Shringaraprakasha 5.p2.18

5.p2.18

Because it makes known a relation possessing its own distinctive quality, which is grasped through a differentiation of actions, and through semantic compatibility, the specific action is [conceived as] the ground for the affix, through 'anu'. Therefore, verbal expression of action is not assumed for it, because the cognition of the action comes from elsewhere. Or rather, the karmapravacanīyas convey a distinction in the form of a relation without adverting to any particular distinction to be attributed to a difference in cause. As in: 'adhi brahmadatte pañcālāḥ' (the Pañcālas are subject to Brahmadatta), 'abhimanyuḥ arjunataḥ prati' (Abhimanyu resembles Arjuna)—the relation of master and property, and resemblance, are understood from the karmapravacanīya. But relations, by delimiting a qualified substratum, make known through their inherent distinction a differentiation of actions. For instance: activities such as governing and collecting taxes from the Pañcālas by Brahmadatta, or perseverance and combat by Abhimanyu resembling Arjuna, are supplied by virtue of the relation in question. Thus this 'anu' delimits the relation between the Saṃhitā and the raining as cause and effect. But because the Saṃhitā has the nature of a specific recitation, and by virtue of its capacity to express an excellence of sound, hearing it is here grasped through the sense, and is not to be understood from the karmapravacanīya. But this is the view: 'anu' has an established capacity to express posteriority, as in 'anurūpādānam'. Since here too it should possess that very meaning, its function as cause is assumed on the basis of its injunction in the sense of cause. Even as posteriority there is no contradiction, since precedence and succession here concern cause and effect. Then this very role as cause, by delimiting a relation, is the ground for differentiating those designated as karmapravacanīyas from [other] nipātas. Why then are 'adhi' and 'pari' in the sense of meaningless, 'su' in the sense of honor, 'ati' in the sense of excess, and 'api' in the senses of word-meaning, possibility, permission, censure, and accumulation, assigned the name karmapravacanīya? For no relation is delimited by these. That is true. However, for the sake of grammatical operations, an explicitly stated karmapravacanīya designation is assigned in the treatise even to items not possessing those characteristics. (5.33)