Indian philosophy

The big questions

1. How many means of knowledge are there, and is being told something one of them?
2. Is there a self that lasts through change, or only a stream of passing events?
3. Is the world many real things, or one reality appearing as many?
4. Does an effect already exist in its cause before it appears?
5. Can reasoning reach beyond the senses, to God, rebirth or a self after death?
6. Can a claim be true from one standpoint and false from another?
7. What is liberation, and is knowledge enough to reach it?

A traveller walking home at dusk stops short: there is a snake on the path. She looks again and sees a coil of rope. Nothing about the rope changed, and for a moment she was certain. What, then, did she see?

The case of the rope and the snake is a stock example across the classical schools, and their answers divide them. For Nyāya, she had a false cognition: snakehood, a real feature known from earlier sightings of real snakes and revived by memory, was misplaced onto the rope. The Prābhākara school of Mīmāṃsā held that no cognition is false in itself; she saw something, remembered a snake, and failed to register the difference between the two. The Buddhist idealists of Yogācāra took such errors as evidence that the objects of perception are supplied by consciousness. Advaita Vedānta made the case an emblem: as the snake is superimposed on the rope, so the world of many things is superimposed on the one reality, Brahman.

That is Indian philosophy in miniature. It is also called classical Indian philosophy, darśana (“vision”, from a root meaning “to see”), or ānvīkṣikī, “critical inquiry”, the word that classical authors such as Kauṭilya used for philosophy in the Arthaśāstra. It is not one doctrine but a long argument among rival schools, Hindu, Buddhist, Jaina and materialist, which shared a technical vocabulary, read one another closely and were shaped as much by their opponents as by their founders.

What it asks

The organising question of most schools is how we know: what the means of knowledge, the pramāṇas, are, and how many. Perception and inference are accepted almost everywhere; testimony, comparison, presumption and non-perception are disputed. A school’s list commits it to a view of what there is, so knowledge and reality are argued together: whether there is a lasting self, or ātman; whether an effect exists in its cause before it appears; whether the world is many real things or one reality appearing as many.

Behind much of this lies a practical aim. Most schools, the Cārvākas apart, held that living beings are bound in a cycle of rebirth, saṃsāra, that ignorance of what we really are keeps us there, and that the right knowledge frees us; that freedom is mokṣa or nirvāṇa. The aim does not soften the arguments. The Nyāya-sūtra opens with sixteen topics that include doubt, the members of an argument, the fallacies and the kinds of debate, and the later literature on knowledge and logic is highly technical.

The schools and how they are counted

The usual count, settled by convention over centuries, gives six āstika schools that accept the authority of the Vedas: Nyāya (logic and knowledge), Vaiśeṣika (categories and atoms), Sāṃkhya (spirit and matter), Yoga, Mīmāṃsā (the interpretation of the Veda) and Vedānta (the Upaniṣads’ teaching on Brahman). The nāstika schools do not accept that authority: Buddhist philosophy, Jaina philosophy, the materialist Cārvāka or Lokāyata, and the Ājīvikas and Ajñānins, who did not survive.

The scheme is a later classification. The fourteenth-century Sarvadarśanasaṃgraha, attributed to Mādhava, reviews sixteen systems, from the Cārvākas through the Buddhists, Jainas, Śaiva schools and grammarians to Śaṅkara’s Vedānta, which it places last. Some āstika works give the Veda little weight: the Sāṃkhyakārikā ranks the Vedic means of ending suffering below knowledge of the manifest, the unmanifest and the knower. Vedānta itself divided into rival schools, among them Śaṅkara’s Advaita, Rāmānuja’s Viśiṣṭādvaita and Madhva’s Dvaita.

The main debates

The self. Nyāya held that the self is an enduring substance whose qualities (desire, aversion, effort, pleasure, pain and cognition) come and go. Sāṃkhya held that the true person is puruṣa, pure consciousness, distinct from matter, prakṛti, and that there are many puruṣas. Advaita held that there is one self, identical with Brahman; Rāmānuja answered that selves and the world are real, as the body of Brahman. The Buddhists, following the Buddha’s teaching of no-self, and the Cārvākas both denied any such essence: there is the psychophysical complex and nothing beneath it, a view argued in Nāgasena’s chariot.

Causation. Sāṃkhya’s satkāryavāda holds that an effect already exists in its material cause, since what does not exist cannot be produced. Nyāya and Vaiśeṣika replied with asatkāryavāda: the effect is a new beginning. Śaṅkara held that an effect is nothing apart from its cause, but denied that Brahman really transforms into the world: the world is Brahman appearing as many.

Knowledge. The Cārvākas accepted perception alone, arguing that inference rests on universal connections, vyāpti, which no perception can establish. Nyāya and Buddhist logicians turned the challenge into a fallibilism about inference, and Gaṅgeśa pointed out that the sceptic relies on such generalisations whenever he uses words. Nyāya defined testimony as the statement of an āpta, someone who knows the truth and wants to convey it without deception, and Vātsyāyana added that even a foreigner can be one; Vaiśeṣika and the Buddhists reduced testimony to inference. Mīmāṃsā held that the Veda has no author, and that knowledge is certified by itself (svataḥ prāmāṇya); Nyāya held that certification comes from outside the cognition.

Reality. Nāgārjuna argued that nothing has a reality of its own, the doctrine of emptiness; Yogācāra that there is only consciousness; Nyāya, Vaiśeṣika and Mīmāṃsā defended a realism of many things. The Jainas’ anekāntavāda holds that reality is many-sided and that no single statement captures the whole truth, an idea often illustrated by the blind men and the elephant.

A short history

The oldest Vedic hymns go back to the second millennium BCE, and the earliest Upaniṣads, from the late Vedic period before about 500 BCE, ask what the self is and how it stands to Brahman, the ground of everything (the Chāndogya’s lesson of the salt). The Buddha and Mahāvīra, in the sixth or fifth century BCE, led śramaṇa movements of wandering ascetics that rejected Vedic authority.

Over the following centuries several schools gathered their teaching into root texts of terse aphorisms (sūtras): Kaṇāda’s Vaiśeṣika-sūtra and Akṣapāda Gautama’s Nyāya-sūtra, which scholars place somewhere between the sixth century BCE and the second century CE; such dates are disputed. Then came the great age of commentary and cross-school debate: Nāgārjuna (c. 150–250), Vasubandhu, Dignāga and Dharmakīrti among the Buddhists, Kumārila in Mīmāṃsā, Śaṅkara in the eighth century, Rāmānuja and Madhva after him.

In the fourteenth century Gaṅgeśa’s Tattvacintāmaṇi became the founding text of Navya-Nyāya, the “New Logic” school, whose precise technical language was used in poetics, linguistics and law, and which flourished for about four centuries. From the late nineteenth century, Vivekananda and, later, Radhakrishnan presented Vedānta in English to Indian and Western audiences, and K. C. Bhattacharyya used what he called constructive interpretation to examine the classical systems as one examines the problems of present-day philosophy. Gandhi argued for non-violent resistance, and Ambedkar for the political rights and social freedom of Dalits, a case that led him from Hinduism to Buddhism (modern Indian philosophy).

Counterparts in other traditions

The parallels are close enough to be useful. The Cārvāka attack on inference anticipates Hume’s problem of induction. The dispute between Nyāya and the Buddhists over whether testimony is a source of its own or a kind of inference has the same shape as the later Western dispute between Reid, who treated trust in testimony as basic, and Hume, who held that it rests on evidence. The Buddhist no-self has been read, notably by Mark Siderits, as a reductionism about persons of the kind Parfit defended, and it sits naturally beside Hume’s bundle theory. Buddhist philosophy, which began in India, was carried into China, where it met Chinese philosophy, and on to Korea and Japan.

They need care. Karl Potter argued in 1984 that the Sanskrit terms usually translated “knowledge” do not map onto the analysis of knowledge as justified true belief, while others have used Nyāya’s tools to answer the Gettier problem. A comparison helps when it keeps the difference in view.

What it means now

Vedānta, Buddhism, Jainism and Yoga are living traditions, and many readers come to Indian philosophy looking for a way to live. It is also a working field: four of the ten units of the UGC NET philosophy syllabus are Indian, and nine of the twenty topics in Paper I of the UPSC philosophy optional are Indian schools and thinkers. Since the later twentieth century, notably in the work of B. K. Matilal, who held the Spalding Chair of Eastern Religions and Ethics at Oxford from the 1970s until 1991, philosophers have read Nyāya and Buddhist logic alongside epistemology in the analytic style, on epistemic luck, perceptual content and error.

The part closest to my own work is the theory of śabda, knowledge from words, and what it says about knowing from machines (my research note; What Nyāya Can Teach a Language Model).

How to begin

Begin with primary texts that need no training: the Kaṭha Upaniṣad, the Dhammapada, the second chapter of the Gītā. Then read the Cārvāka chapter of the Sarvadarśanasaṃgraha for the dissent, and the Chāndogya and the Yoga Sūtras before Śaṅkara. The reading path below sets them in order. For a survey, Dasgupta’s history is free; for current scholarship, the Stanford Encyclopedia’s entry on classical Indian epistemology is the place to go next. Leave Navya-Nyāya until the end.