PREMABASESiddhānta-kaumudī
Prakaraṇa 52 · SK 2554–2562

तिङन्तक्र्यादिप्रकरणम्

tiṅantakryādiprakaraṇam · 9 sūtras
📖 This chapter in the textbook →outline ↗
show
SK 2554 · A 3.1.81V
क्र्यादिभ्यः श्ना ॥ २५५४ ॥
kryādibhyaḥ śnā
kryādibhyaḥ śnā — After krī and the rest [the kryādi roots], śnā [comes] [before a sārvadhātuka denoting the agent (kartari … sārvadhātuke — carried forward (anuvṛtti); kartari from kartari śap, SK 2167), in the root's own sense; an exception (apavāda) to śap].

Rule. After krī and the other roots of its class, the affix śnā comes before a sārvadhātuka affix denoting the agent. It keeps the root's own meaning. It is an exception (apavāda) to śap.

"Denoting the agent" (kartari) is carried from SK 2167.

Example.

krī + ti → śnā comes, not śap

→ krī + nā + ti (the ś of śnā is a marker)

→ nā is apit, so it counts as ṅit (SK 2234); the ī of krī takes no guṇa

→ krīṇāti "he buys".

krī + nā + tas → the ending tas begins with a consonant and is ṅit

→ the ā of nā becomes ī (SK 2497)

→ krīṇītaḥ "the two buy".

krī + nā + jhi → jh becomes ant first (SK 2169)

→ krī + nā + anti

→ the ending now begins with a vowel, so the ā of nā is lost (SK 2483)

→ krīṇanti "they buy".

Applies when / blocked by. Why does ant come before the ī? Bhaṭṭoji gives three reasons. The ant rule is later in the order of rules (para). It is nitya: it applies whether or not ī has been made. And it is antaraṅga. The same holds in the ātmanepada. There at replaces the jh of jha (SK 2258) before the ī can come. So the ātmanepada gives krīṇīte, krīṇāte, krīṇate.

The roots of this class take śnā only in the sārvadhātuka. Elsewhere they behave like other roots:

  • liṭ: cikrāya "he bought". Dual cikriyatuḥ: before the kit ending the ī becomes iy.
  • liṭ, second person singular: cikrayitha or cikretha. The iṭ is optional before thal by the Bhāradvāja restriction (SK 2296). The printed text reads cikriyitha; scholars discuss this.
  • liṭ, first person dual: cikriyiva, with iṭ by the kṛādi restriction (SK 2293). Ātmanepada cikriyiṣe.
  • luṭ kretā, lṛṭ kreṣyati, with guṇa.
  • āśīrliṅ ātmanepada kreṣīṣṭa.
  • luṅ akraiṣīt, ātmanepada akreṣṭa.

One commentator, the Bālamanoramā, adds the forms with śnā: loṭ krīṇātu, krīṇīhi; laṅ akrīṇāt, akrīṇīta; vidhiliṅ krīṇīyāt, krīṇīta.

Watch out. The ā of śnā has three fates. Before a consonant-initial ṅit ending it becomes ī. Before a vowel-initial one it is lost. Before the pit ending ti it stays: krīṇāti.

The roots that follow krī in the list:

  • Prī "to satisfy" and "to desire": prīṇāti, prīṇīte.
  • Śrī "to cook".
  • Mī "to injure". After pra, the n of nā becomes ṇ by SK 2530: pramīṇāti, pramīṇītaḥ. Where an e or ai would arise, ā comes instead (SK 2508). liṭ mamau, mimyatuḥ, mamitha or mamātha; ātmanepada mimye; luṭ mātā; lṛṭ māsyati; āśīrliṅ mīyāt, māsīṣṭa; luṅ amāsīt, amāsiṣṭām, ātmanepada amāsta.
  • Ṣi "to bind": sināti, sinīte; liṭ siṣāya, siṣye; luṭ setā.
  • Skuñ, glossed āpravaṇa. Bhaṭṭoji's commentators do not explain the gloss. This root leads into the next rule.
Siddhānta-kaumudī
{$ {!1473 डुक्रीञ्!} द्रव्यविनिमये$} ॥ क्रीणाति । ई हल्यघः SK 2497 क्रीणीतः । ईत्वात्पूर्वं झेरन्तादेशः । परत्वान्नित्यत्वादन्तरङ्गत्वाच्च । एवं झस्याद्भावः । ततः श्नाभ्यस्तयोः - SK 2483 इत्याल्लोपः । क्रीणन्ति । क्रीणीते । क्रीणाते । क्रीणाते । चिक्राय । चिक्रियतुः । चिक्रियिथ । चिक्रेथ । चिक्रियिव । चिक्रियिषे । क्रेता । क्रेष्यति । क्रीतात् । क्रेषीष्ट । अक्रैषीत् । अक्रेष्ट ।{$ {!1474 प्रीञ्!} तर्पणे कान्तौ च$} । कान्तिः कामना । प्रीणाति । प्रीणीते ।{$ {!1475 श्रीञ्!} पाके$} ।{$ {!1476 मीञ्!} हिंसायाम्$} । हिनुमीना SK 2530 प्रमीणाति । प्रमीणीतः । मीनातिमिनोति SK 2508 इत्येज्विषये आत्वम् । ममौ । मिम्यतुः । ममिथ । ममाथ । मिम्ये । माता । मास्यति । मीयात् । मासीष्ट । अमासीत् । अमासिष्टाम् । अमास्त ।{$ {!1477 षिञ्!} बन्धने$} । सिनाति । सिनीते । सिषाय । सिष्ये । सेता ।{$ {!1478 स्कुञ्!} आप्रवणे$} ॥

(1473) ḍukrīñ — in the exchange of goods (dravyavinimaya). krīṇāti [laṭ, prathama-puruṣa singular; śnā, as an apit sārvadhātuka, is treated as ṅit, so the ī takes no guṇa]. By ī haly aghoḥ (SK 2497) [ī for the ā of śnā before a hal-initial kṅit sārvadhātuka]: krīṇītaḥ [prathama-puruṣa dual]. [With krī + ṇā + jhi standing,] the substitution of ant for jhi [by jho 'ntaḥ (SK 2169)] comes before the ī-substitution — because it is later (paratva), because it is nitya (it applies whether or not ī has been made), and because it is antaraṅga. Likewise the substitution of at for jha [by ātmanepadeṣv anataḥ (SK 2258), in the ātmanepada, also comes before ī]. After that [i.e. after the ant- and at-substitutions], deletion of the ā [of śnā, before a vowel-initial kṅit sārvadhātuka] by śnābhyastayor ātaḥ (SK 2483): krīṇanti [prathama-puruṣa plural]. krīṇīte, krīṇāte, [krīṇate] [laṭ ātmanepada, prathama-puruṣa singular, dual, plural]. cikrāya [liṭ, prathama-puruṣa singular]. cikriyatuḥ [prathama-puruṣa dual]. cikrayitha, cikretha [madhyama-puruṣa singular, with and without iṭ: iṭ is optional before thal by the restriction of ṛto bhāradvājasya (SK 2296)]. cikriyiva [uttama-puruṣa dual; iṭ by the kṛādi restriction]. cikriyiṣe [ātmanepada, madhyama-puruṣa singular]. kretā [luṭ]. kreṣyati [lṛṭ]. krītāt. kreṣīṣṭa [āśīrliṅ ātmanepada]. akraiṣīt [luṅ parasmaipada]. akreṣṭa [luṅ ātmanepada]. (1474) prīñ — in satisfying (tarpaṇa) and in desiring (kānti). kānti is desire (kāmanā). prīṇāti, prīṇīte. (1475) śrīñ — in cooking. (1476) mīñ — in injuring. [n becomes ṇ by] hinumīnā (SK 2530): pramīṇāti, pramīṇītaḥ [with the prefix pra]. By mīnātiminoti… (SK 2508), ā [is substituted for the root-final] where an ec [vowel] would arise: mamau [liṭ, prathama-puruṣa singular]. mimyatuḥ [prathama-puruṣa dual]. mamitha, mamātha [madhyama-puruṣa singular]. mimye [liṭ ātmanepada]. mātā [luṭ]. māsyati [lṛṭ]. mīyāt [āśīrliṅ parasmaipada]. māsīṣṭa [āśīrliṅ ātmanepada]. amāsīt, amāsiṣṭām [luṅ parasmaipada, prathama-puruṣa singular, dual]. amāsta [luṅ ātmanepada]. (1477) ṣiñ — in binding. sināti, sinīte. siṣāya [liṭ]. siṣye [liṭ ātmanepada]. setā [luṭ]. (1478) skuñ — in āpravaṇa.

Tattvabodhinī
NSP p. 424–425
एवं झस्याद्भाव इति । ‘आत्मनेपदेष्वनतः’ इत्यनेन ॥—सौत्रा इति । नकारोपधा इत्यपि ज्ञेयाः ॥—विष्टभ्नोतीति । ‘अनिदिताम्—’ इति नलोपः । ‘स्तन्भेः’ इति षत्वम्
Bālamanoramā
अथ क्र्यादयः । अथ श्नाविकरणधातवो निरूप्यन्ते । डु क्रीञिति । क्र्यादिभ्यः श्ना - क्र्यादिभ्यः श्ना.कर्त्रर्थे सार्वधातुके परे क्र्यादिभ्यः श्नाप्रत्ययः स्यात्स्वार्थे इत्यर्थः । शपोऽपवादः । क्रीणातीति । श्नाप्रत्ययस्य अपित्सार्वधातुकत्वेन ङित्त्वादीकारस्य न गुण इति भावः । ई हल्यघोरिति । हलादौ क्ङिति सार्वधातुके ईत्त्वमिति भावः । क्री णा झीति स्थितेई हल्यघो॑रिति ईत्वमाशङ्क्य आह — ईत्वात्पूर्वमिति । नित्यत्वादिति । अकृते कृते च ईत्वे अन्तादेशस्य प्रवृत्तेरिति भावः । एवं झस्येति । क्री णा झ इति स्थिते 'आत्मनेपदेष्वनतः' इत्यदादेशोऽपि ईत्त्वात्पूर्वमित्यर्थः । तत इति । अन्तादेशाददादेशाच्च पश्चादित्यर्थः । अजादौ क्ङिति सार्वधातुके श्नाप्रत्ययस्य आल्लोपः । भारद्वाजनियमात्थलि वेडिति मत्वाऽऽह — चिक्रयिथ चिक्रेथेति । चिक्रियिवेति । क्रादिनियमादिडिति भावः । क्रीणातु । क्रीणीहि । अक्रीणात् । अक्रीणीत । क्रीणीयात् । क्रीणीत ।
translator notes
  • commentary BM glosses the sūtra: 'before a sārvadhātuka in the sense of the agent (kartrarthe sārvadhātuke pare), the affix śnā comes after krī and the rest, in the root's own sense (svārthe)'; 'an exception to śap' (śapo 'pavādaḥ). Followed: bracketed in sutra_en.
  • commentary BM on krīṇāti: since the affix śnā is an apit sārvadhātuka, it is ṅit [by sārvadhātukam apit, SK 2234], so the ī takes no guṇa; on ī haly aghoḥ: ī before a hal-initial kṅit sārvadhātuka. Followed: the brackets on krīṇāti and krīṇītaḥ.
  • commentary BM: with 'krī ṇā jhi' standing, apprehending ī by ī haly aghoḥ, Bhaṭṭoji says 'before the ī-substitution'; nityatvāt = because the ant-substitution applies both when ī has not been made and when it has; 'evaṃ jhasya' = with 'krī ṇā jha' standing, the at-substitution by 'ātmanepadeṣv anataḥ' too comes before ī; 'tataḥ' = after the ant- and at-substitutions; ā-deletion of śnā before a vowel-initial kṅit sārvadhātuka. Followed: the bracketed stem, the gloss of nitya, the rule for adbhāva and the condition of ā-deletion.
  • commentary TB (sk.2554.t1, re-filed on 2026-10-05 from the tail of sk.2546.t1, where it followed 'iti tanādayaḥ'): 'evaṃ jhasyādbhāva iti — by ātmanepadeṣv anataḥ'. Followed: the rule named in brackets for the at-substitution.
  • commentary BM on the liṭ forms: taking iṭ before thal to be optional by the Bhāradvāja restriction, Bhaṭṭoji gives 'cikrayitha cikretha'; cikriyiva — iṭ by the kṛādi restriction. BM then adds forms Bhaṭṭoji omits: krīṇātu, krīṇīhi [loṭ], akrīṇāt, akrīṇīta [laṅ], krīṇīyāt, krīṇīta [vidhiliṅ]. Followed for the brackets and for the reading cikrayitha (see textual note).
  • textual Our text reads 'krīṇāte | krīṇāte |'; the 1915 print has 'krīṇāte | krīṇate |' (NSP p. 415, scan page-0425, read; collation batch03, class A; Wikisource agrees with the print). The prathama-puruṣa plural krīṇate is supplied in brackets in the English.
  • textual Our text and the print (NSP p. 415, scan page-0425, read) both give the madhyama-puruṣa singular as 'cikriyitha'; BM's pratīka is 'cikrayitha cikretha', and thal, being pit, does not take the kit-status that would allow iyaṅ. Wikisource (uttara-2/0075) agrees with BM: 'cikrayitha-cikretha'. The English follows BM's reading cikrayitha. Mūla reading for Ajai's check.
  • textual 'krītāt' stands in both our text and the print (NSP p. 415, scan page-0425, read) between the lṛṭ kreṣyati and the ātmanepada āśīrliṅ kreṣīṣṭa, the slot of the parasmaipada āśīrliṅ, whose expected form is krīyāt; Wikisource (uttara-2/0075) reads 'krīyāt' in this slot. Neither commentary remarks on it. Translated as transmitted and left unlabelled. Mūla disagreement (ours/print krītāt vs Wikisource krīyāt) for Ajai's printed copy.
  • textual The dhātu numbers in our markup run one lower than the print's throughout this prakaraṇa (print 1474 ḍukrīñ … 1479 skuñ; ours 1473 … 1478); given as transmitted. The print heads the prakaraṇa 'tiṅante kryādayaḥ 49' and prints '1474 ḍukrīñ dravyavinimaye' above the sūtra, as our k1 does.
  • textual Filing: the opening of sk.2554.b1 — 'atha kryādayaḥ | atha śnāvikaraṇadhātavo nirūpyante | ḍu krīñ iti' ('now the kryādi roots: the roots with the vikaraṇa śnā are now set out'), BM's heading for the kryādi section and its lemma on ḍukrīñ — was re-filed 2026-10-05 from the tail of sk.2894.b1 (p66) and prepended here. It adds nothing that bears on the English; the rest of sk.2554.b1, covered in the notes above, agrees with Wikisource (uttara-2/0075).
  • translator The kṛādi restriction is kṛsṛbhṛvṛstudrusruśruvo liṭi (SK 2293), read as a restriction (niyama) that other roots take iṭ in liṭ. The meaning of āpravaṇa (skuñ) is not explained by either commentator and is left untranslated.
SK 2555 · A 3.1.82V
स्तन्भुस्तुन्भुस्कन्भुस्कुन्भुस्कुञ्भ्यः श्नुश्च ॥ २५५५ ॥
stanbhustunbhuskanbhuskunbhuskuñbhyaḥ śnuśca
stanbhustunbhuskanbhuskunbhuskuñbhyaḥ śnuś ca — After stanbhu, stunbhu, skanbhu, skunbhu and skuñ, śnu [comes], and [śnā — carried forward (anuvṛtti) by 'ca' from kryādibhyaḥ śnā, SK 2554].

Rule. After stanbhu, stunbhu, skanbhu, skunbhu and skuñ, the class sign śnu comes. By the word ca "and", śnā comes as well. Śnā is carried from SK 2554.

Example.

vi + stanbh + ti → śnu comes

→ vi + stanbh + nu + ti (the ś of śnu is a marker)

→ nu is ṅit, so the penultimate n of the root is lost (SK 415): vi + stabh + nu + ti

→ after vi, the s becomes ṣ (SK 2272): vi-ṣ-tabh-nu-ti

→ with the usual sound changes (the u of nu becomes o before ti; t becomes ṭ after ṣ)

→ viṣṭabhnoti "he stiffens".

With śnā instead: viṣṭabhnāti.

Applies when / blocked by. For skuñ Bhaṭṭoji gives both sets of forms:

  • with śnu: skunoti, skunute;
  • with śnā: skunāti, skunīte;
  • liṭ cuskāva; luṭ skotā; luṅ askauṣīt, ātmanepada askoṣṭa.

The liṭ ātmanepada is transmitted in two readings; scholars discuss this.

The four roots stambhu and the rest are sautra roots. Bhaṭṭoji's commentators add that their penultimate is n. On their meaning there are two views:

  • some say all four mean "to obstruct";
  • Mādhava says the first and third mean "to stiffen", the second "to extract", and the fourth "to hold".

All four take parasmaipada only.

The forms of stanbhu with prefixes:

  • after ava: avaṣṭabhnoti, avaṣṭabhnāti, with ṣ by SK 2273;
  • liṭ avataṣṭambha, with ṣ in the abhyāsa too, by SK 2277;
  • luṅ: aṅ is optional by SK 2291. With aṅ, vyaṣṭabhat, where the ṣ comes across the augment a by SK 2276. Without it, vyaṣṭambhīt.

Stunbhu gives stubhnoti, stubhnāti.

Watch out. The root's n is lost before the ṅit class sign. In the liṭ form avataṣṭambha it stays.

Siddhānta-kaumudī
चात् श्ना । स्कुनोति । स्कुनुते । स्कुनाति । स्कुनीते । चुस्काव । चुस्काव । चुस्कवे । स्कोता । अस्कौषीत् । अस्कोष्ट । स्तम्भ्वादयश्चत्वारः सौत्राः । सर्वे रोधनार्था इत्येके । माधवस्तु प्रथमतृतीयौ स्तम्भार्थौ द्वितीयो निष्कोषणार्थश्चतुर्थो धारणार्थ इत्याह । सर्वे परस्मैपदिनः । नलोपः । विष्टभ्नोति । विष्टभ्नाति । अवष्टभ्नोति । अवष्टभ्नाति । अवतष्टम्भ । जॄस्तम्भु-SK 2291 इत्यङ्वा । व्यष्टभत् । व्यष्टम्भीत् । स्तुभ्नोति । स्तुभ्नाति ॥

By 'ca', śnā [also comes]. skunoti, skunute [with śnu, parasmaipada and ātmanepada]. skunāti, skunīte [with śnā]. cuskāva [liṭ, prathama-puruṣa singular]. cuskave [liṭ ātmanepada]. skotā [luṭ]. askauṣīt [luṅ parasmaipada]. askoṣṭa [luṅ ātmanepada]. The four, stambhu and the rest, are sautra [roots, taught in the sūtra itself; they have n as penultimate]. All of them mean obstructing (rodhana) — so some. But Mādhava says: the first and the third mean stiffening (stambha), the second means extracting (niṣkoṣaṇa), and the fourth means holding (dhāraṇa). All are parasmaipadin. Deletion of the n [by aniditāṃ hala upadhāyāḥ kṅiti (SK 415)]: viṣṭabhnoti, viṣṭabhnāti [with vi; ṣ by stambheḥ (SK 2272)]. avaṣṭabhnoti, avaṣṭabhnāti [with ava; ṣ by avāc cālambanāvidūryayoḥ (SK 2273)]. avataṣṭambha [liṭ; ṣ by sthādiṣv abhyāsena cābhyāsasya (SK 2277)]. By jṝstambhu… (SK 2291), aṅ [comes in luṅ] optionally: vyaṣṭabhat [with aṅ; ṣ by prāk sitād aḍvyavāye 'pi (SK 2276)], vyaṣṭambhīt [without aṅ]. stubhnoti, stubhnāti.

Bālamanoramā
स्तम्भुस्तुम्भुस्कम्भुस्कुम्भुस्कुञ्भ्यः श्नुश्च - सौत्रा इति । नोपधा इत्यपि ज्ञेयम् । नलोप इति ।अनिदिता॑मित्यनेनेति भावः । विष्टभ्नोतीति ।स्तन्भे॑रिति षत्वम् । अवष्टभ्नोतीति ।अवाच्चालम्बनाविदूर्ययो॑रिति षत्वम् । अवतष्टम्भेति ।स्थादिष्वभ्यासेने॑ति षत्वम् । व्यष्टभदिति ।प्राक्सितादिति षत्वम् ।
translator notes
  • commentary BM on 'sautrāḥ': it is also to be understood that they have n as penultimate (nopadhāḥ); 'nalopaḥ' — by aniditām [SK 415]; viṣṭabhnoti — ṣ by 'stanbheḥ'; avaṣṭabhnoti — ṣ by 'avāc cālambanāvidūryayoḥ'; avataṣṭambha — ṣ by 'sthādiṣv abhyāsena'; vyaṣṭabhat — ṣ by 'prāk sitāt'. Followed: all the bracketed rule names.
  • commentary TB, filed in sk.2554.t1: 'sautrā iti — it is also to be understood that they have n as penultimate'; 'viṣṭabhnoti — n-deletion by aniditām; ṣ by stanbheḥ'. Followed, as for BM.
  • textual Our sūtra reads stanbhu-stunbhu-skanbhu-skunbhu (the n-form of the upadeśa, = Wikisource); the 1915 print (NSP p. 415, scan page-0425) and BM's pratīka read stambhu-stumbhu-skambhu-skumbhu (collation batch03, class V). Our kaumudī itself says 'stambhvādayaḥ'. The pratīka is given as in our text.
  • textual Our text reads 'cuskāva | cuskāva | cuskave'; the print has 'cuskāva | cuskave' (NSP p. 415, scan page-0425, read; collation batch03, class A). The duplicate is not rendered.
  • textual cuskave: ours = the print (scan read); Wikisource has cuskuve, the form expected for an ātmanepada liṭ, which is kit and so takes uvaṅ rather than guṇa (collation batch03, class V). Translated as transmitted; mūla reading for Ajai's check.
  • translator The stanbhu forms illustrate both vikaraṇas (śnu: viṣṭabhnoti; śnā: viṣṭabhnāti); stubhnoti, stubhnāti are the corresponding forms of stunbhu.
SK 2556 · A 8.3.77V
वेः स्कभ्नातेर्नित्यम् ॥ २५५६ ॥
veḥ skabhnāternityam
veḥ skabhnāter nityam — After vi, [ṣ comes in place of the s] of skabhnāti, invariably (nityam).

Rule. After vi, ṣ replaces the s of skabhnāti, the root skanbhu. It does so invariably (nityam).

Example.

vi + skanbh + nu + ti → the n of the root is lost before the ṅit nu (SK 415)

→ vi + skabh + nu + ti

→ after vi, ṣ replaces s

→ with the usual sound changes (the u of nu becomes o before ti)

→ viṣkabhnoti.

With śnā: viṣkabhnāti.

Applies when / blocked by. The word nityam is there to stop the "optionally" of SK 2359 from carrying into this rule. So the ṣ is not optional.

The Tattvabodhinī reads skabhnāti as a name for the root itself. It does not mean only the form with śnā. Otherwise the ṣ would not come in viṣkambhitā, where there is no śnā.

Without vi the rule has nothing to do. Skunbhu gives skubhnoti, skubhnāti.

Watch out. The rule is about the root, not about śnā. The ṣ also comes in forms with no class sign, such as viṣkambhitā.

Siddhānta-kaumudī
वेः परस्य स्कभ्नातेः सस्य षः स्यात् । विष्कभ्नोति । विष्कभ्नाति । स्कुभ्नोति । स्कुभ्नाति ॥

In place of the s of skabhnāti [i.e. of the root skanbhu, not only of its śnā-form] after vi, ṣ comes [invariably]. viṣkabhnoti, viṣkabhnāti. skubhnoti, skubhnāti.

Tattvabodhinī
NSP p. 425
—वेः स्कभ्नातेर्नित्यम् । ‘सिवादीनां वा’ इत्यतो वेति नानुवर्तते इति ध्वननार्थमिह नित्यग्रहणम् ॥—विस्कभ्नोतीति । स्कभ्नातेरित्यत्र श्नाविशिष्टं रूपं न विवक्षितं किं तु धातुमात्रम् । अन्यथा विष्कम्भिता विष्कम्भितमित्यादावपि षत्वं न स्यात् । न चेष्टापत्तिः, माधवादिग्रन्थविरोधादिति भावः
translator notes
  • commentary TB (sk.2556.t1): 'nityam' is stated here to signal that 'vā' is not carried forward (nānuvartate) from 'sivādīnāṃ vā' [SK 2359]. Followed: 'invariably (nityam)' in sutra_en and k1.
  • commentary TB (sk.2556.t1) on viṣkabhnoti: in 'skabhnāteḥ' the form qualified by śnā is not intended, but the root as such; otherwise ṣ would not occur in viṣkambhitā, viṣkambhitam and the like; nor can that be accepted as desired (iṣṭāpatti), since it contradicts the works of Mādhava and others. Followed: the bracket in k1.
  • translator skubhnoti and skubhnāti, without vi, are the forms of skunbhu, completing the series of sautra roots begun in SK 2555; the rule does not apply to them. The print (NSP p. 415, scan page-0425) matches our text (collation batch03: clean). BM has no remark on this sūtra.
SK 2557 · A 3.1.83V
हलः श्नः शानज्झौ ॥ २५५७ ॥
halaḥ śnaḥ śānajjhau
halaḥ śnaḥ śānaj jhau — In place of śnā after a hal, śānac [is substituted] when hi follows.

Rule. After a consonant (hal), śānac replaces śnā when the ending hi follows. The ś and c of śānac are markers, so āna remains.

Example.

stabh + nā + hi → stabh ends in a consonant, so śānac replaces śnā

→ stabh + āna + hi

→ hi is lost after a (SK 2202)

→ stabhāna "stiffen!".

Likewise stubhāna, skabhāna, skubhāna.

Applies when / blocked by. One might object that the a of āna exists only because of hi. Could it then cause the loss of hi? The Tattvabodhinī says yes. Because SK 2202 was taught at all, the sannipāta-paribhāṣā is not applied here, so hi is lost.

Some held that śānac comes only after roots in the Dhātupāṭha, not after sautra roots. They gave stabhnīhi. The Tattvabodhinī rejects this as without authority.

The rule replaces śnā only. With śnu, hi stays after the u: stabhnuhi.

The roots that follow:

  • Yu "to bind": yunāti, yunīte; luṭ yotā.
  • Knū "to sound": knūnāti, knūnīte; luṭ knavitā.
  • Drū "to injure": drūṇāti, drūṇīte.
  • Pū "to purify". It begins the next rule.

Watch out. After a vowel, śnā stays before hi: krīṇīhi, not krīṇāna.

Siddhānta-kaumudī
हलः परस्य श्नः शानजादेशः स्याद्धौ परे । स्तभान । स्तुभान । स्कभान । स्कुभान । पक्षे स्तभ्नुहीत्यादि ।{$ {!1479 युञ्!} बन्धने $}। युनाति । युनीते । योता ।{$ {!1480 क्नूञ्!} शब्दे $}। क्नूनाति । क्नूनीते । क्नविता ।{$ {!1481 द्रूञ्!} हिंसायाम्$} । द्रूणाति । द्रूणीते ।{$ {!1482 पूञ्!} पवने$} ॥

In place of śnā coming after a hal, the substitute śānac comes when hi follows. stabhāna [stabh + āna + hi → luk of hi by ato heḥ (SK 2202) → stabhāna]. stubhāna. skabhāna. skubhāna. In the alternative [with śnu]: stabhnuhi, and so on. (1479) yuñ — in binding. yunāti, yunīte. yotā [luṭ]. (1480) knūñ — in sounding. knūnāti, knūnīte. knavitā [luṭ]. (1481) drūñ — in injuring. drūṇāti, drūṇīte. (1482) pūñ — in purifying (pavana).

Tattvabodhinī
alignment inferredNSP p. 425
—हलः श्नः—॥—स्तभानेति । ‘अतो हेः’ इत्यारम्भसामर्थ्यात्संनिपातपरिभाषाया अप्रवृत्तेर्हेर्लुक् । अत एव जहीत्यत्र हिलोपवारणायेयं परिभाषा नोपन्यस्तेत्याहुः । यत्तु कैश्चित् शानजादेशो धातुपाठपटितेभ्यः परस्य श्नो भवति न तु सौत्रेभ्य इति व्याख्याय स्तभ्नीहीत्युदाहृतं तन्निष्प्रमाणं माधवादिग्रन्थविरुद्धं च
Bālamanoramā
हलः श्नः शानज्झौ - स्तभानेति श्नाप्रत्ययस्य शानजादेसे कृते 'अतो हे' रिति लुक् ।
translator notes
  • commentary BM on stabhāna: once śānac has been substituted for the affix śnā, luk [of hi] by 'ato heḥ'. Followed: the bracketed derivation.
  • commentary TB (sk.2557.t1) on stabhāna: luk of hi, because by the very fact that 'ato heḥ' is taught, the sannipāta-paribhāṣā does not operate here; for this reason, they say, this paribhāṣā is not cited to prevent the deletion of hi in jahi. Recorded.
  • commentary TB (sk.2557.t1): some explain that śānac replaces śnā only after roots read in the dhātupāṭha, not after sautra roots, and give 'stabhnīhi' as the example; that is without authority and contrary to the works of Mādhava and others. Followed: stabhāna etc. from the sautra roots stand as Bhaṭṭoji's forms.
  • textual The print (NSP p. 415, scan page-0425) matches our text (collation batch03: clean); the TB OCR 'śraḥ', 'stannīhi' read śnaḥ, stabhnīhi on the scan.
  • translator śānac: ś and c are it-markers, leaving āna. 'In the alternative' refers to the śnu-forms allowed by SK 2555, where hi follows u after a conjunct and is not deleted.
SK 2558 · A 7.3.80V
प्वादीनां ह्रस्वः ॥ २५५८ ॥
pvādīnāṃ hrasvaḥ
pvādīnāṃ hrasvaḥ — Of pū and the rest, a short vowel [is substituted] [before a śit (śiti — carried forward (anuvṛtti) from ṣṭhivuklamucamāṃ śiti, SK 2320)].

Rule. For pū and the roots that follow it, a short vowel replaces the root vowel before a śit affix. "Before a śit" (śiti) is carried from SK 2320.

Example.

pū + nā + ti → śnā has ś as marker, so it is a śit affix

→ the ū of pū becomes short: pu + nā + ti

→ punāti "he purifies"; ātmanepada punīte.

Applies when / blocked by. Only a śit affix brings the short vowel. Elsewhere this rule does not apply: luṭ pavitā, with iṭ, since pū is seṭ, ending in ū.

Lū "to cut": lunāti, lunīte.

Stṝ "to cover" shows what the ṝ-final roots do:

  • with śnā: stṛṇāti, stṛṇīte, with short ṛ by this rule;
  • liṭ tastāra; dual tastaratuḥ, with guṇa although the ending is kit, by SK 2383;
  • luṭ staritā or starītā: the iṭ is optionally long (SK 2391);
  • vidhiliṅ stṛṇīyāt, stṛṇīta;
  • āśīrliṅ stīryāt: ir for ṝ (SK 2390), then the i is lengthened before the consonant (SK 354).

The āśīrliṅ ātmanepada of stṝ needs three rules:

  • the iṭ is optional before liṅ in the ātmanepada (SK 2528);
  • with iṭ: the long iṭ of SK 2391 does not come before liṅ (SK 2529), so stariṣīṣṭa;
  • without iṭ: liṅ is kit by uś ca (SK 2368), so there is no guṇa: stīrṣīṣṭa.

luṅ of stṝ:

  • parasmaipada astārīt, astāriṣṭām. The iṭ of sic is not long here, by SK 2392.
  • ātmanepada with iṭ: astariṣṭa or astarīṣṭa, optionally long by SK 2391.
  • ātmanepada without iṭ: astīrṣṭa.

Kṝ "to injure": kṛṇāti, kṛṇīte; liṭ cakāra, ātmanepada cakare with guṇa by SK 2383.

Vṝ "to choose": vṛṇāti, vṛṇīte; liṭ vavāra, vavare; luṭ varitā or varītā.

  • āśīrliṅ vūryāt: u for ṝ after a labial, by SK 2494. The labial v, made with lips and teeth, counts.
  • āśīrliṅ ātmanepada variṣīṣṭa (with iṭ) or vūrṣīṣṭa (without).
  • luṅ avārīt, avāriṣṭām; ātmanepada avariṣṭa, avarīṣṭa, avūrṣṭa. In avūrṣṭa, sic is kit by uś ca, so there is no guṇa, and u comes.

Dhū "to shake": dhunāti, dhunīte.

  • liṭ second person singular dudhavitha or dudhotha: the iṭ is optional by SK 2279. First person dual dudhuviva.
  • luṭ dhotā or dhavitā.
  • luṅ adhāvīt, with iṭ by SK 2385; ātmanepada adhaviṣṭa or adhoṣṭa.

Bhaṭṭoji now notes: "the roots up to badhnāti (bandh) take parasmaipada."

Śṝ "to injure":

  • liṭ dual śaśratuḥ or śaśaratuḥ. When the short vowel of SK 2495 comes, ṛ becomes r (yaṇ). Otherwise guṇa comes by SK 2383.
  • liṭ first person dual śaśariva or śaśriva. SK 2381 would forbid the iṭ, but the kṛādi restriction overrides it.
  • luṭ śaritā or śarītā; loṭ śṛṇīhi; āśīrliṅ śīryāt; luṅ aśāriṣṭām.

Pṝ "to protect, to fill": liṭ papratuḥ or paparatuḥ; āśīrliṅ pūryāt, with u after the labial p.

Then a list of ṝ-final roots:

  • vṝ "to choose" (some say "to bear");
  • bhṝ "to reproach" (some add "to bear");
  • mṝ "to injure": mṛṇāti, liṭ mamāra;
  • dṝ "to tear": liṭ dual dadaratuḥ or dadratuḥ;
  • jṝ "to grow old" (some read jhṝ, others dhṝ);
  • nṝ "to lead"; kṝ "to injure".

Ṝ "to go":

  • ṛṇāti;
  • liṭ arāṃcakāra, with ām. The Bālamanoramā says the root counts as having a heavy vowel;
  • luṭ aritā or arītā;
  • laṅ ārṇāt, ārṇītām, with the augment ā;
  • āśīrliṅ īryāt; luṅ ārīt, āriṣṭām.

Gṝ "to sound". Then jyā "to grow old". Its y becomes i by saṃprasāraṇa (SK 2412), and i + ā gives i (pūrvarūpa). That leaves ji + nā + ti, where the next rule applies.

Watch out. The short vowel is a substitute for the root vowel, not a guṇa. In stṛṇāti the ṝ becomes ṛ; it does not become ar.

Siddhānta-kaumudī
शिति परे । पुनाति । पुनीते । पविता ।{$ {!1483 लूञ्!} छेदने$} । लुनाति । लुनीते ।{$ {!1484 स्तॄञ्!} आच्छादने स्तृणाति$} । स्तृणीते । तस्तार । तस्तरतुः । स्तरिता । स्तरीता । स्तृणीयात् । स्तृणीत । आशिषि । स्तीर्यात् । लिङ्सिचोः SK 2528 इति वेट् न लिङि SK 2529 वॄत इटो लिङि दीर्घो न स्यात् । स्तरिषीष्ट । उश्च SK 2368 इति कित्त्वम् । स्तीर्षीष्ट । सिचि च परस्मैपदेषु SK 2392 इति न दीर्घः । अस्तारीत् । अस्तारिष्टाम् । अस्तरिष्ट । अस्तरीष्ट । अस्तीर्ष्ट ।{$ {!1485 कॄञ्!} हिंसायाम्$} । कृणाति । कृणीते । चकार । चकरे ।{$ {!1486 वॄञ्!} वरणे$} । वृणाति । वृणीते । ववार । ववरे । वरिता । वरीता । आशिषि उदेष्ठ्यपूर्वस्य SK 2494 वूर्यात् । वरिषीष्ट । वूर्षीष्ट अवारीत् । अवारिष्टाम् । अवरिष्ट । अवरीष्ट । अवूर्ष्ट ।{$ {!1487 धूञ्!} कम्पने$} । धुनाति । धुनीते । दुधविथ । दुधोथ । दुधुविव । धोता । धविता । स्तुसुधूञ्भ्यः - SK 2385 इतीट् । अधावीत् । अधविष्ट । अधोष्ट ॥ अथ बध्नात्यन्ताः परस्मैपदिनः ।{$ {!1488 शॄ!} हिंसायाम्$} । शॄदॄप्रां ह्रस्वो वा SK 2495 इति ह्रस्वपक्षे यण् । अन्यदा ऋच्छत्यॄताम् SK 2383 इति गुणः । शश्रतुः । शशरतुः श्र्युकः किति SK 2381 इति निषेधस्य क्रादिनियमेन बाधः । शशरिव । शश्रिव । शरिता । शरीता । शृणीहि । शीर्यात् । अशारिष्टाम् ।{$ {!1489 पॄ!} पालनपूरणयोः $}। पप्रतुः । पपरतुः । आशिषि । पूर्यात् ।{$ {!1490 वॄ!} वरणे$} । भरण इत्येके ।{$ {!1491 भॄ!} भर्त्सने$} । भरणेप्येके ।{$ {!1492 मॄ!} हिंसायाम्$} । मृणाति । ममार ।{$ {!1493 दॄ!} विदारणे$} । ददरतुः । दद्रतुः ।{$ {!1494 जॄ!} वयोहानौ$} । झॄ इत्येके । धॄ इत्यन्ये ।{$ {!1495 नॄ!} नये$} ।{$ {!1496 कॄ!} हिंसायाम्$} ।{$ {!1497 ॠ!} गतौ$} । ऋणाति । अरांचकार । अरिता । अरीता । अर्णात् । अर्णीताम् । ईर्यात् । आरीत् । आरिष्टाम् ।{$ {!1498 गॄ!} शब्दे$} ।{$ {!1499 ज्या!} वयोहानौ$} । ग्रहिज्या SK 2412 ॥

When a śit follows [— 'śiti' carried forward (anuvṛtti) from ṣṭhivuklamucamāṃ śiti (SK 2320)]. punāti, punīte. pavitā [luṭ; seṭ, being ū-final]. (1483) lūñ — in cutting. lunāti, lunīte. (1484) stṝñ — in covering. stṛṇāti, stṛṇīte. tastāra [liṭ, prathama-puruṣa singular]. tastaratuḥ [prathama-puruṣa dual; guṇa by ṛcchaty ṝtām (SK 2383) although the ending is kit]. staritā, starītā [luṭ; the iṭ optionally long by vṝto vā (SK 2391)]. stṛṇīyāt, stṛṇīta [vidhiliṅ parasmaipada and ātmanepada]. In the āśīrliṅ: stīryāt [ir for ṝ by ṝta iddhātoḥ (SK 2390), lengthened by hali ca (SK 354)]. [In the ātmanepada] iṭ is optional by liṅsicor ātmanepadeṣu (SK 2528); [by] na liṅi (SK 2529), the long vowel [taught by vṝto vā] for the iṭ of vṝ and of ṝ-final roots does not occur before liṅ: stariṣīṣṭa. [Without iṭ,] by uś ca (SK 2368) [liṅ] is kit [so there is no guṇa]: stīrṣīṣṭa. By sici ca parasmaipadeṣu (SK 2392) there is no lengthening [of the iṭ of sic in the parasmaipada]: astārīt, astāriṣṭām [luṅ parasmaipada]. astariṣṭa, astarīṣṭa [luṅ ātmanepada with iṭ, optionally long by vṝto vā], astīrṣṭa [without iṭ]. (1485) kṝñ — in injuring. kṛṇāti, kṛṇīte. cakāra [liṭ]. cakare [liṭ ātmanepada; guṇa by ṛcchaty ṝtām]. (1486) vṝñ — in choosing. vṛṇāti, vṛṇīte. vavāra, vavare [liṭ]. varitā, varītā [luṭ]. In the āśīrliṅ, [by] udoṣṭhyapūrvasya (SK 2494) [u for ṝ after a labial, the dento-labial v included]: vūryāt. variṣīṣṭa, vūrṣīṣṭa [ātmanepada, with and without iṭ]. avārīt, avāriṣṭām [luṅ parasmaipada]. avariṣṭa, avarīṣṭa, avūrṣṭa [luṅ ātmanepada]. (1487) dhūñ — in shaking. dhunāti, dhunīte. dudhavitha, dudhotha [liṭ, madhyama-puruṣa singular; iṭ optional by svaratisūtisūyatidhūñūdito vā (SK 2279)]. dudhuviva [uttama-puruṣa dual]. dhotā, dhavitā [luṭ]. iṭ by stusudhūñbhyaḥ parasmaipadeṣu (SK 2385): adhāvīt [luṅ parasmaipada]. adhaviṣṭa, adhoṣṭa [luṅ ātmanepada]. Now [the roots] ending with badhnāti [bandh] are parasmaipadin. (1488) śṝ — in injuring. In the alternative where the short vowel comes by śṝdṝprāṃ hrasvo vā (SK 2495), yaṇ [r for the ṛ]; otherwise guṇa by ṛcchaty ṝtām (SK 2383): śaśratuḥ, śaśaratuḥ [liṭ, prathama-puruṣa dual]. The prohibition [of iṭ] by śryukaḥ kiti (SK 2381) is overridden by the kṛādi restriction: śaśariva, śaśriva [uttama-puruṣa dual]. śaritā, śarītā [luṭ]. śṛṇīhi [loṭ, madhyama-puruṣa singular]. śīryāt [āśīrliṅ]. aśāriṣṭām [luṅ, prathama-puruṣa dual]. (1489) pṝ — in protecting and filling. papratuḥ, paparatuḥ [liṭ]. In the āśīrliṅ: pūryāt. (1490) vṝ — in choosing. Some [read] 'in bearing' (bharaṇa). (1491) bhṝ — in reproaching (bhartsana). Some [read] 'in bearing' too. (1492) mṝ — in injuring. mṛṇāti. mamāra [liṭ]. (1493) dṝ — in tearing (vidāraṇa). dadaratuḥ, dadratuḥ [liṭ, prathama-puruṣa dual]. (1494) jṝ — in ageing (vayohāni). Some [read] jhṝ; others dhṝ. (1495) nṝ — in leading. (1496) kṝ — in injuring. (1497) ṝ — in going. ṛṇāti. arāṃcakāra [liṭ, with ām]. aritā, arītā [luṭ]. ārṇāt, ārṇītām [laṅ, prathama-puruṣa singular, dual]. īryāt [āśīrliṅ]. ārīt, āriṣṭām [luṅ]. (1498) gṝ — in sounding. (1499) jyā — in ageing. [samprasāraṇa by] grahijyā… (SK 2412).

Tattvabodhinī
NSP p. 425–426
—प्वादीनां ह्रस्वः । ‘ष्ठिवुक्लमुचमाम्—इत्यतः शितीत्यनुवर्तत इत्याह—शिति पर इति ॥—न लिङीति । अनेन वॄत इटो लिङि दीर्घो नेति स्मारितम् ॥—उदोष्ठ्य—। दन्त्योष्ठ्योऽपि ओष्ठ्यग्रहणेन गृह्यत इत्याह—वूर्यादिति ॥—वरिषीष्टेति । ‘लिङ्सिचोरात्मनेपदेषु’ इति वेट् ॥—अवूर्ष्टेति । ‘उश्च’ इति लिङ्सिचोः कित्वाद् गुणाभावे सत्युत्वम्
Bālamanoramā
प्वादीनां ह्रस्वः - प्वादीनां ह्रस्वः। ष्ठिवुक्लमुचमाम् इत्यतः शितीत्यनुवर्तत इत्यभिप्रेत्य शेषं पूरयति– शिति परे इति। पवितेति। ऊकारान्तत्वात् सेड् इति भावः। स्तॄञ् आच्छादने। सेट्। तस्तरतुरिति। कित्त्वेऽपि ‘ऋच्छत्यॄताम्’ इति गुणः। स्तरिता स्तरीतेति। ‘वॄतो वा’ इति दीर्घविकल्पः। आशिषि स्तीर्याद् इति। ‘ॠत इद्धातो’रिति इत्त्वे रपरत्वे ‘हलि चे’ति दीर्घः। स्तॄ षीष्ट इति स्थिते आह– लिङ्सिचोरिति वेड् इति। ‘वॄतो वा’ इति दीर्घविकल्पनिषेधसूत्रं स्मारयति– न लिङीति। वॄत इटो दीर्घो नेत्यर्थः। इडभावपक्षे स्तॄषीष्टेत्यत्र गुणम् आशङ्क्याह– उश्चेति कित्त्वम् इति। लुङि परस्मैपदे सिचि अस्तारिष्टाम् इत्यत्र ‘वॄतो वे’ति इटो दीर्घविकल्पे प्राप्ते आह– सिचि चेति। अस्तरीष्ट अस्तरिष्टेति। ‘वृतो वे’ति दीर्घः। अस्तीर्ष्टेति। ‘लिङ्सिचो’रिति इडभावपक्षे ॠत इत्त्वे ‘हलि चे’ति दीर्घ इति भावः। कॄञ् हिंसायाम्। चकरे इति। ‘ऋच्छत्यॄताम्’ इति गुणः। वॄञ् वरणे इति। आशिषीति। वॄ यात् इति स्थिते ‘उदोष्ठपूर्वस्ये’ति ॠकारस्य उत्त्वम् इत्यर्थः, तत्र दन्त्योष्ठपूर्वस्यापि ग्रहणाद् इति भावः। ॠत उत्त्वे रपरत्वे ‘हलि चे’ति दीर्घ इति बोद्ध्यम्। वरिषीष्ट वूर्षीष्टेति। ‘लिङ्सिचो’रिति वेट्। इडभावे ‘उश्चे’ति कित्त्वाद् गुणाभावे उत्त्वे रपरत्वे ‘हलि चे’ति दीर्घ इति भावः। अवारिष्टाम् इति। ‘सिचि चे’ति इटो न दीर्घ इति भावः। अवरिष्ट अवरीष्टेति। ‘वॄतो वा’ इति सिच इटो दीर्घविकल्प इति भावः। दुधविथ, दुधोथेति। ‘स्वरतिसूतिसूयतिधूञूदित’ इति वेडिति भावः। शॄ हिंसायाम्। सेट्। णलि– शशार। शॄ अतुस् इत्यत्र कित्त्वेऽपि ‘ऋच्छत्यॄताम्’ इति नित्ये गुणे प्राप्ते आह– शॄदॄप्राम् इति। गुणाऽपवादे पाक्षिके ह्रस्वे कृते ॠकारस्य यणि रेफ इत्यर्थः। शशरिथ। पॄधातुरपि शॄधातुवत्। आशिषि पूर्यादिति। ‘उदोष्ठ्यपूर्वस्ये’त्युत्वम् इति भावः। दॄ विदारणे। ‘शॄदॄप्राम्’ इति ह्रस्वविकल्पं मत्वाह– ददरतुः दद्रतुरिति। ॠ गतौ। अरांचकारेति। व्यपदेशिवत्त्वेन गुरुमत्त्वाद् आम्। लङ्याह– आर्णादिति। ज्या वयोहानौ। अनिट्। ज्या ना तीति स्थिते ‘ग्रहिज्ये’ति सम्प्रसारणे पूर्वरूपे च जिनातीति स्थिते–
translator notes
  • commentary BM: intending that 'śiti' is carried forward (anuvartate) from ṣṭhivuklamucamām, [Bhaṭṭoji] completes the remainder: 'śiti pare'. Followed: bracketed in sutra_en and k1.
  • commentary TB (sk.2558.t1): 'śiti' is carried forward (anuvartate) from 'ṣṭhivuklamucamām', hence 'śiti pare'. Followed, as for BM.
  • commentary BM on the stṝ forms: pavitā — seṭ because ū-final; stṝ is seṭ; tastaratuḥ — guṇa by ṛcchaty ṝtām even though kit; staritā/starītā — optional length by vṝto vā; stīryāt — i by ṝta iddhātoḥ, r-augment, length by hali ca; na liṅi recalled as the sūtra that forbids the optional length of vṝto vā for the iṭ; stīrṣīṣṭa — in the iṭ-less alternative, kit by uś ca, so no guṇa; sici ca — blocks the length of iṭ in astāriṣṭām; astarīṣṭa/astariṣṭa by vṝto vā; astīrṣṭa — iṭ-less by liṅsicoḥ, ir and length. Followed: all the brackets on the stṝ, kṝ and vṝ forms.
  • commentary TB (sk.2558.t1): by 'na liṅi' it is recalled that the iṭ of vṝ [and ṝ-final roots] is not long before liṅ; on udoṣṭhya: the dento-labial too is taken by the mention of 'oṣṭhya', hence vūryāt; variṣīṣṭa — optional iṭ by liṅsicor ātmanepadeṣu; avūrṣṭa — since liṅ and sic are kit by uś ca, guṇa does not occur, and u [for ṝ, by udoṣṭhyapūrvasya] results. Followed: '[the dento-labial v included]'.
  • commentary BM on vūryāt: with 'vṝ yāt' standing, u for ṝ by 'udoṣṭhyapūrvasya', since a dento-labial is also included, then r-augment and length by hali ca; vūrṣīṣṭa — without iṭ, kit by uś ca, no guṇa, u, r, length. Followed.
  • commentary BM on dudhavitha/dudhotha: optional iṭ by 'svaratisūtisūyatidhūñūditaḥ'. On śṝ: seṭ; ṇal — śaśāra; in 'śṝ atus', guṇa by ṛcchaty ṝtām being obligatorily due even though kit, śṝdṝprām is stated; when the optional short vowel (an exception to guṇa) is made, yaṇ gives r; śaśaritha; pṝ is like śṝ; pūryāt by udoṣṭhyapūrvasya; dadaratuḥ/dadratuḥ by the optional short vowel of śṝdṝprām. Followed: the brackets on dhūñ, śṝ and dṝ.
  • commentary BM on ṝ gatau: arāṃcakāra — ām because [the root] is guru-possessing by vyapadeśivadbhāva; on laṅ he gives 'ārṇāt'. On jyā: aniṭ; with 'jyā nā ti' standing, samprasāraṇa by grahijyā and pūrvarūpa give 'ji nā ti' — leading into halaḥ (SK 2559). Followed: the reading ārṇāt (see textual note).
  • textual Our text reads 'arṇāt | arṇītām'; the print reads 'ārṇāt | ārṇītām' (NSP p. 416, scan page-0426, read; collation batch03, class A), with the āṭ augment of laṅ; BM's pratīka is 'ārṇāt'. The English follows the print.
  • textual Our text reads 'udeṣṭhyapūrvasya' (SK 2494); the print has 'udoṣṭhyapūrvasya' (NSP p. 415, scan page-0425, read; collation batch03, class A). Translated as udoṣṭhyapūrvasya.
  • textual The print (NSP pp. 415–416, scans page-0425/0426) otherwise matches our text in this block; its dhātu numbers run one higher than ours (print 1483 pūñ … 1500 jyā). BM on this unit ends mid-sentence ('… jinātīti sthite —'), continuing into halaḥ (SK 2559).
  • translator The kṛādi restriction is kṛsṛbhṛvṛstudrusruśruvo liṭi (SK 2293). Where the pvādi group ends is itself discussed at SK 2559 (vṛt after plī). 'badhnātyantāḥ' names the end of the parasmaipadin run by the laṭ form of bandh (1508).
SK 2559 · A 6.4.2V
हलः ॥ २५५९ ॥
halaḥ
halaḥ — [In place of an aṅga (aṅgasya — adhikāra, applied twice) ending in a samprasāraṇa (saṃprasāraṇasya — carried forward (anuvṛtti) from samprasāraṇasya, SK 1004) that comes] after a hal [which is a member of the aṅga], a long vowel [is substituted (dīrghaḥ — carried forward (anuvṛtti) from ḍhralope pūrvasya dīrgho 'ṇaḥ, SK 174)].

Rule. A long vowel replaces an aṅga that ends in a saṃprasāraṇa vowel, when that vowel comes after a consonant belonging to the aṅga. "Saṃprasāraṇa" is carried from SK 1004 and "long vowel" from SK 174.

Bhaṭṭoji's commentators read the heading aṅgasya twice. Once it says the consonant must be part of the aṅga. Once it names the aṅga that is replaced, ending in the saṃprasāraṇa vowel (tadantavidhi).

Example.

jyā + nā + ti → saṃprasāraṇa (SK 2412) and pūrvarūpa

→ ji + nā + ti

→ the i is a saṃprasāraṇa after the consonant j, so it becomes long: jī + nā + ti

→ śnā is a śit affix, so SK 2558 makes the vowel short again: ji + nā + ti

→ jināti "he grows old".

liṭ jijyau, jijyatuḥ.

Applies when / blocked by. The Bālamanoramā gives the examples vītaḥ, śūnaḥ and jīnaḥ. The two commentaries give counter-examples for each condition:

  • no consonant before the vowel: utaḥ;
  • the consonant is not part of the aṅga, as the r of nir: nirutam, durutam;
  • the aṅga does not end in the saṃprasāraṇa vowel: vidhyati.

Watch out. In jināti this rule does fire. Its long vowel is then undone by SK 2558.

The roots that follow:

  • Rī "to go, to howl". Howling (reṣaṇa) is the cry of wolves.
  • Lī "to cling". Where e or ai would arise, ā comes optionally, by SK 2509: liṭ lalau or lilāya; luṭ lātā or letā.
  • Vlī "to choose": vlināti.
  • Plī "to go". Then the marker vṛt. It closes the run beginning with lū (lvādi). Some say it closes the pvādi too.
  • Vrī "to choose", bhrī "to fear" (some say "to bear") and kṣīṣ "to injure". If the pvādi do not close at vṛt, these three take the short vowel of SK 2558. Some say they do not.

Jñā "to know". Before a śit affix jñā becomes jā, by SK 2511: jānāti "he knows". The jā is stated with a long vowel. By the force of that statement, SK 2558 does not shorten it. Scholars discuss this reasoning.

Bandh "to bind":

  • badhnāti: the n is lost before the ṅit śnā (SK 415);
  • liṭ babandhitha, with iṭ. One commentator adds babanddha;
  • luṭ bandhā, bandhārau: the t of tās becomes dh (SK 2280);
  • lṛṭ bhantsyati: b becomes bh (SK 326), and dh becomes t by cartva;
  • loṭ badhāna, with śānac by SK 2557;
  • luṅ abhānsīt, with vṛddhi (SK 2267) and bh for b. One commentator writes abhāntsīt; scholars discuss this;
  • luṅ plural abhāntsuḥ.

The dual luṅ abāndhām needs care:

  • a + bandh + s + tām → vṛddhi → abāndh + s + tām;
  • by SK 12 the later bh rule counts as not yet applied when the earlier rule SK 2281, the loss of sic between two jhal sounds, is applied. So sic is lost first;
  • objection: an affix that is lost still has its effect (pratyayalakṣaṇa, SK 262). Would the lost s not still bring bh?
  • reply: the loss of sic is asiddha for pratyayalakṣaṇa. So no bh comes;
  • the t of tām becomes dh (SK 2280). One dh may then be lost (SK 71): abāndhām. One commentator writes abānddhām.

Vṛ "to serve" (vṛṅ, ātmanepada): vṛṇīte; liṭ vavre, vavṛṣe, vavṛḍhve; luṭ varitā or varītā; luṅ avariṣṭa, avarīṣṭa, avṛta.

Śranth "to release, to gladden". From here on the roots take parasmaipada. The n is lost before śnā: śrathnāti.

On the liṭ of śranth, Bhaṭṭoji reports some views:

  • Haradatta and others: another grammar makes liṭ optionally kit for śranth, granth and some other roots. Then e and the loss of the abhyāsa must also be stated: śrethatuḥ, śrethuḥ.
  • Sudhākara: this kit status holds even for the pit endings: śrethitha. With ṇal, śaśrātha.
  • Mādhava: in the first person, śaśrātha or śaśratha.

Bhaṭṭoji says the basis for this must be sought. Scholars discuss this.

More roots:

  • Manth "to churn".
  • Śranth and granth "to compose". Śranth is read again for its different meaning; its forms are as above.
  • Kunth "to cling" (some say "to afflict"): kuthnāti, liṭ cukuntha. Durga reads the root as kuth: cukotha.
  • Mṛd "to crush": mṛdnāti; loṭ mṛdāna, with śānac.
  • Mṛḍ, also "to crush", and also "to be happy": mṛḍṇāti, with ṣṭutva of the n.
  • Gudh "to be angry": gudhnāti.
  • Kuṣ "to extract": kuṣṇāti; luṭ koṣitā. It leads into the next rule.
Siddhānta-kaumudī
अङ्गावयवाद्धलः परं यत्संप्रसारणं तदन्ताङ्गस्य दीर्घः स्यात् । इति दीर्घे कृते प्वादीनां ह्रस्वः SK 2258 । जिनाति । जिज्यौ । जिज्यतुः ।{$ {!1500 री!} गतिरेषणयोः$} । रेषणं वृकशब्दः ।{$ {!1501 ली!} श्लेषणे$} । विभाषा लीयतेः SK 1509 इत्येज्विषये आत्वं वा । ललौ । लिलाय । लाता । लेता ।{$ {!1502 व्ली!} वरणे$} । व्लिनाति ।{$ {!1503 प्ली!} गतौ$} । वृत् । ल्वादयो वृत्ताः । प्वादयोऽपीत्येके ।{$ {!1504 व्री!} वरणे$} ।{$ {!1505 भ्री!} भये$} । भरण इत्येते ।{$ {!1506 क्षीष्!} हिंसायाम्$} । एषां त्रयाणां ह्रस्वः । केषांचिन्मते तु न ।{$ {!1507 ज्ञा!} अवबोधने$} । ज्ञाजनोर्जा SK 2511 जानाति । दीर्घनिर्देशसामर्थ्यान्न ह्रस्वः । {$ {!1508 बन्ध!} बन्धने$} । बध्नाति । बबन्धिथ । बन्धा । बन्धारौ । भन्त्स्यति । बधान । अभान्सीत् । पूर्वत्रासिद्धम् SK 12 इति भष्भावात्पूर्वं झलो झलि SK 2281 इति सिज्लोपः । प्रत्ययलक्षणेन सादिप्रत्ययमाश्रित्य भष्भावो न । प्रत्ययलक्षणं प्रतिसिज्लोपस्यासिद्धत्वात् । अबान्धाम् । अभान्त्सुः ।{$ {!1509 वृङ्!} संभक्तौ$} । वृणीते । वव्रे । ववृषे । ववृढ्वे । वरिता । वरीता । अवरिष्ट । अवरीष्ट । अवृत ।{$ {!1510 श्रन्थ!} विमोचनप्रतिहर्षयोः$} ॥ इतः परस्मैपदिनः ॥ श्रन्थाति । ।

[— 'saṃprasāraṇasya' (SK 1004) and 'dīrghaḥ' (from ḍhralope…, SK 174) carried forward (anuvṛtti), 'aṅgasya' applied twice —] A long vowel comes in place of an aṅga ending in a samprasāraṇa that comes after a hal which is a member of the aṅga. When the long vowel has been made by this [rule, in ji-nā-ti from jyā], [a short vowel comes by] pvādīnāṃ hrasvaḥ (SK 2558): jināti [jyā + nā + ti → samprasāraṇa by grahijyā… (SK 2412) and pūrvarūpa → ji + nā + ti → long vowel by this rule → jī + nā + ti → short vowel by pvādīnāṃ hrasvaḥ → jināti]. jijyau, jijyatuḥ [liṭ, prathama-puruṣa singular, dual]. (1500) rī — in going and in reṣaṇa. reṣaṇa is the howling of wolves (vṛkaśabda). (1501) lī — in clinging (śleṣaṇa). By vibhāṣā līyateḥ (SK 2509), ā [is substituted] optionally where an ec [vowel] would arise: lalau, lilāya [liṭ]. lātā, letā [luṭ]. (1502) vlī — in choosing. vlināti. (1503) plī — in going. vṛt. The lū-group (lvādayaḥ) is closed. Some [say] the pū-group (pvādayaḥ) too. (1504) vrī — in choosing. (1505) bhrī — in fearing. Some [read] 'in bearing' (bharaṇa). (1506) kṣīṣ — in injuring. These three have the short vowel [by pvādīnāṃ hrasvaḥ, on the view that the pū-group does not close at vṛt]; but in the view of some, not. (1507) jñā — in knowing. [jā for jñā before a śit by] jñājanor jā (SK 2511): jānāti. By the force of the statement [of jā] with a long vowel, no short vowel [comes by pvādīnāṃ hrasvaḥ]. (1508) bandh — in binding. badhnāti [n deleted by aniditāṃ hala upadhāyāḥ kṅiti (SK 415)]. babandhitha [liṭ, madhyama-puruṣa singular, with iṭ]. bandhā [luṭ; the t of tās becomes dh by jhaṣas tathor dho 'dhaḥ (SK 2280)]. bandhārau [prathama-puruṣa dual]. bhantsyati [lṛṭ; bh for b by ekāco baśo bhaṣ… (SK 326), t for dh by car]. badhāna [loṭ, madhyama-puruṣa singular; śānac by halaḥ śnaḥ śānaj jhau (SK 2557)]. abhānsīt [luṅ, prathama-puruṣa singular; vṛddhi of a hal-final root and bhaṣ]. By pūrvatrāsiddham (SK 12), the deletion of sic by jhalo jhali (SK 2281) [comes] before the bhaṣ-substitution. [Objection: once sic is deleted, the affix is s-initial by pratyayalakṣaṇa, so bhaṣ should follow. Response:] bhaṣ does not come on the strength of the s-initial affix through pratyayalakṣaṇa (SK 262), because the deletion of sic is asiddha with respect to pratyayalakṣaṇa. abāndhām [luṅ, prathama-puruṣa dual; dh for t by jhaṣas tathor dho 'dhaḥ]. abhāntsuḥ [prathama-puruṣa plural]. (1509) vṛṅ — in serving (saṃbhakti). vṛṇīte. vavre [liṭ, prathama-puruṣa singular]. vavṛṣe [madhyama-puruṣa singular]. vavṛḍhve [madhyama-puruṣa plural]. varitā, varītā [luṭ]. avariṣṭa, avarīṣṭa, avṛta [luṅ]. (1510) śranth — in releasing (vimocana) and in gladdening (pratiharṣa). From here on, [the roots are] parasmaipadin. [śrathnāti] [with the n deleted].

vārttika
श्रन्थिग्रन्थीत्यादिना कित्त्वपक्षे एत्वाभ्यासलोपावप्यत्र वक्तव्यौ इति हरदत्तादयः

'In the alternative where [liṭ] is kit by "śranthi-granthi…" [a rule of another grammar], e-substitution and deletion of the abhyāsa too must be stated here' — so Haradatta and others.

(वार्तिकम्) ॥ श्रेथतुः । श्रेथुः । इदं कित्त्वं पितामपीति सुधाकरमते । श्रेथिथ । अस्मिन्नपि पक्षे णलि । शश्राथ । उत्तमे तु । शश्राथ शश्रथेति माधवः । तत्र मूलं मृग्यम् । {$ {!1511 मन्थ!} विलोडने$} ।{$ {!1512 श्रन्थ!} {!1513 ग्रन्थ!} सन्दर्भे$} । अर्थभेदाच्छ्रन्धेः पुनः पाठः । रूपं तूक्तम् ।{$ {!1514 कुन्थ!} संश्लेषणे$} । संक्लेशे इत्येके । कुथ्नाति । चुकुन्थ । कुथेति दुर्गः । चुकोथ ।{$ {!1515 मृद!} क्षोदे$} । मृद्नाति । मृदान ।{$ {!1516 मृड!} च$} । अयं सुखेऽपि । ष्टुत्वम् । मृड्णाति ।{$ {!1517 गुध!} रोषे$} । गुध्नाति ।{$ {!1518 कुष!} निष्कर्षे$} । कुष्णाति । कोषिता ॥

śrethatuḥ, śrethuḥ [liṭ, prathama-puruṣa dual, plural]. In Sudhākara's view this kit-status [holds] even for the pit [endings]: śrethitha [madhyama-puruṣa singular]. In this view too, with ṇal: śaśrātha [prathama-puruṣa singular]. But in the uttama [-puruṣa singular]: śaśrātha, śaśratha — so Mādhava. The basis for that must be sought. (1511) manth — in churning (viloḍana). (1512) śranth, (1513) granth — in composing (sandarbha). śranth is read again because its meaning differs; its forms have [already] been stated. (1514) kunth — in clinging (saṃśleṣaṇa); some [read] 'in afflicting' (saṃkleśa). kuthnāti. cukuntha [liṭ]. Durga [reads the root as] kuth: cukotha. (1515) mṛd — in crushing (kṣoda). mṛdnāti. mṛdāna [loṭ, madhyama-puruṣa singular, with śānac]. (1516) mṛḍ — also [in crushing]. This [root] is also [used] in [the sense of] being happy (sukha). ṣṭutva [of the n of śnā]: mṛḍṇāti. (1517) gudh — in being angry. gudhnāti. (1518) kuṣ — in extracting (niṣkarṣa). kuṣṇāti. koṣitā [luṭ].

Tattvabodhinī
NSP p. 426
—हलः । संप्रसारणस्येत्यनुवर्तते । अङ्गस्येत्यवयवपृष्ठी । हल इति पञ्चमी तदेतदाह—अङ्गावयवादिति । अङ्गावयवात्किम् । निरुतम् । दुरुतम् । तदन्ताङ्गस्य किम् । विध्यति । क्षीष् हिंसायाम् । षित्त्वादङ् । क्षिया हेतिक्षियायाम् । क्षियाशीरित्यादौ क्षिया धर्मव्यतिक्रम आचारभेद इति वक्ष्यति ॥—दीर्घनिर्देशसामर्थ्यादिति । ननु दीर्घग्रहणेनाङ्गवृत्तपरिभाषा ज्ञाप्यत इत्याकरे स्थितम् । अन्यथा ज्ञाजनोर्ज्ञ विदध्यात् । ‘अतो दीर्घो यञि’ इति दीर्घसिद्धेः । तथा च दीर्घनिर्देशसामर्थ्यस्योपक्षयात्कथमिह ह्रस्वाभावसिद्धिरिति चेत् । अत्राहुः । ह्रस्वाभावे सत्येव दीर्घनिर्देश उक्तपरिभाषायां ज्ञापकः । सति तु ह्रस्वे पुनरङ्गकार्यं प्रवृत्तमेवेति न ज्ञापकः तथा च ह्रस्वाभावसिद्धौ न किञ्चिद्वाधकमिति । किञ्च ‘अङ्गवृत्ते पुनरङ्गवृत्तावविधिः’ इत्येव परिभाषाशरीरम् । तच्चोक्तज्ञापकात्सिद्धम् । यदि तु ‘भ्यसो भ्यम्’ इति सूत्रे भाष्ये परिनिष्ठितस्येत्येव पाठस्तर्हि स्वतन्त्रमेवेदं वचनं न तु ज्ञापकसिद्धमिति पक्षोऽपि युष्मदस्मच्छब्दगतमनोरमाग्रन्थादवगम्यते । तस्मिंश्च पक्षे दीर्घग्रहणसामर्थ्यं नोपक्षीणमिति सम्यगेवायं ग्रन्थ इति । केचित्तु संज्ञापूर्वकविधेरनित्यत्वान्न ह्रस्व इति व्याख्येयमित्याहुः ॥—बध्नातीति । ‘अनिदिताम्—’ इति नलोपः ॥—अभान्सीदिति । ‘वदव्रज—’ इति वृद्धिः । भष्भावः ॥—अबान्धामिति । ‘झषस्तथोः—’ इति तस्य धत्वम् । ‘झरो झरि’ इति वा लोपः । वृङ् संभक्तौ । भक्तिर्भजनम् । श्रन्थ ॥—कित्वपक्षे इति । श्रन्थिग्रन्थिदम्भिस्वञ्जीनां लिटः कित्वं वेति व्याकरणान्तरमित्युक्तत्वात्
Bālamanoramā
हलः - दिग्योगे पञ्चम्येषा । हलः परस्येति लभ्यते ।संप्रसारणस्ये॑ति सूत्रमनुवर्तते । अङ्गस्येत्यधिकृतमिहानुवृत्तमावर्त्त्यते । एकमवयवषष्ठन्तं हल इत्यत्रान्वेति । अङ्गावयवाद्धल इति लभ्यते । द्वितीयं तु स्थानषष्ठन्तं संप्रसारणेन विशेष्यते । तदन्तविधिः । 'ढ्रलोपे' इत्यतो दीर्घ इत्यनुवर्तते । तदाह — अङ्गावयवादित्यादिना । उदाहरणं तु — वीतः, शूनः, जीनः । हल इति किम् । उतः उतवान् । अङ्गावयवात् किम् । निरुतम् । तदन्ताङ्गस्येति किम् । विध्यति । एषामिति । व्रीभ्रीक्षीषामित्यर्थः । ल्वादय एव वृत्ताः, न तु प्वादय इति मते इति भावः । ज्ञा अवबोधने । शिति जादेशं स्मारयति — ज्ञाजनोर्जेति । दीर्घनिर्देशे॑ति । प्वादित्वेऽपि आकारोच्चारणसामर्थ्यान्न ह्रस्व इति भावः । बन्ध बन्धने । अनिट् । नोपधः । बध्नातीति ।अनिदिता॑मिति नलोप इति भावः । भारद्वाजनियमात्थलि वेडिति मत्वाऽऽह — बबन्धिथ बबन्द्धेति । अकित्त्वान्नलोपो न । अनिट्पक्षे तुझषस्तथो॑रिति थस्य धः । बन्धेति । तासि.झषस्तथो॑रिति तकारस्य धः । भन्त्स्यतीति । 'एकाचः' इति बस्य भष् । धस्य चर्त्वेन तः । बधानेति । 'हलः श्नः शानज्झौ' इति शानच् । अभान्त्सीदिति । हलन्तलक्षणावृद्धिः । भष्भावः । अबान्द्धामित्यत्र सकारपरकत्वाद्भष्भावमाशङ्क्याह — पूर्वत्रत्यादि । ननु कृते सिज्लोपे प्रत्ययलक्षणमाश्रित्य सकारपरकत्वाद्भष् दुर्वार इत्यत आह — प्रत्ययलक्षणेनेति । अबान्द्धामिति ।झषस्तथो॑रिति तस्य धः । श्रन्थ विमोचने । श्रन्थिग्रन्थीत्यादिनेति ।श्रन्थिग्रन्थिदम्भिस्वञ्जीना लिटः कित्त्वं वे॑ति व्याकरणान्तरेणेत्यर्थः । अस्मिन्नपि पक्षे इति ।पितामपीदं कित्त्व॑मिति सुधाकरमतेऽपि प्रथमपुरुषणलि — शश्राथ । उत्तमपुरुषणलि तु शश्राथ शश्रथेति माधव आहेत्यर्थः । तत्र मूलमिति ।कित्त्वपक्षे एत्त्वाभ्यासलोपा॑वित्यारभ्य 'माधव' इत्यन्तसंदर्भे मूलं नास्तीत्यर्थः । संयुक्तहल्मध्यस्थत्वादेवंविधवार्तिकाऽदर्शनाच्चेति भावः । कुष निष्कर्षे ।
translator notes
  • commentary BM: 'halaḥ' is a fifth-case word (pañcamyanta) in connection with direction (digyoga), giving 'after a hal'; the sūtra 'saṃprasāraṇasya' is carried forward (anuvartate); the adhikāra 'aṅgasya' is carried forward and applied twice (āvartyate): one [aṅgasya], a sixth-case word of the part (avayavaṣaṣṭhī), goes with 'halaḥ', giving 'after a hal that is a member of the aṅga'; the other, a sixth-case word of the place (sthānaṣaṣṭhī), is qualified by saṃprasāraṇa, with tadantavidhi; 'dīrghaḥ' is carried forward from 'ḍhralope'. Examples: vītaḥ, śūnaḥ, jīnaḥ; why 'halaḥ'? utaḥ, utavān; why 'aṅgāvayavāt'? nirutam; why 'tadantāṅgasya'? vidhyati. Followed: all the anuvṛttis are bracketed in sutra_en and k1.
  • commentary TB (sk.2559.t1): 'saṃprasāraṇasya' is carried forward (anuvartate); 'aṅgasya' is a sixth-case word of the part (avayavaṣaṣṭhī); 'halaḥ' is a fifth-case word (pañcamī) — hence 'aṅgāvayavāt'; why 'aṅgāvayavāt'? nirutam, durutam; why 'tadantāṅgasya'? vidhyati. Followed, as for BM.
  • commentary BM, at the tail of sk.2558.b1: jyā is aniṭ; with 'jyā nā ti' standing, samprasāraṇa by grahijyā and pūrvarūpa give 'ji nā ti', whereupon this sūtra applies. Followed: the bracketed derivation of jināti.
  • commentary BM on 'eṣām': of vrī, bhrī and kṣīṣ — on the view that only the lū-group is closed [at vṛt], not the pū-group. On jñā: he recalls the jā-substitution before a śit; 'dīrghanirdeśa' — even though [jñā] is in the pū-group, there is no short vowel by the force of pronouncing [jā] with ā. Followed: both brackets.
  • commentary TB (sk.2559.t1) on 'dīrghanirdeśasāmarthyāt': Objection — the source works (ākara) hold that the long vowel [of jā] makes known (jñāpayati) the aṅgavṛtta-paribhāṣā, since otherwise [Pāṇini] would have taught 'ja', the length coming by ato dīrgho yañi; that force being used up, how can it establish the absence of the short vowel here? Reply — the long vowel is a jñāpaka of that paribhāṣā only if there is no short vowel; were there a short vowel, an aṅga-operation would have re-applied and it would not be a jñāpaka; so nothing opposes the absence of the short vowel. Further, if the Bhāṣya on 'bhyaso bhyam' reads 'pariniṣṭhitasya', the paribhāṣā is an independent statement, not jñāpaka-based (a view known from the Manoramā on yuṣmad/asmad), and the force of the long vowel is not used up; so this text is right. Some say it should be explained as: no short vowel, because an injunction preceded by a technical name (saṃjñāpūrvaka-vidhi) is anitya. Recorded; the English keeps Bhaṭṭoji's wording.
  • commentary BM on bandh: aniṭ, n-penultimate; badhnāti — n-deletion by aniditām; taking iṭ before thal to be optional by the Bhāradvāja restriction, [BM gives] 'babandhitha babanddha' — no n-deletion because [thal is] not kit, and in the iṭ-less form dh for th by jhaṣas tathoḥ; bandhā — in tās, dh for t by jhaṣas tathoḥ; bhantsyati — bhaṣ for b by 'ekācaḥ', t for dh by car; badhāna — śānac by 'halaḥ śnaḥ śānaj jhau'; 'abhāntsīt' — vṛddhi of a hal-final [root], bhaṣ. Followed: the brackets on these forms; BM's pratīka reads abhāntsīt (see textual note).
  • commentary BM on 'pūrvatrāsiddham…': apprehending bhaṣ in abāndhām because [the root] is followed by s, [Bhaṭṭoji] says this; 'nanu …' — the objection that, once sic is deleted, bhaṣ is unavoidable through pratyayalakṣaṇa, is answered by 'pratyayalakṣaṇena…'; 'abānddhām' — dh for t by jhaṣas tathoḥ. Followed: the objection/response bracket and the dh.
  • commentary TB (sk.2559.t1): badhnāti — n-deletion by aniditām; abhānsīt — vṛddhi by vadavraja… [SK 2267], bhaṣ; abāndhām — dh for t by jhaṣas tathoḥ, and optional deletion by jharo jhari [savarṇe, SK 71]; vṛṅ saṃbhaktau — bhakti is bhajana. Followed: 'in serving' for saṃbhakti and the dh of abāndhām.
  • commentary TB (sk.2559.t1): kṣīṣ hiṃsāyām — aṅ because ṣit; [its noun] kṣiyā in hetikṣiyāyām, kṣiyāśīḥ and the like, where kṣiyā is transgression of dharma, a deviation of conduct, as will be said [later]. Recorded.
  • commentary BM: 'śranthi-granthi…' = 'śranthi-granthi-dambhi-svañjīnāṃ liṭaḥ kittvaṃ vā' — [a rule] of another grammar (vyākaraṇāntara). Followed: the bracket.
  • commentary TB (sk.2559.t1, tail) on 'kittvapakṣe': because 'the kit-status of liṭ is optional for śranthi, granthi, dambhi and svañji' is said to be [a rule of] another grammar. Followed, as for BM.
  • commentary BM on 'asminn api pakṣe': even in Sudhākara's view that this kit-status holds for the pit [endings] too, with the prathama-puruṣa ṇal — śaśrātha; with the uttama-puruṣa ṇal Mādhava says 'śaśrātha, śaśratha' [OCR of our BM: 'śaśrāya śaśraya', th misread as y; the kaumudī on the print, scan page-0426, reads 'śaśrātha śaśratheti']. On 'tatra mūlam': there is no basis for the whole passage from 'kittvapakṣe ettvābhyāsalopau' up to 'Mādhava', because [the a of śranth] stands between conjunct hals and because no such vārttika is seen. Followed: prathama/uttama labels; BM's wider scope for 'no basis' is recorded, the English keeps Bhaṭṭoji's 'tatra'.
  • textual Our text ends k1 '… || itaḥ parasmaipadinaḥ || śranthāti | |'; the print has 'itaḥ parasmaipadinaḥ | śrathnāti |' (NSP p. 416, scan page-0426, read; collation batch04, class A): the n is deleted before śnā, as in kuthnāti in the same column. The English gives [śrathnāti] in brackets.
  • textual The cross-reference after 'pvādīnāṃ hrasvaḥ' is 2258 in our text and '2258' in the print (scan page-0426, read); the sūtra is SK 2558 (sk2558 above; SK 2258 is ātmanepadeṣv anataḥ). The English gives SK 2558.
  • textual The cross-reference after 'vibhāṣā līyateḥ' is 1509 in our text; the print has '2509' (scan page-0426, read), and 6.1.51 is SK 2509 (SK 1509 is a taddhita sūtra). The English gives SK 2509.
  • textual Under bhrī our text reads 'bharaṇa ity ete'; the print reads 'bharaṇa ity eke' (NSP p. 416, scan page-0426, read; not in the collation report), matching 'bharaṇa ity eke' under vṝ in SK 2558. The English follows the print.
  • textual NSP footnote 1, printed at the foot of p. 416 (scan page-0426) on '[dīrgha]nirdeśasāmarthyāt', is OCR'd into the text of sk.2560.t1 ('niraku1 nirdeśasāmarthya…'). On the scan: 'nirdeśasāmarthyād iti — … aṅgavṛttaparibhāṣā … iti sūcayati iti śekharakāraḥ' — the author of the [Laghuśabdendu]śekhara takes the phrase as indicating something about the aṅgavṛtta-paribhāṣā; the middle words are covered by the digitizer's watermark and cannot be read.
  • textual The source files this sentence as a vārttika block and k3 opens with the editorial tag '(vārttikam)'. The print marks it with the ✻ sign (scan page-0426, read), but it is a dictum ascribed to Haradatta and others, and BM says no such vārttika is seen; it is therefore not labelled '(vārttika)' in the English, and the tag at the head of k3 is not rendered.
  • textual Our text reads 'arthabhedāc chrandheḥ'; the print has 'śranthe[ḥ]' (NSP p. 416, scan page-0426, read; collation batch04, class A); the root is śranth. The English follows the print.
  • textual abhānsīt is kept as transmitted (our text = the print, NSP p. 416, scan page-0426, read; TB's pratīka also abhānsīt). BM's pratīka is abhāntsīt, the form grammatically expected (vṛddhi, bhaṣ, then t for dh by khari ca before the s), parallel to Bhaṭṭoji's own bhantsyati and abhāntsuḥ. Mūla disagreement (ours/print/TB abhānsīt vs BM abhāntsīt) for Ajai's printed copy.
  • translator abāndhām is TB's form (with the optional deletion by jharo jhari); BM writes abānddhām. The dhātu numbers here run one lower than the print (print 1501 rī … 1519 kuṣ).
SK 2560 · A 7.2.46V
निरः कुषः ॥ २५६० ॥
niraḥ kuṣaḥ
niraḥ kuṣaḥ — After kuṣ preceded by nir, [a val-initial ārdhadhātuka takes iṭ (ārdhadhātukasyeḍ valādeḥ — carried forward (anuvṛtti) from SK 2184)] [optionally (vā — carried forward (anuvṛtti) from svaratisūtisūyatidhūñūdito vā, SK 2279)].

Rule. After kuṣ preceded by nir, an ārdhadhātuka affix beginning with a val sound takes iṭ optionally. "Takes iṭ" is carried from SK 2184, and "optionally" from SK 2279.

Example.

nir + kuṣ + tās (luṭ) → tās begins with t, a val sound, so iṭ is optional

→ with iṭ: nir + kuṣ + i + tās

→ with the usual sound changes (u → o, r → ṣ, tās → tā): niṣkoṣitā

→ without iṭ, with the same changes and ṣ + t → ṣṭ: niṣkoṣṭā.

luṅ with iṭ: nirakoṣīt. Without iṭ, ksa comes by SK 2336: nirakukṣat.

Applies when / blocked by. Why "after nir"? Without it the iṭ is not optional: koṣitā, koṣitum.

The roots that follow:

  • Kṣubh "to be agitated": kṣubhnāti, kṣubhnītaḥ. The n does not become ṇ, by SK 792. luṭ kṣobhitā; loṭ kṣubhāna.
  • Ṇabh and tubh "to injure": nabhnāti, tubhnāti. The same roots also take śap: nabhate, tobhate. And śyan: nabhyati, tubhyati.
  • Kliś (kliśū) "to torment": kliśnāti. SK 111 would make the n palatal after the ś (ścutva). But SK 112 (śāt) blocks that substitution after ś. So the n stays. luṭ kleśitā or kleṣṭā; luṅ akleśīt or aklikṣat.
  • Aś "to eat": aśnāti; liṭ āśa. The Bālamanoramā notes there is no nuṭ, since the root does not have two consonants.
  • Udhras "to glean". The u is a marker: dhrasnāti. Some say the u belongs to the root: liṭ udhrasāṃcakāra.
  • Iṣ "to repeat" (repetition or intensity): iṣṇāti.
  • Viṣ "to separate": viṣṇāti; luṭ veṣṭā.
  • Pruṣ and pluṣ "to moisten, serve, fill": pruṣṇāti, pluṣṇāti.
  • Puṣ "to nourish": luṭ poṣitā. It is seṭ. Only the puṣ with śyan is aniṭ.
  • Muṣ "to steal": luṭ moṣitā.
  • Khac "to be born late", after the due time has passed: khacñāti. After c the n of nā becomes ñ by ścutva. Some say this root ends in v: khav. That leads into the next rule.

On iṣ, Bhaṭṭoji reports a view and corrects it. SK 2340 makes iṭ optional before t for iṣ, sah and others. "They say" it means only the iṣ of the tudādi, because sah there is the bhvādi root; on that view iṣṇāti has only eṣitā. Bhaṭṭoji says the option is proper for iṣṇāti too. A vārttika supports him: the option of SK 2340 is withheld before t only from the iṣ that takes śyan.

Siddhānta-kaumudī
निरः परात्कुषोवलादेरार्धधातुकस्य इड्वा स्यात् । निष्कोषिता । निष्कोष्टा । निरकोषीत् । निरकुक्षत् ।{$ {!1519 क्षुभ!} संचलने$} । क्षुभ्नादिषु च SK 792 क्षुभ्नाति । क्षुभ्नीतः । क्षोभिता । क्षुभान ।{$ {!1520 णभ!} {!1521 तुभ!} हिंसायाम्$} । नभ्नाति । तुभ्नाति । नभते तोभते इति शपि । नभ्यति तुभ्यतीति श्यनि ।{$ {!1522 क्लिशू!} विबाधने$} । शादिति श्चुत्वनिषेधः । क्लिश्नाति । क्लेशिता । क्लेष्टा । अक्लेशीत् । अक्लिक्षत् ।{$ {!1523 अश!} भोजने$} । अश्नाति । आश ।{$ {!1524 उध्रस!} उञ्छे$} । उकार इत् । ध्रस्नाति । उकारो धात्ववयव इत्येके । उध्रसांचकार ।{$ {!1525 इष!} आभीक्ष्ण्ये$} । पौनः पुन्यं भृशार्थो वा आभीक्ष्ण्यम् । इष्णाति । तीषसह - SK 2340 इत्यत्र सहिना साहचर्यादकारविकरणस्य तौदादिकस्यैव इषेर्ग्रहणं नतु इष्यतीष्णात्योरित्याहुः । एषिता । वस्तुतस्तु इष्णातेरपि इड्विकल्प उचितः । तथा च वार्तिकम् ॥

[— 'ārdhadhātukasyeḍ valādeḥ' (SK 2184) and 'vā' (from svaratisūti…, SK 2279) carried forward (anuvṛtti) —] A val-initial ārdhadhātuka after kuṣ preceded by nir takes iṭ optionally. niṣkoṣitā, niṣkoṣṭā [luṭ, with and without iṭ]. nirakoṣīt [luṅ with iṭ]. nirakukṣat [without iṭ, ksa by śala igupadhād aniṭaḥ ksaḥ (SK 2336)]. (1519) kṣubh — in being agitated (saṃcalana). [No ṇ for the n by] kṣubhnādiṣu ca (SK 792): kṣubhnāti, kṣubhnītaḥ. kṣobhitā [luṭ]. kṣubhāna [loṭ, madhyama-puruṣa singular]. (1520) ṇabh, (1521) tubh — in injuring. nabhnāti, tubhnāti. With śap: nabhate, tobhate. With śyan: nabhyati, tubhyati. (1522) kliśū — in tormenting (vibādhana). By śāt (SK 112), the palatal substitution (ścutva) [of the n of śnā, by stoḥ ścunā ścuḥ (SK 111)] is prohibited: kliśnāti. kleśitā, kleṣṭā [luṭ, with and without iṭ]. akleśīt, aklikṣat [luṅ, with and without iṭ]. (1523) aś — in eating. aśnāti. āśa [liṭ]. (1524) udhras — in gleaning (uñcha). The u is an it: dhrasnāti. Some [say] the u is part of the root: udhrasāṃcakāra [liṭ]. (1525) iṣ — in repetition (ābhīkṣṇya). ābhīkṣṇya is repetition (paunaḥpunya) or intensity. iṣṇāti. In tīṣasaha… (SK 2340), because of its association (sāhacarya) with sah, only the tudādi iṣ, whose vikaraṇa is a, is taken, not [the iṣ of] iṣyati and iṣṇāti — so they say: eṣitā. In fact, however, optional iṭ is proper for iṣṇāti too; and so there is the vārttika:

vārttika
इषेस्तकारे श्यन्प्रत्ययात्प्रतिषेध इति

(vārttika) 'For iṣ, before t, the prohibition [of the optional iṭ of tīṣasaha…] is [for the iṣ] that takes the affix śyan.'

(वार्तिकम्) ॥{$ {!1526 विष!} विप्रयोगे$} । विष्णाति । वेष्टा ।{$ {!1527 प्रुष!} {!1528 प्लुष!} स्नेहनसेवनपूरणेषु$} । प्रुष्णाति । प्लुष्णाति ।{$ {!1529 पुष!} पुष्टौ$} । पोषिता ।{$ {!1530 मुष!} स्तेये$} । मोषिता ।{$ {!1531 खच!} भूतप्रादुर्भावे$} । भूतप्रादुर्भावोऽतिक्रान्तोत्पत्तिः । खच्ञाति । वान्तोऽयमित्येके ॥

(1526) viṣ — in separating (viprayoga). viṣṇāti. veṣṭā [luṭ]. (1527) pruṣ, (1528) pluṣ — in moistening, serving and filling. pruṣṇāti, pluṣṇāti. (1529) puṣ — in nourishing. poṣitā [luṭ; seṭ]. (1530) muṣ — in stealing. moṣitā [luṭ]. (1531) khac — in bhūtaprādurbhāva. bhūtaprādurbhāva is being born after [the due time] has passed (atikrāntotpatti). khacñāti. Some [say] this [root] ends in v.

Tattvabodhinī
NSP p. 426–427
—निरः कुषः । ‘स्वरति—’ इति सूत्राद्वेत्यनुवर्तते ‘आर्धधातुकस्येद—’ इत्यधिक्रियत एव तदाह—वलादेरित्यादि । निरः किम् । कोषिता । कोषितुम् ॥—निरकु१ निर्देशसामर्थ्यविसिद्धिः अङ्गवृत्तपरिभाषा निष्फलेति सूचयति इति शेखरकारः । क्षदिति । इडभावे क्सः ॥—सहिना साहचर्यादिति । यद्यपि षह षुह चक्यर्थे इति दिवादिरपि सहिरस्ति तथापि सहेति शपा निर्देशाद् भौवादिकेनैव सहिना साहचर्यमित्याहुः ॥—श्यन्प्रत्ययादिति । बहुत्रीहिरयम् । श्यन्प्रत्ययवत इषेस्तकारे परत इड्विकल्पप्रतिषेध इति वार्तिकार्थः । तथा च दैवादिकाद्विकल्पो नेति फलितम् । एवं च ‘तीषसह—इत्यत्र इष्णातेरपि ग्रहणमिति बोध्यम् ॥—पोषितेति । अनिट्केषु पुष्येति श्यना निर्देशादयं सेडिति भावः
Bālamanoramā
निरः कुषः - निरः कुषः ।आर्धधातुकस्येड्वलादे॑रित्यनुवर्तते ।स्वरतिसूती॑त्यतो वेति च । तदाह — निरः परादिति । निरकुक्षदिति । इडभावपक्षे 'शल' इति क्स इति भावः । अश भोजने । आशेति । द्विहल्त्वाऽभावान्न नुडिति भावः । इष आभीक्ष्ण्ये । तासितीषसहे॑तीड्विकल्पमाशङ्क्याह — तीषसहेत्यत्रेति । सहेति शपा निर्देशबलेन भौवादिक एव सहधातुरत्र निर्दिष्टः । तत्साहचर्यात्तौदादिकस्यैव इषेग्र्रहणम्, अकारविकरणसामान्यादिति भावः । इषेस्तकारे श्यन्प्रत्ययादिति । श्यन् प्रत्ययो यस्मादिति बहुव्रीहिः । श्यन्विकरणपठितादिषेस्तकारे परेतीषसहे॑ति विधिर्नेत्यर्थः । इष्णातेस्तकारे इड्विकल्पः फलित इति भावः । प्रूष प्लुषेति । आद्यो दीर्घोपधः । पुष्ट पुष्टौ । पोषितेति । अनिट्सु श्यन्विकरणस्यैव पुषेग्र्रहणादयं सेडिति भावः ।पुषादिद्युतादी॑त्यत्र श्यन्विकरमपुषादेरेव ग्रहणादङ् न । खच भूताप्रादुर्भावे । अतिक्रान्तोत्पत्तिरिति । दशममासादौ उत्पत्तियोग्यस्य एकादशादिमासादिषूत्पत्तिरित्यर्थः । खच्ञातीति । नस्य श्चुत्वेन ञ इति भावः । वान्तोऽयमिति ।खव भूतप्रादुर्भावे॑ इत्येवं दन्त्योष्ठआन्तमेके पठन्तीत्यर्थः ।
Footnotes (NSP 1915 ed.)
NSP p. 427 १ एकाच इति—हरदत्तानुरोधेनेदमुक्तम् । अग्रहीष्टामित्यादौ लावस्थायामडिति पक्षे एकाचो विहितत्वाभावाद्दीर्घाप्राप्त्या असंगतं हरदत्तोक्तमिति बोध्यम् । २ स्वार्थे इति—करणे इत्यस्य तु नानुवृत्तिः, विच्छिन्नत्वादिति भावः ।
translator notes
  • commentary BM: 'ārdhadhātukasyeḍ valādeḥ' is carried forward (anuvartate), and 'vā' from 'svaratisūti…'; hence Bhaṭṭoji's 'niraḥ parāt…'. Followed: both bracketed in sutra_en and k1.
  • commentary TB (sk.2560.t1): 'vā' is carried forward (anuvartate) from the sūtra 'svarati…'; 'ārdhadhātukasyeṭ' is in force as an adhikāra (adhikriyata eva); hence 'valādeḥ'. Why 'niraḥ'? koṣitā, koṣitum. Followed, as for BM.
  • commentary BM on nirakukṣat: in the iṭ-less alternative, ksa by 'śalaḥ'. On aś: āśa — no nuṭ, since [the root] does not have two hals. Followed: the ksa bracket.
  • commentary TB (sk.2560.t1) on '[niraku]kṣat': ksa in the absence of iṭ. Followed, as for BM.
  • commentary BM on 'tīṣasaha…': apprehending optional iṭ in tās by tīṣasaha…, he says this; 'saha' is named with śap, so by that force the bhvādi sah is meant; by association with it only the tudādi iṣ is taken, sharing an a-vikaraṇa. Followed: 'whose vikaraṇa is a'.
  • commentary TB (sk.2560.t1) on 'sahinā sāhacaryāt': although there is also a divādi sah ('ṣaha ṣuha cakyarthe'), sah is named with śap, so the association is with the bhvādi sah — so they say. Recorded.
  • commentary BM on the vārttika: 'śyanpratyayāt' is a bahuvrīhi — [iṣ] from which the affix śyan [comes]; the injunction of tīṣasaha does not apply before t to the iṣ read with śyan as vikaraṇa; the upshot is optional iṭ for iṣṇāti before t. Followed: 'that takes the affix śyan'.
  • commentary TB (sk.2560.t1): 'śyanpratyayāt' is a bahuvrīhi; the vārttika means: for iṣ having the affix śyan, before t, the optional iṭ is prohibited; so the divādi [iṣ] gets no option; and thus iṣṇāti too is taken in tīṣasaha. Followed, as for BM.
  • commentary BM on 'prūṣa pluṣa': the first has a long penultimate (dīrghopadha). On poṣitā: seṭ, because among the aniṭ roots only the puṣ with śyan is taken; nor does aṅ come by puṣādidyutādi…, since only the śyan-class puṣ is taken there. Recorded (the English keeps pruṣ, see textual note); followed for '[seṭ]'.
  • commentary TB (sk.2560.t1) on poṣitā: seṭ, because in the aniṭ list [puṣ] is named with śyan as 'puṣya'. Followed, as for BM.
  • commentary BM on atikrāntotpatti: the birth, in the eleventh month or later, of one fit to be born in the tenth month or so; khacñāti — ñ for the n [of śnā] by ścutva; 'vāntaḥ' — some read [the root] as 'khav bhūtaprādurbhāve', ending in the dento-labial. Followed: 'being born after [the due time] has passed'.
  • textual Our k3 opens with the editorial tag '(vārttikam) ||', which is not rendered; the print marks the vārttika with the ✻ sign before it (NSP p. 417, scan page-0427, read). The print reads 'pruṣa … pruṣṇāti' with short u, as ours (scan read); BM's pratīka is 'prūṣa' and calls it dīrghopadha.
  • textual This unit's editorial_notes (n1) carry two NSP footnotes printed at the foot of p. 417 (scan page-0427): 1 'ekāca iti' belongs to SK 2562, 2 'svārthe iti' to SK 2563. The TB OCR of sk.2560.t1 also contains, after 'niraku-', the text of NSP footnote 1 of p. 416 ('nirdeśasāmarthya… iti śekharakāraḥ'), which belongs to SK 2559 (dīrghanirdeśasāmarthyāt); it is not TB. The print otherwise matches our text in this unit (collation batch04: clean).
  • translator The dhātu numbers run one lower than the print (print 1520 kṣubh … 1532 khac). 'ścutva' is the substitution of the palatal class by stoḥ ścunā ścuḥ (SK 111), which śāt (SK 112) blocks after ś.
SK 2561 · A 6.4.19V
च्छ्वोः शूडनुनासिके च ॥ २५६१ ॥
cchvoḥ śūḍanunāsike ca
cchvoḥ śūḍ anunāsike ca — In place of ch [accompanied by tuk] and of v, ś and ūṭh [respectively] are substituted before a nasal, and [before kvi and before a jhal-initial kṅit (kvijhaloḥ kṅiti — carried forward (anuvṛtti) by 'ca' from anunāsikasya kvijhaloḥ kṅiti, SK 2666)].

Rule. Ś replaces ch together with its augment t (tuk). Ūṭh replaces v. Both happen before a nasal. By ca "and", they also happen before the affix kvi and before a kit or ṅit affix beginning with a jhal. That condition is drawn from SK 2666, a rule taught later.

Example.

khav + nā + ti → the n of nā is a nasal, so ūṭh replaces v

→ the ṭh of ūṭh is a marker: kha + ū + nā + ti

→ a + ū becomes au by vṛddhi (SK 73)

→ khaunāti.

liṭ cakhāva; luṭ khavitā.

khav + nā + hi → khav ends in a consonant, so śānac (SK 2557) might come

→ but ūṭh is the later rule (para), so it applies first: kha + ū + nā + hi

→ the root no longer ends in a consonant, so there is no śānac

→ the ā of nā becomes ī before hi (SK 2497); a + ū → au (SK 73)

→ khaunīhi.

Applies when / blocked by. The c in the rule's cch stands for the augment t before ch. So ś replaces the whole tch. Bhaṭṭoji's commentators give praśnaḥ and viśnaḥ, from pracch and vicch with the affix naṅ. Because the t goes too, it is not heard.

Watch out. Before hi, do not make āna for khav. The order of rules removes the final consonant first.

The roots that follow:

  • Heṭh, also "to be born late": hiṭhṇāti, with ṣṭutva of the n. Some copies read the root as hiṭh; scholars discuss this.
  • Grah "to take". It has a svarita vowel as marker, so it takes both sets of endings. Before the ṅit śnā its r becomes ṛ by saṃprasāraṇa (SK 2412), then pūrvarūpa: gṛhṇāti, gṛhṇīte. The Bālamanoramā adds the liṭ: jagrāha, jagṛhatuḥ, … jagṛhe.
Siddhānta-kaumudī
सतुक्कस्य छस्य वस्य च क्रमाच्छ ऊठ् एतावादेशौ स्तोऽनुनासिके क्वौ झलादौ च क्ङिति । खौनाति । चखाव खविता । शानचः परत्वादूठि कृते हलन्तात्वाभावान्न शानच् । खौनीहि{$ {!1532 हेठ!} च$} । ष्टुत्वम् । हिठ्णाति ।{$ {!1533 ग्रह!} उपादाने$} । स्वरितेत् । ग्रहिज्या - SK 2412 । गृह्णाति । गृह्णीते ॥

In place of ch accompanied by tuk, and of v, the substitutes ś and ūṭh respectively come before a nasal, before kvi, and before a jhal-initial kṅit [— 'kvijhaloḥ kṅiti' drawn in (anuvṛtti) by 'ca' from anunāsikasya kvijhaloḥ kṅiti (SK 2666)]. khaunāti [khav + nā + ti → ūṭh for v → kha + ū + nā + ti → vṛddhi by etyedhatyūṭhsu (SK 73) → khaunāti]. cakhāva [liṭ]. khavitā [luṭ]. [With khav + nā + hi standing,] since [ūṭh] is later (para) than śānac, once ūṭh has been made [the root] no longer ends in a hal, and so there is no śānac [by halaḥ śnaḥ śānaj jhau, SK 2557]: khaunīhi [loṭ, madhyama-puruṣa singular]. (1532) heṭh — also [in bhūtaprādurbhāva]. ṣṭutva [of the n of śnā]: hiṭhṇāti. (1533) grah — in taking. It is svaritet [and so ubhayapadin]. [samprasāraṇa before a kṅit by] grahijyā… (SK 2412): gṛhṇāti, gṛhṇīte.

Tattvabodhinī
alignment inferredNSP p. 427
—च्छ्वोः शूडनु—॥ च्छ्वोरिति सतुकस्यानुकरणम् । चुत्वेन तकारस्थाने चकार इत्याह—सतुकस्येति । एवं च सूत्रे छकारात्पूर्व चकारो ज्ञेयः । सतुकस्य किम् । विच्छप्रच्छाभ्यां नङ्प्रत्यये छकारस्य शकारे कृते तुकोऽश्रवणं यथा स्यात् । अकृतव्यूह-परिभाषा त्वनित्या अस्मादेव सतुकग्रहणात् । तेनातिक्रान्तो भवकन्तमतिभवकानित्यादि सिद्धमित्याहुः । सूत्रस्थेन चश-ब्देन क्विझलोः क्ङितीत्यनुकृष्यत इत्याह—क्वावित्यादि । ग्रहो लिटि दीर्घः ॥ एकाचः किम् । जाग्रहिता । जाग्रहिष्यति । इत्यादौ यङ्छुकि माभूत् । विहितस्येति किम् । ग्राहितम् । न चात्र णिलोपस्य स्थानिवद्भावाद्ग्रहेः परत्वं नेति वाच्यम् । दीर्घविधौ स्थानिवत्त्वनिषेधात् । त्रैपादिकदीर्घविधावेव स्थानिवत्त्वनिषेध इत्यभ्युपगमे तु विहितविशेषणं व्यर्थमेवेत्याहुः । प्रकृतस्येटो ग्रहणात् ग्राहिता ग्राहिष्यत इत्यादौ चिण्वदिटो न दीर्घः ॥—ग्रहीषीष्टेति । वॄत इत्यनुवृत्तेः ‘न लिङि’ इति दीर्घनिषेधो न प्रवर्तते
Bālamanoramā
च्छ्वोः शूडनुनासिके च - च्छ्वोः शूडनुनासिके च । च्छ् व् अनयोद्र्वन्द्वात्षष्ठीद्विवचनम् । छकारात्प्राक्तुकः श्चुत्वेन चकारस्य निर्देशः । 'शूठ' इति छेदः । श् ऊठ् अनयोः समाहारद्वन्द्वात्प्रथमा । चकारेण क्विझलोः क्ङितीत्यनुवर्तते । तदाह — सतुक्कस्येत्यादि । यथा विश्नः प्रश्नः । विच्छधातोः प्रच्छधातोश्च औणादिके नङ्प्रत्यये अन्तरङ्गत्वात्छे चे॑ति तुकि कृते सति सतुक्कस्य छस्य शकार आदेशः । वकारस्य ऊठमुदाहरति — खौनातीति । खव् ना तीति स्थिते वकारस्य ऊठ, ठकार इत् 'एत्येधत्यूठ्सु' इति वृद्धिरिति भावः । खौनीतः खौनन्तीत्यादि । खव् ना हि इति स्थितेहलः श्नश्शानज्झौ॑इतिशानजादेशमाशङ्क्याअह — शानचः परत्वादिति । हेठ चेति । हेठधातुरपि भूतप्रादुर्भावे वर्तते इत्यर्थः । ग्रह उपादाने इति । अदुपधः । स्वरितेत्त्वादुभयपदी । किति ङिति च संप्रसारणं स्मारयति — ग्रहिज्येति । गृह्णाति गृह्णीते इति । श्नाप्रत्ययस्य ङित्त्वाद्रेफस्य संप्रसारणे पूर्वरूपमिति भावः । णलि द्वित्वे कृतेलिटभ्यासस्ये॑ति संप्रसारणे उरदत्त्वे रपरत्वे हलादिशेषे अभ्यासचुत्वे उपधावृद्धिः । जग्राह.अतुसादौ क्ङति परत्वात्ग्रहिज्ये॑ति संप्रसारणे कृते द्वित्वादि । जगृहतुः जगृहुः । जग्रहिथ । जगृहुथुः जगृह । जग्राह-जग्रह जगृहिव जगृहिम । जगृहे जगृहाते जगृहिरे । जगृहिषे जगृहाथे जगृहिढ्वे-जगृहिध्वे । जगृहे जगृहिवहे जगृहिमहे ।
translator notes
  • commentary BM: 'cchvoḥ' is a sixth-case dual (ṣaṣṭhīdvivacana) after a dvandva of ch and v; the c before ch names the tuk [t] that precedes ch, with ścutva; 'śūṭh' is split ś + ūṭh, a first-case word (prathamā) of a samāhāra-dvandva; by 'ca', 'kvijhaloḥ kṅiti' is carried forward (anuvartate); hence 'satukkasya…'. Examples [of ś]: viśnaḥ, praśnaḥ — from vicch and pracch with the uṇādi affix naṅ, tuk being made first by che ca as antaraṅga, ś replaces ch together with tuk. Followed: '[accompanied by tuk]', 'respectively' and the anuvṛtti bracket.
  • commentary TB (sk.2561.t1): 'cchvoḥ' is the imitation (anukaraṇa) of [ch] with tuk, c standing for t by cutva; so the c before ch in the sūtra is to be understood thus. Why 'with tuk'? So that, when ś replaces ch in vicch and pracch with the affix naṅ, the tuk is not heard. The akṛtavyūha-paribhāṣā is anitya precisely because of this mention of tuk; hence atibhavakān etc. are established, they say. By the 'ca' in the sūtra, 'kvijhaloḥ kṅiti' is drawn in (anukṛṣyate). Followed, as for BM.
  • commentary BM on khaunāti: with 'khav nā ti' standing, ūṭh for v, ṭh being it, vṛddhi by 'etyedhatyūṭhsu'; also khaunītaḥ, khaunanti etc. With 'khav nā hi' standing, apprehending śānac by 'halaḥ śnaḥ śānaj jhau', he says 'śānacaḥ paratvāt'. Followed: the bracketed derivations of khaunāti and khaunīhi.
  • commentary BM on 'heṭha ca': the root heṭh too has the sense of bhūtaprādurbhāva. On grah: a-penultimate; ubhayapadin because svaritet; recalling samprasāraṇa before kit and ṅit by grahijyā; gṛhṇāti — samprasāraṇa of the r because śnā is ṅit, and pūrvarūpa. BM then derives the liṭ: jagrāha (ṇal: doubling, samprasāraṇa of the abhyāsa by liṭy abhyāsasya, a for ṛ with r-augment, halādiśeṣa, cutva of the abhyāsa, penultimate vṛddhi); before atus etc., grahijyā first as the later rule, then doubling: jagṛhatuḥ, jagṛhuḥ, jagrahitha, jagṛhathuḥ, jagṛha, jagrāha/jagraha, jagṛhiva, jagṛhima; jagṛhe, jagṛhāte, jagṛhire, jagṛhiṣe, jagṛhāthe, jagṛhiḍhve/jagṛhidhve, jagṛhe, jagṛhivahe, jagṛhimahe. Followed: '[in bhūtaprādurbhāva]', '[and so ubhayapadin]', and the samprasāraṇa bracket; the paradigm is BM's, not Bhaṭṭoji's.
  • textual Our text reads 'heṭha ca … hiṭhṇāti'; the print reads 'hiṭha ca | ṣṭutvam | hiṭhṇāti' (NSP p. 417, scan page-0427, read; not in the collation report, which lists 2561 as clean), so that root and form agree. BM reads 'heṭha'. Our text is thus internally inconsistent (a root heṭh cannot yield hiṭhṇāti). The English gives the root as our text has it and the form as transmitted in both. Mūla disagreement (ours/BM heṭha vs print hiṭha) for Ajai's printed copy.
  • textual The print numbers the roots 1533 hiṭha, 1534 grah (ours 1532, 1533). The OCR page text of scan page-0427 omits the head of this unit; the scan reads 'satukkasya chasya vasya ca kramāc cha ūṭh etāv ādeśau sto 'nunāsike kvau jhalādau ca kṅiti | khaunāti | cakhāva khavitā | śānacaḥ … khaunīhi', matching ours.
  • textual TB on the next sūtra (graho 'liṭi dīrghaḥ, SK 2562) is printed within sk.2561.t1, after the TB on this sūtra.
SK 2562 · A 7.2.37V
ग्रहोऽलिटि दीर्घः ॥ २५६२ ॥
graho'liṭi dīrghaḥ
graho 'liṭi dīrghaḥ — For [the iṭ (iṭ — carried forward (anuvṛtti) from ārdhadhātukasyeḍ valādeḥ, SK 2184, and turned into a sixth-case word)] after [monosyllabic (ekācaḥ — carried forward (anuvṛtti) from ekāca upadeśe 'nudāttāt, SK 2246)] grah, a long vowel [is substituted], but not in liṭ.

Rule. A long vowel replaces the iṭ that is enjoined after grah of one syllable. This does not happen in liṭ. Iṭ is carried from SK 2184 and "of one syllable" (ekācaḥ) from SK 2246.

Example.

grah + tās (luṭ) → iṭ is added: grah + i + tās

→ the iṭ follows grah, outside liṭ, so it becomes long: grah + ī + tās

→ with the usual sound changes (tās → tā): grahītā.

liṭ: jagrah + i + tha → the iṭ stays short → jagrahitha.

Applies when / blocked by.

  • Why "of one syllable"? So the rule does not apply in the yaṅ-luk form jāgrahitā. Scholars discuss this.
  • Why "enjoined after grah"? In the causative form grāhitam, ṇi is lost. The iṭ then stands after grah, but it was not enjoined after grah. So it is not lengthened.
  • āśīrliṅ ātmanepada grahīṣīṣṭa. SK 2529 forbids a long iṭ before liṅ, but only for vṝ and the ṝ-final roots. So it does not stop this length.

The other forms:

  • āśīrliṅ parasmaipada gṛhyāt;
  • luṅ: no vṛddhi, by SK 2299: agrahīt, agrahīṣṭām;
  • luṅ ātmanepada: agrahīṣṭa, agrahīṣātām, agrahīṣata.

Watch out. The length belongs to the iṭ, not to the root. Grah keeps its short a: grahītā, agrahīt.

Here end the kryādi roots.

Siddhānta-kaumudī
एकाचो ग्रहेर्विहितस्येटो दीर्घः स्यान्नतु लिटि । ग्रहीता । लिटि तु जग्रहिथ । गृह्यात् । ग्रहीषीष्ट । ह्म्यन्त - SK 2299 इति न वृद्धिः । अग्रहीत् । अग्रहीष्टाम् । अग्रहीष्ट । अग्रहीषाताम् । अग्रहीषत ॥

[— 'iṭ' (SK 2184) and 'ekācaḥ' (SK 2246) carried forward (anuvṛtti) —] For the iṭ enjoined after monosyllabic grah, a long vowel comes — but not in liṭ. grahītā [luṭ]. But in liṭ: jagrahitha [madhyama-puruṣa singular]. gṛhyāt [āśīrliṅ parasmaipada]. grahīṣīṣṭa [āśīrliṅ ātmanepada; the prohibition of na liṅi (SK 2529) does not apply]. By hmyantakṣaṇa… (SK 2299) there is no vṛddhi: agrahīt [luṅ parasmaipada, prathama-puruṣa singular]. agrahīṣṭām [prathama-puruṣa dual]. agrahīṣṭa [luṅ ātmanepada, prathama-puruṣa singular]. agrahīṣātām [prathama-puruṣa dual]. agrahīṣata [prathama-puruṣa plural]. [Here end the kryādi roots.]

Tattvabodhinī
alignment inferredNSP p. 427
—अग्रहीषातामिति । लुङात्मनेपदद्विवचनम् ॥ इति क्र्यादयः ॥
Bālamanoramā
ग्रहोऽलिटि दीर्घः - ग्रहोऽलिटि दीर्घः ।ग्रहट इति दिग्योगे पञ्चमी । 'आर्धधातुकस्येट्' इत्यत इडित्यनुवृत्तं षष्ठन्तं विपरिणम्यते,एकाच उपदेशेऽनुदात्ता॑ दित्यत एकाचेति च । तदाह — एकाच इत्यादि । एकाचः किम । यङ्लुकि माभूत् । जाग्रहिता । विहितस्येति किम् । ग्रहितम् । णिलोपे कृते ग्रहधातोः परत्वेऽपि विहतत्वाऽभावादिटो न दीर्घः । न च णिलोपस्य स्थानिवत्त्वान्न ग्रहधातोः पर इडिति वाच्यं, दीर्घकित्त्वात्संप्रसारणमिति भावः । लिङस्तङ्याह — ग्रहीषीष्टेति । 'ग्रहोऽलिटी' ति दीर्घः ।न लिङी॑ति इटो दीर्घनिषेधस्तु न, तत्रवृतट इत्यनुवृत्तेः । अग्रहीष्टामिति ।ग्रहोऽलिटी॑ति दीर्घः । अग्रहीषुः । अग्रहीरित्यादि । लुङस्तङ्याह — अग्रहीष्टेति ।ग्रहोऽलिटी॑ति दीर्घः । अग्रहीषतेति । अग्रहीष्ठा इत्यादि सुगमम् । इति क्र्यादयः ।॥ इति बालमनोरमायाम् क्र्यादयः॥अथ तिङन्ते लकारार्थप्रक्रिया ।अथ लकारार्थप्रक्रिया निरूप्यते ।
translator notes
  • commentary BM: 'grahaḥ' is a fifth-case word (pañcamī) in connection with direction (digyoga); 'iṭ', carried forward (anuvṛtta) from 'ārdhadhātukasyeṭ', is changed (vipariṇamyate) into a sixth-case word (ṣaṣṭhyanta), and 'ekācaḥ' [is carried forward] from 'ekāca upadeśe 'nudāttāt'; hence 'ekācaḥ…'. Followed: both anuvṛttis bracketed in sutra_en and k1.
  • commentary BM: why 'ekācaḥ'? So that it does not apply in yaṅluk: jāgrahitā. Why 'vihitasya'? grāhitam [from the ṇic-form; our BM OCR reads 'grahitam', TB (sk.2561.t1) 'grāhitam']: after ṇi-deletion the iṭ follows grah but was not enjoined after it, so it is not lengthened; [the objection that] by sthānivadbhāva of the ṇi-deletion the iṭ does not follow grah is not to be raised [the reason BM then gives is garbled in the OCR and is not rendered]. Recorded.
  • commentary TB, printed under sk.2561.t1: why 'ekācaḥ'? jāgrahitā, jāgrahiṣyati — so that it does not apply in yaṅluk; why 'vihitasya'? grāhitam; nor is it to be said that by sthānivadbhāva of the ṇi-deletion [the iṭ] is not after grah, since sthānivadbhāva is prohibited for a lengthening rule; on the view that the prohibition holds only for lengthening rules of the tripādī, the qualifier 'vihita' is simply superfluous, they say; since the iṭ under discussion is meant, in grāhitā, grāhiṣyate the ciṇvat iṭ is not lengthened. Recorded.
  • commentary BM on grahīṣīṣṭa: length by 'graho 'liṭi'; the prohibition of the iṭ's length by 'na liṅi' does not apply, because 'vṝtaḥ' is carried forward there. agrahīṣṭām and agrahīṣṭa — length by graho 'liṭi; BM adds agrahīṣuḥ, agrahīḥ; 'agrahīṣṭhāḥ etc. are easy'. It closes 'iti kryādayaḥ || iti bālamanoramāyāṃ kryādayaḥ ||'. Followed: the bracket on grahīṣīṣṭa.
  • commentary TB, printed under sk.2561.t1 (tail) and sk.2562.t1: grahīṣīṣṭa — because 'vṝtaḥ' is carried forward, the prohibition of length by 'na liṅi' does not operate; agrahīṣātām — luṅ ātmanepada dual; then 'iti kryādayaḥ' (OCR 'iti myādayaḥ'). Followed, as for BM.
  • textual Our kaumudī ends 'agrahīṣātām | agrahīṣata ||'; the print continues 'agrahīṣata | iti kryādayaḥ ||' and then heads the next section 'tiṅante curādayaḥ 50' before '1535 cura steye' (NSP p. 417, scan page-0427, read; collation batch04, class M). The colophon is supplied in brackets in the English; both BM and TB also close with 'iti kryādayaḥ'.
  • textual NSP footnote 1, printed at the foot of p. 417 (scan page-0427) on 'ekācaḥ' (the marker stands on 'ekāco' in the scan), filed under sk.2560.n1: 'ekāca iti — this is said in deference to Haradatta; in agrahīṣṭām and the like, on the view that aṭ [comes] at the lakāra stage, [the iṭ] is not enjoined after a monosyllable, so the lengthening would not be obtained; hence what Haradatta says is inconsistent.'
  • textual The TB on the first curādi sūtra (SK 2563, 'satyāpapāśa…') that followed TB's colophon in sk.2562.t1 was re-filed to sk.2563.t1 on 2026-10-05.
← all prakaraṇasoutline →