A Comparative Grammar of the Modern Aryan Languages of India v3: Volume 3

A Comparative Grammar of the Modern Aryan Languages of India v3

Language: English

v3.c13 Front matter — 17 parts
v3.c14 Table of Contents — 4 parts
v3.c15 Chapter Ii.—the Simple Tenses. — 3 parts
v3.c16 Chapter Iv.—the Compound Tenses. — 5 parts
v3.c17 Errata — 851 parts
v3.c18 Chapter V — 153 parts
v3.c19 Chapter VI — 92 parts
v3.c20 Index — 137 parts

Volume 3

Front matter

v3.c13.11

HINDI, PANJABI, SINDHI, GUJARATI, MARATHI, ORIYA AND BANGALI.

v3.c13.14

BENGAL CIVIL SERVICE, FELLOW OF THE UNIVERSITY OF CALCUTTA, MEMBER OF THE ROYAL ASIATIC SOCIETY, THE GERMAN ORIENTAL SOCIETY, ETC., ETC.

| | | PAGE |
| § 1. | Structure of the Sanskrit Verb | 1 |
| § 2. | Beginnings of the Analytical System in Sanskrit | 6 |
| § 3. | Conjugations of the Pali Verb | 8 |
| § 4. | Tenses of the Pali Verb | 11 |
| §§ 5. 6. | The Verb in Jaina Prakrit | 16 |
| § 7. | Scœnic Prakrit Verb | 22 |
| § 8. | Apabhrança Verbal Forms | 25 |
| § 9. | The Modern Verbal Stem | 28 |
| § 10. | Phases of the Verb | 29 |
| § 11. | Single and Double Stems | 32 |
| § 12. | Single Neuter Stems from Sanskrit BHÛ Roots | 33 |
| § 13. | The same from other Classes of Sanskrit Roots | 36 |
| § 14. | Modern Neuter Stems from Sanskrit Past Participles | 37 |
| § 15. | Single Active Stems | 40 |
| § 16. | Treatment of Sanskrit Roots ending in a Vowel | 42 |
| § 17. | The Stem DEKH | 45 |
| § 18. | Double Verbs | 46 |
| § 19. | Sindhi Double Stems Differing in the Final Consonant | 48 |
| § 20. | Double Stems Differing in Vowel and Final Consonant | 52 |
| § 21. | Double Stems Differing only in the Vowel | 54 |
| § 22. | Examples and Illustrations | 58 |
| § 23. | Laws of the Formation of Modern Stems | 66 |
| § 24. | The Passive Intransitive | 68 |
| § 25. | The Passive | 71 |
| § 26. | The Causal | 75 |
| § 27. | The Passive Causal | 82 |

v3.c14.4

§ 28. The Causal in a Neuter Sense . . . . . . 84 § 29. Secondary Stems . . . . . . . . . 87 § 30. Reduplicated and Imitative Stems . . . . . 89 § 31. Gipsy Verbal Stems . . . . . . . . 95

Chapter Ii.—the Simple Tenses.

v3.c15.1

§ 32. Classification of Tenses . . . . . . . 99 § 33. The Simple Present or Aorist . . . . . 101 § 34. The Imperative . . . . . . . . . 108 § 35. The Future in Old-Hindi and Gujarati . . . 112 § 36. Type of the Active Verb in Sindhi and Marathi . . 115 § 37. Synopsis of the Simple Tenses in all Seven Languages . 118 § 38. Simple Tenses in the Gipsy Verb . . . . . 119

v3.c15.2

§ 39. Definition of the Participial Tenses . . . . 121 § 40. The Present Participle Active . . . . . 123 § 41. Tenses formed thereby—the Sindhi Future . . . 126 § 42. Marathi Indicative and Conditional Present . . . 127 § 43. Bangali and Oriya Conditional . . . . . 129 § 44. Hindi, Panjabi, and Gujarati Present . . . . 131 § 45. The Past Participle Passive . . . . . . 132 § 46. Early Tadbhava Participles in Sindhi and Panjabi . . 136 § 47. The same in Gujarati and Marathi . . . . 141 § 48. The same in Old and New Hindi . . . . 144 § 49. Tenses formed from the Past Participle . . . 147 § 50. The *Prayogas* . . . . . . . . . 151 § 51. The Future Participle Passive . . . . . 152 § 52. Tenses formed from it in Sindhi, Gujarati, and Marathi . 155 § 53. The Future in Oriya, Bengali, and Eastern Hindi . . 158 § 54. The Hindi and Panjabi Future . . . . . 160 § 55. Marathi Future compared with that in certain Hindi Dialects . . . . . . . . 161 § 56. Synopsis of the Participial Tenses in all Seven Languages 164 § 57. Participial Tenses in the Gipsy Verb . . . . 168

v3.c15.3

CHAPTER III.—THE PARTICIPIAL TENSES.

Chapter Iv.—the Compound Tenses.

v3.c16.1

| | | PAGE |
| § 58. | Definition of the Compound Tenses and Auxiliary Verbs | 170 |
| § 59. | The Root AS—Present Tense | 171 |
| § 60. | Imperfect in Panjabi and Gipsy | 175 |
| § 61. | AS, with a Negative | 178 |
| § 62. | Compound Tenses formed with AS | 179 |
| § 63. | The Root ACHH—Discussion as to its Origin | 180 |
| § 64. | Tenses derived therefrom | 184 |
| § 65. | Compound Tenses formed therewith | 187 |
| § 66. | BHÛ—the Simple Tenses | 194 |
| § 67. | id.—the Participial Tenses | 201 |
| § 68. | Compound Tenses formed therewith | 203 |
| § 69. | STHÂ | 208 |
| § 70. | YÂ | 213 |
| § 71. | Ancillary Verbs Defined | 215 |
| § 72. | Examples of Ancillaries | 216 |

v3.c16.2

CHAPTER V.—OTHER VERBAL FORMS.

v3.c16.3

| | | PAGE |
| § 73. | The Conjunctive Participle | 229 |
| § 74. | The Infinitive | 234 |
| § 75. | The Agent | 238 |
| § 76. | Sindhi Verbs with Pronominal Suffixes | 241 |
| § 77. | Conjugation of Stems ending in Vowels in Hindi, Panjabi, and Sindhi | 246 |
| § 78. | The same in Marathi | 251 |
| § 79. | The same in Bangali and Oriya | 254 |

v3.c16.4

CHAPTER VI.—THE PARTICLE.

v3.c16.5

| | | PAGE |
| § 80. | Adverbs, Nominal and Pronominal | 256 |
| § 81. | Pronominal Adverbs of Time, Place, and Manner | 257 |
| § 82. | Adverbs Derived from Nouns and Verbs | 264 |
| § 83. | Conjunctions | 270 |
| § 84. | Interjections | 272 |
| § 85. | Postpositions | 275 |
| § 86. | Conclusion | 278 |
| | General Index to the Three Volumes | 279 |

Errata

v3.c17.1

PAGE LINE 3, 19, for *this much* read *such*. 10, 9, for मञ्जीत read मञ्जित. 14, 28, for *different* read *difficult*. 19, 21, for पुरयति read पूरयति. 21, 26, for निषोतितव्यं read निषो-दितव्यं. 34, 3 from below, for *Pali* read *Prakrit* 36, 8, for *bhála* read *bhala*. 39, 26, for थत्तिष्ठत् read उत्तिष्ठत्. 44, 2, for *detu* read *dehi*. 47, 26, dele *that*. 50, 16, for मञ्ज read भञ्ज. 50, 18, after word a full stop instead of a comma. 52, 13, for *discharged* read *discharge*. 56, 5, for 90 read 38 (ed. Stenzler). 56, 7, for ॰विश्रम॰ read ॰विषम॰. 57, 28, for हष read हष्ट. 58, 5, after *Pali* a full stop instead of a comma. 61, 25, for *hladid* read *hlatu*. 63, 4, for *masj* read *majj*. 65, 9, for तळ read नळ. 70, 12, for *phrase* read *phase*. 73, 16, for नारी read मारी. 83, 24, for माडाइज्ञ read साडाइज्ञ. 105, 6, for बलें read चलें. 112, last but one, for मरसी read करसी 114, 21, for अम read अम.

v3.c17.2

PAGE LINE 124, 1, dele comma after *hearing*. 134, 6, for छोडिलो read छोडेली. 135, 29, after *dielala* insert *f.*; for *dielat*, etc., read *délat*, etc. 140, 31, for सिडणा read सिउणा. 141, 19, for भति read भीत. 149, 24, for *asmáh* read *asmaḥ*. 154, 29, for जाक्या read जावया. 162, 7, for भाछंलो read माछंलो. 163, 2, the words ‘aorist सुटां’ should be put between brackets. 175, 29, for *Ludhiana* read *Lodiana*. 176, 5 from below, for *Pr.* read *P.* 178, 18, for *Nuanti* read *Nuhanti*. 179, 24, for दखा read देखा. 196, 10, for ‘Daughter’ read ‘Laughter.’ 196, 17, for स्यनि read र्यिनि. 202, 2 from below, for 59 read 60. 214, 4, for *Imperfect* read *Imperative*. 225, 29, for दहती read रहतो. 250, 17, for *Imperfect* read *Imperative*. 254, 5, for *Oriya* read *Aryan*. 257, 3, for नाजा read राजा. 262, 22, for कश्य read कथ्य. 262, 23, for प्रमाणं read प्रमाणं. 262, 23, for लोयं read लोगं. 262, 23, for *its* read *as*. 263, 6, for *H.-D.* read *K.-D.* 267, 9, for संहित read सहित.

v3.c17.3

MODERN ARYAN LANGUAGES OF INDIA.

v3.c17.4

STRUCTURE OF VERBAL STEMS.

v3.c17.5

CONTENTS.—§ 1. STRUCTURE OF THE SANSKRIT VERB.—§ 2. BEGINNINGS OF THE ANALYTICAL SYSTEM IN SANSKRIT.—§ 3. CONJUGATIONS OF THE PALI VERB.—§ 4. TENSES OF THE PALI VERB.—§§ 5, 6. THE VERB IN JAINA PRAKRIT.—§ 7. SCENIC PRAKRIT VERB.—§ 8. APABHRANÇA VERBAL FORMS.—§ 9. THE MODERN VERBAL STEM.—§ 10. PHASES OF THE VERB.—§ 11. SINGLE AND DOUBLE STEMS.—§ 12. SINGLE NEUTER STEMS FROM SANSKRIT BHŪ ROOTS.—§ 13. THE SAME FROM OTHER CLASSES OF SANSKRIT ROOTS.—§ 14. MODERN NEUTER STEMS FROM SANSKRIT PASSIVE PAST PARTICIPLES.—§ 15. SINGLE ACTIVE STEMS.—§ 16. TREATMENT OF SANSKRIT ROOTS ENDING IN A VOWEL.—§ 17. THE STEM DEKH.—§ 18. DOUBLE VERBS.—§ 19. SINDHI DOUBLE STEMS DIFFERING IN THE FINAL CONSONANT.—§ 20. DOUBLE STEMS DIFFERING IN VOWEL AND FINAL CONSONANT.—§ 21. DOUBLE STEMS DIFFERING ONLY IN THE VOWEL.—§ 22. EXAMPLES AND ILLUSTRATIONS.—§ 23. LAWS OF THE FORMATION OF MODERN STEMS.—§ 24. THE PASSIVE INTRANSITIVE.—§ 25. THE PASSIVE.—§ 26. THE CAUSAL.—§ 27. THE PASSIVE CAUSAL.—§ 28. THE CAUSAL IN A NEUTER SENSE.—§ 29. SECONDARY STEMS.—§ 30. REDUPLICATED AND IMITATIVE STEMS.—§ 31. GIPSY VERBAL STEMS.

v3.c17.6

§ 1. THE Sanskrit verb, with its long array of tenses, intricate phonetic changes, and elaborate rules of formation, seems to have been subjected at a very early period to processes of

v3.c17.7

simplification. Indeed, we may be permitted to hold that some, at least, of the forms laid down in the works of Sanskrit grammarians, were never actually in use in the spoken language, and with all due deference to the opinions of scholars, it may be urged that much of this elaborate development arose in an age when the speech of the people had wandered very far away from the classical type. Even if it were not so, even if there ever were a time when the Aryan peasant used polysyllabic desideratives, and was familiar with multiform aorists, it is clear that he began to satisfy himself with a simpler system at a very distant epoch, for the range of forms in Pali and the other Prakrits is far narrower than in classical Sanskrit.

v3.c17.8

Simplification is in fact the rule in all branches of the Indo-European family of languages, and in those we are now discussing, the verb follows this general law. To make this clear, it may be well to give here, as a preliminary matter, a slight sketch of the structure of the verb as it stands in the Sanskrit and Prakrit stages of development.

v3.c17.9

In that stage of the Sanskrit language which is usually accepted as the classical one, the verb is synthetical throughout, except in one or two tenses where, as will be hereafter shown, the analytical method has already begun to show itself. By separating the inflectional additions, and unravelling the euphonic changes necessitated by them, we may arrive at a residuum or grammarian's abstraction called the root. These roots, which have no real existence in spoken language, serve as useful and indispensable pegs on which to hang the long chain of forms which would otherwise defy all attempts at reducing them to order. Some writers have lately thought fit to sneer at the philologist and his roots, and have made themselves merry over imaginary pictures of a time when the human race talked to each other in roots only. These gentlemen set up a bugbear of their own creation for the purpose of pulling it to pieces again. No one, as far as I am aware, has ever asserted that at a given period of the world's history a certain race of men used such words as bhú, gam, or kar, till some one hit on the ingenious device of adding to bhú the word ami, and, modifying bhú into bhava, burst upon his astonished countrymen with the newly-discovered word bhavâmi, "I am." What has been asserted, and truly too, is that in Sanskrit we find a large number of words expressing the idea of "being," in which the consonantal sound bh is followed by various vowels and semivowels, which, according to phonetic laws, spring from the vowel û, and that as, for scientific purposes, some common generic term is required to enable us to include under one head all parts of the verb, we are justified in putting together these two constant unvarying elements, and so obtaining a neat technical expression bhú, to which, as to a common factor, can be referred all the words expressive of "being" in its relations of time, person, and condition. Analysis and arrangement of this sort is an essential part of every science, and the native grammarians had done this much work for us before European skill was brought to bear on the subject.

v3.c17.10

Verbal roots, then, are grammarians' tickets, by which actual spoken words are classified and arranged in groups for convenience of investigation. The roots in Sanskrit are mostly monosyllabic, consisting of a consonant followed by a vowel, as bhú, yá, ní, or of a vowel followed by a consonant, as ad, ish, ubh, or of a vowel between two consonants, as kar, gam, pat. Roots may also consist of a single vowel, as i, and in the place of a single consonant there may be a nexus, as grah, pinj, mlai. Those roots which have more than one syllable are usually of a secondary nature, being in some cases produced by reduplication, as jágar, in others made from nouns, as kumâr.

v3.c17.11

Each verbal root presents six phases or grades of action: active, neuter, passive, causal, desiderative, intensive. All these are distinguished by certain modifications of the letters of the root, and by certain prefixed and affixed syllables. Thus भू *bhú*, “to be,” undergoes the following modifications:

v3.c17.12

| *Active* | |
| *Neuter* | } bhava. |
| *Passive* | bhûya. |
| *Causal* | bhâvaya. |
| *Desiderative* | bubhûsha. |
| *Intensive* | bobhûya. |

v3.c17.13

The causal also is in some cases treated as primary stem, and gives rise to subsidiary forms; thus from *pâtaya* “cause to fall,” is made a passive *pâtya*, whence comes a desiderative causal *pipâtayisha*.

v3.c17.14

Each of these six phases may be conjugated throughout thirteen tenses, in each of which are nine forms representing the three persons of the singular, dual, and plural. It rarely happens in practice that any one verbal root exhibits the whole of these forms, but if we regard the general type, we may fairly say that a Sanskrit verb, as an individual entity, is an aggregate of seven hundred and two words, all agreeing in expressing modifications of the idea contained in the root-syllable, which is the common inheritance of them all.¹ Of the thirteen tenses, nine are conjugated according to certain rules which, with some exceptions, hold good for all verbs in the language, but the remaining four tenses are subject to rules by which they are divided into ten classes or conjugations. These four are the present, imperfect, imperative, and optative; and before we can determine what form a verbal root can take in any of these tenses, we must know what conjugation it belongs to.

v3.c17.15

¹ Namely, 6 phases × 13 tenses × 9 persons = 702. But this is an extreme calculation, for the Subjunctive (Leṭ) is only found in Vedic Sanskrit; and the two forms of the Perfect (Liṭ) may be regarded as variations of the same tense. Thus the number of tenses may be reduced to ten, viz. Present (Laṭ), Imperfect (Laṅ), Optative (Liṅ), Imperative (Loṭ), Perfect (Liṭ), Aorist (Luṅ), Future (Lṛṭ), Conditional (Lṛṅ), Second Future (Luṭ), Benedictive (âśir Liṅ). By this reckoning the number of forms would be 6 × 10 × 9 = 540.

v3.c17.16

Inasmuch also as the Sanskrit grammarians class the active and neuter phases together, we must find out which of these two phases any given verb employs, for the terminations of the tenses and persons are different. Some verbs employ both, but the majority are conjugated only in one of the two, and as there is no rule as to which of the two is to be used, the dictionary is our only guide. The active, or Parasmaipada, as it is called, stands to the neuter, or Âtmanepada, in the same relation as the active in Greek does to the middle voice, and the resemblance is the greater, in that the Âtmanepada, like the middle voice in Greek, uses the terminations of the passive.

v3.c17.17

Although each of the seven hundred and two words which make up the complete typical Sanskrit verb contains the common root-syllable, yet this syllable does not appear in the same form in each word, but is subject to certain euphonic and other influences which affect both the vowels and consonants composing it, and often materially alter its shape. Thus the verbal root KAR, “do,” appears in classical Sanskrit in the following forms:

v3.c17.18

1. कृ Kri, in 1 du. pf. Par. chakriva, 1 pl. id. chakrima, 2 s. pf. Âtm. chakrishe, 1 du., 1 and 2 pl. id. chakrivahe, chakrimahe, chakridhve; in the whole of the 1 aor. Âtm., as akrishi, akrithâh, akrita, etc.; in the pass. part. kritah, and gerund kritvâ, and in the benedictive Âtm., as krishishta, etc.

v3.c17.19

2. क्रि kri, in bened. Par., as kriyâsam, kriyâh, kriyât, etc., and in the passive present, as kriye, kriyase, kriyate, etc.

v3.c17.20

3. कर kar, in pres. Par., as karomi, karoshi, karoti, and before all weak terminations.

v3.c17.21

4. कुर kur, in pres. Âtm., as kurve, kurushe, kurute, and before strong terminations.

v3.c17.22

5. कार kâr, in pf. Par., as chakâra, and 1 aor. Par., as akârsham, also in the causal, as kârayati.

v3.c17.23

6. क्र kr, in 2 and 3 pl. pf. Par., chakra, chakruh, and 1 and 3 s. pf. Âtm. chakre.

v3.c17.24

In the same way the root ÇRU “hear,” appears in some parts of the verb as çri, in others as çru, çrû, çriṇ, and çrâv. In the whole range of verbal roots there is perhaps not one which does not undergo more or less modification in the course of being conjugated.

v3.c17.25

Not only does the root-syllable present itself in various forms in the several tenses, but the terminations of the nine persons differ in each tense, and sometimes one tense will have two sets of terminations. Moreover, the endings of any given tense in one phase, differ from the corresponding ones of the same tense in another phase. Thus the terminations of the present tense are in the active phase

v3.c17.26

| Singular | 1. ami. | 2. si. | 3. ti. |
| Dual | 1. avah. | 2. thah. | 3. tah. |
| Plural | 1. amah. | 2. tha. | 3. nti. |

v3.c17.27

But in the middle phase the same tense ends in

v3.c17.28

| Singular | 1. i. | 2. se. | 3. te. |
| Dual | 1. avahe. | 2. ithe. | 3. ite. |
| Plural | 1. amahe. | 2. dhve. | 3. nte. |

v3.c17.29

This slight outline will suffice to show how vast and intricate are the ramifications of the Sanskrit verb. The reader who has followed the steps by which the noun has been simplified, as shown in the second volume of this work, will not be surprised to find in the present volume how widely the modern verb differs from that of Sanskrit. It was impossible to reduce the verb to anything like the simplicity required by modern speakers without sacrificing by far the greater portion of the immense and unwieldy apparatus of ancient times.

v3.c17.30

§ 2. Owing to the want of a continuous succession of literary documents, such as exists in the case of the modern Romance

v3.c17.31

languages of Europe, it is scarcely possible to trace step by step the changes which have occurred in the verb. It is necessary, however, to make the attempt, and to piece together such evidence as we have, because the modern verb is an undoubted descendant of the ancient one, though only a slight trait here and there recalls the features of its parent, and its structure in many points can only be rendered intelligible by tracing it back to the ancient stock whence it sprung.

v3.c17.32

The first steps in the direction of simplification occur in Sanskrit itself. Many of the elaborate forms cited by grammarians are of very rare occurrence in actual literature, and some of them seem almost to have been invented for the sake of uniformity. Three instances of this tendency in classical Sanskrit may here be noticed.

v3.c17.33

The perfect tense in Sanskrit, as in Greek, is usually formed by reduplication, so we have from √ तप् “burn,” pf. तताप, √ दृश् “see,” pf. ददर्श, just as λείπω makes λέλοιπα and τρέπω, τέτροφα. But there are certain roots which cannot take reduplication, and these form their perfect by an analytical process. The root is formed into a sort of abstract substantive in the accusative case, and the perfect of an auxiliary verb is added to it. The verbs भू “be,” अस् “be,” and कृ “do,” are the auxiliaries principally employed for this purpose. Thus— √ उंद् “wet,” makes pf. उंदां चकार, उंदां बभूव or उंदां आस. √ चकास् “shine,” ” ” चकासां चकार, etc. √ बोधय “explain,” ” ” बोधयां चकार, etc.¹

v3.c17.34

¹ Max Müller's Sanskrit Grammar, p. 172.

v3.c17.35

Another instance of the analytical formation is seen in the future tense made out of the agent of the verb with the present tense of the auxiliary अस् “be.” Thus from √ बुध “know,” comes the agent बोधिता, which with the present of अस् makes S. 1. बोधितास्मि P. 1. बोधितास्मः
2. बोधितासि 2. बोधितास्थ.

v3.c17.36

A third instance is a form of phrase in which the passive past participle is combined with this same auxiliary अस् to form a perfect definite, as आगतोऽस्मि “I have come,” or, as more faithfully represented by other European languages, “je suis venu,” and as we sometimes say ourselves, “I am come.” Here an analytical construction supplies the place of the perfect. Closely allied to this is the frequent habit in writers of the classical style of expressing the same tense by the neuter of the p.p.p. with the subject in the instrumental, as तेन गतं “by him gone,” *i.e.* “he went,” instead of जगाम.

v3.c17.37

These are the first faint indications of a method which, in the course of ages, has developed to such an extent as to constitute the leading principle in the organization of the modern verb. By this system a greater facility for expressing nice shades of meaning is obtained. जगाम may mean “he went,” or, “he has gone,” but by the other system each of these two meanings has a phrase peculiar to itself, गतोऽस्मि meaning “he has gone,” and तेन गतं “he went.” Precisely in the same way the Latin had only *ego amavi* for “I loved” and “I have loved,” but the Romance languages found this insufficient, and they have—

v3.c17.38

v3.c17.39

§ 3. The next step in the reduction of the numerous Sanskrit tenses to a more manageable compass is seen in Pali, originally an Indian Prakrit, but which became the sacred language of the Buddhists of Ceylon, having been carried thither in the middle of the third century¹ before Christ, by Mahendra, son of King Açoka, and spread thence to Burma and Siam.

v3.c17.40

¹ Kuhn, Beiträge zur Pali Grammatik, p. 1. But Turnour, Mahawanso xxix., gives B.C. 307. So also Childers, preface, p. ix.

v3.c17.41

Although the Pali grammarians, in their anxiety to exalt their sacred speech, tell us that the verb has ten conjugations, yet examples of all these are but rarely found.¹ Four of the ten Sanskrit conjugations, the first, fourth, sixth, and tenth, resemble each other very closely even in that language, and are easily brought down to one in Pali. The seventh of Sanskrit also loses somewhat of its peculiar type, which consists in inserting न् between the vowel of the root and the final consonant, or न् before weak terminations. Thus in Skr. √रुध् rudh, “to obstruct,” makes its present रुणद्धि runaddhi, but in Pali, while the न् is retained, the present is rundhati, after the type of the first class.

v3.c17.42

¹ Seven classes are given by Kaccâyana. See Senart, Journal Asiatique, vi. série, vol. xvii. p. 439.

v3.c17.43

Five out of the ten Sanskrit conjugations are thus reduced almost, if not entirely, to one. Of the remaining five, the second of Sanskrit in roots which end in a vowel exhibits some traces of Sanskrit forms, while in those which end in a consonant the types of the first, or Bhû, class prevail. Thus Skr. √या “to go,” pr. याति, Pali also yâti, but

v3.c17.44

Skr. √मृज् “to rub,” pr. मार्ष्टि. Pali majjati, as if from a Skr. *मर्जिति*. √दुह् “to milk,” ,, दोग्धि. ,, dohati. √लिह् “to lick,” ,, लेढि. ,, lehati.

v3.c17.45

The third conjugation occasionally takes the reduplication as in Sanskrit, but in many instances prefers the Bhû type. Thus Skr. √भी “to fear,” बिभेति. Pali भायति. √धा “to hold,” दधाति. ,, दधाति and दहति.

v3.c17.46

The verb dâ, “to give,” which belongs to this conjugation, has special developments of its own, and is discussed in § 16.

v3.c17.47

The fifth, eighth, and ninth classes are very similar even in Sanskrit, for while the fifth adds नु to its root, the eighth adds उ; but as all its roots except one already end in न्, it comes practically to pretty much the same thing as the fifth. The ninth adds न्, ना, and नी to the root before various terminations. Here Pali draws very slight distinctions, making verbs of the fifth class take नु and ना indifferently, and both fifth and ninth appear occasionally in the guise of the first. Thus—

v3.c17.48

v3.c17.49

The reason why the forms of the Bhû conjugation exercise so great an influence, and, like the -as-stem in nouns, so largely displace all the other types, is probably that the first conjugation is by far the largest, containing upwards of nine hundred out of the two thousand roots said to exist in Sanskrit. The second conjugation has only seventy-three, the third but twenty-five, the fourth and sixth about one hundred and forty each. The tenth, it is true, contains four hundred, but it is identical in form with the causal. The fifth has only thirty-three, the ninth sixty-one, while under the seventh class are twenty-five, and under the eighth only nine. These figures, it must be added, are taken from the Dhâtupâtha, a grammarian’s list of roots,¹ which contains many roots seldom, if ever, found in use, so that for all practical purposes the first conjugation covers more than half the verbs in the language. When it is also remembered that the fourth, sixth, and tenth differ but slightly from the first, it is not surprising that the terminations common to these four conjugations should have fixed themselves in the popular mind, and been added by the vulgar to all roots indiscriminately. Nearly all those verbs which retain the type of any conjugation, except the first, are words of extremely common use, which would naturally keep their well-known forms in the mouths of the people in spite of all rules and tendencies to the contrary.

v3.c17.50

¹ Westergaard, Radices Sanskr. p. 342.

v3.c17.51

§ 4. The dual number has entirely disappeared from Pali, and the Âtmanepada, or middle phase, has practically merged into the active, for although Kaccâyana (J. As., vol. xvii. p. 429, sûtra 18) gives terminations for it, yet it is admitted that those of the active may be used instead, and practically it would appear that they are so used. The other phases, as causal, passive, desiderative, and intensive, have their own forms as in Sanskrit.

v3.c17.52

Among the tenses the chief is the present, and it is in Pali that we first find a tendency to retain throughout the whole verb that form of the root which is in use in the present. This tendency grows stronger in the later Prakrits, and becomes an almost invariable rule in the modern languages. Thus—

v3.c17.53

v3.c17.54

Phonetic influences in Sanskrit change this root as regards its final consonant in the different tenses, but Pali, having got hold of the form pach in the present tense, retains it throughout the verb. It is still, however, only a tendency, and not a law, for we find instances in which Pali forms are derived directly from the corresponding tense in Sanskrit. One who should attempt to learn Pali without reference to Sanskrit would find it difficult to understand how the words karoti, kubbati, kayirâ, kâhâmi, akâsi, kattum, could all spring from the same verbal root. It is only when the corresponding Sanskrit forms karoti, kurvate, kuryât,¹ kartâsmi, akârshît, kartum, are put by their side, that the thread which connects them all becomes evident. Just so in the Romance languages, Italian *so, sa, sapete, sanno, seppi*, seem to have very little beyond the initial *s* in common, till it is perceived that they come from the Latin *sapio, sapit, sapitis, sapiunt, sapui*; thus, also, *ho* and *ebbi* can only be seen to be parts of the same verb when their origin from Latin *habeo* and *habui* is recognized. In Spanish there is the same difficulty, as will be seen by comparing *hacer, hago, hice, hare*, and *hecho*, with their Latin originals *facere, facio, feci, facere habeo*, and *factum*. In Portuguese, which seems to be the lowest and most corrupt Apabhrança of the Romance Prakrits, the changes are such as almost to defy analysis. For instance, *ter, tenho, tinha, tive, terei*, correspond to Latin *tenere, teneo, tenebam, tenui, tenere habeo*: also *hei, houve, haja*, to *habeo, habui, habeam*, and *sou, he, foi, seja*, to *sum, est, fui, sit*.¹

v3.c17.55

¹ Or more strictly from an older karyât not in use in classical Sanskrit. Kuhn, Beiträge, 105.

v3.c17.56

¹ Diez, Gramm. d. Romanischen Sprachen, vol. ii. p. 188.

v3.c17.57

The tenses of the Pali verb are eight in number.² These correspond to the tenses of the Sanskrit verb, omitting the periphrastic or second future (luṭ), the benedictive (âçir liṅ), and the subjunctive (leṭ). The present active is almost exactly the same as the Sanskrit as regards its terminations in the Bhû form, and the middle only differs, and even then very slightly, in the 1 and 2 plural. Thus—

v3.c17.58

² The materials for this section are taken chiefly from Kuhn, Beiträge, p. 93 seqq., with some additions from Childers's Dictionary, and a few remarks of my own.

v3.c17.59

Skr. 1 pl. पचामहे. 2. पचध्वे. Pa. 1. पचाम्हे. 2. पचध्वे.

v3.c17.60

In this tense, as in many others, Pali is not very instructive, it clings too closely to the Sanskrit. It is, however, necessary to give a sketch of its forms, because they exhibit the first traces of that gradual change which has led to the modern conjugation. Even when the Pali conjugates a verb according to any class other than the Bhû, it still keeps the personal endings of Sanskrit for that conjugation; thus from √ या “go,” we have—

v3.c17.61

Pa. S. 1. यामि, 2. यासि, 3. याति; P. 1. याम, 2. याथ, 3. यांति.

v3.c17.62

which differs from Sanskrit only in omitting the visarga in P. 1.

v3.c17.63

The imperative follows the type of the present, and may be thus compared with Sanskrit Parasmaipada,

v3.c17.64

Skr. S. 1. पचानि, 2. पच, 3. पचतु; P. 1. पचाम, 2. पचत, 3. पचंतु. Pa. S. 1. पचामि, 2. पचहि, 3. पचतु; P. 1. पचाम, 2. पचय, 3. पचंति.

v3.c17.65

and with the Âtmanepada, thus—

v3.c17.66

Skr. S. 1. पचै, 2. पचस्व, 3. पचतां; P. 1. पचामहै, 2. पचध्वं, 3. पचंतां. Pa. S. 1. पचे, 2. पचस्सु, 3. पचतं; P. 1. पचामसे, 2. पचह्वो, 3. पचं तं.

v3.c17.67

Here the S. 1 Parasmai seems to have arisen from some confusion with the present, as also P. 2. Noteworthy is S. 2, with its ending हि, which, though only found in classical Sanskrit in the second, third, seventh, and ninth conjugations, has crept into all in Pali, and has continued on into the mediæval period, thus Chand

v3.c17.68

तिन सु गल्ह अच्छी कहहि॥

v3.c17.69

“Say thou a good word of them.”—Pr. R. i. 9.

v3.c17.70

where कहहि = Skr. कथय (हि). In Vedic Skr. हि appears in all the conjugations. Of the Âtmane forms P. 1 seems to be derived from an older form, masai. P. 2 should perhaps be read hvo, not vho, in which case it is a regular resultant from Sanskrit dhv.

v3.c17.71

The potential is the Sanskrit optative (lin), thus—

v3.c17.72

PARASMAI.

v3.c17.73

Skr. S. 1. पचेयं, 2. पचेस्, 3. पचेत्; P. 1. पचेम, 2. पचेत, 3. पचेयुस्. Pa. S. 1. पचेय्यामि, 2. ०य्यासि, 3. ०य्य; P. 1. ०य्याम, 2. ०य्याथ, 3. ०य्युं.

v3.c17.74

ÂTMANE.

v3.c17.75

Sk. S. 1. पचेय, 2. पचेथास्, 3. पचेत; P. 1. पचेमहि, 2. पचेध्वं, 3. पचेरन्.

v3.c17.76

Pa. S. 1. पचेय्यं, 2. पचेय्यो, 3. पचेय्य; P. 1. पचेय्याम्हे, 2. ०य्यहो, 3. पचेरं.

v3.c17.77

In this tense the point specially to be noticed is the tendency to simplify not only the root-syllable, but the range of terminations also. Having got the syllables *eyya* as the type of the tense, Pali seeks to avoid all further distinctions, and to use as much as possible the personal endings of the present tense. It sometimes conjugates the potential according to the types of other classes, and in this respect follows the lead of the present less faithfully in this tense than in the imperative. Thus, though in the present and imperative of *kar*, it follows the Sanskrit, and has *karoti*, *karotu*, yet in the potential it treats *kar* as if it belonged to the Bhû class, and has *kareyyâmi* as though from a Sanskrit *kareyam* instead of the actual *kuryâm*. There are other peculiarities about this tense which are not here noticed, as having no bearing upon the subject of the modern languages.

v3.c17.78

The imperfect has been, to some extent, mixed up with the aorist (*luṅ*), and both, together with the perfect, lead us into considerations which are of interest only for Pali itself, not having survived or had any influence on modern developments. They may therefore be passed over as immaterial to our present inquiry.

v3.c17.79

The future, on the contrary, offers many interesting peculiarities, especially, as will be seen hereafter, in reference to Gujarati and some of the rustic dialects of Hindi. The future is a different tense in the modern languages, and every scrap of information which can help to elucidate it deserves special notice. It runs thus in Pali (√ गम् “go”)—

v3.c17.80

Skr. S. 1. गमिष्यामि, 2. ०ष्यसि, 3. ०ष्यति; P. 1. ०ष्यामस्, 2. ष्यथ, 3. ष्यन्ति.

v3.c17.81

Pa. S. 1. गमिस्सामि, 2. ०स्सिसि, 3. ०स्सति; P. 1. ०स्साम, 2. स्सथ, 3. स्सन्ति.

v3.c17.82

Here the only noteworthy feature is the change of ष्य into स्स. The Âtmanepada follows the same rule throughout. Although the tendency to keep that form of the root which exists in the present leads to divergences from the Sanskrit future type, yet instances occur in which the Sanskrit type is preserved. These occur in reference to that very troublesome feature in the Sanskrit verb, the intermediate इ, which is sometimes inserted between the root and the termination, and sometimes not. When it is not inserted, the euphonic laws of Sanskrit require that the final consonant of the root be changed to enable it to combine with the initial consonant of the termination. Thus √ पच् “cook,” when it has to take the future termination ष्यति, becomes पक् and पक् + ष्यति = पक्ष्यति. Here Pali sticks to the form पच्, because it is used in the present and makes its future पचिस्सति as though there had been (as there probably was in colloquial usage) a Sanskrit future पचिष्यति with the intermediate इ inserted.

v3.c17.83

In a certain number of verbs, however, it has two forms, one as above retaining the root-form of the present, and the other a phonetic equivalent of the Sanskrit. Kuhn¹ gives the following examples, to which I add the Sanskrit for comparison.

v3.c17.84

¹ Beiträge, p. 115.

v3.c17.85

Skr. √ लभ् “get,” future लप्स्यते. Pali लच्छति but also लभिस्सति. √ वच् “speak,” ,, वक्ष्यते. ,, वक्खति. √ धा “put,” ,, धास्यति. ,, धस्सति. √ वस् “dwell,” ,, वत्स्यति. ,, वच्छति but also वसिस्सति. √ छिद् “cleave,” ,, छेत्स्यति. ,, छेच्छति ,, छिन्दिस्सति. √ भुज् “eat,” ,, भोक्ष्यति. ,, भोक्खति ,, भुंजिस्सति. √ मुच् “loose,” ,, मोक्ष्यति. ,, मोक्खति ,, मुंचिस्सति. √ श्रु “hear,” ,, श्रोष्यति. ,, सोस्सति ,, सुनिस्सति.

v3.c17.86

The consonantal changes are in accordance with the treatment of the nexus as explained in Vol. I. p. 304. The striving after uniformity is seen, however, in the retention of the alternative forms having the same type as the present, and it is, moreover, worth observing that the forms which reproduce the type of the Sanskrit without the intermediate द् seem by degrees to have been misunderstood. The illiterate masses, and even those better instructed, seem to have missed the issati which so generally indicated to their minds the future tense, and regarded those forms which had not this familiar sound as present tenses. So they made double futures by adding the issa to them. Thus from दृश् “to see,” future द्रक्ष्यति, Pali made a form dakkhati, but the people by degrees took this for a present, and made what to them seemed a more correct future dakkhissati. I mention this here as I shall have occasion hereafter to discuss the much-debated question of the origin of the familiar modern stem dekh “see” (see § 17). Another instance is

v3.c17.87

§ 5. It used to be held that Pali was a descendant of the Mâgadhi dialect of Prakrit, but this opinion is now, I believe, exploded. Though the question is not yet set at rest, it would seem to have been fairly established that Mahendra was a

v3.c17.88

Skr. √ शक् “be able,” future शक्ष्यति. Pa. सक्खति, whence vulgo सक्खस्सति.

v3.c17.89

In one case Pali has a future which points back to a Vedic form:

v3.c17.90

Skr. √ रुद् “weep.” Vedic future रोत्स्यति. Pa. रोच्छति. Classic ditto रोदिष्यति. ,, रोदिस्सति.

v3.c17.91

Occasionally the स्स is softened to ह, as in काहति, काहिति from करिस्सति, Skr. करिष्यति. This is noteworthy with reference to Bhojpuri and the eastern Hindi dialects generally.

v3.c17.92

native of Ujjayin, and that the language which he carried to Ceylon was the ordinary vernacular of his own province.¹ This dialect was not very different from that of Magadha, and Mahendra may have slightly altered the Mâgadhi sayings of the great master, by his Ujjayini pronunciation, while retaining the name Mâgadhi out of deference to the sacred associations which clustered round the birthplace of Buddha.

v3.c17.93

¹ Kuhn, Beiträge, p. 7.

v3.c17.94

Be this as it may, the nearest Indian dialect to Pali seems undoubtedly to be the Prakrit of the Bhâgavatî, a sacred book of the semi-Buddhist sect of Jainas. If Hemachandra, himself a Jain and author of several works on Prakrit, were available for reference, our task would be easier; as yet, however, none of Hemachandra’s writings have been printed or edited. Weber’s articles on the Bhâgavatî are at present our only source of information.²

v3.c17.95

² Fischel’s admirable edition of Hemachandra’s Grammar (Orphanage Press, Halle, 1877) has reached me just as this work is going to press, and too late to be of use for this edition, except for a few hasty notes here and there. Mueller’s Beiträge zur Grammatik des Jainaprakrit came into my hands about the same time. I find it enables me to add a few illustrations to this section, which, however, was written in the latter part of 1875.

v3.c17.96

In the Jaina Prakrit the ten conjugations of the Sanskrit verb are, with few exceptions, reduced to the Bhû type. In this respect it goes further than Pali, treating as verbs of the first conjugation many which in Pali retain the type of other conjugations. The fifth, seventh, and ninth conjugations, which in Sanskrit insert न with certain variations, are all reduced to one head by regarding the न as part of the root, as is also the case with the य of the fourth class. The a inserted between the root and termination of the Bhû class is used throughout, though occasionally weakened to i, or changed to e from some confusion between this and the e = aya, which is the type of the tenth class. The following examples will illustrate the above remarks.

v3.c17.97

Skr. √ हृ “take,” i. हरति Jaina हरति, हरइ. √ विद् “know,” ii. वेत्ति ,, वेदेइ. Pl. 3 विदन्ति ,, वेदेन्ति. √ धा “put,” iii. दधाति with अपि, अपिदधाति ,, पिहेइ “puts on (clothes)” and पिहाइ. √ सिध् “succeed,” iv. सिध्यति ,, सिज्झइ. but आराध् “propitiate,” आराधयति ,, आराहइ. √ आप् “get,” v. आप्नोति. with प्र, प्राप्, प्राप्नोति ,, पाउणइ, the न being treated as part of the root. √ चि “gather,” v. चिनोति ,, चयति, चयइ, but also चि- नाइ, with the same confusion between the नो of v. and ना of viii. as occurs in Pali. √ श्रु “hear,” v. शृणोति with प्रति, प्रतिशृणोति ,, पडिसुणेइ “promises.” √ स्पृश् “touch,” vi. स्पृशति ,, फुसेइ. √ भञ्ज् “break,” vii. भनक्ति ,, भंजइ. √ कृ “do,” viii. करोति ,, करइ. √ ग्रह “take,” ix. गृह्णाति ,, गेण्हइ, here again the ण has passed into the root. √ ज्ञा “know,” ix. जानाति ,, जाणइ.

v3.c17.98

The tenth class being identical with the first is omitted. It will be seen that the present tense is formed throughout on the model of the first conjugation, the Jain words given above being phonetic modifications of words which would be in Sanskrit respectively harati, vedati, dhâti, ârâdhati, prâpanati, chayati, suṇati, bhanjati, karati, grihṇati, and jânati, if all those verbs belonged to the first or Bhû conjugation.

v3.c17.99

It is not so easy to draw out a full verbal paradigm in this dialect as in Pali, because we have as yet no grammars, and are obliged to fall back on the words that occur in a single text. The range of tenses appears to consist of a present (corresponding to the Sanskrit laṭ), imperative (loṭ), potential (liṅ), imperfect and aorist jumbled together as in Pali, and future (lṛit). The perfect (liṭ) seems to be altogether wanting, as it is in the modern languages.

v3.c17.100

The present runs thus:—√ नम “bow.”

v3.c17.101

S. 1. नमामि, 2. नमसि, 3. नमति; P. 1. नमामो, 2. नमह, 3. नमंति. नमेमि, नमेसि, नमेति; नमेमो, नमेंति. नमइ; नमंति. नमेइ.

v3.c17.102

Those terminations which contain the vowel e have crept into the conjugation of all verbs from the tenth, to which that vowel, as shortened from aya, must be held strictly to belong, or to causals. Thus in Bhâg. i. 60, we have phâseti, pâleti, sobheti, tîreti, pûreti, kiṭṭeti, aṇupâlei, árâhei, for Sanskrit स्पर्शयति, पालयति, शोभयति, तारयति, पुरयति, कीर्त्तयति, अनुपालयति, आराधयति, respectively. In the last word the causal form becomes the same as the active given above. Of the imperative we have only the S. 2 and P. 2, which are in fact the only persons which an imperative can properly have. The S. 2 takes the ending हि as in Pali with junction vowels á and e, the P. 2 ends in ह, which, as Weber points out, is from the P. 2 of the present, in Sanskrit थ. Thus—

v3.c17.103

Skr. √ रुच् “shine,” causal रोचय, impv. रोचय, Jaina रोएहि. श्रद्धा “believe,” ,, ,, श्रद्धेहि, ,, सद्दहाहि (pres. सद्दहइ). √ बंध् “bind,” impv. P. 2. बधीत, ,, बंधह.

v3.c17.104

The potential, of which only the S. 3 is traceable, resembles Pali in using the termination eyya with variant ejja. Skr. √ गम् “go,” S. 3. गच्छेत् Jaina गच्छेय्य, गच्छेज्ज. √ ग्रह “take,” ” गृह्णीयात् ” गेहेज्ज.

v3.c17.105

But there exist some old simple forms derived by phonetic changes from the corresponding Sanskrit tense, as kujja = kuryât, dajja = dadyât (Mueller, p. 60).

v3.c17.106

The future resembles that of Pali, thus— S. 1. नमिस्सामि, 2. ॰इस्ससि, 3. ॰इस्सइ; P. 1. ॰इस्सामो, 2. ॰इस्सह, 3. इस्संति.

v3.c17.107

It also appears with a termination ihi produced by weakening स् into ह and the following a to i, thus— Skr. गमिष्यति, Jain गच्छिहिति and गमिहिति.

v3.c17.108

Moreover, there is a trace of the double future like Pali dakkhissati.

v3.c17.109

Skr. √ पद् “go,” with उप, उपपद् “attain,” future उपपत्स्यते, Jaina उववज्जहिति.

v3.c17.110

Here उपपत्स्यते would phonetically become उववच्छइ, and by still further softening उववज्जइ, whence, as if from a present, is formed the future उववज्जइस्सइ and उववज्जहिति.

v3.c17.111

§ 6. The reduction in the number of tenses necessitates a greatly extended use of participles. This is one great step in the transition from the synthetical to the analytical system. The Sanskrit present active participle takes in that language the characteristics of the ten conjugations, and is declined as a noun in three genders. It ends properly in ant, but the nasal is dropped before certain terminations, as

v3.c17.112

| M. | F. | N. |
| पचन् | पचन्ती | पचत् |
| रुधन् | रुधती | रुधत् |

v3.c17.113

The nasal, however, is retained throughout in Jaina Prakrit, thus—

v3.c17.114

Skr. ज्वलन् ज्वलन्ती ज्वलत्. Jaina जलंतो जलंती जलंतं.

v3.c17.115

This peculiarity is worth remembering; much depends on this retention of the nasal, as will be seen when we come to the modern Sindhi and Panjabi verbs.

v3.c17.116

Very great interest attaches to the participle of the future passive, which in Sanskrit ends in तव्य. In verbs which do not take intermediate इ, this ending is added directly to the root with the usual Sandhi changes; but as Prakrit prefers to insert the इ in order to preserve the root-form of the present, it comes to pass that the त of the termination stands alone between two vowels, and in consonance with Prakrit phonetics is elided. The hiatus thus produced is in the Jaina writings filled by य. If to this we add the regular mutation of व्य into व्व, we get from तव्य the form व्व. In its original meaning this participle corresponds to the Latin in ndus, as faciendus, and expresses that which is to be done, as त्वया गन्तव्यं “by thee it is to be gone,” i.e. “thou must go.” In this sense it occurs frequently in Bhâgavatî, as for instance in § 56:

v3.c17.117

Jaina एवं देवाणुप्पिया गंतव्यं, चिट्टियव्यं, निसीतियव्यं, भुंजियव्यं, etc. Skr. एवं देवानुप्रिया गन्तव्यं, स्थातव्यं, निषत्तव्यं, भोक्तव्यं, etc.

v3.c17.118

“Thus, O beloved of the gods, must ye go, must ye stand, must ye sit, must ye eat,” where the last two words postulate a Sanskrit form with the इ inserted, such as निषीतितव्यं, भुंजितव्यं.

v3.c17.119

It is obvious that it would require no great straining of the sense of this participle to make it into an infinitive, and seeing that as early as this Jaina dialect the use of the regular Sanskrit infinitive in तुं has become rare, it follows that recourse should be had to some participial form to supply its place. In this way we find the past passive participle in इत, with the त elided and its place supplied by य, employed in a construction where we should expect the infinitive. Thus Bhâg. § 54, इच्छामि पव्वावियं, मुंडावियं, सेहावियं, सिक्खावियं (Weber, Bhâg. p. 274): “I wish to wander, to take the tonsure, to practise austerities, to learn,” as though from Sanskrit forms प्रव्राजितं, मुण्डापितं, सेधापितं, शिक्षपितं, the three last being causals formed with áp, as is frequently the case with causals in Prakrit, though of course these forms are not found in Sanskrit. In that language the formation of causals by means of प is restricted to a few stems.

v3.c17.120

More will be said on this subject in a subsequent chapter, but it is necessary here to note an early instance of this process which takes a much wider development in later times, the infinitive in Gujarati and Oriya and several participial constructions and verbal nouns being derived from it.

v3.c17.121

§ 7. The scenic Prakrits represent a further step in development. Despite the admittedly artificial character of these dialects, they probably retain forms which were at one time in general use, although that time may not have been the epoch when the dramas were written, and without referring to them, the structure of the modern verb could not be clearly understood. It is expedient to avoid discussing this question, lest attention should be drawn away from the real subject of this work, namely, the modern languages. All this part of the present chapter is merely introductory and is only inserted in order to pave the way for a more intelligent appreciation of the origin and growth of Hindi and its fellows.

v3.c17.122

In the Mâhârâshtri or principal poetical dialect all conjugations are reduced to the type of the first or Bhû class, and the same holds good for the Çauraseni or chief prose dialect. Only here and there do we find faint traces of the peculiarities of other conjugations. Of the six phases only three remain, active, passive, and causal. The passive differs from the active only in the form of the root, the characteristic य of the Sanskrit passive having been worked into the stem, and the terminations of the active being added to it. The Âtmanepada and the dual are of course rejected.

v3.c17.123

Of tenses these dialects have a still more restricted range than the Jaina Prakrits. They have the present, imperative and future, with traces of the potential. The past tense is chiefly formed by the p.p.p. with auxiliary verbs. Thus from √रुच् “shine,”

v3.c17.124

Present S. 1. रोचामि, 2. रोचसि, 3. रोचदि. रोचमि, रोचइ. रोचम्हि,

v3.c17.125

P. 1. रोचामो, ॰मु, ॰म, 2. रोचध, ॰ह, 3. रोचन्ति. रोचम, ॰मु, ॰म्हो, ॰म्ह, ॰त्य, ॰धं. रोचमो, ॰मु, ॰इत्य.

v3.c17.126

Here are observable those first indications of a confusion of forms, and uncertainty in their use, which are always characteristic of that period in languages when the synthetical structure is breaking down into the analytical. In these dialects, as in Jaina Prakrit, the practice exists of inserting प् as a junction vowel; thus we have such forms as कलेमि “I do,” Skr. करोमि, instead of करामि, which would be the regular result of treating कर as a Bhû verb, गच्छेम्ह for गच्छाम, “let us go.” The presence of the ह in S. 1 and P. 1 is accounted for by its being confused with that construction in which the present of अस् is used with a past participle; thus we find कदम्हि “I was made” = Sanskrit कृतोऽस्मि, and पेसिदम्हि “I have been sent” = Skr. प्रेषितोऽस्मि.

v3.c17.127

The imperative has the following forms—

v3.c17.128

S. 2. रोच 3. रोचदु. P. 2. रोचध 3. रोचंतु. रोचाहि रोचउ. रोचह.

v3.c17.129

The S. 2 has also forms रोचस्स, रोचस्सु, pointing to a Sanskrit Âtmane form रोचस्व and P. 2 similarly रोचधं = Skr. रोचध्वं, though neither are used in a middle sense, but are equivalents as regards meaning of the Sanskrit active.

v3.c17.130

The following are a few examples: पेक्खस्स “look thou!” Skr. प्रेक्षस्व. णमस्स “bow thou!” ” नमस्व. वहध “bear ye.” ” वहत. याध “go ye.” ” यात. अवेध “go away.” ” अपेत. ओशूलध “get out of the way!” ” अपसरत. कलेध “do.” ” कुरुत. जगगेध “wake up.” ” जाग्रत.¹

v3.c17.131

¹ Some of these are Mâgadhi Prakrit, but for my present purpose it is not necessary to draw a distinction between Mâgadhi and Çauraseni.

v3.c17.132

The future most usually exhibits the form of the Sanskrit present in इस्स = इस्स. S. 1. रोचिस्सामि, 2. रोचिस्ससि, 3. ºइस्सइ. रोचिस्सं P. 1. ºइस्सामो, 2. ºइस्सध, 3. ºइस्संति. etc. ºइस्सइ.

v3.c17.133

This form is used indifferently with roots of all classes as in Pali, but here also there still subsist some traces of a future formed without the intermediate इ. Vararuchi (vii. 16, 17) gives the following:— Skr. √ श्रु “hear,” fut. श्रोष्यामि. Pr. सोच्छं. √ वच् “speak,” ” वक्ष्यामि. ” वोच्छं. √ गम् “go,” ” [गंस्यामि]. ” गच्छं. √ रुद् “weep,” ” Ved. रोत्स्यामि. ” रोच्छं. √ विद् “know,” ” वेत्स्यामि. ” वेच्छं.

v3.c17.134

These forms are, however, justly regarded as exceptions; for the rule in scenic, as in other, Prakrits is to retain throughout the root-form of the present. The regular type of the future is that in issa-, and the above words have also a future formed in the regular way, सुणिस्सइ, वचिस्सइ, गमिस्सइ, etc. This स्स is softened to इ, and the following vowel is weakened to इ, producing as characteristic the syllables ihi. Thus—

v3.c17.135

√ हस् “laugh,” S. 1. हसिहिमि. 2. हसिहिसि, 3. हसिहिइ, etc.

v3.c17.136

By a forgetfulness of the origin of such forms as सोच्छं, the ordinary future terminations may be added to them too, just like dakkhissati in Pali (§ 4), so that we find सोच्छिस्सति, and सोच्छिहिति.

v3.c17.137

§ 8. While the Maharashtri and Çauraseni dialects are considered the principal ones in the dramas, there are yet others of great importance, such as the Mâgadhi, with its sub-dialects. Among these, however, it is necessary only to notice that called Apabhrança. I do not wish here to touch upon the question whether the dialect called by this name in the dramas really represents the speech of any particular Indian province or not. I assume, for the sake of convenience, that Apabhrança is really a vulgar speech further removed from the classical idiom than Maharashtri or Çauraseni. There may have been half a dozen Apabhranças, probably there were. In this section I am merely seeking to put together examples of verbal forms in a dialect one step nearer to modern times than the principal scenic Prakrits, and having done so, shall go on to my own special subject.

v3.c17.138

The various tenses which in Sanskrit indicate past time have already in Pali and the earlier Prakrits been fused down into one. In scenic Prakrit a further step is taken, and the syllables ia, erroneously written ia in some MSS., are added to the root for all persons of the past tense (Var. vii. 23, 24. Lassen, Inst. Pr., 353). This is probably the neuter of the p.p.p. in Sanskrit, and its use is due to the frequency of the construction with the instrumental. Instead of saying “I saw, I went, I heard,” the people said, “by me seen, gone, heard.” This point is one of great importance in modern Hindi and Gujarati.

First 167 of 1,642 passages. Read on