________________
INFERENCE
91
that the sun's motion is actually the cause of its being noticed at a new place.28 Jayanta elaborates his point at considerable length, but that is not much important.29 For the real difficulty lies with the point which is the common point of Vātsyāyana and Jayanta and is directed against the Buddhist, viz. that there can obtain between two things an invariable concomitance without being based on some causal consideration or other
After thus offering his first set of interpretations Jayanta comes to the second. Those advocating this second set of interpretations feel dissatisfied with the so must importance attached to causal consideration in connection with the first.30 So on their showing the three types of inference in question have to be understood as follows:
(1) When an invariable concomitance is once established between x and y then a thing similar to x acts as probans for inferring another thing correspondingly similar to y, e.g. when invariable concomitance has been once established between smoke and fire then any particular smoke acts as probans for inferring a concerned particular fire. This is the type of inference pūrvavat (lit. 'like the earlier one').
(2) When a situation can obtain in several alternative ways and it is noticed that in a particular case it does not obtain in any of those ways except one, then it can be inferred that in this case it obtains in this remaining way; e.g. fire can be hay-fire, leaf-fire, wood-fire, cowdung-fire so that if it is noticed that a particular fire is neither hay-fire, nor, leaf-fire, nor wood-fire then it can be inferred that it is cowdung-fire. This is the type of inference seşavat (lit. 'like the remaining one').
(3). If x is somehow similar to xl and x' is invariably accompanied by yl, then x is inferred to be accompanied by y which is correspondingly similar to you even if y is something inherently imperceptible; e.g, an act like cutting etc. is invariably accompanied by (i.e. executed through) an instrument like axe etc. while perception is an act, so that it can be inferred that perception is accompanied by (i.e. executed through) an instrument of the form of a sense-organ even if a sense-organ is something inherently imperceptible. This is the type of inference sāmānyatodrsta (lit. 'observed in a general fashion').
In connection with the second set of interpretations Jayanta considers an objection, and the consideration is somewhat revealing.
