRamsey, Bernard Williams, and Donald Davidson spring to mind.Gettier would be an extreme example. In CP, by contrast, considerablereputations always rest upon considerable books.

I have so far been explicating the features of AP in terms of its being aparadigm-guided activity. CP has been characterized only insofar as ithas emerged from the discussion as the mirror image of AP. I now turn toa more direct characterization of it.

There is an ambiguity in Kuhn’s work with regard to practices thataim at knowledge acquisition but are not normal sciences. SometimesKuhn suggests that they are carried out in the absence of paradigms(1970, 11); at others he suggests that they have their own kind ofparadigm, of a character different from that possessed by the sciences(179). Whether the practitioners of these alternatives ways of seekingknowledge should be regarded as being in possession of a paradigm ornot, it is clear that they do not agree on fundamentals to anything like thesame extent that the practitioners of ‘‘normal’’ science do. They do notshare a sense of which problems are important and tractable and whichnot. Thus, they must justify their choice of problems, their methods ofapproach, even their ontology, as well as putting forward solutions tothose problems. This is one reason they more frequently producemonographs than do scientists – able to take much less for granted, theyhave to build their field from the ground up.

Since this is precisely the situation of the Continental philosopher,work in CP proceeds in much the same manner as did work in physicaloptics prior to Newton: ‘‘Being able to take no common body of belieffor granted, each writer on physical optics felt forced to build his fieldanew from its foundations’’ (Kuhn 1970, 13). In the absence of aparadigm, we do not get the segmentation of the field of knowledge wefind in normal science, paralleled by the many subdisciplines of AP.Instead, we get fragmentation: the division of the discipline into rivalschools. This is characteristic of preparadigm science (12), and it is, Ihave argued, characteristic of CP. That is why it is misleading to speak, asDummett does, of ‘‘the phenomenological school.’’ There is no suchschool, if by that is meant an entity which would be isomorphic with theextension of the concept CP.

Explaining the Explanation

That, then, is my suggested explanation of the differences between APand CP. The first took, not the linguistic turn – for that it shares with CP– but the paradigmatic turn, and it modeled itself on normal science. Ittherefore turned itself into a number of highly specialized, technicaldisciplines. The second did not take that turn, and it instead remained a



more ‘‘literary’’ genre: more accessible, more concerned with practicalmatters, and more historically orientated.

This, as it stands, has the virtue of accounting for a number of thecharacteristic features of both AP and CP, but is not yet a fully satisfyingexplanation. To achieve that, we need to go further, and explain why APtook the paradigmatic turn, whereas CP did not. I confess that I do nothave a full explanation. But I do have one suggestion that will be part ofit. This explanation is sociological.

The fact that CP is characteristically advanced in the form of booksthat are relatively accessible (if only in the sense that they have the roomto explain their specialized terminology), while AP is advanced in theform of journal articles that are generally comprehensible only to otherspecialists, is explained by the differences in the audiences to whomContinental and analytic philosophers typically address themselves. Asnormal scientists, analytic philosophers address themselves to fellowspecialists. But Continental philosophers commonly address themselvesto the educated public at large. This is no doubt due, in important part,to the fact that philosophy is taught in high schools across Europe. Anunexpected consequence of this expansion of the potential audience ofphilosophical texts is the relatively small part played in CP by detailedresearch upon esoteric questions, as opposed to more wide-rangingspeculation. The analytic philosopher addresses specialists she knows willshare her technical vocabulary and her sense of what problems she oughtto be concerned with. The Continental philosopher addresses an educatedlayperson she knows will possess at least an outline knowledge of thehistory of Western thought.

That, I suspect, is part of the reason for CP’s frequently notedhistorical sense. In the absence of a shared paradigm, of a shared set ofproblems and exemplary problem solutions, it is the history ofphilosophy that binds CP together. It is the element shared byphilosopher and philosopher, philosopher and audience.

But history also plays a more positive role in CP. It does not simplyprovide the unifying force; it is also the horizon within which all problemsare understood. For the Continental philosopher, the delineation by APof the perennial problems of philosophy is hopelessly naıve: it is absurd,for example, to treat Aristotle on akrasia as a response to the same set ofquestions Donald Davidson addresses in his work in the philosophy ofaction.

The explanation for this emphasis on history is, once again, to be foundin educational practices. And on this contrast Kuhn is illuminating onceagain. His holism would, I suspect, lead him to side on this question withCP: just as the Newtonian concepts of space, time, and mass only partiallyoverlap with their Einsteinian counterparts – since concepts get theirmeaning from their place in a network – so, I suspect, for KuhnAristotelian akrasia and Davidsonian weakness of the will are only


NEIL LEVY296

partially overlapping concepts.5 Why, then, does AP persist in treating thehistory of philosophy essentially as a set of attempts to solve its problems?

Kuhn suggests that the parallel tendency in science is the result of aretrospective illusion, caused by the manner in which science is taught.He sketches an illuminating contrast between the way in which scientistsare initiated into their community and the way education is conducted inother fields. In the humanities, Kuhn suggests, students are typicallytaught, at least in important part, by way of exposure to the original textsof a variety of authors from a variety of historical periods. They are not,therefore, led to adopt one paradigm, one set of canonical puzzles as theproblems of their field, nor (typically) are they under pressure to adoptone set of methods or one approach to tackle these problems. Instead thestudent is confronted with a wide variety of problems, drawn from theentire history of the discipline, and ‘‘has constantly before him a numberof competing and incommensurable solutions to tackle these problems,solutions that he must ultimately evaluate for himself’’ (1970, 165).

When this approach characterizes philosophical education, we canexpect its students to be historically oriented, to disagree amongthemselves as to what the most fundamental problems of philosophyare, and therefore to turn to history itself and to its study to unify theirdiscipline. They will be acutely aware of historical differences and alive tothe subtleties that characterize the approach of individual philosophers.Ask them what they are working on, and, as Mulligan remarks, they willfrequently reply with a proper name: Husserl, Hegel, Aristotle.6

But this is not the kind of approach to education we find in the naturalsciences. Instead, education there takes place largely through thetextbook, not the reading of the original texts of great scientists. Andtextbooks, Kuhn says, are inevitably ‘‘systematically misleading’’ aboutthe history of the sciences (1970, 137); they

refer only to that part of the work of past scientists that can easily be viewed ascontributions to the statement and solution of the texts’ paradigm problems.Partly by selection and partly by distortion, the scientists of an earlier age areimplicitly represented as having worked upon the same set of fixed problemsand in accordance with the same set of fixed canons that the most recentrevolution in scientific theory and method has made seem scientific. (138)

5 This is even clearer with regard to Aristotelian virtue as contrasted to the virtue of thevirtue ethicists: kindness, as has often been pointed out, has no place in the Aristotelianview. Notice, too, that here the many techniques which AP has evolved to limit or eliminateKuhnian incommensurability hardly get a grip at all. For the causal theory of reference tocome into play, for example, we would need to be sure that we were referring to entities orphenomena that exist independently of our views and attitudes toward them – which is atleast not obviously the case with regard to human virtues and weaknesses.

6 Thus Critchley has things exactly backwards in his characterization of CP: CP is notantiscientistic because it is so historical; it is historical because it is antiscientistic (though nodoubt the tendencies are mutually reinforcing).



A complete and accurate representation of the history of the sciences runscounter to the aims of scientific education. The educational institutionaims to initiate the student into the scientific community. It seeks to leadthe student to share the reactions and dispositions of that community, to,as Kuhn puts it, view ‘‘the situations that confront him as a scientist inthe same gestalt as other member of his specialists’ group’’ (189). Thisgoal is most efficiently achieved when the student is able to trust herteachers; when, that is, she has no doubt that the approach they areimparting is uniquely definitive of science. To expose that student to avariety of other approaches to science, then, to other and rival sets ofpuzzles and interpretations of nature, is not simply functionless but mightactually serve to undermine the aims of scientific education. Thetextbook, therefore, teaches little history, and what history it doesmention is simply the story of how scientists came to free themselves fromerror and superstition on their way to building the viewpoint of normalscience.

I suggest that the approach to philosophical education which ischaracteristic of AP is closer to this model than is the approach in CP.Once again, I speak of tendencies; naturally the student of philosophycan never be educated by way of textbooks to anything like the sameextent that the young scientist can be. Nevertheless, the tendency is clear.Students of metaethics are regularly presented with just two or threepages of Hume, for example. Once we have read that ‘‘reason is, andought to be, the slave of the passions,’’ we have located his work as acontribution to the articulation of a canonical problem – that of therelation between the motivational and the cognitive aspects of morality.It goes without saying that this reckless disregard for the historical andcultural context of ideas is anathema to CP.

Assessing the Traditions

If this characterization of the differences between AP and CP is right, ornearly so, it inevitably confronts us with the question as to which is thebetter way of doing philosophy. The balance sheet is mixed.

Of course, as the application of a Kuhnian framework would suggest,we are faced with the difficulty that the two traditions are unlikely toagree on standards of evaluation. Instead, each is likely to possess its ownset of standards, standards by which it will do well while its opponentsscore badly. We need, so far as possible, to find standards that are notquestion-begging.

One such standard immediately suggests itself. If my description iscorrect, we should expect AP to be capable of progress. As a normalscience, it should be able to make rapid progress on the problems it setsitself. Perhaps this should be counted as a point in its favor, but we needto be careful here. While it may indeed be the case that AP – and only


NEIL LEVY298

AP – progresses, that is simply to be expected: progress and theacceptance of a normal scientific paradigm are correlative notions. Not tomake progress within a discipline defined in relation to such a paradigmis to fail, but outside such a discipline the notion of progress simply hasno application. As Kuhn himself notes, this kind of pattern is ‘‘notunfamiliar in a number of creative fields today, nor is it incompatiblewith significant discovery and invention’’ (1970, 13).

On the other hand, if Kuhn’s description of the systematicallymisleading way in which history is represented in normal science isaccurate, we should expect CP to have a clear advantage when it comes tothe history of philosophy (and probably the philosophy of history aswell). I suggest that this is indeed the case. Even such a defender of AP asKevin Mulligan (for whom CP is ‘‘is inherently obscure and obscurantist,often closer to the genre of literature than to that of philosophy; it isdevoid of arguments, distinctions, examples and analysis; it is problem-arm’’ [1991, 115]) finds analytic history of philosophy characteristicallymarred by the failure to understand the context in which ideas weredeveloped (116). Though Mulligan wouldn’t concede it, it is also possibleto believe that there is more than a little truth to the Continentalphilosopher’s claim that the lack of historical sense in AP renders thetreatment of at least some of its problems superficial.

Another point in CP’s favor concerns the ability of the latter toaddress practical questions. There is, I suspect, a trade-off at work here, atrade-off whose existence we might have deduced from Kuhn’s text. Theadoption of a normal scientific paradigm has as a consequence a greatlyincreased specialization of science: a much greater concentration on amuch smaller area. As a result, science becomes increasingly divorcedfrom the day-to-day concerns of non-scientists. ‘‘Frequently . . . re-volution narrows the scope of the community’s professional concerns,increases the extent of its specialization, and attenuates its communica-tion with other groups, both scientific and lay’’ (1970, 170). With theacquisition of a paradigm, AP acquired a set of relatively well-delineatedproblems or puzzles, upon which it was able to focus almost all itsattention and thus to make great progress in solving them. As a result,however, it came – rightly, in my view – to be seen as less and less relevantto the kinds of pressing questions that often drive people to philosophyin the first place.7

7 To the extent that analytic philosophers do address practical questions – engaging inwhat they call applied ethics, for instance – they risk finding themselves in the position of thescientist who writes books: ‘‘More likely to find his professional reputation impaired thanenhanced’’ (Kuhn 1970, 20). There is something of a paradox here. Applied ethics isparadigm AP, in as much as it is one more relatively well-defined subdiscipline. Yet thephilosophers who engage in it are looked down on by others in AP. No such problem arisesfor those working in CP, who tend to engage with practical questions as part of largerprojects.



In CP, this situation is reversed: it addresses questions of greaterrelevance with a much greater frequency than does AP, but does so in a waythat is – from the perspective of AP – rather shallow. I suggest that there is,in fact, a trade-off between relevance and depth, at least depth of the normalscientific kind. (Of course, what I am indicating here is only a tendency.Philosophy cannot follow the normal sciences and leave its own foundationsunexamined; not, at least, without ceasing to be philosophy. For the samereasons, it cannot abandon the examination of the fundamental questionsthat draw people toward it. What I am suggesting is that AP tends tochannel its students away from those questions, and in the direction ofdetailed work on its puzzles, to a greater extent than does CP.)

Thus far, the scales seem finely balanced. AP can legitimately claim tomake progress, but only because it has set itself a relatively tractable set ofproblems to deal with. If AP can claim greater depth and rigor, CP canclaim greater social relevance. And CP can claim a decisive edge when itcomes to writing the history of philosophy. If anything, it is CP that seemsto have the advantage. At this point, however, I would like to register aconcern with regard to CP. Since understanding that concern requiresunderstanding what I take its aim to be, I turn now to sketching that aim.

To characterize the contrasting goals of AP and CP, it is helpful to seethem as alternative responses to cultural modernism. A few landmarks inthe genesis of AP (their dates alone are suggestive):

Begriffsschrift (1879)‘‘Uber Sinn und Bedeutung’’ (1892)‘‘The Refutation of Idealism’’ (1903)‘‘On Denoting’’ (1905)Tractatus Logico-Philosophicus (1922)

These are, it goes without saying, the years of the flowering of modernismin the arts, the years of Mallarme and Eliot, Picasso and Joyce. These arealso the years in which the seminal texts of CP are written:

On the Genealogy of Morals (1887)Logical Investigations (1900)Ideas (1913)Being and Time (1927)The Transcendence of the Ego (1937)

My – tentative – speculation is this: modernity is characterized by twocompeting impulses, which find expression most distinctly in the naturalsciences, on the one hand, and in modern (nonrepresentational) art onthe other. In the one, research is an essentially cumulative enterprise, and,when revolutions disrupt its continuous progress, it rewrites its ownhistory so as to represent even the revolutions as a particularly fruitfulpart of the continuum. In the other, novelty and revolution are actively


NEIL LEVY300

sought, not suppressed. The most able painter hopes, not to perfect analready existing style, but to produce his or her own. As has often beennoted, continuous revolution is characteristic of the avant-garde. This, Isuspect, is not because it is looking for something that it has not yet beenable to find but because revolution is its very goal.

I suggest, therefore, that CP models itself on modernist art, just as APmodels itself on modern science. Hence the dizzying succession ofrevolutions in philosophy that characterize its progress: phenomenology,existentialism, Marxism, structuralism, poststructuralism, nouveau phi-losophie, each attempting, not to build on its predecessors, but to replacethem.

Hence, too, what I take the goal of CP to be. The avant-garde artist, Isuspect, typically has the goal of leading us to see the world anew, from adifferent perspective. Hence the constant need to revolutionize in art, tooverthrow ways of perceiving before they become sedimented intohabitual dispositions. Something like this is, I suspect, the goal of theContinental philosopher. Hence her constant urge to begin again, toquestion the foundations of philosophical systems, particularly of thosesystems that, she believes, shape the common-sense and everydayperception of her entire culture. Thus the problem of social transformationis the constant horizon of her work. This demand that philosophy innovate,that it allow us to think anew, is captured by Foucault’s definition:

What is philosophy today . . . if it is not the critical work that thought brings tobear on itself ? In what does it consist, if not in the endeavor to knowhow and to what extent it might be possible to think differently? (Foucault1986, 8–9)

It is this conception of philosophy to which Lyotard subscribes when hedefines the most important task of the philosopher as being the search fornew vocabularies to express as yet unrepresented experiences. Equally, itis this conception that is at stake in Deleuze and Guattari’s recentdefinition of philosophy not as the analysis but as the invention ofconcepts. As a definition of philosophy in general, I suspect this failshopelessly. As a definition of CP, however, it may be spot on. Newconcepts enable us to see the world anew, through eyes rejuvenated by therevolutionary philosopher.8

8 This description of the goals of philosophy will, no doubt, put one in mind of RichardRorty. For Rorty, too, the aim of philosophy is to invent new vocabularies so as to enable usto play new language games; not to solve puzzles so much as to invent new ones. It is nocoincidence that Rorty, like the Continentals he often appropriates for his own ends, alsosees philosophy as essentially a kind of writing. For him, AP is essentially ‘‘the same sort ofdiscipline as we find in the other ‘humanities’ departments. . . . The normal form of life in thehumanities is the same as that in the arts and in belles-lettres; a genius does something newand interesting and persuasive, and his or her admirers begin to form a school ormovement’’ (Rorty 1982, 217–18). It is because this is Rorty’s conception of philosophy thathe is so widely regarded as an apostate by analytic philosophers.



Thus, whereas AP sets itself the goal of solving its relatively well-delineated problems, CP glories in the fact that it will not define its pro-blems in advance. To do so would be to foreclose too many possibilities,to prevent the thinking of the radically new.

I can now sketch my worry. Normal science, as we have seen, does notseek novelty. Indeed, it will often actually suppress it, until it becomes tooinsistent to be ignored any longer. Nevertheless, and for that very reason,it is, Kuhn claims, peculiarly ‘‘effective in causing them to arise’’ (1970,64). For, just as Davidson showed that disagreement emerges onlyagainst a background of agreement, so novelty only emerges with clarityagainst the background of the expected:

Novelty ordinarily emerges only for the man who, knowing with precision whathe should expect, is able to recognize that something has gone wrong.Anomaly appears only against the background provided by the paradigm. Themore precise and far-reaching that paradigm is, the more sensitive an indicatorit provides of anomaly and hence of an occasion for paradigm change. (Kuhn1970, 65)

My worry, therefore, is that CP’s very insistence on always being open tothe radically other might prevent it from recognizing instances of thealterity it seeks. Seeking alterity everywhere, it might fail to see where itappears most massively. One must have expectations for them to bedisappointed. Or, to use Kuhn’s language, one must have a paradigm inorder to experience its revolutionary overthrow. I see no way to steer amiddle course here, and no way to sublate the opposition either. It seemsthat either our philosophy will seek novelty, and risk never being able tosee it, or it will work to suppress it (and perhaps grasp it all the moreclearly for that).

Whether or not CP does suffer from this fault, the fact that we haveidentified strengths and weaknesses in both styles of philosophy suggestsan obvious course for philosophy to follow in the future. We could hopeto combine the strengths of each: to forge a kind of philosophy with thehistorical awareness of CP and the rigor of AP. Is such a philosophypossible? Kuhn’s work implies that it is not: ‘‘The depreciation ofhistorical fact is deeply, and probably functionally, ingrained in theideology of the scientific professions’’ (138).

If a student is educated historically, if she is exposed to a history that isnot systematically misrepresented, she will not become a normal scientistor an analytic philosopher. If she is to become one, she must acquire theappropriate dispositions. In particular, she needs to learn see her field inthe appropriate way, learning to see her problems ‘‘in the same gestalt’’ asother members of her discipline (189). She must learn ‘‘to group objectsand situations into similarity sets’’ (200). Now, acquiring this ability tosee her field in the appropriate manner requires immersion in its world.The student must be constantly exposed to exemplars of the kind of


NEIL LEVY302

seeing into which she is being initiated. Thus, the kind of educationrequired to turn out normal scientists is antithetical to the kind ofeducation required to turn out people with a sense of history. To achievethe first, we expose the student to examples of the appropriate kind ofprocedure, until she comes to share the intuitions of the group. Exposureto alternative methods, to other ways of seeing the world, would here becounterproductive. But to produce students with a historical sense, wedeliberately expose them to as wide a variety of ways of proceeding aspossible, inviting them to enter the thought styles of each. In this kind ofeducation the student ‘‘has constantly before him a number of competingand incommensurable solutions to these problems, solutions that he mustultimately evaluate for himself’’ (165). Educating students in this way isbound to produce thinkers who disagree among themselves, who sharenot a paradigm but only a set of texts.

If this is correct, we have little reason to be optimistic that AP and CPcould overcome their differences and produce a new way of doingphilosophy that would combine the strengths of both. But we cannevertheless hope that the situation is not as bleak as this application ofKuhn’s work to it suggests. There may yet be a way to steer between thisparticular Scylla and Charybdis. What the details of this middle waymight be, I do not know, but we can point to the increasing signs of ahistorical consciousness among analytic philosophers – evidenced by therecent work of John McDowell and of Hilary Putnam, and the return toAristotle among analytic ethicists, for example – as a sign that it ispossible.

Centre for Applied Philosophy and Public EthicsDepartment of PhilosophyUniversity of MelbourneParkville, Vic [email protected]

Acknowledgments

I would like to thank Armen Marsoobian and an anonymous reviewerfor Metaphilosophy for their very helpful comments on an earlier draft ofthis article

References

Cooper, David E. 1994. ‘‘Analytical and Continental Philosophy.’’Proceedings of the Aristotelian Society 94: 1–18.

Critchley, Simon. 1997. ‘‘What Is Continental Philosophy?’’ InternationalJournal of Philosophical Studies 5, no. 3 (October): 347–64.



Derrida, Jacques. 1995. Points . . . Interviews, 1974–1994. Translated byPeggy Kamuf et al. Stanford: Stanford University Press.

Descombes, Vincent. 1980. Modern French Philosophy. Translated by L.Scott-Fox and L. M. Harding. Cambridge: Cambridge UniversityPress.

Dummett, Michael. 1993. Origins of Analytical Philosophy. London:Duckworth.

Engel, Pascal. 1999. ‘‘Analytic Philosophy and Cognitive Norms.’’Monist 82, no. 2 (April): 218–34.

F�llesdal, Dagfinn. 1997. ‘‘Analytic Philosophy: What Is It and WhyShould One Engage in It?’’ In The Rise of Analytic Philosophy, editedby Hans-Johann Glock, 1–16. Oxford: Blackwell.

Foucault, Michel. 1986. The Use of Pleasure. Translated by R. Hurley.New York: Vintage Books.

Kuhn, Thomas S. 1970. The Structure of Scientific Revolutions. Secondedition. Chicago: University of Chicago Press.

Mulligan, Kevin. 1991. ‘‘On the History of Continental Philosophy.’’Topoi 10, no. 2: 115–20.

Rorty, Richard. 1982. Consequences of Pragmatism. Minneapolis:University of Minnesota Press.

Williams, Bernard. 1985. Ethics and the Limits of Philosophy. London:Fontana/Collins.

FFF. 1996. ‘‘Contemporary Philosophy: A Second Look’’ In TheBlackwell Companion to Philosophy, edited by Nicholas Bunnin and E.P. Tsui-James, 25–37. Oxford: Blackwell.


NEIL LEVY304

Analytic and Continental Philosophy [Explaining the Differences]

Documents