Who else wants to read an Essay on the Carvaka Epistemology?

The entire philosophy of the Carvakas may be said to depend logically on their epistemology or the theory of knowledge. The problems of epistemology are: How far can we know reality?

How does knowledge originate and develop? This last question involves the problem: what are the different sources of knowledge? This problem forms one of the chief topics of Indian epistemology.

Knowledge of reality or valid cognition is called prama and the source of such knowledge is called pramana. The Carvaka holds that perception is the only pramana or dependable source of knowledge.

For establishing this position, he criticises the possibility of other sources of knowledge like inference and testimony which are regarded as valid pramanas by many philosophers.

1. Inference is not certain

If inference is to be regarded as a pramana, it must yield knowledge about which we can have no doubt and which must be true to reality.

But inference cannot fulfil these conditions, because when we infer, for example, the existence of fire in a mountain from the perception of smoke in it.

We take a leap in the dark, from the perceived smoke to the unperceived fire a logician, like the Naiyayika, will perhaps point out that such a leap is justified by the previous knowledge of the invariable concomitance between smoke and fire and that the inference stated more fully would be: all cases of smoke are cases of fire, this (mountain) is a case of smoke, therefore, this is a case of fire.

The Carvaka points out that this contention would be acceptable only if the major premise, stating the invariable relation between the middle term (smoke) and the major (fire), were beyond doubt.

But this invariable relation (vyapti) can be established only if we have knowledge of all cases of smoke and presence of fire.

This, however, is not possible, as we cannot perceive even all the cases of smoke and fire existing now in different parts of the world, to speak nothing of those which existed in the past or will exist in the future.

No invariable, universal relation (vyapti) can, therefore, be established by perception. Neither can it be said to be based on another inference, because it will involve a petitio principii, since the validity of that inference again has to be similarly proved.

Nor can this vyapti be based on the testimony (sabda) of reliable persons (who state that all cases of smoke are cases of fire).

For, the validity of testimony itself requires to be proved by inference. Besides, if inference always depended on testimony, no one could infer anything by himself.

But it may be asked: though it is not possible to perceive all individual cases of smoke and fire, is it not possible to perceive the constant class-characters (samanya) like ‘smokeness’ and ‘fireness’ which must be invariably present in all instances of smoke and fire respectively?

If so, then can we not say that we at least perceive a relation between smokeness and fireness and with its help infer the presence of fire, wherever we perceive smoke?

The Carvaka replies that even if we grant the perception of a relation between smokeness and fireness, we cannot know therefrom any invariable relation between all individual cases of smoke and fire.

To be able to infer a particular fire, we must know that it is inseparably related to the particular smoke perceived.

In fact, it is not possible even to know by perception what ‘smokeness’ or the class-character universally present in all particular instances of smoke is, because we do not perceive all cases of smoke.

What is found has to be universally present in the unperceived ones. The difficulty of passing from particulars to the universal, therefore, remains here as before.

Rut it may be asked: if we do not believe in any fixed universal law underlying the phenomena of the world, how we would explain the uniformities that experienced objects Possess.

Why fire is always experienced to be hot and water to him cool? The Carvaka reply is that it is due to the inherent natures (svabhava) of things that they possess particular characters.

No supernatural principle need be supposed to account for the properties of experienced objects of nature. There is neither any guarantee that uniformity perceived in the past would continue in future.

A modern student of inductive logic would be tempted to ask the Carvaka: ‘But can we not base our knowledge of the invariable relation between smoke and fire on a causal relation between them?’

The Carvaka reply would be that a causal relation, being only a kind of invariable relation, cannot be established by perception owing to the same difficulties.

The Carvaka would further point out that a causal or any other invariable relation cannot be established merely by repeated perception of two things occurring together.

For one must be certain that there is no other unperceived condition (upadhi) on which this relation depends.

For example if a man perceives a number of times fire accompanied by smoke and on another occasion he infers the existence of smoke on the perception of fire, he would be liable to error, because he failed to notice a condition (upadhi), namely, wetness of fuel, on the presence of which alone fire is attended with smoke.

So long as the relation between two phenomena is not proved to be unconditional, it is an uncertain ground for inference.

And un conditionality or absence of conditions cannot be established beyond doubt by perception, as some conditions may always remain hidden and escape notice.

Inference or testimony cannot be used for proving this unconditionality without a petitio principii because its validity also is being questioned here.

It is true that in life we very often act unsuspectingly on inference. But that only shows that we act uncritically on the wrong belief that our inference is true. It is a fact that sometimes our inference comes true and leads to successful results.

But it is also a fact that sometimes inference leads to error as well- Truth is not then an unfailing character of all inferences; it is only an accident, and a separable one, that we find only in some inferences.

Inference cannot be regarded, therefore, as a pramana a sure source of valid cognition.

2. Testimony is not a Safe Source of Knowledge

But can we not regard the testimony of competent persons as a valid and safe source of knowledge? Do we not very often act on knowledge received from authority? The Carvaka replies that testimony consists of words (sabda).

So far as words are heard through our ears, they are perceived. Knowledge of words is, therefore, knowledge through perception and is quite valid.

But insofar as these words suggest or mean things not within our perception, and aim at giving us knowledge of those unperceived objects, they are not free from error and doubt.

Very often we are misled by so called authority. The authority of the Vedas, for example, is held in high esteem by many.

But in reality the Vedas are the works of some cunning priests who earned their living by duping the ignorant and the credulous.

With false hopes and promises the Vedas persuade men to perform Vedic rites, the only tangible benefit of which goes to the priests who officiate and enjoy the emoluments.

But will not out knowledge be extremely limited and practical life sometimes impossible, if we do not accept the words of the experienced and do not depend on expert advice?

The Carvaka reply is that insofar as we depend on any anthority, because we think it to be reliable, the knowledge obtained is really base on inference; because our belief is generated by a mental process hke this: This authority should be accepted because it is reliable, reliable authority should be accepted.

Being based on inference, knowledge derived from verbal testimony or authority as precarious as inference. And as in the case of inference, so often act on knowledge derived from authority on the belief that it is reliable.

Sometimes this belief accidentally to successful results, sometimes it does not. Therefore, authority or testimony cannot be regarded as a safe and valid source of knowledge.

As neither inference nor authority can be proved to be reliable, preception must be regarded as the only valid source of knowledge (pramana).