Response to Roy W. Perrett's Review of Classical Indian Philosophy of Mind: The Nyaya Dualist Tradition

I am thankful to Roy W. Perrett for a substantial review of my book Classical Indian Philosophy of Mind (Philosophy East and West 52 [1] [January 2002]). I certainly appreciate the many positive things he has to say about the book. In the short space available here, however, I confine myself to addressing very briefly some critical points raised by him. I hope that the discussion that follows will generate more light than heat.

Perrett has questioned my observation that the Nyāya theory of causation is Hume-like. He argues that a distinctly non-Humean type of (non-logical) necessity seems to be implied by the Nyāya view that a cause is an invariable (niyata) and non-superfluous (ananyathāsiddha) antecedent of an effect. It is unclear what Perrett means by non-logical necessity and what justifies attributing this to a causal condition as understood in the Nyāya. Still, I shall offer the following remarks.

First, let us look at the meaning of being invariable. Something invariable (niyata) is not variable (aniyata) or deviant (vyabhicārin). So invariant means non-deviant. What is deviation? If something is present with something some of the time and without that something at other times, it is deviant or variable. For example, some animals have horns and some do not. Being horned, then, is a variable and deviant accompaniment of being an animal. In such a case the claim that there is invariable accompaniment is false. It is false because what is claimed to be invariant is the negatum (pratiyogin) of an absolute absence (atyantābhāva) in the locus of the other thing. Something invariable is different from that. It is not the negatum of an absolute absence in the locus of something (tadvannisṣṭhātyantābhāvāpratiyogin). Thus, invariance boils down to not being the negatum of an absolute absence in the locus of something. Something invariant is also called a pervader (vyāpaka), and this analysis is also offered for the concept of a pervader: a pervader, too, is something that is not the negatum of an absolute absence in the locus of something. Three key notions involved here are difference, the negatum of an absolute absence, and locus. Necessity is not implied in any of these. For example, a book is different from a table. It is not implied that a book is necessarily different from a table; in fact, a book could serve as a table in some situations. Similarly, if there is absence of the book on the table, the book is the negatum of an absolute absence belonging to the table. This does not imply necessary exclusion; the book may be put on the table. Again, necessity is not implied in the concept of a locus. For example, when there is a book on a table, the latter is the locus of the former. But the relation is not necessary and ends if the book is removed from the table. I do not imply that these three concepts are irrelevant for understanding necessity. On the contrary, these concepts are useful for unpacking the nature of necessity where there is necessity. But the three notions do not necessarily imply necessity either severally or jointly.

Now let us look at non-superfluity. Non-superfluity is based on economy (lāghava). Three kinds of economy may be distinguished. First, there is economy in constitution (śarīra). Of two antecedents the one that is analyzable into fewer constituents is more economical. For example, for a substance to be perceived (according to the Nyāya view) it should have intermediate magnitude and should also be made of many substances. Although both are necessary conditions, only the former is accepted as a causal condition of the perception of a substance and not the latter, on the ground of economy in constitution. Second, there is economy in relation (sambandha). Of two antecedents the one that is more directly related to the effect is more economical. For example, a wheel is accepted as a causal condition of...