As pointed out earlier, an inferential cognition, i.e., the conclusion of an inference, is true if the paramarsa, which causes it, is true. A paramarsa is the cognition to the effect that the paksa (the subject of inference) is characterised by the hetu (the inferential mark) which is pervaded by the sadhya (the inferable property). The paramarsa is true if and only if paksadharmatajnana, i.e., the cognition to the effect that the inferential mark characterizes the subject of inference, and vyaptijnana, i.e., the cognition to the effect that the inferential mark is pervaded by the inferable property, are true. These are again true if and only if (i) the inferential mark exists in the subject of inference (paksa-sattva) and (ii) the inferential mark exists in a thing (other than the subject of inference) which is known to be a locus of the inferable property (sapaksa-sattva) and/or the inferential mark does not exist in anything which is known to be characterised by the absence of the inferable property (vipaksasattva). The characteristic of paksa-sattva is the truth-condition of  paksa-dharmatajnana. The characteristics sapaksa-sattva and vipaksa-sattva, on the other hand, are disjunctively the truth conditions of vyaptijnana. And since the truth of paksadharmatajnana and vyaptijnana are necessary conditions for the truth of  paramarsa, the truth of the inferential cognition (the conclusion of the inference) ultimately depends on the three above mentioned characteristics of the hetu. If the hetu of an inference possesses those three characteristics, the inference is sure to yield a true conclusion.

The last two characteristics of a legitimate hetu, i.e., abadhitatva  and  asatpratipaksitatva  seem to have a direct relevance to the truth of the conclusion or anumiti. Like any other cognition an inferential cognition can be regarded as true according to the Naiyayikas if it corresponds to the objective situation. For example, if we infer that a particular hill possesses fire then that inferential cognition can be regarded as true if that particular hill actually possesses fire. Now this is not possible unless the hetu is abadhita and asatpratipaksita. If a hetu was badhita then the objective situation would be such that the inference drawn on the basis of such a hetu could not correspond with it. To be badhita, is to be contradicted, but when a hetu is said to be badhita, what is contradicted is not the hetu itself but the proposition which is to be drawn on its basis. That means, the contradictory of the conclusion is true and it corresponds with the objective situation, hence the conclusion from a badhita hetu cannot correspond to the objective situation. In our example, the hetu, i.e., the smoke would be badhita if that particular hill did not possess fire. That means, it is the contradictory of the proposed conclusion which corresponds to the objective situation and is thus true, and so the proposed conclusion itself, i.e., ‘That hill possesses fire’ cannot be true.

If a hetu is satpratipaksita then there is no certainty that the conclusion drawn from it would be true. A hetu is said to be satpratipaksita if another hetu which is capable of proving the opposite conclusion can be found. In such a case only one of the contradictory conclusion can be true, so there is the possibility of the conclusion, which is drawn on the basis of the first hetu, being false. So in order to ensure the truth of the conclusion, a hetu must possess the characteristics of asatpratipaksitatva.

It may be pointed out here that if the inferer entertains a belief that the hetu upon which he is going to draw an inference is not abadhita and asatpratipaksita then he cannot proceed with his inference. That means, if the inferer believes rightly or wrongly that the hetu he is employing is badhita and/or satpratipaksita then he would refrain from drawing the inference. If a hetu is believed to be badhita it would be logically inconsistent to use such a hetu for drawing an inference. For an attempt to draw a conclusion which is already contradicted, on the basis of some evidence (pramana) cannot be rationally justified. Similarly, to proceed to draw a conclusion when one is aware of the presence of some premise (hetu) tending to yield just the opposite conclusion is not rationally justifiable. One can proceed with the inference only after being convinced that the premise (hetu) which has been alleged to prove the opposite conclusion is incapable of proving it. That is the reason why we have taken the absence of the belief that the hetu is badhita and also the absence of the belief that the hetu is satpratipaksita to be pre-conditions for an inference.

We may now sum up the points made so far about the Nyaya theory of inference.

First, the Nyaya theory of inference upholds a different notion of validity, such that an inference cannot be valid without leading to a conclusion which is actually true. A purely formal concept of validity is not found in Indian logic. But still if we look for a corresponding concept in the Nyaya theory of inference we may say that on the Nyaya theory any inference that can be expresses in the following schema may be treated as formally valid:

                                The paksa has the hetu,
                                The hetu is pervaded by the sadhya,
            Therefore, The paksa has the sadhya,

provided the hetu in the premises 1 and 2 stand for things belonging to the same class, and the sadhya in the premise 2 and in the conclusion stand for things belonging to the same class. But the Indian logicians are not interested in this sort of validity, it would be a trivial concept for them. Any possible inference in their system of logic belongs to this form, even fallacious arguments have the same form. Only an inference which belongs to this form and has a legitimate (sat) hetu is valid in their sense of the term and leads invariably to a true conclusion.

Secondly, it specifies three sets of conditions: (i) conditions for the very possibility of an inference or pre-conditions  (ii) causal conditions of an inference; and (iii) conditions for the validity of an inference.

Thirdly, the conditions for validity, as formulated in the Nyaya theory of inference, are truth-ensuring. So on this theory every valid inference has a true conclusion as well as true premises. The Nyaya philosophers’ argument in brief is as follows:

An inference cannot be valid unless its hetu is legitimate.

An inferential mark cannot be legitimate unless it satisfies the five (four in some cases) conditions for legitimacy enumerated by the Nyaya philosophers, or, in other words unless the hetu possesses the five characteristics of a legitimate hetu.

If the hetu or the inferential mark is legitimate (sat), then the premises are true. And the premises being true, the conclusion which follows from them must also be true.

Thus with the fulfilment of the condition of the legitimacy of a hetu, which is the same as the condition for the validity of an inference, the truth of the premises as well as the conclusion is guaranteed.

 

Next..