9.6 — Nyaya and Vaisheshika: Logic and Atoms

These two are the least glamorous of the six and the most rigorous, and they are where Indian philosophy does the work that Aristotle does in the West (10.4).

Nyaya — the science of reasoning

Nyaya means rule, method, or the correct way of going about something. The system is about how anything is established.

The four valid means of knowledge

Nyaya's foundational question: what actually counts as a way of knowing something?

It accepts four, and rejects others that rival schools accept.

Perception — pratyaksha. Direct sensory contact. And Nyaya is careful: perception must be non-erroneous and definite.

Inference — anumana. Reasoning from what you see to what you do not.

Comparison — upamana. Learning what a word refers to by being told it is like something you know. A gavaya is like a cow. Then you see one and identify it. A minor category and a genuinely distinct one.

And testimony — shabda. Reliable verbal report.

The inclusion of testimony is the interesting one. Nyaya argues that most of what anybody knows comes from being told by somebody trustworthy, and that pretending otherwise is dishonest. Which is obviously correct — you have not personally verified the existence of Antarctica — and which raises the hard question of how you assess reliability. Nyaya's answer is that a reliable speaker is one who knows the matter and intends to state it truly, and much of the tradition's later work is about how you tell.

The five-part inference

And Nyaya's most distinctive contribution: a formal structure for an argument, in five parts.

The standard example, used for two thousand years:

1. The claim. There is fire on the hill.

2. The reason. Because there is smoke.

3. The general rule with an example. Wherever there is smoke there is fire, as in a kitchen.

4. The application. This hill has smoke.

5. The conclusion. Therefore this hill has fire.

Compare Aristotle's three-part syllogism (10.4) — major premise, minor premise, conclusion.

The Indian version has two extra components and they are doing something.

The example is required. You cannot assert a general rule abstractly; you must produce a case where the connection is observed. This makes the system less purely formal and more empirical than the Greek one.

And a negative example is often required too — a case where the reason is absent and the conclusion is absent. Which is, in effect, a demand for a control case.

The fallacies

Nyaya catalogues the ways an inference fails, and the list is sharp.

The reason is too broad — present in cases where the conclusion does not hold. There is fire because there is something.

The reason contradicts the claim.

The reason is present in neither similar nor dissimilar cases, so nothing can be concluded.

The reason is defeated by a stronger counter-argument.

And the reason assumes what is being proved.

This is a real theory of fallacy, developed independently, and it stands comparison with anything before the modern period.

Why Nyaya exists

And the point, which is easy to lose in the technicality.

Nyaya is not logic for its own sake. Its stated purpose is liberation, and its chain is: false knowledge produces attachment and aversion, which produce action, which produces rebirth and suffering.

Break the first link — false knowledge — and the whole thing unravels.

So the most technical school in Indian philosophy is technical because it thinks bad reasoning is the root of human suffering, which is a position worth respecting even if you do not share the metaphysics.

Vaisheshika — the categories and the atoms

The name comes from vishesha, particularity, because the system's distinctive move is that individual things have an irreducible particularity.

The categories

Everything that exists falls into six (later seven) categories.

Substance, quality, activity, universality, particularity, and inherence — plus, added later, absence.

The last two are the interesting ones.

Samavaya, inherence, is the relation between a thing and its qualities — between the cloth and its colour, which are not two separate objects sitting next to each other. A genuinely hard problem, addressed directly.

And abhava, absence, is treated as a real category. The absence of a pot in a room is a fact about the room, and Nyaya argues it can be perceived. This is a distinctive and defensible position and it has close parallels in modern discussions.

The atoms

Vaisheshika holds that the four material elements — earth, water, fire, air — are composed of indivisible atoms, anu.

The reasoning is a proper argument rather than a guess. Divide a material thing. Divide the parts. If division could continue without limit, then a mountain and a mustard seed would both contain infinite parts and would be equal in size, which is absurd. Therefore division terminates.

This is the same argument the Greek atomists made (10.1), and the two traditions appear to have reached it independently.

Atoms are eternal, indivisible, and have no parts. They combine in pairs, then in triads, and larger structures build up from there. The smallest perceptible thing is said to be a combination of three pairs — the size of a mote in a sunbeam.

How close is this to modern atomic theory? Honestly: in the general idea, close. In the specifics, not at all — four kinds of atom corresponding to the classical elements, combining in a fixed hierarchy, with no experimental basis. It is a good philosophical argument and not a scientific result, and Volume III covers what atoms actually turned out to be.

Panini

And one more thing belongs in this chapter, because it is the single most impressive intellectual achievement of ancient India and it is not usually filed as philosophy.

Panini's Ashtadhyayi, probably fourth or fifth century BCE, is a complete generative grammar of Sanskrit in about 4,000 rules.

And "generative" is exact. The rules are ordered, they apply in sequence, later rules override earlier ones under stated conditions, and the system produces every valid form of the language from a set of roots and affixes.

It uses metarules — rules about how the rules apply — and a compressed technical notation of its own.

It is, functionally, a formal system with an ordering discipline and conflict resolution, and it predates anything comparable in the West by well over two thousand years. Linguists and computer scientists study it, and the comparison to a programming language's specification is made routinely and is not a stretch.

This is the achievement that makes the claims in 9.1 about invented ancient science unnecessary. Panini is real, it is verifiable, and it is astonishing.

What to take

Three things.

Testimony is a valid means of knowledge and pretending otherwise is dishonest. Almost everything you know came from somebody telling you, and the real work is assessing reliability, not pretending to independence.

An argument needs an example, and ideally a negative one. The five-part form's insistence on a case where the rule is observed — and a case where it fails — is a better discipline than the abstract syllogism, and it is exactly what a modern reader should demand of a claim.

And bad reasoning is treated as the root of suffering rather than as an intellectual failing. Nyaya's whole rigour exists because it thinks getting things wrong is what keeps people trapped, which is a more serious reason for caring about logic than most.

What to carry from this page

Four ways of knowing: perception, inference, comparison and testimony. Most of what you know is the fourth.

Demand the example. A general rule without an observed case, and without a case where the rule fails, is not established.

The atomic argument was made independently in two civilisations from the same reasoning about divisibility.

And Panini's grammar is a formal generative system from the fourth century BCE. When somebody tells you ancient India had aircraft, tell them about the grammar instead.

Next: 9.7 — the least appealing school in the six, which turns out to have produced the idea at the centre of the Bhagavad Gita.