Nyāya on selves (A guest post by Sikander Gilani)

(I received the following post from Sikander Gilani, who is a Philosophy PhD student at Austin. They will surely be happy to read your questions or feedback below.)

Nyāya-Vaiśeṣika holds that there are self substances that bear psychological properties and initiate movements in the bodies they are associated with, which can be animal (including human) or plant bodies. Selves are eternal and immaterial, and incarnate in various bodies over the course of a cycle of rebirths. The Nyāya-sūtra affirms this substantialist position on the self in more than one sūtra, the earliest and most important being sūtra 1.1.10:

icchādveṣaprayatnasukhaduḥkhajñānāni ātmano liṅgam iti |
Desire, aversion, effort, pleasure, pain, and knowledge are the marks of the self. (Vidyābhuṣana, 1913, p. 5)

In classical Indian logic, a mark (liṅga) is a property to which one has epistemic access through any means of knowledge (pramāṇa).* The presence of the mark justifies an inference to the presence of some other property, namely, the property to be proved (sādhya) (Phillips, 2012, pp. 50-71). In this argument the sādhana is a set of psychological qualities (desire, aversion, etc.), and the sādhya is an enduring substantial self. A natural next question to ask is how one can infer from the sādhana to the sādhya. Later philosophers, including Nyāya and Vaiśeṣika commentators, attempt to answer that question.

Argument from diachronic unity
Vātsyāyana is one such commentator. He elaborates upon Nyāya-sūtra 1.1.10 in the form of an argument from diachronic unity, that is, from the fact that there is mental unity over time, to the conclusion that there is a substantial self that persists over time and serves as the locus of such unity. I have reconstructed his argument as follows:

  1. Desire, aversion, etc. require memory and recognition.
  2. Desire, aversion, etc. exist.
  3. So, memory and recognition exist.
  4. If there were no continuing selves having desires, aversions, etc., then at any one moment the content of a cognition would be no more connected with prior cognitions in the same person than with those in other persons.
  5. Memory and recognition amount to a connection between present and past cognitions in the same person that does not exist between cognitions in distinct persons.
  6. Therefore, there are continuing selves having desires, aversions, etc.**

Premise 1 in the argument is supported by the observation that desire, aversion, and effort require memory and recognition so that they can fix on the appropriate objects: for example, we feel aversion to what has caused pain in the past, but in order to do so there must be recognition that it is the same type of object. Pain and pleasure do not require memory – though knowledge of what is painful and pleasurable of course do – which means that this argument, at least as currently stated, does not establish all six qualities as marks of the soul, but only desire, aversion, effort, and knowledge.
Premise 2 is supported by introspection, and is uncontroversial.
Premise 3 is a sub-conclusion that follows deductively from the first two premises.
Premise 4 is partly supported by the observation that we do not remember what happens to others as we remember what happens to us, and we are not averse to or desirous of things on the basis of others’ experiences of pleasure and pain in the way that we are of our own.
In its current form premise 4 is not defensible, as clearly there is no logical necessity of an enduring self for diachronic mental unity. There is also no reason to think – and Vātsyāyana does not provide us with any – that there is some other kind of necessity either, such as necessity due to the laws of nature.
Alternative explanations of memory and recognition are at least logically consistent and are, to the extent of our knowledge, at least as consistent with the laws of nature as substantialist theses. These alternatives are explanations of memory and recognition in terms of a transmission of the relevant information from one entity to another, where the entities are not substantial selves. For example, a contemporary scientific materialist may try to explain memory and recognition in terms of the storage of information in the brain, and the later retrieval of that information through material processes that at no point involve substantial selves.*** The Buddhists explain it in terms of the transmission of information from one fundamental entity (dharma), which exists for a moment (kṣaṇa), to the next dharma. As long as such explanations are possible, the mere fact that diachronic unity exists in individual minds in the form of memory and recognition does not conclusively prove that substantial selves exists.
Can there still be reasons to prefer the substantialist thesis, however, even if the existence of one does not follow from the existence of memory and recognition in the way that Vātsyāyana claims? In my opinion, yes. It is not memory or recognition, but identity, or the sense of self, that should be taken as at least one liṅga of the ātman. In fact, identification may be involved in other types of synchronic and diachronic cognitive unity, but that is not a position I will explicate or defend here. Identification encompasses everything from our sense of ownership over our body and body-parts, to our sense of certain ancestries, personality traits, or situational properties belonging to us. Identification is by its very nature ‘selfal’, that is, selfness infuses the identification experience, and when there is identification with some property or object, there is a representation of the target of identification bearing a certain unique relationship with the self. I would go so far as to say that it is impossible for the self not to exist in some form, given that identification exists. But even if we do not accept this, a presumption of non-skepticism, which is independently plausible or required for any fruitful philosophical practice, can help us here and give us prima facie reason to accept the self exists based on the identification experience. The self cannot be the same as identification, as identification represents that some entity bears a certain relationship with the self. This means that the self exists and is something other than the identification experience.
From here making the case that the self is an individual substance or an object is not difficult. Reflection on experience as well as on the nature of self-talk suggests that selves are particulars and concrete, and that they are not properties, events, states, or processes. For example, selves are not predicated to objects, rather, properties are predicated to selves. We do not talk about selves happening the way that we do with events or processes, rather we talk about selves existing. And we do not talk about selves obtaining in the way that we do about states or facts. Even granting for now the possibility of being systematically deluded in our self-talk, the principle of non-skepticism allows these facts about our experience and everyday speech to be prima facie evidence in favour of the claim that selves are objects.

* The Naiyāyikas consider perception (pratyakṣa), testimony (śabda), inference (anumāna), and analogy (upamāna) to be the four pramāṇas.
** This reconstruction is partly based on the exposition of Vātsyāyana provided by Dasti and Phillips in their The Nyāya-sūtra (2017, pp. 75-77).
*** This point has also been made by Parfit (1984, pp. 223-24) and Siderits (2003, p. 23).

Dasti, M., and Phillips, S. (2017). The Nyāya-sūtra: Selections with Early Commentaries. Hackett: Indianapolis.
Parfit, D. (1984). Reasons and Persons. New York: Oxford University Press.
Phillips, S. (2012). Epistemology in Classical India: The Knowledge-Sources of the Nyāya School. New York: Routledge.
Siderits, M. (2003). Personal Identity and Buddhist Philosophy: Empty Persons. London: Ashgate.
Vidyābhuṣana, M.S.C. (1913). The Nyāya Sūtras of Gotama. Panini office, Bhuvaneshwari Ashram, Bahadurganj.

About elisa freschi

My long-term program is to make "Indian Philosophy" part of "Philosophy". You can follow me also on my personal blog: elisafreschi.com, on Academia, on Amazon, etc.

2 thoughts on “Nyāya on selves (A guest post by Sikander Gilani)

  1. I really enjoyed your post Sikander. I’m wondering if your argument can allow for the Nyāya-Vaiśeṣika tradition to hold a conception of the self that disavows notions of the true self as a being distinct and separate entity (like common understood teachings of Vedanta school hold). It seems like a necessary implication of what you are arguing is that the self has to be conceived as distinct and separate (since that would be in line with our everyday experiences of identification).

    • Hi Sanjay,

      Thanks for your comment. When you talk about the self being distinct, what are you thinking of it as being distinct from? If you’re thinking of selves as being distinct from each other, i.e. if you’re thinking of the view that there are multiple selves rather than a single self (as is held in Advaita Vedanta), then yes, Nyāya-Vaiśeṣika does hold that. My own view is that there is something that binds together all the selves, but there is nevertheless a multiplicity of selves. My view is perhaps closer to the Viśiṣṭādvaita view than the Nyāya-Vaiśeṣika view.

      If you mean the view that selves are a distinct primary substance, i.e. not made out of/the same as other substances, then Nyāya-Vaiśeṣika also holds that, and has various arguments to support it. My own argument from identification does not establish this, however, as for all I’ve established above selves may be material objects. However, I endorse the Naiyāyika argument from the uniqueness of psychological qualities (made in Nyāya-sūtra 3.2.57), and I think there are other arguments that give us good reason to believe that the self is non-material.

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>