Praudhamanorama p158

p158

[Regarding the text] "and in dvandve ca": for that prohibition applies to the aggregate, but not to the parts, as will be explained. And the point is that without the rule of extending the name to that which ends in it (tadantavidhi), there is no possibility of the technical name applying to the aggregate. [Objection:] What is the benefit of the technical name applying to that which ends in it? And one cannot say that smāi and the like are the benefit in paramasarvasmai and the like, because that is established merely by the tadantavidhi within the governing section on the base (aṅgādhikāra); therefore he says: "by this." Nor should it be said that here too, since 'prātipadikāt' continues, the tadantavidhi occurs when sarvanāman qualifies it; for there is the statement "a prohibition applies in the rule regarding compounds and affixes." This is the meaning. [Regarding] "and akac": by the word 'and' (ca), the suffix tasil is included in paramasarvataḥ.
jasaḥ śī (Pā. Sū. 7.1.17). [Regarding the statement] "because of having multiple phonemes": not, however, because of having the marker ś, this is the meaning. [Regarding the statement] "after that which ends in a non-a/ā vowel": the meaning is, after a base ending in the vowel a. Therefore, there is no underapplication in yeṣām, keṣām, teṣām, and the like.
[Regarding the statement] "of ām prescribed after a pronoun": however, it is not said 'occurring after a pronoun', because the augment suṭ would undesirably apply in dakṣiṇottarapūrvāṇām. For the prohibition "dvandve ca" (Pā. Sū. 1.1.31) applies to the aggregate, not to the constituent part. And this ām is not prescribed after that [part], but only after the aggregate; hence there is no suṭ. As for the authority for resorting to its being a qualification of what is prescribed, it is the very usage in the Bhāṣya: "dakṣiṇottarapūrvāṇām".
[On the word] "pañcatriṃśat": [it can be analyzed as] *pañca ca triṃśacceti* ("five and thirty"), a dvandva compound. Or, [it can be analyzed as] *pañcādhikāṃstriṃśaditi* ("thirty increased by five"), a tatpuruṣa compound. As for what commentators explaining the older text's statement, "pratyāhāras are forty-three" (*pratyāhārā eracatvāriṃśat*), said—namely: "This is not a tatpuruṣa with elision of the middle member; for if that were so, the original accent of the prior member (*pūrvapadaprakṛtisvara*) prescribed by *rājanyabahuvacanadvandve'ndhakavṛṣṇiṣu* (Pā. 6.2.34) [and] *saṃkhyā* (Pā. 6.2.35) would not apply. Therefore, this is only a samāhāra-dvandva compound; and the neuter gender does not occur in forms like *ekaviṃśati* because gender is dependent on worldly usage"—that is questionable. For it is an accepted consequence that there is a difference in accent according to the difference in compound. And accordingly, explaining the vārttika on the rule *bahuvrīhau saṃkhyeye* (Pā. 2.2.25)—"Addition of a numeral tatpuruṣa for the sake of *nistriṃśa* etc., since elsewhere there is elision of the additional member"—the Bhāṣyakāra states: "*ekaviṃśatiḥ* means twenty increased by one (*ekādhikā viṃśatiḥ*); [likewise] *dvāviṃśatiḥ*." [On the words] "hence alone" (*ata eva*): the meaning is, because for two [referents], only the dual is proper.