Pramāṇa theory
from 200 BCE The Nyāya Sūtra; the debate runs for two millennia
Before asking whether a claim is true, ask which recognised means of knowing it came through — and whether that means was operating properly.
- Active
- from 200 BCE
- Where
- North India
- Also
- pramāṇa
- Status
- In use
- Tradition
- Non-Western procedures
What it is
A pramāṇa is a means by which knowledge is acquired. Indian philosophy makes the list of admissible means the first question rather than a late one: perception, inference, comparison and testimony on the Nyāya account, with other schools admitting fewer or more.
The method is to sort a claim by its source before assessing it. Each means has its own conditions for operating correctly, and most disputes turn out to be about whether one of those conditions was met rather than about the claim itself.
How it works
Identify the claim. Determine which means it arrived through. Check that means against its own conditions — perception requires unimpaired contact, inference requires a reliable pervasion, testimony requires a competent and honest speaker. Where a claim is defended by more than one means, check each separately.
The schools differ on the list itself, and that disagreement is substantive rather than terminological: admitting testimony as a means, as Nyāya does and Cārvāka does not, changes what can be known at all.
The moves, in order
- Identify the claim
- Determine which means of knowledge it came through
- Check that means against its own conditions of correct operation
- Where several means are invoked, check each separately
- Accept the claim only so far as some means supports it
The procedure
Proposition Three of the four grounds for the boundary claim are the same ground.
- The claim is that the fence line is the boundary. Four grounds are offered.
- First: the surveyor pegged it there in 1994. This is testimony — it rests on the surveyor's competence and honesty, and it can be checked by the conditions testimony requires: the field notes exist and the surveyor was registered.
- Second: the fence has been there ever since. This is perception, and it is sound as far as it goes — but what is perceived is a fence, and that the fence marks the boundary is not perceived.
- Third: both parties have always treated it as the boundary. This is inference from behaviour to belief, and the pervasion it needs — that people treat as a boundary only what is one — plainly does not hold.
- Fourth: the neighbouring lots are all fenced on their survey lines. This is comparison, and it establishes a likeness, not this lot's line.
Therefore Only the first ground reaches the boundary itself; the other three reach the fence, the parties' beliefs and the neighbours' lots respectively — so the case rests entirely on the 1994 field notes and stands or falls with them.
Sorting the grounds by means of knowledge did the work. Nothing was refuted; three of the four were simply shown not to be about the thing in dispute.
What it is good for
Any dispute where several grounds are offered for one conclusion. Sorting by source reveals how many independent grounds there really are, which is usually fewer than the list suggests.
Where it fails
The list itself cannot be established by the method. Asking which pramāṇa establishes the list of pramāṇas either runs in a circle or regresses, and the Indian tradition argued about this for centuries without resolution — it is the same difficulty the verification principle ran into, arrived at two thousand years earlier.
The conditions are also stated in terms that need judgement. Whether a speaker is competent, whether a pervasion is reliable, whether perception was unimpaired — each of these is a further question the framework does not settle.
Common misconceptions
That it is a theory of justification in the Western sense. It is closer to a theory of sources: the question is what produced the cognition, not what reasons the knower can give for it.
Where it came from
The Nyāya Sūtra of around the second century, though the vocabulary is older. Every subsequent school defines itself partly by which pramāṇas it admits.
What became of it
Continuously developed for two thousand years, and now studied in comparative epistemology alongside Western reliabilism, which it anticipates in important respects.
Lasting influence
It structures the whole of classical Indian philosophy — a school's list of pramāṇas determines what it can argue for.
Why it still matters
Alive as a live tradition and increasingly cited in analytic epistemology.
Put another way
Asking not whether the reading is right but which instrument it came off.
Texts
Gautama, Nyāya Sūtra
The founding text. Book I sets out the four means and the conditions on each.
Where to start
Matilal's Perception is the standard bridge for readers coming from Western epistemology.
The surprising bit
The disagreement between schools is not about how many sources of knowledge there happen to be but about what may count as one, and admitting testimony makes the difference between a philosophy that can use scripture and one that cannot.
In their own words
The means of valid knowledge are perception, inference, comparison and testimony
Who held it
Bimal Krishna Matilal
Connections
- Parallel to
- Verification
The verdict
The right first question about any claim, resting on a list the method cannot itself justify.