In lieu of an abstract, here is a brief excerpt of the content:

  • Reply to Stephen Phillips
  • Arindam Chakrabarti

Much as I am honored by Stephen Phillips' detailed defense, in the face of my methodological "refutation," of the Nyāya thesis that a raw perception of the qualifier is a necessary causal factor for some (not all) determinate perception of an entity as qualified, I am not fully convinced that my deeper qualms about the very idea of immaculate perception unimpregnated by predicative structure have all been adequately addressed. Instead of a detailed response, which I intend to undertake elsewhere, I wish to underscore the central worry behind my fifth (which Phillips calls "good") and seventh (which Phillips calls "weak") objections.

When I claimed that perception is introduced, taxonomically, as a variety of veridical awareness, I had in mind the opening line of Gaṅgeśa's section on the definition of perception:

sā ca pramā caturvidhā, pratyakṣānumityupamitiśābdā-bhedāt

And such veridical awareness is of four kinds: perception, inferential knowledge, knowledge from similarity, and knowledge from words.1

This seems to be flatly incompatible with the admission of a subclass of perceptions that are not veridical awarenesses!

Why does Phillips say that I am simply mistaken? True, taxonomic difficulties cannot overturn empirical evidence. But recall that there is no direct apperceptive evidence for "raw perception." The sole evidence is a generalization from a whole series of mediate (parokṣa) awarenesses that can happily afford to be instrumentally and crucially caused by other awarenesses. On the other hand, immediacy, which is the hallmark of a perception, consists of not having another awareness in its causal nexus. My seeing of a wall as white is direct, rather than inferential, analogical, or testimonial, to the extent that I did not have to go through any other awareness first in order to arrive at this visual perception. An inductive generalization from other mediate (cognition-generated) cognitions should not have such inexorable probative power over our account of immediate sensory cognition as to force us into admitting a neither-true-nor-false third kind of raw awareness!

Phillips seems to have overlooked the metaphor of "spelling" that I used in this context while giving my second objection. If the step-by-step picture of building up the content of perception that Gaṅgeśa assumes is correct, then each element of the perceptual content of the predicative perception that this is a flower—the universal flowerhood, the actual physical object identified as "this," and qualification as a cementing relation, in this case inherence—would first have to be cognized in the unconnected "raw" manner! Now, even if we somehow swallow the idea of seeing a universal barely and not in relation to any one of its instances, it simply makes no [End Page 114] sense to talk about a bare, raw, unstructured perceptual acquaintance with qualification or inherence! Logically and structurally one could argue that unless you have "got" a, b, and R first, you cannot have the complex aRb. But this is admittedly a process of logical postmortem dissection of a whole piece of direct qualificative perception. And the point of my metaphorical warning—"You need to have 'L' and 'O' and 'D' and 'N' to spell 'London,' but you do not need to spell 'L' or 'O' or 'D' or 'N' first. Single letters simply cannot be spelled!"—was this: there are all sorts of different ways of "getting." You need to "get" the qualifying feature somehow in order to see or hear or taste something as qualified by that feature, but you do not therefore need to see or hear or taste that feature alone first in a bare, raw, unqualified manner! I do believe, like a good Naiyāyika, that high-pitchedness (or fifthness, in the octave) can be heard although it is a sound-universal. But I refuse to admit that before I can hear a sung note as high-pitched or (the fifth) I must somehow hear that universal directly without hearing it in any note!

The central point, once again, is this: all awareness is intentional, according to Nyāya. There are only three kinds of intentional roles that our awarenesses assign to their objects: the role of...