After this chapter you will be able to derive the taddhita words for a long chain of senses. kaṣāyeṇa raktaṃ vastram "a cloth dyed with kaṣāya" is one word, kāṣāyam. indro devatāsya "Indra is its deity" is aindram. kākānāṃ samūhaḥ "a collection of crows" is kākam. vyākaraṇam adhīte "he studies grammar" is vaiyākaraṇaḥ. You will also derive the words for places: udumbarāḥ santy asmin deśe "a country with udumbara trees" is audumbaraḥ, and the river with sugarcane is ikṣumatī. You will know which special affix a given base takes in each sense, and the stem rules those forms need. You will know why the name of a country such as pañcālāḥ is plural, and why Pāṇini says that rule need not have been taught at all. You will follow, step by step, how Bhaṭṭoji derives the words pāṇini and pāṇinīya. There are 110 sūtras, and every one is worked on a concrete word.
kaṣāyeṇa raktaṃ vastram → kāṣāyamThis chapter opens a long chain of single senses. Each is set up by a heading sūtra that names a case and a meaning. The affix is aṇ and the rest, and special rules give other affixes for particular bases. Bhaṭṭoji's commentators call the stretch up to SK 1278 the section of "dyed with" and the following senses. The first sense is "dyed with that". On the running example, a cloth dyed with kaṣāya, a mineral dye, is kāṣāyam.
Rule. After a word in the third case (instrumental) that names a rāga, a dye, the affixes aṇ etc. come in the sense "dyed with that". A rāga is "that by which something is dyed".
Example.
kaṣāyeṇa raktaṃ vastram "a cloth dyed with kaṣāya"
→ the affix goes after the instrumental word kaṣāya → kaṣāya + aṇ
→ aṇ is ṇit, so first-vowel vṛddhi (SK 1075): a → ā → kāṣāya + a
→ the final a of the bha stem is deleted by yasyeti ca (SK 311) → kāṣāyam.
The printed text adds the form māñjikam. Bhaṭṭoji's commentators read māñjiṣṭham "dyed with madder" (mañjiṣṭhā, a particular dye), which is probably the right reading.
Applies when / blocked by. The base must name a dye. devadattena raktaṃ vastram "a cloth dyed by Devadatta" takes no affix: Devadatta dyed the cloth, but he is not a dye.
Rule. After lākṣā "lac" and rocanā (a yellow pigment), ṭhak comes in place of aṇ, in the sense "dyed with that".
Example.
lākṣayā raktaḥ "dyed with lac" → lākṣā + ṭhak
→ ṭh → ik (SK 1170), k is a marker, so vṛddhi (SK 1076): ā stays ā → lākṣā + ika
→ the final ā is deleted (SK 311) → lākṣikaḥ.
Likewise rocanā → raucanikaḥ, with vṛddhi o → au.
Applies when / blocked by. Four vārttikas extend the rule:
śākalikaḥ, kārdamikaḥ. Bhaṭṭoji reports that the Vṛttikāra (the author of the old running commentary) allows aṇ here too: śākalaḥ, kārdamaḥ. His commentators note that this option is not in the Bhāṣya.nīlyā raktam → nīlam.pītakam, with no vṛddhi.hāridram, māhārajanam. Bhaṭṭoji's commentators note that añ differs from aṇ only in accent.Bhaṭṭoji's commentators close the section on "dyed with" here.
puṣyeṇa yuktam ahaḥ → pauṣam ahaḥ, but adya puṣyaḥ (with lup)A nakṣatra is a lunar mansion, one of the star-groups the moon passes through. This group names a time after the nakṣatra the moon stands in. The affix normally appears (pauṣam ahaḥ "the Pauṣa day"). But when no part of the day is specified, it disappears by lup (adya puṣyaḥ "today is Puṣya").
Rule. After a third-case word naming a nakṣatra, aṇ etc. come in the sense "a time joined with that". Bhaṭṭoji's commentators explain that "joined with Puṣya" means joined with the moon standing in Puṣya.
Example.
puṣyeṇa yuktam ahaḥ "a day joined with Puṣya" → puṣya + aṇ
→ vṛddhi u → au → pauṣya + a
→ the y is dropped by a vārttika (tiṣyapuṣyayor nakṣatrāṇi yalopaḥ, cited by Bhaṭṭoji's commentators), and the final a by SK 311 → pauṣam ahaḥ.
For a night, the feminine takes ṅīp (SK 470) → pauṣī rātriḥ "the Pauṣī night".
Rule. The affix of SK 1204 disappears by lup when no internal division of the time is understood. The time here is the sixty-daṇḍa day (a full day and night). Its internal divisions are day and night.
Example.
adya puṣyaḥ "today is Puṣya" → puṣya + aṇ → no division of the day-and-night is meant → lup of aṇ → no affix, so no vṛddhi → puṣyaḥ.
Applies when / blocked by. Bhaṭṭoji raises an objection. The full-moon day joined with Puṣya is also a whole day. Why then is it pauṣī and not puṣyā? The answer is a jñāpaka, a wording that reveals a principle. SK 1225 names the full-moon days as phālgunī, kārttikī, caitrī, with the affix kept. This shows that the lup of this rule does not reach the full-moon day.
Watch out. Śravaṇa is the exception. SK 1225 names its full-moon as śravaṇā, with no affix and in the feminine. By that very form there is lup even for the full-moon, and gender does not follow the base (as SK 1294 would require). And a nipātana does not block the regular form either, so śrāvaṇī is also correct. Whether nipātanas block is a point scholars discuss.
Rule. After śravaṇa and aśvattha (a name of the nakṣatra Aśvinī, according to Bhaṭṭoji's commentators), the affix undergoes lup when a name (saṃjñā) is meant.
Example.
śravaṇena yuktā rātriḥ "a night joined with Śravaṇa" → used as a name → lup of aṇ → śravaṇā rātriḥ.
aśvatthena yukto muhūrtaḥ "a muhūrta (a period of the day) joined with Aśvattha" → lup → aśvatthaḥ.
Applies when / blocked by. Bhaṭṭoji says the rule is made for the case where a division of time is understood. A night is such a division, so SK 1205 could not give lup here. With no name meant there is no lup: śrāvaṇī, āśvatthī.
Rule. After a dvandva (a compound of "X and Y") of nakṣatra names, cha comes in the sense "a time joined with them". This holds whether or not a division of time is understood.
Example.
tiṣyaś ca punarvasū ca → tiṣyapunarvasū; tābhyāṃ yuktam ahaḥ "a day joined with Tiṣya and Punarvasu"
→ tiṣyapunarvasu + cha → ch → īy → tiṣyapunarvasu + īya
→ the final u takes guṇa by or guṇaḥ (SK 847): u → o → av → tiṣyapunarvasavīyam ahaḥ.
Also rādhānurādhīyā rātriḥ "a night joined with Rādhā and Anurādhā".
Applies when / blocked by. Why does the lup of SK 1205 not remove this cha? Bhaṭṭoji's commentators explain. SK 1205 stands between SK 1204 and this rule. A rule placed in between acts only on what comes before it. So its lup reaches the aṇ of SK 1204, not this cha.
vastraiḥ parivṛto rathaḥ → vāstro rathaḥTwo more instrumental senses follow. A sāman (a chant of the Sāmaveda) is named after the seer who "saw" it. A chariot is named after what covers it. On the running example, a chariot covered with cloths is vāstro rathaḥ. The group ends with kaumāra, a form set down as it stands.
Rule. The instrumental tena is still in force. After a third-case word, aṇ etc. come in the sense "seen by that", when what is seen is a sāman.
Example. Bhaṭṭoji gives the phrase vasiṣṭhena dṛṣṭaṃ sāma "a sāman seen by Vasiṣṭha", with no derived form after it.
A vārttika (asminn arthe 'ṇ ḍid vā vaktavyaḥ) says that in this sense aṇ is optionally ḍit, marked with ḍ. A ḍit affix deletes the ṭi of the base, its last vowel with what follows.
uśanasā dṛṣṭam "seen by Uśanas" → uśanas + aṇ
→ as ḍit: the ṭi as is deleted → uśan + a → vṛddhi → auśanam
→ not ḍit: → auśanasam.
Rule. After kali, ḍhak comes in the sense "seen by that", of a sāman.
Example.
kalinā dṛṣṭam "seen by Kali" → kali + ḍhak → ḍh → ey, k brings vṛddhi → kāli + eya
→ the i is deleted (SK 311) → kāleyaṃ sāma.
Watch out. Bhaṭṭoji's commentators call this statement a vārttika. The Siddhānta-kaumudī numbers it as a sūtra.
Rule. After vāmadeva, ḍyat and ḍya come in the sense "a sāman seen by him". Both are marked with ḍ.
Example.
vāmadevena dṛṣṭaṃ sāma → vāmadeva + ḍyat → the ḍ deletes the ṭi, the final a → vāmadevyam.
Watch out. Bhaṭṭoji quotes a verse of the Bhāṣya. Objection: yasyeti ca (SK 311) already deletes the final a, so why make the affixes ḍit? Answer: for the accent. An accent rule (yayatoś cātadarthe, A 6.2.156) acts on words ending in ya or yat after the negative nañ. Bhaṭṭoji's commentators explain the principle: a rule that names a bare affix does not cover the same affix with an extra marker. So ḍya and ḍyat are not "ya" and "yat", and avāmadevya does not take that accent.
Rule. After a third-case word, aṇ etc. come in the sense "covered with that", when what is covered is a chariot.
Example.
vastraiḥ parivṛto rathaḥ "a chariot covered with cloths" → vastra + aṇ
→ vṛddhi a → ā, final a deleted → vāstro rathaḥ.
Applies when / blocked by.
vastreṇa parivṛtaḥ kāyaḥ "a body covered with cloth" takes no affix.chātraiḥ parivṛto rathaḥ "a chariot surrounded by pupils" takes no affix.Rule. After pāṇḍukambala, the affix ini comes in the sense "a chariot covered with that". A pāṇḍukambala is a coloured blanket used as a royal coverlet.
Example.
pāṇḍukambalena parivṛtaḥ → pāṇḍukambala + ini → the final a is deleted → pāṇḍukambalin → nominative pāṇḍukambalī.
Watch out. The same form would already come from the possessive ini (A 5.2.115), called matvarthīya "in the sense of matup" (matup is the possessive affix -mat). Bhaṭṭoji says this rule is there to exclude aṇ.
Rule. After dvaipa and vaiyāghra, añ comes in the sense "a chariot covered with that".
Example.
dvīpin, a tiger (vyāghra) according to Bhaṭṭoji's commentators → its product, its hide, is dvaipam → dvaipena parivṛto rathaḥ → dvaipa + añ → dvaipo rathaḥ.
Likewise vaiyāghraḥ, a chariot covered with a tiger's hide.
Watch out. The base already has a vṛddhi vowel, so añ changes nothing visible. Bhaṭṭoji's commentators say it differs from aṇ in accent.
Rule. The form kaumāra is a nipātana. It is used when "having had no previous husband" is to be expressed of the woman. The instrumental tena no longer applies.
Example.
apūrvapatiṃ kumārīṃ patir upapannaḥ "a husband who has married a maiden with no previous husband" → kumārī + aṇ → kaumāraḥ patiḥ.apūrvapatiḥ kumārī patim upapannā "a maiden with no previous husband who has taken a husband" → kaumārī bhāryā, the feminine taking ṅīp.Watch out. Bhaṭṭoji's commentators note that only the woman must have had no previous partner. Whether the man had a previous wife does not matter.
udaśviti saṃskṛtaḥ → audaśvitkaḥ / audaśvitaḥNow the base is in the seventh case (locative), "in that". The senses are "set down in a vessel", "lying on", and "prepared in". The running example is udaśvit, buttermilk. Something prepared in it is audaśvitkaḥ or audaśvitaḥ. The k of the first form comes from SK 1221.
Rule. After a locative word naming an amatra, a vessel, aṇ etc. come in the sense "taken out [and set down] in it". Bhaṭṭoji's commentators say this locative tatra runs up to SK 1223, not including it.
Example.
śarāve uddhṛtaḥ "taken out into an earthen dish" → śarāva + aṇ → vṛddhi, final a deleted → śārāva odanaḥ "rice so served".
Watch out. Why the locative, when "taking out" is a separation? Bhaṭṭoji explains that uddharati here means setting down after taking out. The dish is where the rice is set down. Bhaṭṭoji's commentators add that the Vṛttikāra takes uddhṛta here as "left over after eating".
Rule. After sthaṇḍila "bare ground" in the locative, aṇ etc. come in the sense "one who lies there", when the lying is under a vrata, a vow. The vow must be understood from the expression as a whole.
Example.
sthaṇḍile śete "he lies on the bare ground" (by vow) → sthaṇḍila + aṇ → sthāṇḍilo bhikṣuḥ "a mendicant who sleeps on the bare ground".
Rule. After a locative word, aṇ comes in the sense "prepared in it", when the thing prepared is food. Saṃskṛta "prepared" means given a particular quality by cooking and the like.
Example.
bhrāṣṭre saṃskṛtāḥ "prepared in a frying-pan" → bhrāṣṭra + aṇ → bhrāṣṭrā yavāḥ "pan-parched barley".
aṣṭasu kapāleṣu saṃskṛtaḥ "prepared in eight potsherds" → the sense makes aṣṭan + kapāla a dvigu → the aṇ that caused the dvigu disappears by SK 1080 → no vṛddhi → aṣṭakapālaḥ puroḍāśaḥ "a sacrificial cake in eight potsherds".
Watch out. Bhaṭṭoji's commentators cite this form as aṣṭākapāla, with a long ā. The step by SK 1080 is theirs; Bhaṭṭoji gives the form only.
Rule. After śūla "spit" and ukhā "pot", yat comes in the sense "food prepared in it". It is an apavāda of aṇ.
Example.
śūle saṃskṛtam → śūla + yat → yat has no vṛddhi marker; the final a is deleted → śūlyaṃ māṃsam "spit-roasted meat".
ukhāyāṃ saṃskṛtam → ukhā + yat → ukhyam.
Rule. After dadhi "curd", ṭhak comes in the sense "food prepared in it".
Example.
dadhni saṃskṛtam "prepared in curd" → dadhi + ṭhak → ṭh → ik, vṛddhi a → ā → dādhi + ika
→ the i is deleted (SK 311) → dādhikam.
Watch out. Bhaṭṭoji's commentators explain the locative. The preparing is done with salt and the like; the curd is only the place where it happens.
Rule. After udaśvit "buttermilk", ṭhak comes optionally in the sense "prepared in it". In the other option the affix is aṇ.
Example.
udaśviti saṃskṛtaḥ →
audaśvitaḥ.audaśvit + ṭha… → the replacement of ṭh is settled by the next rule, SK 1221.Rule. The ṭh of an affix like ṭhak is replaced by k, not ik, after a stem ending in is, us, uk or t. So the affix is heard as ka. Here uk is a pratyāhāra, a short name for the vowels u, ṛ and ḷ. So a stem ending in any of these vowels counts.
Example.
udaśvit (it "swells with water") + ṭhak → the stem ends in t → ṭh → k → audaśvitkaḥ.
Applies when / blocked by. Only the is and us taught by name (pratipadokta), as affixes, count. Bhaṭṭoji's commentators name them as the is and us of the uṇādi list.
āśiṣā carati "he moves by means of a blessing" → āśis comes from the root śās; its is is not the affix is → āśiṣikaḥ. The printed text writes āsiṣā.uṣā carati "he moves at dawn" → auṣikaḥ.A vārttika (doṣa upasaṃkhyānam) adds dos "arm": dorbhyāṃ carati → dauṣkaḥ.
Rule. After kṣīra "milk", ḍhañ comes in the sense "prepared in it".
Example.
kṣīre saṃskṛtā yavāgūḥ "gruel prepared in milk" → kṣīra + ḍhañ → ḍh → ey, vṛddhi ī → ai → kṣaireya
→ feminine with ṅīp → kṣaireyī yavāgūḥ.
Applies when / blocked by. From SK 1217 only "prepared" is carried here, not "food". So the rule also covers gruel.
pauṣī paurṇamāsy asmin → pauṣo māsaḥNow the base is in the first case (nominative). A month is named after the full-moon day that falls in it. On the running example, the full-moon day pauṣī (from SK 1204) gives the month pauṣaḥ. Two rules give ṭhak instead for some bases.
Rule. After a first-case word naming a paurṇamāsī, a full-moon day, aṇ etc. come in the sense "in which that is [the full-moon]". The word iti in the sūtra restricts the rule to names.
Example.
pauṣī paurṇamāsy asmin "in which the full-moon is pauṣī" → pauṣī + aṇ
→ the base already has a vṛddhi vowel; the ī is deleted (SK 311) → pauṣo māsaḥ "the month Pauṣa".
Applies when / blocked by. Bhaṭṭoji's commentators give a counter-example. A fifteen-night period in which the pauṣī full-moon falls takes no affix, because it is not so named.
Rule. After āgrahāyaṇī and aśvattha, ṭhak comes in the sense of SK 1223.
Example.
agre hāyanam asyāḥ "she at whose front the year stands" → āgrahāyaṇī, the full-moon after which the year begins. It has aṇ in its own sense, because the prajñādi list is an ākṛtigaṇa (an open list), and then ṅīp. The ṇ comes by SK 857.
āgrahāyaṇī paurṇamāsy asmin → āgrahāyaṇī + ṭhak → āgrahāyaṇikaḥ māsaḥ.
aśvatthena yuktā paurṇamāsī → aśvatthaḥ → with ṭhak → āśvatthikaḥ.
Watch out. Why is the full-moon called aśvattha, with lup? SK 1205 does not reach full-moons. Bhaṭṭoji answers: the sūtra's own form aśvattha is a nipātana, so here there is lup even for the full-moon.
Rule. After phālgunī, śravaṇā, kārttikī and caitrī, ṭhak comes optionally in the sense of SK 1223. In the other option the affix is aṇ.
Example.
phālgunī paurṇamāsy asmin → with ṭhak → phālgunikaḥ; with aṇ → phālgunaḥ māsaḥ.
Likewise śrāvaṇikaḥ / śrāvaṇaḥ, kārtikikaḥ / kārtikaḥ, caitrikaḥ / caitraḥ.
Watch out. This sūtra is the jñāpaka used in SK 1205. Its forms show that the full-moon days keep their affix. Bhaṭṭoji's commentators close the group of "joined with" here.
indro devatāsya → aindraṃ haviḥ; with a stem rule, pitaro devatā asya → pitryamAn oblation or a mantra is named after its devatā, its deity. The base is a first-case word, and the affix is aṇ and the rest. Many deity names take special affixes: ghan, gha, cha, ṭyaṇ, yat, ḍhak, ṭhañ. On the running example, an oblation to Indra is aindraṃ haviḥ. The word for "whose deity is the Fathers" needs a stem rule, SK 1234: pitṛ + yat → pitryam.
Rule. After a first-case word, aṇ etc. come in the sense "that is its deity". A devatā is the particular recipient intended when a substance is given up in offering. It is also the one praised by a mantra.
Example.
indro devatāsya "Indra is its deity" → indra + aṇ → vṛddhi i → ai, final a deleted → aindraṃ haviḥ "an oblation to Indra".
For the mantra-deity: aindro mantraḥ "a mantra to Indra".
Also pāśupatam (from paśupati) and bārhaspatyam (from bṛhaspati).
Applies when / blocked by. Bhaṭṭoji's commentators explain "particular recipient". A priest who receives the remainder of an oblation, or a brāhmaṇa who receives a gift, is not a devatā.
Watch out. In the Vedic sentence āgneyo vai brāhmaṇo devatayā "the brāhmaṇa, by deity, belongs to Agni", no brāhmaṇa is offered to Agni. Bhaṭṭoji says the affix there is in the śaiṣika sense, one of the "remaining" senses taught later. It is ḍhak by the vārttika sarvatrāgnikalibhyām ("after agni and kali in every sense"), which you met in chapter 26.
Rule. Ka (a name of Prajāpati, the creator Brahmā) takes the substitute i in place of its final, together with the affix of "whose deity is that".
Example.
ko devatāsya "Ka is its deity" → ka + aṇ → with the affix, a → i → ki + a
→ now two rules could act. Yasyeti ca (SK 311) would delete the i. First-vowel vṛddhi would turn i into ai.
→ vṛddhi is the later rule, so it wins (paratvāt) → kai + a
→ ai becomes āy before a vowel → kāyaṃ haviḥ.
Applies when / blocked by. The same reasoning explains śrāyam "whose deity is Śrī", where no i-rule is involved: śrī + aṇ → vṛddhi first → śrai → śrāyam. Bhaṭṭoji's commentators note that the Bhāṣya solves the case of ka another way. Bhaṭṭoji's reason is needed anyway for śrāyam.
Rule. After śukra, ghan comes in the sense "whose deity is that".
Example.
śukro devatāsya → śukra + ghan → gh → iy, so the affix is heard as iya; no vṛddhi marker
→ the final a is deleted → śukriyam.
Rule. After aponaptṛ and apānnaptṛ, gha comes in the sense "whose deity is that".
Example.
aponaptṛ + gha → gha → iya → ṛ becomes r before the vowel → aponaptriyam. Likewise apānnaptriyam.
Watch out. The deities are really Aponapāt and Apānnapāt. The forms in -naptṛ are a nipātana that holds only together with this affix. That is why the ritual direction, which has no affix, keeps the base form: aponapāte apānnapāte 'nubrūhi "recite for Aponapāt, for Apānnapāt".
Rule. Cha also comes after the same two words, in the same sense.
Example.
aponaptṛ + cha → cha → īya → aponaptrīyam. Likewise apānnaptrīyam.
Watch out. Why two sūtras, and not one naming gha and cha together? With two bases and two affixes in one rule, they would pair in order (yathāsaṃkhya): the first base with gha only, the second with cha only. Splitting the rule (yoga-vibhāga) lets both affixes follow both bases.
Applies when / blocked by. A vārttika (śatarudrād ghaś ca) gives gha, and by its ca also cha, after śatarudra:
śataṃ rudrā devatā asya "whose deity is the hundred (countless) Rudras" → śatarudriyam, śatarudrīyam.
Śatarudra here is a dvigu, so SK 1080 would drop the affix. It does not. Bhaṭṭoji says gha and cha are expressly taught, and that teaching would be pointless if they were dropped.
Rule. After mahendra, gha and aṇ come, and by ca also cha, in the sense "whose deity is that".
Example.
mahendro devatāsya →
mahendriyaṃ haviḥmāhendrammahendrīyam.Rule. After soma, ṭyaṇ comes in the sense "whose deity is that". It is heard as ya and is ṇit.
Example.
somo devatāsya → soma + ṭyaṇ → vṛddhi o → au, final a deleted → saumyam.
Watch out. The ṭ marker is there for the feminine. A ṭit affix takes ṅīp (SK 470). Bhaṭṭoji's commentators add that the y of the affix is then dropped by halas taddhitasya. So saumī ṛk "a verse whose deity is Soma".
Rule. After vāyu, ṛtu, pitṛ and uṣas, yat comes in the sense "whose deity is that".
Example.
vāyur devatāsya → vāyu + yat → the final u takes guṇa → o → av before the y → vāyavyam.
Likewise ṛtu → ṛtavyam.
The forms for pitṛ and uṣas are given under SK 1234.
Rule. An aṅga (a base before an affix) ending in ṛ takes the substitute rīṅ in two cases. One is before a y that belongs neither to a kṛt affix (one added to a root) nor to a sārvadhātuka affix (a verbal affix of the present-tense type). The other is before the affix cvi, an affix taught in a later chapter.
Example.
pitaro devatā asya "the Fathers are its deity" → pitṛ + yat (SK 1233)
→ the y is a taddhita, neither kṛt nor sārvadhātuka → ṛ → rī → pitrī + ya
→ the ī is deleted by yasyeti ca (SK 311) → pitryam.
The other form of SK 1233: uṣā devatāsya → uṣas + yat → uṣasyam. Here rīṅ plays no part. Bhaṭṭoji's commentators note that the stem is bha before y, so its s stays.
Watch out. Why a long ī, if it is deleted at once? Bhaṭṭoji's commentators say the long vowel is needed for other forms, such as pitrīyati with the affix kyac, another affix taught in a later chapter.
Rule. After dyāvāpṛthivī, śunāsīra, marutvat, agnīṣoma, vāstoṣpati and gṛhamedha, cha comes, and by ca also yat, in the sense "whose deity is that".
Example.
dyāvāpṛthivī devatā asya →
dyāvāpṛthivīyamdyāvāpṛthivyam.Likewise śunāsīrīyam, śunāsīryam. The printed text has śūnāsīryam with a long ū.
Rule. After agni, ḍhak.
Example.
agnir devatāsya → agni + ḍhak → ḍh → ey, vṛddhi a → ā, the i deleted → āgneyam.
Watch out. Bhaṭṭoji's commentators note that by the vārttika sarvatrāgnikalibhyām this ḍhak holds in every prāgdīvyatīya sense, not only "whose deity is that".
Rule. This is an atideśa. After words for times, the affixes of "whose deity is that" are the same as those taught later for the sense bhava, "existing in". Only the affix prescribed there for a given base comes.
Example.
māso devatāsya "whose deity is the month" → the bhava rule kālāṭ ṭhañ (SK 1381) gives ṭhañ → māsikam.prāvṛḍ devatāsya "whose deity is the rainy season" → the bhava rule prāvṛṣa eṇyaḥ (SK 1388) gives eṇya → prāvṛṣeṇyam.Rule. After mahārāja and proṣṭhapada, ṭhañ comes in the sense "whose deity is that". Mahārāja here is Vaiśravaṇa, according to Bhaṭṭoji's commentators.
Example.
mahārājo devatāsya → mahārāja + ṭhañ → ik, vṛddhi → māhārājikam.
proṣṭhapado devatāsya → vṛddhi o → au → prauṣṭhapadikam.
agnāmarutau devatā asya → āgnimārutamA deity can be a pair, named by a dvandva. Bhaṭṭoji here reads three rules from the seventh book on the vṛddhi of such compounds. Bhaṭṭoji's commentators call this an incidental topic. Then two vārttikas return to the main line, and a sūtra of nipātanas gives the kinship words. On the running example, an oblation to Agni and the Maruts is āgnimārutam, with vṛddhi in both members.
Rule. In a dvandva of deity names, the first vowel of both members takes vṛddhi, before an affix marked with ñ, ṇ or k.
Example.
agniś ca marut ca → the compound is agnāmarutau, with ā at the join by A 6.3.26 (also worded devatādvandve ca), which replaces the final vowel of the first member with ā
→ agnāmarutau devatā asya → aṇ is intended
→ before a vṛddhi-bringing affix, id vṛddhau (A 6.3.28) gives i in place of that ā → agnimarut + a
→ vṛddhi of both members: the a of agni → ā, the a of marut → ā → āgnimārutam.
The steps are those of Bhaṭṭoji's commentators; Bhaṭṭoji gives the form.
Rule. Indra does not take this vṛddhi when it is the second member.
Example.
somendrau devate asya → somendra + aṇ → vṛddhi of soma only → saumendraḥ [caruḥ] "an offering to Soma and Indra".
Applies when / blocked by. As the first member, indra does take vṛddhi: indrāgnī → aindrāgnaḥ.
Rule. Varuṇa does not take this vṛddhi when it comes after a long vowel.
Example.
indrāvaruṇau devate asya → the compound has ā at the join → indrāvaruṇa + aṇ
→ varuṇa follows the long ā → no vṛddhi for it; indra → aindra → aindrāvaruṇam.
Applies when / blocked by. Why "after a long vowel"? In āgnivāruṇīm anaḍvāhīm ālabheta "one should offer a draught-cow sacred to Agni and Varuṇa", id vṛddhau has made the vowel before varuṇa a short i. So both members take vṛddhi.
Two vārttikas follow, returning to the main topic:
nāvayajñikaḥ kālaḥ "the time of the new-grain sacrifice"; pākayajñikaḥ. The affix ṭhañ is named by Bhaṭṭoji's commentators.pūrṇo māso 'syāṃ vartate "the full moon is present in it" → paurṇamāsī tithiḥ, the full-moon day. Some take paurṇamāsī as a bahuvrīhi with its own aṇ, and scholars discuss whether this vārttika is needed.Rule. The words pitṛvya, mātula, mātāmaha and pitāmaha are nipātanas. Vārttikas state how each is formed and in what sense.
Example.
pitur bhrātā "father's brother" → pitṛ + vyat → pitṛvyaḥ.mātur bhrātā "mother's brother" → mātṛ + ḍulac → the ḍ deletes the ṭi ṛ → mātulaḥ.mātuḥ pitā "mother's father" → mātāmahaḥ; pituḥ pitā → pitāmahaḥ.mātāmahī, pitāmahī.avisoḍham, avidūsam, avimarīsam. The affixes are read with s, so the s does not become ṣ.tilapiñjaḥ, tilapejaḥ "barren sesame".tilpiñjaḥ.Watch out. Bhaṭṭoji's commentators note that the Bhāṣya forms mātāmaha differently, with an ā substitute and an affix mahac. They also differ on how far the argument from the s reaches beyond avisoḍham. Scholars discuss both points.
kākānāṃ samūhaḥ → kākam; with a stem rule, garbhiṇīnāṃ samūhaḥ → gārbhiṇamA samūha is a collection. The base is a sixth-case (genitive) word, and the default affix is aṇ. Bhaṭṭoji's commentators say the sense runs as far as SK 1260. The derived words are neuter by usage. On the running example, a collection of crows is kākam. A collection of pregnant women is gārbhiṇam; that form needs two stem rules, bhasyāḍhe and SK 1245.
Rule. After a sixth-case word, aṇ etc. come in the sense "a collection of that".
Example.
kākānāṃ samūhaḥ "a collection of crows" → kāka + aṇ → vṛddhi (already ā), final a deleted → kākam.
bakānāṃ samūhaḥ "a collection of cranes" → vṛddhi a → ā → bākam.
Watch out. Bhaṭṭoji's commentators explain the choice of examples. Kāka and baka fall under none of the special rules that follow. They are animate, not gotra words, and have no affix of their own. So they show the plain aṇ.
Rule. After bhikṣā "alms" and the rest of its list, aṇ comes in the sense "a collection of that".
Example.
bhikṣāṇāṃ samūhaḥ → bhikṣā + aṇ → vṛddhi i → ai, final ā deleted → bhaikṣam.
garbhiṇīnāṃ samūhaḥ "a collection of pregnant women" → garbhiṇī + aṇ
→ bhasyāḍhe treats the bha stem as masculine (puṃvadbhāva): the feminine ṅīp is removed → garbhin + a
→ now nas taddhite (SK 679) would delete the ṭi in. SK 1245 stops it.
Applies when / blocked by. This aṇ overrides other affixes. Bhikṣā is inanimate and would take ṭhak by SK 1256. Garbhiṇī has an unaccented first vowel and would take añ by SK 1253. Bhaṭṭoji's commentators give these reasons.
Rule. A stem ending in in keeps its own form (prakṛtyā) before an aṇ that does not mean "offspring". So the ṭi-deletion of SK 679 does not occur.
Example.
garbhin + aṇ (from SK 1244) → this aṇ means "collection", not offspring → in stays → vṛddhi → gārbhiṇam.
yuvatīnāṃ samūhaḥ "a collection of young women" → the feminine is removed, leaving yuvan → yuvan + aṇ → yauvanam.
This rule does not keep the an of yuvan, since yuvan ends in an, not in. Yuvan keeps its an by a separate rule, an (A 6.4.167), which Bhaṭṭoji's commentators name. Bhaṭṭoji cites yauvanam here only beside the añ form yauvatam.
Applies when / blocked by. Take yuvatī instead as the feminine of yuvat, formed with śatṛ, the affix of the present participle (heard as -at). That word has an unaccented first vowel, so it takes añ (SK 1253), not aṇ. With añ this rule does not apply → yauvatam.
Rule. After a gotra word, and after ukṣan, uṣṭra, urabhra, rājan, rājanya, rājaputra, vatsa, manuṣya and aja, vuñ comes in the sense "a collection of that".
Example. The affix vuñ is heard as aka; that step is SK 1247, where the forms are worked.
Watch out. "Gotra" here is the ordinary gotra, meaning any offspring at all. It is not the technical gotra of SK 1089.
Rule. The nasalised affix-parts yu and vu are replaced by ana and aka, respectively.
Example.
glucukāyanīnāṃ samūhaḥ "a collection of female descendants of Glucuka" → glucukāyanī + vuñ (SK 1246)
→ vu → aka → vṛddhi u → au, the final ī deleted → glaucukāyanakam.
ukṣṇāṃ samūhaḥ "a collection of oxen" → ukṣan + vuñ → aka, the ṭi an deleted, vṛddhi → aukṣakam.
Applies when / blocked by. In rājanya + aka and manuṣya + aka the y would be dropped by SK 1082. A vārttika keeps it: prakṛtyā 'ke rājanyamanuṣyayuvānaḥ, "before aka, rājanya, manuṣya and yuvan keep their form" → rājanyakam, mānuṣyakam. Whether SK 1082 could ever reach these two words is a point scholars discuss.
A second vārttika (vṛddhāc ceti vaktavyam) adds the word vṛddha "old man" → vṛddhānāṃ samūhaḥ → vārdhakam.
Rule. After kedāra "field", yañ comes, and by ca also vuñ, in the sense "a collection of that".
Example.
kedārāṇāṃ samūhaḥ →
kaidāryamkaidārakam.Applies when / blocked by. A vārttika (gaṇikāyā yañiti vaktavyam) gives yañ only, not vuñ, after gaṇikā "courtesan" → gāṇikyam.
Rule. After kavacin "armoured man", ṭhañ comes in the sense "a collection of that". By ca, ṭhañ also comes after kedāra.
Example.
kavacināṃ samūhaḥ → kavacin + ṭhañ → ik → the ṭi in is deleted (SK 679) → vṛddhi → kāvacikam.
kedāra + ṭhañ → kaidārikam.
Rule. After brāhmaṇa, māṇava and vāḍava, yan comes in the sense "a collection of that".
Example.
brāhmaṇānāṃ samūhaḥ → brāhmaṇa + yan → no vṛddhi marker → brāhmaṇyam.
Likewise māṇavyam, vāḍavyam.
Applies when / blocked by. A vārttika (pṛṣṭhād upasaṃkhyānam) adds pṛṣṭha (a kind of stotra, a chant) → pṛṣṭhyam.
Rule. After grāma "village", jana "person" and bandhu "kinsman", tal comes in the sense "a collection of that". Tal is heard as tā, and its derivatives are feminine.
Example.
grāmāṇāṃ samūhaḥ → grāma + tal → grāmatā. Likewise janatā, bandhutā.
Applies when / blocked by. Three vārttikas follow.
gajatā, sahāyatā.ahan + īna → the ṭi an is deleted → ahīnaḥ, a sacrifice whose soma-pressings take a series of days. When no sacrifice is meant, ahan takes añ as a member of the khaṇḍikādi list (SK 1254). Then its ṭi stays, by the restriction ahnaṣ ṭakhor eva (SK 789) → āhnaḥ "a collection of days".Rule. Before an affix marked with s (a sit affix), what precedes is called pada.
Example.
parśūnāṃ samūhaḥ "a collection of ribs" → parśu + ṇas (the vārttika under SK 1251)
→ ṇas is sit, so parśu is a pada, not a bha stem
→ or guṇaḥ (SK 847) acts only on a bha stem, so the u takes no guṇa
→ vṛddhi a → ā, and u → v before the vowel → pārśvam.
saktūnāṃ samūhaḥ → sāktukam (ṭhak with ka by SK 1221)The collection sense continues. Words with an unaccented first vowel take añ. Inanimate things take ṭhak. A few lists take ya and other feminine affixes. On the running example, a collection of saktu "barley-meal" is sāktukam. The stem ends in u, one of the uk vowels, so the ṭh becomes k by SK 1221.
Rule. After a word whose first vowel is anudātta (unaccented), añ comes in the sense "a collection of that".
Example.
kapotānāṃ samūhaḥ "a collection of pigeons" → kapota has its accent on the middle vowel, so its first vowel is unaccented → kapota + añ → vṛddhi → kāpotam.
mayūrāṇāṃ samūhaḥ "peacocks" → māyūram.
Watch out. Bhaṭṭoji's commentators note why both examples name animals. For an inanimate base, the later ṭhak of SK 1256 wins.
Rule. After khaṇḍikā and the rest of its list, añ comes in the sense "a collection of that".
Example.
khaṇḍikānāṃ samūhaḥ → khaṇḍikā + añ → vṛddhi, final ā deleted → khāṇḍikam.
Watch out. Bhaṭṭoji's commentators say añ is taught here for an accent on the first vowel. One adds that it also blocks the ṭhak of SK 1256.
Rule. This is an atideśa. After names of caraṇas, the students of a Vedic school, the collective affixes are the same as those taught later for the sense dharma, "the custom of".
Example.
kaṭhānāṃ samūhaḥ "a collection of Kaṭhas" → the dharma rule gotracaraṇād vuñ gives vuñ → kāṭhakam.chandogānāṃ samūhaḥ "a collection of Sāman-singers" → the dharma rule chandogaukthika… gives ñya → chāndogyam.The two rules are named by Bhaṭṭoji's commentators.
Rule. After words for acitta, inanimate things, and after hastin "elephant" and dhenu "milch-cow", ṭhak comes in the sense "a collection of that".
Example.
saktūnāṃ samūhaḥ "a collection of barley-meal" → saktu + ṭhak
→ the stem ends in u, one of the uk vowels, so ṭh → k by SK 1221, not ik → vṛddhi a → ā → sāktukam.
dhenūnāṃ samūhaḥ → likewise → dhainukam.
hastināṃ samūhaḥ → hastin + ṭhak → ik, the ṭi in deleted → hāstikam.
Rule. After keśa "hair" and aśva "horse", yañ and cha come respectively, optionally, in the sense "a collection of that".
Example.
keśānāṃ samūhaḥ → with yañ → kaiśyam; otherwise ṭhak (SK 1256, hair being inanimate) → kaiśikam.
aśvānāṃ samūhaḥ → with cha → aśvīyam; otherwise aṇ → āśvam.
Rule. After pāśa "snare" and the rest of its list, ya comes in the sense "a collection of that". The derivatives are feminine by usage.
Example.
pāśānāṃ samūhaḥ → pāśa + ya → final a deleted, feminine ṭāp → pāśyā.
Likewise tṛṇyā (grasses), dhūmyā (smokes), vanyā (woods), vātyā (winds).
Rule. After khala "threshing-floor", go "cow" and ratha "chariot", ya comes in the sense "a collection of that".
Example.
khalānāṃ samūhaḥ → khalyā.
gavāṃ samūhaḥ → go + ya → o → av before the y → gavyā.
rathānāṃ samūhaḥ → rathyā.
Watch out. These three could have been listed in the pāśādi list. Bhaṭṭoji's commentators say they are taught apart so that they alone carry into SK 1260.
Rule. Ini, tra and kaṭyac come, in order, after khala, go and ratha, in the sense "a collection of that".
Example.
khala + ini → khalin → feminine ṅīp → khalinī.go + tra → feminine ṭāp → gotrā.ratha + kaṭyac → feminine ṭāp → rathakaṭyā.Applies when / blocked by. A vārttika (khalādibhya inir vaktavyaḥ) gives ini after a list headed by khala. The list is an ākṛtigaṇa → ḍākinī, kuṭumbinī.
śibīnāṃ viṣayo deśaḥ → śaibaḥThe collection sense has ended, but the genitive tasya goes on. A region is named after the people whose domain it is. Then come a verse-group named after its first metre, a battle named after its purpose or its warriors, and games and actions named in the feminine. On the running example, the region that is the domain of the Śibis is śaibaḥ.
Rule. After a sixth-case word, aṇ etc. come in the sense "the thoroughly frequented domain (viṣaya) of that", when that domain is a region.
Example.
śibīnāṃ viṣayo deśaḥ "the region that is the domain of the Śibis" → śibi + aṇ
→ vṛddhi i → ai, the final i deleted (SK 311) → śaibaḥ.
Applies when / blocked by. It must be a region. devadattasya viṣayo 'nuvākaḥ "the Vedic lesson that is Devadatta's field" takes no affix.
Rule. After rājanya and the rest of its list, vuñ comes in the sense "the region that is the domain of that".
Example.
rājanyānāṃ viṣayo deśaḥ → rājanya + vuñ → aka (SK 1247) → rājanyakaḥ.
Rule. After the bhaurikyādi list, vidhal comes; after the aiṣukāryādi list, bhaktal. The sense is "the region that is the domain of that". The derivatives are neuter by usage.
Example.
bhaurikīṇāṃ viṣayo deśaḥ → bhaurikividham. Likewise bhaulikividham.
aiṣukārīṇāṃ viṣayo deśaḥ → aiṣukāribhaktam. Likewise sārasāyanabhaktam.
Rule. After the name of a metre in the first case, aṇ comes in the sense "that is the beginning of this", when "this" is a pragātha, a group of verses.
Example.
paṅktir ādir asya "of which the paṅkti metre is the beginning" → paṅkti + aṇ → vṛddhi, final i deleted → pāṅktaḥ pragāthaḥ.
Applies when / blocked by. A vārttika (svārtha upasaṃkhyānam) adds aṇ in the sense of the base itself (svārthe), adding no new meaning → triṣṭub eva "just the triṣṭubh" → traiṣṭubham.
Rule. The words so 'sya "that … of this" are carried forward. After first-case words for a purpose or for warriors, aṇ etc. come in the sense "that is the purpose of this" or "they are the warriors of this", when "this" is a battle.
Example.
subhadrā prayojanam asya saṃgrāmasya "Subhadrā is the purpose of this battle" → subhadrā + aṇ → saubhadraḥ.
bharatā yoddhāro 'sya saṃgrāmasya "the Bharatas are the warriors of this battle" → bhārataḥ.
Rule. After a first-case word, ṇa comes in the sense "that is the weapon in this", when "this" is a game.
Example.
daṇḍaḥ praharaṇam asyāṃ krīḍāyām "a game in which the stick is the weapon" → daṇḍa + ṇa → vṛddhi → dāṇḍa → feminine ṭāp → dāṇḍā.
Likewise muṣṭi "fist" → mauṣṭā.
Watch out. Why ṇa, not aṇ? Bhaṭṭoji's commentators explain: aṇ would bring ṅīp in the feminine (SK 470), giving -ī. With ṇa the feminine takes ṭāp: dāṇḍā.
Rule. After a first-case word ending in ghañ and naming an action, ña comes in the sense "that action is in this". The derivative is feminine. Ghañ is a kṛt affix that makes action nouns such as pāta "falling".
Example. The forms are worked under SK 1268.
Applies when / blocked by. The sūtra names a kṛt affix. By the reading of such names, a ghañ-word preceded by a gati (a verbal prefix) or a kāraka word (a participant in the action) is also covered. So a compound like śyenapāta "the swoop of hawks" can take ña.
Rule. The augment (āgama) mum is added to śyena "hawk" and tila "sesame", when pāta follows as the second member and ña follows pāta.
Example.
śyenapāto 'syāṃ vartate "in which the swoop of hawks takes place" → śyenapāta + ña (SK 1267)
→ mum after śyena → śyenaṃpāta + a
→ vṛddhi on the first vowel of the compound, e → ai → feminine ṭāp → śyainaṃpātā mṛgayā "the hunt".
tilapāto 'syāṃ vartate "in which the dropping of sesame takes place" → tailaṃpātā svadhā "the ancestral rite".
Applies when / blocked by. For other words there is no mum: daṇḍapāto 'syāṃ tithau vartate "the day on which staffs fall" → dāṇḍapātā tithiḥ.
Watch out. A svadhā rite is not a game. Bhaṭṭoji's commentators take this to show that "game" from SK 1266 is not carried into SK 1267.
vyākaraṇam adhīte veda vā → vaiyākaraṇaḥA student or a knower is named after his subject. The base is a second-case (accusative) word. On the running example, one who studies or knows vyākaraṇa, grammar, is vaiyākaraṇaḥ. The rules that follow give ṭhak, vun and ini after particular subjects.
Rule. After a second-case word, aṇ etc. come in two senses: "he studies that" and "he knows that". The word tad is repeated so that the two senses stay separate.
Example.
vyākaraṇam adhīte veda vā "he studies or knows grammar" → vyākaraṇa + aṇ
→ the first vowel stands after the y of the prefix vy- → by na yvābhyām (A 7.3.3), named by Bhaṭṭoji's commentators, that vowel takes no vṛddhi; instead the augment ai comes before the y → vaiyākaraṇa
→ final a deleted → vaiyākaraṇaḥ.
Watch out. Bhaṭṭoji's commentators distinguish the two senses. To study is to receive the text from the teacher's mouth. To know is to understand its meaning.
Rule. Ṭhak comes in the senses "studies" and "knows" after three kinds of word: names of particular kratus (sacrifices), the ukthādi list, and words ending in sūtra.
Example.
agniṣṭoma + ṭhak → āgniṣṭomikaḥ; vājapeya → vājapeyikaḥ. Only names of particular kratus count, not the word kratu itself. In its primary meaning such a name gives the "knower" sense. When it stands for the text that expounds the sacrifice, it gives the "student" sense.aukthikaḥ "one who studies or knows that text".nyāya → naiyāyikaḥ (with the ai augment), vṛtti → vārttikaḥ, lokāyata → laukāyatikaḥ.Applies when / blocked by. A string of vārttikas follows.
sāṃgrahasūtrikaḥ; but kalpasūtra → kālpasūtraḥ with aṇ.vāyasavidyikaḥ, āśvalakṣaṇikaḥ, pārāśarakalpikaḥ, and gaulakṣaṇikaḥ (printed gaulakṣaṇakaḥ).āṅgavidyaḥ, kṣātravidyaḥ, dhārmavidyaḥ. And trividhā vidyā "knowledge of three kinds" is trividyā → traividyaḥ.yāvakrītikaḥ.vāsavadattikaḥ.aitihāsikaḥ, paurāṇikaḥ.sarvavedān adhīte "he studies all the Vedas" → sarvavedaḥ; sarvatantraḥ; savārttikaḥ. With the dvigu luk of SK 1080 → dvitantraḥ.pūrvapadikaḥ, uttarapadikaḥ. Ṣikan after pathin preceded by śata or ṣaṣṭi → śatapathikaḥ, feminine śatapathikī; ṣaṣṭipathikaḥ, ṣaṣṭipathikī. The ṣ brings ṅīṣ in the feminine.Watch out. Bhaṭṭoji's commentators analyse savārttikaḥ in two different ways, and scholars discuss which is right.
Rule. After krama and the rest of its list, vun comes in the senses "studies" and "knows". The list is krama, pada, śikṣā, mīmāṃsā.
Example.
kramam adhīte veda vā → krama + vun → vu → aka (SK 1247) → no vṛddhi marker → kramakaḥ.
Rule. After anubrāhmaṇa, ini comes in the senses "studies" and "knows". An anubrāhmaṇa is a text resembling a brāhmaṇa.
Example.
anubrāhmaṇam adhīte → anubrāhmaṇa + ini → final a deleted → anubrāhmaṇin → anubrāhmaṇī.
Watch out. The possessive ini would give this form anyway. Bhaṭṭoji says the rule is there to block aṇ, as at SK 1212.
Rule. After vasanta and the rest of its list, ṭhak comes in the senses "studies" and "knows". Here vasanta "spring" names a text describing the spring.
Example.
vasanta + ṭhak → vāsantikaḥ.
atharvāṇam adhīte "he studies the Atharvan" → atharvan + ṭhak → ātharvaṇikaḥ.
Watch out. Why is the ṭi an not dropped in ātharvaṇika by SK 679? Because the form is a nipātana in the sūtra dāṇḍināyana… (SK 1145), which you met in chapter 27.
paṇ → paṇa → paṇī → pāṇinaḥ → pāṇiniḥ → pāṇinīyam → pāṇinīyaḥSome student-words show no student-affix at all. SK 1274 drops it after an affix meaning "proclaimed". Bhaṭṭoji uses this rule to derive a famous name. The running example is the whole chain from the root paṇ "to praise" to pāṇinīya, "a student of Pāṇini's grammar". It runs through SK 1274–1276. Bhaṭṭoji's commentators close the long section of single senses at SK 1278.
Rule. The student- or knower-affix disappears by luk when it follows an affix in the sense prokta, "proclaimed (taught) by".
Example. Bhaṭṭoji begins the chain here.
paṇanam "praising" → paṇ + ka (a vārttika allows ka in the sense of ghañ) → paṇaḥ "praise"
→ so 'syāstīti "he has that" → paṇa + ini (the possessive) → paṇin, nominative paṇī
→ tasya gotrāpatyam "his gotra descendant" → paṇin + aṇ.
This aṇ means offspring, so SK 1245 does not keep the in. The ṭi-deletion of SK 679 would be due. SK 1275 stops it.
Watch out. Why ka and not ghañ? Bhaṭṭoji's commentators explain that ghañ would lengthen the a, giving pāṇa.
Rule. Gāthin, vidathin, keśin, gaṇin and paṇin keep their own form before aṇ, even an aṇ meaning offspring.
Example.
paṇin + aṇ (from SK 1274) → in stays → vṛddhi → pāṇinaḥ, the gotra descendant
→ after the gotra word, the yuvan affix iñ (SK 1095) → pāṇina + iñ → final a deleted → pāṇiniḥ.
Rule. A yuvan aṇ or iñ disappears by luk after a word ending in a gotra affix of four kinds:
Example. In each pair the father (gotra) and the son (yuvan) have the same form.
kauravyaḥ pitā, kauravyaḥ putraḥ.śvāphalkaḥ pitā, śvāphalkaḥ putraḥ.vāsiṣṭhaḥ pitā, vāsiṣṭhaḥ putraḥ.taikāyaniḥ pitā, taikāyaniḥ putraḥ.Applies when / blocked by.
kauhaṅaḥ is the father, and with iñ the son is kauhaṅiḥ.vāmarathyasya chātrāḥ "the pupils of Vāmarathya" → vāmarathāḥ. The affix for pupils is not a yuvan affix, so this rule does not remove it.Example (continued: the running chain).
Why is the iñ of pāṇini not dropped? The aṇ of pāṇina might look like a seer's affix. But "ārṣa" takes in only the seer-aṇ taught by name (pratipadokta). The aṇ of pāṇina is the general aṇ → the iñ stays → pāṇiniḥ.
→ pāṇininā proktam "what was proclaimed by Pāṇini" → pāṇini is vṛddha (its first vowel is a vṛddhi vowel) → cha by vṛddhāc chaḥ (SK 1337) → pāṇini + īya → pāṇinīyam.
→ The aṇ of iñaś ca (SK 1333) does not come instead. That rule, as it will say, needs an iñ in the gotra sense; the iñ of pāṇini is yuvan.
→ pāṇinīyam adhīte "he studies the Pāṇinīya" → aṇ → luk by SK 1274 → pāṇinīyaḥ.
Watch out. Does the luk matter, if aṇ would bring no visible vṛddhi? Yes: it changes the accent and the feminine. Bhaṭṭoji's commentators explain: with the luk, the feminine takes ṭāp → pāṇinīyā, not -ī. Bhaṭṭoji's commentators question his reason for keeping the iñ of pāṇini, though not the form itself; scholars discuss this.
Rule. The student- or knower-affix also disappears by luk after a word that names a sūtra-text and has k as its upadhā (penultimate sound). Here the base need not be in the sense "proclaimed".
Example.
aṣṭāv adhyāyāḥ parimāṇam asya "that which has eight chapters as its measure" → aṣṭaka, Pāṇini's sūtra-text
→ tad adhīyate vidanti vā "they study or know it" → aṣṭaka + aṇ → k is the penultimate → luk → aṣṭakāḥ.
Watch out. Bhaṭṭoji's commentators comment on a further restriction of this rule, with counter-examples, that the printed text does not contain. Scholars discuss what the fuller text said.
Rule. This is a niyama, a restricting rule. Names of chandas (Vedic hymn) and brāhmaṇa texts that end in a "proclaimed" affix are tadviṣaya, "bound to that". They are never used without the student- or knower-affix.
Example.
kaṭhena proktam adhīyate "they study what Kaṭha proclaimed"
→ Kaṭha is a pupil of Vaiśampāyana, so the "proclaimed" affix is ṇini → it undergoes luk by kaṭhacarakāl luk (SK 1487) → kaṭha
→ the student-affix aṇ → luk by SK 1274 → kaṭhāḥ "the Kaṭhas".
udumbarāḥ santy asmin deśe → audumbaraḥNow four senses are set up for place-names. They are "that exists in this", "founded by that", "the dwelling of that" and "not far from that". The ca of SK 1282 joins them, so every affix taught after it serves all four. Such an affix is called cāturarthika, "of the four senses". On the running example, a country with udumbara fig-trees is audumbaraḥ.
Rule. After a first-case word, the prescribed affix comes in the sense "that exists in this", when the derived word is the name of a country. Bhaṭṭoji's commentators call this an apavāda of matup, the possessive affix -mat.
Example.
udumbarāḥ santy asmin deśe "in this country there are udumbara trees" → udumbara + aṇ
→ vṛddhi u → au, final a deleted → audumbaraḥ.
Watch out. Bhaṭṭoji's commentators say the words "when it is the name of a country" carry into the next three sūtras.
Rule. After a third-case word, aṇ etc. come in the sense "brought into being (founded) by that", the derivative being a place-name.
Example.
kuśāmbena nirvṛttā "founded by Kuśāmba" → kuśāmba + aṇ → vṛddhi → feminine ṅīp → kauśāmbī nagarī "the city Kauśāmbī".
Rule. After a sixth-case word, aṇ etc. come in the sense "the dwelling of that", the derivative being a place-name.
Example.
śibīnāṃ nivāso deśaḥ "the region where the Śibis dwell" → śibi + aṇ → śaibaḥ.
Watch out. The same form śaibaḥ came at SK 1261, for the "domain" sense. Bhaṭṭoji's commentators report the difference: a domain is owned, while a dwelling is mere residence.
Rule. Aṇ etc. also come in the sense "not far from that", the derivative being a place-name.
Example.
vidiśāyā adūrabhavaṃ nagaram "the town not far from the Vidiśā" → vidiśā + aṇ → vṛddhi i → ai, final ā deleted → vaidiśam.
Watch out. The ca "and" brings the three earlier senses together with this one. So every affix taught from here to SK 1311 is cāturarthika. It may come in whichever of the four senses fits.
kakṣatur asmin asti → kākṣatavamThe first cāturarthika affixes depend on the shape of the base. A u-final stem takes añ, and so do some many-voweled words and some well-names. Other lists take aṇ back. Then one long sūtra gives seventeen affixes after seventeen lists. On the running example, a place where kakṣatu grows is kākṣatavam.
Rule. After a stem ending in u, añ comes in the four senses. It is an apavāda of aṇ. Bhaṭṭoji's commentators say this añ runs as far as SK 1289.
Example.
kakṣatur asmin asti → kakṣatu + añ
→ vṛddhi a → ā; the final u takes guṇa by or guṇaḥ (SK 847): u → o → av → kākṣatavam.
Applies when / blocked by. For a river, matup (SK 1304) wins, because it is the later rule → ikṣavaḥ santy asyām → ikṣumatī.
Rule. After a word ending in matup, whose base (aṅga) has many vowels (bahvac), añ comes in the four senses, not aṇ.
Example.
sidhrakāvān asmin asti → sidhraka has three vowels → sidhrakāvat + añ → vṛddhi → saidhrakāvatam.
Applies when / blocked by. Ahi "snake" has only two vowels → ahimat + aṇ → āhimatam.
Watch out. Why does the sūtra say aṅga? Without it, "many vowels" would describe the whole matup-word. Ahimat itself has three vowels, and would wrongly take añ.
Rule. After a many-voweled word, añ comes in the four senses when a kūpa, a well, is named.
Example.
dīrghavaratreṇa nirvṛttaḥ "made by Dīrghavaratra" → dīrghavaratra + añ → vṛddhi ī → ai → dairghavaratraḥ kūpaḥ.
Rule. For wells on the northern bank of the Vipāś river, añ comes. This rule is for words that do not have many vowels.
Example.
dattena nirvṛttaḥ "made by Datta" → datta has two vowels, so SK 1285 cannot apply → north of the Vipāś → datta + añ → dāttaḥ kūpaḥ.
Applies when / blocked by. For wells on the southern side the affix is aṇ only.
Rule. After saṅkala and the rest of its list, añ comes in the four senses. "For wells" has ceased here.
Example.
saṅkalena nirvṛttam "made by Saṅkala" → sāṅkalam. Likewise puṣkala → pauṣkalam.
Rule. Añ comes when a place of feminine gender in Sauvīra, in Sālva, or in the East is named.
Example.
dattāmitreṇa nirvṛttā "founded by Dattāmitra" → dattāmitra + añ → vṛddhi → feminine ṅīp → dāttāmitrī nagarī.vaidhūmāgnī.mākandī.Rule. After suvāstu and the rest of its list, aṇ comes in the four senses. It is an apavāda of the añ of SK 1283.
Example.
suvāstor adūrabhavam "what is near the Suvāstu" → suvāstu + aṇ → vṛddhi u → au, and guṇa of the final u → av → sauvāstavam.
Likewise varṇu → vārṇavam. The printed text reads the base as varmu.
Watch out. Why name aṇ, when the default would come back anyway? To block matup for a river, which would win as the later rule (SK 1304). So a river named after the Suvāstu is sauvāstavī.
Rule. After roṇī and words ending in it, aṇ comes in the four senses. It overrides the well-añ of SK 1285.
Example.
roṇī + aṇ → vṛddhi o → au, final ī deleted → rauṇaḥ.
ājakaroṇī + aṇ → ājakaroṇaḥ.
Rule. After a stem with k as its upadhā (penultimate sound), aṇ comes in the four senses. It is an apavāda of añ.
Example.
karṇacchidrakeṇa nirvṛttaḥ "made by Karṇacchidraka" → kārṇacchidrakaḥ kūpaḥ.
kṛkavāku + aṇ → vṛddhi ṛ → ār, guṇa of u → av → kārkavākavam.
triśaṅku + aṇ → traiśaṅkavam.
Rule. Seventeen affixes come after seventeen lists, in order, in the four senses. Bhaṭṭoji's commentators say the word ādi "and the rest" goes with each list, so this is really seventeen rules.
Example. One form for each pair:
1. arīhaṇādi + vuñ: arīhaṇena nirvṛttam "founded by Arīhaṇa" → ārīhaṇakam.
2. kṛśāśvādi + chaṇ: kārśāśvīyam.
3. ṛśyādi + ka: ṛśyakam.
4. kumudādi + ṭhac: kumudikam.
5. kāśādi + ila: kāśilaḥ.
6. tṛṇādi + sa: tṛṇasam.
7. prekṣādi + ini: prekṣī.
8. aśmādi + ra: aśmaraḥ.
9. sakhyādi + ḍhañ: sākheyam.
10. saṅkāśādi + ṇya: sāṅkāśyam.
11. balādi + ya: balyam.
12. pakṣādi + phak: pākṣāyaṇaḥ. A gaṇa-sūtra inside this list, pathaḥ pantha ca, gives phak after pathin with pantha as its substitute → pānthāyanaḥ.
13. karṇādi + phiñ: kārṇāyaniḥ.
14. sutaṅgamādi + iñ: sautaṅgamiḥ.
15. pragadyādi + ñya: the stem is pragadin → prāgadyaḥ.
16. varāhādi + kak: vārāhakaḥ.
17. kumudādi + ṭhak: kaumudikaḥ.
Watch out.
vārīhakaḥ and kaumudīkaḥ for the last two. Other printings have vārāhakaḥ and kaumudikaḥ, the forms the affixes give.pañcālānāṃ nivāso janapadaḥ → pañcālāḥ (lup, plural as of the base)A janapada is a country named after its people. For it the cāturarthika affix disappears by lup. SK 1294 then says that gender and number follow the base. This explains why the country pañcālāḥ is plural. Pāṇini next says that this rule, and several definitions of the older teachers, need not be taught at all. Usage settles them. The group ends with more lup rules and the six forms of śarkarā.
Rule. When a janapada is named, the cāturarthika affix disappears by lup.
Example.
pañcālānāṃ nivāso janapadaḥ "the country where the Pañcālas dwell" → pañcāla + aṇ (SK 1281) → lup of aṇ → no vṛddhi → pañcāla. Its number is settled by SK 1294.
Rule. When there is lup of an affix, the gender (vyakti) and number (vacana) are those of the base (yukta), not those of the thing meant. This is called yuktavadbhāva.
Example.
pañcāla (from SK 1293) → the base pañcālānām is masculine plural → the country-name is masculine plural too, although janapada is singular → pañcālāḥ.
Likewise kuravaḥ, aṅgāḥ, vaṅgāḥ, kaliṅgāḥ.
Rule. The rule of SK 1294 need not be taught (aśiṣya), because names (saṃjñā) are authoritative.
Example. pañcālāḥ is simply the name of the country, used in the plural. Its plural is learned from usage, with no rule of grammar. Bhaṭṭoji's commentators compare dārāḥ "wife", which is masculine plural by usage.
Watch out. Here Pāṇini rejects a rule that stands in his own text. Bhaṭṭoji's commentators disagree on why it is there: one calls it a rule of the older teachers, the other a rule Pāṇini framed in deference to them. Scholars discuss this.
Rule. The lup itself need not be taught either, because the connection (yoga) is not perceived. By "the lup" are meant two rules, according to Bhaṭṭoji's commentators: SK 1293 and SK 1301. Bhaṭṭoji explains yoga as the meaning of the word's parts.
Example. No one hears pañcālāḥ as "the country connected with the Pañcāla kings". It is understood straight away as the name of a country. So no affix ever arises, and there is nothing for lup to remove.
Rule. If the connection were authoritative, the word would not be used once the connection was gone.
Example. A country called pañcāla is still so called when a king of another line rules it. Bhaṭṭoji's commentators draw the conclusion: pañcāla is a purely conventional name.
Rule. The older teachers' statement "the meaning of the affix is principal" also need not be taught, because meaning is known from another authority: usage.
Example. Bhaṭṭoji gives no word here. His commentators state the rejected definition in full: of stem and affix, the affix meaning is the thing qualified, and the stem meaning is its qualifier. Pāṇini's point is that speakers already understand a derived word this way, with no rule to tell them.
Rule. The same holds for time and for the upasarjana (the subordinate member). The older teachers' definitions of these need not be taught, because they are known in common usage.
Example. Two such definitions, quoted by Bhaṭṭoji:
Rule. The qualifiers of a lup-sense also take the gender and number of the base, except a qualifier that is a jāti word (a word for a class).
Example.
pañcālā ramaṇīyāḥ "Pañcāla is pleasant" → ramaṇīya qualifies the country, and takes the base's plural.
godau ramaṇīyau → the qualifier takes the dual of godau.
Applies when / blocked by. A jāti word keeps its own number: pañcālā janapadaḥ "Pañcāla is a country", godau grāmaḥ "Godau is a village".
Three vārttikas follow.
harītakyāḥ phalāni "the fruits of the harītakī" → harītakyaḥ, feminine like the tree but plural like the fruits.khalatikasya parvatasyādūrabhavāni "forests near Mount Khalatika" → khalatikaṃ vanāni, singular like the mountain but neuter.cañcā abhirūpaḥ "a handsome Cañcā", said of a man.Watch out. In the Aṣṭādhyāyī this sūtra (A 1.2.52) stands right after SK 1294. The Siddhānta-kaumudī reads it only here, after A 1.2.57.
Rule. After varaṇā and the rest of its list, the cāturarthika affix also disappears by lup. This rule is for places that are not janapadas.
Example.
varaṇānām adūrabhavaṃ nagaram "the town not far from the Varaṇās" → lup of aṇ → gender and number of the base (SK 1294) → varaṇāḥ, feminine plural.
Rule. After śarkarā "gravel", the lup of the cāturarthika affix is optional.
Example. The forms are given under SK 1303.
Rule. After śarkarā, ṭhak and cha also come in the four senses.
Example. Śarkarā is also read in the kumudādi and varāhādi lists of SK 1292, so ṭhac and kak come too. The vā of SK 1302 brings back the default aṇ, with optional lup. Six forms result:
śarkarā → lup of aṇśārkaram → aṇśārkarikam → ṭhakśarkarīyam → chaśarkarikam → ṭhacśārkarakam → kak.ikṣavaḥ santy asyām → ikṣumatī; naḍa → naḍvān / naḍvalaḥ / naḍakīyamThe chapter ends with matup and related affixes. A river is named with matup: the river with sugarcane is ikṣumatī. The reed naḍa shows three more affixes: naḍvān with ḍmatup, naḍvalaḥ with ḍvalac, and naḍakīyam with cha and the augment kuk.
Rule. When a river is named, matup comes as a cāturarthika affix. It is heard as mat; u and p are markers.
Example.
ikṣavaḥ santy asyām "in which there are sugarcanes" → ikṣu + matup → ikṣumat → feminine ṅīp → ikṣumatī.
Applies when / blocked by. It wins over the añ of SK 1283, as the later rule.
Rule. After madhu and the rest of its list, matup also comes in the four senses. This rule is for places that are not rivers.
Example.
madhu + matup → madhumat → nominative madhumān.
Rule. After kumuda, naḍa and vetasa, ḍmatup comes. The ḍ deletes the ṭi of the base.
Example.
kumuda + ḍmatup → the final a goes → kumud + mat. The m of the affix then becomes v, by rules taught later:
kumudvān, naḍvānvetasvān.Applies when / blocked by. A vārttika (mahiṣāc ceti vaktavyam) adds mahiṣa "buffalo" → mahiṣmān, the name of a country.
Watch out. Which rule gives the v of naḍvān is a point scholars discuss.
Rule. After naḍa "reed" and śāda (mud, or grass), ḍvalac comes in the four senses.
Example.
naḍa + ḍvalac → the ḍ deletes the ṭi, the final a → naḍvalaḥ.
śāda → śādvalaḥ.
Rule. After śikhā, valac comes in the four senses. It blocks aṇ for a place so named.
Example.
śikhā + valac → no ḍ, so the ā stays → śikhāvalam.
Rule. After utkara and the rest of its list, cha comes in the four senses.
Example.
utkara + cha → īya → the final a is deleted → utkarīyaḥ.
Rule. After naḍa and the rest of its list, cha comes, and the base takes the augment kuk (a k added at its end).
Example.
naḍa + kuk + cha → naḍak + īya → naḍakīyam.
Applies when / blocked by. Two gaṇa-sūtras inside the list:
kruñcakīyaḥ.takṣakīyaḥ.Rule. After bilvaka and the rest (words of the naḍādi list with kuk already added), cha disappears by luk when a further taddhita follows.
Example.
bilvā yasyāṃ santi "the place where there are bilva trees" → bilva + kuk + cha (SK 1310) → bilvakīyā
→ tasyāṃ bhavāḥ "those who live there" → aṇ in the sense bhava → before it, the whole īya disappears by luk → bilvak + a → vṛddhi → bailvakāḥ.
Likewise vetrakīyāḥ → vaitrakāḥ.
Watch out.
| SK | Sūtra | What it does | Example |
|---|---|---|---|
| 1202 | tena raktaṃ rāgāt | Aṇ etc. after a dye-word: "dyed with that". | kāṣāyam |
| 1203 | lākṣārocanāṭ ṭhak | Ṭhak after lākṣā, rocanā; vārttikas for śakala, kardama, nīlī, pīta, haridrā, mahārajana. | lākṣikaḥ, nīlam, pītakam, hāridram |
| 1204 | nakṣatreṇa yuktaḥ kālaḥ | Aṇ etc.: "a time joined with a nakṣatra". | pauṣam ahaḥ, pauṣī rātriḥ |
| 1205 | lub aviśeṣe | Lup of that affix when no part of the day is meant. | adya puṣyaḥ |
| 1206 | saṃjñāyāṃ śravaṇāśvatthābhyām | Lup after śravaṇa, aśvattha in a name. | śravaṇā rātriḥ |
| 1207 | dvandvāc chaḥ | Cha after a nakṣatra dvandva. | tiṣyapunarvasavīyam |
| 1208 | dṛṣṭaṃ sāma | Aṇ etc.: "a sāman seen by"; optionally ḍit. | auśanam / auśanasam |
| 1209 | kaler ḍhak | Ḍhak after kali. | kāleyam |
| 1210 | vāmadevāḍ ḍyaḍḍyau | Ḍyat, ḍya after vāmadeva. | vāmadevyam |
| 1211 | parivṛto rathaḥ | Aṇ etc.: "a chariot covered with". | vāstro rathaḥ |
| 1212 | pāṇḍukambalād iniḥ | Ini after pāṇḍukambala. | pāṇḍukambalī |
| 1213 | dvaipavaiyāghrād añ | Añ after dvaipa, vaiyāghra. | dvaipo rathaḥ |
| 1214 | kaumārāpūrvavacane | Nipātana kaumāra: no previous husband. | kaumāraḥ patiḥ, kaumārī |
| 1215 | tatroddhṛtam amatrebhyaḥ | Aṇ etc. after a vessel: "set down in it". | śārāva odanaḥ |
| 1216 | sthaṇḍilāc chayitari vrate | Aṇ etc.: "who lies on the ground by vow". | sthāṇḍilo bhikṣuḥ |
| 1217 | saṃskṛtaṃ bhakṣāḥ | Aṇ: "food prepared in". | bhrāṣṭrā yavāḥ |
| 1218 | śūlokhād yat | Yat after śūla, ukhā. | śūlyam, ukhyam |
| 1219 | dadhnaṣ ṭhak | Ṭhak after dadhi. | dādhikam |
| 1220 | udaśvito 'nyatarasyām | Optional ṭhak after udaśvit. | audaśvitaḥ |
| 1221 | isusuktāntāt kaḥ | Ka for ṭha after is, us, uk, t. | audaśvitkaḥ, dauṣkaḥ |
| 1222 | kṣīrāḍ ḍhañ | Ḍhañ after kṣīra, even for gruel. | kṣaireyī |
| 1223 | sāsmin paurṇamāsīti | Aṇ etc.: the month named after its full-moon. | pauṣo māsaḥ |
| 1224 | āgrahāyaṇyaśvatthāṭ ṭhak | Ṭhak after āgrahāyaṇī, aśvattha. | āgrahāyaṇikaḥ |
| 1225 | vibhāṣā phālgunīśravaṇākārttikīcaitrībhyaḥ | Optional ṭhak after these four. | phālgunikaḥ / phālgunaḥ |
| 1226 | sāsya devatā | Aṇ etc.: "whose deity is that". | aindraṃ haviḥ |
| 1227 | kasyet | Ka → ki; vṛddhi beats SK 311. | kāyam, śrāyam |
| 1228 | śukrād ghan | Ghan after śukra. | śukriyam |
| 1229 | aponaptrapānnaptṛbhyāṃ ghaḥ | Gha after aponaptṛ, apānnaptṛ. | aponaptriyam |
| 1230 | cha ca | Cha after the same; vārttika for śatarudra. | aponaptrīyam, śatarudriyam |
| 1231 | mahendrād ghāṇau ca | Gha, aṇ, cha after mahendra. | mahendriyam, māhendram |
| 1232 | somāṭ ṭyaṇ | Ṭyaṇ after soma. | saumyam, saumī |
| 1233 | vāyvṛtupitruṣaso yat | Yat after vāyu, ṛtu, pitṛ, uṣas. | vāyavyam, ṛtavyam |
| 1234 | rīṅ ṛtaḥ | Rī for final ṛ before a non-kṛt, non-sārvadhātuka y. | pitryam |
| 1235 | dyāvāpṛthivī…gṛhamedhāc cha ca | Cha and yat after these deity-words. | dyāvāpṛthivīyam / dyāvāpṛthivyam |
| 1236 | agner ḍhak | Ḍhak after agni. | āgneyam |
| 1237 | kālebhyo bhavavat | Time-words take their bhava affixes. | māsikam, prāvṛṣeṇyam |
| 1238 | mahārājaproṣṭhapadāṭ ṭhañ | Ṭhañ after mahārāja, proṣṭhapada. | māhārājikam |
| 1239 | devatādvandve ca | Vṛddhi of both members of a deity-dvandva. | āgnimārutam |
| 1240 | nendrasya parasya | No vṛddhi for final indra. | saumendraḥ |
| 1241 | dīrghāc ca varuṇasya | No vṛddhi for varuṇa after a long vowel; vārttikas navayajña, pūrṇamāsa. | aindrāvaruṇam, paurṇamāsī |
| 1242 | pitṛvyamātulamātāmahapitāmahāḥ | Kinship nipātanas, with vārttikas. | pitṛvyaḥ, mātulaḥ, mātāmahaḥ |
| 1243 | tasya samūhaḥ | Aṇ etc.: "a collection of that". | kākam |
| 1244 | bhikṣādibhyo 'ṇ | Aṇ after the bhikṣā list. | bhaikṣam |
| 1245 | inaṇy anapatye | In kept before a non-offspring aṇ. | gārbhiṇam, yauvanam |
| 1246 | gotrokṣoṣṭra…ajād vuñ | Vuñ after gotra words, ukṣan etc. | (forms at 1247) |
| 1247 | yuvor anākau | Yu → ana, vu → aka. | aukṣakam, rājanyakam |
| 1248 | kedārād yañ ca | Yañ, vuñ after kedāra; yañ after gaṇikā. | kaidāryam, gāṇikyam |
| 1249 | ṭhañ kavacinaś ca | Ṭhañ after kavacin, kedāra. | kāvacikam |
| 1250 | brāhmaṇamāṇavavāḍavād yan | Yan after these; vārttika pṛṣṭha. | brāhmaṇyam |
| 1251 | grāmajanabandhubhyas tal | Tal after these; vārttikas gaja, ahan, parśu. | janatā, ahīnaḥ |
| 1252 | siti ca | Before a sit affix, the stem is pada. | pārśvam |
| 1253 | anudāttāder añ | Añ after an anudātta-initial word. | kāpotam |
| 1254 | khaṇḍikādibhyaś ca | Añ after the khaṇḍikā list. | khāṇḍikam |
| 1255 | caraṇebhyo dharmavat | School names take their dharma affixes. | kāṭhakam, chāndogyam |
| 1256 | acittahastidhenoṣ ṭhak | Ṭhak after inanimate words, hastin, dhenu. | sāktukam, hāstikam |
| 1257 | keśāśvābhyāṃ yañchāv anyatarasyām | Optional yañ / cha after keśa / aśva. | kaiśyam, aśvīyam |
| 1258 | pāśādibhyo yaḥ | Ya after the pāśa list. | pāśyā |
| 1259 | khalagorathāt | Ya after khala, go, ratha. | gavyā |
| 1260 | initrakaṭyacaś ca | Ini, tra, kaṭyac after the same. | khalinī, gotrā, rathakaṭyā |
| 1261 | viṣayo deśe | Aṇ etc.: "the region that is the domain of". | śaibaḥ |
| 1262 | rājanyādibhyo vuñ | Vuñ after the rājanya list. | rājanyakaḥ |
| 1263 | bhaurikyādyaiṣukāryādibhyo vidhalbhaktalau | Vidhal / bhaktal after two lists. | bhaurikividham |
| 1264 | so 'syādir iti cchandasaḥ pragātheṣu | Aṇ after a metre: "that begins this pragātha". | pāṅktaḥ |
| 1265 | saṃgrāme prayojanayoddhṛbhyaḥ | A battle named after its purpose or warriors. | saubhadraḥ, bhārataḥ |
| 1266 | tad asyāṃ praharaṇam iti krīḍāyāṃ ṇaḥ | Ṇa: a game named after its weapon. | dāṇḍā |
| 1267 | ghañaḥ sā 'syāṃ kriyeti ñaḥ | Ña after a ghañ-word: "that action is in this". | (forms at 1268) |
| 1268 | śyenatilasya pāte ñe | Mum after śyena, tila before pāta + ña. | śyainaṃpātā |
| 1269 | tad adhīte tad veda | Aṇ etc.: "studies / knows that". | vaiyākaraṇaḥ |
| 1270 | kratūkthādisūtrāntāṭ ṭhak | Ṭhak after kratu names, ukthādi, sūtra-final words; many vārttikas. | āgniṣṭomikaḥ, naiyāyikaḥ |
| 1271 | kramādibhyo vun | Vun after krama etc. | kramakaḥ |
| 1272 | anubrāhmaṇād iniḥ | Ini after anubrāhmaṇa. | anubrāhmaṇī |
| 1273 | vasantādibhyaṣ ṭhak | Ṭhak after vasanta etc. | vāsantikaḥ, ātharvaṇikaḥ |
| 1274 | proktāl luk | Luk of the student-affix after a prokta affix. | pāṇinīyaḥ, kaṭhāḥ |
| 1275 | gāthividathikeśigaṇipaṇinaś ca | These keep in even before an offspring aṇ. | pāṇinaḥ, pāṇiniḥ |
| 1276 | ṇyakṣatriyārṣañito yūni lug aṇiñoḥ | Luk of a yuvan aṇ / iñ after four kinds of gotra word. | kauravyaḥ, vāsiṣṭhaḥ |
| 1277 | sūtrāc ca kopadhāt | Luk of the student-affix after a k-penult sūtra-name. | aṣṭakāḥ |
| 1278 | chandobrāhmaṇāni ca tadviṣayāṇi | Such text-names are never used without the student-affix. | kaṭhāḥ |
| 1279 | tad asminn astīti deśe tannāmni | Place-sense 1: "that exists in this". | audumbaraḥ |
| 1280 | tena nirvṛttam | Place-sense 2: "founded by that". | kauśāmbī |
| 1281 | tasya nivāsaḥ | Place-sense 3: "the dwelling of that". | śaibaḥ |
| 1282 | adūrabhavaś ca | Place-sense 4: "not far from that"; later affixes are cāturarthika. | vaidiśam |
| 1283 | orañ | Añ after a u-final stem. | kākṣatavam |
| 1284 | matoś ca bahvajaṅgāt | Añ after a matup-word with a many-voweled base. | saidhrakāvatam |
| 1285 | bahvacaḥ kūpeṣu | Añ after a many-voweled word, for wells. | dairghavaratraḥ |
| 1286 | udak ca vipāśaḥ | Añ for wells north of the Vipāś. | dāttaḥ |
| 1287 | saṃkalādibhyaś ca | Añ after saṅkala etc. | sāṅkalam |
| 1288 | strīṣu sauvīrasālvaprākṣu | Añ for feminine place-names in these regions. | dāttāmitrī |
| 1289 | suvāstvādibhyo 'ṇ | Aṇ after suvāstu etc., even for a river. | sauvāstavam, sauvāstavī |
| 1290 | roṇī | Aṇ after roṇī and words ending in it. | rauṇaḥ |
| 1291 | kopadhāc ca | Aṇ after a stem with k as penult. | traiśaṅkavam |
| 1292 | vuñchaṇ…kumudādibhyaḥ | Seventeen affixes after seventeen lists. | ārīhaṇakam, sākheyam |
| 1293 | janapade lup | Lup of the cāturarthika affix for a janapada. | pañcāla |
| 1294 | lupi yuktavad vyaktivacane | After lup, gender and number of the base. | pañcālāḥ |
| 1295 | tad aśiṣyaṃ saṃjñāpramāṇatvāt | SK 1294 need not be taught: names are authoritative. | pañcālāḥ |
| 1296 | lub yogāprakhyānāt | Lup need not be taught: no connection is perceived. | pañcālāḥ |
| 1297 | yogapramāṇe ca tadabhāve 'darśanaṃ syāt | Were it connection-based, the name would lapse with the connection. | pañcāla under another king |
| 1298 | pradhānapratyayārthavacanam arthasyānyapramāṇatvāt | "The affix meaning is principal" need not be taught. | — |
| 1299 | kālopasarjane ca tulyam | Nor the older definitions of time and upasarjana. | adyatana |
| 1300 | viśeṣaṇānāṃ cājāteḥ | Qualifiers follow the base too, except jāti words; three vārttikas. | pañcālā ramaṇīyāḥ |
| 1301 | varaṇādibhyaś ca | Lup after varaṇā etc., for non-janapadas. | varaṇāḥ |
| 1302 | śarkarāyā vā | Optional lup after śarkarā. | śarkarā / śārkaram |
| 1303 | ṭhakchau ca | Ṭhak, cha after śarkarā: six forms in all. | śārkarikam, śarkarīyam |
| 1304 | nadyāṃ matup | Matup for a river. | ikṣumatī |
| 1305 | madhvādibhyaś ca | Matup after madhu etc., not for rivers. | madhumān |
| 1306 | kumudanaḍavetasebhyo ḍmatup | Ḍmatup after kumuda, naḍa, vetasa; vārttika mahiṣa. | naḍvān, vetasvān |
| 1307 | naḍaśādāḍ ḍvalac | Ḍvalac after naḍa, śāda. | naḍvalaḥ, śādvalaḥ |
| 1308 | śikhāyā valac | Valac after śikhā. | śikhāvalam |
| 1309 | utkarādibhyaś chaḥ | Cha after utkara etc. | utkarīyaḥ |
| 1310 | naḍādīnāṃ kuk ca | Cha with the augment kuk after naḍa etc. | naḍakīyam |
| 1311 | bilvakādibhyaś chasya luk | Luk of the whole cha before a further taddhita; kuk stays. | bilvakīyā → bailvakāḥ |