The Study of Indian Epistemology: Questions of Method—A Reply to Matthew Dasti and Stephen H. Phillips

I would like to thank the editors of Philosophy East and West for courteously asking me if I would like to respond to Matthew Dasti and Stephen Phillips' very thoughtful remarks about the review I wrote of Phillips' translation and commentary on the pratyakṣa chapter of Gaṅgeśa's Tattvacintāmaṇi, prepared in collaboration with N. S. Ramanuja Tatacharya (Phillips and Tatacharya 2004). Let me begin by reaffirming what I said at the beginning of my review, that the book is "a monumental and momentous achievement, one whose importance cannot be understated." I have indeed enormous admiration for the magnitude of their achievement and respect for the contribution they have made through this translation to the field of Navya Nyāya and Indian philosophical studies.

My review focused on the interpretative remarks in Phillips' Introduction, not on the translation itself. In my review, I said that I did not understand how two claims Phillips makes about Gaṅgeśa's account of the notion of a pramāṇa, "knowledge source" (p. 7), could be held together. The first is that "pramāṇas are viewed as natural processes, as part of the universe's causal web" (p. 7), and the second is that "pramāṇa is to be understood as an infallible process" (p. 8). I asked whether infallibilism about knowledge sources is compatible with naturalism in epistemology, and doubted that it was. Dasti and Phillips respond with a defense of the claim that "pramāṇas are factive."

I admit to being initially puzzled by this response. I did not use the term "factive" in my review, and as far as I can see it is only used once in the book's Introduction, when Phillips writes that "Nyāya takes an externalist or factive point of view" (Phillips and Tatacharya 2004, p. 8). The term "factive," according to the dictionary, is used with reference to verbs that assign the status of a fact to their clausal object—for example, knows or regrets. It has been used in recent epistemological literature by Timothy Williamson to refer to truth-entailing propositional attitudes: "Knowing is in that sense a factive attitude; one knows p only if p is true" (Williamson 2000, p. 21). I am not aware of it being used with reference to knowledge sources, however, and was not at first very sure what it means to claim that "pramāṇas are factive." If the claim is that verbs like "perceive," "infer," and "inform" are factive, then I, of course, am as firmly committed to the truth of that as anyone. If the claim is that the mental states corresponding to perception, inference, and informative testimony (i.e., pratyakṣa-pramā, anumiti, and śabda-pramā) are factive in Williamson's sense, then naturally I agree with that, too.

Fortunately, Dasti and Phillips do tell us what they mean by their unconventional use of the term: they say that they want to "reassert an understanding of pramāṇa as factive, that is, as knowledge sources that are inerrant." The term pramāṇa is factive in the sense that it would be wrong to say "My cognition is pramāṇa-produced but it is false." In this use, then, the word "factive" means "inerrant," and I take it that "inerrant" means the same as the word that was used in the book itself, "infallible." Now, however, we have moved from an uncontroversial point about the success grammar of epistemic verbs to a much more controversial claim about the nature of the causal processes that underlie the activity of knowing. Let me give an example of why I think that this introduces a real philosophical shift. Suppose that we are watching Roger Federer serve in a game of tennis. Sometimes the ball lands just inside or on the line, and the serve is in; sometimes it lands just outside and the ball is out. We can certainly introduce a factive verb to describe the successful service actions, for example "ace," a verb whose grammar implies that to...