1 REALISM AND THE REASONS OF THE HEART In a footnote to his seminal article The Natural Ontological Attitudean article

which begins with the sentence Realism is dead Arthur Fine offers a pregnant analogy between realism and religion. In support of realism there seem to be only those reasons of the heart which, as Pascal says, reason does not know. Indeed, I have long felt that belief in realism involves a profound leap of faith, not at all dissimilar from the faith that animates deep religious convictions. I would welcome engagement with realists on this understanding, just as I enjoy conversation on a similar basis with my religious friends. The dialogue will proceed more fruitfully, I think, when the realists finally stop pretending to a rational support for their faith, which they do not have. Then we can all enjoy their intricate and sometimes beautiful philosophical constructions (of, e.g., knowledge, or reference, etc.) even though to us, the nonbelievers, they may seem only wonder-full castles in the air. In various recent articles, I have tried to expand on Fines analogy between realism and religion. I have been suggesting that we see realism as an etiolated version of the religious attempt to bow down before a non-human power to which human beings owe respect. I have also urged that we see the idea that natural science is privileged over other parts of culture as a version of the priestly project of claiming respect from

2 other humans because of ones purported special relation to such a power. As I see it, the great divide in contemporary philosophical thinking is between representationalists who believe that there is an intrinsic nature of non-human reality to be held in human thought and antirepresentationalists who believe that scientific, like moral, progress is a matter of finding ever more effective ways for enriching human life. Representationalists are necessarily realists, since if they did not believe that there was a Way the World Is In Itself they would have no reason to insist that culture is divided between areas where there is a fact of the matter and areas in which there is not. Whereas realists find pathos in the thought of the gap which seperates human thought from its non-human object, we antirerpesenationalists find pathos in thinking of the distance which seperates us from a utopian human futurefrom the world in which our remote descendantsl have developed far better ways of dealing with the non-human environment, presently unimaginable artistic genres, and far kinder and more decent social institutions and customs. On my account of modern times, the pathos of mans seperation from God has been succeeded, among the intellectuals, by these two alaternative forms of pathos. If you do not like the term pathos, the word Romance will do as well. Or one could use Thomas Nagels term the ambition of transcendence. The important point is not the choice of terms but the recognition that both sides in contemporary philosophy are

3 trying to find a suitable replacement for religion. Representationalism and realism are no more or less passtionately irrational than antirepresentationalist humanism. Neither side is ever going to have a knockdown argument, any more than Enlightenment secularism had a knockdown argument against traditional theism. . The heartfelt conviction that there just must be a non-human authority to which we can resort has been, for a very long time, woven into the common sense of the West. It is a conviction common to Socrates and to Luther, to scientists who say they love truth and fundamentalists who say they love Christ. Reweaving the network of shared beliefs and desires which makes up Western culture so as to get rid of this conviction will take centuries, or perhaps millenia. This reweaving, if it ever occurs, will result in an inabilty to share the intuitions which, in our culture, are pumped up by the cosmological argument for the existence of God and by the argument that only correspondence to the intrinsic nature of reality can explain the success of natural science. I think that it is a bit misleading to say, as Fine does in the quotation with which I began, that such arguments as these cannot provide rational support for the view being defended. What counts as rational support, like what counts as valid inference, is a matter of what people are willing are accustomed to take as rational or as valid, and this in turn is what coheres with their intuitions. Those intuitions are determined by the

4 culture in which they have been raised, and there is no way to shortcircuit culture by appealing either to Reason or to Nature. The battle is always between an old culture and one striving to be born. As an example of what I mean, consider the theist who is told that the term God, as used in the conclusion of the cosmological argument is merely a name for our ignorance. Consider the realist who is told that his explanation for the success of science is no better than Molieres doctors explanation of why opium puts people to sleep. Both are, in most cases, equally unfazed. Even if they go so far as to admit that their opponents objection to their favorite argument admits of no refutation, they insist that it produces no conviction. For they feel the need for answers to questions like What came before the Big Bang? and Why does science succeed? So they typically fall back on rhetorical questions such as If not God, what? If not correspondence to reality, what? They think that anyone who does not share the need to answer these questions, anybody who just shrugs them off, is being irrational. Realists think that it is irrational for Fine to shirk the attempt to explain the success of science, just as theists think it irrational of Hume, Kant and Steven Weinberg to speculate about the nature of the First Cause. It is often said that religion was refuted by showing the incoherence of the concept of God or by showing that there is no evidence for the existence of God. It is often said that realism can be refuted by showing

5 the incoherence of the notions of World or correspondence as the realist uses them. Both claims seem to me almost entirely wrong. No one accustomed to throw around a term like the will of God or mindindependent World in complex and sometimes persuasive arguments is ever going to be persuaded that his concepts are incoherent. A concept is just the use of a word, and useful words cannot be denied coherence simply because their users can be forced into tight dialectical corners. Insofar as religion is dying, it is because of the attractions of a humanist culture, not because of flaws internal to the discourse of theists. Insofar as realism is dying, it is because of the attractions of a culture which is more deeply and unreservedly humanist than that offered by the arrogant scientism which is the philosophical position most frequently associated with the ideals of the Enlightenment. For all these reasons, it seems best to stop saying that either religion or realism is unsupported by rational argument, and instead to say that the notion of rational support is not apropros when it comes to proposals to retain, or to abandon, intuitions as deep-lying and long-lasting as those to which theists, realists, and we anti-representationalists appeal. Where argument always seems to fail, as James rightly says in The will to believe, the reasons of the heart should have their way. But this does not mean that the human heart always has the same reasons, asks the same questions, and hopes for the same answers. The gradual growth of

6 secularismthe gradual increase in the number of people who do not find theism what James called a live, momentous and forced option, is testimony to the hearts malleability. Only when the sort of cultural change I envisage is complete will we be able to start doing what Fine suggests--enjoying such beautiful and intricate theistic or realistic constructions as Spinozas Ethics and Kripkes Naming and Necessity as aesthetic spectacles. Only when realism is no longer a live, momentous and forced option for us, only when Dummett can no longer find an audience for his view of the history of philosophy, will we be able to hear questions about the mind-independence of the real as having the quaint charm of questions about the consubstantiality of the Persons of the Trinity. In the sort of culture which I hope our remote descendants may inhabit, the philosophical literature about realism and anti-realism will have been aestheticized in the same way as we have aestheticized the debate between Occamites and Scotists. Dummett capitalizes on a hope which has burned brightly since Platothe hope that we can divide up the culture into the bits where the non-human is encountered and acknowledged, the hard areas, and the softer areas in which we are on our own. This need to divide culture into harder and softer areas is, I think, the most familiar and pervasive contemporary form of the idea that there is something to which human beings are responsible save other human beingssomething like God or

7 Truth or Reality. The idea of a hard area of culture is the idea of an area in which this responsibility is salient. Dummetts suggestion that a lot of philosophy is about the question of whether bivalence obtains for a certain range of sentences amounts to the claim that philosophers have a special responsibility for telling us where the hard ends and the soft begins. A great deal of Fines work is devoted to casting doubt on the need to draw that line. He is the philosopher of science who has done most to deflate the arrogance embodied in Quines quip that philosophy of science is philosophy enough. Much of what he has written gears in nicely with the writings of two other contemporary philosophersDonald Davidson and Robert Brandomwho are trying to put all true sentences on a referential par, and thereby to erase the line between the hard and the soft. Fine, Davidson and Brandom have helped us understand how to stop thinking of intellectual progress as a matter of increasing tightness of fit with the non-human world and how to picture it instead as our being forced by that world to reweave our networks of belief and desire in ways that make us better able to cope. Humanism, in the sense I am using the term, will only triumph when we no longer discard the question Do I know the real object, or only one of its appearances? and replace it with the question Am I using the best possible description of the situation in which I find myself, or can I find a better one?

I see Fines NOA papers as fitting in nicely with Davidsons claim that we can make no good use of the notion of mind-independent reality and with Brandoms Sellarsian attempt to interpret linguistic meanings as a matter of the rights and responsbilities of participants in a social practice. The writings of these three philosophers blend together, in my imagination, to form a sort of manifesto for the kind of antirepresentationalist movement in philosophy whose aspirations I have just outlined. Occasionally, however, I come across passages, or lines of thought, in Fines work, which are obstacles to my syncretic efforts. There are two sets of passages in his 1984 paper (The Natural Ontological Attitude) which arouse doubts; passages about reference and passages about method. An example of the first reads as follows: When NOA counsels us to accept the results of science as true, I take it that we are to treat truth in the usual referential way, so that a sentence (or statement) is true just in case the entities referred to stand in the referred-to relations. Thus NOA sanctions ordinary referential semantics and commits us, via truth, to the existence of the individuals, properties, relations, processes, and so forth referred to by the scientific statements that we accept as true. (p. 130)

9 Reading this passage leaves me uncertain of whether, like Davidson, Fine wants to read all the sentences we accept as truethe ones accepted after reading literary critics as well as after reading scientific textbooksas true just in case the entities referred to stand in the referred-to relations Davidson thinks that the sentence Perserverance makes honor bright is true in this way, just as much as F=MA. But Davidson thinks this because he does not think that reference has anything to do with ontological commitment, and indeed has no use for the latter. Fine does seem to have a use for this notions, and I suspect he drags in ordinary referential semantics because he thinks that reference and ontological commitment have something to do with each other. I think it would be more harmonious with the overall drift of his thinking to mock that Quinean idea rather than to try to rehabilitate it. NOA, Fine says, tries to let science speak fo itself, and it trusts in our native ability to get the message without having to rely on metaphysical or epistemological hearing aids. (And not, p. 63) So why drag in a semiotic hearing aid such as ordinary referential semantics? If, as Fine recommends, we stop trying to conceive of truth as a substantial something, a something that will then act as limit for legitimate human aspirations (And not, p. 56) , as a goal of inquiry, should we still say that we are committed, via truth, to the existence of this or that? Whey

10 should we conceive of existence as something which we sometimes commit to and sometimes refrain from committing to? If we give up the notion that we are trying to correspond to reality, as he suggests we do, will we still ask ourselves questions like Am I committed to the existence of X? As support for my suggestion that the notion of ontological commitment is one Fine could get along nicely without, let me cite another of his instructive remarks about the analogy between religion and realism. Fines answer to the question Do you believe in X?, for such Xs as electrons and dinosaurs and DNA, is I take the question of belief to be whether to accept the entities or instead to question the science that backs them up. (Afterword, p. 184) Then, in response to the objection But does not believe in mean that they really and truly exist out there in the world? Fine says that he is not sure it does. He points out that those who believe in the existence of God do not think that is the meaning, at least not in any ordinary sense of really and truly out there in the world. I take the point of the analogy to be that people who talk about God as the unquestioningly and unphilosphically religious do not need to distinguish between believing in God and talking the way they talk. To say that they believe in God and that they talk the talk are two ways of describing the same phenomenon. Similarly, Fine is saying, for a physicist to say that to say that she believes in electrons and to say that she does

11 not question the science behind electron-talk are two ways of describing the same situation. When Kant or Tillich ask theists whether they are not really talking about a regulative ideal or a symbol of ultimate concern, rather than something that they think really exists, they are justifiably irritated. Physicists should be equally irritated when asked whether they think that statements about electrons are true or merely empirically adequate. The theist sees no reason to resort to demonstrations of existence, or analyses of the meaning of is when applied to God for he talks God-talk into his life as true in exactly the manner, in her capacity as physicist, she talks electron-talk into her life as true. It accords with the overall humanist position I outlined in the first section of this paper to say there are no acts called assent or commitment which we can perform that will put us in a different relation to an object than simply talking about that object in sentences whose truth we have taken into our lives. To use a word in a certain way and to believe in the existence of the referent of that word is the same phenomenon. The idea that we might be in a better position to figure out what to believe by first finding out what there really and truly isby finding the right ontology--is an impossible attempt to separate thought and language from action. Looked at another way, this idea epitomizes a confusion between existential commitment and a profession of satisfaction with a way of

12 speaking or a social practice. An existential commitment, as Brandom nicely says in MAKING IT EXPLICIT, is a claim to be able to provide an address for a certain singular term within the structured space provide mapped out by certain canonical designators. (See MIE, pp. 444ff.) To deny the existence of Pegasus, for example, is to deny that a continuous spatiotemporal trejectory can be traced out connecting the region of space-time occupied by the speaker to one occupied by Pegasus. To deny that Sherlock Holmes fairy godmother exists is to deny that she can be related to Conan Doyles text in the way that Moriarity and Mycroft can. And so on for other structured spaces, such as those of mathematics. The point of putting the matter Brandoms way is to make clear that metaphysical discourse, the discourse of ontological commitment, does not provide us with a such a structured space. It is, instead, a discourse in which we express our like or dislike, our patience or impatience with, various linguistic practices. We do so by pretending that our decision about whether to engage in those practices can be based upon reflection on the desirability of certain ontological commitments. But this is fantasy. We talk first, and commitment falls out of the talk. Similarly, as Davidson points out, we talk first, and reference falls out of the attempt to interpret what we are saying. Fine and Davidson seem to me entirely right in saying that we need to stop the pendulum swinging back and forth between an analysis of trut

13 of correspondence and an analysis of it as acceptance. I entirely agree with Fines claim that we should be neither realists nor antirealists, and should acknowledge that the concept of truth cannot be explained or given an account of without circularity (And not, p. 62). But there seems to me at least the potential for a division between Find and Davidson when it comes to reference. If one takes Davidsons line on reference, one will not find it natural to hook it up with the notion of ontological commitments or attitudes. One will lots of sentential attitudes, but no ontological ones. Davidson urges that we not treat reference as a concept to be given an independent analysis or interpretation in terms of non-linguistic concepts. (Inquiries, p. 219) Rather, reference is, he says, a posit we need to implement a theory of truth (bid., p. 222) For Davidson, a theory of truth for a natural language does not explain reference, at least in this sense: it assigns no empirical content directly to relations between names or predicates and objects. These relations are given a content indirectly when the T-sentences are. (ibid, p. 223) If one assumes that a theory which permits the deduction of all the T-sentences is all we need in the way of what Fine calls ordinary referential semantics, then the notion of reference becomes hard to tie up with that of ontological commitment. It seems equally unnatural to ask whether I am committed either to the existence of jejuneness or of magnetic monopoles. The notion of

14 ontological attitude, or ontological commitment, seems to have no function in the life of someone who takes the results of both physics and literary criticism in the same way as we accept the evidence of our senses (to use Fines phrase). Perhaps, however, Fine would agree both with Davidson about the nature of the notion of reference and with me about the need to treat literary criticism and physics as producing truth, and reference, of exactly the same sort. That he would is suggested by his saying that those who accept NOA are being asked not to distinguish between kinds of truth or modes of existence or the like, but only among truths themselves in terms of centrality, degrees of belief, and the like. (NOA, 127) This last quotation chimes with Fines remark that NOA is basically at odds with the temperament that looks for definite boundaries demarcating science from pseudo-science, or that is inclined to award the title scientific like a blue ribbon on a prize goat. (And not, p. 62) It chimes also with the last paragraph of his recent Presidential Address to the APA, in which he says that the first false step in this whole area is the notion that science is special and that scientific thinking is unlike any other. (Viewpoint, p. 19) If Fine would carry through on these remarks by saying that there is no more point in using notions like reference and ontological attitude in connection with physics than in connection with literary criticism, then he and I might agree that nobody should ever

15 bother to have an ontology. Nobody should have the sort of worries Quine had, about having more things in his ontology than there are in heaven and earth. To give up on the project of dividing culture into the hard and the soft areas would lead one to give up on the project of listing the things that there are in heaven and earth, a list presumed to be shorter than the list of expressions in our language which can be nominalized and thereby reified. To take this line would mean that Einstein would not have been worrying about a serious question when he, as Fine puts it, wanted to claim genuine reality for the cnetral theoretical entities of the general theory, the four-dimensional space-time manifold and associated tensor fields. Fine goes on to say that had Eintsteins claim been right then space and time cease to be real. But if Fines ultimate view is what I think it ought to be, he should hesitate to say that the war between Einstein, the realist, and Bohr, the nonrealist, was not just a sideshow in physics, nor an idle intellectual exercise. He should say rather that an idle intellectual exercise was the outward and visible form of a serious discussion about what young physicists should and should not spend their time looking for. Nor, if Fines view were what I would like it to be, would he himself express the hope that quantum theory is at least consistent with some kind of underlying reality. He would translate this ontologicalsounding remark into some sort of suggestion about what lines of research

16 physicists will find profitable and which not. He would, in short, translate remarks about what non-human things there really are into remarks about what human beings should do to improve the human future. One more example of a passage which makes me wonder whether Fines view coincides with my own can be be found in the Afterword to THE SHAKY GAME. There Fine says that It is no possible to have a global characterization of scientific productsneither as constructions, nor as externally real things, nor as generally reliable models. If science is genuinely open, we need to go local and particularist. We need to look at each case and see what there is to say about the character of scientific products and representations and whether any general characterization is needed at all. (p. 188) On the view I am suggesting, there is no need for a local characterization either. For once we have decided that a scientific product (a set of statements, a new device, a new hormone, a new vocabulary) is useful for this and useless for that, and once we know how to get, or avoid getting, similar products by steering students into this disciplinary matrix rather than that, it is not clear what else we need to do. It is not clear why we should want any further characterization, for the only characterizations which might come to mind are likely to be variations on the worn-out themes of hardness and softness: characterizations such as

17 made found, instrumental and real. So although I applaud the spirit of Fines remark that the realists determinate, external world is just one more social construction, I balk at the letter. The realists conception of inquiry, like the theists conception of mans position in the universe, is no more or less constructed than is the human self-image which will, I hope, dominate the culture of our remote descendantsa self-image in which non-human authority plays no role, and in which the notion of love of truth has become interchangeable with that of love of conversation. I would rather say that we shall only escape the need for a hardsoft distinction when we have abjured the made-found distinction, and that therefore never start asking, about some given scientific product, what our share and what natures share in its production has been. This is why I am dubious about the Fines claim that Constructivism is a useful antidote to realism; its attention to science in action deepens our understanding of the social in science and the myriad ways in which science is open (p. 188). It seems to me that one can follow Latour around on his tours of laboratory life without adopting anything remotely like constructivism. I take the moral of Davidsons doctrine of triangulation to be that everything we can ever speak of is a much constructed, and as much real, as anything else. If that doctrine is right, we do not need Latours notion of quasi-object, because there are no objects which are not quasi..

18 Before leaving the topics of reference and ontological commimtment, let me remark that the passage I quoted about ordinary referential semantics has been seized upon by Alan Musgrave tp ridicule Fines claim to have a position distinct from that of the realist. Musgrave would have had less ammunition, I think, if Fine had not only omitted this passage but had been more explicit in admitting that NOA, as Jared Leplin has lately said, is not an alternative to realism and antirealism, but a preemption of philosophy altogether, at least at the metalevel. (Leplin, p. 174) Leplin is right, in my opinion, to say that Fines idea that scientific theories speak for themselves, that one can read off of them the answers to all legitimate philosophical questions abouit science, cannot be squared with the rich tradition of philosophical debate among scientists over the proper interpretation of theories. So I think that the Fine should not attempt to take the Einstein-Bohr debate at face value, nor to rehabilitate notions like ontological commitment. He should grant to Leplin that Philosophy of science in the role of interpreter and evaluator of the scientific enterprise, and realism in particular, as such a philosophy of science, are superfluous. (Leplin, p. 139) He should say that we felt the need for such an interpreter, evaluator, and public-relations man only so long as we thought of natural science as privileged by a special relation to non-human reality, and of scientists as the priests of the modern age. .

19

I said earlier that there is a second set of passages in Fines writings which made me wonder if I am really entitled to include him in my syncretic projects. These are the passages in which he uses the term method. Sometimes Fine seems to take this notion more seriously than I think it deserves. At the beginning of The Natural Ontological Attitude Fine says that Boyd and Putnam have focused on on the methods embodied in scientific practice, methods teased out in ways that seem to me accurate and preceptive about ongoing science, methods that lead to scientific success. (NOA, p. 113) Here Fine speaks in the same idiom as Philip Kitcher who, in his book THE ADVANCEMENT OF SCIENCE, undertakes to to defend realism on the basis of an examination of what he calls processes that are well-designed for promoting cognitive progress (Kitcher, p. 192) Kitchers argument for the residual truth of rationalism in the philosophy of science is based on the claim that we can isolate such processes and attribute the achievement of cognitive progress to their use. To make his case, he would have to find a middle ground between the substantive explanations of events put forward by scientists, and the disciplinary matrices which shape themselves around acceptance of those explanations, on the one hand, and uttterly general and completely uncontroversial methodological platittudes on the other. I do not think

20 that he, or any other writer on scientific method I have read, succeeds in this task. When Kitcher commends, for example, elimination of alternatives through the production of inconsistency, forced retreats that open problems, and employment of background constraints (Kitcher, p. 288), he is not recommending anything that deserves to be called a method or a cognitive process. He is describing an activity which all human beings perform every day. When, on the other hand, Kitchers examples of superior cognitive strategies are more specific, theya re simply summaries of the arguments by means of which some victorious scientific theory triumphed. Kitcher fails to give us an interesting difference the difference between a process of reasoning and and an explanatory schema, between a method a good idea. I suspect that Fine became more dubious about the notion of method, and less sympathetic to such efforts as Kitchers, in the years that separate his 1984 article from his Presidential Address to the APA. For in the latter we find him making fun of the idea that There are universal principles governing the procedures that make for objectivity, principles whose absence would lead to something that involves relativisim and irrationalism. (Viewpoint, p. 13) His attitude in this paper chimes with that in the Afterword, where he says that NOA is an attitude of trust in two senses, the first of which he describes as follows:

21 NOA trusts that scientists are serious people trying to do good work and that scientific procedures are reliable ways of conducting that serious enterprise. (the contrast here is with antiscience attitudes, ranging from general skepticism to the derisive contempt toward science that one sometimes encounters in the humanities.) If one trusts that scientists are serious and reliable, then one will not be inclined to ask what methods they employ. One will not try to find something intermediate between their practices and their successes something which can be called a cognitive strategy or a process of reasoning, and evaluated for reliability in a general, methodological, way. So one will not distinguish between epistemic and nonepistemic or rational and irrational factors in scientific research, any more than between artistic and non-artistic or creative and noncreative factors in artistic production. One will settle for saying that we call a scientists results true, and instances of cognitive progress, when we think they compare well with those of her competitors, just as we praise an artists productions when they compare well with his. The same goes when one regards various other groups of people as serious and reliable: for example, those politicians, literary critics, lawyers, theologians, philosophers, hairdressers, carpenters, bird-watchers, astrologers or engineers whom one has gotten in the habit of trusting. One may have no use for, and therefore neither trust nor distrust

22 concerning, one or another of these groups of experts. One may feel derisive contempt for such a group if, having once been tempted to rely on their results, one was severely disappointed. But ones attitude toward any such group will not be determined by any methodological critique of their procedures of justification. If I attribute a blissfully happy marriage, as well as spectacular financial and professional success, to faithfully following my favorite astrologers advice, I will not ask for an account of how his procedures fit in with the nature of things, any more than I ask how the expert bird-watcher can tell the Great Yellowlegs from the Lesser from a hundred yards away. I shall trust them both, and no mere philosopherand certainly no specialist in comparative methodologist--is going to make me dubious. My trust is determined by their past utility, not by ones grasp of the procedures of justification which are employed within the group. As I see it, the notion of method is a residue of the idea that natural scientists, unlike literary critics, politicians and chicken-sexers, have found a way to cut through appearance to reality. Viewed etymologically, a method is a road that takes one from subject to object. If one stops thinking of knowledge as a relation between mind or language and an object, and instead thinks of it as the grasp of more true sentences, then the notion of such a road becomes less plausible. Instead of thinking of how we can get access to a kind of object, we view the

23 object as whatever it is we are trying to talk about. If we talk about it, we are never without access to it, and our attempt to say more useful things about it is not a matter of getting closer to it or getting a clearer view of it or fitting it better. I think that the idea of scientific method and the idea that such disciplines as philosophy or political science or law should become more scientific are symptoms of our having been held capture by the picture of an abyss which seperates human minds from non-human reality If we instead get our pathos, romance, and ambition of transcendence by contemplating the gap between ourselves and a utopian human future, we should stop lamenting our lack of method and scientificity, and also stop trying to reveal the natural scientists secrets of success. If it were not for the public relations role of philosophy of science--a role which was necessary in the seventeenth and eighteenth centuries, but which should have doffed in the course of the nineteenthI doubt that we should ever have been told that geologists, astrophysicists, evolutionary biologists, and botanists all used the same method. We should not have thought that there was a natural kind called natural science which had a subject and method of its own. We should have assumed that one became a good scientist by going into somebodys lab or seminar room and getting the hang of it. We should have found it natural that no scientist would want his apprentices taking time off to

24 study Cohen and Nagels INTRODUCTION TO LOGIC AND SCIENTIFIC METHOD, or any other work which advised them about what cognitive processes to use. In arguing, as I have in the past, that natural science is not a natural kind, I am of course speaking out of dense ignorance. All I am entitled to say is that I have not yet encountered a work of philosophy of science which gave me reason to think that geology, astrophysics and the rest are any more closely linked to one another than to medicine, law and engineering. It is always possible that Paul Gross is right when he says that people who have not practiced science cannot understand the nature of science. Perhaps Daniel Dennett is right that what he calls flatfooted ignorance of the proven methods of scientific truth-seeking and their power is responsible for my holding the philosophical views that I do, and for my lack of what he calls the intellectual leverage provided by scientists faith in truth. Dennett claims that on the view I espouse it is all just conversations, and [that] there are only political or historical or aesthetic grounds for taking one role or another in an ongoing conversation I do indeed want to substitute the notion of conversation for that of method. Agreeing as I do with Brandom that conversation is the highest good for discursive creatures, I think that to take conversation seriously to trust, at least initially, that ones interlocutors are serious and reliable

25 inquirersis the best way to display the virtue Dennett calls love of truth. Dennett thinks my writings discourage people from acquiring this virtue. But he is wrong to suggest that, on my view, only such traditionally soft grounds as might be labeled political or historical or aesthetic are available. I think that the obvious grounds for being on one side or another of a conversation about choosing between scientific theoires are specifically scientific grounds. But I do not think that one can explicate specifically scientific in any way that accords with the traditional project of philosophy of science. One can identify a scientific ground for taking a conversational side only by immersing oneself in the disciplinary matrix within which a particular conversation is going on, and by making the current state of that conversation part of your life. To love truth in a given area of culture is, ninety-nine times out of a hundred, simply to immerse oneself in one or another conversation among people one trusts as serious and reliable, and then do ones best to meet the expectations of ones interlocutors. In the hundreth case, of course, the love of truth may lead one to break out of the disciplinary matrix within which that conversation has been going on, and to try to set up a new one. But I cannot see that either fidelity to disciplinary expectations or the courage to effect a break-out can be correlated with ones view on the philosophical questions over which Dennett and I, or Fine and Leplin, disagree.

26 Dennett, and many other critics of my work, seem to think that scientists and desirably hard-minded philosophers argue, whereas literary critics, bird-watchers, astrologers, and politicians simply chat. But astrologers, like the the Oxbridge and Sorbonne Aristotelians of the seventeenth century. do argue. One should grant them all the clarity and rigor they claim, even if one agrees with Bacon and Descartes that the topics they argue about are not worth discussing. The argument-chat distinction is perfectly real, and most of us can tell which we are doing when. But it does not map onto the traditional hard-soft distinction. Argument between Heidegger and Sartre, or between T. S. Eliot and Harold Bloom, are just as unchatty and just as serious as arguments about Tyler Burge and Donald Davidson, or between Einstein and Bohr. Dennett says that we have created a technology of truth and that its name is science, I would respond that its name is high culture. This is the area of culture which contains those disciplinary matrices in which power and money matter less than achieving free agreement on the answers to questions unintelligible to the vulgar. Philosophy is one prominent part of high culture, so defined. One of the differences between Dennett and myself is about whether Kant, Husserl, Russell on others who wanted to put philosophy on the secure path of a science improved or damaged the quality of philosophical conversation. I think that these effects were harmful, and in my utopian human future they

27 would cease to be made. More specifically, I think that the question What should the method of philosophy be? should cease to be asked. Just as the way to become a good scientist is to go into somebodys laboratory and try to get in the swing of things, so the way to become a good philosopher is to throw oneself into the works of some well-reputed philosopher and try to talk his or her talk. This will work no matter whether the good philosopher in question is Kant or Hegel, Heidegger or Davidson, Nietzsche or Mill. You get into the philospohical conversation by discussing what such people discussed, and in particular by answering or being persuaded by their critics How good a philosopher you get to be as a result of throwing yourself into one philosophical conversation rather than another is not determined by your initial choice, but of how conversable you are in later life: how well you can fuse your own conversational horizons with those of lots of other philosophers, philosophers trained in different matrices and different countries. In particular, there is no point in adivising young people, looking for a philosopher whom they can heroize and imitate, that they should attend to the methods being used. If I were told that Davidson uses a good method, and Heidegger a bad one, or the converse, I would be baffled. They seem to me to be distinguished not by methods but by contexts of discussion, and further distinguished by the fact that they, but not most of the other membesr of their respective cohorts, happened to

28 come up with some brilliantly original ideasideas which wer no more the result of applying a method than were Einsteins or Bohrs. .The terms analytic method and phenomenological method strike me as of no use. I would not know how to teach either method to a student. Neither notion helps explains what differentiates Davidson from all the secondrate hacks who also call themselves analytic philosophers, or Heidegger from all those slavish disciples of Husserl who never had the guts to break out on their own. Davidson entered philosophy by getting in on a conversation between Quine and Carnap. Heidegger entered the field by imagining a conversation between the Marburg neo-Kantians and such figures as Kierkegaard and Nietzsche. Both men turned out to have some original ideas. The search for a criterion which would make one of them a good clear, rigorous philosopher and the other a bad, unclear, unrigorous philosopher seems to me pointless,. Such a search, once undertaken will always become a public relations exercise on behalf of ones favored candidate. The notions of clarity and rigor, like that of rational support, are determined by cultural expectations. Differing expectations about what to get out of physics accounted for the fact that, as Kuhn recounts, it took a hundred years to domestic a nonAristotelian mechanics. Differing expectations about what to get out of philosophy typically take even longer to be resolved.

29 . Leplin has one set of expectations about what counts as philosophy of science, Fine another. Fine is, as Leplin suggests, staging a break-out from a familiar disciplinary matrix. So, in less obvious ways, are Brandom and Davidson; if there work were taken to heart, huge piles of books and articles in the philosophy of language would be thought to contain nothing but clever answers to pointless questions. The decisions participants in the relevant conversations come to about whether to follow these three men into a non-representationalist philosophical utopia, will not be made on what Dennett calls political or aesthetic or historical grounds. They will be made on philosohical grounds, in the sense that they will be the outcome of conversations in which only people who have read quite a lot of philosophy will be able to take part. But they will not be made by following a good or a bad philosophical method, any more than you and I decide whom to vote for, whom to propose marriage to, or where to go to school by following a good or a bad method. They will be made in the way human beings have always made such decisions when faced with a choice between two sets of people, both made up serious, reliable, trustworthy interlocutors. When the trustworthy disagree, we are on our own. But this is not to say that are choices become irrational or aesthetic or political. They remain conversational.

30 In the areas of culture traditionally called hard, most of the participants in the conversation eventually agree about what decision was best. In other areas they do not. But this is not to be explained by the existence of contact with the non-human in the former areas and lack of such contacts in the latter. It is to be expalined by the fact that in the former there is more agreement about what kind of product a given disciplinary matrix is expected to produce. The sources of such agreement are to the province of historians and sociologists, not of epistemologists or ontologists.

Richard Rorty May 3, 1999