Reviewed by David Carrier

The aim of this paper is to evaluate Clark Glymour's important recent book Thinking Things Through [1] from a somewhat unexpected vantage point. I want to consider the broad view of philosophy he develops; then take issue with one point of detail, the discussion of what he calls primitivism. I ask questions about a concern Glymour mentions, but does not in depth develop, the relation of his inquiry to politics or, more broadly, to what he calls questions about "how we can best conduct our lives" [2]; and, finally, I say something about the relevance of Glymour's book to visual art.

I think Thinking Things Through is an obviously magnificent book, an analysis whose clarity, comprehensiveness and range are worthy of the highest praise. Under the guise of providing a mere textbook, Glymour offers the best account I have read of some very fundamental philosophical issues. Identifying him as thinker who inspires argument from someone like myself with very different interests is to praise him very highly; an extremely decisive writer, in dealing with points of detail, he never loses track of the broad themes. Glymour's topics fall outside my areas of professional competence; I am not competent to argue for or against his broad thesis about the nature of scientific explanation, whose evaluation in any case depends upon mastery of the details of Glymour's program. Nor do I have an alternative to his approach, and so my critical questions mostly are genuine questions to which I have no real answers. I always dislike reviewers of my work who pick on points of mere detail. Here, in arguing in with one small point, I intend to raise large issues. In order to write this paper, I have to pretend to be more confident than I have any right to be. I identify Glymour as a "logical positivist," using that useful term in a loose descriptive way, not as a term of abuse as is sometimes done. As he says in the first sentence of his earlier book Theory and Evidence, there are many differences between him and the earlier philosophers associated with that school of thought [3]. Very often the history of philosophy is studied and taught without explaining why that history is worth studying. Reading commentaries on Descartes or Spinoza, it is easy to feel that the commentators are pursuing this project for its own sake, like working at a very complex cross-word puzzle, without asking what is the interest of setting all these details in place. Since the classical texts are difficult, it is natural to treat them this way; but what proper motivation of discussion demands is some explanation of why such exegesis is worthwhile. In art history, the situation is different. Giotto's successors learned techniques unkown to him, and yet, this does not diminish the aesthetic value of his work. His paintings have intrinsic value. In literature, similarly, great texts deserve scrutiny because of their artistic value; we need not ask, if this author's way of thinking, or morality, is acceptable today; we need not accept Jane Austen's view of things, or Flaubert's, to think their novels deserving of attention. But in philosophy we seek truth, and so the situation is different. To the extent that Descartes' or Spinoza's argumentation is bound up with the limits of the science of their day, why should we take an interest in their writings, except insofar as we are interested in doing history of philosophy for its own sake?

Descartes believed that only spiritual substances could think. Observing that the most elaborate known mechanical devices had relatively few parts, he concluded that no such apparatus could think.

"It is not conceivable that . . . a machine should produce different arrangements of words so as to give an appropriately meaningful answer to whatever is said in its presence. . . . it is for all practical purposes impossible for a machine to have enough different organs to make it act in all the contingencies of life in the way in which our reason makes us act."

"Automatons never answer in word or sign, except by chance, to questions put to them . . . [4]"

Descartes could not imagine twentieth century technology and so his claims are merely of historical interest. Aestheticians often treat Kant's writing on aesthetic experience with the greatest respect, although he often discussed things he had never seen, but only knew about from books.

"Bold, overhanging, and . . . threatening rocks, thunderclouds piled up the vault of heaven, borne along with flashes and peals . . . the boundless ocean rising with rebellious force . . . make our power of resistance of trifling moment in comparison [5]."

As Alain Corbin has explained, going to the beach had became fashionable; and so even a non traveler could easily pick up these much discussed ideas about the sublimity of the ocean [6]. When Kant made very strong claims about the universality of aesthetic judgment, to me, though not to many of my fellow aestheticians, it seems fantastical that a provincial eighteenth century writer who never saw many pictures could understand aesthetic judgment as it is practiced in our present day museum culture. "Hardly any painters, except the Venetian and especially the Flemish," Hegel writes, "have become perfect masters of colour; both of these groups lived near the sea in low-lying country intersected by fens, streams and canals" [7]. In his history of the High Renaissance, Sydney J. Freedberg elaborates on just this point: "The atmosphere of this sea-borne city heightens the existence of seen things. Colour is deepened in the damp-saturated air and sharpened by the sea-reflected light, which also may make complicating interactions among colours [8]." Although he never went to Italy, Hegel's account anticipates modern discussions of Venetian art. One claim fundamental for Glymour is that philosophy should not be especially, or centrally concerned with its history. In Thinking Things Through, Glymour shows interest in philosophy's history insofar as it leads towards the present. In that way, philosophy for him is like the sciences; physicists are not primarily interested in the history of physics. Looking ahead to the present, the history of philosophy, as Glymour tells it, is the story of the movement towards present day computer science and the ways in which it has utterly transformed traditional philosophy of mind. The central figures in Thinking Things Through are Aristotle, Ramon Lull and some other medievals, Descartes, Leibnitz, Boole, and Frege in what certainly is, in part, a nonstandard history of philosophy, because their work anticipates our concerns. "The human brain is a biological computer and the cognitive activities of humans are produced by computational procedures within this biological computer" [9]. These philosophers are important because their work led us towards this view. There is not reason to care much about the details of the argumentation of Kant, Hegel, Nietzsche or many, many other figures who are thought by historians of philosophy to be highly important; these philosophers were on the wrong track. By contrast, Lull who discovered in the thirteenth century that "reasoning can be done by a mechanical process" [10] was headed in the right direction.

The argument for this position assembles a great mass of materials in a way that does justice to the great achievements of cognitive psychology and computer science. What then would be the argument for an alternative position to Glymour's, for the view that the study of history of philosophy is intrinsically valuable? This is the natural challenge he proposes to his critics: write an alternative history as rich, suggestive and plausible as his. The great philosophical texts, it might be argued, embody deep wisdom because they have withstood the test of time, demonstrating their capacity to educate many generations. When present-day American conservatives give this sort of argument, I always am disappointed that they do not have the courage of their convictions and demand that Latin and Greek be brought back into school system so that students can study canonical writers in the original languages. The notion that the test of time by itself establishes anything about the validity of a way of thinking is puzzling [11]. If a building is old and remains standing, that shows it was well constructed. That a philosophical theory is venerable does not by itself show that theory to be true, or even plausible. There is no reason that a false view might not endure for a very long time. It would be very easy to compile a large collection of all the foolish things Plato, Aristotle and the other canonical figures had to say about science, morals and many other issues. The marvelous feminist Luce Irigaray has devoted a book, Speculum of the Other Woman, to the comically misguided accounts of philosophers discussing women [12]. They were men of their time, writers whose claims today have to be read selectively, if they are to seem at all plausible.

Doing philosophy, it might be said, is so bound up in its history that this activity is impossible to perform without an acute historical sensitivity. My hunch is that such a vision can only be defended by showing how philosophy amounts essentially to something more than the philosophy of science; I mean by that, how philosophy does something other, and different than science. This anti-scientific view has been defended in our day by such otherwise extremely diverse figures as Heidegger and Wittgenstein. In a one page analysis, supplemented by a long footnote, Glymour discusses what he calls primitivism, the rejection of "the scientific description of the world as a place of things, events, and processes that are in themselves indifferent to human concerns, and in which the emergence of human consciousness and intentionality constitute phenomena to be explained" [13]. His primitivists include Wilhelm Dilthey; John Dewey; Heidegger; and Sartre, Maurice Merleau-Ponty, Paul de Man --- with qualifications, Camus --- and, in his bibliography, Philosophy and the Mirror of Nature , the book by his former Princeton colleague, the former analytic philosopher who nowadays identifies himself as a pragmatist, Richard Rorty. What links all these philosophers together is concern with "the sort of anthropomorphic conception of the world that we use in our everyday lives, a conception in which we think of things in terms of their utility to us and others and their significance as symbols." Primitivism, Glymour correctly observes, is a view of philosophy held by many non-professional philosophers---by literary critics for example. I would characterize primitivism in a perhaps different way: it is the residue of a religious world view, what is left behind of such anthropomorphic ways of thinking when their theological justification is abandoned.

Glymour does not claim to offer much of a positive argument against primitivism. Nor, it must be said, does he characterize that position in detail, or show any interest in its nuances. The strongest argument against primitivism made in Thinking Things Through is the positive demonstration of the efficacy of a scientific world view. Inevitably the philosophers who have sought to resist logical positivism seem to be fighting a rearguard action, offering ever weaker positions as science advances. Like many of us discussing something we detest, Glymour is brief. He claims not just that primitivism is wrongheaded, but that because it "emphasizes social authority rather than the autonomy of individuals and . . . denigrates rationality . . . ," it is politically pernicious [14]. Allowing that some philosophers he admires had dreadful political views---Frege was, in unpublished work, an anti-Semite---Glymour seems to claim that primitivism tends to be associated with bad political positions. Insofar as a philosopher rejects a scientific world view, Glymour suggests, he or she will tend to have nasty politics.

I do not think that Marxists can be primitivists, for they are committed to the view that it is possible to provide a very full explanation of historical events. (Whether that Marxist explanation be correct is another story.) For the positivist, Marxists have the right attitude, but a bad theory. Phenemenologists are a different matter; they do have an anti-scientific attitude. Glymour's identification of "description of the world as a place of things, events, and processes that are in themselves indifferent to human concerns" obviously raises complex questions about the evaluation of art. In literature or painting, the world is presented in terms of human concerns, and so it is natural that interpretation of such an account would do so from the vantage point of such concerns. What he is perhaps criticizing is taking an aesthetic attitude towards the world---replacing an objective view of things with looking at it in terms of its meaning in our lives. The art critic Baudelaire adapted a determinedly anti scientific world view, and that led him to write great poetry and wonderful art criticism [15]. But if Baudelaire's view is treated not as an aesthetic fiction, but an account of how the world really is, then it could have disastrous effects.

With some philosophers, Frege for example, or a contemporary figure such as the analytic philosopher of language Donald Davidson, no amount of study of their best known philosophical work would reveal their political positions. Other philosophers, Hume for example, wrote both about narrowly philosophical themes, epistemology and metaphysics, and political history. We philosophers tend not to read Hume's history; complex argumentation would be needed to relate his skepticism to his view of politics. In still other cases, such figures with sweeping systems---Kant or Hegel for example---offer political analysis. What then is striking about Glymour's list of primitivists is how varied are their political positions. Dewey and Rorty are champions of American democracy; Heidegger tried to attach himself to the Nazis, who hardly needed his esoteric theorizing; Merleau-Ponty, after the early Marxist phrase to which Glymour alludes, turned away decisively from that tradition; Sartre, after his early apolitical phase turned into a odd kind of Marxist. And Paul de Man, whose nasty early anti-Semitic writings were discovered after his death, was closely associated with Jacques Derrida, a leftist critical of Marxism whose political views are hard to summarize briefly. Heidegger's primitivism, Glymour says, "emphasizes the authority of the community over the individual" [16]. Heidegger gives this authority to tradition, to the great texts. But does not logical positivism also in one way supersede the individual, giving authority to the collective project, the scientific investigation? What remains difficult is determining whether there be an internal connection between such a philosophers' primitivism and his politics. Theodore Kisiel's very full account The Genesis of Heidegger's Being & Time suggests that Heidegger's intensive study of classical and scholastic philosophy did not in any way prepare him to grasp German politics of the 1920s [17].

Not involved with racist doctrines, Heidegger was a bookish The aim of this paper is to evaluate Clark Glymour's important recent book Thinking Things Through [1] from a somewhat unexpected vantage point. I want to consider the broad view of philosophy he develops; then take issue with one point of detail, the discussion of what he calls primitivism. I ask questions about a concern Glymour mentions, but does not in depth develop, the relation of his inquiry to politics or, more broadly, to what he calls questions about "how we can best conduct our lives" [2]; and, finally, I say something about the relevance of Glymour's book to visual art.

Part Two

The primitivist, we might say, is like a navigator who tries to steer with a defective compass; he may lead us in the right direction, but only by accident. Rejecting primitivism thus yields a minimal constraint; it rules out positions that are based upon views not true to the facts. A logical positivist will be rational politically; and that, in this century, is not an insignificant achievement. A case study worth consideration is the phenomenology of Merleau-Ponty, for he, drawing on a great deal of psychology of perception, gave a defense of primitivism. Following Husserl, he writes in the Preface to Phenomenology of Perception (1962), he seeks to re-achieve "a direct and primitive contact with the world" and endow "that contact with a philosophical status" [19]. His book concludes with a quotation from Saint-Exupery about the need to acknowledge individual choice, and so to accept our freedom. What is unclear, still, is whether this general starting point need yield any particular political conclusion. Wittgenstein, a figure closer to the world of logical positivism, in his later work certainly takes a position closely allied with primitivism. Emphasizing the dividing line between philosophy and science, and insisting that it is unbridgable he, in a different way than the phenomenologists, limits the boundaries of scientific inquiry. But Wittgenstein's way of thinking does not have any obvious immediate political implications.

None of this argumentation says anything about whether Glymour's rejection of primitivism is mistaken. All I am questioning is his claim that primitivism yields inherently pernicious political views. Primitivism includes many rather different ways of thinking, and so may suggest a plurality of political perspectives. But there is no reason to think that primitivism as such must have bad political consequences, for it is not clear that it necessarily has any political consequences. Suppose the general view of philosophy's role presented in Thinking Things Through were correct. What political position then follows?

I do not believe that the adaptation of a scientific description of the world can provide any substantial political guidance. The reasoning for that claim is twofold. First, I do not see in Glymour's book any claim that adopting a scientific world view yields any particular political position. Second, when I look at the best known substantial accounts of these issues, I do not find much reliance upon the details of a particular general philosophical position such as Glymour adapts. Political liberalism, John Rawls says öassumes that, for political purposes, a plurality of reasonable yet incompatible comprehensive doctrines is the normal result of the exercise of human reason within the framework of the free institutions of a constitutional democratic regime [20].ä

Even if we suppose that such political regimes are associated with a scientific world view, still the scientist's work does not tell us what kind of politics to adopt. Christians, Jews, Muslims and believers in astrology could, under conditions Rawls specifies, form a just community. I see no reason why a community of phenomenologists or pragmatists could not adopt Rawls' principles. (Marxists would not, but they, I have argued, are not primitivists.) Even if we imagine every citizen to be Glymour's ideally rational agent, a person who never makes logical mistakes, nothing in the history of philosophy as presented in Thinking Things Through suggests that such an agent would be capable of developing a substantial moral viewpoint [21]. Identifying truthful scientific theories does not tell us how to behave in society, except, perhaps, as it demands that we create a society in which the pursuit of scientific truth is possible. And there is no reason to associate that task only with rational democratic politics.

Glymour rejects cultural relativism, which he defines as the assertion "that the beliefs of all human communities are equally well founded and that none are more or less true than others" [22]. For my present purposes, we need only imagine that there exist more than one well founded system of beliefs, and that the aim of politics is to provide a good way of adjudicating disputes between people holding in good faith competing beliefs. Maybe the co-existence of such communities is inconsistent with genuine general adaptation of a scientific world view; perhaps ultimately views of truth will converge. Even so, for the present, one central goal of politics is to resolve conflicts.

And so here, seeking to identify the political force of anti-primitivism, I fall back on a different line of argument. Taking as given Glymour's view about the falsity of primitivism, what follows for politics? Since it seems impossible to identify an internal link between Glymour's philosophy and political theory, let us adapt another strategy, historical analysis. Glymour's view of philosophy is a natural product of the extraordinary recent successes of computer science. Of course, parts of the modern scientific world view have been in place since Galileo and Newton; but what is central to Glymour's own logical positivist position is the achievements relevant to philosophy of mind provoked by computer science. Were we not all living with inexpensive, convenient word processing, then Glymour's analysis of computability would be a highly academic affair. As it is, what gives great general interest to the discussion is the existence of such computers. And so it is natural to ask: what is the political effect of these machines? In this indirect fashion, I seek to return to understand the political implications of Glymour's philosophy.

Since computers are part and parcel of an anti-primitivist world view, we might expect them to promote a rational political regime. But because computers are so spectacularly good at controlling information, it is very easy to imagine how they could be the basis of a authoritarian society. The philosopher who has said the most suggestive, perhaps wrong-headed, things about this issue is Michel Foucault, and so it is unfortunate that Glymour is allergic to his ways of thinking. Insofar as computers are associated---so recent American experience teaches---with radical economic inequalities, it is hard to see how they could have any inherent connection with democracy. Many diverse scenarios about the effects of computers can readily be written. Twenty years ago, no one could have imagined our present situation, and so it seems hopelessly optimistic to trust in any projection about the future. This is why I am skeptical about any attempt to derive a view of politics from Glymour's logical positivism. I do not see how any discussion of the conditions of theory testing in science, the seeking of rational grounds for our beliefs, generates any particular view of what narrowly we call politics, or what more broadly we can call with Glymour "how we can best conduct our lives." Since the further development of computer technology is about as inevitable as any social process that can be imagined, what is the point of asking whether this is a good thing? It will happen, whether we like it or not; since philosophical reflection is very unlikely to change the world, what is demanded of the social philosopher is an understanding of the process.

In Thinking Things Through art has little place. This of course could be a mere accident of Glymour's personal history. When in the last chapter he describes some influential philosophical work, he adds, with winning modesty, that these examples are "chosen simply because I know of them" [23]. What might be the implications of his account for the philosophy of art history? When Glymour points to a number of new sciences---computer science, Bayesian statistics, cognitive science---as "all informed by developments in philosophy within the last 100 years" [24], it is natural for me to add art history to that list. Michael Podro has traced the process in which German philosophy was the starting point for modern art history [25]. Since Vasari, there had of course been extensive commentary on painting. But only when philosophical ways of thinking were developed was it possible for art history to become an academic discipline. It is possible to imagine art history developing according to models provided by philosophy of science. Like the scientist, the historian of art forms a hypothesis that which may be testable; like that scientist, he or she considers alternative explanations of relevant features of art, seeing the most economical, plausible and far-ranging account. Just as, finally, a scientist might test a theory by generalizing, so too an art historian could test an interpretative approach by working with new examples. Insofar as historians of art are concerned to explain representation-making, it seems natural to expect them to enlist the aid of scientific psychologies of perception.

But what is possible does not always happen, and I would find it very difficult to cite much recent influential work in art history that develops according to this model---with one exception, and that is the most famous living art historian, a man who for complicated political reasons is somewhat much out of fashion, Sir Ernst Gombrich. He is, in interesting ways, a potential ally for Glymour [26]. Anyone Jewish and interested in psychoanalysis in 1920s Vienna certainly was aware of the political power of primitivist thinking. When then later, as a emigre in London, Gombrich developed both a broad view of art's history and an account of modernism, he laid great stress on the importance of controlling the irrational by reliable means. Gombrich has a resolutely rationalist theory of art history. A novel form of representation is tested, he argues, in much the way in which a scientific theory is tested. Gombrich was much influenced by his great friend Karl Popper, both in rejecting Hegelianism in any form, including Marxism, and in thinking of testability as a key condition for a good art-historical

The history of representational art, according to Gombrich, is like the history of science as it might be told by a logical positivist. When a theory is developed, it then is critically tested; and the history of artmaking is the history of ever better ways of representing the visual world. These systems of representation can be tested by observation. The specific affinity of Gombrich's account to Glymour's views comes in Gombrich's worries about the political implications of artistic primitivism. Art that revels in the irrational, Gombrich fears, is the art of a society that may readily become irrational. Recent art history, in its concern with feminism and gender issues, with multiculturalism, with many forms of leftist politics, with poststructuralism, and with modernism and postmodernism has, not entirely fairly, identified Gombrich as the enemy [27]. Nothing would attract more hostile commentary from many sides than to speak of art history as a proto- science. The logical positivist worldview presented in Thinking Things Through would also, I fear, seem alien, repugnant and reactionary to many of my colleagues in art history. Since Glymour has contempt for his opponents, he hardly is prepared to listen to their arguments; they, in turn, are unlikely to respond sympathetically to his analysis. Seeing how large is the gap between his view of philosophy and the concerns of art historians, and also most professors in literature and cultural studies, it is hard to know how genuine discussion is possible; recognizing this is not necessarily a bad thing, for fruitless discussion is a waste of time.

Here we come to my only real criticism of Glymour. What is odd is his concern with the possible political influence of a few academic primitivists. In the recent literature, conservatives worry a great deal about the social effect of esoteric literary debates concerning deconstruction and about the support from the (U.S.) National Endowment for the Arts for a few politically controversial artists. Reading such accounts, you would think that the way literary criticism is studied at Yale or Irvine, or the way a small sector of the public looks at postmodernist art had a large effect on the entire culture. Surely the dominant ideology of our culture is logical positivism. There are, however, some primitivists who have real political effect---I mean the religious fundamentalists who, be they Christian, Jewish or Muslim, do reject "the scientific description of the world as a place of things, events, and processes that are in themselves indifferent to human concerns." In my view, the natural enemy of the logical positivist is religion and so, if Glymour is concerned with politics, it should be his real target.

Reference and Notes

1. Clark Glymour, Thinking Things Through: An Introduction to Philosophical Issues and Achievements (Cambridge, Mass.: MIT, 1992).

2. Glymour [1], p. 371.

3. Clark Glymour, Theory and Evidence (Princeton: Princeton Univ. Press, 1980), p. ix.

4. Rene Descartes, Selected Philosophical Writings , Trans. John Cottingham, Robert Stoofhoff, and Dugald Murdoch (Cambridge, UK: Cambridge Univ. Press, 1988), pp. 4-5; The Philosophical Writings of Descartes. Vol. III, The Correspondence , Trans. John Cottingham and Anthony Kenny (Cambridge, UK: Cambridge Univ. Press, 1991), p. 99.

5. Immanuel Kant, The Critique of Judgement , Trans. James C. Meredith (Oxford; Oxford Univ. Press, 1952), p. 110.

6. Alain Corbin, The Lure of the Sea. The discovery of the seaside 1750-1840 , Tr. Joselyn Phelps (London: Penguin Books, 1994).

7. G. W. F. Hegel, Aesthetics. Lectures on Fine Art , Trans. T. M. Knox (Oxford: Clarendon Press, 1975), vol. II, p. 839.

8. S. J. Freedberg, Painting in Italy. 1500 to 1600 (Harmondsworth, Middlesex: Penguin Books, 1971), pp. 76-7.

9. Glymour [1], p. 216.

10. Glymour [1], p. 71.

11. See my review of A. Savile, The Test of Time, The J. of Philosophy, LXXXI, 4 (l984):226-30.

12. Luce Irigaray, Speculum of the Other Woman , Trans. Gillian C. Gill (Ithaca: Cornell Univ. Press, 1985).

13. Glymour [1], p. 242.

14. Glymour [1], p. 375, ft 2.

15. See my High Art. Baudelaire and the Origins of Modernism (University Park, PA: Pennsylvania State Univ. Press, 1996).

16. Glymour [1], p. 242.

17. Theodore Kisiel, The Genesis of Heidegger's Being & Time (Berkeley and Los Angeles: University of California Press, 1993).

18. Maurice Merleau-Ponty, Humanism and Terror. An Essay on the Communist Problem , Trans. John O'Neill (Boston: Beaton Press, 1969 (originally pub. 1947).

19. Maurice Merleau-Ponty, Phenomenology of Perception , Trans. Colin Smith (London: Routledge & Kegan Paul, 1962), p. vii.

20. John Rawls, Political Liberalism (New York: Columbia University Press, 199 ), p. xvi.

21. Glymour [1], p. 215.

22. Glymour [1], p. 237.

23. Glymour [1], p. 370.

24. Glymour [1], p. 5.

25. Michael Podro, The Critical Historians of Art (New Haven: Yale Univ. Press, 1984).

26. See my "Gombrich on Art Historical Explanations," Leonardo , XVI, 2 (l983):91-6; and 'The Big Picture. David Carrier Talks with Sir Ernst Gombrich," Artforum , February 1996: PP. 66-9, 106, 109.

27. Eric Fernie, Art History and Its Methods. A Critical Anthology (London: Phaidon, 1995), pp. 223-26; "[W]hile many in Britain consider him a traditional figure, in France he is ranked as a radical thinker alongside figures such as the deconstructionists" (p. 224). The one well-known figure whose general attitude towards scientific accounts in art history is similar to Gombrich's is his former student Michael Baxandall; see my review, M. Baxandall, Shadows and Enlightenment and E. H. Gombrich, Shadows: The Depiction of Cast Shadows in Western Art, The J. of Aesthetics and Art Criticism (1995).