Pramāṇa · Search · About · API

Pali Literature and Language c7: Part II. Grammar of Pāli

Pāli Literature and Language c7

Language
English

Contents

Part II. Grammar of Pāli

c7.2

Literature (general, grammars, dictionaries, chrestomathies).

c7.3

E. Burnouf et Chr. Lassen, Essai sur le Pâli. Paris 1826.

c7.4

E. Burnouf, Observations grammaticales sur quelques passages de l'Essai sur le Pâli de Burnouf et Lassen. Paris 1827.

c7.5

B. Clough, Compendious Pali Grammar with a copious Vocabulary in the same Language. Colombo 1824.

c7.6

J. Alwis, Lecture on the Buddhist Scriptures and their Language, the Pali (1861). Reprinted in JPTS. 1883, p. 39 ff.

c7.7

J. Minayeff, Grammaire Pâlie, traduite par St. Guyard. Paris 1874.—Pāli Grammar, a phonetic and morphological Sketch of the Pāli Language, with an introductory Essay on its Form and Character by J. M., 1872 ; transl. from Russian into French by M. St. Guyard, 1874. Rendered into English by Ch. G. Adams, 1882.

c7.8

Fr. Müller, Beiträge zur Kenntnis der Pāli-Sprache, I. II. III. Stzber. d. Wiener Ak. d. W., phil.-hist. Cl. Vol. 57, 1867, p. 7 ff.; p. 243 ff. ; Vol. 60, 1868, p. 533 ff.

c7.9

R. C. Childers, A Dictionary of the Pali Language. London 1875 (4. ed. 1909).

c7.10

E. Kuhn, Beiträge zur Pāli-Grammatik. Berlin 1875.

c7.11

V. Trenckner, Pāli Miscellany, Vol. I. London 1879. Cf. also Trenckner's Notes to the Milinda-pañha, revised and edited by D. Andersen, JPTS. 1908, p. 102 ff.

c7.12

A. Torp, Die Flexion des Pāli in ihrem Verhältnis zum Sanskrit. Christiania 1881.

c7.13

O. Frankfurter, Handbook of Pāli, being an elementary Grammar, a Chrestomathy, and a Glossary. London and Edinburgh 1883.

c7.14

E. Müller, A simplified Grammar of the Pali Language. London 1884.

c7.15

Morris, Notes and Queries, JPTS. 1884, p. 69 ff. ; 1885, p. 29 ff. ; 1886, p. 94 ff. ; 1887, p. 99 ff. ; 1889, p. 200 ff. ; 1891/3, p. 1 ff.

c7.16

E. Müller, A Glossary of Pāli Proper Names, JPTS. 1888, p. 1 ff.

c7.17

Morris, Contributions to Pali Lexicography, Academy 1890 91.

c7.18

V. Henry, Précis de Grammaire Pâlie, accompagné d'un choix de textes gradués. Paris 1894.

c7.19

Tha Do Oung, Grammar of the Pali Language (after Kaccāyana). Vol. I: Sandhi, Nāma and Kāraka, and Samāsa ; Vol. II: Taddhita, Kita, Uṇādi, Ākhyāta, Upasagga and Nipāta particles ; Vol. III: Dictionary of Pali word-roots ; Vol. IV: Chandam etc. Akyab 1899-1902.

c7.20

M. Bode, Index to Pāli Words discussed in translations, JPTS. 1897-1901, p. 1 ff.

c7.21

H. H. Tilbe, Pāli Grammar. Rangoon 1899 (Student's Pāli Series).

c7.22

J. Takakusu, A Pāli Chrestomathy, with Notes and Glossary giving Sanskrit and Chinese Equivalents. Tokyo 1900.

c7.23

D. Andersen, A Pāli Reader with Notes and Glossary. Part I: Text and Notes ; Part II: a Pāli Glossary including the words of the Pāli Reader and of the Dhammapada. Copenhagen, London and Leipzig 1901, 1907.

c7.24

R. O. Franke, Pāli und Sanskrit, in ihrem histor. und geogr. Verhältnis auf Grund der Inschriften und Münzen. Strassbourg 1902.

c7.25

H. H. Tilbe, Pāli First Lessons. Rangoon 1902 (Student's Pāli Series).

c7.26

J. Gray, Elementary Pāli Grammar or Second Pāli Course. Calcutta 1905.

c7.27

J. Gray, First Pāli Delectus or Companion Reader to the Pāli Course. Calcutta 1905.

c7.28

E. Windisch, Über den sprachlichen Charakter des Pali. Actes du XIVe Congrès Internat. des Orientalistes 1, Sect. I, p. 252 ff. Paris 1906.

c7.29

Ch. Duroiselle, A Practical Grammar of the Pāli Language. Rangoon 1906.

c7.30

Ch. Duroiselle, School Pāli Series. I : Reader ; II : Vocabulary. Rangoon 1907.

c7.31

K. P Johansson, Pāli-Miszellen; Le Monde Oriental 1907-08, p. 85 ff.

c7.32

Nyānatiloka, Kleine systematische Pāli Grammatik. Breslau 1911 (Veröffentlichungen d. D. Pāligesellschaft 5).

c7.33

K. Seidenstücker, Elementargrammatik der Pāli Sprache. Leipzig 1916.

c7.34

Sri Buddhatta Sthavira, The New Pāli Course, Part 1. Colombo 1937

c7.35

Valuable preliminary work for a Pāli dictionary is contained also in: St. Konow, Lexicographical Notes, Words beginning with H, JPTS. 1907, p. 152 ff. ; St. Konow and D. Andersen, Lexicography, Words beginning with S, JPTS. 1909, p. 1 ff. ; T. W. Davids and W Stede, The Pali Text Society's Pali-English Dictionary, London, PTS. 1921 ; Dines Andersen and Helmer Smith, A Critical Pāli Dictionary.

c7.36

Other monographs on Pāli Grammar and Vocabulary will be mentioned in proper places in the following.

A. Phonology

1. Sound-system and Accent

c7.s1.t1.1

§ 1. For the writing of Pāli there are used in the Orient various scripts : in Ceylon the Sinhalese, in Burma the Burmese, in Siam the Kamboja script. The Bangkok edition of the Tipiṭaka is printed in Siamese letters.

c7.s1.t1.2

§ 2. The sound-system of Pāli consists of the following:

c7.s1.t1.3

1. Vowels : *a, ā, i, ī, u, ū, e, o*, as well as the nasal vowels *aṃ, iṃ, uṃ*.

c7.s1.t1.4

2. Consonants : Gutturals : *k, kh, g, gh, ṅ*. Palatals : *c, ch, j, jh, ñ*. Cerebrals : *ṭ, ṭh, ḍ, ḍh, ṇ*. Dentals : *t, th, d, dh, n*. Labials : *p, ph, b, bh, m*. Liquids : *r, l, ḷ, ḷh*.¹ Semi-vowels : *y, v*. Sibilant : *s*. Aspiration : *h*.

c7.s1.t1.5

¹ There is no special sign for this sound. It is usually indicated by the ligature *ḷ+h*.

c7.s1.t1.6

Here should be noticed : 1. The vowels *e, o* are of middle length, in closed syllables they are short and in open syllables they are long.—2. The sign of nasalisation (*ṃ*) corresponding to the anusvāra and anunāsika of Sanskrit is called Niggahīta by Pāli grammarians. In Ceylon at present the Niggahīta is pronounced as a guttural nasal.—3. The consonant *ḷ* stands for intervocalic *ḍ*, likewise *ḷh* for *ḍh*. In the MSS. *l* and *ḷ* are always confused. Some spellings seem to be arbitrary. Thus *kāḷa* ‘black’ is mostly written with *ḷ*, although it is equivalent to Sanskrit *kāla*, probably to distinguish it from *kāla* ‘time.’ Surely *ḷ* should be written everywhere where an etymological connection with cerebrals is apparent.—4. The *h* is, where it stands alone, a consonant. In combination with *y, r, l, v*, or with the nasals, it appears to have been pronounced in a particular manner. The grammarians in this case call it *orasa* ‘spoken in the breast.’²

c7.s1.t1.7

² Minayeff, P Gr. p. 2.

c7.s1.t1.8

§ 3. In its sound-system Pāli stands in Prākritic stage when compared with Old Indian. Also Prākrit has given up the vowels ṛ, ḷ, as well as the diphthongs ai, au, in all the dialects, and the vowel ṛ in most of the dialects excepting Apabhraṃśa. Moreover Prākrit like Pāli possesses the cerebral consonant ḷ and the middle-length vowels e, o. Most Prākrit dialects have like Pāli only the dental s; the cerebral ṣ is found in none of them, and the palatal ś is absent in most.¹

c7.s1.t1.9

¹ Pischel, PkrGr. § 45.

c7.s1.t1.10

The original diphthongical character of e, o is brought to light by the circumstance that in Sandhi a + i becomes e and a + u becomes o. Cf. *macchassevodoke gataṃ* = *macchassa iva udake go* ‘like the course of fish in water’ Jā. 1.295⁸. There was even a tendency to use them (instead of ai, au) as *vṛddhi*-vowels. Thus are found forms originated inside Pāli, such as *tepiṭaka* ‘devoted to the Tipiṭaka’ DhCo. III. 384,¹⁸ Mhvs. 5.84 from *tipiṭaka*, and *opadhika* ‘relating to the substratum of existence’ Vv. 34.21, 24 = S. I. 233¹⁵·²¹ from *upadhi*. This strengthening into e, o may take place even in those cases where originally there was no i, u. Thus *pothujjanika* ‘relating to common people’ Vin. I.10¹² from *puthujjana*, where the u corresponds to Skr. ṛ (*pṛthagjana*). Cf. *sosānika* ‘relating to cemetery’ Pu. 69²⁷, DhCo. I. 69⁴ from *susāna* = *śmaśāna*. In *gelañña* ‘illness’ D.II. 99¹⁶, JāCo. II. 31²⁰ from *gilāna* = Skr. *glāna*, and in *sovatthika* ‘bringing prosperity’ Vv. 18.7 from *suvatthi* = Skr. *svasti*, e and o are the strengthened forms of the svarabhakti-vowels i, u originated within Pāli. Cf. also *veyyāvacca* ‘rendering service’ Vin. I.28²⁰ from *viyāvaṭa* = Skr. *vyāpṛta*; *veyyākaraṇa* ‘answering’ D. I. 51⁵ from *viyākaroti* = Skr. *vyākaroti*.

c7.s1.t1.11

In the same way *a* has been strengthened into *ā* within Pāli: *sākhalya, lla* ‘friendliness’ M.I. 446¹⁹, Jā. IV. 57⁵ from *sakhila*; *bhākuṭika* ‘with frowning eye-brows’ Vin. III. 181⁶ from *bhakuṭi* = Skr. *bhrakuṭi* (or *bhṛkuṭi*).

c7.s1.t1.12

§ 4. Nothing has been handed down to us about the nature of Pāli accent. It is, however, improbable that the ancient Indian accent was still in force. Rather, as Jacobi has suggested also for Prākrit,² the Sanskritic accent was the rule in Pāli.³ This is suggested by the changes of vocalism in Pāli, such as the weakening of a vowel after the accented syllable (in the Skr. form) or its strengthening in the main tonic syllable. Cf. § 19 ff.

c7.s1.t1.13

² Different view of Pischel, KZ. 34.568 ff., 35.40 ff.; PkrGr. § 46. Against him Jacobi, ZDMG. 47.574 ff., KZ. 35.578 ff. Cf. Grierson, ZDMG. 49.395 ff. ; Michelson, IF. 23.231.

c7.s1.t1.14

³ On this cf. Jacobi, ZDMG. 47.574.

2. The Law of Mora

c7.s1.t2.1

§ 5. In Pāli, as generally in Middle Indian,¹ a syllable can contain only one mora or two moras but never more. The syllable is thus either (1) open with short vowel (one mora), or (2) open with long vowel (2 moras), or (3) closed with short vowel (2 moras). Every syllable with a nasal vowel is considered as closed. Long nasal vowels do not occur. Due to this law, where Skr. has long vowel before double-consonance (i.e. in closed syllable), Pāli has there either (a) short vowel before double-consonance² or (b) long vowel after the following double-consonance simplified.

c7.s1.t2.2

¹ R. O. Franke, P. und Skr. p. 90 f.

c7.s1.t2.3

Examples of (a) : jiṇṇa ‘old, exhausted’ = jīrṇa. Likewise maṃsa ‘flesh’ = māṃsa, and the final in nadiṃ ‘the river’ (acc.) = nadī. The vowels e, o are short in such cases: semha ‘catarrh’ = śleṣman ; oṭṭha ‘lip’ = oṣṭha.—Examples of (b) : lākhā ‘lac’ = lākṣā ; dīgha ‘long’ = dīrgha. In case of the vowels e, o, the orthography in the mss. varies not infrequently, such as apekkha and apēkhā ‘expectation’ = apekṣā ; upekkhā and upēkhā ‘indifference’ = upekṣā ; vimokkha and vimōkha ‘deliverance’ D. II.70²⁸, A. IV. 306¹¹ = vimokṣa.

c7.s1.t2.4

§ 6. The Law of Mora has led to various changes.³ Due to it: 1. there can be in Pāli long vowel before single consonant where Skr. has short vowel before double-consonance: sāsapa ‘mustard seed’ (instead of *sass-) Dh. 401, S. II. 182¹⁷, DhCo. I. 107³ = sarṣapa ; vāka ‘bark’ (instead of *vakka) D. I.167², Vin. III. 34²⁸, JāCo I. 304² = valka ; nīyāti ‘goes away’ = niryāti.⁴ 2. Pāli shows short vowel before double-consonance where originally there was long vowel before a single consonant: abahati ‘draws out’ Th. 1. 162, 1007, Sn. 334, Jā. II. 95⁵ = ābṛhati; niḍḍa (sic!) ‘nest’ Dh. 148 (AMāg. neḍḍa) = nīḍa; udukkhala ‘mortar and pestle’ Vin. III. 6²⁶, D. II. 34¹⁶, Jā-Co. I. 502²⁰ (AMāg. udukkhala beside udūhala) = udūkhala ; kubbara ‘ pole of a carriage ’ Vv. 64.2, A. IV. 191⁶, VvCo. 269⁵ =kūbara, kūvara ; pēttika ‘ fatherly ’ D. II. 232¹⁰, Vin. III. 16³⁵, JāCo. II. 59¹³ for *pētika = paitṛka (after which has been latterly formed mettika). Cf. also mahabbala, mahapphala = mahābala, mahāphala.— 3. As the short nasal vowel has two moras like the long, a nasal not infrequently appears in the place of a pure long vowel¹ and vice versa : mamkuṇa ‘ bug ’ JāCo. III. 423¹³ instead of *māk-, *makk- = matkuṇa ; saṃvarī ‘ night ’ D.III. 196²² (verse), Jā. IV. 441⁶ instead of *sāv-, *sabb- = śarvarī ; suṃka ‘ toll ’ (AMāg. ussuṃka) instead of *sūka, *sukka = Skr. śulka ; ghaṃsati ‘ rubs ’ instead of *ghās-, *ghass- = gharṣati ; vidaṃsentī ‘ she who shows ’ Th2. 74 instead of *vidās-, *vidass- = vidarśayanti. On the other hand : vīsati vīsaṃ ‘ twenty ’ = viṃśati ; sīha ‘ lion ’ = siṃha ; sārambha ‘ audacity ’ (beside saṃrambha Dāṭhāvs. 4.34) = saṃrambha, and other words with saṃ-.

c7.s1.t2.5

² In P. sutti, therefore, there have coincided, e.g., Skr. śukti ‘oyster-shell’ (Vin. II. 106¹¹,¹²) and sūkti ‘good speech’ (Saddhammopāyana 340, 617).

c7.s1.t2.6

³ For analogous phenomena in Pkr. see Pischel, PkrGr. § 62-65, 90, 74-76.

c7.s1.t2.7

⁴ Cf. also svatanāya ‘for next day’ (in Buddhistic Skr. śvetanāya etc.) from Skr. śvastana. Johansson, Monde Oriental, 1907-08, p. 106 ff.

c7.s1.t2.8

¹ The tendency towards nasalisation is often in evidence in mss. : naṃgara instead of nagara, gaṃchi (gañchi) instead of gacchi etc. Cf. uḷumpa ‘ raft ’ D.II. 89¹⁵ = uḍupa.

c7.s1.t2.9

§ 7. Sometimes a long vowel is retained before double-consonance. Thus particularly in contractions, as in sājja = sā ajja Th2.75, yathājjhāsayena ‘ according to desire ’ JāCo. IV. 243⁸. Further, in derivatives such as dussīlya from dussīla. Cf. bālya ‘ stupidity ’ DhCo. II. 30³, variant reading of balya. These are evidently cases of learned orthography. There are also sporadical cases like dābbī (name of a plant) Abhp. 586 = dārvī ; dātta ‘ sickle ’ Abhp. 448 = dātra (none of these, forms, however, is quotable from texts).² Quite frequent is svakkhāta ‘ well proclaimed ’ Vin. I. 12²⁴ etc. = su-ākhyāta.

c7.s1.t2.10

² Childers gives also ājjava ‘ straightness ’ beside ajjave. Sākya Bodhivs. 22¹¹, 27²⁹ instead of Sakya (according to the Colombo edition), Sakka or Sakiya is perhaps a wrong reading. The form Sākiya also is not quite correct.

c7.s1.t2.11

§ 8. It is also due to the action of the Law of Mora that even in the case of the separation a consonant-group by a Svarabhakti vowel (§ 29 ff.) a long vowel preceding the consonant-group is regularly shortened.³ The two one-mora syllables in these cases represent one two-mora syllable : suriya ‘ sun ’ (instead of *sūyya) = sūrya ; but sūriya ‘ heroism ’ Jā. I.282¹⁷, purposely differentiated from the preceding ;

c7.s1.t2.12

³ As a rule it is different in Pkr., cf. AMāg. sūriya vīriya etc But still AMāg., JMāh. veruliya = vaidūrya (P. veḷuriya). In Pāli we have sīliya, Jā. III. 7¹ only metri causa.

c7.s1.t2.13

pakiriya gerund of pakirati ‘loosens (the hair)’ = prakirya. In words like cetiya = caitya and Moriya = Maurya we have therefore to consider e and o as short vowels.—The insertion of the svarabhakti-vowel however does not disturb the length of a following vowel: thus gilāna ‘ill’ = glāna. Only in the case of the originally monosyllabic words itthī ‘woman’ = strī, sirī ‘fortune’ = śrī and hirī ‘modesty’ = hrī does this law act in some measure.¹ In compounds these words have short vowel² : itthiratana ‘jewel of a woman’ D. I. 89³, hirimana ‘of modest disposition’ D. II. 78⁴ etc. Also sirimant, hirimant and occasionally sassirika,³ ahirika etc.

c7.s1.t2.14

¹ On the same words in Pkr. see Pischel, PkrGr. § 98, 147.

c7.s1.t2.15

³ sassirika JāCo. I. 504[^29]. Smp. 300[^23], nissirika JāCo. VI. 456[^1].

3. The Vowels ĕ ĭ ŏ

c7.s1.t3.1

§ 9. Occasionally e appears for a before double-consonance: pheggu ‘empty, worthless’ M. I. 194⁵, S. IV. 168² etc. = phalgu ; seyyā ‘bed’ = śayyā.⁴ The word ettha ‘here’ is probably not = atra, but=*itra, Av. iθra,⁵ and is therefore to be classed with § 10.2. In the same way heṭṭhā ‘below’ is derived not from adhasthāt, but from an *adheṣṭhāt⁶ as is shown by the cerebrals -ṭṭh-.

c7.s1.t3.2

² Lengthening takes place only metri causa, thus sirimant Thl. 94, siridhara Mhvs. 5.16.4.

c7.s1.t3.3

⁴ There might have been intermediate forms like *phiggu, *siyyā (Pkr. sejjā with the frequent variant reading sijjā). Cf. mimjā ‘marrow’ (§6.3) = majjā. For Pischel’s explanation see PkrGr. § 101. Jacobi, KZ. 35.573 ascribes the i in sijjā to the influence of the palatal.

c7.s1.t3.4

⁵ So already Lassen, Instit. Linguae Pracriticae 129. Cf. E. Kuhn, Beitr. p. 21. Pischel, § 107 thinks of connection with Ved. itthā I would however like to point out the v. l. ubhayettha of ubhayattha DhCo. I. 29[^13].

c7.s1.t3.5

⁶ Pischel § 107. Not so Johansson, IF. 3.218 ; Monde or. 1907, 93 and Wackernagel, KZ. 43.293.

c7.s1.t3.6

§ 10. 1. The vowels i, u are lengthened in the flexional endings -īhi, -ūhi and -īsu, -ūsu of i- and u-declensions (§ 82).—2. Not infrequently i and u become ĕ and ŏ before double-consonance⁷: Veṇhu (J Māh. Viṇhu) D. II. 259²² = Viṣṇu⁸ ; nekkha ‘gold ornament’ Sn. 689, A. I. 181¹³ (beside nikkha Vin. I. 38¹⁶) = niṣka ; koccha ‘bundle, ball, fleshy part’ Vin. II. 149³², 266²³ = kūrca (cf. §62.1) ; oṭṭha ‘camel’ M. I. 80¹³, Vin. IV.

c7.s1.t3.7

⁷ Frequently also in Pkr., Pischel, § 119, 122, 125.

c7.s1.t3.8

⁸ Also in S. I. 52[^10] should be read Veṇhu instead of Veṇḍu.

c7.s1.t3.9

7¹⁶ = uṣṭra ; vokkamati ‘ gets deranged ’ D. I. 230¹⁴, M. III. 117²³, JāCo. I. 23²¹ = vyutkramati ; Okkāmukha (proper name) = Ulkāmukha.¹ In words like rāmaṇeyya ‘ charming ’ = rāmaṇīya, ‘dakkhiṇeyya ‘ worthy of veneration ’ = dakṣiṇīya, an intermediate stage with *-iyya has to be imagined.—The change into e takes place even in the case of such i as is derived from ṛ. Thus in the verb gheppati ‘ grasps, ’ if it is derived from *ghippati < *ghṛpyati like Pkr. gheppaï as Pischel (PkrGr. § 107) has suggested.—The double-consonance following after ĕ, ŏ may be secondarily simplified according to the Law of Mora with concomitant protraction of e, o: Uruvelā (place-name) through *-vellā, *-villā = Uruvilvā ; oja ‘ strength ’ D. II. 285¹⁰, M. I. 124³², DhCo. I. 107¹⁶ through *ojjā, *ujjā = ūrjā. In vihesati ‘ injures, insults ’ Ud. 44³⁰, 45⁸ (beside vihiṃsati) the intermediate steps *vihisati, *vihissati *vihĕssati have to be imagined. Out of original ṛ this i,e has been developed in paligedha ‘ desire ’ A. I. 66¹⁰, palyedhin A. III. 265⁷ through *-gĕddha, *-gĕddhin, *-giddha, *-giddhin = gṛddha, gṛddhin.

c7.s1.t3.10

¹ But always ukkā ‘ flame, meteor ’ D. I. 49³¹, JāCo. IV. 290²³ : D. I. 10¹⁷ = ulkā. The name Okkāka = Ikṣvāku, which is connected by the Buddhists with ikṣu ‘ sugarcane ’ is in fact derived from *Ukkhāka (*ukkhu side-form of ucchu). The analogical influence of Okkāmukha was also effective.

c7.s1.t3.11

§ 11. Intermediate stages with double-consonance have to be imagined where in open syllable ī, ū have become e, o: edi, edisa (erisa), edisaka, edikkha (erikkha) ‘ such one ’ Sn. 313, Vin. I. 195¹¹, Mhvs. 5.56 (beside īdisa etc. Mhvs. 5.93) through *iddi, *ĕddi etc. = īdṛś, īdṛśā, īdṛkṣa²; āveḷā ‘ garland ’ Vv. 36.2, JāCo. I. 444⁶, 501²⁹ (Pkr. āmeḷā) through *āvĕḍā, *āvĕḍḍā *āviḍḍā, = āpīḍā³; gaḷoci (a plant ; Pkr. gaḷoi) Abph. 581 through *gaḷŏcci, *gaḷucci = guḍūcī; jambonada ‘ gold ’ Dh. 230, Vv. 84. 17 through *-ŏnnudu, *-unnada = jambūnada. Mahesi ‘ queen ’ from mahiṣī⁴ is remarkable.

c7.s1.t3.12

² Cf. AMāg. eddaha beside erisa etc. Pischel, PkrGr. § 121, 122. Pāli has however only kīdi, kīdisa etc.

c7.s1.t3.13

³ Pischel, § 122, 248.

c7.s1.t3.14

⁴ E. Kuhn, Beitr. p. 24 suspects influence of mahesi ‘ wise man. ’ Cf. here also gahetvā, netvā which have been wrongly explained by Minayeff, PGr. § 16, p.

4. Representation of the Vowels ṛ, ḷ

c7.s1.t4.1

§ 12. The vowel ṛ is represented by a, i, u in P.—even in initial position, which is not the case in Pkr.¹ The quality of the vowel is largely influenced by the neighbouring sounds (cf. § 16); *u* appears mostly after labials.—1. *a* stands for *ṛ*: *accha* ‘bear’ Jā. VI. 507⁵, JāCo. VI. 538²¹ =*ṛkṣa*; *pasada* ‘spotted antelope’ Jā. VI. 587³¹ =*pṛṣata*; *vaka* ‘wolf’ Sn. 201, JāCo. I. 336¹⁷ (verse)=*vṛka*; *hadaya* ‘heart’ =*hṛdaya*.—2. *i* stands for *ṛ*: *ikka* ‘bear’ Jā. VI. 538¹ (Co.=*accha*)=*ṛkṣa* (cf. § 62.2); *iṇa* ‘debt’ (AMāg. *aṇa*) Sn. 120, D. I. 71³¹, JāCo. I. 321²⁰ =*ṛṇa*; *vicchika* ‘scorpion’ D. I. 9⁸, Vin. II. 148⁹ =*vṛścika*; *sipāṭikā* ‘seed-house’ M. I. 306² =*sṛpāṭikā*.—3. *u* stands for *ṛ*: *uju* or *ujju* (Jā. VI. 518⁸) ‘straight’ =*ṛju*; *usabha* ‘bull’ Dh. 422, S. I. 75³², JāCo. I. 336²⁰ =*ṛṣabha* (beside *vasabha*=*vṛṣabha*); *pucchati* ‘asks’ =*pṛcchati*; *muḷāla* JāCo. I. 100⁷ and *muḷāli* Jā. VI. 530¹⁶ ‘lotus-stalk’ =*mṛṇāla*; *pāvusa* ‘rainy season’ Thl. 597 f., Jā. VI. 202²⁷ =*prāvṛṣa*.—4. Sometimes representation varies: cf. above, beside *accha* the dialectical form *ikka* ‘bear’; *vṛddhi* has been differentiated into *vaḍḍhi* ‘blessing’ and *vuddhi* ‘growth’; *mṛga* has been differentiated into *maga* ‘animal’ Sn. 275, Thl. 958, S. I. 199²¹ and *miga* ‘gazelle’ passim.² Beside *iṇa* there is *anaṇa* ‘debtless’ Th2. 2, M. II. 105¹⁶ and *sāṇa* (=*sa-aṇa*) ‘indebted’ M. III. 127⁷,⁹, S. II. 221¹, probably through vowel-assimilation³ (but *sayiṇa* or *saiṇa* Mhvs. 36.39). Beside *kaṇha* ‘black’ =*kṛṣṇa* there is found as variant reading *kiṇha* D. I. 90¹⁵, S. IV. 117⁶. Skr. *pṛthivī* appears as *pathavī*, *paṭhavī*, *puthavī*, *puthuvī*, *puṭhuvi*; here the region from where the MSS. are derived is be taken into consideration: *pathavī*, e.g., is the orthography of the Burmese MSS. Moreover cf. *pitughātaka*, *mātughātaka* ‘parricide, matricide’ Vin. I. 88²⁰ with *pitipakkhato*, *mātipakkhato* ‘from paternal side, from maternal side’ etc.

c7.s1.t4.2

¹ In Pkr. initial *ṛ* usually becomes *r*+vowel, Pischel PkrGr. § 56. In P. we have only *iruveda* Dpvs. 5. 62. or *irubbeda* DCo. I. 247¹⁹= *ṛgveda*. This word is however an artificial formation.

c7.s1.t4.3

² The form *miga* is found in its general meaning ‘animal’ in *sākhāmiga* ‘monkey’ Jā. III. 98¹⁴, *migacakka* ‘animal magic’ D. I. 9¹⁰, DCo. I. 9⁴.

c7.s1.t4.4

³ Trenckner, Notes p. 76 (JPTS. 1908, p. 129).

c7.s1.t4.5

§ 13. In some cases the *ṛ*-vowel becomes consonant: *brahant*, *brahā*° ‘big’ Thl. 31, Jā. III. 117²³ =*bṛhant*; *brūheti* ‘devotes himself to a cause’ Dh. 285, Ud. 72¹⁷ (verse), JāCo. I. 289¹¹=*bṛṃhayati*, *vṛṃhayati*. *Vṛ* becomes *ru* in *rukkha* ‘tree’ =*vṛkṣa*⁴ and in *pāruta* ‘ covered, concealed ’ Th1. 153, S. I. 167²⁷, JāCo. I. 347⁸ = prāvṛta and apāruta ‘ opened up ’ Vin. I. 7⁴ (verse), D. II. 217¹⁵, JāCo. I. 264⁴ = apāvṛta.

c7.s1.t4.6

⁴ Pischel, PkrGr. § 320 connects *rukkha* with Skr. *rukṣa*. Cf. Wackernagel, Altind. Gr. I, § 184 b. The side-form *rakkha* is found in Jā. III. 144¹⁵.

c7.s1.t4.7

§ 14. The vowel ḷ is represented by u: kutta¹ ‘ clipped ’ D. I. 105 ⁹, DCo. 1. 274¹⁷ = kḷpta.² In the same way also kuttaka (a kind of woollen cover) D. I. 7¹⁰, Vin. I. 192⁸, II. 163²⁴, perhaps ‘ shorn cover.’³ Further kutta, kutti ‘ behaviour, procedure ’ = kḷpta, kḷpti in itthikutta, purisakutta A. IV. 576¹⁹, JāCo. I. 296²¹ etc., saṃnatavirakutti Jā. V. 215¹⁶ where kutta = kappana, just as mata = maraṇa. In DhsCo. 321¹¹ it is explained by kiriyā.

c7.s1.t4.8

¹ E. Leumann, GGA. 1899, Nr. 8, p. 594.

c7.s1.t4.9

² Cf. Rhys Davids, Dial., I. 130, f.-n. 2.

c7.s1.t4.10

³ Differently explained by Buddhaghosa as quoted by Rhys Davids and Oldenberg, VT. II. 27, f.-n. 4.

5. Diphthongs and their Representation

c7.s1.t5.1

§ 15. The diphthongs e, o are as a rule preserved; ai and au have become e and o: Erāvaṇa (name of Indra’s elephant) = Airāvaṇa; metti ‘ friendship ’ = maitrī; ve (interj.) = vai. orasa ‘ derived from the breast ’ = aurasa; pora ‘ urban ’ = paura; ratto ‘ at night ’ = rātrau

c7.s1.t5.2

Not infrequently e and o are shortened into i and u before double-consonance ⁴; this shortening may take place even where the double-consonance is of secondary origin (according to § 6.2): 1. i from e = original e: paṭivissaka ‘ neighbour ’ M. I. 126⁵, DhCo. III. 155¹¹ from *-vessaka = prativeśya-ka; pasibbaka ‘ bag ’ Vin. III. 17¹⁰, JāCo. III. 10²¹ etc. through *-sebbaka = prasevaka. The word ubbilla ‘ pleasant surprise ’ M. III. 159⁴ with its numerous derivatives belongs to the root vell with ud. Also dvinnam, ubhinnam are traced by E. Kuhn⁵ to *dvenam, *ubhenam which are directly derived from the Nom. (*ubhe instead of ubho is due to analogy with dve⁶).—2. i from e = original ai: issariya ‘ rulership ’ = aiśvarya; sindhava ‘ horse from Sind ’ = saindhava.—3. u from o = original o: akuppa, asamkuppa ‘ unshakable ’ Th1. 182, 649 = -kopya; tutta ‘ spur ’ Cp. III. 5.2, D. II. 266⁵ (verse) = tottra⁷; sussam (v.l. sossam) ‘ I shall hear ’

c7.s1.t5.3

⁴ Cf. for Pkr. Pischel, § 84.

c7.s1.t5.4

⁵ Beitr. p. 28.

c7.s1.t5.5

⁶ Cf. Ś. duveṇaṃ, duvehi, duvesu.

c7.s1.t5.6

⁷ E. Müller, PGr. p. 12

c7.s1.t5.7

Sn. 694 = śroṣyāmi ; gunnam Gen. Pl. of go ‘cow’ = gonām.—4. u from o = original au : usukka ‘zeal’ = autsukya; khudda ‘honey’ Jā. VI. 532³⁰, D. III. 85¹⁶=kṣaudra ; ludda ‘horrible, diabolical’ Sn. 247, Vv. 84.⁵, M. II. 97²⁶ = raudra¹ ; assumha ‘we heard’ (§ 159. III) = aśrauṣma. In ussāva ‘dew’ JāCo. II. 11¹², DhCo. III. 338¹⁷ (AMāg. ussā and osā)=avaśyāya the u is derived from o<ava.

c7.s1.t5.8

6. Influence of Neighbouring Vowels or Consonants on the Vowels

c7.s1.t5.9

§ 16. Vowels are not infreqneutly influenced by neighbouring vowels. Here we have the beginnings of a ‘vowel-assimilation’ in Pāli.² 1. Influence of following vowels: (a) i becomes u before a following u: usu ‘arrow’ (also in AMāg.) = iṣu ; ucchu ‘sugar-cane’ (AMāg. ucchu beside ikkhu) = ikṣu ; kukku (a measure of length) A. IV. 404²¹, Vin. I. 254³⁶=kiṣku³ ; susu ‘young’ = śiśu (nasalised form in sumsumāra ‘crocodile’ = śiśumāra). In kukkusa ‘powder in rice-ears’ Vin. II. 280²⁰ = kiknasa⁴ the intermediate steps are *kikkasa, *kikkusa (§19.2). Hereto belongs nuṭṭhubhati, -hati ‘spits out’ Vin. II. 175⁷, JāCo. I. 459², II. 105²³ (beside niṭṭhubhati Ud. 50¹⁸. -hati DhCo. II. 36⁷) from a root *stubh with ni⁵.—(b) a becomes u before following u: sumugga ‘basket’ JāCo. I. 265²⁸ (beside samugga) = samudga ; usūyā, usuyyā ‘envy’ S. I. 127⁸, JāCo. I. 444⁵, D. II. 243² (§ 6.2) = asūyā.—(c) a becomes i before following i: sirimsapa ‘snake’ = sarisṛpa ; timissā ‘moonless night’ JāCo. III. 433¹⁰ = tamisrā.⁶—(d) u becomes a before following a in kappara (AMāg. koppara) ‘knuckle’ Vin. III. 121⁹. JāCo. I. 293¹ =kūrpara.

c7.s1.t5.10

¹ H. Lüders, GN. 1898, p 1.

c7.s1.t5.11

² Trenckner, Notes, p. 75 f. For Pkr. see Pischel, § 117. On the vowel-assimilation in modern dialects see Grierson, ZDMG. 49. 400 ff. ; Geiger, LSprS. p. 43ff.

c7.s1.t5.12

³ E. Müller, Pgr. p. 9.

c7.s1.t5.13

⁴ VT. III. 367, f.-n. 4.

c7.s1.t5.14

⁵ Pischel, PkrGr. § 120.

c7.s1.t5.15

⁶ Also timisā D. II. 175¹⁷, M. III. 174²⁶, which presupposes a *tamiṣā.

c7.s1.t5.16

§ 17. There is further 2. the influence of the preceding vowel. (a) a becomes u after preceding u: uluṅka ‘ladle’ JāCo. I. 235²³, III 71²²=udaṅka ; kuruṅga (a kind of gazelle) JāCo. I. 178¹⁵ = kuraṅga ; pukkusa (designation of a low caste) Jā. III. 194³⁰, M. III. 169²⁸, Pu. 51²³, Milp. 5¹²=pukkaśa ; puthujjana ‘ordinary ‘ covered, concealed ’ Th1. 153, S. I. 167²⁷, JāCo. I. 347⁸ = *prāvṛta* and *apāruta* ‘ opened up ’ Vin. I. 7⁴ (verse), D. II. 217¹⁵, JāCo. I. 264⁴ = *apāvṛta*.

c7.s1.t5.17

§ 14. The vowel *ḷ* is represented by *u*: *kutta* ‘ clipped ’ D. I. 105 ⁹, DCo. I. 274¹⁷ = *kḷpta*.² In the same way also *kuttaka* (a kind of woollen cover) D. I. 7¹⁰, Vin. I. 192⁸, II. 163²⁴, perhaps ‘ shorn cover ’.³ Further *kutta*, *kutti* ‘ behaviour, procedure ’ = *kḷpta*, *kḷpti* in *itthikutta*, *purisakutta* A. IV. 576¹⁹, JāCo. I. 296²¹ etc., *saṃnatavīrakutti* Jā. V. 215¹⁶ where *kutta* = *kappana*, just as *mata* = *maraṇa*. In DhsCo. 321¹¹ it is explained by *kiriyā*.

c7.s1.t5.18

² Cf. Rhys Davids, Dial., I. 130, f.-n. 2.

c7.s1.t5.19

³ Differently explained by Buddhaghosa as quoted by Rhys Davids and Oldenberg, VT. II. 27, f.-n. 4.

c7.s1.t5.21

§ 15. The diphthongs *e*, *o* are as a rule preserved; *ai* and *au* have become *e* and *o*: *Erāvaṇa* (name of Indra’s elephant) = *Airāvaṇa*; *metti* ‘ friendship ’ = *maitri*; *ve* (interj.) = *vai*. *orasa* ‘ derived from the breast ’ = *aurasa*; *pora* ‘ urban ’ = *paura*; *ratto* ‘ at night ’ = *rātrau*

c7.s1.t5.22

Not infrequently *e* and *o* are shortened into *i* and *u* before double-consonance⁴; this shortening may take place even where the double-consonance is of secondary origin (according to § 6.2): 1. *i* from *e* = original *e*: *paṭivissaka* ‘ neighbour ’ M. I. 126⁵, DhCo. III. 155¹¹ from *-vessaka* = *prativeśya-ka*; *pasibbaka* ‘ bag ’ Vin. III. 17¹⁰, JāCo. III. 10²¹ etc. through *-sebbaka* = *prasevaka*. The word *ubbilla* ‘ pleasant surprise ’ M. III. 159⁴ with its numerous derivatives belongs to the root *vell* with *ud*. Also *dvinnam*, *ubhinnam* are traced by E. Kuhn⁵ to *-dvenam*, *-ubhenam* which are directly derived from the Nom. (*ubhe* instead of *ubho* is due to analogy with *dve*⁶).—2. *i* from *e* = original *ai*: *issariya* ‘ rulership ’ = *aiśvarya*; *sindhava* ‘ horse from Sind ’ = *saindhava*.—3. *u* from *o* = original *o*: *akuppa*, *asaṃkuppa* ‘ unshakable ’ Th1. 182, 649 = *-kopya*; *tutta* ‘ spur ’ Cp. III. 5.2, D. II. 266⁵ (verse) = *tottra*⁷; *sussam* (v.l. *sossam*) ‘ I shall hear ’

c7.s1.t5.23

¹ E. Leumann, GGA. 1899, Nr. 8, p. 594.

c7.s1.t5.24

⁴ Cf. for Pkr. Pischel, § 84.

c7.s1.t5.25

⁵ Beitr. p. 28.

c7.s1.t5.26

⁶ Cf. Ś. *duveṇam*, *duvehi*, *duvesu*.

c7.s1.t5.27

⁷ E. Müller, PGr. p. 12.

c7.s1.t5.28

Sn. 694 = śroṣyāmi ; gunnam Gen. Pl. of go ‘cow’ = gonām.—4. u from o = original au : usukka ‘zeal’ = autsukya; khudda ‘honey’ Jā. VI. 582³⁰, D. III. 85¹⁶=kṣaudra ; ludda ‘horrible, diabolical’ Sn. 247, Vv. 84.⁵, M. II. 97²⁶ = raudra¹ ; assumha ‘we heard’ (§ 159. III) = aśrauṣma. In ussāva ‘dew’ JāCo. II. 11¹², DhCo. III. 338¹⁷ (AMāg. ussā and osā)=avaśyāya the u is derived from o<ava.

c7.s1.t5.29

¹ H. Lüders, GN. 1898, p. 1.

6. Influence of Neighbouring Vowels or Consonants on the Vowels

c7.s1.t6.1

§ 16. Vowels are not infrequently influenced by neighbouring vowels. Here we have the beginnings of a ‘vowel-assimilation’ in Pāli.² 1. Influence of following vowels: (a) i becomes u before a following u: usu ‘arrow’ (also in AMāg.) = iṣu ; ucchu ‘sugar-cane’ (AMāg. ucchu beside ikkhu) = ikṣu ; kukku (a measure of length) A. IV. 404²¹, Vin. I. 254³⁶=kiṣku³ ; susu ‘young’ = śiśu (nasalised form in sumsumāra ‘crocodile’ = śiśumāra). In kukkusa ‘powder in rice-ears’ Vin. II. 280²⁰ = kiknasa⁴ the intermediate steps are *kikkasa, *kikkusa (§19.2). Hereto belongs nuṭṭhubhati, -hati ‘spits out’ Vin. II. 175⁷, JāCo. I. 459², II. 105²³ (beside niṭṭhubhati Ud. 50¹⁸, -hati DhCo. II. 36⁷) from a root *stubh with ni⁵.—(b) a becomes u before following u: sumugga ‘basket’ JāCo. I. 265²⁸ (beside samugga) = samulga ; usūyā, usuyyā ‘envy’ S. I. 127⁸, JāCo. I. 444⁸, D. II. 243² (§ 6.2) = asūyā.—(c) a becomes i before following i: sirimsapa ‘snake’ = sarīsṛpa ; timissā ‘moonless night’ JāCo. III. 433¹⁰ = tamisrā.⁶—(d) u becomes a before following a in kappara (AMāg. koppara) ‘knuckle’ Vin. III. 121⁹. JāCo. I. 293⁷ = kūrpara.

c7.s1.t6.2

² Trenckner, Notes, p. 75 f. For Pkr. see Pischel, § 117. On the vowel-assimilation in modern dialects see Grierson, ZDMG. 49. 400 ff. ; Geiger, LSprS. p. 43ff.

c7.s1.t6.3

³ E. Müller, PGr. p. 9.

c7.s1.t6.4

⁴ VT. III. 367, f.-n. 4.

c7.s1.t6.5

⁵ Pischel, PkrGr. § 120.

c7.s1.t6.6

⁶ Also timisā D. II. 175¹⁷, M. III. 174²⁶, which presupposes a *tamiṣā.

c7.s1.t6.7

§ 17. There is further 2. the influence of the preceding vowel. (a) a becomes u after preceding u: uluṅka ‘ladle’ JāCo. I. 235²³, III 71²²=udaṅka ; kuruṅga (a kind of gazelle) JāCo. I. 178¹⁵ = kuraṅga ; pukkusa (designation of a low caste) Jā. III. 194³⁰, M. III. 169²⁸, Pu. 51²³, Milp. 5¹²=pukkaśa ; puthujjana ‘ordinary (not converted) people ' = *pṛthagjana*.—(b) *i* becomes *u* after preceding *e*: *arañjara* ' water-pot ' Abhp. 456 = *aliñjara* ; *kākaṇikā* (small coin) JāCo. I. 120²⁰, DhCo. III. 108¹² = *kākiṇikā* ; *pokkharaṇī* ' lotus-tank ' = *puṣkariṇī* ; *sākhalla*, *-lya* ' friendship ' (§ 3) from *sakhila*.¹—(c) *u* becomes *a* after preceding *a*: *āyasmant* ' venerable ' = *āyuṣmant*; *matthaluṅga* ' brain ' Kh. 3 JāCo. I. 493¹⁰ = *mastuluṅga*; *sakkhali*, *-likā* ' ear-lobe ' (JPTS. 1909, p. 17) = *śaṣkulī*.—(d) *a* becomes *i* after preceding *i*: *siṅgivera* ' ginger ' = *śṛṅgavera* ; *nisinna* ' sitting ' (but *pasanna*, *saṃsanna*) = *niṣaṇṇa*.²

c7.s1.t6.8

¹ Similarly *kosajja* ' slowness ' Dh. 241, A. I. 11²⁹ (from *kusīta*) as opposed to *kausīdya*.

c7.s1.t6.9

² Also in *pathavi* (-th-) and *puthuvi* (§ 12.4) = *pṛthivī* we have vowel-assimilation; *puthavi* in a cross-form. Not so Pischel, PkrGr. § 115.

c7.s1.t6.10

§ 18. The influence of consonants on vowels comes to light in the fact that 1. the vowel *u* appears by preference in the neighbourhood of labials, and that 2. *i* appears by preference in the neighbourhood of palatals. Ad 1 : Of derivatives from the root *majj* with *ni* and *ud* we have *nimujjati* ' drowns ', *ummujjati* ' floats ', *rimujjā*, *ummujjā*, *nimugga* etc. Cf. further *saṇimujjanī*, *-muñjanī* ' broom ' DhCo. III. 168⁶ beside *saṃmajjani* DhCo. III. 7¹⁹ = *saṃmārjani*. Also *mutu* ' thought ' Sn. 714, 793, M.I. 3²² etc., *muti* ' thought ' Sn. 846, *mutimā* Sn. 321, Jā. IV. 76¹⁰ = *matimān* should be regarded only as dialectical side-forms of *mata* etc. The existence of a root *mu* however seems to be guaranteed by the Fut. Pass. Part. *motabba* and the verbal noun *motar* A. II. 25¹⁶⁻¹⁸.—Ad 2: *mimjā* ' marrow ' (see p. 65, f.-n. 4) from *majjā*; *jigucchati* ' conceals ' D. I. 213²³, JāCo. I. 422²⁰, *jigucchā* D. I. 174¹⁹ as opposed to *jugupsate*, *jugupsā*³; *bhiyyo* ' more ' from *bhūyas*. On *seyyā* see § 9.

c7.s1.t6.11

³ Forms like *jeguccha* ' contrary ' Vin. I. 58²⁸ etc. and *jegucchim* ' disgusting ' Vin. III. 3¹, JāCo. I. 390¹⁴ etc. are new formations. Cf. § 3.

7. Influence of Accent on Vocalism

c7.s1.t7.1

§ 19. In words of three or four syllables, which on the evidence of Skr. had the accent on the first syllable, the vowel of the second syllable is often reduced. In most cases *i* appears as the reduced vowel; after labials appears frequently, though not always, *u* instead of *i*: 1. After the accent-syllable *a* becomes *i*: *caṃdimā* ' moon ' = *candramās*⁴; *carima* ' following, last ' Thl. 202 = *carama* ; *parima* ' the highest ' M. III. 112¹⁵ = *parama* ; *puttimā* N. Sg. ' endowed with sons ' Sn. 33, 34 = *putramān* ; *majjhima* ' middle ' = madhyama¹; saccika ‘true’ Milp. 226¹⁷ = satyaka. Cf. the forms ahimkāra, mamimkāra ‘self-consciousness’ M. III. 32³⁴ beside ahamk- mamamk-. In the same way should be judged the future forms like dakkhisi ‘you will see’, kāhisi ‘you will do’ (beside dakkhasi, kāhasi), chisi ‘you will go’, etc. as opposed to Skr. drakṣyasi, *karṣyasi, cṣyasi.—2. After the accent-syllable a becomes u: navuti ‘ninety’ (AMāg. naüim)=navati; pāpuraṇa ‘mantle’ S. I. 175 f., DhCo. III. 1⁹ through pāvuraṇa M.I. 359¹³ (AMāg. pāüraṇa)=prāvaraṇa; sammuti ‘consent’ (beside sammata)=sammati (cf. above §18.1). In the flexion-al system (§ 92.3) brahmunā, brahmuno; kammunā, kammuno (also AMāg. kammuṇā, -ṇo); addhunā, addhuno²=brahmaṇā, -ṇas; karma-ṇā, -ṇas; adhvanā, -nas. After non-labials there is u in ajjuka (name of a plant) Abhp. 579=arjaka³; kukkusa (§ 16. 1a)=kiknasa; pckhuṇa ‘wing’ Th1. 211, 1136, JāCo. I. 207¹⁰=preṅkhaṇa ‘swing’⁴; sajjulasa ‘resin’ Vin. I. 202¹=sarjarasa.—3. Occasionally after the accent-syllable i becomes u and u becomes i: rājula (a reptile) Abhp. 651=rājila; geruka ‘reddish chalk’ Vin. I. 48⁸ (AMāg. geruya beside Māh. geria)=gairika; pasuta ‘intent on something’ Th1. 28, D. 1. 135²⁵, JāCo. III. 26⁴=prasita. Further muditā ‘softness’ M. I. 370⁹, S. V. 118²⁵ (beside mudutā A .I. 9²⁹)=mṛdutā. On suṇisā see § 31.2.

c7.s1.t7.2

⁴ A different but very far-fetched explanation is given by Pischel, PkrGr. § 108.

c7.s1.t7.3

¹ This should not be regarded as “ Samprasāraṇa ” as E. Kuhn, Beitr. p. 54 suggests, because the i is not derived from ya; the y is contained in ijk. At the most one can say that at an earlier stage of the language the a after y in madhyama had a pronunciation leaning towards i.

c7.s1.t7.4

² The same view should be taken as in f.-n. 1.

c7.s1.t7.5

³ Subhūti, Abhp.-Sūci under this word gives ajjaka beside ajjuko.

c7.s1.t7.6

§20. Unstressed short vowels, particularly immediately after the accent, are sometimes syncopated: jaggati ‘watches’ (§142.⁴) is to be traced from jāgarati through *jāgªrati; oka ‘water’ Dh. 34, 91 from udaka through *ŏkka, *ukka, *utka, *udªka; agga ‘house’ (in uposath-agga, khuragga, bhattagga etc.) from agāra through *agªra, *agara. Syncope is in evidence also in the verbal ending -mhe (beside -mahe).⁵ Finally, there is a number of onomatopoetic words in which syncope may be clearly traced: cicciṭāyati ‘rustles’ Vin. I. 225²⁵, S. I. 169⁹, Sn. S. 14, Pu. 36³² beside ciṭiciṭāyati; sassara, babbhara M.I. 128³⁵ (JPTS. 1889, p. 209) for *sarªsara, *bharªbhara beside sarasara, bharabhara. Enclicis khalu has become kho through *khªlu, *kkhu.⁶

c7.s1.t7.7

⁴ Pischel, § 89.

c7.s1.t7.8

⁵ E. Kuhn, Beitr., p. 94.

c7.s1.t7.9

⁶ In Pkr. Saurasenī and Māgadhī we have kkhu which causes the shortening of prceding e, o. Pischel § 94, 148. Whence the o in Pāli kho?

c7.s1.t7.10

§ 21. Weakening of the syllable preceding the accented one is found in *kahāpaṇa* (a coin) (Pkr. *kăhāvaṇa*) = *kārṣāpaṇa*; perhaps also in *nigrodha* ‘ficus indica’ = *nyagrodha* and in *susāna* ‘burial ground’ from *śvaśāna*, a side-form of *śmaśāna*¹. Yet however similar instances of *samprasāraṇa* are found also in the syllable with main accent. It is perhaps due to the weakening of the syllable preceding the accented one that *dvi-* appears as *du-* in compounds, e.g., *dujivha* ‘double-tongued’ Jā. V. 82². Under the accent we have the regular forms *dvi-*, *di-*, e.g. *dipada* ‘biped’³. The two types having later crossed each other we have on the one hand *duvidha* ‘double’ = *dvividha*, and on the other *dvibhūmiko* ‘consisting of two stories’ JāCo. II. 18⁴.—In forms like *thapeti* (*uṭṭhāpeti* etc.) as opposed to *sthāpayati* we have analogical formations after the type *jñapayati* etc. (§ 180.1); similarly in *kiṇati* ‘buys’, as opposed to *krīṇāti*, after the types *mināti*, *lunāti*⁵.

c7.s1.t7.11

² Also Pkr. has *du-*, *do-* beside *di-*, *bi-*. Pischel, § 436.

c7.s1.t7.12

³ A different explanation is given by Michelson, IF. 23. 127.

c7.s1.t7.13

⁴ Even new nominal stems are formed in this way; Skr. *āgas* ‘sin’ through *āgo* becomes *āgu*, inflected like *madhu*.

c7.s1.t7.14

⁵ A different explanation in Pischel, PkrGr. § 103.

c7.s1.t7.15

§ 22. The effect of accent is perceptible also in the shortening of unstressed final syllables. Thus *o* becomes *u* in *asu* ‘that’ (§ 109) through *aso* (thus in AMāg.) = *asau*; *udāhu* ‘or’ = *utāho*; *sajju* ‘immediately’ Dh. 71 (from which *sajjukam*) through *sajjo* = *sadyas*; *hetu* (in *kissa hetu* ‘what for ?’) through *heto* = *hetos*⁶. With later nasalisation: *-khattum* (adv. numeral suffix) through *-khatto* = *-kṛtvas* and *adum* ‘that’ through *ado* = *adas*. Cf. § 66. 2 *b*. Qualitative change (reduction) is in evidence in *saddhiṃ* ‘together with’ = *sārdham*, *sakkhi(ṃ)* or *succhi* ‘before one’s eyes’ (certainly not = *sākṣāt*, but) = *sākṣam* (AMāg. *sakkham*⁷); *saniṃ* ‘slowly’ Mhvs. 25. 84 (not = *śanais*, but) = *śanam*. Reduction to *u* under the influence of an *u* of the preceding syllable (§ 17. 2 *a*) is found in *puthu* ‘separate’ Thl. 86, Milp. 4 = *pṛthak*. The enclitic *-svid* has become *-su*, *-ssu*: *kiṃ-su*, *kena-ssu* etc. S. I. 36 ff. We have moreover *-si* in *kaṃ-si* DhCo. I. 91⁸. Cf. § 111.¹.

c7.s1.t7.16

¹ Pischel, PkrGr. § 104. Johansson (IF. 25. 225 ff.) separates *susāna* from *śmaśāna* and derives the former from *śavaśayana* ‘burial ground’.

c7.s1.t7.17

⁶ Pischel, § 114.

c7.s1.t7.18

§ 23. In a series of words even from the beginning the long second syllable was shortened. This is evidently due to the shifting of the accent to the first syllable. Examples: *alika* ‘false’ Sn. 239, S. I. 189², Rasav. II. 83¹⁴=alīka; gahita ‘seized’=gṛhita; pannavant ‘intelligent’ Thl. 70, Vin. I. 60¹=prajñāvant; pānīya ‘water’ beside pānīya (pānīyāni D. I. 148⁴, pānīyāni JāCo. I 450⁸) =pānīya (AMāg., JMāh. pānīya); vammika ‘ant-hill’ JāCo. I. 482⁵ beside vammika JāCo. III. 85⁸=vălmika; sāluka ‘lotus-root’ Vin. I. 246¹⁶=śālūka. Similarly dutiya ‘second’, tatiya ‘third’=dvitīya, tṛtīya¹.—In other cases, where the vowel of the second syllable was originally short, qualitative change of the vowel took place as a result of this shifting of accent: Pajjunna (name of the god of rains) D. II. 260²⁵, JāCo. I. 331²¹=Parjanya; mutiṅga ‘drum’ D. I. 79¹³, Vin. I. 15¹⁰ (Pkr. muiṅga)=mṛdaṅga. Cf. meraya ‘intoxicating drink’ Dh. 247, D. I. 148²⁰=maireya.

c7.s1.t7.19

§ 24. The effect of the new expiratory accent is perceptible also in the occasional lengthening of the vowel of the first syllable: ājira ‘courtyard’ Mhvs. 35.3=ajira; ālinda ‘terrace in front of a house’ D. I. 89³⁰, Vin. I. 248³, DhCo. I. 26⁴=alinda; perhaps we have also to include here ānubhāva ‘power’ JāCo. I. 509²³=anubhāva. This explanation of the lengthening of the vowel of the initial syllable often however remains doubtful, as in āroga ‘in good health’ JāCo. I. 408¹ (reading uncertain) beside aroga=aroga; pātibhoga ‘surety’ Ud. 17¹⁰, Iv. 1⁸, JāCo. II. 93¹⁴, which in meaning is difficult to connect with pratibhoga; pāṭiyekka ‘individually’ JāCo. I. 92²⁴ beside pacceka=pratyeka.—Gemination of consonants may take the place of lengthening of vowels: ummā ‘flax’ in ummāpuppha (a precious stone) A. V. 61²¹=umā; kummagga ‘evil path A. III. 420²⁹, Pu. 22⁷, Milp. 390⁸, kunnadi ‘small (intermittent) river’ Thl. 145, S. I. 109⁵, JāCo. III. 221¹¹, kussubbha ‘small pool’ S. V. 63⁸ (beside kusubbha)=ku+mārga, nadī, śvabhra; mukkhara ‘garrulous’ Minayeff, Prātimokṣa 59 (beside mukhara S. I. 203³⁴, JāCo. III. 103⁵) =mukhara².

c7.s1.t7.20

¹ Pischel, KZ. 35. 142, PkrGr. § 82, 91 assumes the basic forms *dvitýa, *tṛtýa. But see Jacobi, Kalpasūtra (AbhKM. VI. 1), p. 103, f-n., as well as KZ. 35. 570 f.

c7.s1.t7.21

² The example suddiṭṭha quoted in J PTS. 1909, p. 193 is=su-uddiṭṭha; sakkāya is not=svakāya, but=satkāya (Franke, D., p. 54, f.-n. 4) and cikkhalla ‘earthen,’ Vin. II. 122³¹ not=cikhala but=*caikhalya > *cēkhalla > *cekkhalla (E. Müller, PGr. p. 19).

c7.s1.t7.22

8. Samprasāraṇa and the Loss of Syllable through Contraction

c7.s1.t7.23

§ 25. Through Samprasāraṇa: 1. yā becomes ī also in stressed syllable: thīna ‘sloth’=styāna; dviha, tiha ‘period of two (three)

c7.s1.t7.24

days ’ D. I. 190¹⁵, A. I. 140¹⁵ = dvyaha, tryaha; visīveti ‘ dries out at fire ’ JāCo. II. 68¹⁶, DhCo. I. 225² = viśyāpayati. Of frequent occurrence is viti- = vyati-, vi-ati-, e.g., vitivatta = vyativṛtta etc. Cf. niṅka (a kind of deer) Abhp. 619 = nyaṅku. Instead of ī there is e in (sam)pavedhati ‘ shakes ’ Sn. 928, D. II. 22² from root vyath. Often ya is retained: vyasaha ‘ misfortune ’, vyādha ‘ hunter ’ etc. In cajati ‘ gives up ’ = tyajati, majjhantika ‘ relating to noon-time ’ from madhya etc. we have the assimilation of y to the preceding consonant.—2. vā becomes ū in sūna ‘ dog ’ Abhp. 519 from the stem śvān.¹. Before double-consonance ū becomes o through u (§ 10): sotthi ‘ welfare ’ (beside suvatthi) = svasti ; soppa ‘ sleep ’ (beside supina) S. I. 110³² (verse) = svapna; sobbha ‘ tank ’ (cf. kussubbha § 24) = śvabhra. The form ko ‘ where? ’ S. I. 199¹⁶ (verse), Vin. I. 36²⁴ (beside kvam, kuvam, kva-ci) is probably a sandhi-form before double-consonance. Before single consonants there is o instead of ū, e.g., in sopāka (AMāg. sovāga) ‘ man of low caste ’ = śvapāka and soṇa ‘ dog ’ ². Moreover vā is often retained in the assimilation of v, e.g., assattha ‘ ficus religiosa ’ (AMāg. however has assottha etc.) = aśvattha. —3. Quite peculiar is dosa in which Skr. doṣa ‘ fault ’ and dveṣa ‘ hatred ’ have coincided. Cf. dosaniya ‘ deserving hatred ’ A. III. 169²⁸ = dveṣaṇīya.

c7.s1.t7.25

¹ In turita ‘ hastening ’ and kuthita ‘ boiled,’ as opposed to tvarita and krathita, we have in fact no Samprasāraṇa of the Pāli stage. We are to assume here older basic forms *turita, *kuthita.

c7.s1.t7.26

² The intermediate stages seem to have been iya and uva; thus dvyaha : *dviyaha : dviha, śvān- : *suvān- : sūna, soṇa. Cf. soṇṇa ‘ gold ’ beside suvaṇṇa.

c7.s1.t7.27

§ 26. Through contraction aya can become e and ava can become o³, clearly through the intermediate stages ayi: aï, avu: aü (§ 19). 1. aya becomes e in jeti ‘ wins ’ (beside jayati) = jayati, etc.; ajjhena ‘ study ’ Sn. 242, M. III. 1¹³ = adhyayana. Facultatively also in causatives and other verbal stems in aya, such as moceti, katheti⁴. Further terasa ‘ thirteen ’ = *trayadaśa, tevisa(ti) = *trayaviṃśati; aya is retained in nayana ‘ eye,’ sayana ‘ bed ’ etc. (but seṇḍsana ‘ bedstead and seat ’ beside sayanāsana Sn. 388 f., Dh. 185).—2. ava becomes o in odhi ‘ limit ’ D. II. 160³², JāCo. II. 18²¹ = avadhi; oma ‘ lowly ’ Sn. 860, A. III. 359²⁸ (verse) = avama; poṇa ‘ sinking, inclined ’ Vin. II. 287¹⁹, Ud. 53⁹ etc. = pravaṇa; loṇa ‘ salt ’ = lavaṇa; hoti ‘ is, becomes ’

c7.s1.t7.28

³ Pischel, PkrGr. § 153 f.

c7.s1.t7.29

⁴ In the same way may be explained also bhāyāmi ‘ I fear ’ and palāyati ‘ flies ’ beside which are found also bhemi and paleti. Cf. § 138 and 139, 1.

c7.s1.t7.30

(beside *bhavati*) and many other forms. Also facultatively *o* = the prefix *ava-* (*orodha* ‘harem’ = *avarodha*) and *vo* = the prefix *vyava-*, *vi-ava*, (*vosita* ‘fulfilled’ Dh. 423 = *vyavasita*). Cf. *uposatha* (Pkr. *posaha*) = *upavasatha*. *Ava* is retained in *lavana* ‘harvest,’ *savana* ‘hearing’ etc. But *loṇa* ‘salt’ = *lavaṇa*.

c7.s1.t7.31

§ 27. Further cases of contraction are 1. *aya* becomes *ā*: *paṭisallāna* ‘meditation’ D. II. 9¹⁰, JāCo. II. 77¹¹ etc. = *pratisamlayana*; *sotthāna* ‘welfare’ Sn. 258, A. IV. 271²⁰ (verse) = *svastyayana* (§ 25.2).—2. *āya* becomes *ā*: *vehāsa* ‘atmosphere’ D. I. 9¹⁰, JāCo. I. 445²⁰ etc. = *vaihāyasa*; *upaṭṭhāka* ‘attendant’ Vin. I. 72¹⁷, JāCo. I. 357⁴ etc. = *upasthāyaka* (but fem. *upaṭṭhāyikā* Thūpavs. 81²⁹); *Kaccāna* (beside *Kaccāyana*), *Moggallānu* n. pr. = *Kātyāyana*, *Maudgalyāyana* etc.¹ Very frequently at the end of a word *-āya* is contracted into *-ā*, such as *sayaṃ abhiññā* ‘on the strength of one’s own knowledge’ instead of *-ññāya* = *abhijñāya* Ger.; *apaṭipucchā* ‘without hearing’ Vin. II. 3³ instead of *-cchāya* I. Sg. f.; *esanā* ‘(goes) in search of’ JāCo. II. 34¹⁶ instead of *-nāya* D. Sg. m.; *chamā* ‘on the earth’ instead of *-māya* Loc. Sg. f.² Particularly in the first syllables of words *āya* is likely to be retained: *vāyasa*, *jāyati* etc.—3. *āva* becomes *o* in *aṭidhona(cārin)* ‘(committing) transgressions’ Dh. 240 = *atidhovana*.³ But *āva* is retained in the first syllables: *pāvaka*, *sāvaka*.—4. *avā* becomes *ā* in the *yāgu* ‘rice-gruel’ A. III. 250¹² etc. = *yavāgū*; *avā* remains uncontracted in *kavāṭa*, *pavāḷa*, as *ayā* in *dayālu* etc.—5. *ayi* and *avi* become *e*: *acchera* ‘miraculous’ Vv. 84. 12 through *acchayira* (beside *acchariya*) = *āścarya*; similarly *ācera* ‘teacher’ Jā. IV. 248⁹ (beside *ācariya*) = *ācārya*; *macchera* ‘envy’ Dh 242, DhCo. III. 2¹ etc. = *mātsarya*⁴; *thera* ‘venerable priest’ = *sthavira*; *hessati* ‘will be’ (§ 154.2) = *bhaviṣyati*⁵.—6. *āyi* becomes *e* in the technical term *acceka(cīvara)* ‘(garment) given at an unusual hour’ Vin. III. 260³³ beside *accāyika* ‘pressing’ M. II. 112²⁷, JāCo. I. 338³¹ = *atyāyika*. Beside this *e* we have *ī* in *pāṭihīra* ‘sign of miracle’ D. I. 193³, Mhvs. 5.188 through *pāṭihāyira* (beside pāṭihāriya) = prātihārya; similarly (a)samhirā ‘(not) to be won’ Jā. V. 81¹⁷, A IV. 141¹¹ etc. = (a)samhārya.—7. iya is changed into ī (ī) in kittaka ‘how much?’ Smp. 304¹ = *kiyattaka. According to § 10. 2 is to be explained ettaka ‘so much’ = *iyattaka.—8. Isolated cases of contraction are found moreover in koṭṭha in the names of birds rukkhakoṭṭhasakuna JāCo. III. 25²⁹ if it is = koyaṣṭi² and mora ‘peacock’ (the same form also in Pkr. ; in Pāli also mayūra D. III. 201²²) = mayūra³.

c7.s1.t7.32

¹ Also *pācittiya* ‘transgression requiring penance’ Vin. IV. 1 ff., if it is related to *prāyaścittika* Sylv. Lévi, JAs. Ser. X, t. 20, p. 506 ff. derives it from a *prāk-citta*. ² In analogy with these cases an inorganic *ya* has sometimes been added to a final *ā*: *senāya caturaṅginī* Jā. V. 322¹⁹. ³ D. Andersen, P. Gl. under the word. ⁴ Thus through metathesis in the intermediate step. Cf. E. Kuhn, Bietr. p. 55; E. Müller, PGr. p. 41 f.; V. Henry, Précis de Gramm. Pâlie § 88.4. Pischel, PkrGr. § 176 however suggests epenthesis. ⁵ On *hohiti*, *hotabba*, *hotum* see § 151, 206.

c7.s1.t7.33

¹ Not so Pischel, § 153.

c7.s1.t7.34

² Fausböll, Five Jāt. p. 38.

c7.s1.t7.35

³ Uncontracted mayūkha ‘beam of light’ as opposed to Pkr. Māh. moha.

c7.s1.t7.36

§ 28. As in Pkr.,⁴ so also in P. the prepositions upa- and apa- (through *uva-, *ava-) may become ū- and o-: 1. upa- becomes ū-, as I think, in ūhadeti ‘besmears with dung’ = upahadati and ūhasaṇa ‘smiling at somebody’ Milp. 127²¹. Cf. Pkr. ūhasia in Hem. = upahasita⁵.—2. apa- becomes o- in ovaraka ‘inner apartment (of a house)’ Vin. I. 217¹⁷, VvCo. 304¹⁴ = apavaraka; oṭṭapati ‘feels shame’ A. III. 2¹⁶ (oṭṭappa ‘shame’, oṭṭappin or oḷṭāpin ‘shameful’) from the root trap with apa⁶. Presumably also in (pacc)osakkati ‘falls back’ D. I. 230²¹, JāCo. I. 383⁸, Mhvs. 25.84 (AMāg. paccosakkai) from root svaṣk with (prati)apa⁷.

c7.s1.t7.37

⁴ Hem. I. 173; Pischel, PkrGr. § 155.

c7.s1.t7.38

⁵ It may be thought that also ūhanti in the meaning ‘defiles’ is derived from han with upa. Yet this is rendered improbable by Vin. I. 78¹², where ummihati stands parallel to it. It is certain that ūhanti=han with ud and it means ‘conquers, annihilates.’ Instead of ūhananti M. I. 243²³ we have in fact upahananti (with v.l. uh-) in the parallel passage S. IV. 56¹⁹.

c7.s1.t7.39

⁶ E Müller, PGr., p. 43.

c7.s1.t7.40

⁷ In Jā. III. 83⁶ we have avasakkati, but with the variant apa- in the Burmese MSS. The word oggata. Tbl. 477 (used about the sun) may be = apagata or aragata.

9. Increase of Syllables through Svarabhakti

c7.s1.t8.1

§ 29. Only the consonant-groups containing r, l, y, v or a nasal are separated by svarabhakti⁸. An exception is to be found in kasaṭa ‘bad, false’ A. I. 72⁸, JāCo. II. 96²², Milp. 119¹³ etc. = kaṣṭa. This is perhaps a dialectical expression. In Pkr. we have the Paiśācī form kasaṭa⁹. —The added vowel appears mostly in the inside of words. In initial position it is found in *itthī* ‘woman’ = *strī* and in *umhayati*, -te ‘smiles’¹ Jā. II. 131²², JāCo. III. 44¹⁴ = *smayate*.—Beside forms with the added vowel there are often those showing assimilation of the consonant-groups. The latter are archaic and are found particularly in the gāthās. In the commentary they are explained by the forms with the added vowel, which therefore must have been the current forms. Thus we have in Jā. III. 151⁵ *asi tikkho va maṃsamhi*; the Co. replaces *tikkha* ‘sharp’ = *tīkṣṇa* by *tikhina*. Regarding the action of the law of mora on the quantity of a long vowel preceding a consonant-group separated by svarabhakti, see above § 8. In verse the svarabhakti-vowels are often ignored as also in Pkr.² Cf. Dh. 10 *arahati* = *arhati*; Dh. 25 *kayirātha* = *kayrātha*; Th1. 477 *suriyasmim* = *suryasmim*; Th2. 49 *puriso* = *purso* etc. The *i* of *itthī* is always metrically justified; beside it however there is found in verses the form *thī*: Sn. 76¹, Jā. I. 295⁸, Jā. V. 81¹⁶.

c7.s1.t8.2

⁸ For Pkr. cf. Jacobi, KZ. 23.594; Pischel, PkrGr. § 131.

c7.s1.t8.3

⁹ Vararuci X. 6, Hemacandra IV. 314; Grierson, ZDMG. 66.52²¹; Pischel, § 132; St. Konow, ZDMG. 64. 114³⁶.

c7.s1.t8.4

¹ Wenzel, Academy 1890, II, p. 177.

c7.s1.t8.5

² It is however not right to change the orthography for the sake of metre.

c7.s1.t8.6

§ 30. Of all the added vowels the most frequent is *i* (both prothetic and anaptyctic): 1. In the group *ry*: *iriyati* ‘moves’ M. I. 74⁶, A. III. 451⁶ (substantive *iriyā*) = *iryate*, *īryā*; *mariyādā* ‘frontier’ Mhvs. 34.70 = *maryādā* etc. In the same way are formed the passives like *kariyati* from *karoti* (also *karīyati* D. I. 52²⁷), *vāriyati* ‘is held back’ = *vāryate*.—2. In other combinations with *y*: *kālusiya* ‘darkening’ DCo. I. 95¹⁰ = *kāluṣya*; *jiyā* ‘bow-string’ D.II. 334²⁰, Mhvs. 14.4 = *jyā* etc. In the same category are to be included also the passives like *pucchiyati* ‘is asked’ = *pṛcchyate*. Also in *hiyyo* (AMāg. *hijjo*) ‘yesterday’ = *hyas* we have svarabhakti with secondary reduplication of *y*.—3. In other combinations with *r*: *vajira* ‘thunder, diamond’ (AMāg. *vaira*) Dh. 161, D. I. 95⁸, Milp. 118²¹ = *vajra*. On *sirī*, *hirī* see § 8. Svarabhakti by *i* is in evidence also in *purisa* ‘man’ (cf. § 29). The basic form is *pūrṣa*. In the popular dialects the form with the svarabhakti vowel *i* was preferred to the Skr. form *puruṣa* with *u*³. From *pūrṣa* through *pussa*, *possa* is derived also P. *posa* Sn. 110 (and otherwise in verses).—4. In consonant-groups with *l*: *pilakkhu* (name of a tree) Jā. III. 24²⁶ = *plakṣa*; *hilāda* ‘joy’ Attanagaluvs. 1. 11 = *hlāda* etc. But we have invariably *sukka* ‘white’ = *śukla*.—5. In consonant-groups with nasals: *sineha* ‘affection’ = *sneha*; *tasiṇā* ‘thirst’ Dh. 342 f. (beside *taṇhā*) = *tṛṣṇā*. On the other hand there occur only *kaṇha* ‘black’ = *kṛṣṇa* (Pkr. *kaṇha*, *kasiṇa*, *kasaṇa*) and *nagga* ‘naked’ = nagna (AMāg. nagiṇa, nigiṇa). In flexion we have rājinā, rājino beside raññā, rañño = rājñā, rājñas. On gini from *agini, agni see § 66.¹. On mihita see § 50.6.

First 173 of 935 passages. Read on

Linked texts (102)