1. This is a remarkably strenuous issue, largely neglected by analytic philosophers (and others as well). I have already suggested that Chomsky gives the matter very scant attention, intruding in effect a sense of "knowledge" of the deep structure of language (that we are unaware of) as an extension or analogue of the "ordinary" sense of "know"—which he does not bother to analyze. This, the most interesting part of his theory philosophically, is nowhere explicitly defended, except to say that there is really no other alternative. Piaget, however, offers the sketch of an alternative that is not unreasonable in its general outlines, except that it is an eccentrically ''structuralist" alternative. There is a very revealing exchange between Chomsky and Piaget in a volume devoted to a debate between them, that exposes the arbitrariness of each. See Massimo Piattelli-Palmarini (ed.), Language and Learning: The Debate Between Jean Piaget and Noam Chomsky (Cambridge: Harvard University Press, 1980), particularly the opening papers by Piaget and Chomsky. On Piaget's view, see, also, Jean Piaget, Structuralism, trans. and edit. Chaninah Maschler (New York: Basic Books, 1970), which confirms that Piaget mysteriously supposed that the sequence of the developmental phases of our cognitional powers is somehow triggered by the external environment; Piaget, however, does not explain (Chomsky catches him out in this) why the developmental sequence is invariant. See, also, C. H. Waddington, The Strategy of the Genes: A Discussion of Some Aspects of Theoretical Biology (New York: Macmillan, 1920), and Evolution of an Evolutionist (Ithaca: Cornell University Press, 1975). On his side, Piaget effectively challenges Chomsky's assumption that there is a determinate disjunction between what is "innate" and what is "acquired." Vygotsky criticizes the early Piaget in a telling way for the implicit solipsism of his account. See L. S. Vygotsky, Thought and Language, edit and trans. Eugenia Hanfman and Gertrude Vakar (Cambridge: MIT Press, 1962), ch. 2. In a rather subtle way, Jerome Bruner draws attention to the likelihood of prelinguistic socialized invariants on which first-language learning depends—which, if granted, could easily admit forms of grammatical regularity short of the invariances of a "universal gram-

mar" (a hard-wired, species-specific grammar, the full meaning of which is hardly clear)—given that there is no obvious way to account for "informational" invariances of the sort in question. For a glimpse of Chomsky's view, see Noam Chomsky, Rules and Representations (New York: Columbia University Press, 1980). On Bruner, see Jerome Bruner, Acts of Meaning (Cambridge: Harvard University Press, 1990), ch. 3.

2. An excellent recent specimen is provided in John R. Searle, The Rediscovery of the Mind (Cambridge: MIT Press, 1992). It obviously admits social phenomena in its account, but effectively construes social significance in terms of some prior psychological endowment, which is itself identified with the brain's mode of functioning. This is precisely what I call "solipsistic." There is no convincing way to generate the "social" or ''collective" (for instance, in terms of language) from solipsistic mental resources. Searle promises a new book on the "social character of the mind" but he does not acknowledge the "anthropomorphized" nature of his present speculations about consciousness. So he holds that the mental is a feature of the brain. At best, the present argument is premature. I should add that Husserl is, in my opinion, guilty of a similar mistake, although it is frankly never quite clear just how Husserl distinguishes between the "psychological" and the "subjective." He clearly means to absorb the social in the (subjective) work of transcendental phenomenology. But it is doubtful that he succeeds; if he had succeeded, he would not be able to disjoin the psychological and the subjective, and then the search for the apodictic would have been compromised. See Edmund Husserl, Cartesian Meditations, especially the Fifth Meditation. As I have already remarked, the generic version of this trick appears already in Kant's first Critique .

3. Fodor's indebtedness to Chomsky's conception is apparent in Jerry A. Fodor, The Language of Thought (New York: Thomas Y. Crowell, 1975). See, also, Fodor's contribution, "On the Impossibility of Acquiring 'More Powerful' Structures," in Massimo Piartelli-Palmarini (ed.), Language and Learning, which confirms that the thread runs through all of Fodor's work. In effect, both Chomsky and Fodor are Platonists, who have interpreted "recollection" as genetic invariance with respect to "knowing."

4. I mean, by "equilibration," a theorizing strategy, not an epistemic source. For its more or less standard use, see Nelson Goodman, "The New Riddle of Induction," in Fact, Fiction, and Forecast, 2d ed. (Indianapolis: Bobbs-Merrill, 1965); and John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971).

5. The reference is to Thomas Nagel's well-known paper, "What Is It Like to Be a Bat?" in Mortal Questions (Cambridge: Cambridge University Press, 1979). Nagel has made a complete about-face on the analysis of mind, but his query does not improve our sense of the question. The reason, plainly, is that only humans can know what it is like to be a bat! Humans may make empirical mistakes about the matter, as perhaps they do in wondering whether lobsters feel pain on being boiled alive. But only humans can correct their conjectures, by interpolating within a range of cases that are reasonably clear-cut. But if this holds for Nagel's conjecture, it holds as well for Chomsky's and Fodor's. The truth is, their conjectures presuppose (but nowhere establish) something like

Hjelmslev's doctrine: that is, that for the process of natural-language discourse there must be a closed generative system adequate to it. They do not show the modal necessity for such a supposition. Nor do they notice the "anthropomorphic" nature of their own speculation. See Hjelsmlev, Prolegomena to a Theory of Language . Searle is at his most effective, in The Rediscovery of the Mind, in combatting Chomsky.

6. The essential pivot of Gadamer's hermeneutics is directed against the so-called Romantic hermeneuts, who claimed that the objective rule for determining the meaning of a linguistic (or literary) utterance was the speaker's (or author's) intention (in the intuitive psychological sense). Gadamer's thesis is simply that speakers' intentions can be reclaimed only in a constructivist sense—from the vantage of our present reflexive practice—and hence, that that recovery is subject to two conditions: one, that "authorial intent" is a function of what, from our present vantage, we determine to be the intentional ethos or tradition within which relevant utterances are uttered; the other, that the meaning or significance of what is uttered is tacitly affected by our horizonal interests and sense of significance. The first condition implicates the historicity of discerning authorial intent; the second implicates the historicity of discerning the significance of whatever is designated in accord with the first. Together, they mark the joint play of what Gadamer calls Horizontverschmelzung (the fusion of horizons) and wirkungsgeschlichtliches Bewusstsein (effective-historical consciousness). Together, they deny the possibility of closure and uniquely correct interpretations under the terms of the hermeneutic circle. See Gadamer, Truth and Method . Gadamer's entire theory may be fairly construed as "post-Heideggerean," meaning both that Gadamer was specifically influenced by the theory of time and history offered in Heidegger's Being and Time, and that, as a consequence, Gadamer avoids metaphysical fixities (notably in the human sphere). The upshot is that Gadamer's theory, like Heidegger's, emphasizes questions of "authentic" existence rather than of methodological objectivity. This, however, leaves the question of objectivity unresolved but still relevant. Moreover, Gadamer's own themes require an answer that he nowhere supplies. See Hans-Georg Gadamer, "Interview: Historicism and Romanticism,'' in Hans-Georg Gadamer on Education, Poetry, and History, Applied Hermeneutics, ed. Dieter Misgeld and Graeme Nicholson, trans. Lawrence Schmidt and Monica Reuss (Albany: SUNY Press, 1992).

7. Rightly seen, the theorem is the essential pivot for the defeat of the entire classical tradition of epistemology: from Descartes and Locke, through Hume, through Kant, up to Husserl— a fortiori, to all the naturalisms of the twentieth century. For (10.9) signifies that any and all accounts of the solipsistic or species-wide resources of cognizing agents are projected from within the terms of our lebensformlich competence. Applied to Quine, for instance, it raises a question (which Quine never addresses) of why "analytical hypotheses" are not called into play in every cognitive claim, why there is an interval reserved for certain "stimulus-meaning" or "holophrastic" utterances that are not subject to the influence of "analytical hypotheses." See Quine, Word and Object and Pursuit of Truth . This may be all the reassurance Donald Davidson needed for his claims

in "A Coherence Theory of Truth and Knowledge." There is a loose analogy here between Davidson's use of Quine and his use, elsewhere, of Tarski.

8. I cannot find any clear version of the theory of Forms in Plato's dialogues, and I cannot find clear evidence that Plato is committed to the theory of Forms. There's no question that he broaches the matter and appeals to myths involving the Forms. But I also cannot find any evidence that he repudiates the doctrine, or that Aristotle is mistaken in claiming that his own doctrine is closer to that of Socrates than is Plato's. See, for instance, Aristotle, Metaphysics . Aristotle says that Socrates "did not make the universals exist apart." By a similarly motivated reflection, I cannot see that Wittgenstein actually formulates his notion of Lebensformen sufficiently explicitly to answer relevant epistemic questions, although the clues he offers are more robust than Plato's with regard to the Forms. It would not take much to interpret Plato along lines not altogether distant from Wittgenstein's—if one cared. The conservative bent of Wittgenstein's thought is captured in J. C. Nyíri, Tradition and Individuality: Essays (Dordrecht: Kluwer, 1992).

9. The theorem draws attention, among other things, to two features of collective entities: one, that the (Intentional) predicables attributed to persons, artworks, and the like are, intrinsically, subject to changes due to historicity; another, that their "natures" are, accordingly, no more determinate than Intentional attributes can be. Notions like period style, for instance, are inherently informal. This bears on the fortunes of the various forms of hermeneutics. For an instant sense of the difference between the "semiotic" and "Romantic" views of style or genre, see Nelson Goodman, "The Status of Style," in Ways of World-making (Indianapolis: Hackett Publishing Co., 1978); and E. D. Hirsch Jr., Validity in Interpretation (New Haven: Yale University Press, 1967). Hirsch, in Appendix II, explicitly condemns Gadamer's "Heideggerean'' innovation. For Gadamer's use of "prejudice," see Gadamer, Truth and Method . To grasp the sense of (10.14) is to realize: (a) that the "nature" of any particular possessing collective, Intentional properties may be affected by pertinent changes in other particulars sharing similar properties (the "tragic" cast of Sophocles's Antigone may be affected by the later history of Shakespeare's Hamlet ); and (b) the cognitive competence of the individual members of the same society to understand one another is a function of their sharing a common Lebensform, in a sense that parallels the mutual interpretability of artworks possessing any of a range of associated genres or styles within a common cultural tradition. I have explored these connections further in my Interpretation Radical But Not Unruly . I should add, of course, that I certainly do not regard Goodman as a "Romantic" hermeneut; only that, like the Romantics, Goodman lacks a sense of historicity. (Goodman's theory is a kind of ahistorical semiotics.)

10. See Iris Murdoch, Metaphysics as a Guide to Morals (London: Chatto and Windus, 1992), ch. 4. See, also Alexander Nehamas, Nietzsche, Life as Literature (Cambridge: Harvard University Press, 1985).

11. I take the acquisition of a first language to be an empirical mystery. I cannot see how it can be analogized, as Chomsky does, to the acquisition of a second language. That is simply Platonism. The idea plays a large part in Chomsky's debate with Nelson Goodman and Hilary Putnam, for instance, who are

noticeably flat-footed in response to Chomsky. But the image is the wrong one, for, on Chomsky's own view, "universal grammar" (which is said to be innate) cannot function apart from the acquisition of a first language in the usual intuitive sense. Chomsky nowhere discusses that, and the supposed possession of a universal grammar is nowhere convincingly shown to function criterially with respect to the acquisition of a "natural" language. Wittgensteinian Lebensformen, on the other hand, bear only on the fait accompli, not on the process of the acquisition. See the symposium on Chomsky's "innateness hypothesis" in Synthese 17 (1967), to which Noam Chomsky, Nelson Goodman, and Hilary Putnam contributed.

12. For Bourdieu's use of the term, see Pierre Bourdieu, Outline of a Theory of Practice, trans. Richard Nice (Cambridge: Cambridge University Press, 1977), The Logic of Practice, trans. Richard Nice (Stanford: Stanford University Press, 1980), and In Other Words: Essays Towards a Reflexive Sociology (Stanford: Stanford University Press, 1990).

13. I find Marx unwaveringly clear—implicitly—about this theorem. It is the key both to the joke about "Robinsonades" and to the criticism of Pierre Joseph Proudhon. See Karl Marx, Grundrisse: Foundations of the Critique of Political Economy, trans. Martin Nicolas (New York: Random House, 1973), Introduction. Marx's emphasis is invariably on the historicized nature of collectively enabled cognitive powers—what Marx clearly means by " praxis ." The term is Aristotle's, originally; also, Marx does not discuss the notion in a sustained and systematic way. For what seems to me to be the most convincing elaboration of what Marx may have had in mind, see Karel Kosík, Dialectics of the Concrete: A Study on Problems of Man and World, trans. Karel Kovanda and James Schmidt (Dordrecht: D. Reidel, 1976). Here you have the clue regarding my deformation of Wittgenstein's Lebensformen and Gadamer's horizonal sense of "tradition." For, of course, neither Wittgenstein nor Gadamer has anything to say about the dynamics of social history.

14. On the very different behaviorisms of Skinner and Pavlov, see B. F. Skinner, Science and Human Behavior (New York: Macmillan, 1953), and About Behaviorism (New York: Knopf, 1974); and I. P. Pavlov, Conditioned Reflexes, trans. G. V. Anrep (London: Humphrey Milford, 1927), and Lectures on Conditioned Reflexes, trans. W. H. Grant (New York: International Publishers, 1928). Of course, Pavlov is neither a reductionist nor an eliminativist regarding human minds or selves.

15. I draw attention, here, to two important themes: first, epistemology is inseparable from the metaphysics of persons; second, epistemology is inseparable from moral philosophy. Both are ignored in analytic philosophy. Within the terms of symbiosis, theorem (10.29) strengthens the artifactual standing of both truth-claims in general and claims of moral objectivity in particular. This is the consequence of conceding that knowing is inseparable from what we are able to do as agents . I take this to be the most abstract consequence of endorsing something like Marx's notion of praxis . The issue will occupy us in chapter 11. I may perhaps say that Rorty's pretense to restore historicity to philosophy is nowhere more transparent than in his advocacy of Davidson's ahistorical epistemology and in his disjunction between the private and public spheres of in-

terest among human agents. These moves are simply incoherent on the acceptance of anything like (10.29). See Richard Rorty, "Pragmatism, Davidson and Truth," and Contingency, Irony and Solidarity .

16. For a sense of various standard ways of distinguishing between the two, see Chisholm, Theory of Knowledge ; and Fred I. Dretske, Seeing and Knowing (Chicago: University of Chicago Press, 1969). I offer these as specimen views only. Chisholm's favors a foundationalism I cannot endorse, and Dretske's an externalism I cannot endorse.

17. Hume, who is a very likeable and ingenious philosopher, is almost never taken to task by his analytic admirers for the hopeless muddle that lies at the very heart of his theory. He offers his "official" account in terms of sensory "impressions" and "ideas," but, in doing so, he appeals to our sensibilities in discerning these . Now, Hume nowhere develops the issue of how to understand the continuous perceptual and cognitive competence of selves in virtue of which (alone) his entire argument makes sense. One might suppose Hume had missed the question in some way or other. But I know of no empiricist who has seriously considered the bearing of Hume's doctrine on the existence and nature of selves! (If anything, they typically—and wrongly—take Hume to deny that there are selves.) By parity of reasoning, I find most analytic accounts of perception and cognition (however they depart from empiricism) to neglect in general (unless they are out and out Kantian) a theory of selves as cogniscient agents apt for the perceptions attributed to them. In this sense, Hume's influence among analytic philosophers is "clinically" instructive. See Hume, A Treatise of Human Nature .

18. Causal accounts of perception and knowledge are very popular among analytic philosophers. Nevertheless, I know of no explicit, sustained, and convincing account that comes to terms with the ineliminable propositional element of cognitive states. Two points need to be stressed: one, to the effect that the admission of the propositional ingredient inevitably leads to the ineliminability of legitimative matters; the second, to the effect that naturalizing epistemology leads to a causal theory of perceptual (and similar sorts of) belief. The second seems to free matters for a causal theory, but it does not really do so for we have no satisfactory account of the causal conditions of intentional states —certainly none that could be said to behave in nomologically regular ways. Clearly, a causal theory of belief is bound to implicate a model of rationality (along lines already sketched, to which I shall return in a moment). Also, the judgment that suitably caused beliefs constitute knowledge remains a nagging question that the causal theory cannot itself resolve. See, for specimens, H. P. Grice, "The Causal Theory of Perception," in Studies in the Way of Words (Cambridge: Harvard University Press, 1989); and Alan White, with H. P. Grice, in "Symposium: The Causal Theory of Perception," in Proceedings of the Aristotelian Society, suppl. vol. 4 (1961). On the most recent version of the naturalist's causal account of knowledge, see Alvin I. Goldman, "A Causal Theory of Knowing," in Liaisons, and Epistemology and Cognition (Cambridge: Harvard University Press, 1986), ch. 3.

19. I take Wittgenstein's account of pain avowal to betray the fact that he, too, is an externalist. This suggests that he does not construe his Lebensformen

in the way I have deliberately exploited. For, if he had, he would have construed persons and their perceptions as artifacts of our Lebensformen, and then he would have had to be hospitable to the idea that avowals of pain could function reportorially (in their distinctive way) every bit as much as constative utterances about sensory perception. See Wittgenstein, Philosophical Investigations . For an appreciation of Wittgenstein's account, see George Pitcher, The Philosophy of Wittgenstein (Princeton: Princeton University Press, 1964), ch. 12.

20. On Brentano's view, see Brentano, Psychology from an Empirical Standpoint, bk. 2.

21. Freud treats the unconscious as a theoretical posit invoked in explanatory contexts. It plays no reportorial role in first-person contexts. See Sigmund Freud, "The Unconscious," in Collected Papers, vol. 4, trans. under supervision of Joan Riviere (New York: Basic Books, 1959). It helps, therefore, in clarifying the fact that "belief" often plays an explanatory role in ordinary contexts, where it is "unconscious" in a familiar sense. Chomsky's version of the "unconscious" (''innate," "species-specific") does no explanatory work that I can see. It has no variable or variably structured function in different linguistic contexts. It amounts to no more than an en bloc pronouncement that whatever we rightly claim are the invariant and exceptionless grammatical structures of natural language are innately present in the biological resources of humans. It does not explain how these structures work and it does not confirm that there are any such structures. See Chomsky, Rules and Representations, and Knowledge of Language .

22. Here I generalize in a way linked to Locke's theory, except that Locke (who is of course profoundly "pre-Kantian") does not realize that construing persons forensically alerts us to the possibility that both persons and their perceptions may be artifacts of a deeper process. Locke treats persons as independent existents— but forensically; whereas I treat them as artifactual existents— hence forensically. This suggests why Locke has no difficulty reconciling his "Cartesian" view of natural rights with his empiricism. It also explains why, although Hume is cleverer, Locke is more solid and more plausible. That Hume is on his way to some sort of constructivism is clear, but Kant's example shows that it needn't have led to the doctrine of symbiosis or to that of historicity. So the novelty of (10.40) becomes clearer. See Locke, An Essay Concerning Human Understanding, and, also, Two Treatises of Government, edit. Peter Laslett (Cambridge: Cambridge University Press, 1960).

23. I take Putnam to have had a strong disposition favoring the collective, lebensformlich, constructivist theory of persons the post-Kantian tradition has championed, without quite grasping the full import of that line of reasoning. See Hilary Putnam, "The Meaning of 'meaning'," in Philosophical Papers, vol. 2 (Cambridge: Cambridge University Press, 1973). It helps to explain the collapse of Putnam's epistemology and metaphysics in The Many Faces of Realism, while not yet abandoning the vestiges of a Kantian externalism.

24. A very reasonable account of the need to posit theoretical entities is offered in Ian Hacking, Representing and Intervening: Introductory Topics in the Philosophy of Science (Cambridge: Cambridge University Press, 1983). For an account along different lines, see Cartwright, How the Laws of Physics Lie .

Van Fraassen, who is more of a purist in the way of positivism than Cartwright, tries to interpret the laws of nature in empiricist terms: except that, in his appealing but self-defeating candor, van Fraassen admits that there is no principled demarcation between theoretical and perceptual distinctions. See Bas C. van Fraassen, The Scientific Image (Oxford: Clarendon, 1980).

25. I am not certain how, precisely, the notion of the "folk-theoretic" became a term in the eliminationist idiom. I feel sure that it has its immediate sources in Wilfrid Sellars, "The Language of Theories" and "Philosophy and the Scientific Image of Man," in Science, Truth and Reality ; and in papers like that by Feyerabend, "Materialism and the Mind-Body Problem," Review of Metaphysics 17 (1963). But it appeared in what seemed an instantly established usage in Stephen P. Stich, From Folk Psychology to Cognitive Science: The Case Against Belief (Cambridge: MIT Press, 1983); and is associated in my mind with the views of such figures as Churchland, A Neurocomputational Perspective ; and Daniel C. Dennett, Brainstorms: Philosophical Essays on Mind and Psychology (Montgomery: Bradford Books, 1978). But to be able to affirm (10.45) on the strength of the foregoing argument is to expose the rather surprising laxness of the entire company of eliminativists, who fail to a man to explain precisely how they mean to dispose of the ''subject" or "self" or cognizing "agent." Feyerabend suggests that we wait as long as the "folk" tradition has held sway; Churchland says that anything like (10.45) is simply "empirically" false; Stich says what he would replace the folk idiom with but admits he cannot see how to do so quite yet.

26. The double lesson of the theorem is that "objectivity" is an artifact of lebensformlich practices and, as such, belongs to the realism of the cultural, but only (or initially) in a holist way. I construe holism here in a sense akin to that of Wittgenstein's appeal to Lebensformen (or, a "language game" within the terms of a Lebensform ), that is, acts and utterances in which we are entitled to claim a sense of congruity with our enabling practices but not yet with rules or criteria in virtue of which our claims may be judged to be right or valid. We simply "know how to go on" within our "form of life," and whatever we assign as the criteria of correctness will, similarly, conform with those encompassing practices. See Ludwig Wittgenstein, Remarks on the Foundations of Mathematics, trans. G. E. M. Anscombe (Oxford: Basil Blackwell, 1956), and On Certainty . The important point to bear in mind is that Quine's "holism," in Word and Object and Pursuit of Truth, is altogether different—a heterdox and doubtful interpretation of Duhem's more interesting holism. Quine's, I'm afraid, is incoherent: it tries to combine the "indeterminacy of translation" doctrine (which looks at first glance like a generalization of Duhem's theory) with some sort of minimal empiricism (which is supposed to be metaphysically neutral, as in the way of "holophrastic sentences"). But Duhem's emphasis is applied only to the empirical testing of theories, without our being able to say precisely which propositions within those theories are being tested (and why that is so), whereas Quine's maneuver is meant to subvert the pretensions of something like a "folk" science and metaphysics. See Duhem, The Aim and Structure of Physical Theory .

27. The hermeneutic circle is the central theme of so-called hermeneutic philosophies. Originally, it was applied to the interpretation of texts for which

canons might be claimed: sacred, legal, and, more recently, literary. In that form, in the tradition from Schleiermacher to, say, E. D. Hirsch, it simply signifies the part/whole relation of interpretable texts: the meaning of any part depends on the meaning of the whole, and the meaning of the whole depends on the meaning of its parts. In post-Heideggerean hermeneutics—Gadamer's, preeminently—societal life itself conforms with the terms of the hermeneutic circle. In that sense, the "circle" signifies the consensual nature of interpretation, the absence of ahistorical canons, the historicized and openended nature of interpretation itself. I am simply usurping the term in that sense, in a way that, contrary to Gadamer's usage, permits us to reconsider what objectivity may mean within the human sciences and human studies. My own argument emphasizes, of course, that the objectivity accorded the natural sciences is also affected. (I shall come to this in a moment.) For a sense of the "Romantic" view, see Hirsch, Validity in Interpretation ; for the post-Heideggerean, see Gadamer, Truth and Method .

28. I have isolated the "hermeneutic circle" for particular mention. But there are a great many different influential views of the relationship of interpreting agents and interpretable materials that trade on the same part/whole relationship. I see the same theme in Marx's notion of praxis and in Roland Barthes's semiotized (poststructuralist) literary criticism; also, of course, in Foucault's archaeologies. See Roland Barthes, S/Z: An Essay, trans. Richard Miller (New York: Hill and Wang, 1974); also, Foucault, The Order of Things .

29. Ernest Jones's Hamlet and Oepidus (New York: Norton, 1949) is a well-known example, literal-minded though it is. A more interesting case is offered by George Thompson, Aeschylus and Athens (London: Lawrence and Wishart, 1941), who shows us how to construe the Oresteia in Marxist terms. I view these examples both metaphysically and epistemologically. The idea that these plays are simply being cleverly interpreted misses the essential challenge. My thesis is that the cultural world is "permeable," "porous," and (as I shall say in a moment) "labile''—metaphysically—because of its Intentional properties, and that objectivity in the human sciences is affected—consensually—because, in the human world, perceiver and perceived are one (10.51)-(10.56). I hold that persons and artworks are very similar, ontologically, with regard to the fixity of their "natures."

30. I find it difficult to account for Dennett's blunder. See Dennett, Content and Consciousness . Dennett does not correct his mistake in The Intentional Stance . At bottom, Dennett is an eliminationist, but he is also put off by the crudity and haste of the usual eliminationists. Hence, he keeps lengthening the postponement of the "inevitable" eliminativist coup. But he never supplies the decisive argument.

31. For a remarkable disconfirmation of Freud's speculations about Leonardo da Vinci, see Meyer Schapiro, "Leonardo and Freud: An Art-Historical Study," Journal of the History of Ideas 17 (1956). See Sigmund Freud, "Leonardo da Vinci and a Memory of His Childhood," in The Standard Edition of the Complete Psychological Works of Sigmund Freud, trans. under the general editorship of James Strachey, vol. 11 (London: Hogarth Press and the Institute of Psycho-analysis, 1957). Shapiro's superb command of the art-historical materials as well as what pertains to Freudian psychoanalysis shows, by example,

something of the recoverability of a sense of objective claims within the human sciences. Here, I should say, somewhat against Schapiro's view of his own procedure, that where he disconfirms Freud's conjectures (about the import of treating Mary and St. Anne as being of the same age), the evidence leads to a more decisive finding than when he attempts to support his own affirmative conjectures about the psychoanalytic import of further details (for instance, about the alleged homosexual import of treating Jesus and John the Baptist as children of about the same age). I don't mean this in a merely art-historical sense—that is, in a merely evidentiary sense. What I have in mind is that truth and falsity play asymmetrical roles in a relativistic logic (chapter 4) and that art-historical arguments favor, in my opinion, a relativistic rather than a bivalent logic.

32. Actions and historical events are extraordinarily difficult to individuate—hence, difficult to reidentify through redescriptions. For one thing, they have Intentional identities, and, for another, there are no clear criteria for deciding the aptness of alternative descriptions for particular actions and particular historical events. Donald Davidson has offered a remarkably confident thesis, congruent with a moderate physicalism and sympathetic with supervenience, for identifying actions in terms of minimal physical movements ("primitive actions"). But this supposes that there is a legible relationship between actions and movements favoring the rule, "one movement, one action" (rather like William's policy of "one body, one person''). If, however, cultural emergence is, as I have argued, a sui generis form of emergence, the thesis is at the very least question-begging. See Donald Davidson, "Actions, Reasons and Causes" and "Agency," in Essays on Actions and Events (Oxford: Clarendon, 1980). Alvin I. Goldman has, in A Theory of Action (Englewood Cliffs: Prentice-Hall, 1970), constructed an alternative theory, rather on the extravagant side, one that threatens paradox here and there, that nevertheless allows for indefinitely many different actions to be conceptually linked to a single bodily movement. My own view is that neither of these accounts can be said to have isolated anything that is modally necessary. Neither is particularly plausible. And neither conforms very closely to the actual practice of natural-language discourse. The important point is that any policy falling between Davidson's and Goldman's options will upset the effectiveness of an extensional treatment of actions— a fortiori, an extensional treatment of historical events.

33. The cultural world, I am arguing, depends essentially on the role of individual persons as competent agents. I have shown how entities other than culturally apt human persons may, anthropomorphically, be construed as agents. The most interesting extension involves collective agents: societies, classes, families, clans, peoples, and the like. The Annaliste school has, in its effort to make history a science, been attracted to the possibility that individual human agents may be marginalized or eliminated. But this is to misunderstand (in a way not altogether distant from the structuralist temptation) the conceptual relationship between individual agents and the collective features of the culture human aggregates share. See Fernand Braudel, On History, trans. Sarah Matthews (Chicago: University of Chicago Press, 1980). War, for instance, can only (on the usual theories) be fought by collective agents, but, on the argument, there are no collective agents, they are fictions (10.72). Hence, wars implicate

ideologies by which aggregates of individuals believe they are serving the interests of a collective agent.

34. I believe I have invented a unique "relational" notion, suited exclusively for cultural phenomena, but comparable in an interesting way with the member/class and instance/kind "relationship." I take it that classes, kinds, and types do not exist but may mark real attributes. The notion of "tokens'' and "types" originates with Peirce, I believe. Peirce's account treats types as universals of some sort. I construe the "relation" as one "between" individuals—one that is heuristically introduced. See Joseph Margolis, Art and Philosophy (Atlantic Highlands: Humanities Press, 1980). The principal alternative interpretations appear in Wollheim, Art and Its Objects; and Wolterstorff, Works and Worlds of Art . Both of these views admit real universals.

35. See John Searle, Minds, Brains and Science (Cambridge: Cambridge University Press, 1984).