dravya and avayavin in Nyāya and Vaiśeṣika

As a result of a recent conversation with Roy Perrett, I had a question.
(Incidentally, many of you will know, but some may not, that Roy Perrett’s ‘Introduction to Indian Philosophy’, Cambridge University Press, has been out since the beginning of the year. It will be a very important book for students and teachers of Indian Philosophy classes, for people working in other areas of Philosophy, and for people doing research on Indian Philosophy.)

The question — about Nyāya and Vaiśeṣika ontology:
When we speak of a pot, say (no surprises there as to my choice of example!), can we characterize it just as a substance possessing qualities, i.e. by appeal to two things: substance and qualities (and a relation of inherence between them), or are we dealing with three things? By ‘three things’ I mean the substance, the qualities and the ‘whole’ (avayavin). I.e. is the pot-avayavin firmly distinguished from the pot-dravya or can we equate them? (Roy suggested that we have to distinguish them since the whole is composite and divisible, but the substance is neither.)

4 thoughts on “dravya and avayavin in Nyāya and Vaiśeṣika

  1. Hi Alex. Thanks for your great posts today! (I will respond to your other post later.) I probably don’t understand exactly what Professor Perrett is getting at, but the notion of an avayavin as a third kind of thing (as opposed to merely being one sort of substance) seems to be an addition to the classical theories that departs somewhat significantly from their own statements. The avayavin simply is a substance. And the classical commentators agree that central to its status as a substance is that it is indivisible qua being a distinct sort of substance. Once you start breaking it down, it no longer exists. Or better put, the question of dividing it up makes no sense. This is discussed in the well known nyAyasUtras 2.1.32-36.

    Of course, we must ask which naiyAyikas we are talking about here, but for the early tradition, I’d say with some confidence that avayavins are indeed not divisible. They are made of parts, yes, but as a unification of these parts, which generates a unified cognition of a single thing, they are not actually divisible. In his book on Nyaya’s theory of mind, KK Chakrabarti has a good summary of this argument from the indivisibility of a substance.

    What I’ve found pretty interesting in my recent readings of some of the classical commentators is their explicit defense of the importance of conjunction in debates over avayavins, since conjunction is the guNa which makes avayavins possible. I can post a few translation snippets if curious.

  2. Dear Alex, from the Vaisheshika point of view as I understand it, the question sounds as a confusion of the two levels of concepts. Dravya, guna and samavaya are padarthas, the modes of reality of the first order, while avayava-avayavin, as well as dharma-dharmin are the general relations that may take as its members different padarthas. For example, a pot may be an avayavin because it has its constituting parts – the atoms of earth, which are also dravyas, at the same time it may also be a dharmin as it has its qualities, like color. The pot as a whole is divisible up to its parts, when we look at the pot as a bearer of gunas we must have in mind that these gunas also reside in atoms. This follows from the V. doctrine of pilupakavada which is about changing the color of the pot after the contact of fire with its atoms. If by things you mean ontological categories, I would say that a pot may exemplify four categories – dravya, guna, samanya-vishesha (it is a bearer of potness and it is for this very function that we may call it an avayavin, because potness resides in the whole) and samavaya.

  3. I am grateful to Alex for his kind words about my latest book, and for our very stimulating philosophical conversations. (I am also grateful to him for pointing out an egregious error in my book: on p.205, ll.6-8 “former (combined)” should be amended to read “second”.) I’m not sure, however, that what he quotes me as saying in his latest post brings out quite as perspicuously as it might an issue I personally took to be of particular philosophical interest in our conversations.

    Alex is right that at one point I suggested that one way of distinguishing
    between the avayavin and a simple dravya is that the former is composite
    and the latter impartite. We certainly cannot equate a partite whole
    with such an impartite substance. I don’t believe I’m confused here about the official Nyāya-Vaiśeṣika line on which ontological category wholes occupy, but this does need a little qualification.

    First of all, as Matthew points out, “impartite” here is not the same as “divisible”. Wholes in Nyāya-Vaiśeṣika have parts, but are not thereby divisible into separable (as opposed to distinguishable) parts. Secondly, “composite” might be misleading as a description of a whole: for Nyāya-Vaiśeṣika, a whole is not constituted out of its parts, but rather inheres in its
    parts. Thirdly, Naiyāyikas distinguish between eternal and non-eternal
    substances. The nine dravyas of the standard Nyāya padārtha list are
    all eternal substances and as such are all simples (i.e. partless).
    But Nyāya also allows for non-eternal compound substances (like pots),
    which are non-simples. Whereas no simple substances inhere in
    anything, non-eternal substances (which have simple substances as
    their parts) do inhere in other things – namely, their impartite
    parts. So a pot is a compound substance that inheres in its simple
    parts (its atoms, for instance), but those atoms are simple substances
    and they do not inhere in anything at all. Finally, Naiyāyikas espouse
    a very radical variety of mereological essentialism according to which a compound substance (like all wholes) is destroyed immediately
    one of its parts is eliminated: thus a piece of cloth is a non-simple
    substance which ceases to exist when even a single thread is removed. Hence all non-simple, compound substances (like the pot) are not only non-eternal, they presumably also must be only short-lived.

    So, as a matter of exegesis, it seems to me to be roughly true that, according to the Nyāya-Vaiśeṣika categorical ontology, wholes are supposed to be substances (albeit compound ones). But if, as philosophers, we are not content to rest with mere exegesis, then we need to dig a little deeper and ask whether all this is metaphysically defensible. In our conversations Alex seemed somewhat skeptical that in the end the Nyāya position on wholes could survive serious philosophical scrutiny without having to compromise its official line that wholes are substances. Personally, I was drawn in a different direction: it seems to me that one rather difficult thing the traditional Nyāya mereology requires is the philosophical coherence of the challenging notion of partless structure, i.e. structure that does not require the presence of separable parts. After all, even if partite wholes are not supposed to be divisible into separable parts, they do have structure and that structure is somehow dependent on the parts the whole inheres in, for the removal of any part destroys that whole. But it is not at all clear how we are to make sense of such a metaphysics of structure. And the Nyāya texts on avayavin I know seem to me to be unilluminating on this issue.

    • Dear Professor Perrett, thank you for this clarification, (and for your participation on the blog).Your project is much clearer now, and I do sympathize with your concerns. To me, the strength of the Nyaya position is the view that structured wholes have causal capacities that their parts lack, and thus there is something irreducibly more than just the parts. And further that conjunction is a metaphysically distinct part of reality that allows for structure.

      The weakness is using existing tools like inherence to explain mereology, and most problematically, as you note, a radical sort of idea that any change in parts destroys the whole.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>