Pramāṇa · Search · About · API

Pali Literature and Language c7.s1.t13: 14. Consonant-groups

Pāli Literature and Language c7.s1.t13

Language
English

14. Consonant-groups

c7.s1.t13.1

Combination of Two Consonants

c7.s1.t13.2

§ 48. Consonant-groups may be divided by svarabhakti according to § 29. They however remain undivided, 1. if they consist of similar consonants or of a mute with the corresponding aspirate, or 2. if they consist of a nasal with a homorganic mute. In the derivatives from *pañca* are found however also forms with *ṇṇ*, *ṇṇ*, *ññ*: *pannarasa* ‘fifteen, fifteenth’ (rarely *paṇṇa-*), *paṇṇuvisa* ‘twenty-five’ Jā. III. 138²⁰ beside *pañcadasa* Sn. 402 and *pañcavīsa* Dpvs. 3.29, and only *paṇṇāsa(m)* DhCo. III. 207¹² or *paññāsa(m)* ‘fifty’ = *pañcāśat*.¹ Assimilation of a mute to a preceding nasal is to be found in *ārammaṇa* (§ 45) = *ālambana*.

c7.s1.t13.3

¹ Cf. Pkr. AMāg. JMāh. *paṇṇarasa*, *paṇṇāsam* etc. Pischel, § 273. E. Kuhn, KZ. 33.478 tried to explain this phenomenon through dissimilation.

c7.s1.t13.4

§ 49. Consonant-groups containing *h* have to be dealt with separately: 1. Metathesis takes place in the case of groups *h* + nasal, *y* or *v*.² Thus *hṇ*, *hn*, *hm*, *hy*, *hv* become respectively *ṇh*, *nh*, *mh*, *yh*, *vh*. Examples: *pubbaṇha* ‘forenoon’ D. I. 109²⁹, DhCo. III. 98²⁰ = *pūrvāhṇa*. Similarly *aparaṇha* ‘afternoon’ = *aparāhṇa*, and after them also *sāyaṇha* ‘evening’ = *sāyāhṇa*; *cinha* ‘sign’ Abhp. 55 (beside *cihana* with svarabhakti Abhp. 879) = *cihna*; *jimha* ‘crooked, false’ Jā. III. 111¹⁷, A. V 299⁴ = *jihma*; *veyhā* ‘movable chair’ Jā. VI. 500¹³ from *vāhya*. Similarly in Future Passive Participles: *sayha* ‘that which is to be endured’ Sn. 253 = *sahya*; in Gerunds: *āruyha* from root *ruh* with *a* ‘to mount’ = *āruhya*; in Passives: *duyhati* ‘is milked’ Milp. 41⁴ = *duhyate*.³ Also *jivhā* ‘tongue’ D. I. 21¹⁹ etc. = *jihvā*. In compounds we have *bavhābādha* ‘ill’ M. II. 94²⁰ = *bahvābādha*; *bavhodaka* ‘containing much water’ Thl. 390.—2. The combination *hr* undergoes various changes.⁴ Initially we find *h* in *hesati* ‘neighs’ Dāṭhāv. 44, *hesā*, *hesita* ‘neighing’ Rasav. II. 98¹⁶, Mhvs. 23.72 = *hreṣate*, *hreṣā*, *hreṣita*, but we have *r* in *rassa* (as opposed to Māg. *hassa*) ‘short’ Dh. 409, JāCo. I. 356¹⁸ etc. = *hrasva*. Both svarabhakti and metathesis are in evidence in *rahada* = *hrada* (§ 47.2).

c7.s1.t13.5

² As in Pkr.; Pischel, § 330, 332.

c7.s1.t13.6

³ In Pkr. *hy* becomes *jjh*, initially *jh*; Pischel, § 331.

c7.s1.t13.7

⁴ For *hl* may be quoted only *kallahāra* ‘white water-lily’ Dpvs. 16.19 = *kahlāra*; cf. § 30.4.

c7.s1.t13.8

§ 50. There should further be mentioned the groups consisting of a sibilant followed by a nasal. As in Pkr.,⁵ so in Pāli too, there takes place in these cases, as a rule, metathesis with concomitant mutation of the sibilant into *h*. Frequently there are found parallel forms with svarabhakti which took place in some cases already in the original sound-group and in some cases after metathesis and mutation of *s* into *h*. 1. *śn* becomes *ñh* (Pkr. *ṇh*): *pañha* (AMāg. *paṇha*) ‘question,’

c7.s1.t13.9

⁵ Pischel, PkrGr. § 312 ff.

c7.s1.t13.10

= *praśna*; *pañhipaṇṇi* (sic! not *paṇhi-*) (name of a plant) Abhp. 584 = *préniparṇi*.—2. *śm* becomes *mh*: *amhanā* ‘with the stone’ Sn. 443 = *aśmanā*; *amhamaya* ‘stony’ Dh. 161. Besides also *asmā* Jā. III. 29¹⁷. Sometimes *sm* = *śm* is retained in Pāli: *Kasmīra* = *Kīśmīra*; *rasmi* (Pkr. *rassi*) ‘ray, rein’ Dh. 222, M.I. 124²¹, JāCo. I. 444 (beside *raṃsi* Sn. 1016, Vv. 52.5) = *raśmi*; *vesma* ‘house’ Abhp. 206 = *veśman*. Initially *ś* is assimilated to *m* (*m* < *mm*) in *massu* (AMāg. *maṃsu*) ‘beard’ D. I. 60¹⁷ etc. = *śmaśru*.—3. *ṣṇ* becomes *ṇh*: *uṇha* ‘hot, heat’ D. Andersen Pāli Gl. sub voce = *uṣṇa*; *uṇhīsa* ‘diadem’ D. II. 179¹, Dpvs. 12.1 = *uṣṇīṣa*; *kaṇha* ‘black, demoniac’ Thl. 140, Vin. III. 20³⁰ etc. = *kṛṣṇa*; *taṇhā* ‘thirst’ (beside rarer *tasiṇā* § 30.5) = *tṛṣṇā*; *tuṇhī* ‘silent’ = *tūṣṇīm*; *Veṇhu* (§ 10) = *Viṣṇu*; *suṇhā* ‘daughter-in-law’ through *suṣṇā* from *snuṣā*,¹ beside *suṇisā* (§ 31.2).—4. *ṣm* becomes *mh*: *gimha* ‘summer’ Dh. 286, Vin. I. 79²⁹, JāCo. I. 390²⁸ = *grīṣma*; *semha* (AMāg. *sembha*, *simbha*) ‘phlegm’ = *śleṣman*; *tumhe*, *tumhākaṃ* etc. = *yuṣme*, *yuṣmākam* etc. (§ 104). *sm* = *sm* is retained in *usmā* ‘warmth’ D. II. 335¹⁵ (beside *usumā* § 31.2) = *uṣman*; *āyasmant* ‘venerable’ = *āyuṣmant*; *bhesma* (sic!) ‘horrible’ Abhp. 167 = *bhaiṣma*.—5. *sn* becomes *nh* in: *nhāyati* ‘bathes,’ *nhāna* ‘bath’ etc. mostly in verses, beside the forms appearing in prose such as *nahāyati*, *nahāna* (*sunhāta*, *sunahāta* ‘well-bathed’ D. I. 104²⁷, M. II. 120¹³, S. I. 71¹¹) = *snāyati*, *snāna*; *nhāru* Vin. I. 25¹ beside the more frequent *nahāru* Sn 194, M. I. 429²² etc. connected with *snāyu* (§ 46.3).—6. *sm* becomes *mh* in: *vimhaya* ‘astonishment’ Mhvs. 5.92, *vimhita* Mhvs. 6. 19 = *vismaya*, *vismita*; *amhe*, *amhākaṃ* etc. (§ 104) = *asmān*, *asmākam* etc. There are moreover the Pāli forms *asme* JāCo. III. 359²¹ (verse), *asmākaṃ* Sn. 102 in which *sm* has been retained. It is retained moreover in *asmi* (beside *amhi*) ‘I am’, in the endings *-smā* of Abl. Sg. (beside *-mhā*) and *-smiṃ* of Loc. Sg. (beside *-mhi*), in *bhasma* ‘ash’ Dh. 71, S. I. 169²⁵ = *bhasman*. There is svarabhakti in initial *sm* in the case of *sumarati* ‘remembers’ Dh. 324¹ = *smarate*; beside it also *surati* with assimilation. Similarly *sita* ‘smile’ M. II. 45⁴, DhCo. III. 479⁷ beside *mihita* JāCo. VI. 504³⁰ = *smita*.

c7.s1.t13.11

¹ H. Jacobi, *Erzählungen in Māhārāṣṭrī*, p. XXXII. 3 explains Pkr. *suṇhā* through metathesis out of *ṣṇuṣā*. Not so Pischel, § 148.

c7.s1.t13.12

§ 51. Moreover, in so far as no svarabhakti intervenes, the assimilation of consonant is characterised by the rule that the consonants of lesser power of resistance are assimilated to those of greater resisting power. The power of resistance diminishes in the order: mutes—

c7.s1.t13.13

sibilants—nasals—l, v, y, r. Thus an r is assimilated to a mute or a sibilant, both when it precedes or follows it. Where a mute is combined with a mute, or a nasal with a nasal, the first consonant is assimilated to the second.

c7.s1.t13.14

The following details should also be noticed: 1. If the consonant-group contains an aspirate, the aspiration appears at the end of the new group after completed assimilation: kh + y becomes kkh, k + th becomes tth. Aspiration of the resultant group is normally caused also by the presence of a sibilant in the original group: s + t becomes tth.—2. In initial position, there remains only one of the assimilated consonants, which is normally the second one: thus tth becomes th. In compounds however the double-consonance normally appears again, and occasionally also in external sandhi. Cf. § 67, 74.1.—3. Wherever according to the laws of assimilation the sound-group vv would originate, there appears in P., in contrast to the other Middle Indian dialects, always bb¹; initially however only v.—4. Also certain other qualitative changes are concomitant with assimilation: dentals, as well as n, are palatalised by a following y before the effectuation of assimilation. Sometimes also k is palatalised in the combination k + s.—5. Between m and a following liquid there is introduced in the first instance the slide-sound b². Only after that there takes place assimilation or separation through the svarabhakti-vowel. Examples: amba ‘mango’ Vv 81. 16, JāCo. I. 450²⁰ etc. through *ambra = āmra; ambaṭaka (§ 42. 1) = ămrātaka; tamba ‘red, copper’ Vv. 32 3, M. III. 186¹⁵, JāCo. I. 464⁷ etc. through *tambra = tāmra; Tambapaṇṇi (Ceylon) JāCo. I. 85¹¹ = Tāmraparṇī. Svarabhakti in ambila ‘sour, acid’ JāCo. I. 349³⁰ etc. through *ambla = amla. In this way is to be explained also gumba ‘mass, bush’ D. I. 84¹⁶, Thl. 23 etc. = gulma, with metathesis, from *gumla, *gumba³.

c7.s1.t13.15

¹ Similarly, according to § 6. 2, yobbana (Pkr. jovvaṇa) ‘youth’ Dh. 155 f., D. I. 115 ¹⁶ = yauvana.

c7.s1.t13.16

² Also in Pkr.; Pischel, § 295.

c7.s1.t13.17

³ No metathesis has taken place in Pkr. gumma; here assimilation has been direct.

c7.s1.t13.18

§ 52. Progressive assimilation takes place 1. in the combination of mute with mute: chakka ‘collection of six’ M. III. 280 ³³ = ṣaṭka; satthi ‘thigh’ Thl. 151, Vv. 81. 17, JāCo. II. 408 ⁵ = sakthi; mugga ‘bean’ Jā III. 55⁴, D. II. 293²⁰, JāCo. I. 274 ²⁴ = mudga;

c7.s1.t13.19

ugg āta ‘blow’ Vin. I. 192²=udghāta.—2. In the combination of sibilant with mute (with aspiration of the resultant group): acchera (§ 27. 5)=āścarya ; nikkha, nekkha (§ 10)=niṣka¹; apphoṭeti ‘claps the hands’ JāCo. VI. 486¹⁷ (apphoṭana, apphoṭita)=āsphoṭayati. Initially: khalati ‘stumbles’ Thl. 45, Milp. 187¹²=skhalati; thaneti ‘thunders’ D. II. 260²⁵ (verse), thanita JāCo. I. 470¹²=stanayati, stanita; phassa ‘touch’=sparśa. There is no assimilation in bhasta (§ 40. 1 a)=basta (cf. bhastā ‘bellows’ M. I. 128²¹=bhastrā); vanaspati ‘tree’ Jā. I. 329⁶, S. IV. 302²³=vanaspati.—3. In the combination of liquid with mute, sibilant or nasal: kakka (a precious stone) VvCo. 111²⁵=karka (karketana); kibbisa ‘sin’ Jā. III. 34¹³, M. III. 165⁵=kilbiṣa; vāka (§ 6. 1)=valka; kassaka ‘farmer’ D. I. 61¹⁶=karṣaka²; ūmi ‘wave’ JāCo. I. 498⁶, Milp. 3⁷ (§ 5 b)=ūrmi; kammāsa ‘spotted’ D. II. 80²⁴, A. II. 187²⁶=kalmāṣa.—4. In the combination of nasal with nasal: ninna ‘deep, low’ Dh. 98, S. IV. 191¹, JāCo. II. 3⁶=nimna; ummūleti ‘uproots’ JāCo. I. 328⁹=unmūlayati.—5. In the combination of r with l, y, v: dullabha ‘difficult to attain’=durlabha; ayya ‘venerable’ Vin. II. 290²⁸, D. I. 92¹³, JāCo. III. 61¹³ (beside ariya with svarabhakti according to § 30. 1)=ārya³; udiyyati ‘is heard, resounds’ Thl. 1282=udiryate⁴; niyyāti ‘goes away’ D. I. 49³¹ (niyyāna, niyyānika)=niryāti etc.; niyyāma ‘sailor’ JāCo. IV. 137¹⁰=niryāma; niyyāsa ‘resin’ Mhvs. 29. 11, Thūpavs. 57¹⁸=niryāsa; saṃkiyati ‘is mixed up, defiled’ S. III. 71¹⁶, A. IV. 246¹⁸ (§ 5b) through *-kiyyati= saṃkīryate; kubbanti ‘they make’ Jā III. 118¹⁰ (§ 51.3)=kurvanti; sabba ‘all’=sarva; dubbuṭṭhi(kā) ‘drought’ D. I. 11⁷, JāCo. VI. 487⁶ =durvṛṣṭi.⁵ In verbs of the type jīryati, pūryate we have mostly r (instead of yy, y) as the result of the regressive assimilation of ry. Thus (pari)pūrati ‘is filled’ Dh. 121 f., Jā. I. 498²², JāCo. I. 460²⁷=pūryate, beside the analogically formed passives -hīrati, -bhīrati (§ 175). Cf. the doublets jīyati ‘is digested, becomes old’ and jīrati=jīryati, -te (§ 137).

c7.s1.t13.20

¹ Similarly dukkha ‘sorrow’ = duḥkha.

c7.s1.t13.21

² Cf. ghaṃsati ‘rubs’ (§ 6.3) = gharṣati.

c7.s1.t13.22

³ In Pkr. ry becomes yy only in Māg., otherwise jj; Pischel, § 287.

c7.s1.t13.23

⁴ The similar form miyyati (and miyati) ‘dies’, which cannot be connected with Skr. mriyate, is derived from a *mīryate.

c7.s1.t13.24

⁵ In analogy with it there appears bb also in subbuṭṭhikā.

c7.s1.t13.25

§ 53. Regressive assimilation takes place 1. in the combination of mutes with nasal: ubbigga ‘anxious’ Jā. I. 486¹⁰, JāCo. III. 197¹⁴= udvigna; soppa (§ 25.2) = svapna; abhimathati ‘rubs, grinds’ Dh. 161, S. I. 127¹⁴ = abhimathnāti; chaddan ‘veil, cover’ in vivattacchadda (of a Buddha) D. I. 89⁹, JāCo. I. 56¹⁴ = chadman. On the other hand jñ becomes ññ¹ through progressive assimilation: paññā, paññāṇa ‘knowledge’ Sn. 1136, Dh. I. 124⁴ = prajñā, prajñāna; raññā, rañño Instr. and Gen. Sg. of rājan = rājñā, rājñas. In initial position jñ becomes ñ: ñatti ‘request’ Vin. I. 56¹⁴ etc. = jñapti. For āṇā ‘order’ see § 63.2. The assimilation observed in the form rummavati = rukmavati² quoted by E. Kuhn, Beitr. p. 46 from Vuttod. would also be progressive.—2. In the combination of mutes with liquids: takka ‘whey’ JāCo. II. 363¹⁰ = takra; udda ‘otter’ Vin. I. 186²¹, JāCo. III. 51²⁶ = udra; sobbha (§ 25.2) = śvabhra; sukka ‘white’ = śukla. In initial position there appears in these cases only a single mute: kaya-vikkaya ‘purchase and sale’ D. I. 5¹⁰ = krayavikraya; tāṇa ‘protection’ Dh. 288, M. III. 165⁵ = trāṇa; bhātar ‘brother’ = bhrātar. Sometimes the combination mute + r remains unchanged: nigrodha (§ 21) = nyagrodha; tatra ‘there’ Thl. 31, Vin. I. 10¹², D.I. 76²³ (beside tattha Thl. 183) = tatra; citra ‘multicoloured’ Jā. VI. 497¹⁶, D. I. 7²² (beside citta Dh. 151) = citra; bhadra ‘happy’ S. I. 117²⁴ etc. (beside bhadda D. II. 95¹⁷ etc.) = bhadra; udriyati ‘is split’ S. I. 113¹⁵, D. 1. 96¹⁷ (udraya ‘fruit, reward’ S. II. 29¹², A. I. 97³¹) from *uddriyate for uddiryate³. The Part. Pres. atriccham Jā. I. 414⁶, III. 207¹⁵ is explained in the commentary by atra atra icchanto ‘desiring this and that’; cf. atriccha ‘desirous’ JāCo. III. 206¹⁰, atricchatā ‘covetousness’ JāCo. III. 222⁶. In atrajā ‘son, daughter’ Jā. III. 181⁴, Dpvs. 18.29 through *attajā = ātmajā the tr originated through folk-etymology.—3. In the combination of mutes with semi-vowels (dental + y will be discussed in § 55): sakka ‘capable’ = śakya; vuccati ‘is said’ = ucyate; kuḍḍa ‘wall’ D. I. 78³, S. III. 238²⁵ = kuḍya; pajjalati ‘burns’ D. II. 163²⁰ etc. = prajvalati; labbha ‘attainable’ Jā. III. 204²⁷, M. II. 220¹³ = labhya; cattāro ‘four’ = catvāras; addhan ‘way’ = adhvan; saddala ‘grassy’ Thl. 211, Ja. VI. 518²¹ = śādvala. In initial position there appears only the single mute: kaṭhita (§ 42. 2) = kvathita; dija (poetic term) ‘bird’

c7.s1.t13.26

¹ In Pkr. jñ mostly becomes ṇṇ; Pischel, § 276.

c7.s1.t13.27

² Another rumma is to be found in rummavāsī ‘irregularly dressed’ Jā. IV 380¹, 334³; cf. also rummi Jā. IV. 322¹ (Com. = anañjitamaṇḍita)

c7.s1.t13.28

³ The verb udrabhati, -bheti ‘tears off’ M. I. 306¹²,¹⁵ (in 307² we have udrahey-yum) is derived from a root *drabh or *drah (Skr. darh) with ud

c7.s1.t13.29

D. II. 258²² (verse), Jā. II. 205¹⁵ = dvija; dhanita ‘resounding’ DCo. I. 177¹, Milp. 344², JPTS. 1887, p. 26¹⁸ = dhvanita¹. But we find initially b for dv in bārasa ‘twelve’, bāvisati, battimsa (§ 116. 2). Sometimes the combination mute + semi-vowel remains unchanged: vākya (poetic term) ‘word, speech’ D. II. 166⁵ (verse) etc. = vākya²; ārogya ‘good health’ Jā. I. 366²⁴, D. I. 11⁹, JāCo I. 367³ = ārogya; kvam ‘where, how’, kvaci ‘somewhere’ = kva, kvacit; the gerundial suffixes -tvā, tvāna; dve ‘two’ (beside duve), dvidhā D. II. 341³, dvedhā Vin. I. 97⁵ etc. = dve, dvidhā, dvedhā etc. In compositional combination dv, dv become bb³ through vv due to progressive assimilation (cf. § 55, 57): ubigga (see § 53. 1) = udvigna; ubbilla (§ 15.1) from root vell with ud; ubbāsiyati ‘becomes depopulated’ Mhvs. 6. 22 (Pass. of the Caus. of root vas with ud); ubbatteti ‘anoints’ Thūpavs. 39¹¹ = udvartayati; ubbinnaya ‘against the Vinaya’ Vin. II. 306²⁰ = *udvinaya; ubbejitar ‘one who causes excitement’ Pu. 47¹⁷ from root vij with ud; tabbam sika ‘descended from this family’ Mhvs. 37.89 (= Cūlavs. 37. 39 ed. Colombo) from tadvams-. Similarly chabbaṇṇa ‘six-coloured’ Mhvs. 17. 48 = ṣaḍvarṇa; chabbisati ‘twenty-six’ = ṣaḍvimśati.

c7.s1.t13.30

¹ Minayeff, PGr. p. 49, § 3; Morris, Transactions Congr. of Or., London 1892, I. 482 f.

c7.s1.t13.31

² For the proper name Sākya cf. p 64, f -n. 2.

c7.s1.t13.32

§ 54. Regressive assimilation takes place 4. in the combination of sibilant with liquids or semi-vowels: missa ‘mixed’ Thl. 143, JāCo. III. 95⁴ = miśra; avassam ‘necessary’ DhCo. III. 170²³ = avaśyam; vayassa ‘friend’ JāCo. II. 31⁹ = vayasya; assa ‘horse’ = aśva; palissajati (§ 44) = pariṣvajate⁴. In initial position there is only s: sota ‘stream’ = srotas; semha (§ 5) = śleṣman; sandana ‘chariot’ Jā. VI. 511³ = syandana; seta ‘white’ = śveta. Initially sv is retained in sve ‘to-morrow’ (beside suve) = śvas, svātanāya ‘for next day’⁵ and in forms like svakkhāta ‘well-proclaimed’ from su-ākhyāta, svāgata ‘welcome’ Vv. 63. 4, D. I. 179¹⁶ (variant reading sāg-) = svāgata.—The combination ṣy becomes h in future forms like ehisi ‘you will go,’ ehiti (beside essasi, essati) = eṣyasi, eṣyati. Similarly kāhāmi ‘I shall do,’ kāhasi, kāhati through *kassāmi, *kāsāmi from *karṣyāmi. See § 153. 1.—5. In the combination of nasal or liquid *l* with semi-vowels (*n*, *ṇ* + *y* will be discussed in § 55): *sammannati* ‘agrees’ Vin. I. 106⁴, II. 295¹¹ from root *man* (*manve*, not = *manye*) + *sam*; *samannesati* ‘seeks’ D. I. 105²⁵, S. I. 194²⁸ from root *i* with *sam-anu* and similar compounds with *anu*; *kiṇṇa* ‘ferment’ Abhp. 533 = *kiṇva*; *ramma* ‘graceful’ Thl. 63, Dpvs. 1. 69 = *ramya*; *kalla* ‘ready, possible’ Vin. I. 16¹, D. I. 157²⁶, S. IV. 25² = *kalya*; *billa* (a kind of fruit) A. V. 170²⁶ = *bilva*; *bella* (the same fruit) Jā. III. 77²⁴ (besides *beluva* M. II. 6³⁵) = *bailva*; *khallāṭa* ‘bald’ (in *Khallāṭanāga* Mhvs 33. 29) = *khalvāṭa*; *pallaṅka* ‘seat with cross-legs’ = *paryaṅka* presupposes a *paryaṅka* just as *pallattha* = *paryasta* presupposes a *palyattha*. The combination *nv* is retained in *anvadeva* ‘afterwards’ D. II. 172²⁵, M. III. 172²⁹; *anveti* ‘follows’ = *anveti*; *anvaya* ‘progeny’ D. II. 261⁹ (verse), M. I. 69⁵ (besides *durannaya* ‘difficult to follow’ Dh. 92, Jā. II. 86²) = *anvaya*, etc. Similarly *my* in *°hamya*, *°kamyatā* ‘wishing something, desire for something’ Vin. IV. 12²⁴. Thl. 1241 = *°kāmya*, *°kāmyatā*; *ly* in *malya* ‘flower’ Vv. 1. 1, 2. 1 = *mālya*, etc. In flexion we have, *e.g.*, *pipphalyā* (Gāthā-language !) Vv. 43. 6 (I. Sg. of) *pipphalī* ‘pepper’.—6. In the combinations *vy*, *vr* which become *bb* (through *vv*): *paribbaya* ‘expenditure’ JāCo. I. 433¹⁸ = *parivyaya*; *udayabbaya* (in composition) ‘origin and decay’ Thl. 10, 23 = *udaya + vyaya*; *tibba* ‘sharp’ Dh. 349, S. I. 110¹⁸ = *tīvra*; *patibbatā* (in composition) ‘devoted to the husband’ Jā. VI. 533⁷ = *pativratā*. Initially we have *v* in: *vapayanti* ‘they go away, disappear’ Vin. I. 2⁵ from root *i* with *vi-apa*; *vāḷa* ‘beast of prey, snake’ Jā. VI. 497¹³, JāCo. I. 99¹⁴ = *vyaḍa*; *vata* ‘religious observance’ Vv. 84. 24, S. I. 201²⁹, JāCo. III. 75¹ = *vrata¹*; *vo-* (§ 26.2) = *vyava*. Also in composition as in *udayavaya* A. II. 45²⁶ besides *udayabbaya* (see above). Frequently however *vy* is retained as in: *vyāseka* ‘mixing up’ DCo. I. 183²⁴, *vyāsiñcati* ‘is mixed’ S. IV. 78⁷ from root *sic* with *vi-ā*; *vyeti* ‘bifurcates’ Thl. 170 (*vyagā*) = *vyeti*; *vyāvaṭa* (§ 38.5) = *vyāpṛta*, etc. In manuscripts from hinter-India *by-* is written for *vy-* in these cases. Medially we have *vy* in *paṭhavyā* Dpvs. 5. 2 (Loc. Sg.) besides *paṭhaviyaṃ*, and in the composition *udayavyaya* D. II. 35¹⁵ (var. reading *-bb-*).

c7.s1.t13.33

³ In Pkr. too dv becomes vv in combinations with ud; Pischel § 298 (towards the end).

c7.s1.t13.34

⁴ The verbs osakkati, paccosakkati (§ 28.2) are to be explained through *ossakkati.

c7.s1.t13.35

⁵ Johansson, Monde Oriental 1907/8, p. 106 f.

c7.s1.t13.36

¹ The word *sorata* ‘kind’ and its abstract noun *soracca* should not be derived from *suvrata* which has become *subbata* in Pāli, but from *saurata*, *sauratya* (the opposite view is expressed in JPTS. 1909, p. 233).

c7.s1.t13.37

§ 55. Palatalisation takes place in the combination of dentals (including *ṇ*) with *y*, as well as in the combination of *ṇ* with *y* (cf. § 51.4): *sacca* ‘true’ = *satya*; *racchā* ‘street’ JāCo. I. 425¹² (besides *rathiyā* D. I. 83⁶)=*rathyā*; *chijjāti* ‘is split’ = *chidyate*; *dvcjjha* ‘falsity, uncertainty’ A. III. 403¹⁰ = *dvaidhya*; *añña* ‘another’ = *anya*¹. In flexion: *jaccā* Jā. III. 395⁹, Sn. 136 (besides *jātiyā* Ins. Sg. from *jāti* ‘birth’; *najjā* Vin. I. 1⁶ (besides *nadiyā*) G. Sg. from *nadī*. Initially: *cajati* ‘leaves’ = *tyajati*; *jotati* ‘lightens up’ JāCo. I. 53⁴ = *dyotate*; *ñāya* ‘method’ D. II. 21² etc. = *nyāya*. Examples of *ññ* from *ny*: *kammañña* ‘ready for use’ A. I. 9³¹, Vin. I. 182³¹ (besides *kammaṇiya* D.I. 76¹⁴, Vin. III. 4¹⁹) = *karmanya*; *piññāka* ‘oil-cake’ D. I. 166²², Pu. 55²⁵ = *piṇyāka*, etc. The rule seems to apply also in the case of the combination cerebral + *y*: *vekurañjā* <*vaikuraṇḍya* from *vikuraṇḍa* ‘without testicle’ (Skr. *kuraṇḍa*). But *addha* ‘rich’ = *ādhya*. When *ud* precedes a word beginning with *y*, the combination *dy* becomes *yy* through progressive assimilation (cf. § 53.3, 57): *uyyāna* ‘garden’ = *udyāna*; *uyyutta* ‘alert’ DhCo. III. 451¹⁵ = *udyukta*, etc.²

c7.s1.t13.38

¹ In Pkr. *ny* becomes *ṇṇ*, which is also written as *nn* in the Jaina works; Pischel, § 282.

c7.s1.t13.39

§ 56. The Skr. sound-group *kṣ* requires special treatment. 1. Where Skr. *kṣ* = Indo-Iranian *kṣ* or *śṣ*, it is represented as in Prākrit by *kkh* or *cch*. Pischel’s hypothesis, according to which Pkr. *kkh* should be derived from Indo-Iranian *kṣ* = Avestan *xṣ*, and Pkr. *cch* from Indo-Iranian *śṣ* = Avestau *š*, although both have coincided in *kṣ* in Skr.,³ can be as little proved from the actual state of things in Pāli as from that in Pkr. Rather it seems that *kkh* and *cch* appear quite promiscuously, sometimes in accordance with, but as often in opposition to, the indication of the Avestan language. Sometimes even Pāli and Pkr. do not agree with each other, and not infrequently both forms are found side by side also in P. as in Pkr. (a) We have *kkh* in P. *dakkhiṇa* (similarly Pkr., but Avestan *dašina*) = *dakṣiṇa*; *makkhikā* ‘fly’ (Av. *maxši*, but Pkr. *macchiā*⁴) = *makṣikā*. Initially: *khudā* (§ 49.² b) = *kṣudhā* (Av. *šuδa*. Pkr. *khuhā* and *chuhā*) etc. (b) We find *cch* in *kaccha* (similarly in Māh., *kakkha* in AMāg. JMāh. : Av. *kaša*) ‘axis, arm’ Sn. 449, Vin. I. 15¹⁰, JāCo. II. 88¹⁸ = *kakṣa*; *tacchati* (Pkr.

c7.s1.t13.40

² In Pkr. we get *jj*, which however (in analogy with *vv* from *dv*) is not directly derived from *dy*, but from *yy*.

c7.s1.t13.41

³ Pischel, GGA. 1881, p. 1322; PkrGr. § 316 ff.

c7.s1.t13.42

⁴ Only S. *immakkhia* = *nirmakṣika*.

c7.s1.t13.43

takkhai and tacchai, Av. tašan) ‘to shape’ D. II. 341¹, JāCo. I. 247¹⁸ = takṣati, etc. Initially: chārikā ‘ash’ Ud. 93⁹, D. II. 164⁹ = kṣārikā, etc. (c) Sometimes in P. kkh and cch alternate in one and the same word: acchi ‘eye’ Abhp. 149 besides the usual akkhi = akṣi (Pkr. akkhi and acchi, Av. aši); ucchu (§ 16.1a) = ikṣu (Pkr. ucchu, in AMāg. also ukkhu) besides Okkāka (p. 66, f.-n. 1) through *Ukk-, *Ukkh- = Ikṣvāku; accha (§ 12.1) besides ikka (for *ikkha, § 12.2 and § 62.2) = tkṣa (AMāg. accha and riccha, Av. arəša). A differentiation in meaning has come about in chaṇa ‘festival’ JāCo. I. 423⁹ and khaṇa ‘moment’ Thl. 231, Vin. I. 12¹² = kṣaṇa, as well as in chamā ‘earth’ Sn. 401, M. III. 164²⁵ and khamā ‘forgiveness’ Abhp. 161 (also khamā ‘earth’ Abhp. 994) = kṣamā.¹—2. Where Skr. kṣ corresponds to the Indo-Iranian sonant-group ǰž = Av. γž, there appears in P. ggh, jjh and in Pkr. jjh.² Cf. paggharati ‘drips’ Thl. 394, D. II. 347¹⁸ = prakṣarati. Similarly uggharati Thl. 394, D. II. 347¹⁸. Initially we have jh: jhāma (so also in AMāg.) ‘burnt’ S. IV. 193¹⁷, JāCo. I. 238¹⁴, DhCo. I. 118²⁴ = kṣāma; jhāyati (AMāg. jhiyāi) ‘burns (intrans.)’ Ud. 93⁹, A. I. 137¹², JāCo. I. 485⁹ = kṣāyati; jhāpeti ‘burns (transitive)’ D. II. 159²⁴, JāCo. III. 164⁹ etc. = kṣapayati.

c7.s1.t13.44

§ 57. Finally the Skr. sound-groups ts and ps have to be mentioned. Both become cch in P.: kucchita ‘despicable’ VvCo. 215¹ = kutsita; maccharin ‘jealous’ Dh. 262, Vv. 52.26, JāCo. I. 345¹⁸ = matsarin; vacchatara ‘ox’ D. I. 127¹², A. IV. 41¹¹ = vatsatara; accharā ‘nymph’ = apsaras; jigucchā, jigucchati (§ 18.2) = jugupsā, jugupsate. Skr. icchati and ipsate have coincided in icchati ‘wishes’. Through dialectical influence ts appears as th initially in tharu (also AMāg. tharu, besides charu) ‘handle, sword’ A. III. 152³², JāCo. III. 221¹ = tsaru. The representation of ps by ch in initial position is quite regular: chāta ‘hungry’ Jā. III. 199², JāCo. I. 345²⁹ = psāta.—In composition, when t (d) at the end of a word combines with an initial Skr. ś or s, the resulting sound-groups *ts (= Skr. cch), ts become ss through progressive assimilation. In rare cases there appears cch³, mostly in combinations with ud. Cf. § 53.3, 55. Examples for ts: ussada ‘friction’ DhCo. I, 28¹⁵ (verse), JāCo. IV. 188¹³ from Skr. utsādana (ucchādana); ussanna ‘increased’ Vin. I. 71³⁴, DhCo. III. 425¹⁰= utsanna; ussava ‘festival’ JāCo. III. 87³=utsava; ussahati ‘exerts’ D. I. 135²⁰, JāCo. II. 19²⁵, ussāha Vin. I. 58¹⁹, ussoḷhi Dhs. 13, 22, Vbh. 217⁷=utsahate, utsāha, *utsoḍhi; ussiñcati ‘exhausts’ JāCo. I. 450¹¹=utsiñcati; ussuka ‘eager’ Dh. 199, ussukka (§ 15.4)=utsuka, autsukya; ussūra ‘evening’ DhCo. III. 305¹²=utsūra. Also tassāruppa ‘corresponding to that’ M. III. 163¹⁹ from tat + sār-. Examples for *ts (Skr. cch): ussaṅkin ‘coward’ Vin. II. 190²³ from root śaṅk with ud; ussīsaka (JMāh. ūsīsāa) ‘head-end (of bed)’ JāCo. II. 410²⁰, Mhvs. 30.77=ucchīṣaka (ud + śīrṣa); ussussati ‘dries up’ S. I. 126²=ucchuṣyati (śuṣ with ud). On the other hand we find in P. cch for ts in ucchaṅga ‘lap’ Jā. I. 308⁵, JāCo. II. 412⁹=utsaṅga; ucchādana ‘annihilation’ D. I. 76¹⁸, S. IV. 83²⁷, ‘friction’ (besides ussada, see above) D. I. 7¹⁹, DCo. I. 88¹²=utsādana; cch appears also for *ts in ucchiṭṭha ‘leaving’ Jā. VI. 508⁷, DhCo. III. 208²=ucchiṣṭa (śiṣ with ud).

c7.s1.t13.45

¹ khudda ‘small’ Thl. 43, Vin. II. 287³¹ and chuddha ‘despised’ Dh. 41 are of course not both = kṣudra as Childers explains them in his dictionary; the latter rather = kṣubdha We have moreover P. khubbhati Jā. VI. 489¹³ and khobheti JāCo. I. 501³¹.

c7.s1.t13.46

² Pischel, § 326.

c7.s1.t13.47

³ So also in Pkr.; Pischel, § 327a.

c7.s1.t13.48

§ 58. Under the influence of the general laws of assimilation, groups of more than two consonants are reduced to combinations of two.—1. Where a nasal preceding a mute stands at the beginning of a group, it remains according to § 48, and the following consonants are assimilated and simplified: ānañca ‘infinity’ is derived from ānantya through *ānañcca (§ 55); randha ‘hole’ Jā. III. 192²⁹, A. IV. 25¹⁵, DhCo. III. 376¹⁰ is derived from randhra through *randdha; kaṅkhā ‘doubt’ from kāṅkṣā through *kaṅkkhā.—2. When a heavy consonant (mute or sibilant) stands between light consonants (nasal, liquid, semi-vowel), at first the first light consonant is assimilated to the heavy one: macca ‘man’ is derived from martya through *mattya, *matya; paṇhi(kā) ‘heel’ D. II. 17¹⁹, JāCo. I. 491¹⁰ from pārṣṇi through *passṇi, *paṣṇi; akamha ‘we did’ (§ 159. III) is likewise derived from akārṣma. Svarabhakti is in evidence in vaṭuma ‘path’ D. II. 8⁵, S. IV. 52²¹ through *vaṭṭma, *vaṭma=vartman, as well as in pāsaṇi(?) ‘heel’ Abhp. 277 besides paṇhi.—3. In the same manner, assimilation and simplification of the first two consonants take place at first in those cases where a light consonant stands at the end of the group, and two heavy consonants or one light and one heavy consonant stand at the beginning of the group: oṭṭha (§ 10) through *uṭṭhra, *uṭhra=uṣṭra; tikkha ‘sharp’ Jā. III. 151⁵, S. I. 191¹⁰, Vin. I. 6³⁸ through *tikkhṇa, *tikhṇa=tīkṣṇa; dāṭhā ‘tooth’ besides daṭṭhā Milp.

c7.s1.t13.49

Combination of more than two Consonants

c7.s1.t13.50

150⁹ = *damṣṭrā*. Svarabhakti however may take place, though mostly in the last stage: *tikhina* ‘sharp’ JāCo. II. 18²¹ etc. besides *tikkha*; *pakhuma* ‘eyelid’ D. II. 18²⁸, Th2Co. 255¹⁴ = *pakṣman*; *sukhuma* ‘tender’ = *sūkṣma*¹. Taking into consideration also § 57, we are then in a position to understand also *ussāpeti* ‘raises’ Vin. III. 203¹ = *ucchrā-payati* (root *śri* with *ud*), (*sam*) *ussita* = (*sam*) *ucchrita*, (*sam*) *ussaya* = (*sam*) *ucchraya*. The *v* is retained in Gerunds like *mutvā* = *muktvā*, *patvā* = *āptvā* with *pra*, *vatvā* = *uktvā*. Similarly *y* is retained in forms like *ratyā* ‘at night’ = *rātryām*, *ratyo* ‘the nights’ = *rātryas*,³ as well as in *agyantarāya* ‘hindrance through fire’ Vin. I. 112³⁷, *agyāgāra* D. I. 101²², Vin. I. 24²¹ from *aggi* + *antarāya* (*āgāra*).—4. Groups consisting only of heavy consonants are found in composition in forms like *uggharati* (§ 56.2) from root *kṣar* with *ud* and *nicchubhati* ‘thrusts out’ Bu. 11.15, Jā. III. 512¹¹, Milp. 130¹⁹ from root *kṣubh* with *nis*³. In these cases *kṣ* at the beginning of the root at first became *gh* and *ch* respectively, to which then the final consonant of the preposition was assimilated.

c7.s1.t13.51

¹ In *hammiya* ‘hut’ Vin. I. 58²⁰, II. 146²⁹ = *harmya*, as well as in *abhikkhaṇam* ‘repeatedly’ M. I. 129²⁵ = *abhīkṣṇam* the Svarabhakti has taken place in the first stage.

c7.s1.t13.52

³ Wrongly E. Müller, PGr. p 9.

c7.s1.t13.53

§ 59. Some details: 1. The sound-groups *kṣn*, *kṣm*, *tsn* may be treated as *ṣn*, *ṣm*, *sn*,⁴ and thus according to § 50. 3-5 they may be changed into *ṇh*, *mh*, *nh*: *saṇha* ‘tender’ = *ślakṣṇa*; *tiṇha* ‘sharp’ D. I. 56³², JāCo. III. 89¹⁴ (besides *tikkha*, *tikhina*) = *tīkṣṇa*; *abhi-ṇham* ‘repeatedly’ Jā. I. 190⁵, A. V. 87²⁹. Pu. 48⁹, JāCo. II. 39¹¹ (besides *abhikkhaṇam*) = *abhīkṣṇam*; *abhiṇhaso* ‘continuously’ Th1. 2⁵ = *abhīkṣṇaśas*; *pamha* ‘eyelid’ Th2. 383, VvCo. 162²⁷ (besides *pakhuma*) = *pakṣman*; *juṇhā* (with cerebralisation, for *junhā*, as also Pkr. *joṇhā*) ‘moonlight’ = *jyotsnā* (cf. § 15.3).—2. *ts* is treated as in composition in *dosina* ‘clear’ D. I. 47¹⁰, JāCo. I. 509⁶ (besides *juṇha*) through **dossna*, **dosna* = *jyautsna* (cf. § 63.3) and in *kasina* ‘whole’ Abhp. 702 = *kṛtsna*. In the same manner is treated *cch* in *kasira* ‘difficult’ (besides *kiccha*; cf. *kicchena kasirena* ‘with much labour’ Vin. I. 195⁶, JāCo. I. 338²⁹ etc.) = *kṛcchra*.—3. Besides *uddham* ‘upwards’ = *ūrdhvam* there is also *ubbham*⁵ Th1. 163. Here *dhv* is treated like *dv* in composition (§58. 3).—4. From Skr. *dṛṣṭvā* Gerund of root *dṛś* ‘to see’ we get Pāli *disvā*, as also AMāg. *dissā*.¹

c7.s1.t13.54

² The sentences *ratyā ruccati caṇḍo*, *ratyo amoghā gacchanti* are quoted by Subhūti, Nāro. p. 38.

c7.s1.t13.55

⁴ Also in Pkr.: Pischel, § 312.

c7.s1.t13.56

⁵ Also in Pkr. *ubbha* besides *uddha*; Pischel, § 300

c7.s1.t13.57

15. Sporadical Phonological Aberrations in Sound-groups

c7.s1.t13.58

§ 60. One of these sporadical phonological phenomena in sound-groups is the representation of a sonant aspirate (cf. § 37) by *h* in the group *bhy*, which became *yh* through metathesis in *tuyhaṃ* = *tubhyaṃ*. But it is perhaps formed in analogy with *mayhaṃ* = *mahyam*. The group *dhv* became *vh* in the ending of the 2. Pl. Pres. Med. *-vhe* = *-dhve*. Moreover in some words *h* alternates with an aspirated media after a nasal. Thus in *-sumbhati* ‘strikes’ Jā. VI. 549⁹, VvCo. 212²², Pv. III. 1. 7, 8, Th2. 302 and *-sumhati* Jā. III. 185⁹, JāCo. III. 435²¹; *vambheti* ‘shames,’ *vambhanā* D. I. 90²⁵, M. I. 523³⁶, A. V. 150⁷, Vin. IV. 6¹, DhCo. IV. 38¹¹ and *vamheti*, *vamhanā* JāCo. I. 454²⁰, 356³. The roots underlying these forms seem to have been *sumbh*, *vambh*.² Besides *rundhati* ‘encloses’ JāCo. I. 409²⁰, *saṃnirundhati* M. I. 115³² we have *-rumbhati* JāCo. I. 62³¹, II. 341¹⁰ and *-rumhati* JāCo. II. 6²⁷, VvCo. 217²⁷. There was thus clearly a root *rubh* (*rumbh*) beside *rudh* (*rundh*).³ Similarly we have further *(saṃ)ūhanti* ‘removes’ D. II. 254¹⁸, M. II. 193⁸, *(saṃ)ūhata* Th1. 223, which stand for *(saṃ)ūdhanti*, *(saṃ)uddhanti*, *(saṃ)uddhata* from root *han* with *saṃ-ud*.

c7.s1.t13.59

¹ Pischel, § 354.

c7.s1.t13.60

² Not so R. O. Franke, WZKM. 8. 331.

c7.s1.t13.61

³ Cf. Māh., AMāg. *rumbhaī* and *rundhaī*; Fausböll, Ten Jātakas, p. 93; E. Kuhn, Beitr. p. 42; Pischel, PkrGr. § 507.

c7.s1.t13.62

§ 61. 1. Softening of tenues (cf. § 38) sometimes occurs after a nasal: *nighaṇḍu* ‘vocabulary’ D. I. 88⁵, A III. 223¹⁹ = *nighaṇṭu*; *gandha* ‘book’ Mhvs. 34. 66 besides *gantha* DhCo. I. 7¹⁸ = *grantha*; the interjection *handa* = *hanta*; *aḍḍhuḍḍha* ‘three and a half’ Vin. I. 34¹⁰ instead of *-uṭṭha*. But *puñjati* ‘rubs off’ JāCo. I. 318⁵ for *puñchati* JāCo. I. 392¹¹ etc. = *proñchati* is perhaps merely a graphic error. The group *kkh* was softened in *sagghasi* ‘you will be able to’ instead of *sakkhasi* Sn. 834 = *śakṣyasi*. On *leḍḍu* see § 62. 2. —2. Hardening of media (cf. § 39) is found in *bhiṅkāra* ‘jug’ (however with the frequent variant reading with *g*) D. II. 172²¹, Dpvs. 11. 32 = *bhṛṅgāra*; *tippa* ‘sharp’ Jā. VI. 507⁷, M. I. 10²⁹, Milp.

c7.s1.t13.63

148¹⁸ besides tibba Dh. 349, S. I. 110¹⁸ = tīvra¹; vilāka ‘slim’ Jā. IV. 19²⁹, V. 215¹⁹ through *vilakka, *vilagga = vilagna.

c7.s1.t13.64

§ 62. Unetymological aspiration or de-aspiration of sound-groups (cf. § 40) is not rare.—1. Unetymological aspiration: siṅghāṭaka (also AMāg. siṅhāḍaka) ‘road-crossing’ = śṛṅgāṭaka; Khaṇḍapura (name of a city) Sāsvs. 81²⁵ from Skanda (name of the war-god); pipphala ‘Ficus Religiosa’ Abhp. 909 = pippala; pipphalī ‘pepper’ Jā. III. 85²⁴, S. V. 79²², Vv. 43. 6 = pippalī. Such aspiration is often caused by r: acchi ‘light’ S. IV. 290²⁷ besides acci = arcis; koccha (§ 10) = kūrca.² Sometimes the r occupies the second position in the original sound-group: tattha ‘there’ (besides tatra, § 53.2) = tatra; sotthiya ‘Brahman’ (besides sottiya) = śrotriya; paripphoseti ‘sprinkles’ M. III. 243¹⁷ from root pruṣ with pari (paripphosaka ‘besprinkled all around’ D. I. 74³, M. II. 15¹³ etc.). Initial aspiration: khiḍḍā ‘play’ (beside kiḷā) through *khiḍā = krīḍā; phāsu(ka) ‘comfortable,’ if it is connected with Vedic prāśu.³—2. De-aspiration: lodda (§ 44) = lodhra, rodhra; babbu(ka) ‘cat’ Jā. I. 480¹ = babhru; bunda ‘root’ Abhp. 549 (with concomitant metathesis) = budhna (cf. bondi ‘body’ Pv. IV. 3. 32); muccati ‘coagulates’ Dh. 71, DhCo. II. 67²² = mūrcchati.⁴ The form milāca ‘forest-dweller’ Jā. IV. 291³ is perhaps a variant of milakkha (§ 34) and derived from *milacca, *milaccha. The form ludda(ka) ‘hunter’ DhCo. III. 31⁷, Mhvs. 28. 41 = lubdha(ka) is perhaps due to contamination with ludda ‘cruel’ (§ 15.4, 44). On aṭṭa = artha see § 64.1. Not infrequently the expected aspiration (according to § 51.1) does not take place in groups containing a sibilant. Thus, when the sibilant is first in the group: saccessati ‘will interrupt (?)’ A. IV. 343²⁵ from root saśc; kukku (§ 16.1 a) = kiṣku; catukka ‘a collection of four, crossing of roads’ Dāṭhāvs. 1. 58, JāCo. III. 44⁶, Milp. 1¹⁸ = catuṣka; nippesika ‘cheat’ (?) D. I. 8³⁰ = naiṣpeṣika; bappa ‘tear’ Abhp. 260 = bāṣpa⁵; maṭṭa ‘polished’ D. II. 133⁸ (besides maṭṭha Vv. 84. 17, DhCo. I. 25⁹) = mṛṣṭa; (abhi)vaṭṭa ‘he who has showered rain’ JāCo. I. 487²⁸, Milp. 176¹ (besides vaṭṭha, vuṭṭha)= vṛṣṭa; (pacc)osakkati (p. 97, f.-n. 4) from root svaṣk; takkara ‘thief’ Abhp. 522=taskara; samtatta ‘horrified’=samtrasta; as well as Indapatta (name of a city) (beside -pattha)=Indraprastha.¹ In leḍḍu ‘clod of earth’ M.I. 123³⁴, JāCo. III. 16¹⁸ through *leṭṭhu, *leṭṭu =leṣṭu² there is concomitant softening of the group (§ 61.1). In composition, the expected aspiration is missed in: niccala ‘immovable’=niścala; duccarita ‘bad deed’=duścarita; duttara ‘difficult to cross’=dustara; namakkāra ‘obeisance’ JāCo. II. 35¹ =namaskāra; also majjhatta ‘impartial’ JāCo. I. 300¹⁸, Mhvs. 21.14=madhyastha. Missing aspiration in sound-groups with the sibilant in second position: dhaṅka ‘crane’ Thl. 151, Jā. II. 208¹⁹, ‘crow’ S. I. 207²⁹ (verse), VvCo. 334³⁰ through *dhaṅkha=dhvāṅkṣa; ikka (§ 12.2)=ṛkṣa; Okkāka (p. 66, f.-n. 1) from Ikṣvāku; Takkasilā (name of a city)=Takṣaśilā. Expected aspiration missing in initial position: kuḍḍa³ ‘small’ D. II. 146¹², 169⁹, Jā. V. 102²⁴ (cf. § 64.1) besides khudda=kṣudra; culla, cūla (also AMāg., JMāh. culla) ‘small’ (for chulla)=kṣulla, which in itself is very probably a popular form of the word kṣudra.

c7.s1.t13.65

¹ On account of its meaning the form cannot be derived from tṛpra.

c7.s1.t13.66

² Similarly to be considered perhaps also dubbhati ‘injures’ Thl. 1129, S. I. 225¹⁰ etc. (dūbhin ‘malignant’ Jā. III. 73²⁸, mittadubbhika ‘treacherous’ Mhvs. 4. 1.), which I am inclined to connect not with druh—from it is derived mittaddu S. I. 225²¹ (verse), Mhvs. 4. 3—but with durv dūrvati (dhurv).

c7.s1.t13.67

³ Not so Pischel, PkrGr. § 208.

c7.s1.t13.68

⁴ Besides it muccheti ‘sounds, plays (on the lyre)’ JāCo. III. 188²² = mūrchayati.

c7.s1.t13.69

⁵ According to Mārkaṇḍeya in Pischel § 305 the form bappa in the sense of ‘tears’ is said to have been current also in Śaurasenī.

c7.s1.t13.70

¹ E. Müller, JPTS. 1888, p. 12.

c7.s1.t13.71

² Also in Pkr. leḍu etc. beside leṭṭhu; Pischel, § 804.

c7.s1.t13.72

³ The reading of the MSS. is however uncertain.

c7.s1.t13.73

§ 63. Change of consonant-classes in sound-groups (cf. § 41): 1. Guttural appears for palatal perhaps in bhisakka ‘physician’ M.I. 429⁴, A. III. 238⁶, Milp. 247¹¹ as against bhesajja ‘medicine’.—2. Cerebral appears for palatal in āṇā ‘order’ JāCo. I. 369²³ etc. (āṇāpeti, āṇatta, āṇatti)=ājñā (ājñāpayati⁴). But we have also aññā ‘highest knowledge’, aññātar ‘one who knows well’, aññāya Gerund from root jñā with ā. Similarly to be explained also paṇṇarasa, paṇṇuvīsa, paṇṇāsa (§ 48.2).—3. Dental appears for palatal in uttiṭṭha (Skr. ucchiṣṭa) ‘rest of a meal’ Milp. 213 f. (uttiṭṭhapatta ‘alms-bowl with grains of food attaching to it⁵’), a dialectical side-form of ucchiṭṭha (§ 57). For initial jy there is d instead of j in dosina (§ 59.2)=jyautsna.

c7.s1.t13.74

⁴ Similarly Pkr. āṇā āṇāvei.

c7.s1.t13.75

⁵ VT. I. (=SBE. XIII), p. 152, f.-n. 1. Wrong interpretation in SBE. XXXV, p. 4.

c7.s1.t13.76

§ 64. Cerebralisation of dental-groups is the most frequent case of the change of place of articulation (cf. § 42). 1. Under the influence of r: thus rt, rd, rdh become ṭṭ, ḍḍ, ḍḍh: aṭṭa ‘pained’ Thl. 1106, Vin. I 121⁹, JāCo. I. 265⁷ = ārta; kevaṭṭa ‘fisherman’ D. I. 45²⁰, JāCo. III. 171¹² = kaivarta; chaḍḍeti ‘throws away’ = chaṛdayati; vaḍḍhati ‘increases’ (with numerous derivatives) = vardhate. Besides vuddha, vaddha ‘grown up, old’ Jā. I. 177¹, D. I. 90⁴ there is buḍḍha, vuḍḍha (§ 46.1); beside vuḍḍhi (§ 12.4) there is vaḍḍhi. In aṭṭa ‘law-suit’ (de-aspiration according to § 62.2) as opposed to aṭṭha ‘property etc.’¹ = artha the change in sound has been accompanied by a change in meaning. Forms both with tt and ṭṭ are to be found side by side among the derivatives of the root vart: always vaṭṭati when it signifies ‘it is proper’, but vattati signifies ‘becomes, originates etc.’; similarly vaṭṭa ‘round, circle’, but vaṭṭa ‘duty, responsibility’, both = vṛtta. The spelling with ṭṭ is preferred also in vaṭṭati in composition, when the sense of rolling is emphasised: āvaṭṭati pavaṭṭati ‘rolls hither and thither’ D. II. 140¹, JāCo. VI. 504¹⁵ etc. Hence also āvaṭṭa ‘turn, whirl’, saṃvaṭṭa ‘overturning, annihilation’, vivaṭṭa ‘renewal (of an aeon)’ D.I. 14²⁶, Vin. III. 4²⁵. The group nt became ṇṭ under the influence of an original ṛ in vaṇṭa ‘stalk’ Jā VI. 537²², D. I. 46¹³, tālavaṇṭa ‘fan’ Vin. II. 137⁷, JāCo. I. 265²⁹ = vṛnta, tālavṛnta. Cf. also alla ‘damp’ D. II. 332²⁶, Vin. I. 109³ etc. through *aḍḍra, *aḍḍa = ārdra.—2. Under the influence of a sibilant: ṭhāti, ṭhahati ‘stands’, ṭhāna ‘place’, saṃṭhāna ‘figure’, paṭṭhāya ‘beginning from,’ kūṭaṭṭha ‘standing firm as a rock’ etc. from root sthā, sthāna, saṃsthāna, prasthāya, kūṭastha etc.—3. Irregular cerebralisation is found in jaṇṇuka ‘knee’ (perhaps this form should be corrected) JāCo. VI. 332¹⁶ besides jannu(ka) = jānu, as well as kaviṭṭha (§ 38.5) besides kapittha. Cf. also the cerebrals in daḍḍha (§ 42 3) = dagdha.

c7.s1.t13.77

¹ On the other hand the usual spelling is aṭṭhakathā ‘commentary.’

c7.s1.t13.78

16. Metathesis in Sound-groups and Loss of Syllable through Haplology.

c7.s1.t13.79

§ 65.1. Metathesis in sound-groups takes place in combinations of h with nasal or semi-vowel (§ 49, 60), and further in combinations of sibilant with nasal, which become nasal + h (§ 50). The sibilant is retained, inspite of the metathesis, in ramsi ‘ray’ = raśmi. On the metathesis of ry into yr with concomitant insertion of svarabhakti see § 47.2; for gumba = gulma see § 51.5; for bunda = budhna see § 62.2. Unique is the case of gadrabha ‘ass’ D. II. 343¹⁴, JāCo. II. 96²⁴ as opposed to gardabha.

c7.s1.t13.80

2. Haplology and the loss of syllable caused by it is in evidence in aḍḍhatiya (for *aḍḍhatatiya) ‘three and a half’ JāCo. II. 93⁴ (also aḍḍhateyya Vin. I. 39²⁴, DhCo. I. 95⁹); viññāṇañcāyatana (for viññāṇānañcāy-) ‘sphere of infinite knowledge’ M. III. 106¹³ etc. As sporadic cases may be mentioned: pavissāmi (for pavississāmi) ‘I shall enter’ Cp. I. 9.56, JāCo. II. 68²⁰, sossi (for sossasi) ‘you will hear’ Jā. VI. 423⁸, vipassi (for vipassasi) ‘you apprehend’ Th2. 271, gacchisi (for gacchissasi) ‘you will go’ Th1. 356. Cf. also sakkhī Jā. V 116⁵ for *sakkhisi (Co: sakkhissasi), āsādum Jā. V. 154¹⁹ (Co: āsāditum). It is however impossible to decide whether the optatives of ṭ-roots such as jeyyam from ji, neyyam from nī should be regarded as “haplologies” for *jayeyyam, *nayeyyam.

Linked texts (58)