Vaiyākaraṇa-siddhānta-kaumudī — tiṅantajuhotyādiprakaraṇam

Learner edition · Prakaraṇa 46

46. Tiṅanta juhotyādi-prakaraṇa — Verbs: the hu class, whose roots are doubled before the endings

This chapter covers the third class of roots in the Dhātupāṭha, the juhotyādi, "the roots beginning with juhoti". Its first root is hu "to offer". In these roots ślu deletes the vikaraṇa śap, and before that ślu the root is doubled. So hu gives juhoti "he offers". After this chapter you will be able to derive the class's key forms. They include juhoti, juhvati and juhavāṃcakāra; bibhitaḥ or bibhītaḥ; and piparti, pipūrtaḥ and papratuḥ. They also include bibharti and mimīte; the imperatives jahāhi, jahihi, jahīhi; dattaḥ, dehi, dhattaḥ and dhatse. Last come nenekti, nenijāni, iyarti, and jajātaḥ beside jajñati. As in chapters 44 and 45, Bhaṭṭoji walks the list of roots in order, from 1083 hu to 1106 gā. He teaches each rule at the root where it first bites, so the rules come from five different books of the Aṣṭādhyāyī. The chapter has 16 sūtras, SK 2489 to SK 2504.

Before you start

From chapter 1 you know a dhātu (a root) and the Dhātupāṭha, the list of roots. You know an it, a marker sound that is not kept, and upadeśa, a form as first taught. You know anuvṛtti (carrying a word forward) and adhikāra (a heading).
From chapter 1 you know guṇa (a, e, o), vṛddhi (ā, ai, au), hrasva (short) and laghu (a light vowel). You know that the rules of the Tripādī, the last three quarters of the Aṣṭādhyāyī, are asiddha, "treated as not having taken effect" (SK 12).
From chapter 3 you know SK 70: a vowel that replaces ṛ is followed by r (raparatva). You know SK 49, sthānivadbhāva: a substitute is treated like what it replaced. You also know SK 50, acaḥ parasmin pūrvavidhau: a substitute for a vowel is treated like that vowel for a rule about what precedes it.
From chapter 5 you know SK 123: n becomes anusvāra before a jhal.
From chapter 8 you know SK 260: luk, ślu and lup are names for the deletion of an affix.
From chapter 11 you know SK 326: in a one-vowel, jhaṣ-final portion of a root, a baś becomes bhaṣ before s or dhv. You know SK 354, hali ca: lengthening before a consonant. You know SK 378, coḥ kuḥ: a palatal becomes a velar before a jhal. You know SK 426: the two copies of a doubled root together are named abhyasta.
From chapter 44 you know the lakāras, the eighteen endings, parasmaipada and ātmanepada, and the three persons, prathama, madhyama and uttama. You know sārvadhātuka and ārdhadhātuka affixes, and the vikaraṇa śap.
From chapter 44 you know doubling (SK 2175, 2177) and the abhyāsa, the first copy. You know its usual operations:
only the first consonant stays (SK 2179);
the vowel is shortened (SK 2180);
a velar or h becomes a palatal (SK 2245);
an aspirate loses its aspiration (SK 2182).
From chapter 44 you know that a sārvadhātuka affix not marked with p counts as ṅit (SK 2234), and that a kit or ṅit affix blocks guṇa. You know āṭ, the pit augment of the loṭ uttama (SK 2204), and jus for jhi after an abhyasta (SK 2226). You know the ām-liṭ with a helping verb (SK 2306), iyaṅ (SK 2290), and the ghu roots dā and dhā (SK 2373).
From chapter 44 you also know these rules, cited here by number:
SK 2363: loss of the penultimate of jan, khan and others before a kṅit;
SK 2383: guṇa of ṝ-final roots in liṭ;
SK 2387: v for the u of hu before a vowel;
SK 2390: i for a root's final ṝ.
From chapter 45, the class of ad, you know the rules for the doubled roots of that class. SK 2479 gives at for the jh of an ending after an abhyasta. SK 2481 gives guṇa before jus. SK 2482 gives i for the ā of daridrā. SK 2483 deletes the ā of an abhyasta before a vowel-initial kṅit. SK 2471 gives dehi. SK 2425 turns hi into dhi.

A. hu: ślu, doubling, ām with śluvadbhāva, guṇa by paratva — running example: `hu + śap + ti → hu + ti → hu-hu-ti → ju-ho-ti → juhoti`; `juhu + jhi → juhu + ati → juhvati`

Three rules make the class what it is. SK 2489 deletes śap after these roots by ślu. SK 2490 doubles the root before that ślu. SK 2491 lets four roots take ām in liṭ, and doubles them there too. The running example is hu "to offer". hu + śap + ti loses śap, is doubled, and becomes juhoti "he offers". Before the vowel of ati, the u becomes v: juhvati "they offer".

SK 2489 · जुहोत्यादिभ्यः श्लुः · juhotyādibhyaḥ śluḥ (A 2.4.75)

जुहोत्यादिभ्यः श्लुः

juhotyādibhyaḥ śluḥ · A 2.4.75

{$ {!1083 हु!} दानादनयोः $}। आदाने चेत्येके । प्रीणनेऽपीति भाष्यम् । दानं चेह प्रक्षेपः । स च वैधे आधारे हविषश्चेति स्वभावाल्लभ्यते । इतश्चत्वारः परस्मैपदिनः ॥ शपः श्लुः स्यात् ॥

Bhaṭṭoji first gives the root entry: 1083 hu, "in giving and in eating". Some add "taking". The Bhāṣya, Patañjali's commentary, adds "pleasing". From here on, four roots are parasmaipadin: they take the parasmaipada endings. They are hu, ñibhī, hrī and pṝ.

Rule. After hu and the roots that follow it, ślu replaces śap. ślu is a name for deletion (SK 260). So śap is simply removed. The word śapaḥ "of śap" is carried forward from SK 2423.

Example.

hu + śap + ti (laṭ, prathama singular)

→ ślu for śap (this rule) → hu + ti.

The next rule doubles the root.

Applies when / blocked by. "Giving" here means casting (prakṣepa). Bhaṭṭoji adds two limits. The casting is into a receptacle enjoined by rule, such as the sacred fire. And what is cast is an oblation (haviṣ). These limits come from established usage (svabhāva), not from the rule. One of Bhaṭṭoji's commentators gives an illustration. A stone thrown into the fire by mistake or in anger is a casting. It is not an offering (homa).

Watch out. hu takes only the parasmaipada endings. One of Bhaṭṭoji's commentators therefore calls the ātmanepada form hoṣye incorrect. But juhvānāḥ, made with the participle affix cānaś, is correct.

SK 2490 · श्लौ · ślau (A 6.1.10)

श्लौ

ślau · A 6.1.10

धातोर्द्वे स्तः । जुहोति । जुहुतः । अदभ्यस्तात् SK 2479 इत्यत् । हुश्नुवोः - SK 2387 इति यण् । जुह्वति ॥

Rule. Before ślu, the root is doubled. "Two" (dve) is carried forward from SK 2175. "Of the root" (dhātoḥ) is carried forward from SK 2177. So doubling, which chapter 44 taught for liṭ, also happens wherever ślu has replaced śap.

Example.

hu + ti → doubled (this rule) → hu-hu-ti

→ the first copy is the abhyāsa; its usual operations make it ju → ju-hu-ti

→ guṇa of the second u before the pit ending tip → juhoti "he offers".

juhu + tas → tas is not pit, so it counts as ṅit, and there is no guṇa → juhutaḥ.

juhu + jhi → the stem is abhyasta, so at replaces the jh (SK 2479) → juhu + ati

→ v for u by SK 2387 → juhvati "they offer".

Applies when / blocked by. Before ati, the general rule would give uvaṅ (uv, SK 271). One of Bhaṭṭoji's commentators calls the v of SK 2387 an exception (apavāda) to uvaṅ. So the form is juhvati, not juhuvati.

SK 2491 · भीह्रीभृहुवां श्लुवच्च · bhīhrībhṛhuvāṃ śluvac ca (A 3.1.39)

भीह्रीभृहुवां श्लुवच्च

bhīhrībhṛhuvāṃ śluvacca · A 3.1.39

एभ्योलिट्याम्वा स्यादामि श्लाविव कार्यं च । जुहवांचकार । जुहाव । होता । होष्यति । जुहोतु । जुहुतात् । हेर्धिः । जुहुधि । आटि परत्वाद्गुणः । जुहवानि । परत्वात् जुसि च SK 2481 इति गुणः । अजुहवुः । जुहुयात् । हूयात् । अहौषीत् ।{$ {!1084 ञिभी!} भये$} । बिभेति ॥

Rule. After bhī, hrī, bhṛ and hu, the affix ām is optionally added before liṭ. When ām is added, the root is treated as if ślu followed. This is śluvadbhāva, "treatment as before ślu". So the root is doubled before ām, just as before ślu.

The words ām and liṭi are carried forward from SK 2306. In the sūtra the four roots stand in the sixth case. One of Bhaṭṭoji's commentators says it has the sense of the fifth case, "after".

Example.

hu + liṭ → ām (this rule) → hu + ām

→ treated as before ślu: doubled → juhu + ām

→ guṇa of the u, then av before the vowel → juhavām

→ the helping verb kṛ in liṭ follows → juhavāṃcakāra "he offered".

Without ām, ordinary liṭ: juhāva.

More forms of hu:

luṭ hotā, lṛṭ hoṣyati; loṭ juhotu or juhutāt.
juhu + hi → hi becomes dhi (SK 2425) → juhudhi "offer!".
vidhiliṅ juhuyāt; āśīrliṅ hūyāt; luṅ ahauṣīt, with vṛddhi before sic.

Applies when / blocked by. Two forms show a conflict of rules. Bhaṭṭoji settles both by paratva: of two rules that both apply, the one later in the Aṣṭādhyāyī wins.

loṭ uttama singular: juhu + āni. The augment āṭ is pit (SK 2204), so the ending is not ṅit. guṇa is therefore due. But the v of SK 2387 is also due before the vowel. guṇa is the later rule and wins → juhavāni.
laṅ prathama plural: a + juhu + jhi → jus for jhi, the stem being abhyasta (SK 2226) → ajuhu + us. Again SK 2387 is due. But guṇa before jus (SK 2481) is the later rule and wins → ajuhavuḥ.

Watch out. Write juhavāni and ajuhavuḥ, not juhvāni and ajuhvuḥ. The v of SK 2387 loses to guṇa in both.

Bhaṭṭoji then gives the next root: 1084 ñibhī "in fear". The ñi is an it. Its present is bibheti "he fears", with ślu and doubling as for hu.

B. bhī and hrī: optional i before a hal-initial kṅit — running example: `bi-bhī + tas → bibhitaḥ / bibhītaḥ`

One rule belongs to bhī. Before a consonant-initial ending that is kit or ṅit, the root's ī may become short i. The running example is the present dual. bi-bhī + tas gives either bibhitaḥ or bibhītaḥ "they two fear". The root hrī "to be ashamed" follows, and the group ends with the root entry for pṝ.

SK 2492 · भियोऽन्यतरस्याम् · bhiyo 'nyatarasyām (A 6.4.115)

भियोऽन्यतरस्याम्

bhiyo'nyatarasyām · A 6.4.115

इकारः स्याद्धलादौ क्ङिति सार्वधातुके । बिभितः । बिभीतः । बिभ्यति । बिभयांचकार । बिभाय । भेता ।{$ {!1085 ह्री!} लज्जायाम्$} । जिह्रेति । जिह्रीतः । जिह्रियति । जिह्रयांचकार । जिह्राय ।{$ {!1086 पॄ!} पालनपूरणयोः$} ॥

Rule. For bhī, i optionally replaces its ī. This happens before a sārvadhātuka affix that begins with a consonant (hal) and is kit or ṅit (kṅit).

Four words are carried forward (anuvṛtti):

it "the vowel i", from SK 2482;
hali "before a consonant", from SK 2497, taught later in this chapter;
kṅiti "before a kit or ṅit", from SK 2363;
sārvadhātuke "before a sārvadhātuka", from SK 2467.

Example.

bhī + tas (laṭ, prathama dual) → ślu and doubling (SK 2489, 2490) → bhī-bhī-tas

→ abhyāsa operations → bi-bhī-tas

→ tas begins with t, a consonant, and is ṅit → i optionally (this rule)

→ bibhitaḥ or bibhītaḥ "they two fear".

Applies when / blocked by. One of Bhaṭṭoji's commentators shows each condition with a form where it fails:

not sārvadhātuka: bebhīyate, bibhiye;
not consonant-initial: bibhī + ati → at for jh (SK 2479) → y for ī → bibhyati "they fear";
not kṅit: bibheti, since tip is pit and takes guṇa.

Some hold that "consonant-initial kṅit" adds little here; scholars discuss this.

More forms of bhī: liṭ bibhayāṃcakāra, with ām and doubling as before ślu (SK 2491), or bibhāya; luṭ bhetā.

Watch out. The rule names bhī alone. The next root, 1085 hrī "in being ashamed", has no such option. Bhaṭṭoji gives:

laṭ jihreti, jihrītaḥ, jihriyati. In jihriyati, at comes (SK 2479), then iyaṅ for the ī.
liṭ jihrayāṃcakāra (by SK 2491) or jihrāya.

Bhaṭṭoji then gives the root entry 1086 pṝ "in protecting and in filling". Its vowel is long ṝ.

C. pṝ (and ṛ): i in the abhyāsa, u for ṝ after a labial, the short vowel in liṭ — running example: `pṝ-pṝ-ti → pir-pṝ-ti → pi-pṝ-ti → pi-par-ti → piparti`; `pi-pṝ-tas → pi-pur-tas → pipūrtaḥ`; `pa-pṝ-atus → papratuḥ / paparatuḥ`

The root pṝ "to protect, to fill" needs three rules. SK 2493 gives its abhyāsa the vowel i: piparti "he fills". SK 2494 turns its ṝ into u after the labial p, where guṇa does not come: pipūrtaḥ "they two fill". SK 2495 lets the ṝ become short before a kit ending of liṭ: papratuḥ beside paparatuḥ "they two filled".

SK 2493 · अर्तिपिपर्त्योश्च · artipipartyoś ca (A 7.4.77)

अर्तिपिपर्त्योश्च

artipipartyośca · A 7.4.77

अभ्यासस्य इकारोऽन्तादेशः स्यात् श्लौ ॥

Rule. For ṛ "to go" and pṝ "to fill", i replaces the final of the abhyāsa. This happens before ślu. The sūtra names the two roots by their present forms, arti and piparti.

Three words are carried forward, all from rules taught later in the Kaumudī:

abhyāsasya "of the abhyāsa", from SK 2620;
it "the vowel i", from SK 2496;
ślau "before ślu", from SK 2502.

Example.

pṝ + ti → ślu and doubling (SK 2489, 2490) → pṝ-pṝ-ti

→ i for the vowel of the abhyāsa, followed by r (SK 70) → pir-pṝ-ti

→ only the first consonant of the abhyāsa stays (SK 2179) → pi-pṝ-ti

→ guṇa of the second part before tip, followed by r → pi-par-ti → piparti "he fills, he protects".

Applies when / blocked by. The i needs ślu. So liṭ, which has no ślu, keeps the ordinary abhyāsa: papāra. The form of the other root, iyarti "he goes", is given under SK 2503.

SK 2494 · उदोष्ठ्यपूर्वस्य · ud oṣṭhyapūrvasya (A 7.1.102)

उदोष्ठ्यपूर्वस्य

udoṣṭhyapūrvasya · A 7.1.102

अङ्गावयवौष्ठ्यपूर्वो य ॠत्तदन्तस्याङ्गस्य उत्स्यात् । गुणवृद्धी परत्वादिमं बाधेते । पिपर्ति । उत्वम् । रपरत्वम् । हलि च SK 354 इति दीर्घः । पिपूर्तः । पिपुरति । पपार । किति लिटि ॠच्छात्यॄताम् SK 2383ल इति गुणे प्राप्ते ॥

Rule. u replaces the ṝ of an aṅga ending in ṝ, when a labial (oṣṭhya, a sound made with the lips, such as p) stands before the ṝ. The labial must be part of the aṅga itself. By SK 70 the u is followed by r. An aṅga is the stem to which an affix is added; here pipṝ before tas.

The word ṝtaḥ "of an ṝ" is carried forward from SK 2390, which gives i for such an ṝ. This rule gives u instead after a labial. The heading aṅgasya "of the aṅga" is read twice here. Once it says the labial belongs to the aṅga. Once it says the aṅga ends in ṝ.

Example.

pi-pṝ + tas (laṭ, prathama dual) → tas is ṅit, so there is no guṇa

→ u for ṝ, followed by r (this rule) → pi-pur-tas

→ the u is lengthened before a consonant by hali ca (SK 354) → pipūrtaḥ "they two fill".

pi-pṝ + ati → at for jh; the ending is ṅit, so no guṇa → u, with r → pipurati "they fill".

Applies when / blocked by. guṇa and vṛddhi are later rules (para), so they set this rule aside when they apply:

pi-pṝ + ti → guṇa before the pit tip → piparti, not pipūrti;
liṭ pa-pṝ + a (ṇal) → vṛddhi → papāra "he filled".

Why must the labial be part of the aṅga? Bhaṭṭoji's commentators give samīrṇaḥ, from ṝ "to go" with the prefix sam and the participle affix kta. The m is a labial, but it belongs to the prefix, not to the aṅga. So u does not come. The root takes i by SK 2390 instead. Otherwise the form would be samūrṇa.

Watch out. In liṭ before a kit ending, SK 2383 makes guṇa due for an ṝ-final root. The next rule is stated for that case.

SK 2495 · शॄदॄप्रां ह्रस्वो वा · śṝdṝprāṃ hrasvo vā (A 7.4.12)

शॄदॄप्रां ह्रस्वो वा

śṝdṝprāṃ hrasvo vā · A 7.4.12

एषां किति लिटि ह्रस्वो वा स्यात् । पक्षे गुणः । पप्रतुः । पप्रुः । पपरतुः । पपरुः । परिता । परीता । अपिपः । अपिपूर्ताम् । अपिपरूः । पिपूर्यात् । पूर्यात् । अपारीत् । अपारिष्टाम् । ह्रस्वान्तोऽयमिति केचित् । पिपर्ति । पिपृतः । पिप्रति । पिपृयात् । आशिषि प्रियात् । अपार्षीत् । पाणिनीयमते तु तं रोदसी पिपृतमित्यादौ छान्दसत्त्वं शरणम् ।{$ {!1087 डुभृञ्!} धारणपोषणयोः$} ॥

Rule. For śṝ "to injure", dṝ "to tear" and pṝ, a short vowel (hrasva) optionally replaces the ṝ. This happens before a kit ending of liṭ. In the other option, guṇa comes by SK 2383. śṝ and dṝ are roots of the kryādi class.

The word liṭi "before liṭ" is carried forward from SK 2388. The word kiti "before a kit" is not found in any rule before this one. One of Bhaṭṭoji's commentators explains why it is supplied. This rule is an exception to the guṇa of SK 2383, and that guṇa is for the kit liṭ.

Example.

pa-pṝ + atus (liṭ, prathama dual; atus is kit)

→ short ṛ for ṝ (this rule) → pa-pṛ + atus → r for ṛ before the vowel (yaṇ) → papratuḥ.

Or guṇa (SK 2383) → pa-par + atus → paparatuḥ.

Likewise the plural: papruḥ or paparuḥ.

Applies when / blocked by. Why not simply make the guṇa optional? Without guṇa, yaṇ would have to give papratuḥ. But another rule would come first: i by SK 2390, or u by SK 2494. These operate on the aṅga, and an operation on the aṅga is stronger than one on sounds, like yaṇ. So Bhaṭṭoji's commentators say the short vowel must be taught directly.

More forms of pṝ:

luṭ paritā or parītā, the iṭ optionally long (SK 2391);
laṅ apipaḥ, apipūrtām, apiparuḥ (jus, with guṇa by SK 2481);
vidhiliṅ pipūryāt; āśīrliṅ pūryāt; luṅ apārīt, apāriṣṭām.

Watch out. Some teachers read the root with short ṛ, as pṛ. On that view: piparti, pipṛtaḥ, piprati; vidhiliṅ pipṛyāt; āśīrliṅ priyāt; luṅ apārṣīt, the root being aniṭ (without iṭ). Bhaṭṭoji keeps Pāṇini's reading with long ṝ. On that reading the Vedic phrase taṃ rodasī pipṛtam, with short ṛ, would need u. So the only way to account for it is that it is Vedic (chāndasa).

Bhaṭṭoji then gives the root entry 1087 ḍubhṛñ "in holding and in nourishing". Its ñ makes it ubhayapadin: it takes both sets of endings. One of Bhaṭṭoji's commentators says the marker ḍu gives the noun bhṛtrimam.

D. bhṛñ, māṅ, ohāṅ: i in the abhyāsa; ī for the ā of an abhyasta — running example: `bhṛ-bhṛ-ti → bi-bhṛ-ti → bibharti`; `mā-mā-te → mi-mā-te → mimīte`

Three roots take i in the abhyāsa before ślu: bhṛñ "to bear", māṅ "to measure" and ohāṅ "to go" (SK 2496). So bhṛ gives bibharti "he bears". Then SK 2497 turns the ā of a doubled ā-final root into ī before a consonant-initial kṅit ending. So mā gives mimīte "he measures".

SK 2496 · भृञामित् · bhṛñām it (A 7.4.76)

भृञामित्

bhṛñāmit · A 7.4.76

भृञ् माङ् ओहाङ् एषां त्रयाणामभ्यासस्य इत्स्यात् श्लौ । बिभर्ति । बिभृतः । बिभ्रति । बिभृध्वे । श्लुवद्भावाद् द्वित्वेत्वे । बिभरामास बभार । बभर्थ । बभृव । बिभृहि । बिभराणि । अबिभः । अबिभृताम् । अबिभरुः । बिभृयात् । भ्रियात् । भृषीष्ट । अभार्षीत् । अभृत ।{$ {!1088 माङ्!} माने शब्दे च$} ॥

Rule. i replaces the vowel of the abhyāsa of three roots: bhṛñ, māṅ and ohāṅ. This happens before ślu.

The sūtra names only bhṛñ, but in the plural. One of Bhaṭṭoji's commentators reads the plural as "bhṛñ and the roots after it". Three words are carried forward:

abhyāsasya "of the abhyāsa", from SK 2620;
trayāṇām "of the three" and ślau "before ślu", from SK 2502.

Some obtain "three" in another way; scholars discuss this.

Example.

bhṛ + ti → ślu and doubling → bhṛ-bhṛ-ti

→ i for the vowel of the abhyāsa (this rule), with the usual abhyāsa operations → bi-bhṛ-ti

→ guṇa before the pit tip → bibharti "he bears".

bibhṛ + tas → ṅit, no guṇa → bibhṛtaḥ.

bibhṛ + jhi → at for the jh of jhi, the stem being abhyasta → r for ṛ before the vowel → bibhrati "they bear".

Ātmanepada, madhyama plural: bibhṛdhve.

Applies when / blocked by. Why "before ślu"? In ordinary liṭ there is no ślu, so there is no i: babhāra "he bore".

But bhṛ is one of the four roots of SK 2491. With ām, it is treated as before ślu. So it is doubled and takes i: bibharāmāsa.

Why "three"? The fourth root, ohāk, has no i: jahāti.

More forms of bhṛ:

liṭ babhartha, babhṛva, both without iṭ;
loṭ bibhṛhi: hi is not pit, so it is ṅit, and there is no guṇa;
loṭ bibharāṇi: āṭ is pit, so guṇa comes;
laṅ abibhaḥ, abibhṛtām, abibharuḥ (jus, with guṇa by SK 2481);
vidhiliṅ bibhṛyāt;
āśīrliṅ bhriyāt, with riṅ (SK 2367); ātmanepada bhṛṣīṣṭa, kit by SK 2368, so no guṇa;
luṅ abhārṣīt; ātmanepada abhṛta, with sic lost by SK 2369.

Bhaṭṭoji then gives the root entry 1088 māṅ "in measuring, and in sound". Its ṅ makes it ātmanepadin: it takes only the ātmanepada endings. By this rule its stem is mi-mā.

SK 2497 · ई हल्यघोः · ī haly aghoḥ (A 6.4.113)

ई हल्यघोः

ī halyaghoḥ · A 6.4.113

श्नाभ्यस्तयोरातईत्स्यात्सार्वधातुके क्ङिति हलि न तु घुसंज्ञकस्य । मिमीते । श्नाभ्यस्तयोः - SK 2483 इत्यालोपः । मिमाते । मिमते । प्रण्यमास्त ।{$ {!1089 ओहाङ्!} गतौ $}। जिहीते । जिहाते । जिहते । जहे । हाता । हास्यते ।{$ {!1090 ओहाक्!} त्यागे$} । परस्मैपदी । जहाति ॥

Rule. ī replaces the ā of śnā and of an abhyasta stem. This happens before a sārvadhātuka affix that begins with a consonant and is kṅit. It does not happen in a ghu root, that is, dā or dhā (SK 2373). śnā is the vikaraṇa seen in lunāti "he cuts".

Three items are carried forward:

śnābhyastayor ātaḥ "of the ā of śnā and of an abhyasta", from SK 2483;
kṅiti, from SK 2363;
sārvadhātuke, from SK 2467.

Example.

mā + te (laṭ, prathama singular) → ślu, doubling, and i in the abhyāsa (SK 2496) → mi-mā-te

→ te begins with t and is ṅit → ī for ā (this rule) → mimīte "he measures".

Before a vowel-initial ending the ā is lost instead (SK 2483):

mi-mā + āte → mimāte "they two measure".

mi-mā + jha → at for jh → mi-mā + ate → mimate "they measure".

Applies when / blocked by. One of Bhaṭṭoji's commentators shows the conditions:

with śnā: lunīte;
not kṅit: lunāti, since tip is pit;
a ghu root: dhattaḥ, dattaḥ, with no ī.

The same commentator adds a point about SK 2462 (ghumāsthā…), a rule that also names mā. It does not act here, since it is limited to ārdhadhātuka affixes.

Bhaṭṭoji also gives luṅ praṇyamāsta, with the prefixes pra and ni.

Then come two roots spelled hā:

1089 ohāṅ "in going". Its ṅ makes it ātmanepadin. It takes i in the abhyāsa by SK 2496. The h of the abhyāsa becomes a palatal (cutva), as in chapter 44. Its forms are like those of māṅ: laṭ jihīte, jihāte, jihate; liṭ jahe; luṭ hātā; lṛṭ hāsyate.
1090 ohāk "in abandoning". Its o and k are it, and it is parasmaipadin. It is not one of the three of SK 2496, so its abhyāsa keeps a: jahāti "he abandons".

Watch out. ohāk is named in the following rules by its present form, jahāti.

E. ohāk (jahāti): i or ī, ā/i/ī before hi, loss before y; then dā and dhā — running example: `ja-hā + tas → jahitaḥ / jahītaḥ`; `ja-hā + hi → jahāhi / jahihi / jahīhi`; `ja-hā + yāt → jahyāt`

The root ohāk "to abandon" (jahāti) has its own three rules. Before a consonant-initial kṅit ending its ā becomes i or ī (SK 2498). Before hi it may also stay ā (SK 2499). Before a y-initial ending it is lost (SK 2500). The group then gives the ghu roots dā "to give" and dhā "to hold", where the ā is lost and does not become ī: dattaḥ, dehi.

SK 2498 · जहातेश्च · jahāteś ca (A 6.4.116)

जहातेश्च

jahāteśca · A 6.4.116

इत्स्याद्वा हलादौ क्ङिति सार्वधातुके । पक्षे ईत्वम् । जहितः । जहीतः । जहति । जहौ ॥

Rule. For jahāti too, i optionally replaces the ā before a consonant-initial kṅit sārvadhātuka. In the other option, ī comes by SK 2497.

Four words are carried forward:

it "the vowel i", from SK 2482;
anyatarasyām "optionally", from SK 2492;
kṅiti, from SK 2363;
hali, from SK 2497.

Example.

hā + tas → ślu, doubling, abhyāsa operations → ja-hā + tas

→ i (this rule) → jahitaḥ, or ī (SK 2497) → jahītaḥ "they two abandon".

ja-hā + ati → a vowel-initial ending, so the ā is lost (SK 2483) → jahati "they abandon".

liṭ jahau.

Applies when / blocked by. One of Bhaṭṭoji's commentators shows two conditions:

not kṅit: jahāti, since tip is pit;
not sārvadhātuka: hīyate.

SK 2499 · आ च हौ · ā ca hau (A 6.4.117)

आ च हौ

ā ca hau · A 6.4.117

जहातैर्हौ परे आ स्यात् चादिदीतौ । जहाहि । जहिहि । जहीहि । अजहात् । अजहुः ॥

Rule. Before hi, ā replaces the ā of jahāti. By ca "and", i and ī come as well. So there are three options.

jahāteḥ runs on from SK 2498. One of Bhaṭṭoji's commentators says that ca draws in (anukarṣa) the i of SK 2482 and the ī of SK 2497.

Example.

ja-hā + hi (loṭ, madhyama singular)

→ ā (this rule) → jahāhi

→ or i → jahihi

→ or ī → jahīhi "abandon!".

laṅ ajahāt, ajahuḥ.

Applies when / blocked by. hi begins with a consonant and is ṅit. So SK 2497 and SK 2498 alone would give only ī or i. This rule keeps the long ā as a third option.

Watch out. All three are correct. One of Bhaṭṭoji's commentators says that is why the poet Bhaṭṭi uses all three in one line: jahihi jahīhi jahāhi rāmabhāryām.

SK 2500 · लोपो यि · lopo yi (A 6.4.118)

लोपो यि

lopo yi · A 6.4.118

जहातेरालोपः स्याद्यादौ सार्वधातुके । जह्यात् । एर्लिङि SK 2374 । हेयात् । अहासीत् ।{$ {!1091 डुदाञ्!} दाने$} । प्रणिददाति । दत्तः । ददति । दत्ते । ददौ । घ्वसोः -SK 2471 इत्येत्वाभ्यासलोपौ । देहि । अददात् । अदत्ताम् । अददुः । दद्यात् । देयात् । अदात् । अदाताम् । अदुः । अदित ।{$ {!1092 डुधाञ्!} धारणपोषणयोः $}। दानेऽप्येके । प्रणिदधाति ॥

Rule. The ā of jahāti is deleted (lopa) before a sārvadhātuka affix beginning with y. This rule is an exception to SK 2498.

Three words are carried forward:

jahāteḥ "of jahāti", from SK 2498;
ātaḥ "of the ā", from SK 2483;
sārvadhātuke, from SK 2467.

Example.

ja-hā + yāt (vidhiliṅ, prathama singular) → yāt begins with y → the ā is lost (this rule) → jahyāt "he should abandon".

Applies when / blocked by. The affix must be sārvadhātuka. The āśīrliṅ is not, so the rule does not act. There e comes by SK 2374: heyāt "may he abandon". luṅ ahāsīt.

Bhaṭṭoji then gives the root entry 1091 ḍudāñ "in giving". Its forms show the ghu exception of SK 2497:

praṇidadāti, with ṇ by SK 2285;
da-dā + tas → dā is ghu, so there is no ī → the ā is lost by SK 2483 → dattaḥ "they two give";
dadati; ātmanepada datte; liṭ dadau;
da-dā + hi → e for the ā, and loss of the abhyāsa (SK 2471) → dehi "give!";
laṅ adadāt, adattām, adaduḥ; vidhiliṅ dadyāt; āśīrliṅ deyāt;
luṅ adāt, adātām, aduḥ, with luk of sic (SK 2223);
luṅ ātmanepada adita, with i by SK 2389 and sic lost by SK 2369.

Next comes 1092 ḍudhāñ "in holding and in nourishing". Some add "giving". praṇidadhāti. Before tas, the stem is doubled and its ā is lost, as for dā: da-dhā + tas → dadh + tas. The next rule continues this form.

F. dhā: dh (a bhaṣ) for the d of dadh — running example: `da-dhā + tas → dadh + tas → dhadh + tas → dhattaḥ`

After doubling and the loss of its ā, dhā "to hold" has the stem dadh. SK 2501 makes the first d aspirate before t, th, s and dhv. So dadh + tas becomes dhattaḥ "they two hold". The running example follows that d to dh, while the root's dh becomes t.

SK 2501 · दधस्तथोश्च · dadhas tathoś ca (A 8.2.38)

दधस्तथोश्च

dadhastathośca · A 8.2.38

द्विरुक्तस्य झषन्तस्य धाञो बशो भष् स्यात्तथयोः स्ध्वोश्च परतः । वचनसामर्थ्यादालोपो न स्थानिवदिति वामनमाधवौ । वस्तुतस्तु पूर्वत्रासिद्धीये न स्थानिवत् । धत्तः । दधति । धत्थः । दध्वः । धत्ते । धत्से । धद्ध्वे । धेहि । अधित ॥ अथ त्रयः स्वरितेतः ॥{$ {!1093 णिजिर्!} शौचपोषणयोः$} ॥

Rule. Take dadh, the root dhā once doubled, ending in a jhaṣ (a voiced aspirate such as dh). Its baś (a voiced unaspirated stop, here the d of the abhyāsa) is replaced by a bhaṣ (the matching voiced aspirate, dh). This happens before t or th. By ca, it also happens before s and dhv.

The words jhaṣantasya, baśaḥ and bhaṣ are carried forward from SK 326. So is sdhvoḥ "before s and dhv", through the ca. One of Bhaṭṭoji's commentators notes that dadha in the sūtra is the doubled root. It is not the separate root dadha "to hold".

Example.

dhā + tas → ślu, doubling, abhyāsa operations → da-dhā + tas

→ ā lost (SK 2483), dhā being ghu (SK 2497 does not act) → dadh + tas

→ dh for the d of the abhyāsa, t following (this rule) → dhadh + tas

→ the root's dh becomes t by cartva → dhattaḥ "they two hold".

Likewise dhatthaḥ before th.

Ātmanepada: dhatte; dhatse before s; dhaddhve before dhv.

Applies when / blocked by. No t, th, s or dhv follows in two forms, so there is no aspirate:

dadhati "they hold", with at and loss of the ā;
dadhvaḥ "we two hold".

Why "ending in a jhaṣ"? In dadhāti the ā stays, so the stem does not end in a jhaṣ, and no aspirate comes.

Other forms: dhehi "hold!", with e and loss of the abhyāsa (SK 2471); luṅ ātmanepada adhita.

Watch out. There is a puzzle here. The loss of ā (SK 2483) is a substitute. By acaḥ parasmin pūrvavidhau (SK 50), a substitute for a vowel is treated like that vowel for a rule about what precedes it. So the loss of ā might count as the ā. Then dadh would not end in a jhaṣ, and no t would follow it. The rule would never apply.

Vāmana and Mādhava answer by vacanasāmarthya, "the force of the statement". The rule must apply somewhere, so the loss is not treated as the ā.
Bhaṭṭoji's own answer is simpler. This rule is in the Tripādī (A 8.2), the asiddha zone of SK 12. For a rule of that zone, a substitute gets no sthānivadbhāva.

One of Bhaṭṭoji's commentators calls the first route needless trouble once the second is accepted.

Bhaṭṭoji then announces three roots whose it-vowel is svarita (svaritet). svarita is one of the three accents of a vowel. Such roots take both sets of endings. The first is 1093 ṇijir "in cleansing and in nourishing".

G. ṇij, vij, viṣ: guṇa of the abhyāsa, no light-penultimate guṇa; the Vedic run ghṛ … jana — running example: `nij-nij-ti → ni-nij-ti → ni-nej-ti → ne-nej-ti → nenekti`; `nenij + āni → nenijāni` (not *nenejāni)

Three roots, ṇij "to cleanse", vij "to separate" and viṣ "to pervade", take guṇa in the abhyāsa before ślu (SK 2502): nenekti "he cleanses". SK 2503 then forbids guṇa of the root's light penultimate vowel before a vowel-initial pit ending: nenijāni "let me cleanse". The group ends with a run of Vedic roots, from ghṛ to jana, where Bhaṭṭoji also gives iyarti "he goes".

SK 2502 · निजां त्रयाणां गुणः श्लौ · nijāṃ trayāṇāṃ guṇaḥ ślau (A 7.4.75)

निजां त्रयाणां गुणः श्लौ

nijāṃ trayāṇāṃ guṇaḥ ślau · A 7.4.75

णिज्विज्विषामभ्यासस्य गुणः स्यात श्लौ । नेनेक्ति । नेनिक्तः । नेनिजति । निनेज । नेक्ता । नेक्ष्यति । नेनेक्तु । नेनिग्ध ॥

Rule. guṇa replaces the vowel of the abhyāsa of ṇij, vij and viṣ. This happens before ślu. The plural nijām means "the three beginning with nij". abhyāsasya is carried forward from SK 2620.

Example.

nij + ti → ślu and doubling → nij-nij-ti

→ only the first consonant of the abhyāsa stays (SK 2179) → ni-nij-ti

→ guṇa of the second part before the pit tip → ni-nej-ti

→ guṇa of the abhyāsa (this rule) → ne-nej-ti

→ g for j (SK 378) → ne-neg-ti

→ k for g before t (cartva) → nenekti "he cleanses".

neniktaḥ "they two cleanse"; nenijati "they cleanse", with at.

Applies when / blocked by. Why "before ślu"? liṭ has no ślu, so the abhyāsa takes no guṇa: nineja "he cleansed".

More forms: luṭ nektā; lṛṭ nekṣyati; loṭ nenektu. In loṭ madhyama singular, hi is not pit, so it is ṅit. There is no guṇa of the root, and hi becomes dhi. The abhyāsa still keeps its e. The j becomes g by SK 378, as in nenekti: nenigdhi "cleanse!".

SK 2503 · नाभ्यस्तस्याचि पिति सार्वधातुके · nābhyastasyāci piti sārvadhātuke (A 7.3.87)

नाभ्यस्तस्याचि पिति सार्वधातुके

nābhyastasyāci piti sārvadhātuke · A 7.3.87

लघूपधगुणो न स्यात । नेनिजानि । अनेनेक् । अनेनिक्ताम् । अनेनिजुः । नेनिज्यात् । निज्यात् । अनिजत् । अनैक्षीत् । अनिक्त ।{$ {!1094 विजिर्!} पृथग्भावे$} । वेवेक्ति । वेविक्ते । विवेजिथ । अत्र विज इट् SK 2536 इति ङित्वं न । ओविजी इत्यस्यै तत्र ग्रहणात् । णिजिविजी रुधादावपि ।{$ {!1095 विष्लृ!} व्याप्तौ$} । वेवेष्टि । वेविष्टे । लृदित्त्वादङ् । अविषत् । तङि क्सः । अजादौ क्सस्याचि SK 2337 इति अल्लोपः । अविक्षत । अविक्षाताम् । अविक्षन्त ॥ अथ आगणान्ताः परस्मैपदिनश्छान्दसाश्च ॥{$ {!1096 घृ!} क्षरणदीप्त्योः$} । जिघर्म्यग्रिं हविषा घृतेन । भृञामित् SK 2496, बहुलं छन्दसि SK 3598 इति इत्वम् ।{$ {!1097 हृ!} प्रसह्यकरणे$} । अयं स्रुवो अभिजिहर्ति होमान् ।{$ {!1098 ऋ!} {!1099 सृ!} गतौ$} । बहुलं छन्दसि SK 3598 इत्येव सिद्धे अर्तिपिपर्त्योश्च SK 2493 इतीत्वविधानादयं भाषायामपि । अभ्यासस्यासवर्णे SK 2290 इतीयङ् । इयर्ति । इयृतः । इय्रति । आर । आरतुः । इडत्त्यर्ति - SK 2384 इति नित्यमिट् । आरिथ । अर्ता । अरिष्यति । इयराणि । ऐयः । ऐयृताम् । ऐयरुः । इयृयात् । अर्यात् । आरत् । ससर्ति ।{$ {!1100 भस!} भर्त्सनदीप्त्योः$} । बभस्ति । घसिभसोर्हलिच SK 3550 इत्युपधालोपः । झलो झलि SK 2281 इति सलोपः । बब्धः । बप्सति ।{$ {!1101 कि!} ज्ञाने$} । चिकेति ।{$ {!1102 तुर!} त्वरणे$} । तुतोर्ति । तुतूर्तः । तुतुरति ।{$ {!1103 धिष!} शब्दे$} । दिधेष्टि । दिधिष्टः ।{$ {!1104 धन!} धान्ये$} । दधन्ति । दधन्तः । दधनति ।{$ {!1105 जन!} जनने$} । जजन्ति ॥

Rule. An abhyasta stem does not take guṇa of its light penultimate vowel (laghūpadha). This holds before a pit sārvadhātuka affix that begins with a vowel. guṇaḥ is carried forward from SK 2346, and laghūpadhasya "of the light penultimate" from SK 2189.

Example.

nenij + āni (loṭ, uttama singular) → āṭ is pit (SK 2204), so the ending is not ṅit

→ guṇa of the light penultimate i would come (SK 2189)

→ this rule forbids it → nenijāni "let me cleanse".

Applies when / blocked by. One of Bhaṭṭoji's commentators shows the conditions:

not abhyasta: dveṣāṇi;
not vowel-initial: veveṣṭi;
not sārvadhātuka: nineja.

He adds that "pit" is in the sūtra for the sake of the rule after it, SK 2443 (uto vṛddhir luki hali), taught in chapter 45.

More forms of ṇij:

laṅ anenek, aneniktām, anenijuḥ;
vidhiliṅ nenijyāt; āśīrliṅ nijyāt;
luṅ anijat, with aṅ because the root is marked with ir, or anaikṣīt;
luṅ ātmanepada anikta, with sic lost by SK 2281.

Bhaṭṭoji then takes the other two roots.

1094 vijir "in separation": vevekti; ātmanepada vevikte. liṭ vivejitha, with iṭ, takes guṇa. SK 2536 (vija iṭ) makes an iṭ-initial ending after vij ṅit-like, but that rule takes only the root ovijī. ṇij and vij are also listed in the rudhādi class.
1095 viṣḷ "in pervading": veveṣṭi; ātmanepada veviṣṭe. Its ḷ gives aṅ in luṅ: aviṣat. In the ātmanepada, ksa comes (SK 2336), and its a is lost before a vowel (SK 2337): avikṣata, avikṣātām, avikṣanta.

Bhaṭṭoji now says that the roots to the end of the class are parasmaipadin and Vedic (chāndasa).

1096 ghṛ "in sprinkling and in shining": jigharmy agniṃ haviṣā ghṛtena. The i of the abhyāsa comes from SK 2496 read with bahulaṃ chandasi (SK 3598), "variously in the Veda". The ordinary form is jagharti.
1097 hṛ "in acting by force": ayaṃ sruvo abhijiharti homān. Here too the i of the abhyāsa is Vedic.
1098 ṛ and 1099 sṛ "in going". SK 3598 alone would give ṛ its i in the Veda. So SK 2493 names ṛ to show that it is used with ślu in ordinary speech (bhāṣā) as well. The derivation:

ṛ + ti → ślu and doubling → ṛ-ṛ-ti

→ i for the abhyāsa (SK 2493) → i-ṛ-ti

→ guṇa of the second part → i-ar-ti

→ iyaṅ for the abhyāsa before an unlike vowel (SK 2290) → iyarti "he goes".

iyṛtaḥ (ṅit, no guṇa); iyrati (at, then r for ṛ).

Further forms of ṛ:
liṭ āra, āratuḥ, and āritha with iṭ always (SK 2384);
luṭ artā; lṛṭ ariṣyati, with iṭ by SK 2366;
loṭ iyarāṇi; laṅ aiyaḥ, aiyṛtām, aiyaruḥ;
vidhiliṅ iyṛyāt; āśīrliṅ aryāt, with guṇa by SK 2380;
luṅ ārat, with aṅ (SK 2382) and guṇa (SK 2406).
From sṛ: sasarti.
1100 bhasa "in reviling and in shining": babhasti. In the dual:

babhas + tas → the penultimate a is lost (SK 3550) → babhs + tas

→ s is lost between jhals (SK 2281) → babh + tas

→ t becomes dh (SK 2280), and bh becomes b (jaśtva) → babdhaḥ.

Plural bapsati.

1101 ki "in knowing": ciketi.
1102 tura "in hastening": tutorti, tutūrtaḥ (lengthening by SK 354), tuturati.
1103 dhiṣa "in sound": didheṣṭi, didhiṣṭaḥ.
1104 dhana "in grain", that is, in acquiring grain: dadhanti, dadhantaḥ, dadhanati. In dadhantaḥ the vowel before the nasal is not lengthened; scholars discuss the reason.
1105 jana "in janana" (arising or producing): jajanti.

Watch out. In jigharmi and abhijiharti (and jigāti, SK 2504) the i of the abhyāsa is Vedic, by SK 2496 read with SK 3598. SK 2496 itself covers just three roots.

H. jan (and gā): ā for the final before a jhal-initial san or kṅit — running example: `ja-jan + tas → ja-jā-tas → jajātaḥ`; `ja-jan + ati → jajn-ati → jajñati`

The last rule gives ā for the final n of jan, san and khan before certain affixes. The affix must begin with a jhal (a stop, sibilant or h). So ja-jan + tas gives jajātaḥ, but ja-jan + ati keeps its n and gives jajñati. The chapter closes with the Vedic root gā "to praise".

SK 2504 · जनसनखनां सञ्झलोः · janasanakhanāṃ sañjhaloḥ (A 6.4.42)

जनसनखनां सञ्झलोः

janasanakhanāṃ sañjhaloḥ · A 6.4.42

एषामाकारोऽन्तादेशः स्याज्झलादौ सनि झलादौ क्ङिति च । जजातः । जज्ञति । जजंसि । जजान । जजन्यात् । जजायात् । जन्यात् । जायात् ।{$ {!1106 गा!} स्तुतौ$} । देवाञ्जिगाति सुम्नयुः । जिगीतः । जिगति ॥

Rule. For jan, san and khan, ā replaces the final. This happens in two settings:

before the affix san beginning with a jhal;
before a kṅit affix beginning with a jhal.

āt "the vowel ā" is carried forward from SK 2982. jhali kṅiti is carried forward from SK 2428. One of Bhaṭṭoji's commentators explains how the two parts fit. The carried jhal qualifies san. The sūtra's own jhal qualifies kṅit.

Example.

jan + tas → ślu, doubling, abhyāsa operations → ja-jan + tas

→ tas begins with t, a jhal, and is ṅit → ā for n (this rule) → ja-jā-tas → jajātaḥ "they two are born / produce".

ja-jan + ati → ati is ṅit but begins with a vowel, so no ā

→ the penultimate a is lost (SK 2363) → jajn + ati

→ n becomes ñ after j (ścutva) → jajñati "they are born / produce".

jajaṃsi, with anusvāra (SK 123); liṭ jajāna.

Applies when / blocked by. Before a y-initial ending the ā is optional, by SK 2319:

vidhiliṅ jajanyāt or jajāyāt;
āśīrliṅ janyāt or jāyāt.

Why must san begin with a jhal? One of Bhaṭṭoji's commentators gives jijaniṣati, sisaniṣati, cikhaniṣati. There san begins with iṭ, so no ā. With a jhal-initial san: siṣāsati. The Bhāṣya does without the word "jhal" here; scholars discuss this.

One of Bhaṭṭoji's commentators says jan covers two roots: jana "to produce" and janī "to arise".

The last root of the class is 1106 gā "in praising", a Vedic root.

devāñ jigāti sumnayuḥ "desiring favour, he praises the gods". The abhyāsa has i by SK 3598.
jigītaḥ "they two praise", with ī by SK 2497.
jigati "they praise", with at and the ā lost by SK 2483.
One of Bhaṭṭoji's commentators adds luṅ agāsīt.

The printed text closes: iti juhotyādayaḥ, "Here end the juhotyādi roots."

Terms used in this chapter

juhotyādi — "the roots beginning with juhoti": the third class of the Dhātupāṭha, headed by hu (SK 2489).
dhātu / Dhātupāṭha — a root / the list of roots; the numbers 1083–1106 are its serial numbers.
it — a marker sound, not kept in the finished form, such as the ñi of ñibhī.
anuvṛtti — carrying a word forward from another sūtra into this one; here often from a rule taught later in the Kaumudī.
adhikāra — a heading whose word is read into later rules, such as aṅgasya (SK 2494).
vikaraṇa — an affix placed between a root and its ending, such as śap or śnā.
śap — the default vikaraṇa a before a sārvadhātuka in the agent sense.
ślu — a name for the deletion of an affix (SK 260); here it removes śap (SK 2489).
parasmaipadin / ātmanepadin / ubhayapadin — a root that takes the parasmaipada endings / the ātmanepada endings / both.
Bhāṣya — Patañjali's great commentary on the Aṣṭādhyāyī.
prakṣepa — casting, the sense of "giving" in hu (SK 2489).
haviṣ — an oblation.
svabhāva — here, established usage, from which a sense is known without a rule (SK 2489).
homa — an offering into the sacred fire.
cānaś — a participle affix; it makes juhvānāḥ correct (SK 2489).
abhyāsa — the first of the two copies made by doubling.
abhyasta — the two copies of a doubled root taken together (SK 426); such a stem takes at and jus.
guṇa / vṛddhi — the vowels a, e, o / ā, ai, au.
pit / apit — an affix marked with p / one not so marked; a sārvadhātuka that is not pit counts as ṅit.
kit / ṅit / kṅit — an affix marked with k / with ṅ / with either; such an affix blocks guṇa.
at — the substitute a for the jh of an ending after an abhyasta (SK 2479).
yaṇ — the semivowels y v r l, replacing i u ṛ ḷ before a vowel.
uvaṅ — the substitute uv for u before a vowel (SK 271).
apavāda — an exception: a rule that sets aside a more general one.
ām — the affix that some roots take before liṭ, with a helping verb after it (SK 2306, 2491).
śluvadbhāva — treatment of a root before ām as if ślu followed, so that it is doubled (SK 2491).
para / paratva — later in the order of the Aṣṭādhyāyī / the principle that the later of two conflicting rules wins (SK 2491, 2494).
āṭ — the pit augment ā of the loṭ uttama endings (SK 2204).
jus — the substitute us for jhi after an abhyasta (SK 2226).
sic — the affix s of luṅ.
sārvadhātuka / ārdhadhātuka — the endings and ś-marked affixes after a root / the other affixes after a root.
hal — a consonant; "hal-initial" means beginning with a consonant.
iyaṅ — the substitute iy for i or ī before a vowel: for a root's final vowel by SK 271 (jihriyati), for the abhyāsa by SK 2290 (iyarti).
raparatva — the rule that a vowel replacing ṛ is followed by r (SK 70).
oṣṭhya — a labial, a sound made with the lips (SK 2494).
aṅga — the stem before an affix.
kryādi — the class of roots beginning with krī; śṝ and dṝ belong to it (SK 2495).
hrasva — a short vowel (SK 2495).
iṭ / aniṭ — the augment i / a root that does not take it.
chāndasa / chāndasatva — Vedic / being Vedic; a form explained only as Vedic usage (SK 2495).
bahulaṃ chandasi — SK 3598, "variously in the Veda": the source of the Vedic i in jigharmi and jigāti.
bhāṣā — ordinary, non-Vedic speech (SK 2503).
riṅ — the substitute ri for ṛ before certain affixes (SK 2367).
cutva — replacement of a velar or h by a palatal, as in the abhyāsa of jihīte.
śnā — the vikaraṇa seen in lunāti and lunīte (SK 2497).
ghu — the name of the roots dā and dhā (SK 2373); they are excepted from SK 2497.
lopa — deletion (SK 2500).
luk — deletion of an affix, as of sic in adāt.
anukarṣa — "drawing along": a ca in a sūtra that pulls in items from an earlier one (SK 2499).
jhaṣ / baś / bhaṣ — voiced aspirates (jh bh gh ḍh dh) / voiced unaspirated stops (b g ḍ d) / the voiced aspirates that replace a baś (SK 2501).
cartva — replacement by an unaspirated voiceless stop, as dh by t in dhattaḥ.
jaśtva — replacement by an unaspirated voiced stop, as bh by b in babdhaḥ.
sthānivadbhāva — treatment of a substitute like what it replaced (SK 49; for a vowel's substitute, SK 50, as at SK 2501).
vacanasāmarthya — "the force of the statement": the argument that a rule must apply somewhere, or it would be pointless (SK 2501).
Tripādī — the last three quarters of the Aṣṭādhyāyī (A 8.2–8.4), whose rules are asiddha (SK 12).
asiddha — treated as not having taken effect.
pūrvatrāsiddha — the asiddha zone of SK 12; a substitute gets no sthānivadbhāva for a rule of this zone (SK 2501).
Vāmana and Mādhava — earlier commentators whose answer Bhaṭṭoji reports at SK 2501.
svarita — one of the three accents of a vowel.
svaritet — a root whose it-vowel is svarita; it takes both sets of endings.
laghūpadha — having a light penultimate vowel (SK 2503).
irit / ḷdit — a root marked with ir / with ḷ; both may take aṅ in luṅ.
aṅ — an affix a that replaces the luṅ affix in some roots.
ksa — the luṅ affix sa of certain roots (SK 2336).
rudhādi — the class of roots beginning with rudh.
kutva — replacement of a palatal by a velar (SK 378), as j by g in nenigdhi.
jhal — a stop, a sibilant or h: every consonant except nasals and semivowels.
san — an affix added to a root, as in siṣāsati (SK 2504).
ścutva — replacement of a dental by a palatal next to a palatal, as n by ñ in jajñati.
anusvāra — the nasal ṃ, as in jajaṃsi (SK 123).
prathama / madhyama / uttama — the three persons of the endings: "he", "you", "I".
laṭ, liṭ, luṭ, lṛṭ, loṭ, laṅ, vidhiliṅ, āśīrliṅ, luṅ — the lakāras, as in chapter 44.

Summary

SK	what it does	example
2489	after hu and the roots after it, ślu deletes śap	`hu + śap + ti → hu + ti`
2490	before ślu, the root is doubled	`hu-hu-ti → juhoti`; `juhvati`
2491	after bhī, hrī, bhṛ, hu, optional ām in liṭ, treated as before ślu	`juhavāṃcakāra` / `juhāva`; guṇa wins in `juhavāni`, `ajuhavuḥ`
2492	bhī: optional i before a hal-initial kṅit sārvadhātuka	`bibhitaḥ` / `bibhītaḥ`
2493	ṛ and pṝ: i for the abhyāsa before ślu	`piparti`, `iyarti`
2494	u for an aṅga-final ṝ after a labial of the aṅga; guṇa and vṛddhi win	`pipūrtaḥ`; but `piparti`, `papāra`
2495	śṝ, dṝ, pṝ: optional short vowel before a kit liṭ	`papratuḥ` / `paparatuḥ`
2496	bhṛñ, māṅ, ohāṅ: i for the abhyāsa before ślu	`bibharti`, `mimīte`, `jihīte`; but `babhāra`
2497	ī for the ā of śnā or an abhyasta before a hal-initial kṅit; not ghu	`mimīte`, `jigītaḥ`; but `dattaḥ`
2498	jahāti: optional i there, else ī	`jahitaḥ` / `jahītaḥ`
2499	jahāti before hi: ā, i or ī	`jahāhi` / `jahihi` / `jahīhi`
2500	jahāti: ā lost before a y-initial sārvadhātuka	`jahyāt`; but `heyāt`
2501	doubled dhā (dadh): d → dh before t, th, s, dhv	`dhattaḥ`, `dhatse`, `dhaddhve`; but `dadhvaḥ`
2502	ṇij, vij, viṣ: guṇa of the abhyāsa before ślu	`nenekti`, `veveṣṭi`; but `nineja`
2503	an abhyasta takes no light-penultimate guṇa before a vowel-initial pit sārvadhātuka	`nenijāni`
2504	jan, san, khan: ā for the final before a jhal-initial san or kṅit	`jajātaḥ`; but `jajñati`

← textbook contents reading page (Technical) →

46. Tiṅanta juhotyādi-prakaraṇa — Verbs: the hu class, whose roots are doubled before the endings

Before you start

A. hu: ślu, doubling, ām with śluvadbhāva, guṇa by paratva — running example: hu + śap + ti → hu + ti → hu-hu-ti → ju-ho-ti → juhoti; juhu + jhi → juhu + ati → juhvati

SK 2489 · जुहोत्यादिभ्यः श्लुः · juhotyādibhyaḥ śluḥ (A 2.4.75)

SK 2490 · श्लौ · ślau (A 6.1.10)

SK 2491 · भीह्रीभृहुवां श्लुवच्च · bhīhrībhṛhuvāṃ śluvac ca (A 3.1.39)

B. bhī and hrī: optional i before a hal-initial kṅit — running example: bi-bhī + tas → bibhitaḥ / bibhītaḥ

SK 2492 · भियोऽन्यतरस्याम् · bhiyo 'nyatarasyām (A 6.4.115)

C. pṝ (and ṛ): i in the abhyāsa, u for ṝ after a labial, the short vowel in liṭ — running example: pṝ-pṝ-ti → pir-pṝ-ti → pi-pṝ-ti → pi-par-ti → piparti; pi-pṝ-tas → pi-pur-tas → pipūrtaḥ; pa-pṝ-atus → papratuḥ / paparatuḥ

SK 2493 · अर्तिपिपर्त्योश्च · artipipartyoś ca (A 7.4.77)

SK 2494 · उदोष्ठ्यपूर्वस्य · ud oṣṭhyapūrvasya (A 7.1.102)

SK 2495 · शॄदॄप्रां ह्रस्वो वा · śṝdṝprāṃ hrasvo vā (A 7.4.12)

D. bhṛñ, māṅ, ohāṅ: i in the abhyāsa; ī for the ā of an abhyasta — running example: bhṛ-bhṛ-ti → bi-bhṛ-ti → bibharti; mā-mā-te → mi-mā-te → mimīte

SK 2496 · भृञामित् · bhṛñām it (A 7.4.76)

SK 2497 · ई हल्यघोः · ī haly aghoḥ (A 6.4.113)

E. ohāk (jahāti): i or ī, ā/i/ī before hi, loss before y; then dā and dhā — running example: ja-hā + tas → jahitaḥ / jahītaḥ; ja-hā + hi → jahāhi / jahihi / jahīhi; ja-hā + yāt → jahyāt

SK 2498 · जहातेश्च · jahāteś ca (A 6.4.116)

SK 2499 · आ च हौ · ā ca hau (A 6.4.117)

SK 2500 · लोपो यि · lopo yi (A 6.4.118)

F. dhā: dh (a bhaṣ) for the d of dadh — running example: da-dhā + tas → dadh + tas → dhadh + tas → dhattaḥ

SK 2501 · दधस्तथोश्च · dadhas tathoś ca (A 8.2.38)

G. ṇij, vij, viṣ: guṇa of the abhyāsa, no light-penultimate guṇa; the Vedic run ghṛ … jana — running example: nij-nij-ti → ni-nij-ti → ni-nej-ti → ne-nej-ti → nenekti; nenij + āni → nenijāni (not *nenejāni)

SK 2502 · निजां त्रयाणां गुणः श्लौ · nijāṃ trayāṇāṃ guṇaḥ ślau (A 7.4.75)

SK 2503 · नाभ्यस्तस्याचि पिति सार्वधातुके · nābhyastasyāci piti sārvadhātuke (A 7.3.87)

H. jan (and gā): ā for the final before a jhal-initial san or kṅit — running example: ja-jan + tas → ja-jā-tas → jajātaḥ; ja-jan + ati → jajn-ati → jajñati

SK 2504 · जनसनखनां सञ्झलोः · janasanakhanāṃ sañjhaloḥ (A 6.4.42)

Terms used in this chapter

Summary

A. hu: ślu, doubling, ām with śluvadbhāva, guṇa by paratva — running example: `hu + śap + ti → hu + ti → hu-hu-ti → ju-ho-ti → juhoti`; `juhu + jhi → juhu + ati → juhvati`

B. bhī and hrī: optional i before a hal-initial kṅit — running example: `bi-bhī + tas → bibhitaḥ / bibhītaḥ`

C. pṝ (and ṛ): i in the abhyāsa, u for ṝ after a labial, the short vowel in liṭ — running example: `pṝ-pṝ-ti → pir-pṝ-ti → pi-pṝ-ti → pi-par-ti → piparti`; `pi-pṝ-tas → pi-pur-tas → pipūrtaḥ`; `pa-pṝ-atus → papratuḥ / paparatuḥ`

D. bhṛñ, māṅ, ohāṅ: i in the abhyāsa; ī for the ā of an abhyasta — running example: `bhṛ-bhṛ-ti → bi-bhṛ-ti → bibharti`; `mā-mā-te → mi-mā-te → mimīte`

E. ohāk (jahāti): i or ī, ā/i/ī before hi, loss before y; then dā and dhā — running example: `ja-hā + tas → jahitaḥ / jahītaḥ`; `ja-hā + hi → jahāhi / jahihi / jahīhi`; `ja-hā + yāt → jahyāt`

F. dhā: dh (a bhaṣ) for the d of dadh — running example: `da-dhā + tas → dadh + tas → dhadh + tas → dhattaḥ`

G. ṇij, vij, viṣ: guṇa of the abhyāsa, no light-penultimate guṇa; the Vedic run ghṛ … jana — running example: `nij-nij-ti → ni-nij-ti → ni-nej-ti → ne-nej-ti → nenekti`; `nenij + āni → nenijāni` (not *nenejāni)

H. jan (and gā): ā for the final before a jhal-initial san or kṅit — running example: `ja-jan + tas → ja-jā-tas → jajātaḥ`; `ja-jan + ati → jajn-ati → jajñati`