तद्धिताधिकारे प्राग्घितीयप्रकरणम्

Covers SK 1626–1660 (35 sūtras): all present. The spine is Aṣṭādhyāyī 4.4.75–4.4.109, in order, with two departures. First, the SK splices in 8.2.79 (na bhakurchurām, SK 1629) after dhuro yaḍḍhakau, because it completes that rule's derivation. Second, the SK omits 4.4.106 (ḍhaś chandasi), a Vedic rule it keeps for the vaidikī section (SK 3455). So SK 1657 is 4.4.105 and SK 1658 is 4.4.107.

taddhitādhikāre prāgghitīyaprakaraṇam — outline (p32, SK 1626–1660, all 35 sūtras)

This prakaraṇa is the yat heading. prāg ghitād yat (1626) makes yat the default affix for every sense taught before tasmai hitam (5.1.5, SK 1665). The prakaraṇa itself ends at 4.4.109 (SK 1660). SK 1661–1664 (5.1.1–5.1.4) fall in p33. As in p31, the rules come in runs, and the case of the base shifts from run to run:

• second case, tat (carried from 1627 tad vahati): "draws / conveys". Later senses in the run: pierces, obtains, has gone under.
• first case (the second-case tat recast): "is visible in this", "of this [the root is] to be pulled up".
• third case: joined with; crossed by, equal to, attained by, slain by, gained by, measured by; made by. Also one fifth-case rule: "not departed from".
• sixth case: pleasing to, binding of; the karaṇa, talk or ploughing of.
• seventh case, tatra sādhuḥ (1650): "good at / fit for that". Then, after sādhu lapses (1658), "dwells in" and "lain in".

From saṃjñāyāṃ dhenuṣyā (1641) onward, the heading saṃjñāyām runs to the end of the pāda (BM). The commentators use it again and again to restrict the derivatives to their established referents: vayasya is a friend (not an equal-aged enemy), urasya is a son, tīrtha means a teacher.

Most sūtras name a new sense and leave yat in place. The rest are apavādas for particular bases:

• ḍhak (with yat), kha, khañ, aṇ, ṇa, ṇya, ñya, ṭhak, ṭhañ, ḍhañ;
• ya, a separate affix without the t-marker, taught three times (1641, 1657, 1660) to avoid the tit-svara;
• one luk (1631).

Only the kaumudī blocks are translated (SK-TRANSLATE §25). The readings cited as BM (Bālamanoramā) and TB (Tattvabodhinī) are the ones recorded in the unit notes. Where the p32 audit (.work/sk-run/p32-audit-fixes.md) corrects a reading, this outline follows the corrected reading. This applies to 1629, 1630, 1633, 1637, 1640, 1643, 1647–1649 and 1650–1657. Ordering: SK order throughout; nothing is reordered.

Affix shapes used below:

• yat → ya. The t is for accent (tit-svara), and yat brings no vṛddhi. A final a/ā/i/ī of the stem is deleted before it by yasyeti ca (6.4.148, SK 311): ratha → rathyaḥ.
• ḍh → ey and kh → īn, by āyaneyīnīyiyaḥ… (7.1.2, SK 475).
• ṭh → ik by ṭhasyekaḥ (7.3.50, SK 1170), but ṭh → k after a stem in u, by isusuktāntāt kaḥ (7.3.51, SK 1221): sāktukāḥ.
• ñ-, ṇ- and k-marked affixes (ḍhak, khañ, aṇ, ṇa, ṇya, ñya, ṭhak, ṭhañ, ḍhañ) bring ādivṛddhi (7.2.117–118, SK 1075–1076). kha and ya do not.

A. The yat heading and "draws it" (tad vahati) — running example: rathaṃ vahati → rathyaḥ

• 1626 prāg ghitād yat 4.4.75 — adhikāra: yat governs the senses taught before tasmai hitam (SK 1665), and that sūtra itself is excluded. BM: "hita" names the sūtra that contains it. No TB on this heading.
• 1627 tad vahati rathayugaprāsaṅgam 4.4.76 — yat after second-case ratha, yuga and prāsaṅga, in the sense "draws that". The base is the thing drawn; the derivative names the animal that draws it.
• ratha + yat → a deleted: rathyaḥ.
• yuga → yugyaḥ.
• A prāsaṅga is a piece of wood fastened on the shoulder of young steers being broken in: prāsaṅgyaḥ. BM speaks of horses and cites Amara; the kaumudī's vatsa is kept.
• tad vahati is the limit of p31's ṭhak heading (prāg vahateḥ, SK 1548).

B. dhur, the yoke-pole, and its compounds — running example: dhuraṃ vahati → dhuryaḥ / dhaureyaḥ

• 1628 dhuro yaḍḍhakau 4.4.77 — after second-case dhur, yat and ḍhak, drawing. With yat, the u of dhur would be lengthened by hali ca (8.2.77, SK 354), since a consonant (the y) follows the r. The kaumudī breaks off here, and the next sūtra completes the sentence.
• TB: dhuro ḍhak ca would have been the easier wording, with yat coming from the heading.
• 1629 na bhakurchurām 8.2.79 — spliced in here. The penultimate is not lengthened in a bha, or in kur and chur. Before yat, dhur is a bha (yaci bham, SK 231), so the u stays short: dhuryaḥ.
• The kaumudī also lists dhaureyaḥ, the ḍhak-form of 1628: ḍh → ey, and ādivṛddhi u → au.
• dhaureyaḥ does not need this prohibition. With eya, no consonant follows the r, so hali ca could never apply. The prohibition is needed only for dhuryaḥ.
• BM draws the anuvṛtti of upadhā from rvor upadhāyāḥ (8.2.76). TB draws it from upadhāyāṃ ca (8.2.78) and gives the counter-example pratidīvnaḥ.
• 1630 khaḥ sarvadhurāt 4.4.78 — kha after second-case sarvadhurā ("the whole yoke"), drawing. sarvadhurāṃ vahati → sarvadhurīṇaḥ: kha → īna, and the ā is deleted. The n → ṇ is not stated by the kaumudī.
• The digital source prints sarvadharām. NSP 1915 reads sarvadhurām, and BM and TB agree, so the corruption is confirmed.
• BM and TB: sarvā dhūḥ is a tatpuruṣa. It takes the samāsānta a by ṛkpūrabdhūḥ… (5.4.74, SK 940), then ṭāp.
• 1631 ekadhurāl luk ca 4.4.79 — after ekadhurā, kha and, by ca, luk: ekadhurīṇaḥ and ekadhuraḥ. What the luk deletes is disputed:
• BM: kha is deleted, optionally.
• TB: the heading's yat is deleted, and kha, being expressly enjoined, is always heard.
• The translation stays neutral ("with luk").

C. The last "draws / conveys" rules: cart, plough, bride — running example: śakaṭaṃ vahati → śākaṭaḥ

• 1632 śakaṭād aṇ 4.4.80 — aṇ after second-case śakaṭa ("cart"). With ādivṛddhi, and the a deleted: śākaṭo gauḥ ("an ox that draws a cart").
• TB: tasyedam's aṇ would already give this form. The rule is undertaken all the same, and by ārambhasāmarthya that yields tadantavidhi. Hence dvaiśakaṭaḥ, where the dvigu luk by dvigor lug anapatye (SK 1080) does not occur.
• No BM on this sūtra. The BM block filed under 1632 is BM on SK 1621.
• 1633 halasīrāṭ ṭhak 4.4.81 — ṭhak after second-case hala and sīra ("plough"): hālikaḥ and sairikaḥ.
• TB: the identical sūtra 4.3.124 (SK 1504) already stands in the section opened by tasyedam (SK 1500). This rule's fruit is again tadantavidhi: dvaihalikaḥ, dvaisīrikaḥ.
• 1634 saṃjñāyāṃ janyāḥ 4.4.82 — yat after second-case janī ("bride"), "conveys her", as a name. janyā is one who brings the bride [to the bridegroom].
• TB takes it as the mother's female companion. BM reads the plural janyāḥ: the bridegroom's friends (Amara). The kaumudī's singular is translated with a neutral "one who".

D. Second case, other verbs: pierces, obtains, gone under — running example: pādau vidhyanti → padyāḥ śarkarāḥ

• 1635 vidhyaty adhanuṣā 4.4.83 — yat after a second-case word, "pierces it", provided a bow is not the instrument. pāda + yat → pad for pāda by padyaty atadarthe (6.3.53, SK 991), a step BM gives and the kaumudī does not: padyāḥ śarkarāḥ ("pebbles that pierce the feet").
• BM's counter-example is dhanuṣā coraṃ vidhyati.
• TB instead supplies "presumed" (saṃbhāvyate) and says that an explicit dhanuṣā is excluded by sāpekṣatva anyway.
• 1636 dhanagaṇaṃ labdhā 4.4.84 — yat after dhana and gaṇa, "obtainer": dhanyaḥ, gaṇyaḥ. labdhā ends in tṛn. By na lokāvyaya… (2.3.69, SK 627), a tṛn-form bars the sixth case of its object, so the second case of the sūtra stands.
• 1637 annāṇ ṇaḥ 4.4.85 — ṇa after anna, [ending in the second case], "obtainer". With ādivṛddhi: annaṃ labdhā → ānnaḥ.
• 1638 vaśaṃ gataḥ 4.4.86 — yat after second-case vaśa ("will"), "has gone under": vaśyaḥ, one who moves by another's will (parecchānucārī; BM's lemma reads -anusārī). vaśa is from vaś "desire", with ap by the vārttika vaśiraṇyor upasaṃkhyānam.

E. First case: "visible in this", "of this"; the saṃjñā heading begins — running example: mūlam ābarhi yeṣām → mūlyāḥ mudgāḥ

• 1639 padam asmin dṛśyam 4.4.87 — yat after pada ("footprint") in the first case, "is visible in this". BM: the carried second-case tat is recast (vipariṇamyate). padyaḥ kardamaḥ is mud in which a footprint shows, i.e. not too dry.
• Contrast 1635's padyāḥ, which comes from pāda with the pad substitute.
• 1640 mūlam asyābarhi 4.4.88 — yat after mūla ("root") [in the first case], "of this [the root] is ābarhi", i.e. to be pulled up (utpāṭana): mūlyāḥ mudgāḥ, mudga beans whose roots are to be pulled up. The bracketed gloss follows BM.
• TB: plants that cannot be gathered without uprooting, i.e. thoroughly dry.
• The source reads ābarh- and is followed. NSP 1915 reads āvarh- throughout, which agrees with TB's root vṛhū, against BM's bṛhū. This is a b/v textual variant.
• 1641 saṃjñāyāṃ dhenuṣyā 4.4.89 — nipātana, as a name. dhenu takes the augment ṣuk and the affix ya in its own sense: dhenuṣyā, a cow placed in pledge, given to the creditor for milking.
• The affix is ya, not yat, to avoid the tit-svara (TB, BM).
• BM: saṃjñāyām is an adhikāra from here to the end of the pāda.

F. Third case (and one fifth-case rule): joined with, crossed by, not departed from, made by — running example: nāvā tāryam → nāvyam

• 1642 gṛhapatinā saṃyukte ñyaḥ 4.4.90 — ñya after third-case gṛhapati ("the sacrificer"), "joined with". Ādivṛddhi, and the i deleted: gārhapatyaḥ (agniḥ), a particular fire.
• TB: because of the saṃjñā heading, the word does not extend to the āhavanīya and other fires.
• 1643 nauvayodharmaviṣamūlamūlasītātulābhyas tāryatulyaprāpyavadhyānāmyasamasamitasammiteṣu 4.4.91 — yat after eight third-case bases, in eight senses respectively (yathāsaṅkhya). mūla is read twice, and not reduced by ekaśeṣa, so that the count comes out right (BM).
• nau, "to be crossed" → au → āv before y (vānto yi pratyaye, SK 63): nāvyam.
• vayas, "equal" → vayasyaḥ, a friend (saṃjñā).
• dharma, "to be attained" → dharmyam.
• viṣa, "to be slain" → viṣyaḥ.
• mūla, "to be gained" → mūlyam, the profit over the capital.
• mūla, "equal" → mūlyaḥ.
• sītā ("furrow"), "joined with" (samita) → sītyam kṣetram, a ploughed or levelled field. The kaumudī reads saṃmitam, and NSP 1915 prints the same, so the reading is secure. It still departs from the sūtra's samita, which BM and TB gloss as saṅgata.
• tulā ("balance"), "measured" → tulyam.
• 1644 dharmapathyarthanyāyād anapete 4.4.92 — yat after dharma, pathin, artha and nyāya, which are in the fifth case by propriety (BM), "not departed from": dharmyam, pathyam, arthyam, nyāyyam.
• This dharmyam has a different sense from the one in 1643.
• TB: by the saṃjñā heading, pathya is only what keeps to the path laid down in śāstra.
• 1645 chandaso nirmite 4.4.93 — yat after third-case chandas, "made by". The kaumudī glosses chandas as desire (icchā): chandasyam, made at will. TB: not "metre", because of the saṃjñā heading.
• 1646 uraso 'ṇ ca 4.4.94 — aṇ and, by ca, yat after third-case uras ("breast"), "made". The referent is a son, by the saṃjñā heading: aurasaḥ (aṇ, ādivṛddhi) and urasyaḥ (yat).

G. Sixth case: pleasing to, binding of, and the karaṇa / talk / ploughing of — running example: hṛdayasya priyaḥ → hṛdyaḥ

• 1647 hṛdayasya priyaḥ 4.4.95 — yat after sixth-case hṛdaya, "pleasing to it". hṛd replaces hṛdaya before yat by hṛdayasya hṛllekha… (6.3.50, SK 988): hṛdyo deśaḥ.
• The digital source quotes the rule as "hṛdasya". NSP 1915 reads hṛdayasya, so the corruption is confirmed.
• TB: priya is agentive, "that which pleases", and the sixth case is objective.
• 1648 bandhane carṣau 4.4.96 — yat after sixth-case hṛdaya, in the sense bandhana, when the Veda (ṛṣi) is meant. With the same hṛd: hṛdyaḥ, a vaśīkaraṇa-mantra.
• bandhana is the means of binding (TB: lyuṭ in the instrument sense): that by which the heart is bound.
• TB: the rule does not apply when a seer such as Vasiṣṭha is meant.
• 1649 matajanahalāt karaṇajalpakarṣeṣu 4.4.97 — yat after sixth-case mata, jana and hala, in the senses karaṇa, jalpa ("talk") and karṣa ("ploughing") respectively.
• matyam: the karaṇa of knowledge, whether the act (bhāva) or the means (sādhana).
• janyaḥ: people's talk. This is distinct from 1634's janyā.
• halyaḥ: halasya karṣaḥ, the ploughing of the plough. The source prints karpaḥ. The sūtra's karṣa and TB's karṣaṇa require karṣa.
• TB: tadantavidhi gives dvihalyaḥ.

H. Seventh case: good at / fit for that (tatra sādhuḥ) — running example: sāmasu sādhuḥ → sāmanyaḥ

• 1650 tatra sādhuḥ 4.4.98 — yat after a seventh-case word, in the sense sādhu. Here sādhu means "good at / fit for": proficient (BM pravīṇa) or fit (TB yogya). It does not mean "good for" (hita), because that is the sense of tasmai hitam (SK 1665). The examples:
• agryaḥ ("good at the front");
• sāmanyaḥ, where the stem keeps its -an (prakṛtibhāva) by ye cābhāvakarmaṇoḥ (6.4.168, SK 1154);
• karmaṇyaḥ, likewise;
• śaraṇyaḥ.
• 1651 pratijanādibhyaḥ khañ 4.4.99 — khañ (→ īna, ādivṛddhi) after pratijana etc.
• pratijanaṃ sādhuḥ → prātijanīnaḥ, good towards every person. BM and TB: pratijanam is a distributive avyayībhāva.
• Also sāṃyugīnaḥ, sārvajanīnaḥ, vaiśvajanīnaḥ.
• 1652 bhaktāṇ ṇaḥ 4.4.100 — ṇa after seventh-case bhakta ("cooked rice"): bhakte sādhavaḥ → bhāktāḥ śālayaḥ, rice grains fit for cooked food. An NSP footnote glosses the word as niṣphalīkṛtāḥ, perhaps "husked".
• 1653 pariṣado ṇyaḥ 4.4.101 — ṇya after seventh-case pariṣad ("assembly"): pāriṣadyaḥ.
• The source prints pāripadyaḥ. NSP 1915 reads pāriṣadyaḥ.
• By yogavibhāga at pariṣadaḥ, ṇa also comes, with vṛddhi: pāriṣadaḥ. The source and the NSP OCR both print pariṣadaḥ; skk-tei reads pāriṣadaḥ, and grammar requires the vṛddhi.
• A variant reading has parṣad: pārṣadaḥ.
• This is not SK 1594 (4.4.44), which has the same wording and gives pāriṣadyaḥ in the sense "joins as a member".
• 1654 kathādibhyaṣ ṭhak 4.4.102 — ṭhak after kathā etc.: kathāyāṃ sādhuḥ → kāthikaḥ. BM calls this sūtra clear, and there is no TB.
• 1655 guḍādibhyaṣ ṭhañ 4.4.103 — ṭhañ after guḍa etc.
• guḍe sādhuḥ → gauḍika ikṣuḥ: sugar-cane fit for jaggery.
• saktu ends in u, so ṭh → k (SK 1221): sāktukā yavāḥ, barley fit for groats.
• TB lists the gaṇa (guḍa, kulmāṣa, saktu, apūpa, …).
• 1656 pathyatithivasatisvapater ḍhañ 4.4.104 — ḍhañ (→ eya, ādivṛddhi) after pathin, atithi, vasati and svapati.
• pathi sādhuḥ → pātheyam, what is fit for the road.
• ātitheyam.
• vāsateyī rātriḥ, a night fit for dwelling (vasati = vasana).
• svāpateyaṃ dhanam. TB: svapati is the owner of wealth (sva), i.e. a rich man (OCR ātmyas ≈ āḍhyaḥ), and svāpateya is wealth fit for him. The same form is the svāgatādi example at SK 1549, where the aic augment is barred.
• 1657 sabhāyā yaḥ 4.4.105 — ya after sabhā [ending in the seventh case]: sabhyaḥ, good in an assembly.
• TB: the affix is ya, not yat, to avoid the tit-svara.
• 4.4.106 ḍhaś chandasi follows in the Aṣṭādhyāyī but is omitted here (SK 3455).

I. Seventh case, sādhu lapsed: dwells in, lain in — running example: samāne udare śayitaḥ → samānodaryaḥ

• 1658 samānatīrthe vāsī 4.4.107 — the kaumudī says sādhur iti nivṛttam: the sense sādhu has lapsed, and the seventh case continues (BM). yat after samānatīrtha, "dwells in".
• vāsin is a ṇini-form.
• tīrtha is only "teacher", by the saṃjñā heading (TB).
• sa replaces samāna by tīrthe ye (6.3.87, SK 1015; BM and TB): satīrthyaḥ, a fellow-student.
• 1659 samānodare śayita o codāttaḥ 4.4.108 — yat after seventh-case samānodara, "lain in". śayita = sthita, "situated". The o is udātta, an apavāda of the tit-svara: samānodaryaḥ, a brother.
• BM: this is the form when vibhāṣodare (6.3.88, SK 1016) does not give sa.
• 1660 sodarād yaḥ 4.4.109 — ya after sodara (sa for samāna by vibhāṣodare, SK 1016), in the same sense: sodaryaḥ.
• TB: o codāttaḥ does not carry over. BM: with no t-marker, the word is end-acute by pratyayasvara.
• BM and TB: Murāri's sodara is a bahuvrīhi, with sa for saha (SK 849), so sahodara is also correct.
• The prāgghitīya section closes here. The colophon, printed in NSP, is missing from source.json. prāk krītāc chaḥ (SK 1661) opens p33.