Quine’s Naturalized Epistemology, Epistemic Normativity and the Gettier Problem Dr. Qilin Li (liqilin@gmail.com; liqilin@pku.edu.cn) The Department of Philosophy, Peking University Beiijing, P. R. China 100871 W. V. Quine is one of the most prominent advocates of the naturalistic approach to epistemology and he argues that epistemology should be naturalized and transformed into a sub-discipline of psychology and hence a chapter in science. In his famous paper “Epistemology Naturalized,” Quine starts to compare epistemology with the logical and set-theoretical studies of the foundations of mathematics. Similar to the studies of the foundations of mathematics, as Quine suggests, epistemological studies, which are “concerned with the foundation of science” (Quine 1968, p. 69), can be divided into two sorts— one is conceptual and the other is doctrinal. Quine further argues that the conceptual ones are concerned with meaning of our material object concepts, which are clarified by reducing them into sense experience concepts; and, on the other hand, the doctrinal ones are concerned with truth of our material object beliefs, which are established by deducing them from the premises about observations. According to Quine, both sorts of studies in traditional epistemology are doomed, since traditional epistemology shows no real advantages over sciences, especially psychology. Thus, Quine advocates that we should abandon this traditional epistemology; but this does not imply the death of epistemology per se, since epistemology can still go on “in a new setting and a clarified status” (Quine 1 1968, p.82). We should embrace this new epistemology, i.e., naturalized epistemology, which “falls into place as a chapter of psychology and hence of natural science” (Quine 1968, p.82), in other words, unlike traditional epistemology, new epistemology “is contained in natural science, as a chapter of psychology” (Quine 1968, p.83). Quine thinks the change from old epistemology to new one is significant, because the relations between science and its foundations can only be more productively investigated in a scientific, causal way, because scientific theories are based upon the evidence that is the totality of our sensory simulations. When the evidence is understood in the above way, it seems quite reasonable for Quine to “make free use of empirical psychology” (Quine 1968, p.83) to provide epistemologists with reliable, causal analyses of human sensory simulations. With respect to naturalism in philosophy, there is always one prominent challenge— the problem of normativity, viz., how it is possible for a fully naturalized philosophical theory to save some room for the concept of normativity. If a fully naturalized philosophical theory only provides us with some descriptions of the causal mechanism without solving the problem concerning “whether we ought to …,” it seems that such a philosophical theory is philosophically less attractive. For instance, in ethics, it is quite common to suggest that scientific theories concerning emotional engagement and moral judgment are philosophically uninteresting if they only provide us with more detailed account of the relations between emotional engagement and moral judgment 1 without any further specification of whether we ought to 1 For instance, Greene, et al (2001) provide a scientific discussion about the influential role of emotional engagement in moral judgment. 2 perform certain acts under some moral situations. Similarly, some philosophers suggest that Quine, as an important advocate of naturalized epistemology, has to confront with a similar difficulty. For instance, Jaegwon Kim (1988) suggests that Quine’s proposal of naturalized epistemology in untenable, since it requires us to dispense with the normativity in epistemology. According to Kim, evidence as well as rationality plays important roles in Quine’s naturalized epistemology and yet both of them are normative concepts. Kim acknowledges that scientific theories provide us with more detailed accounts of our sensory simulations, causation, natural laws, etc., which definitely deepen our understanding of the concepts of them; however, scientific theories do not tell us when and where simulations, causation, natural laws, etc., can be taken as evidence— the concept of evidence is distinct from and therefore irreducible to the above naturalist concepts. Siding with Davidson (1973, 1974, 1975), Kim argues that Quine cannot plausibly talk about belief attribution, belief identification and belief classification within Quine’s fully naturalized epistemology, because belief attribution, belief identification and belief classification presuppose subjects’ rationality that in turn is a normative concept in epistemological studies. According to Kim, if epistemology is fully naturalized in the way that Quine suggests, we have to completely abandon the concept of epistemic normativity and the ensuing new Quinean “epistemology” would be radically different and therefore philosophically unattractive. The defenders of Quinean naturalistic approach to epistemology, however, disagree with Kim. They think Kim’s objection to Quine’s proposal of naturalized 3 epistemology is based upon certain misunderstanding, viz., that Quine’s naturalized epistemology save no room at all for the concept of epistemic normativity. These defenders (for instance, Hilary Kornblith 1980, 1994, 2002) suggest that Quine’s naturalistic approach to epistemology should be interpreted seriously. Re-consider Quine’s famous remark as follows: Epistemology still goes on, though in a new setting and a clarified status. Epistemology, or something like it, simply falls into place as a chapter of psychology and hence of natural science. It studies a natural phenomenon, viz., a physical human subject. This human subject is accorded a certain experimentally controlled input-certain patterns of irradiation in assorted frequencies, for instance-and in the fullness of time the subject delivers as output a description of the three dimensional external world and its history. The relation between the meager input and the torrential output is a relation that we are prompted to study for some what the same reasons that always prompted epistemology; namely, in order to see how evidence relates to theory, and in what ways one’s theory of nature transcends any available evidence. (Quine 1968, pp. 82-83.) According to Kornblith, Quine does not suggest that we have to completely abandon the concept of epistemic normativity— what Quine does is to undermine foundationalist program in epistemology. Kornblith suggests that there are two crucial questions in epistemology (cf. Kornblith 1994, p. 1): Question 1: How ought we to arrive at our belief? Question 2: How do we arrive at our belief? In Knornblith’s words, Quine shows us that “the only genuine questions there are to ask about the relation between theory and evidence and about the acquisition of belief are psychological questions” (Kornblith 1994, p. 4). What Quine reveals is that foundationalist program in epistemology fails to engage with the genuine 4 epistemological question, which concerns the relevance and relation between questions 1 and 2. In other words, Quine’s paper establishes that “question 2 is relevant to question 1 because it holds all the content that is left in question1” (Kornblith 1994, p. 4) and therefore the plausible answer to question 1 cannot be rationally developed independently from plausible answer to question 2. Thus, it seems that Quine’s naturalistic approach to epistemology can not only save enough room for the concept of epistemic normativity but also provide us with rectification of certain misconceptions concerning epistemic normativity. Thus, scientific, natural theories can help philosophers with some hints of the solutions to some philosophical puzzles. Take epistemic skepticism for example. As some philosophers suggest (Bernard Williams 2005, for instance), epistemic skeptics assume certain normative view of the criterion for one to tell whether the state of affairs S is obtaining. Epistemic skeptics think that, in order for one to tell whether S, she/he has to satisfy both conditions as follows: (a) One can tell that S when S; and (b) One can tell that not-S when not-S. (Williams 2005, p.298.) According to epistemic skeptics, one cannot have knowledge of the external world because she/he cannot tell whether she/he is suffering from some skeptical situations, say, being a brain in a vat (hereafter, BIV for short). Since one has no criterion for telling whether she/he is a BIV, since she/he cannot tell that she/he is a BIV when the BIV situation obtains (and hence she/he cannot satisfy the conjunction of (a) and (b) simultaneously). Williams argues, however, that the above view of criterion is 5 falsified by the scientific studies of “the symptoms of anoxia (lack of oxygen), against which high-altitude pilots have to be on their guard” (Williams 2005, p. 299). As scientific theories reveal, there are two crucial symptoms of anoxia: one is blue finger-nails and the other is over-confidence that makes one neglect such things as blue finger-nails. But Williams goes on to remark: On a rather idealized version of this phenomenon, it might well be that A could not tell that he was anoxic when he was; but it would surely be paradoxical to suggest that therefore A could not tell that he was not anoxic when he was not (for instance, A is you, now). (Williams 2005, p. 299) In this view, the supposed view of criterion that is held by epistemic skeptics is falsified and therefore rejected by the scientific considerations. In other words, the scientific studies of the symptoms of anoxia reveals to epistemologists that a plausible account of criterion should engage more with condition (b) rather than the conjunction of (a) and (b). If epistemic skeptics’ view of criterion is abandoned, epistemologists need not seriously treat those skeptical puzzles that come with the assumption of epistemic skeptics’ view of criterion. It seems that Quine’s insight of the relation between the answer to question1 and the answer to question 2 is nicely illustrated by the above example. In this sense, Kim’s objection to Quine’s proposal becomes less challenging, since the concept of epistemic normativity still survive in Quine’s naturalized epistemology. It probably can be further argued that Quine’s naturalized epistemology is able to help philosophers in developing reasonable, robust account of epistemic normativity. Quine’s naturalistic proposal can also fit into the causal theories of knowing (for 6 instance, Goldman 1967). Quine’s naturalistic proposal can also be cashed out within a “knowledge-first” framework. As some epistemologists suggest (Williamson 2000, for example), knowing is a factive state of mind. Quinean naturalized epistemology, with the assistance of scientific studies, is able to provide us with a causal theory of knowing, which scientifically specifies how the very state of mind (i.e., knowing) is caused by the perceptual simulations from the objects. If the epistemic normativity that is implied by knowing is explainable from Quinean naturalist perspective, Kim’s worries concerning evidence and rationality would be resolved respectively within the “knowledge-first” framework. 2 Bearing the above consideration in mind, we may be tempted to conclude that Quine’s naturalized epistemology is sufficient for the concept of epistemic normativity. It is to be shown, however, that the above optimistic view of Quine’s naturalized epistemology neglects some crucial tension between naturalism and normativity of knowing, which is revealed by the consideration of the Gettier problem (cf. Gettier, 1963). Alvin L. Goldman’s fake barn case is a perfect example for the current purpose. Let us compare two situations as follows: (Situation 1) Henry is driving through a normal countryside and sees a barn on his left side. The barn is fully in view. Since he has excellent eyesight and has no doubt about the identity of the barn, Henry says, “That is a barn.” (Situation 2) Henry is driving through the countryside and sees a barn on his left side. The barn is fully in view. Since he has excellent eyesight and has no doubt about the identity of the barn, Henry says, “That is a barn.” But unknown to Henry, this countryside is full of cleverly constructed fake barn facades, which make travelers invariably fail to tell them from 2 For instance, Williamson’s “E=K” can be used to resolve Kim’s objection from the consideration of normative elements in the concept of evidence. 7 real barns. But, fortunately, Henry has not encountered any fake barn facades and the object that he sees is the only genuine barn in the district. Evidently, Henry in situation 2 is gettiered and therefore has no knowledge that the object he sees is a barn. Given the same type of the causal simulations in both situations, Henry in situation 1 should have the same type of the state of mind with the one in situation 2. In both situation, the relevant causal mechanism and causal interaction between the subject and the object should also be identical in type. As Goldman concedes, such a comparison between the above two situations cannot be handled merely by a causal theory of knowing (cf. Goldman 1976, p. 773), no matter how detailed the causal story is developed. If the above argument is correct, the Gettier problem would in turn reveal certain inefficiency of Quinean naturalized epistemology— it fails to account for the most crucial concept, viz., knowing, in epistemology. The normativity involved in the concept of knowing cannot be fully captured by the naturalized, causal theory of knowing. And any naturalized epistemology that fails to provide us with a satisfactory account of the Gettier problem is definitely considered as unattractive. For many philosophers (for instance, DeRose, 1995, Goldman 1976, Nozick 1981, Sosa 1999), the normativity involved in the concept of knowing can only be handled by the counterfactual analysis of the subjunctive conditionals, which is purely a philosophical investigation that cannot be naturalized by a scientific, causal theory. 3 The crucial point revealed by the Gettier problem is that there is certain normative 3 It should be clarified that my argument neither concerns the debate between the sensitivity condition and safety condition on knowing nor depends upon any specific philosophical theory of counterfactual analysis. 8 element in the concept of knowing that is missing in Quinean naturalized epistemology. If a fully naturalized “epistemology” cannot provide us with a plausible theory of knowing that is able to handle the normative element in the concept of knowing with respect to the Gettier problem, I think, we should side with Kim and suggest that the supposed naturalized “epistemology” is no longer genuine epistemology. The above observation, however, implies that we have to be cautious when we consider the prospect of epistemology, which may be probably a hybrid of certain Quinean naturalist insight on the relation and relevance of the answers to both question 1 and question 2 as well as philosophical (i.e., non-naturalized) studies of the normative elements in the concept of knowing. In this sense, when epistemologists investigate epistemic normativity, they would concern both naturalistic and philosophical factors— on the one hand, naturalistic factors cast some significant constraints upon philosophical construction concerning the concept of epistemic normativity and help epistemologist in resisting some misleading temptation (for instance, the skeptics’ view of criterion); on the other hand, the philosophical studies of the concept of epistemic normativity cannot be exhausted by scientific results and therefore cannot be completely naturalized (or replaced by psychological studies). The complexity concerning the normativity in epistemological studies in turn reveals the uniqueness of the concept of epistemic normativity, which is distinct from the normativity in the ethical studies. Reference 9 Davidson, Donald. (1973). “Radical Interpretation,” in [Davidson 1984], pp. 125-140. ____. (1974). “Belief and the Basis of Meaning,” in [Davidson 1984], pp. 141-154. ____. (1975). “Thought and Talk,” in [Davidson 1984], pp. 155-170. ____. (1984). Inquiries into Truth and Interpretation (New York, NY: Oxford University Press, 1984). DeRose, Keith. (1995). “Solving the Skeptical Problem,” The Philosophical Review, Vol. 104, No. 1 (January, 1995), pp. 1-52. Gettier, Edmund L. (1963). “Is Justified True Belief Knowledge,” Analysis, Vol. 23, No. 6 (June, 1963), pp. 121-123. Goldman, Alvin L. (1967). “A Causal Theory of Knowing,” The Journal of Philosophy, Vol. 64, No. 12 (June, 1967), pp. 357-372. ____. (1976). “Discrimination and Perceptual Knowledge,” The Journal of Philosophy, Vol. 73, No. 20 (November, 1976), pp. 771-791. Greene, Joshua D., R. Brian Sommerville, Leigh E. Nystrom, John M. Darley and Jonathan D. Cohen. (2001). “An fMRI investigation of Emotional Engagement in Moral Judgement,” Science, Vol. 293, No. 5537 (September 14, 2001), pp. 2105-2108. Kim, Jaegwon. (1988). “What Is ‘Naturalized Epistemology’?” Philosophical Perspectives, Vol. 2 (1988), pp. 381-405. Kornblith, Hilary. (1980). “Beyond Foundationalism and the Coherence Theory,” The Journal of Philosophy, Vol. 77, No. 10 (October, 1980), pp. 597-612. ____. (1994). “What Is Naturalistic Epistemology?” in [Kornblith (ed.) 1994], pp. 114. ____. (2002). Knowledge and Its Place in Nature (New York, NY: Oxford University Press, 2002). ____. (ed.) (1994). Naturalizing Epistemology, 2nd edition (Cambridge, MA: The MIT Press, 1994). Nozick, Robert. (1981). Philosophical Explanations (Cambridge, MA: Harvard University Press, 1981). Quine, W. V. (1968). “Epistemology Naturalized,” reprinted in [Quine 1969], pp. 6990. ____. (1969). Ontological Relativity and Other Essays (New York, NY: Columbia University Press, 1969). ____. (1992). Pursuit of Truth (Cambridge, MA: Harvard University Press, 1992). Sosa, Ernest. (1999). “How to Defeat Opposition to Moore,” Philosophical Perspectives, Vol. 13 (1999), pp. 141-153. Williams, Bernard. (2005). Descartes: The Project of Pure Enquiry, revised edition (New York, NY: Routledge Publishing Ltd., 2005). Williamson, Timothy. (2000). Knowledge and Its Limits (New York, NY: Oxford University Press, 2000). 10