Praudhamanorama p176

p176

On "iti pakṣe tu": By this he hints at the optionality of the ablative compound. To explain: in the Bhāṣya, three purposes were mentioned—bebhiditā, māthitikaḥ, apīpacat. And that is established otherwise. In "ekāca upadeśe", precisely because [the condition] qualifies what is enjoined, the prohibition of iṭ does not apply in bebhiditā. In the analytical paraphrase 'churned milk is his merchandise', when ṭhak [is added], when ika is substituted for ṭha, [and] when elision of a [occurs] by "yasyeti ca", the substitution of ka by "isusuktāntātkaḥ" will indeed not occur, because there is no ṭha. Nor can it be argued that there is the nature of ṭha by treatment as the original element, because it is an operation conditioned on a single letter (alvidhi); for in "ṭhasyekaḥ", the [vowel] a in both the original element and the substitute is for the sake of pronunciation. Also, apīpacat is established even without treating as the original element the elision of ṇi and the single substitute for caṅ and the final a. On account of association with the rule, jhi gets the substitution of jus from abhyasta only when [belonging to] laṅ. And it should not be said: why should it not be [belonging to] luṅ alone, due to association with sic? For subsequent association is stronger than prior association; this is the general direction.
On "pūrvatrāsiddhe na": And we have stated that this is established by reasoning; for since the Tripādī is suspended (asiddha), when an operation belonging there is to be performed, the extension rule "acaḥ parasmin" does not apply. On "tasya": of that, namely of 'pūrvatrāsiddhe'. And thus the meaning is that treatment as the original element must be stated in three places: in elision of the first member of a conjunct consonant: cakryatra; in substitution of l: nigālyate; in substitution of ṇ: māṣavapanī. Here, because the elision of a by "yasyeti ca" is treated as the original element, the letter n does not stand at the end of a nominal stem, and so substitution of ṇ does not occur.
As for what they say: 'Even if there is somehow no treatment as the original element, in yūṣṇaḥ and the like "raṣābhyām" is of no use, because it is established by retroflexion (ṣṭutva); and thus the mention of ṣ is for the sake of subsequent rules—this is established in the source text; and therefore this was advanced as a hasty statement'—it is that very claim which is a hasty statement. To explain: the mention of ṣ stands here even for the sake of subsequent rules. And thus, in yūṣṇaḥ, puṣṇāti, and the like, by what could the application of the substitution of ṇ by "raṣābhyām" be prevented? If one says: 'by retroflexion', no, because that is suspended (asiddha). Nor can it be said that it is not suspended (siddha) on the ground that the statement would otherwise be vacuous, because it has purpose in puṣṭiḥ and the like. If one says that its part, namely 'ṇa in connection with ṣ', is vacuous, no, because that is not stated separately; and also because the fact that the statement 'after ṣ' is not vacuous here is equal. For in the source text what is intended is only that without the mention of ṣ here, the target form does not fail to be established, but not that "raṣābhyām" does not apply in puṣṇāti and the like; this is the general direction.
On "nalopaḥ prātipadikāntasya" (Pā. Sū. 8.2.7): On "luptaṣaṣṭhīke": [it is a genitive whose ending is elided] because it belongs to a sūtra, but not as the prior member of two genitive tatpuruṣa compounds, since a compound is impossible because of dependence, and because the nominal stem qualifies the inflected word; this is the purport. On "tadantasya": it is a genitive tatpuruṣa; the meaning is: 'of the letter n that is the end of that'. And it has been said by Haradatta: 'The word na, which lacks a case ending, is the qualificand of antasya; likewise, the word prātipadika qualifies pada.' As for what was said by the older grammarian: 'of the n that is at the end of a nominal stem, [and] at the end of an inflected word'; and as for what has been commented: 'antasya is connected with both'—that is burdened with heavy assumption, because it entails the repetition of antasya and results in two syntactically non-concordant genitives. If, however, the inflected word is qualified by na just as by prātipadika through an identical relation, then, by the rule of the term ending therewith through a qualifier, it would be possible to dispense with the Sūtrakāra's mention of anta.