________________
56
INDIAN LOGIC
motion is posited in the sun, and so on and so forth, the point being that a cause cannot produce the effect concerned unless it possesses the 'capacity to produce this effect; (that a capacity' thus posited is not open to perception is understood but from this very understanding it follows that it is not open to inference either, an inference always requiring an invariable concomitance established on the basis of perception).' (2). The fifth sub-type is illustrated by the case where the knowledge that a certain person (named Caitra, say) is alive and yet absent inside his house remains unaccounted for unless it is posited that he is present somewhere outside his house; (this is a case where a piece of knowledge had through the pramāna called "abhava (=absence)' remains unaccounted for unless another piece of knowledge is posited.)3 (3) The sixth sub-type is illustrated by the case where the heard sentence 'fat Devadatta does not eat during day time' necessitates the positing of the sentence Devadatta eats during night-time'; (this is a case where a piece of knowledge had through verbal testimony remains unaccounted for unless another piece of knowledge is posited).* Of these three groups the first is essentially described by what has been just said about it, but the remaining two are further elaborated at considerable length. They might be taken up one by one. Thus it is argued in various ways that the knowledge that Caitra is alive and yet absent inside his house necessitating the knowledge that he is present somewhere outside his house is not a case of inference. First it is contended that in this case it is impossible to point out as to what can be the locus-of-inference. what the probans, what the probandum (various alternatives are considered and rejected)." As a matter of fact, the difficulty thus urged is flimsy; for one can eassily infer: *x is present somewhere outside his house, because x is alive and yet absent inside his house.' The real difficulty is that here the concerned relation of invariable concomitance is not an ordinary one, that is, not one based on observation but one based on logical principle, and when the Mimāmsaka comes to that point he actually says that he is ready to treat the present case as a case of inference provided it is conceded that the concerned