Praudhamanorama 60_366

60_366

Nor should one say: let there be the application of a nominal case ending conveying objecthood and so forth from that which presents the qualificand in the parallel case, and from that which presents the qualifier in the case illustrated, because verbal roots and finite verbs are not nominal stems (prātipadika). For the Naiyāyikas, however, this fault would apply. For in that case, just as one says 'dhāvantaṃ mṛgaṃ paśya' ('see the running deer'), one might also say 'dhāvati mṛgam'. Nor should one say that because [the affixes] śatṛ and śānac are obligatory in coreferential relation with words not ending in the first case ending, such a usage does not occur; for even so, the non-occurrence of the second case ending would be difficult to explain. If it is objected: in that case there is definitely a split of sentences (vākyabheda), because in 'paśya' one supplies the object 'tam' ('him'); and as for 'pacati bhavati' and the like, it is simply undesirable in the sense of 'pāko bhavati' ('cooking takes place')—[we reply:] do not say so. For supplying words involves excess weight (gaurava), and that lacks authoritative support because it contradicts the words of the Sages; and your mere personal disinclination is of no consequence. Moreover, in 'apakṣīd devadatto 'vehi' ('know that Devadatta has cooked') and the like, there is no possibility at all of affixes like śatṛ. And the single-sentence status is agreed upon by all, including the Mahābhāṣya and the vārttikas. Nor can one say that in the case of a finite verb, a cognition of another action qualified by an action simply does not arise; for such a cognition is unavoidable for those who are properly trained in that very way. Nor can one say that correctness is not intended in such a meaning, because it is accepted by the very authorities who set out to teach correctness, and because it is necessary in forms like 'purīm avaskanda' ('he raided the city'), which are derived under the heading of relationship between verbal roots. If it is objected: on our established view, if even an action expressed by a finite verb were accepted to have agenthood or objecthood, it would assume the nature of a substantial entity (asattva)—[we say:] no; for that nature is established merely by the absence of gender and number and of kārakas other than agent and object, or because objecthood is not admitted to be a qualifying attribute (prakāra). This is elaborated elsewhere.
On 'abhihite tu' ('when expressed, however'): If it is objected—in that case, in 'pakvam odanaṃ bhuṅkte' ('he eats cooked rice') also, the second case ending would not occur—[we say:] do not say so; for here there are two object capacities determined respectively by the root pac ('cook') and the root bhuj ('eat'); and the second case ending occurs taking into account the unexpressed capacity determined by the principal action. And it has been stated by Bhartṛhari: