Richard Rorty’s recent book Philosophy and the Mirror of Nature* is an original and sustained attack on the idea that it can be the aim of philosophy, or even of science, to represent the world accurately. Neither activity can reveal, as he sometimes puts it, a vocabulary in which the world demands to be described. The book is remarkable for its learning and for its powers of critical exposition. At the same time, some of it is slapdash, and its program for what philosophy should do when robbed of its traditional conceptions of truth and objectivity is, to put it mildly, schematic.

The present volume consists of twelve already published essays written between 1972 and 1980, together with a new introduction. The jacket says that it fills in the details of the story told in Philosophy and the Mirror of Nature, but it actually does something more interesting than that. It reveals Rorty’s attitude toward questions bearing on the central theories of the earlier book, and offers his view of other philosophers and traditions, including Heidegger, about whom he says that he would now want to revise his view upward. The essays here also have a hero, who was less explicitly seen as one in The Mirror of Nature. This is John Dewey; and the pragmatism of the title is above all that of Dewey.

Rorty claims to free Dewey from dated associations, and to find him already waiting at the end of a road on which Michel Foucault and Jacques Derrida are now traveling. Dewey’s “chief enemy,” Rorty writes, “was the notion of Truth as accuracy of representation, the notion later to be attacked by Heidegger, Sartre, Foucault. Dewey thought that if he could break down this notion, … we would be receptive to notions like Derrida’s—that language is not a device for representing reality but a reality in which we live and move.”

The new book shares some failings with The Mirror of Nature, and at the end we still do not know much about how philosophy should go on without its old illusions. Rorty’s style also provides some minor irritations, such as his tendency to parade lists of great names and of turning points in the history of philosophy (something that could be a legacy from his early days at the University of Chicago). But the essays are wide-ranging, informed, and above all interesting. Rorty has an unsettling vision of philosophy, science, and culture, and it matters to what extent he is right. Like others who have a large view, he sometimes seems to the analytical critic to have run different questions together. What is not always true in such cases, but is usually true of Rorty, is that when he has run different questions together each of them turns out to have its own interest.

“It is impossible to step outside our skins—the traditions, linguistic and other, within which we do our thinking and self-criticism—and compare ourselves with something absolute.” That is one of Rorty’s central these. Or, rather, it is several theses. The least contentious is that we cannot think about the world without describing it in some way: the world cannot present itself uncategorized. Moreover, there is no way in which the world simply describes itself, or presents itself in terms that could not themselves be the subject of inquiry, reflection, and alternative proposals. Those claims, in themselves, are not too upsetting. They still allow us to think that there is an independent world that we are trying to describe, and that what it is actually like can control the success of our descriptions.

Rorty’s pragmatist, however, reaches much more drastic conclusions than this, and claims (so it seems) that all we can ever do is compare one description with another. He denies that “deep down beneath all the texts, there is something which is not just one more text but that to which various texts are trying to be ‘adequate.”‘ He does not think that we can say anything substantial about the purposes served by our descriptions, against which we might test them. Moreover, in addition to this, Rorty has a further, historicist thesis, according to which the categories that any human group uses are a function of its time, and are essentially formed through historically localized tradition. The historicist thesis plays a large part both in The Mirror of Nature and in the present essays.

If one says that any human thought is inescapably immersed in the traditions of its period, what counts as “a period” is an important question; and, in particular, what tradition performs this basic function for us. Rorty is not very definite about this. In The Mirror of Nature sometimes it is the period of “Western man” or “modern Western man.” In the present book the relevant item, at least once, is “human thought since 1600.” The question particularly presses, because Rorty is so insistent that we cannot, in philosophy, simply be talking about human beings, as opposed to human beings at a given time. In the course of a perceptive discussion of Thomas Nagel and Stanley Cavell, both philosophers who (in different terms) hope to recover from the tradition deep philosophical questions that relate to human experience as such, Rorty precisely contrásts the approach of taking some philosophical problem and asking, as they might, “What does it show us about being human?” and asking, on the other hand, “What does the persistence of such problems show us about being twentieth-century Europeans?” (his emphases).


The historicist ideas do provide one fairly natural way of interpreting Rorty’s main thesis, but they do not merely follow from it, even in its most radical form. Basically, he accepts the historicist outlook because he believes that the history of philosophy has itself led us to it. He thinks that Dewey, Quine, Wittgenstein, Heidegger, and Derrida are the true descendants in their various styles of Hegel and the nineteenth-century philosophers who reacted against the Kantian claim that philosophy could discover “the a priori structure of any possible inquiry. …” As he interprets them, they have led us to see that there is not much more to be said about the ways in which we describe the world than that they are the ways that suit us, now. Of course philosophy, traditionally, has tried to say more than that. It has tried to overcome what it has seen as deep and persistent problems about the relation of our thought and action to the world. For Rorty these writers have, accordingly, led us to a point at which traditional philosophy should end.

Sometimes Rorty takes a slightly different turn in his insistence on historical self-consciousness and in his rejection of general groundings for his or any other method. That we should see philosophy and other intellectual activities in the way he commends is not a lesson of where we have come to in history—something that we should rationally conclude from it—but simply a product of that history. We are where we now are, and that is how we, now, go on.

Rorty is not a relativist. He has as crisp a view as any positivist in agreeing, for instance, that it was a good thing that the world that was based on religious conceptions and authority has passed, and he cheerfully describes a certain attitude as “merely a relic of pre-Galilean anthropomorphism.” But is he really in a position to dismiss relativism and the problems associated with it, as he does in one essay? The sort of dialectic in which Rorty’s self-conscious historicism places him is one in which everyone can try to undercut everyone else by asking others whether they have allowed for the ways by which their own consciousness has evolved the very thesis they are advancing. Self-consciousness and reflective awareness, when made into the distinctive attitude of a sophisticated philosophy, make it revolve ever faster; the owl of Minerva, robbed by later skepticism of Hegel’s flight plan to the transcendental standpoint, notoriously finds itself flying in ever-decreasing circles.

Rorty’s procedures, in these respects, are an odd mixture. Sometimes he seems quite knowing about the status of his own thoughts (though he is not as quick on the turn as the French writers, such as Derrida, whom he most admires, or as the poststructuralist critics are, who need these reflexes to keep alive). At other times, he seems to forget altogether about one requirement of self-consciousness, and like the old philosophies he is attempting to escape, naively treats his own discourse as standing quite outside the general philosophical situation he is describing. He thus neglects the question whether one could accept his account of various intellectual activities, and still continue to practice them.

Some of the nastier problems of this sort arise with his treatment of the natural and biological sciences. Rorty’s characteristic tone about science is that there is nothing in the least special or particularly interesting about it.

Pragmatism … does not erect Science as an idol to fill the place once held by God. It views science as one genre of literature—or, put the other way around, literature and the arts as inquiries on the same footing as scientific inquiries. … Physics is a way of trying to cope with various bits of the universe; ethics is a matter of trying to cope with other bits.

In a similar vein he says, in an article called “Method, Social Science, and Social Hope,” that it simply turned out that the Galilean picture of the universe worked better than, say, an Aristotelian picture, but that there is no “epistemological moral” to be drawn from this. In particular, he argues, it is a confusion to think that the success of physics since Galileo is somehow connected with the fact that it regards the universe as “infinite and cold and comfortless,” and it is a mistake to look for any scientific method that explains scientific success. Indeed, the question “What makes science so successful?” is for him a bad question. He applauds T.S. Kuhn’s notion of Galilean science “as exemplifying the power of new vocabularies rather than offering the secret of scientific success.”


In the essay just referred to, and to a lesser degree in The Mirror of Nature, Rorty runs together two questions. One is whether the success of science invites or permits any interesting description of what the success of science consists of. The other question is whether, from its previous success, we can derive any general methods to secure its future success. The questions are distinct. Karl Popper, for instance, who, like Rorty, thinks that there is not much to be said about the second question beyond banal recipes of rational procedure, also believes, unlike Rorty, that there is something to be said about the objective progress of science in finding out what the world is really like.

It is harder than Rorty supposes to throw away conceptions of the aim of science such as Popper’s, and it is harder in more than one way. It is harder, first of all, because it is not clear what Rorty wants us to put in their place. Science “copes,” “is successful,” its vocabularies have “power”; but they have power or success in doing what? In generating predictions, Rorty is sometimes rash enough to say, and that means better predictions. Here we find we are being taken on an old-fashioned philosophical ride. Doesn’t “better” mean, for instance, “true”? On Rorty’s view there is no point in getting off at that stop: “Truth is simply a compliment paid to sentences seen to be paying their way.” But what is it that we see when we see that they are paying their way?

This is a very old subject of debate, and it is still going on in orthodox analytical philosophy. Rorty’s pragmatist does not want to win that debate or to continue it, but rather to opt out of the whole thing, to change the subject. Some analytical philosophers will say that he can’t do that. But Rorty is surely right in saying that much philosophical achievement has consisted simply in changing the subject, and if the pragmatist changes this subject, he changes it.

To me the weakness of Rorty’s position lies in something else, that he sees all this as a matter simply for philosophy; he sees changing the subject as making a move within, or out of, philosophy. This seriously neglects the extent to which the descriptions that he dislikes come from within science itself. Science itself moves the boundaries of explanation and of what is explained, just as it moves the boundary of what counts as observation. It was always a mistake for philosophers to contrast in any absolute way the “observable” and the “theoretical” in science, since theory creates and constitutes new forms of observation. Scientific theory explains, moreover, how such an elaborately constructed image as an electron micrograph can be the record of an observation.

The sharp distinction between theory and observation was a mistake made by Rorty’s enemies, the positivists, who celebrated science for its respect for brute fact. But such criticisms of the positivists turn against Rorty himself, because they are an example of something which, it seems, he should regard as impossible, namely of science explaining the reliability of its own observations. Similarly science can often explain the truth of its conjectures. Advances in scientific theory quite often, in fact, involve explaining why some predictions of previous theories were true, while others failed. Not all scientific advance does that—no recipe fits all scientific advance—but it is one important phenomenon that gives substance to the idea of objective scientific advance.

More generally, it is an important feature of modern science, not mentioned by Rorty, that it makes some contribution to explaining how science itself is possible, and how creatures that have the origins and characteristics it says we have can understand a world that has the properties it says the world has. To say that such achievements as evolutionary biology and the findings of the neurological sciences, for example, are trivial, and that any old theory could do what they do, is simply a mistake (though it is true that limitless numbers of theories could deal with the same questions trivially). These ideas contribute, from within scientific reflection itself, to an image of the objects of science which Rorty says we should not have; they contribute, that is, to a conception of the world as it is, independently of our inquiries. That conception may be an illusion, but if it is, it is not the product of a simple philosophical error to be explained in a line or two of reference to Kant and his successors; and, above all, it is not simply a product of philosophy.

Correspondingly, it is not just a question of philosophy whether it is hard to give up that conception of the world. The other sense in which it will prove hard to give up is one in which it will be hard to give it up even if it is an illusion. It will be hard to give up for those working in science.

There is an important contrast here, which Rorty seems not to see, between scientific inquiry on the one hand and Rorty’s interesting ideas about the future of philosophy. In a revealing passage he says that “pragmatism denies the possibility of getting beyond the … notion of ‘seeing how things hang together’—which, for the bookish intellectual of recent times, means seeing how all the various vocabularies of all the various epochs and cultures hang together.” That may be a program for the successor of philosophy, or for the literary studies from which he does not want that successor to be distinct, but it is certainly no program for science. The sense that one is not locked in a world of books, that one is confronting “the world,” that the work is made hard or easy by what is actually there—these are part of the driving force, the essential consciousness of science; and even if Rorty’s descriptions of what science really is are true, they are not going to be accepted into that consciousness without altering it in important ways—almost certainly for the worse, so far as the progress of science is concerned.

But if that is so, then a dreadful problem confronts the pragmatist: whether his ideas can be, in their own terms, “true” at all: For the pragmatist to say that his formulations are true presumably means simply that they work out: and what reasons have we to think that the pragmatists’ sentences about science will work out better in the practice of scientists than scientists’ sentences do? The point here is not that scientists have self-revealing knowledge of what they are up to, but merely that the scientists’ sentences help to keep them going—and that, for the pragmatist, is all that can matter.

Indeed, there is a question, whether the pragmatist can even appropriately say many of the things that Rorty says. Here there is a problem that was seen more clearly by Wittgenstein than by any of the other philosophers whom Rorty admires, certainly more clearly than by Rorty himself. If it is impossible to provide grounds for, or get beyond, what, at a very general level, we naturally say; and if philosophy, as traditionally understood, tried to go beyond that, and so should now end; why should it not simply end, so that all we should say is what anyway we naturally say? In The Mirror of Nature there are passages to the effect that we have merely found it overwhelmingly “convenient” to say that physics describes a world which is already there, rather than, for instance, that the world changes in relation to our descriptions. But if that is overwhelmingly convenient, and the only consideration can be what is convenient, then what everyone should be saying is simply: physics describes a world that is already there. So why does Rorty go on telling us not to say that?

Here the Rortian pragmatist, like the follower of Wittgenstein, is likely to say something to the effect that without the startling reminders he provides one may be misled, and succumb to false images of our situation. Misled by what? The answer often is—by philosophy, or by similarly irresponsible kinds of discourse. Wittgenstein often gives this answer (though he also gives the materials for some better ones): it is what underlies his famous remark that philosophy occurs “when language goes on holiday,” a remark which, one might say, is, like some others of his, deeply shallow. In fact, the “misleading” impressions are encouraged not just by philosophy but by such activities as pursuing physics. So unless science itself is revealed as an unnatural or holiday activity, it is part of our nature, and not simply a product of philosophy, that we should be “misled.”

But then there is a real problem of what content is left, on the pragmatist’s assumptions, to saying that we are misled at all, and of what basis he can have for saying it, unless he tries to reoccupy the kind of transcendental standpoint, outside human speech and activity, that is precisely what he wants us to renounce.

There is, then, more than one question about how to read Rorty’s descriptions of scientific activity, if we accept his view of what such descriptions can be. There is a different set of problems about the self-understanding, and the future, of philosophy. The problems are different, in particular, because Rorty expects science to continue—its “discoveries form the basis of modern scientific civilization. We can hardly be too grateful for them.” But philosophy should come to an end; or rather, as he often puts it, “Philosophy” should, where the upper case stands for philosophy as a distinct Fach or professional undertaking. There will be room for a kind of post-Philosophical philosophy, a kind of cultural criticism, for which there is no very special expertise. Occasionally Rorty’s speculations about the future of this activity strike a Marxian-utopian note; the nonprofessional inheritor of Philosophy will be a new Renaissance polymath doing literary criticism in the morning and history in the afternoon, and doing them in a spirit of Nietzschean gaiety.

Yet here again there is a problem about how this activity is supposed to coexist with a consciousness of its own nature. It is hard to see how these new forms of intellectual life can thrive for long, when they are at the same time so professedly second-order, derivative, and parasitic on the activities of those in the past who have taken themselves to be doing Philosophy in its own right. “Philosophers could be seen as people who work with the history of philosophy and the contemporary effects of those ideas called ‘philosophic’ upon the rest of culture. …” The reference to the history of philosophy, and the quotes around “philosophic,” immediately reveal the inherited identity that backs up this image. Even the Nietzschean gaiety relates to the use of these figures of the past; it is with approval, I think, that he says of Derrida (one of those who recognize where we really are) that he “does not want to comprehend Hegel’s book; he wants to play with Hegel.”

I doubt, in fact, whether Rorty has extracted from the ruins, as he sees it, of Philosophy any activity that will sustain a post-Philosophical culture of the kind that he sketches. It is not very realistic to suppose that we could for long sustain much of a culture, or indeed keep away boredom, by playfully abusing the texts of writers who believed in an activity which we now know to be hopeless.

Rorty’s views, however, affect more than the future of philosophy considered as a distinct activity. They raise important questions about the significance for culture in general of certain intellectual ideals—above all, a certain image of truthfulness—which philosophy, in some of its styles, particularly cultivates.

One of Rorty’s aims is to overcome the division between scientific and literary culture; he refers surprisingly often to the late Lord Snow, associating with him various distinctions that are considerably subtler than any that occurred to Snow himself. At the same time, he wants to overcome the divide between two kinds of contemporary philosophy, broadly called “analytical” and “Continental.” I have already said that so far as the future of culture is concerned, the first of these aims is not going to be realized in Rorty’s terms, since the business of engaging in scientific research, and the intellectual motivations that people have for doing so, are so totally unlike making comparisons within a web of texts that even if (in some sense that Rorty still needs to explain) that is what science really is, the activity will, so long as it flourishes, reject that description of itself.

But that was a point about describing the aims of science, not of adopting a specific methodology, and even if science successfully continues with the conception of itself as discovering what is really there—if, that is to say, science continues—this leaves open most questions about its connections with any wider cultural or social conceptions of rationality. Here the other divide, between the two kinds of philosophy, comes into the picture. There is something in what Rorty says when he claims that analytical philosophy and Continental philosophy have been the public-relations agencies of science and of literature respectively.

There is something in it, though much is left out. Positivism apart, analytical philosophy has not been committed to the supremacy of science, or to validating science’s laudatory images of itself, and all of this Rorty himself explains very well, both here and in The Mirror of Nature. (He says, incidentally, in the preface to that book, that it could as well have been written in a Heideggerian as in an analytical style, and it is simply a matter of his own experience that he has chosen the latter. He may believe this claim, but I doubt that anyone else does.) But it is certainly true that the discourse of analytical philosophy, its argumentative procedures, are more continuous with those of scientists. It seems to its practitioners more responsible, more consequential, less open to arbitrariness, whimsicality, and rhetoric than other styles of philosophy, and I suspect that it seems so to scientists as well, insofar as it does not seem to them, along with most other philosophy, merely pointless.

If Rorty is right, there is nothing to these contrasts at all, and analytical philosophy’s claim to greater intellectual virtue, of a kind that has some general cultural significance, is simply baseless. It merely mistakes articulateness for clarity of perception and argumentativeness for rationality. It derives no prestige from its relation to science, both because there is no methodology that it can share with science, and because science isn’t in any case what this philosophy generally thinks science is. Its characteristic neglect of the imagination is not a contribution to objectivity but a self-inflicted limitation. If analytic philosophy is like anything else at all, it most resembles the activities of lawyers under an adversarial system, and its admired skills are mainly the forensic skills of courtroom debate.

“Forensic” might be thought at least minimally a compliment, but the complimentary element is missing. Granted a legal system, forensic practices can be thought to assist justice. But without any analogous system of rules, without any accepted standards of argument and evidence, the forensic practices of philosophy will be left, for Rorty, only with the worst aspects of the adversarial system. Thus analytical philosophy is not more rationally organized than any other sort of philosophy; it merely employs a different kind of rhetoric, and uses different methods to bully opponents.

Rorty has made a vigorous and entirely serious challenge, which raises a question more important than merely how to do philosophy. That question can never in the end be that important, and Rorty himself criticizes some philosophers he admires, such as Heidegger, for overrating the significance of philosophy itself for civilization. But the value of philosophical styles of argument go beyond the value of philosophy, because of the virtues that they try to express. No one has to believe that the questions of philosophy are the most important questions there are, or that philosophy can discover what mankind should be doing. But analytical philosophy does hold that it offers a very abstract example of certain virtues of civilized thought: because it gives reasons and sets out arguments in a way that can be explicitly followed and considered; and because it makes questions clearer and sorts out what is muddled.

On this view, analytical philosophy asserts important freedoms, both to pursue the argument and, in its more imaginative reaches, to develop alternative pictures of the world and of human life. It is both a creative activity and an activity pursued under constraints—constraints experienced as, among others, those of rational consistency. Its experience of those constraints, and the terms in which it approves those who most imaginatively work within them, is one where its spirit overlaps with the sciences. Both in this philosophy and in the sciences, the ideal is the old Socratic ideal that mere rhetoric and the power of words will not prevail.

This is the image of philosophy and its virtues that Rorty radically criticizes. He seems to me, however, very unradical and excessively optimistic in his picture of an intellectual community that has got rid of this image. Certain “conversational constraints” (of roughly Habermas’s kind: he does not say much about them) will keep things together as much as anything ever does, and we shall just have to do whatever we can to sustain traditions of open-mindedness and receptiveness to new considerations. He does not want us to get too excited or unnerved by, for instance, Foucault’s vision of discourse as a network of power relations. “‘Power’ and ‘culture,”‘ he writes, “are equipollent indications of the social forces which make us more than animals—and which, when the bad guys take over, can turn us into something worse and more miserable than animals.” When the bad guys take over: there are at least four different ways of intelligibly stressing the words in that phrase, and each of them expresses an equally shallow way of thinking about what happens to a society when rational civility collapses.

No more than in The Mirror of Nature does Rorty give many indications of how discourse should go on when freed of the illusions of truth and objectivity. In the general cultural context, he is just as optimistically neglectful as he was in the case of science about the effects of everyone’s coming to believe what he has to say—effects which the pragmatist, least of all, can afford to neglect. But that still does not mean that he is wrong, except possibly by his own standards of what it is to be wrong, and his challenge to the standards of what analytical philosophy calls clarity and rationality remains one to be taken seriously.

His kind of questioning has great force in a field that he himself does not take up, that of moral philosophy. Analytical moral philosophy has now revived the activity of theorizing about what is right and wrong rather than merely analyzing ethical terms. What this activity urgently requires, and has never yet managed to provide, is some coherent understanding of the relations of such theory to practice, where this includes the relations of the theorists to the rest of society. In the work of such philosophers as Peter Singer, it seems merely to be assumed that the virtues of an intellectual theory, such as economy and simplicity, translate into a desirable rationality of social practice. That represents a Plantonic rationalism of the most suspect kind. There is no advance guarantee of ways in which humane and just social practices may relate to philosophical theory of any sort. That is just one application of the question that Rorty rightly presses, of the relations between the discursive virtues of analytical philosophy on the one hand, and desirable forms of social rationality on the other.

These essays along with The Mirror of Nature should encourage philosophers, and not only philosophers, to ask and pursue that question. There are lessons to be learned from the new and unexpected forms in which Rorty puts the question. There is also something to be learned from the weak parts of his account. The two major weak points are the inadequacy (certainly the pragmatic inadequacy) of his account of science, and the very weak indications that he gives of the nature of a post-Philosophical culture. Perhaps this double weakness expresses a strength in the traditional idea that philosophy and science can share a conception of truthfulness that is not merely an application of the will to power.

An account of the relation of science to culture should still start, it seems to me, from that impression which so powerfully affects its practitioners, and which is so dismissively treated by Rorty: that science offers one of the most effective ways in which we can be led out of the web of texts, the archive of discourse in which Rorty finds himself imprisoned along with the “bookish intellectuals of recent times.” In his optimistic dealings with Foucault, Rorty quotes the dreadful sentence “Man is in the process of perishing as the being of language continues to shine ever brighter upon our horizon.” I suspect that unless we keep the sense (cherished but misinterpreted by empiricism) that science finds ways out of the cell of words, and if we do not recover the sense that pursuing science is one of our essential experiences of being constrained by the truth, we shall find that the brightness of language on the horizon turns out to be that of the fire in which the supremely bookish hero of Canetti’s Auto-Da-Fé immolated himself in his library.

This Issue

April 28, 1983