...Previous
To begin with, we shall point out some basic differences that the Nyaya theory of inference has from the theory of inference in the western formal logic.
In the western formal logic, validity is conceived as something which is independent of truth. On a theory of pure formal logic an inference can be valid without yielding a true conclusion and an inference can yield a true conclusion without being valid.
Even when a semantical concept of validity is used in formal logic, the formal logic is not concerned with the actual truth or falsity of the premise or conclusion. It recognizes merely that a false conclusion cannot follow validly from true premises. So according to the theory of the western formal logic, if the premises are true and the rules of inference are valid then the conclusion must be true, or that, if the conclusion is true then the inference which led to the conclusion must be a valid one.
The Nyaya theory of inference on the other hand rests on a notion of validity on which validity without truth is impossible. According to the Nyaya theory a valid inference is sure to yield a true conclusion; and the condition for validity of an inference, on their theory, coincides with the condition for the truth of the premises, so a valid inference contains only true premises. So in the Nyaya theory of inference a valid inference with either false premises or false conclusion is not possible. Thus Nyaya provides a theory of inference which establishes valid inferences to be invariably a source of true belief or knowledge and thus proves itself to be adequate for epistemology.
The Nyaya school admits four pramanas1 or sources of knowledge of which anumana or inference is one. They also admit that the greater part of our knowledge comes through inference.2 Now from the very definition of pramana, i.e., a pramana is the means to true cognition (pramakaranam pramanam),3 it follows that an anumana or inference in order to be a pramana must lead to a true cognition.
According to the Naiyayikas, an inference can necessarily yield a true conclusion if it is a valid one. And an inference is valid only if the hetu or the inferential mark is a legitimate (sat) one. They have mentioned a number of characteristics which a hetu or inferential mark must have if it is to be a legitimate or sat hetu. We shall elaborate that point later. What we should note here is that the two conditions of yielding a true conclusion and having a legitimate inferential mark, must both be fulfilled to make an inference a pramana. And normally none of these two conditions can be fulfilled without the other being fulfilled too. If the conclusion to which an inference leads is false then the inferential mark could not be a legitimate one, and in that case the inference is not a pramana but a pramanabhasa (pseudo pramana). If on the other hand, the inferential mark (hetu) is legitimate then the conclusion drawn on the basis of it is true and the inference would be a pramana.
Before proceeding further with the Nyaya theory of inference, we should discuss briefly a few preliminaries about anumana or inference as well as some terminological matters.
In Indian epistemology as such, an inference is not analysed into premise/s and conclusion. The Indian epistemologists maintain that an inference which is a cognitive process, admits of a causal analysis. In an inference a number of cognitive states function as causal conditions which lead to another cognitive state which is the effect. Now this effect in the relevant case is called anumiti, which is roughly what is normally called conclusion in the western terminology. The causal conditions are all called anumana, though sometimes in a technical sense, only one among those causal conditions is singled out and called anumana.4 This will be explained later.
Next..