Grammar Of Apabhramsa

Author(s): Madhusudan Mishra
Publisher: Vidyanidhi Prakashan
Catalog link: https://jainqq.org/explore/023436/1

Page #1

A GRAMMAR
OF APABHRAMSA
MADHUSUDAN MISHRA

Page #2

GRAMMAR
of APABHRAMŠA
MADHUSUDAN MISHRA
M.A. Ph. D. Diploma in Applied Linguistics
Deputy Director (Acd.) Rashtriya Sanskrit Sansthan,
New Delhi
Vidyanidhi Prakashan
DELHI (INDIA)

Page #3

Published by : VIDYANIDHI PRAKASHAN 1548, Gali No. 10, (Near Shri Maha Gauri Mandir) Khajuri Khas, Delhi-110094
Author
First Edition-1992
Price : 100.00
Printed by: Jeetu Printer D/256, Bhajan Pura, Delhi-53

Page #4

PREFACE
This introduction to the Apabhramsa language is exclusively based on the examples supplied by Hemacanda in his grammar (sūtras: 4,329-448). Though his sūtras have been of help in comparison, all the forms in the examples supplied by him have been individually parsed by myself with great care. That is why, I have been able to say many things more than Hemacanda has said, specially on syntax. The Apabhramsa syntax has not been taken into account by him, believing it to be like Sanskrit. Even Pischel has not touched syntax in his lengthy grammar, because he has confined himself to the Prakrit grammarians. He does not show any acquaintance with the Apabhramsa texts. Nobody has cared to think that the immediate successors of the various Apbh dialects are almost completely analytic and, therefore, the parents must show the salient features of their children in their own character. There is no doubt that much is buried in the gap between Apbh and the modern vernaculars. This is one of the dark periods in the linguistic history of India. I have tried to go into the darkness to collect some buried elements. But much more remains to be explored
The modern vernaculars have rightly been said to have evolved from the corresponding Apbh dialects. But one should not deny the influence of the migrating communities from one region to the other. The Gujrati language bears the stamp of the trihotra Brahmins from Mithila to such an extent that the western language sounds like eastern. This migration is attributed to Parasurāma in the Purāṇas. Similarly, all the dialects of Bihar are full of non-Magadhan elements, including the catholic Maithili. The socalled Bajjika dialect of a large part

Page #5

(iv)
of Tirhut Division of Bihar, believed to be the meeting point of Maithili and Bhojpuri, has nothing to do with the Magadhi A pbh. It is an import from the Kanyakubja country by an influencial migrating community, later merging with the native stock. They speak ha:ti for I am ha:te (familiar), ha:ta : (ordinary) and ha:tin (polite) for you are, and ha: 1 (ordi. nary), ha:tin (polite) for he is. They speak half ǝ, → and 2ǝ which are not to be found in other Bihari dialects. However, this is not a place to discuss these points.
Throughout my study of the Apbh language, I have been of the view that no grammar was written while Apbh was spoken by the common people, and the poets had also begun to use this language when it was outdated and ceased to be spoken in the form as presented by the poets. That is why, some words are very obscure in meaning and some grammatical forms are obscure even to the grammarians. But this is also not a point to be discussed here.
I have also given examples from other Apbh texts, but they are negligible. It was my desire not to restrict this grammar to a particular dialect, but I am not sure I have been able to do it.
The idea of writing an Apabhramsa grammar came into my mind when I found that the compilation of an Apabhramsa dictionary (entitled Sadda-sayaru), which I had originally planned, was likely to be a stupendous task on account of the unavailability of the printed Apbh texts, specially those with English or Hindi translation.
Though I have written half a dozen of books, I did not have so much thrilling experience in writing them as in the preparation of this grammar. In a life full of ups and downs, there are only few occasions for genuine smiles, but this book gave me wonderful experiences. For a western student of Indology, who has all facilities at his disposal, namely a sincere teacher or a guide, a good library, decent papers and pencils, etc., it would have been a routine experience, but for me it was quite otherwise I was at the same time a student and a teacher. I was

Page #6

(v)
also learning the language as I was writing this grammar. It was my own novel approach to the writing of grammar of that language which I had not studied as a part of my curriculum. I had not also studied it in the manner of German or French by doing exercises. This language had come to me without making me aware of it.
I used to come from the office in the evening, almost tired and exhausted. After a cup of tea, I took up the Apbh dohās of Hemacanda's grammar with the Sanskrit-chāyā of Pt. Saligrama Upadhyāya and parsed every word of the dohas. I do not know, in how many months it was completed. The whole material comprised nearly 5000-7000 sheets of paper. Then I made two groups of verbs and non-verbs. First of all, the non-verbs were arranged according to cases, the indeclinables forming a sub-group. Although I had only the Sanskrit cases in mind, I realised that much of the vast structure was demolished by this time, and the demolition was still going on. The Sanskrit-chāyā often compelled me to smile, realising that the poor descendant of the old royal dynasty still tried to be called by the kingly name for which he had lost all the rights. The construction like वृष्टिरभविष्यच्चेत्सुधान्यमभविष्यत् was forgotten long ago, but लज्जेज्जन्तु वयंसियहु जइ भग्गा घर एन्तु was rendered in chāyā by the conditional of Sanskrit. Really speaking, the writing of an Apbh grammar by a student of the vedic language looks like the writing of the history of a ruined empire. The Apbh language is very poor as regards the tenses and moods. Had it not been the legacy of the popular folk, it would have been also poor in participles.
The exploration of the texts supplied by Hemacanda by myself gave me more light than Hemacanda himself could give through his codified sutras. Although much of what belonged to the Sanskrit syntax came down to Apbh, the latter tended The to be more like the later vernacular than like its ancestor. spoken form of Apbh, which was as good as the early vernaculars, used to reflect in the Apbh poetry. While the poets would have been still writings fears af, they themselves would have been paraphrasing the same sentence as

Page #7

(vi)
fgangseff in their daily speech. The reinforcement of the absolutives by kar-i which was a feature of the vernaculars, are visible in the Apbh poetry : हिअडा फुट्टि तडति करि ।
I just want to point out that the poet's poetic language is about hundred years older than his speech, but even then his daily speech is reflected in his poetry. Therefore, if Apbh syntax has been ignored by the grammarians, they have thrown a big burden on our shoulders. The fore-runner of the Hindi sentence structure must be looked into Apbh, an 1 we are likely to discover the structure like a ra af between Apbh ओइ जाहिं and Hindi वे जाते हैं. The often quoted भल्ला हुआ जु feur afgfor 12 g is only half way between Hindi and Apbh, because the spoken form of Apbh would have had मारिया गया for मारिया.
Therefore I have added a full chapter on syntax, and that is the result of my own spade-work in the language. This spadework has taught me the language, and if some more annotated texts would have been available, I would have written something more. However, I believe that every vernacular should have an Apbh Grammar comparable to it. Then only the full picture of Apbh can be reconstructed.
Unless Apbh is explored extensively, nobody can venture to write an historical grammar of the modern vernaculars. With whatever amount of knowledge I have about Apbh, I shall try to write a historical grammar of Hindi. But whereas the Hindi morphology is more or less influenced by Parsian, its sentence patterns are influenced by the European languages. A typical Hindi sentence is : It is a copy of the English sentence: I am not going to come. But that too is influenced by the French sentence like: Il va venir (He is going to come=He will come). As far as I know, the Apbh language is quite free from borrowing. It is a typically indigenous language.
à
If this little book is received amicably by the students of Apbh, I shall try to write a more comprehensive Apbh. grammar. For the time being let it suffice.

Page #8

(vii)
As I have applied my own mind and labour in the preparation of this grammar, I need not thank so many persons. But I have utilized the Sanskrit chāyā of the Apbh dohās by Pt. Shaligram Upadhyay who amply deserves my thanks. To some extent I have taken help from Pischel's Prakrit grammar, who also deserves my thanks.
Lastly, I am thankful to Shri Badri Nath Tiwari of the Vidyanidhi Prakashan, Delhi for taking up the publication of this book.
Vasanta-pañcami, Magha śukla 5,2048
8.2.1992
Madhusudan Mishra

Page #9


Page #10

CONTENTS
Para
Page jïi-viii
1-7
1-- 3 4- 6 7-8 9-10
11
8-16
Preface Chapter 1
Origin of Apabhramsa Three levels of the Vedic speech Pali Prakrit Apabhramsa
Modern Vernaculars Chapter II
Historical Apabbramsa Phonology Phonelic changes in Pali, Prakrit and Apabhramsa Sounds of Apabhramsa
Historical Survey of Sounds Chapter III
Declension Parts of speech, number, gender, case Declension of nouns and adjectives
of Pronouns
of Numerals Chapter IV
Conjugation Voice, person, tense, mood Verbal bases Conjugation of verbs Secondary verbs Passive Causals
13-15 16-19
17—29
20-40 41-49 50-59 00-64
39-45
65-73 74-75 76-83 84 85-92 93-96

Page #11

Para
Page
97–98 99-114
46-49
117 118
119
50-59
Denominatives
Participles Chapter V Word-formation, compounds
and indeclinables Word-formation Compounds
Indeclinables Chapter VI
An outline of syntax Word-order Numbers Concord Pronouns Government of cases Tenses and moods Passive Present participles Past participles Infinitive Absolutive Appendix Post-Apabhramsa Vernaculars Hindi grammar Awadbi grammar Index
120-122 123 124 125 126-136 137–138 139 140 141-146 147–148
149
60-65
66–67

Page #12

CHAPTER ONE
Origin of Apabhramsa
1. The Apabhramśa language is that phase of the middle Indo-Aryan, the various dialects or the regional variations of which gradually evolved as the modern vernaculars of Nortb India. It is still a synthetic and inflexional language, though. very much simpler and much more varied than its ancestor.
The vedic people offered prayers to their deities in a more: or less homogenous popular language. But gradually, in wake of various social changes, there was the rise of the priestly families. The prayers were then preserved orally for the posterity. Although, with the march of time, the language: was changing, the daily language of the priests was static due: to the constant influence of the older language of the prayers, banded down by oral tradition.
(a) Thus, in course of time, two levels of speech could be distinguished in the Vedic society: one was the conservative language of the priests and the other was the ever-changing language of the common people.
2. The language of the prayers, however, could not be continued in its old and archaic form in the daily speech for a long time. Many obsolete and esoteric forms were gradually thrown out and then a simplified form of the Vedic language had evolved, which was younger than the language of the hymns and older than the language of the common people. Thus, towards the end of the Vedic period itself, three levels of language could be distinguished :

Page #13

A Grammar of Apabhramsa
(a) One was the older and archaic language of the hymns, which had ceased to be spoken even by the learned priests in their daily life.
(b) The other was the spoken language of the priests and some socially elevated groups, wbich was very conservative and little prone to change. It was a secular language corresponding to the sectarian language of the hymns. The hymns were however composed in the older language itself even by the younger generation of the priests, because it was after all their literary language, their second or learnt language in the modern sense of the term.
(c) The third was the language of the common people which can only be reconstructed on the basis of the later form of the language.
In phonology, the spoken language was not much different from the language of the higher social order. Only the final stops were very weakly pronounced and the visarga was almost unheard. The diphthongs tended to be pronounced as monophthongs.
The declension al pattern was simple. In plural, the dative and genitive as well as instrumental and ablative had largely fallen together. In singular also, the dative and genitive had largely fallen together and instrumental and ablative were beginning to do so. The pronominal endings were gradually replacing the nominal endings in ablative and locative singular.
The conjugational system was also simple. There was emphasis on a uniform a-conjugation. There was a tendency to take the third person plural present forms as basic, so that the other forms were based on these stems. Thus there was sunva-ti (for su-no-ti) on the basis of sunva-nti, and so on.
Both in declension and conjugation there was no dual.
The verb as (be) was not restricted to the root class, but it was also conjugated according to other classes. It was also conjugated outside the present system and took the past participal suffix •ta (forming s-ta), which may be seen in use more than two thousand years later in Hindi (in the form of thā).

Page #14

Origin of Apabhramsa
This language was liberal and very often crossed its own social boundary to bring something from other communities. The various non-Vedic forms were the results of conceptual imports from across the border, which were also passed on to the language of the upper class.
3. When the Buddha was preaching his dhamma in the sixth century B.C., the third level of speech had undergone great changes and had developed regional variations.
(a) The spoken language of the higher social order, the second level of speecb, was called bhāṣā.
(b) The language of the hymns, the socalled first level, was largely antiquated by this time and adequate schooling was necessary to understand it.
4. The Buddha preached his dhamma in the lower idiom called Māgadhi and also asked his disciples to use their own dialects. Any reference to the higher idiom was resented by the radicals in the church.
5. About less than two hundred years later, towards the end of the fourth century B.C., the bhāṣā, the language of the higher social order was described by the grammarian Pāṇini.
(a) We do not find any mention of the lower idioms, which were given respectable position by the Buddha, though we frequently find the archaic language of the hymns being compared with the bhāṣā.
6. Māgadhi. the lower idiom, in which the Buddha had preached his dhamma, or something nearest to it, may be seen in the verses said to have been uttered by him in the Buddhist canonical texts.
While editing the earlier texts in the different Buddhist councils the older forms were not replaced by the younger ones. That is why, when the canonical texts were finally edited and the grammars began to be written, the older and younger forms were given equal importance and were treated as variants. It was in this way that the form of the language was fixed and a creole was evolved, which was made the medium of instruction in the monasteries.

Page #15

A Grammar of Apabhramsa
(a) From the sixth century A.D. commentaries had begun to appear on the Buddhist canonical texts which were then called Pāli. The commentaries, called açthakathā, were written in the same language as the texts, but Pali gradually became the name of the language itself in which those texts and commentaries were written.
7. During the same period as that of the Buddhist canonical texts another non-vedic order was established by the Jainas, i.e., the followers of Jina, the victorious (Tirthankaras). But theirs was a personal faith, practised individually by different social groups. They did not feel the necessity of mass congregation and establishment of monasteries where religious principles could be discussed by the different groups of saints speaking different dialects. They did not also have any antipathy for Sanskrit and were least concerned with languagemania. But they were very particular about preserving the teachings of the ancients for the posterity. These were codified in different dialects spoken by them. But they could not develop a creole like Pali. That is to say, the dialects of the Jaipa canons could not be creolised, though they considered them to be a single language. It was later thought to be the language of the gods and received a peculiar name Ardhamagadbi. It is much younger and more diachronic than Pali and has a rustic or unpolished look.
(a) About three hundred years later, the Māgadbı spoken by the Buddha had undergone further change which was recorded, with some regional variations, in the inscriptions of Asoka.
The language of Asoka's inscriptions is in no way the continuation of the creole Pali, which was something more and something other than the Buddha's Māgadhi, but certainly represents a stage posterior to it. Therefore it is generally called early Prakrit. The Ardhamāgadhi also may be said to belong to this age.
8. The bhāṣā, as described by Pāṇini, later ceased to change. Some time later it was called Sam-s-kệta (polished or ornamented) to distinguish it from the speech of the common

Page #16

Origin of Apabhramsa
people, called Präkrta. During the centuries wbich followed, it became the court language of the kings dominated by the Brahmin courtiers. The most classical pieces were produced in it when it had ceased to be the spoken language even in highly educated families and reguired adequate schooling to be well-versed in it...
(a) When Sanskrit had begun to demand a great amout of schooling and the Sanskrit poetry had become the product of genius, the less genuine poets turned to the language of the common people for their poetic expression.
A look at the language of the Asokan inscription convinces us that it could not have been suitable for poetic expressions, because a good deal of refinement was reguired in it. With what amount of labour we do not know, but from the first century A.D. we find the same uncouth Prakrit having been cultivated as perfectly literary language. Though this Prakrit was well understood by a man of literary taste, it was widely removed from the speech of the common people. It was not a spoken Prakrit.
(b) This Prakrit had several regional variations. We may mention at least three of them, namely : Sauraseni (western), Mahārāşğrı (southern) and Māgadhi (easterm). But in the dramatic works, the regional variations of the Prakrits were ignored and a very artificial socio-ethnic and stylistic distribution was imposed on them : Sauraseni was said to be the language of the Prakrit prose, Mahārāştri the language of the Prakrit poetry and Māgadhi the language of the people of the lower social order. It is very curious to learn that a fisherman near Hastināpura spoke the language of Magadha in Kalidāsa's Sakuntala.
9. The spoken form of the Prakrit, however, went on changing. Therefore, the literary Prakrit also began to demand a great amount of schooling for poetic creation. The new generation of the poets then began to patronise the spoken language for it. By this stage the same Prakrit had evolved as Apabbra mśa with the same regional variations. A form of it

Page #17

A Grammar of Apabhramsa
is recorded, first of all, in some isolated stanzas of Kālidāsa's drama Vikramorvasiya towards the end of the fourth cenrury A.D. It was, however, not the spoken language of Kālidāsa but certainly about hundred years older than him. Although we know about this Apabhramśa from Patañjali himself, it was not available to us in complete sentences as here.
10. The following are the distinguishing features of the Apabhramśa language :
(a) There is no neuter gender in Apbh, which was so well marked in Prakrit.
(b) The nominative-accusative form was characterised by - u for a - base, on account of which it was ukāra-bahulā. bhāṣā. Otherwise, as a rule, the nom-acc. form was without any ending in sg. and pl.
(c) The genitive had begun to lose its ending in nouns, thus. becoming identical with nom - acc. However, its adjective used to indicate its genitive case.
(d) The final anusvāra of Prakrit was thrown back on the preceding vowel in the from of nasalisation : Skt. phalani, Pkt. phalāim, Apbh phalās. Gradually, the final -ās too was shortened to - ai (later contracted to -ế grand -ã art in Hindi for fem. words only).
(e) Many phonetic changes which were only sporadic in Prakrit and Pali had become general. Some changes had stopped, while some other types of changes were emerging.
(f) Though the vocabulary was the same as was available to Prakrit through the latest phonetic changes, even Prakrit words were taking new shapes. For example, the conjunct consonants were being reduced to one with the consequent lengthening of the preceding vowel : kārya> kajja> kajju or kāju).
During the following centuries Apbh also assumed literary form and dissociated itself from the language of the common people, which went on changing.
From the beginning of the present millenium, the languages. of India gave up their inflexional and synthetic character and

Page #18

Origin of Apabhramsa
7
became isolating and analytic. A totally different version of the Indo-Aryan emerged, which we now know as modern vernaculars.
11. As we pass from one literary language to the other, e.g., from Sanskrit to Prakrit, from Prakrit to Apabbramsa and from Apbh to the modern vernaculars, we always encounter a dark age, which comprises the spoken form of the literary languages of the later period.
During the dark period major changes are taking place in the language on the tongues of the people. Even a bhāṣā-kavi. who is supposed to write in the language of the people, does not take notice of the changes taking place in the language. His poetic language is always about hundred years older than his speech.
(a) As we stand on the outer wall of the Apbh dialects and look at the adjacent walls of the early forms of the modern vernaculars, we are full of confusing thoughts. We note with surprise that Apbh still preserves all the characteristics of an inflexional language and even the earliest phase of Hindi is completely isolating, there being postpositions in place of case-endings, periphrastic construction in place of verbal endings, and so on. These elements are buried in the gaps during which the language was changing. An attempt will be made to excavate the buried elements in the following pages.
(b) On the other hand, some loose elements of the earlier stage were gradually getting attached with the preceding form, which appear as endings in the modern vernaculars. The notable among them is the future ending-gā of the Hindi verbs, which was just an emphatic particle, namely ghā, of the Vedic people.

Page #19

CHAPTER TWO
Historical Apabhramsa Phonology
12. The Apabbramsa language, as already seen, is the continuation of the popular language of the Vedic society.
Its morphology and syntax will be discussed in the following chapters. Phonologically, it is not much different from Prakrit and Pali, which had already suffered notable changes.
(a) In pali : (i) The diphthongs ai and au had become e and o respecti
vely. (ii) ș was replaced by a, i and u in different positions. (iii) The conjunct consonants were assimilated in the follow
ing way : 1. The conjunction of two different stops resulted in the doubling of the last; e.g. khadga (sword)> khagga, etc.
2. The conjunction of semivowel/nasal and stops resulted in the doubling of the stops; e.g. alpa (few)> appa, nagna (nacked)> nagga, etc.
3. The conjunction of semivowel and nasal resulted in the doubling of the nasal; e.g. dharma (law)> dhamma, etc.
4. The conjunction of two different nasals resulted in the doubling of the last; e.g. janma (birth)>jamma, nimna (low) ninpa, etc.

Page #20

Historical Apabhramsa Phonology
5. But the conjunction of two semivowels resulted in the doubling of the first, except when it is r; e.g. kāvya (poetry) kabba, kārya (work)> kajja Please note that v is doubled as bb and y as jj.
6. A stop following a sibilant is aspirated and doubled; e.g. suşka (dry)> sukkha, etc.
7. A sibilant followed by a semivowel is doubled: e.g. tasya tassa, etc.
8. A sibliant followed by a nasal is voiced to h and then follows the nasal; e.g. grişma (summar) gimha, etc.
9. A dental is palatalised by y; e.g. satya (true) sacca, adya (today)> ajja, etc. (iv) Some individual changes may be summarised thus :
jñ>ñ (ñ), ny>ñ (ñ), kş>ch or kh, hy>yh, hv> vb. (v) The consonants did not change singly as a rule, but
there was a tendency of sibilants changing into h or ch :
trayodaśa (13)> teraba, șaț (6)> cha, etc. Upto Pali, this is the whole story. (b) In Prakrit : (i) e and o were pronounced short ě and ở before conjunct
consonants. (ii) The intervocalic k, g, c, j, t, d, p, y and v were lost,
leaving only the vowels following them; e.g. kāka (crow) > kāa, rājā (king) >rāā, kati (how many), kapi
(monkey), kavi (poet)> kai, etc. (iii) The intervocalic kh, gh, th, dh, ph. and bh were reduced
to h; e.g. mukha (mouth)> muba, megha (cloud)>
meha, etc. These were the major changes in addition to those enumerated above.
(c) In Apabhramśa : (i) The intervocalic m and n were reduced to v and the
nasalisation was thrown back on the preceding vowel;

Page #21

10
(ii) There was increase in the number of sibilants changing into h.
(iii) The gap created by the loss of the intervocalic plain stops (k, g, c, j, t, d, p, v) was generally filled up by placing -y- or -v-; e.g. gatā gaā> gayā, etc.
13. There are 42 sounds in Apabhramsa, 10 vowels and 32 consonants.
(a) The table of sounds is presented below :
Voiceless
AFFINITE
k
t
ट
A Grammar of Apabhramsa
e.g. grāma (village) gama gava गाँव sthāna (place)> thnathava, ja etc.
t
त
P
प
-m
kh
ख
ch
=
th
थ
ph
फ
S
स
Consonants
Vowels short a
2015 to
g
ग
b
ब
y
य
r
44
i u ĕ Ŏ अ इ उ ए ओ long ä 1 ū e O आ ई ऊ ए ओ
h
घ
Voiced
भ
dh
ढ
dh
ध
bh
भ
ǹ
ङ
ñ
ञ
n
ण
5 st 8 #
र ल व
n
न
m
म
STOPS
1 V SEMIVOWELS
SPIRANTS
(b) A comparison with the table of Sanskrit sounds may be very interesting.

Page #22

Historical Apabkramsa Phonology
11
14. The vowels are distinguished as short and long, but not in the manner of Sanskrit, which does not have short ě and o and does have the diphthongs e ai o au, that is, ai āi au au.
Orthographically, a few cases of short ș are available, but it hardly deserves to be called a vowel both in function and pronunciation. Thus grhanti (they take) is just an imitation of the Sanskrit form gļhộanti, and so on.
(a) These vowels are generally nasalised also, and this pasalisation has phonemic value, because the forms are generally distinguished only on account of this : kara-hi (you do): kara-bì (they do), etc.
(d) By the end of the Apbh period, the assimilated conjunct consonants had been simplified and the preceding short vowel was lengthened, thus weakening the final vowel which was. ultimately dropped : Skt. rātri (night)> Pkt. ratti Apbh rattirāti> Hindi rāt, etc.
(c) Occasionally the short vowels are lengthened and the long vowels are shortend for the sake of metre : reha for rehā (line), sāmalā for samala (dark), etc. Thus the length of vowels has no phonemic value in written Apbh.
15. Often even the short vowels are pronounced shorter : meha (cloud) is pronounced meha, kantu (husband) is pronounced kantu, kari (having done) is pronounced kari, and so on. Sometimes it is heard and nothing more. It is the starting point of the total loss of the final short vowel in Hindi words.
(a! We may observe a gradual shortening of the long vowels until its total loss in the beginning of Hindi : Skt āśā (hope)> Pkt. āsā Apbh āsa> Hindi ās, etc.
(d) On the other hand, in other cases, the short vowels tend to be pronounced long. Some Hindi words like sāhū from sādhu (merchant) are the results of this tendency.
16. Now a brief historical survey of the vowels may be presented here :
a is a simple original vowel. But often it represents Skt. s in initial and medial position : třņa (straw)>taña, etc. Very rarely, it also replaces other vowels : rātri +ratti+di=ratta-di

Page #23

12
A Grammar of Apabhramsa (night), etc. It also represents a weakened e : kantehi> kantahi, etc.
ā is a simple original vowel. But very often it is a result of the sandhi of two continuous a towards the end of the Apbh period : ghotaka (horse)> ghodaa ghodā, etc. It was during this period that baccă-type of words of Hindi had begun to emerge.
A long ā is shortened in any part of the word without any phonetic justification : bhāra (burden)> bhara, sthāpaya (place) >thava, etc.
i is an original short vowel. But it also represents Skt. ș: bhṛtya (servant)> bhicca, etc. Often it represents ě and e : vasina sabi-e, etc. This again may be lengthened for the sake of metre: bițți-e> bițți-e from bitti (daughter), etc. Even two short i have merged to form i in the prehistory of Apbh : gami-bi•i-gamibi (will go), etc.
u is an original short vowel. But it also represents Skt. ?: tņņa (straw) ►uņa, etc. Even a long ū is shortened to u before a vowel : hūāhuā (it was).
ū is an original long vowel. It is shortened to u before the case-endings : vahu-hi+vabū (bride), etc.
ě is a secondary vowel, generally representing e before the conjunct consonants :ěkka for eka (one), ěvva far eva (just), etc. But sometimes it is the elevated form of i: -itví (for-tva) + ěppi, etc.
e is an original vowel, but it also represents the prehistoric combination of a and i. The contraction of ay to e is wellknown : kathaya-ti-kahe-i, etc.
• is a secondary vowel, representing o before the conjunct consonants : chờll (cut). etc. Sometimes u is elevated to Ó : pustaka (book)- potthaa, etc.

Page #24

Historical Apabhramsa Phonology
13
o is an original vowel, but it also represents a prehistoric combination of a and u: mayūra (peacock) maūra +mora, etc. Also av is contracted to o : bhava-ti → ho-i, etc.
Quite often any vowel is seen in place of any other vowel without any phonetic justification. These cases may be ignored, because they cannot be put under any rule. Of course, in many cases, the reduction of a vowel is self-explained. Thus pekkh pikkh in pikkh-evi only because of the heavy suffix.
17. The consonants are classified as stops, semivowels and spirants, and arranged according to their qualities and places of utterance:
(a) Horizontally, the first series is guttural, produced by the back of the tongue pressed against the throat; the second series is palatal, produced by the middle of the tongue presseed against the hard palate; the third series is cerebral, produced by the front of the tongue turned upwards against the roof of the palate; the fourth series is dental, produced by the tip of the tongue pressed against the upper teeth; and the fifth series is labial, produced by the lower tip pressed against the upper lip.
(b) Vertically, the first two lines are voiceless and the next three are voiced; and among them, the first and third lines are plain, the second and fourth lines are aspirate, and the fifth line is nasal.
(c) Among the semivowels, the first is palatal, the second cerebral, the third dental and the fourth labio-dental, but all are voiced.
(d) Among the spirants, s and m are voiceless and his voiced, but all are aspirate.
18. From very ancient times in the history of the IndoAryan languages there has been a tendency to have only similar conjunct consonants in the popular speech. Whenever there were dissimilar conjunctions of consonants, they were assimi. lated. They have been detailed already.
(a) A consonant representing two original consonantsis

Page #25

A Grammar of Apabhramsa
regularly doubled, but not initially : satya (truth) -sacca, but tyāga (renunciation)--cāga, and so on.
(b) An aspirate undergoes assimilation medially by the corresponding unaspirate : akşi (eye)-akkhi, but kşamā (forgiveness) →khamā; madhya (middle)-majjha, but dhyāna (meditation) -jbāna. and so on. There is no doubling even after a nasal : kāňkşā (desire) → kamkhā, etc.
(c) On the other hand, the assimilated two nasals are often dissimilated by inserting a glide sound : āmra (mango)→amma
+amba, etc. But this tendency is as old as the Vedic society. The word sundara (beautiful) comes from sū-nara through sunnara, ni-tamba (hip) comes from ni-tāmra (coppery), and so on.
19. A historical survey of the consonants may be presented here :
k is an original consonant, but very rarely it is the devoiced form of g. kh is its corresponding aspirate, which is often reduced to h or voiced to gh: muha for mukha (mouth), sugha for sukha (happiness), etc. It has also come from şk/sk/kş in the middle of a word : suska-sukkha (dried), aksi-akkhi (eye), etc. g is original, but often also the voiced form of some original k. gh is original but it is often reduced to h, except when representing an original kh. The nasal å never occurs alone.
cis an original consonant, but it also represents a platalised t: tyāga>cāga, etc. k has often been platalised to c: kirāta → cilāa. ch is original; but it also represents a palatalised th and Skt sibilants : şaț (six)+cha, etc. It has also come from sc, ts and ps. j is original, but it also represents a palatalised d: adya (today) -ajja. A y used to change into j initially, but later even a medial or final single y used to result as jj, specially the suffixal y (in passive, optative and gerundive form): kriya kijja-, diya-, dijja-. jh is hardly original. It generally represents a palatalised dh and ks : dhyāna— jhāna, dhvani-jhuni, kşiņa -jbina, etc. The nasal ñ never occurs alone.

Page #26

15
Historical Apabhramsa Phonology
t is original as well as secondary. Often the r sound in the surrounding has been lost, cerebralising t: kệta (done) → kața, etc. It is also voiced to d, which appears as flap sound s in Hindi : ghoțaka-ghoďaa-gbodā Hindi IST. țh also is voiced to ạh, which results as the flap sound șh in Hindi : pathati (reads) →padha-i, Hindi 9671. d, coming from t and d, also without any phonetics jutification, changes intor and 1, specially in the numerals. The nasal n is a common Apbh sound, more numerous than n, specially medially.
t is an original sound. It is voiced to d in intervocal position, but more often it is lost, the gap then filled by y: gata →gaas gaya. th has been rarely preserved, because it has either been reduced to h or voiced to dh. d is original, but very often it is dropped, the gap then filled by y. th also has a tendency to be weakened to h.n also is original, but replaced by n in the middle of a word. It is also thrown back on the preceding vowel in the form of nasalisation, its place then taken by -v- : sthāna> thậna thấya.
p is an original sound. But its change into v is well-known in Prakrit. In Asokan Prakrit -tvā appears as-tpă in absolutive forms. In Apbh the absolutive suffixes -ěppi and -ěppinu thus
epresent the very old Vedic -tvi and "tvina of the popular language. The Apbh base pa- in pai for tvayā tells the same story. Similarly, p in-ppana has developed from-tvaņa through - tpapa. On the other hand, p continues to be changed to v till the last days of Apbh. Thus ěppi and ěppiņu also appear as evi and eviņu. ph is an original sound, but it is often voiced to bh and weakened to h. b is generally an original sound, but it also results from p through voicing : paitra (from pitr Father) > bețța. bh is an original sound, but it is also weakned to h. Rarely bh also represents sm and hv: smarati> mharai> bharai (remembers), jihvā>jivhā> jibb (tongue). etc. The change of -hy- to bb is known to Vedic : jar-hvar-i-ti> jarbhuriri (stumbles). m has never suffered any disfigurement, but for the first time in Apbh it is thrown back on the preceding vowel in the form of nasalisation, its place then taken by -v- : grāma (village)>gāma> gãva, etc.
Towards the closing period of Apbh and beginning of Hindi, y bad appeared in place of k, g, c, j, d, P, v and begun to

Page #27

16
A Grammar of Apabhramsa
contract, resulting in a : prakața> paada> payaļa, vacana - vaana> vayaņa> vän, etc.
y is both original and secondary. It has appeared to fill up hiatus.
r is original, but in Hindi numerals it represents d. It also changes into 1.
1 is original, but it often represents r and also n : limbaka. for nimbaka. It has also secondarily resulted from d.
s stands for all the sibilants, but sometimes it itself changes. into h : divasa> diaha (day).
h is original, but has also secondary origin from kb, gh, th, dh, ph, bh and sibilants.
ņ is an outgoing sound of Apbh in final position. It is often sent back on the preceding vowel in the form of nasalistion. Thus phalāni> phalāim phalās etc. But often the orthographic representation of this nasalisation by anusvāra creates. confusion.

Page #28

CHAPTER THREE
Declension
20. In Apbh the declension of nouns and adjectives (forming one group since Sanskrit) and pronouns have almost fallen together, though that was quite distinct earlier. That is to say, there are no strictly separate sets of endings for nouns-adjectives and pronouns. In this process, generally the pronominal endings have survived at the cost of the regular endings of the nouns and adjectives.
(a) The predominance of the pronominal endings is already evident in Pali, which shows that they were quite lively elea ments of the popular language of the Vedic people.
21. There are two numbers, namely singular and plural. The dual was lost already in Pali.
22. There are two genders, namely masculine and femi. nine. The grammarians have also referred to a neuter gender, but that is based on the nominative-accusative pl. ending •ās, later even -ai, representing the older neuter ending -āni, which was however generalised for all words in Apbh: rayanās (ratnāni) :: khalāi (khalān).
(a) As a matter of fact, the neuter gender had disappeared from Apbh. The concrete nouns of neuter gender had become masc. and the abstract nouns bad become fem. The ending -ailai was attached to words without any distinction in the beginning, but in Hindi it was restricted to the fem., that is, to the originally neuter abstract nouns.

Page #29

18
A Grammar of Apabhramsa
(i) The ending -ẽ added to fem. nouns ending in consonants, ū and a in Hindi is its true representative. The ending -yã also, added to fem. nouns in -1 in Hindi, has developed from the same original ending.
23. The masc noun or adjective is changed into fem. by means of the suffix -ā or -i or both : gaa (went)> ga(y)ā or gai, etc. When the fem. is extended by another suffix, the long -1 is shortened : gai + ā = gaia, etc.
(a) When a word is extended by .da and is then also changed into fem. •ļi, the preceding vowel, whatever originally it is, becomes a-: rātri (night)>ratta di, gosit di=gora-și (fair girl), etc.
24. The masc. nouns had lost all endings in sg. and pl. when used as subject and object in Apbh. Only the words ending with a- used to show -u finally in the early Apbh.
(a) In course of time even the possessive forms lost the endings. It was followed by the locative forms, and then by the rest of the cases.
(i) By the beginning of the modern vernaculars, all casesendings were lost.
25. It was in Apbh itself that new masc. nouns ending in ā had begun to develop by the merger of two a sounds: hiaa (heart)> hiā, ghodaa (horse)> ghodā, etc.
(a) But the contact with the Persian language had already started with the beginning of the early Hindi and the opposition of the forms like baccā from Skt. apatya (child) and Persian bachché had come to surface. The Hindi form bocca was adopted as sg. and the Persian form bachché was adopted as pl., vocative and postpositional form. This was perhaps the cause of the solitary opposition of sg. and pl. forms in masc. words in ā in Hindi
(b) But this opposition is limited to desi words. The borrowing from Sanskrit does not show this opposition : sg. rājā :: pl. rājā.
26. In Apbh the masc. nouns generally end in short a, i, and u, but there are a few words ending in long i representing the Skt. Possessives with -in, -min and -vin. As already noted,

Page #30

Declension
19 a few words in long a bad newly developed.
27. The fem. nouns generally ended in ā, i į and u ū, but in declension the long vowels are shortened before the endings : sahi> sahi-e, bālā bāla-hi, etc.
(a) On metrical grounds also the short vowels are lengthened and long vowels are shortened, at least in the written Apbh.
28. As regards the cases, the nominative and accusative had fallen together in masc., both in sg. and pl., having no ending at all. But the pl. forms were often distinguished in fem. by special endings, a feature coming down to Hindi also.
29. But there is a pl. ending -ho in vocative both for masc. and fem. It represents the Vedic pl. in -as-aḥ : janāsaḥ, etc.
(a) It has come down to Hindi also in the form of -o: bacc-o, log-o, devi-y-o, etc.
30. The instrumental and locative had fallen together in pl. : taru-bî aggi-bs, etc.
31. In fem. the ablative and genitive had fallen together, but in masc. the distinction was maintained.
32. There was a tendency to distinguish masc. declension from the fem. in instrumental sg. and genitive. The dative was long ago merged with genitive.
33. The locative forms are very clearly distinguished in sg. and pl : in fem., there is -hi in sg. and -hi in pl.: in masc. there 'is -hĩ both in sg. and pl., if the base is a- in sg. and ě- or i- in pl.; otherwise the ending is -e, -i with a base and -hi elsewhere in sg. and -hĩ in pl.
34. The endings of the different cases then may be seen in the following tables :
sg.
pl.
Nominative
Accusative
1-2
-
-0, -U

Page #31

20
A Grammar of Apabhramsa
Vocative
v - - -ho -ho Instrumental 3 -ě
hĩ -hĩ Ablative
5 che che chũ hũ Genitive
4–6 cho che chã -hũ Locative
7 -e, -i -hi -hi. -hi
-hi 35. These endings were gradually behaving like loose appendices after the bases, because they were destined to go in course of time. That is why, perhaps, the final vowel of the base was shortened or lengthened at the will of the metre.
36. The genitive forms were the first to lose endings in Apbh, of course after the nominative and accusative, but in some later instances even the locative forms are seen without endings.
37. However, the development of postpositions and coming into being of the postpositional bases are buried in the gap between Apbh and Hindi.
38. Perhaps the postpositions for the genitive had developed first of all; then for nominative in passive construction. Once this tendency had begun, there was a rain of postpositions which assumed the offices of all the cases in Hindi.
39. After the development of postpositions, the older group of cases was reduced to three in Hindi :
sg.
m. f. m. f. 1. Direct or
subject-object - - - -ē, -yã 2. Postpositional
-O -O 3. Vocative 40. On account of difference between masc. and fem. forms their declension may be shown separately in case of nouns and adjectives. In case of pronouns the distinction was lost for the most part.
41. We may begin with masc. words.
17
-
-

Page #32

Declension
(a) kanta- (husband).
1-2
V
3.
5.
6.
7.
sg.
kantu
कन्तु
kanta
कन्त
kantě
कन्तें
kanta-he
prae
kanta-ha
कन्तह
kante
apra
pl
kanta
कन्त
kanta-ho
aprast
kantě-hi
aprafa
kanta-hu
कन्तहुं
kanta-hã
कन्तहूं
kantě-hi
aprafe
21
Notes: The NA sg. form kanto is still found in very late Apbh, although it belongs to Prakrit. But kantu is a typical early Apbh form and is found throughout the history of the language. The youngest Apbh from is kanta. In NA pl. kantā is becoming rare and kantāī as well as kantaï are occasionally found, but they were destined to go with the fem. words of abstract neuter origin. The V pl. from kanta-ho has survived in Hindi (cf. log-o!). In Isg. kantena is the oldest form, which was also reduced to kantina, kantim and even kanti. The I pl. form is akin to the Vedic kante-bhiḥ, but another form kanta-hi. is the result of the weakening of the base. A rare form is kanti-hĩ. It is very surprising that I pl. and L pl. have fallen together, though of different origin. In Ab sg. there is also an additional form kanta-hu. The typical Gsg. form in Apbh is kanta-ha, but the older forms kantaho kantasu and kanta-ssu are found throughout the history of the language. In pl. the older form is kanta-hã. In L sg. the older form is kante, but it was later weakened to kantě and kanti. In some cases-hĩ of the pronominal declension is also seen (kanta-hi). However, it is a pl. ending when the base is kante-.

Page #33

22
A Grammar of Apabhramsa b. The declension of aggi (fire) and taru (tree) runs closely parallel :
sg.
pl.
1-2.
aggi अग्गि
t aru
तरु
taru तरु
taru
aggi अग्गि
तरु
taru-ē
तरुए
aggi
अग्गि aggi-ho
अग्गिहो aggi-hi अग्गिहिं aggi-hũ
अग्गिहुँ aggi-hã
अग्गिहुँ aggi-hi अग्गिहि
aggi-ē
अग्गिएँ aggi-he अग्गिहे aggi अग्गि aggi-hi अग्गिहि
taru-he
Tree
taru-ho
तरुहो taru-bi तरुहि taru-hũ
तरुहुँ taru-hã
तरह taru-hĩ तरुहिं
taru
taru-hi तरुहि
Notes : The Isg. form is also abbreviated to aggim tarum, and less commonly, the older form aggiņa taruņa is also seen. The Gsg. form has usually no ending, though the older form with-sa, -ssa, -ha may be met with. The Gpl.has also an additional form with hũ, perhaps through confusion. Very rarely. even Lpl. has an additional form with -hũ, so that 5-6-7 pl. forms are identical.' Probably, this was the cause of the disappearance of the case-endings.
42. As already remarked in the beginning, some words ending in ā, e g. ghoạā (horse), had developed in Apbh, but in declension they had fallen together with a-base.
(a) Similarly, all the consonantal bases of Sanskrit had lost the final consonants and reduced to one of the three bases in Apbb. Thus svāmin (master) was either nothing more than sāmi-or extended to sāmia-, declined like aggi— cr kanta
43. As a rule, any word may be extended by -a or -da in Apbh, and this was clearly with a view to simplifying the declension. Sometimes this extension was demanded by the metre.

Page #34

Declension
(a) The extended form generally ends with -u in Nsg.: kanta kantau, mahāra-a mahāra-u (my ). But some times the extend - ed form takes the ending -u, which is regarded by the grammarians as a neuter form : tucchaū (thin), bhaggaū ( wounded), etc.
44. Though the NA show the ending -ū in sg. and ai/ai in pl., this feature is a rare one and also without any regard to the original gender of the word. Even though the Apbh words were distributed as masc. and fem., the neuter endings had survived. Sometime later the sg. ending-ũ disappeared and the pl. ending -ãi / ai assumed two forms : ē and yã. These were reserved for the feminine words in Hindi.
45. Now we can take up the declension of the fem - words.
(a ) bālā (girl), sahi (girl friend) vahu (bride).
sg
pl.
sahi
सही
1-2 bālā
बाला
V. bāla-e sahi-e
बालए सहिए
3. bāla-e sahi-e बालए
सहिए
5-6. bāla-he
वाल
7. bala-hi
बालहि
sahi-he
सहि
sahi-hi
सहिहि
vahu
वहू
vahu-e
bālā-o
बालाओ
हु
vahu-hi
बहुहि
bāla-ho
बाल हो
वहुए
vahu-e
bala-hi
वहुए
बालहिं
bahu-he väla-hu
223
वालहु
bāla-hĩ
बालहि
sahi-o सहिओ
vabu-o
बहुओ
sahi-ho vahu-ho
सहिहो
वहुहो
sahi-hi vahu-hi
सहहि
वहुहि
sahi-hu vahu-hu
हि
वहु हु
sahi-hi
vahu-hi
सहि
वहुहि
Note: The ablative and genitive have fallen together in fem. declension. There is an additional form bālā-u as well as sahi-u (but vahu-u is naturally impossible) in NA pl. and, while the former has retained the long vowel the latter has shortened it. But, as a matter of fact, the length of a vowel of a base is unimportant in Apbh. It is generally determined by metre.
46. In very late Apbh. the fem. a base also takes the

Page #35

A Grammar of Apabhramsa
ending -ï for hi in Lsg., where the base vowel too is shortened: siya-i for siya-hi.
24
47. The quantity of the final vowel of the ending also is dependent on metre. Thus sometimes the metre requires bāla-hě and sometimes bāla-he.
48. During the dark period Hindi and other vernaculars, the stems appeared in the language. continued in the language we do not know. Perhaps the postpositions, which had already grown up, were waiting for the moment and usurped the offices of the endings on the next morning.
49. The masc. and fem. Hindi words remained content with their original forms in sg.. but in pl. the masc. words bore endings before the post-positions and as vocative, and the fem. words bore endings also as subject-object.
(a) Some masc. words ending with a also take e-form as subject-object in pl. and as vocative and before postpositions in sg. This was probably due to the influence of the Persian language.
This is very briefly about the declension of the nouns and adjectives.
Let us see their forms.
(a) First person
sg
preceding the emergence of endings were lost and bare How long this state of affairs
50. The pesonal pronouns have expected forms in Apbh. The notable peculiarity is that 2-3 have fallen together in sg. and 5-6 in both.
1.
2-3.
hau
20
mai
मई
Pronouns
1-2. amhe,
अम्हे
3. ambe-hi
arafe
pl.
amhai 37755

Page #36

Declension
5-6. mahu,
महु
mai
मई
7.
1.
.3.
2-6.
Notes: The Nsg. form hau is probably derived from hakam, as may be imagined from the Magadhi hake. Otherwise, the sg. base is as old as Sanskrit.
The pl. base amha-would have suffered only the expected phonetic change towards the end of Apbh. On the other hand, the sg. forms must have been influenced by the counterpart in the second person. Even hau was replaced by maï, and the base mujjha spread over the rest of the sg. forms. Similarly, the pl. base amha spread over all the forms.
(b) Second person
sg.
tuhu
तुहं
7.
paï
पइं
majjhu मज्भु
tau
तउ
paï,
पइं
tai
तइं
amha-hã
322
tujjha
तुज्झ
amhā-su
अम्हासु
tai
तइं
tumhe,
तुम्हे
tumhe-hi
grafe
tumha-hã
तुम्ह
25
pl.
tumhā-su
तुम्हासु
Notes: The Nsg. form is difficult to be explained in its last part, unless we derive it from *tusum from the base tuşma in *tuşme, sg. for yuşme. The base pa-in sg. represents the old base tva-through the stage tpa-. In sg., though 5-6 have fallen together, tuha seems to be restricted to 6.
tumhai तुम्हइं
Towards the end of Apbh, tujjha seems to have spread over the sg. forms in oblique casse, pushing back tai pai to Nsg., though tuhu also continued dialectally. Thus tū survives in Hindi, Punjabi, etc., and taï in Avadhi, Brajabhāsā, etc.
51. From Apbh itself we see that the final a in these forms is often lengthened: dekkhu amhārā kantu, etc. Thus the

Page #37

26
A Grammar of Apabhiarmsa
Hindi forms are self-explained.
52. Other pronouns in Abbh are the following: Demonstrative : sa-/ ta-; eha-/ ea.. oha-/ 0-, aya-. Relative - :jaInterrogative : ka-, but more frequently later, kavaņa- from
*ka tana for ka-tama. Pronominal adjectives: anna-, savva and sāha , etc.
53. The demonstrative base sa- is found only in Nsg., and the grammarians also point ou: the distinction of gender, probably on the basis of the earlier forms. As a rule, the distinction of gender was gradually dying in Apbh. It was seen only rarely in sg., never in pl.
54. The declensional forms of ta-are :
sg.
1-2, su m.
sā f. HT tā-e f.
tē m.
te-hi
ताए
तेहिं
ta-hũ
तद
ta-he f.
taches
ta-hã तहां ta-ho m. TET ta-hi तहिं
ta-hã ताह tě-hi तेहिं
Notes : In Nsg. masc, the older form so also is found. Similarly, in pl. the older form te also is found. Generally tam appears as obj. in sg. Perhaps it is tã. The shortening and lengthening of the vowels also are quite common. In Gsg. the older forms tāsu and tassu also appear.
(a) By the end of the Apbh period, the ta- base disappeared, in spite of the fact that it was represented in all cases. Really it belonged to the artificial kavi-bhāsā.
55. Another demonstrative base aya. is said to be repre

Page #38

Declension
27
sented in all cases, but it was also the artificial creation of the poets and was destined to go from the popular language.
56. The demonstrative base eha- (from eşa-h), this, is found only in NA sg. and pl., and o-only in pl. on the analogy of eba- : 1-2. eha ei and *oba
oi ओइ
The grammarians distribute eho as m., eha as f. and ebu as n., but it is difficult to agree with them. These are the older and younger forms. There are also forms like ehaữ and eu.
57. The relative pronoun ja- is represented in all cases : 1-2. ju m. jā f. jam जु
जा जे
जि
11
ज
Sales de color
je-hi
fe ja-hũ
ja-hã जहाँ jā-su m, जासु ja-hị
ja-he f
jā-hã जाह je-bi जेहिं
7.
जहि
The older form jo is used in Nsg. masc. and je in pl.
58. The interrogative pronoun ka- is represented in all cases, but gradually it was supplemented and replaced by the restrictive kavana : 1-2. ko m. kā f. kavana
ke को
का
कवण 3. kavanena
kavaņěhĩ कवणेण
कवणेहि kabă kihe
kahũ कहाँ किहे

Page #39

काह
28
A Grammar of Apabharmśa 6. kāsu m. kahe f.
kāhã .. कासु कहे 7. kahi, kavanahi
kavanehi कहिं कवणहिं
कवणेहि The grammarians also refer to the form kāĩ (why). The indefinite koi Page #40

Declension
duve dunni donni in 1-2, běhí in 3, duhũ in 5-6- 7; and perhaps dubs also in 3. The form donni is evidently on the analogy of tiņņi, but duhũ in 7 is certainly irregular.
(b) The numerals form 5 to 10, namely pamca cha satta ațţha nava daha, are regularly declined on this pattern.
61. The declensional forms of decades from 20 onwards. are not found. They are : visa 20, tisa 30, cattālisa 40, pannāsa. 50, saçțbi 60, sattari 70, assi 80 and navai 90.
62. When simple numerals are compounded with decades, there are some changes on both sides. For example, daha be.. comes -raha before bā- for 12, sattari becomes hattari before ba-for 72, and so on.
63. Among the higher units of numbers, the frequently used ones are : saya 100, sahassa 1000, lakkha 100000, etc.. Later only sahassa seems to be replaced by the Persian bazār. They have the regular forms in declension : saya vāra (hundred times), dukkha sayāl (hundreds of calamities), lakkhehi, etc.
64. The ordinal numbers are : pahilā from pra-thila for pra-tara 1st, bia duijja from dvidya for dvi-tiya 2nd, tra tijja. from tặ-dya for tr-tiya 3rd, cauttha 4th and chațțha 6th. Other ordinals are made by -ma : pamcama 5th, sattama 7th, etc.

Page #41

Conjugation
65. The verbs are conjugated in sg. and pl. in active. There is also a passive voice in Apbh, but there are only active endings for both. The passive affix-ijja- had also generally lost its passive force towards the closing period of Apbh.
66. There are three persons, namely: First person, Second person, and Third person.
67 As far as verbal conjugation is concerned, there is really only one tense, namely the persent.
(a) Though very rare and sporadic instances of the forms of the past tense are found, it has actually disappeared from the Apph conjugation. It was exclusively expressed by the past participles, which have come down to Hindi, participating in some major tense-formations.
(b) Though the forms of the future tense too are found in early Apbh they had ultimately disappeared from conjugation.
68. There are three moods, namely subjunctive, imperative and optative, but they are attached to first person, second person and third person respectively. Therefore, they may be collectively called modal forms.
(a) A conditional mood had developed in Apbh from the present participles, which has also come down to Hindi.
69. Thus only the present and modal endings are available in Apbh.
70. The present tense has been having only one variety from the Vedic language, and it has continued upto the end of the life of the Apbh. The past tense had several varieties in the Vedic language, which gradually diminished till their total

Page #42

Conjugation
31
extinction on the door of Apbh. The furture tense, though of secondary origin from the aorist bases in the Vedic, had a luxurious life in Sanskrit. but it disappeared towards the last phase of Apbh.
71. The subjunctive, imperative and optative moods had distinct forms and functions in the early history of Sanskrit, their respective connotations being requisition, command and wish. From Sanskrit itself we observe a gradual restriction of these ideas with first, second and third persons, which is firmly established in Apbh and has come down to Hindi.
(a) The subjunctive forms become extinct at an early period in the history of the Sanskrit language, and the formal distinction between imperative and optative continued only upto Pali. From the beginning of the Prakrit language, we come across only imperative forms with the traces of optative in third and second person sg. The distinction of forms was lost and the different connotations were attached to the respective persons.
72. Now we come to the question how the modal forms survived in Hindi at the cost of the present forms.
(a) In Pali the present and imperative forms were identical in first person and in second person pl.
Present
Imperative -ti -anti
-tu
antu 2. --si --tha
- tha 1. -mi mo
-mi
- mo. (b) In Prakrit there was an attempt to bring out fresh opposition and, at least, the 1. sg. was redrafted : 3. -i -anti
-antu 2. - si
-- ha
-hi
- ha 1. -mi - mo
-mu
-mo The middle endings could be seen only in Imp. 2sg., which was - su Page #43

-U
A Grammar of Apabharmsa 73. Tne typical Apbh forms, however, begin with the following present and imperative endings : 3. i -hi
-hũ 2. -hi
-hu -hi
-ha 1. cũ -hũ
-hi (a) Though standing confusion in 2pl. was eliminated by having -hu in place of -ha in present, a fresh confusion arose in 2sg. due to the change of -si into -bi There was another confusion in 1sg., where the earlier oppostion of -mi and -mu was lost and both came out as -ũ.
Elsewhere the opposition was maintained. The unphonetic introduction of -h- in some endings appears to be intentional.
(b) Apbh had also some additional endings -e, often weakened to -i, and -u in Imp. 2sg., which may be traced to the optative 2sg. e and Imp. 2sg. -hu <-su (Pali-ssu. Skt. -sva) respectively. There must have been an additional ending -e even in 3sg. derived from the older optative -e.
The reduction of 2sg. -bi to -i and its sandhi with the preceding a- would have obliged the 3. and 2 sg. to be similar asfollows: Present
Imperative 3. -e
-hi 2. e -hu
-ha -hũ
-hi The indentical picture of the present and imperative in sg. and the confused state of affairs in pl. would have been repug. nant to the speakers of the language in the dark period.
The predominance of the modal concept ousted the present, which sought refuge in the periphrastic constructiou.
(c) The Hindi language inherited only the modal endings in conjugation. They are :
sg.
-hũ
2sg. has also a subjunctive from with -e.

Page #44

Conjugation
33
74. Only limited verbal bases were handed down to Pali from the popular language of the Vedic people. They were : a- and aya-, often contracted to -e-. There is perhaps a soli. tary example of the base ho- from bhava-. There were also some mixed bases of irregular formation, because of their direct origin from Sanskrit.
(a) In Prakrit also there are only three bases, namely a-, e- (representing the earlier aya-) and o-(in the solitary example ho-).
(b) In Apbh the same verbal bases have been handed down with the greater predominance of the a- base.
75. With the exception of a few verbs in e and o, all verbs in Apbh may be treated as ending in consonants, so that -a or -e may be added to any of them, although the latter is attached preferably to a polysyllabic verb : dekkh (see) dekkha-/dekkhe- cor (steal)
corepekkh (see) pekkha
vajj (leave) vajjekar(do) kara
mār (kill)
mārehar (take) hara
sammān (respect) sammāņele (take)
ho (be) . ho(a) A good number of passive bases of Sanskrit, which were used as such even in Prakrit, gradually lost their passive force and came down as simple verbs in Apbh :
śak-ya-> sakka-(can), etc.
le
Some earliest of them may be traced to the div-class of Sanskrit conjugation :
kup-ya- (be angry) kuppa-, trutt-ya (break) tuţta-, etc.
Some bases may be traced to the suffix-ya, but they neither belong to the div-class nor to the passive :
rucca- (be pleasing to) Page #45

A Grammar of Apabhramba tuş-ya> tussa-> tūsa (be satisfied) ruş-ya> russa-> rūsa- (be angry) puş-ya> pussa-> pūsa- (bear), etc.
(b) Much later, even the typical Apbh base in -ijja- or -ějja- had begun to appear as simple verbs, when the passive force was completely lost from the form and periphrastic construction had already appeared in the spoken form of the language :
rakkhejja- (keep away) lajj-ejja-be ashamed), etc.
(c) It is very interesting to note that the root class of Sanskrit has managed to stay in the language, of course with the usual phonetic change :
jā (go) from yā ļbā (stand) from sthā, etc. In Apbh this group has been joined by a number of verbs with a- base in Sanskrit and Prakrit because of the loss of the medial syllable :
khād-a-ti (eats)> khāa -> khā .i. etc. (d) The greatest innovation of Apbh is that the old past participles of Sanskrit have often been used as verbs in their phonetically modified forms:
lag.na (stuck to) > lagga (stick to) mrta (dead) > mua (die) pra- viş- ța (entered)> paițţha (enter), etc.
In some cases even the past participles have suffered unexbected changes. Thus there is pahucca -i (reaches) from prabbū-ta through an intermediate form pra-bbūtya-ti.
76. So far we have discussed the bases as well as the endings of the Apbh verbs. Now we can take up the forms in the different tenses and moods.
Let us take the verb kar (do), base kara :

Page #46

Conjugation
35
Present
kara-i कर
kara-hi करहि kara-ũ करउँ
kara-bi करहिं kara-hu करहु kara-hũ करहे
1.
(a) As already stated in the beginning, these forms of the present succumbed to the force of the modal forms, which built their citadel on these ruins.
(b) Only the verb aha -i (is) seemed to survive in the spoken form of Apbh and came down to Hindi as an auxiliary element in some forms.
(c) In the eastern dialects, the literary form as-ti suffered palatalisation, as we see in a:chi (Maithili), âche (Bengali), etc.
Past
77. As already stated earlier, the past tense had disappeared from the Apbh conjugation.
Very rarely, the forms like so-hi-a (heard)=Skt. a- śraus -It were used in the literary Apbb, but they were considered to be highly archaic.
(a) The typical past forms in Apbh represented the older past participles, made by the suffix -ia (Skt. -ita) : mil -ia (met), cal-ia (started to go), etc.
(b) However, the older forms with -ta are as much common as the former, even in late Apbh, of course with the usual phonetic change : patta Page #47

36
A Grammar of Apabharmsa
(c) These Apbh past participles were further extended by -a, which often appeared as -u in 1 -2 sg.: dittha -a. cal-ia-a, etc.
Even the fem. forms are thus extended. the final i then being shortened : gaa>gai, then gaitā=gaiā (went), mua>mui, then muitā=mui-ā, etc. on metrical ground, the final a is again shortened.
(d) When these forms were extended by -da, they had adjectival force as seen in the later dialects : sutta-da (asleep> rutala), etc.
(e) In Hindi, the past participal suffix is -ā.
Future
78. As already remarked in the beginning, the future tense is the vanishing element of the Apbh conjugation.
In Pali and Prakrit the future infix is either - issa/ssa-or -bi-, but in Apbh the former was simplified as - sa -
In the early stage of Ahbh, the future forms used to be borrowed from Prakrit. The typical Apbh forms, if made from kar (do), may be : 3. kar-i-bi-i
kar-i-hi-hi करिहिइ
करिहिहिं 2. kar-i-hi-hi
kar-i-hi-hu करिहिहि
करिहिहु 1. kar.i-hi-ũ
kar-i-hi-hũ करिहिउँ
करिहिहुँ The forms from ho (be) are : 3. ho-sa-i
ho-sa-hi होसइ
होसहिं 2. ho-sa-bi
ho-sa-hu होसहि
होसहु

Page #48

Conjugation
1. ho-sa-ũ होसउँ
(a) In some cases the connecting vowel is -e- sah-sa-i (he will tolerate), etc.
(b) But there was also a stage at which great confusion arose due to merging of infix and ending in various ways. Thus ho-hi-i was contracted to ho-i, gam-i-hi-i to gamihi, and so on. This contraction is abundant in Awadhi: karihi, dharihi, etc.
(c) As we come towards the end of Apbh, we meet with very peculiar forms which neither belong to the current Apbh nor represent the development of some earlier forms, but they are just poets' creations on some obscure models; e.g. pāv-i-su (I shall get), kar-i-su (I shall do), (I shall enter), etc.
y pais-1-su
sg.
3. kara-u
करउ
ho sa hũ होस हुँ
2. kara-hi
arfe
79. Though even the Prakrit modal forms appear in Apbh, the typical Apbh forms, if made from kar (do), may be :
1.
37
The Modal Forms
kara-ũ
करउँ
pl.
kara-hu
करहुँ
kara-ha
करह
kara-hi
कहि
The 2 sg. has also some additional forms: af kar-i, ne kar-u, kar-e.
was
80. Later there was a little change in 3sg., which replaced by -e. In 2 sg. even the bare stem began to be used. A little later phase of the language is represented thus:

Page #49

38
A Grammar of Apabhramba
3. kar-e
करे
2. kar-e,
kar
kara-hũ करहुँ kara-hu करहु kara-hi करहिं
करे
1. kara-ũ
करउँ
This is the story of the Apbh language as recorded.
81. In the spoken form of Apbh, the 3 and 1 pl. became identical and contracted to - ?. Even 2pl. and 1sg. suffered similar contraction. This picture of the dark period is fully represented in Hindi : 3. kar-e
kar-ě
2. kar-e,
करे 1. kar-ú
करूं
kar कर
kar-o करो kar ē
करें
82. On the body of these modal forms, an auxiliary element gā began to be affixed, and thus there was the beginning of the future forms in Hindi.
(a) This gā, with its original short form ga, is even now used in various dialects. In a dialect of the Tirhut division of Bihar, ga is used after a subjunctive form, wbile in some western dialects it is used with present for emphasis.
(b) It is a very old element, already used by the Vedic people, not only in their popular language but also in the language of the hymns. There it was ghā, used with a subjunctive form to express a future idea : ā ghā tā gaccbān uttarā yugāni (Those future ages will come).
(c) In the eastern Apbh dialects -v- had appeared between the base and the endings. Thus there was the beginning of b-future in all the Magadban dialects : Oriya karibi,

Page #50

Conjugation
Bengali kariba, Maithili karab.
83. The conditional mood, the forms of which were made from the future base in Sanskrit, was lost long ago. But, for the first time in Apbh, we see the emergence of new conditional forms from two present participles, connected with jai (if) and to (then ) : लज्जेज्जन्तु वयंसिअहु जइ भग्गा घरु एन्तु (I would have been put to shame if he would have come back wounded).
39
(a) This conditional mood has been
Hindi.
Of the simple verbs in Apbh, this is the whole story.
Secondary Verbs
84. The secondary verbs have been sharply reduced in Apbh. Only the passives and causals are quotable extensively. The denominatives are rare.
handed down to
Passive
(a) The forms of the secondary verbs are hardly quotable outside the present tense and past participles or the past tense.
85. In the older passive forms of Apbh, the Sanskrit forms appear to have been used with the usual phonetic changes chijja- Page #51

A Grammar of Apabhramśā
(a) Sometimes the original roots are recognisable only by the first letter : d-ijja (be given), k-ijja- (be done), etc. These forms, however, had lost their passive force towards the end of Apbb.
(b) The passive base like gheppa- (be taken) presupposes the Vedic gbạp from gặbh through metathesis of aspiration and voicing.
88. The forms, if made from jān (know), base jāņ-iaare; 3. jān-ia-i
jān.ia-hi जाणिअइ
जाणिअहिं 2. jān-ia-hi
jān-ia-hu Erfurstfee
जाणिअहु 1. jān-ia-ũ
jān-ia-hũ जाणिअउँ
जाणिअहुँ (a) In Awadhi the forms like dijjai suffered contraction : dijjā, etc.
(b) The modal forms are rare.
89. In the last days of Apbh the passive base with -ijjahad lost its passive force, so that it had begun to be used as ordinary or simple verb. Thus lajjejja-was not more (or less) than lajj-, rakkh- ejja- was the same as rakkha-, and so on.
90. Simultaneously, we find the beginning of a new passive construction by the Apbh infinitive in -aņaha and the present forms of the verb jā (go): 37FUTE FT FTF (It can not be said), TFFTE FTS (It cannot be eaten), etc.
(a) There is hyper-sanskritisation of this construction in the Purāņic and Buddhist texts written during this period : TOTIE 7 ufa=tera, 11 7 7 fa=n Pud, etc.
91. Still later, the passive construction had started by combining the past participle of the verb in question with jā (go): 2174 FT (It is eaten), etc.

Page #52

Conjugation
41
(a) There is hyper-Sanskritisation of this construction in the Purāņic and Buddhist texts written during this period : TETT 7 orfa=#retra, etc.
92. The passive construction like this mark the beginning of Hindi.
Causals
93. The older forms of the causals are just recastings of the Sanskrit forms on Prakrit pattern : संमानय>संमाणे- (show respect), 2014>2501- (desert), #TCU> HTT- (kill), etc.
94. But the typical Apbh casual bases are formed by -āva : नच्चाव- (cause to dance), हराव- (cause to take away), देवखाव(show), etc. 95. The examples of the forms are :
Pr. 3.777793 naccāva-i forafg naccāva-hĩ, etc.
Past : ArfF3 (killed), alfa (caused to be done), etc. 96. Towards the beginning of Hindi, -āva- was further reduced to -ā.
Denominatives
97. Any word ending in a -could be used as verb with aor e. base : fate> tikkbe-i (makes sharp), 377> karāla i (hardens), etc.
98. In some cases, the word ending in a- turns into I, and then it is led by the verb ho- (be) : Too (powder)> qutats (is crushed to pieces), 1837 (small) 5 E39 (was shortened), etc.
Participles
99. The participles are the most interesting elements in Apbh. Instead of getting out of use like others, they had begun to assume additional offices in the language.

Page #53

A Grammar of Apabhramsa
100. The present participles in -anta were no doubt adjectives, but in the last phase of Apbh itself they had begun to express condition, which also came down to Hindi.
101. The past participles in -ia were also adjectives, but in Apbh they were utilised exclusively to express the past tense.
102. It was only the future passive participles or gerundives which could not survive with its host of early forms. Perhaps only - ijja is the sole survivor, representing some *-eyya of the group panāyya, etc.
Present Participles
103. The present participles are made by-anta, fem. anti, which is added even to passive bases : jo (see) jo-anta, gan (count) gan-anta, kar (do) karanta, dams (see) dams-ijj-anta. lajj (be ashamed) lajj-ějj-anta, etc. In some cases of the contracted verbs in e and o, the affix is simply ·nta : e (come) enta, ho (be) ho-nta, etc.
(a) The present participles are also extended by -a (fem. - 1). Then the final i of the fem. is shortened: honta-a (becoming), gaṇanti-ā, etc. The final à is again shortened before the case-endings : uddāvantiă-e from udļāvantiā, etc.
Past Participles
104. As already said, the past participles had largely or exclusively assumed the office of the past tense in Apbh. They are made by-ia : mil-ia (met), cal-ia (started to go), kar-ia (did), etc.
105. In early Apbb, however, the Sanskrit forms with ta are very common, of course with the usual phonetic change : prāp-ta> patta (reached), sam-āp-ta> samatta (ended), dpsta> dițba (was seen), gata> gaa, f. gai (went), mộta Page #54

Conjugation
etc. In case of extension. the fem. -i is shortened : gai-a, muia, etc.
Future Participles
107. The future participles in Apbh are made by ·ějja or -ijja, which represent the Vedic -eyya : ca-ijja (fit to be given up), ho-ijja (fit to be), etc.
(a) The other rare suffixes are -jěyva or -ěvva : kar-iěvva (to be done), sah-ěvva (to be tolerated), etc. It is also abbreviated as eva : so-eva (to be slept), jagg-eva (to be waked), etc.
(b) In the older literature, often -tavva also is found, but that belongs to Prakrit.
108. The present and past participles of Apbh have participated in some major tense formations in Hindi, besides their own participal use with huh. But the nasal element, of the present participle (jānta, honta, etc.) was lost in all the verna. culars except Punjabi (cp. Hindi जाता है, Awadhi जात अहै : Punj. tar).
Infinitives
109. There were a dozen of infinitive forms of the Vedic language. In Sanskrit they had been reduced to a single form with-tum. Perhaps they were in still larger number in the popular language of the Vedic people. Therefore the middle Indic dialects show them in varieties of forms, not known to Sanskrit.
(a) There are six suffixes in Apbh to express the infinitive idea : -evam, -evi, •eviņu, -ěppi, -ěppiņu and .iu. These suffixes belong to some earlier phase of the literary language. In the later phase three more suffixes were added : -aņa, -aņabã and anahi.
(b) All these suffixes are represented in one or the other modern vernaculars.
(c) The suffixes -ěppi, later -evi, and -ěppiņu, later -eviņu,

Page #55

44
A Grammar of Apabhramsa are also used as absolutives. They are doubtless the development of -tvi and tvīnam.
(d) The examples are : d-evam (to give), pāl-avi (to protect), sam-var-eviņu (to cover), j-ěppi (to conquer), ca-ěppiņu (to give up), bhajj-iu (to break), kar-aņa (to do), bhujj-aņaha (to get), etc.
110. Towards the end of Apbh, the first six suffixes went out of use, and the last two merged together as -anā.
(a) In Hindi, .na is the sign of infinitive.
Absolutives
111. It has been pointed out above that the suffixes-ěppi and -ěppiņu as well as -evi and -eviņu are used both for infinitives and absolutives. We may also postulate an intermediate stage *-etpi between -ěppi and the Vedic -itvi. And, needless to say, -evi appears to be the latest in development.
112. The use of the forms like kar-evi, kar-eviņu, etc. both for the infinitive and absolutives may be explained from the fact that the infinitive forms like -tum, -tave and the absolutive form -tvā in the Vedic may be traced back to only one base in tu.
113. In some cases. -ěppi is also abbreviated as -pi : gampi, etc.
(a) Sometimes a weaker form with -evi also is seen : pi-avi (after drinking), etc.
(b) But the real weak form of ovi is -ivi : jbā-ivi (after thinking), děkkh-ivi (after seeing), etc.
114. This -ivi was further reduced to -i, which is the youngest suffix in Apbh : kar-i mār-i, paitthi, etc.
Upto Apbh, this is the whole story, and this was carried on to Awadhi.

Page #56

Conjugation
115. Towards the beginning of Hindi, -i appeared to be too weak to express the absolutive idea. Then the form in question with i was reinforced by another absolutive form kar-i: mār-i kar-i (having killed), paitthi kar-i (having entered), etc. may be the examples of the dark period.
45
(a) Finally, i was dropped from both the forms, so that kar began to be recognised as absolutive suffix in Hindi: mar kar, paeth kar, etc.

Page #57

CHAPTER FIVE
Word-formation, Compounds
and Indeclinables
116. In the first chapter we have discussed the origin of Apbh from its immediate predecessor Prakrit and its development into the modern vernaculars. In the second chapter we have discussed the Apbh sounds as far as they have evolved from Sanskrit and middle Indic and merged with Hindi and 0.) er modern vernaculars. In the third chapter the declension wi nouns, pronouns and adjectives has been taken up as briefly a3 possible. In the last chapter we have taken up the verbs in all their details.
In this chapter we should have taken up only the indeclinables. But it could have ended in one or two paragraphs. Therefore we are going to discuss all the little things like wordformation and compounds. In the next chapter we will fully discuss the Apbh syntax.
Word-formation
117. In the literary Apbh we find that the words of Sanskrit origin are freely used with the appropriate phonetic changes. But there are a few primary suffixes also by which

Page #58

Word-formation, Compounds and Indeclinables
47 words are derived from the Apbh verbs. The most common suffixes are :
-ira, showing habit : bham-ira (traveller), etc. -aņa-a, making agent nouns : mār-aņaa (killer), boll-anaa
(speaker), bajj-aņaa (drummer), etc. The suffixes making participles have already been discussed.
(a) The secondary suffixes are more numerous in Apbh, and all of them represent the historical development of some vedic suffixes : -a is the word-extending suffix, without any meaning (=Skt.
-ka): dittha-a (seen), gaa-a (gone), bhagga-a (wounded) etc. The fem. form is extended by -ā : gai-ā, mui-ā,
etc. -da also is a similar suffix: diaha-da (day), mār-ia-da
(killed), etc. Sometimes there is extension both by •a and -da : hiaa-da, then also hiaa-da-a and hia-da-a (heart), etc.
When the fem. -di ia added, the preceding vowel, whatever it is, becomes a- : gori (fair girl) +ại=gora-ci, ratti+di == ratta-di (night), etc.
-ā and -1 make fem. : dbanā (lady), gai (gone), etc.
-1 (Vedic -tāti) make abstract nouns: dhan-āi (lady-hood), badd-āi (greatness), etc.
-ima (Skt. -iman) makes abstract nouns: vamk-ima (crookedness), mu::s-ima (manhood), etc.
-ma (Skt. -maya) denotes made of : vajja-ma (made of thunderbolt).
ttana and -ppaņa (Vedic -tvana) make abstract nouns : badda-ttaņa and badda-ppaņa (greatness). The shorter form -pā is seen in Hindi.
-āra is a possessive suffix with some pronominal bases : tumh-āra (your), amh-āra/mah-āra (our), etc.

Page #59

. A Grammar of Apabhramba
Compounds
118. In literary Apbh there is plenty of compounds, generally of three or more words : agalia-neha-nivatta (a man of unending affection), etc. But in ordinary Apbh, there are examples of two words coming in comopounds : sasi-rāhu, etc. Some examples are just the recastings of the Sanskrit compounds : giri-simga (hill-top), haya-vihi (of doomed fate). etc.
(a) In Apbh itself we see such features as are usual in Hindi compounds. For example, the shortening or loss of the final vowel of the first member : jarā-khanļitā> jară-khamdı (tattered). dūra +uļļāņa=dūr-uddāņa (high flight), etc. Compare Hindi : pāns+gbāț -pan-ghas (water-bank), ghoṣāt savār=ghur savās (rider), etc.
(b) The types of compounds may be brought under four heads :
1. The socalled Tatpuruşa of the Sanskrit grammarians : giri-simga (bill-top), joana-lakkha (lakhs of miles), dukkha-saya (hundreds of miseries), kāla-kheva (passing of time), etc.
2. The socalled Karma-dhāraya of the Sanskrit grammarians: khara-patthara (hard-stone), kadu-pallava (bitter leaf), etc.
3. The socalled Dvandva of the Sanskrit grammarians : uttha-vaisa (rising and sitting), sasi-rāhu (moon and Rāhu), etc.
4. The socalled bahuvrihi of the Sanskrit grammarians : tosiasa mkara (who has appeased the Lord Siva), catta -mkusa (who disregards god), etc.
Some archic compounds based on Sanskrit models are found, but they hardly deserve any notice here.
Indiclinables
The indiclinables may be distinguished as underived and derivable.
119. Some of the underived indiclinables are just adverbs,

Page #60

Word-formation, Compounds and Indeclinables
developed from Sanskrit by the usual phonetic change : ajja (today), embai/embabi (thus), kaba vi (any how), kivã/kās/kai (why), cia (just), ji (just), to (then), to vi (even then), tti (thus), pacchai (later), na vi (not at all) vi (even), sahũ (with) hu (indeed), etc.
Some others are conjunctions : aha (and, if), aha...aha (either...or), aha va (or), jai vi (even if), jau (if), dhrũ...tram (because...so), nāí (as if), ju (that), etc.
Some of them are meaningless : ghaî, khāī, etc. Some others indicate manner, location, quantity, limit :
Manner : kema, jema, tema; kevã, jevã, tevã, kivã, jivã, tivã, kiha, jiha, tiha; kidha, jidha, tidha; etc.
Location : ketthu, jetthu, tetthu, etthu Limit: jāvā, tāvā; jāmabi, tāmahĩ; jāma, tāma; etc.
(a) Some indeclinables are loose combinations : vāra i vāra (again and again), ěkka i vāra (only once), etc.
These and other indeclinables are to be learnt from the dictionaries as regards their various meanings and idiomatic usage.

Page #61

CHAPTER SIX
An Outline of Syntax
120. Apbh is the last mile-stone in the march of the IndoAryan language from the inflexional to the isolating type. On the one side there is Apbh of fully inflexional character and on the other there is Hindi (or any other modern vernacular) of completely isolating type.
We cannot also imagine the embarassment of the speakers in the socalled dark period when the old forms were being left out and new ones adopted.
121. The gradual change in morphology was responsible for simplification in the sentence pattern. A sentence like kara-i would have emerged as karanta ahai before its transition to the present Hindi pattern. Similarly, the absolutive paitthi would have passed through the stage paitthi kari, the future karibi through karibi gā, and so on, before the emergence of the corresponding Hindi sentences.
But, instead of dallying with these hypothetical sentences, we should deal directly with the Apbh sentences and compare them with those of the modern vernaculars.

Page #62

An Outline of Syntax
51
Word-order
112, The word-order in Apbh is much more like Prakrit than like Hindi. That is mainly due to the inflexional character of the former.
(a) As a rule, the subject heads the sentence and the verb ends it. The object is kept nearest to the verb, unless other cases and particles of importance intervene. This rule is, however, not strictly followed in the metrical portion of the literature. Some typical examples are : FroT 061 Tag =One gets the fruit afhgfyra fa qfiets The master deserts even a good servant. ATG ATT. tg fag=The time passed as in the heaven. 363 fa53 U Top One spoke non-sense. अहरै अहरु न पत्त =The lips did not reach the lips. केसरी न लहइ बोड्डिअ वि=The lion does not get even a shell. =सायरु उप्परि TTTg=The sea keeps straw at the surface. .
(b) The attributive adjective generally precedes its substantive, and a genitive form precedes even this adjectives : धवलु विसूरइ समिअहो गरुआ भरु पिक्खेवि =The white bull is sorry to see the heavy burden of the master.
Numbers
123. The sg. and pl. numbers are well-marked even in Apbh. Generally the sg. and pl. forms are distinguished only in oblique cases. The subject and object show their numbers through their verbs : मल्लजुज्झ ससिराहु करहिं=The moon and Rahu are wrestling. 3176 Haarfa=The servants get respect.
(a) The adjectives also show the numbers quite usefully in case of possessive forms of nouns which are generally without endings : अइमत्तहं चत्तंकुसहं गय कुंभई=The temples of the intoxicated and chainless elephants. बढ चिन्तन्ताहं =Of the fools who are thus thinking.

Page #63

52
A Gram mar of Apabhramsa
Concord
124. The adjectives have to agree with their substantives in number, gender and case : Are HETTB=My husband. de HETZT ping=See my husband.
(a) When the subject and object are without endings, the agreement is made clear by the pronouns and verbs : 256T ET 13=Those are big houses. 93 fa ger=These are those horses.
(b) Sometimes the agreement is just to be guessed : 7 TOT गोवइ अप्पणा=Who conceals his own qualities. बालहे जाया विसम TUT=The girl has developed heavy breasts.
Pronouns
125. The varieties of pronouns are already on decline in Apbh. The distinction of geneders is almost totally lost, because a few existing distinct forms are colourless. The grammarians try to show the existence of even neuter forms, but their formulations are contradicted at many places.
(a) There is co-relation of jo - so (who -he): afsauto fa जो मिलइ सहि सोक्खहं सो ठाउ=Who meets even after hundred years, friend, he is the place of joy. This correlation is found in other cases also : 77 TOT as sequr TEEB af ff1573 =Who conceals his own qualities, I offer myself to him.
(b) There is a similar correlation of the pronominal adverbs : fafaqat ar73 75 fa fou at fa Ď SITUTE 3T50=Even if my lover is an offender, bring him today even then. Similarly, jivă - tivã (as - so), jahi - tahi (where - there), etc.
(c) The relative pronoun ju (that) acts as conjunction, often connecting two sentences : भल्ला हुआ जु महारा कन्तु मारिया (It was fine that my husband was kllled). In Hindi it has been replaced by the Persian ki. In Apbh it also acts as pure rela

Page #64

53
An Outline of Syntax tive pronoun : T ITEOS =Whatever else is thin.
Government of cases
124. In Apbh the forms of subject and object had, for the most part, fallen together both in sg. and pl. Therefore, the subject and object could be recognised in the sentence only by position and context. As a rule, the subject heads the sentence and the object immediately precedes or follows the verb : मुणि फलु लहन्ति =The sages get the fruit. कन्तु पयासइ मग्गु=My husband enlightens the path. 31778 fra hafa=Servants get respect.
127. The objects are used adverbially as in Sanskrit : GTSFS Fart=I am looked at hundred times. Hufe 75 a17= You speak again and again.
(a) As already pointed out, subjects and objects are formally indistinguishable : Arfa afyoz fa afTETE FANS TATS =The master deserts even a good servant and gives respect to rogues.
128. The instrumental forms are used to express the subject in passive sentences, when the object determines the form of the verb : Hoģhog forf FFG-A fish is devoured by another fish. जे महुदिन्ना दिअहडा दइएँ पवसन्तेण-The dates which were given to me by my departing husband, CT HE TË afer=Dear, I have prevented you. faets hg #furet gå=My daughter, I have told you. Fifah Erforate=Manhood is recognised.
(a) Gradually, the passive construction lost its ground, when the passive bases in -ijja- lost their force and began to be used as simple verbs. Then the instrumental forms became unnecessary, and the status of many pronominal forms was changed. The most remarkable was the shifting of the instr. sg. mas and tai/pai to nominative position. In Awadhi mai and tai were adopted as subject, and Hindi required the assistance of an obscure particle ne in the past tense of the transitive verbs : # (I said).

Page #65

54
A Grammar of Apabhramsa
(b) After such shifting of the instrumental forms to nominative position, the instrumental ending was lost in other cases : तिण सम गणइ=He counts as straw. चलिम वीर हम्मीर 9T37777 ÀGUTT 743=When the brave Hammira started to go, the earth trembled due to the heaviness of his feet.
Its resistance for survival was excellent and this case, along with the locative, was the last to die on the door of the modern vernaculars.
129. From the very beginning of the Indo-Aryan, it has been standing for the instrument of the action : बंकेहिं लोएणहिं जोइज्जउँ=I am looked at with oblique eyes. अग्गिण उण्हउ होइ जगू =The word is heated through fire. FOTTUT GEHT 9F=The house has been burnt by the fire.
130. Sometimes it shows cause ; 78 TOT 973451 at af Ho fotor, 75 PTT STEFAT at a Afcorfu=If the enemies fed away, it was due to my husband; if our own soldiers fied away, it was due to my husband being killed.
131. Sometimes utility is expressed : gră argou To Bayu our go-What is the use of the son when he is born? What is the harm when he is dead? at a afro que=Some purpose is served by that fire.
132. Often price is indicated : 174 parelle Sofra-Elephants are purchased for lakhs.
(a) Even some particles govern this form : faute Without you. मई सहुँ=With me. कालखे काइं= What is the use of passing time?
133. The ablative forms stand for the preposition 'from': वच्छहे गण्हइ फलई जणु =One gets fruits from the tree. गिरिसिंगहें पडिअ=Fallen from the hill-top. गिरिहे सिलायलु तरुहे फलु=Rocks from the mountain and fruits from the tree.
(a) Like the instrumental, it expresses cause too : at forma 779pug=Then why am I looked at ?

Page #66

55
An Outline of Syntax
134. Genitive, in which the dative was merged long ago and which was the first to lose endings in Apbh, expresses possession, purpose, etc. : q7F1 TT Others qualities. IfHTET 7=The burden of the master.
(a) As a rule, it stands for any preposition of English : Tas are feug=I offer myself to him. itfa3 TL 7 DESTES ETUT TT ATT 7 ET1=To whom the life is not dear? To whom the wealth is not desirable ?
(b) Often it stands for other cases : मझु खमेज्जउ भविय जणु =Noble men may forgive me. pag HET3 ATE =My husband is angry with him. The farat FETEST=Which dates were given to me.
135. Towards the end of Apbh, the genitive forms had started losing endings. Generally the nouns lost the endings and the adjectives retained them : geš a far qogget fael=While the fools are thus thinking, it becomes morning. अइमत्तहं चत्तंकुसहं गय कुंभई=The temples of very intoxicated and chainless elephants. रज्जु विहीसण सवलु वि देवी=Giving the whole kingdom to Vibhışaņa.
(a) The absolutive use of genitive is also known to Apbh: रामहु तिहिं पुरि रज्जु करन्तहु=While Rama was ruling in that city. fou utafa ETH=While I was looking at my lover's face.
136. The locative forms are used in the sense of in, on : Firafe T afourate=Who is praised in hundred of battles. भाइरहि मग्गेहि तिहिवि पयट्टइ=The river Ganges moves on three paths.
(a) Even the locative forms are used absolutively : aruf f7373 fango TERE 7 32get=In spite of the sea being filled with pure water, you do dot get even a single drop.
(b) Though the absolutive construction has continued in Hindi, the place of the genitive and locative has been taken by

Page #67

56
A Grammar of Apabhramsa
the oblique form connected by the postposition ke : uske rahte =While he is there.
Tenses and moods
137. The present, past and future forms express the action in the respective tenses : मल्लजुज्झ ससिराहु करहिं=The moon and Rahu are wrestling. भल्ला हुआ जु मारिया=It was fine that he was killed. freg TPHET 27781=The night will pass in slumber.
(a) The forms like thet are the youngest remanants of Abph, which are seen even in Awadhi : उर अपराध न एकउ घरिही =He will not mind a single crime. This seems to be a point where the periphrastic forms with gā would have appreared in the dark period.
(b) The present forms also express the desire of the speaker : aces715 Titag=I would see how many steps does he go.
(c) In historical narration the present forms often express the past happening: जाइ कालु सुरलोए जिह-The time passed as in heaven.
138. Of the model forms
(a) The third person expresses desire or request : HT= Let him go. 13 = Let it be.
(b) The second person expresses order : FAT3 TAHTE=Donot prevent him from going. | 3TTE 3456 = Bring him today. aaHETT forg=See my husband.
(c) The first person expresses the will of the speaker : ang=Let us see. grš=Let me go.
Passive
139. The passive construction is still very lively in Apbh. The forms are found both in present and past. Ho Hoga

Page #68

An Outline of Syntax forferegg=A fish is devoured by another fish. afar favors=I am being offered. जोइज्जउँ=I am looked at. दइएँ पवसन्तेण दिअहडा दिन्ना=Dates were given by my departing husband. मई नुहुँ atfo=You have been prevented by me.
(a) At some period in the history of the language, the passive force was lost from the passive bases. Then new passive forms began to appear by the combination of the past participle and jā (go). A sentence like kahia jāi is the predecessor of the Hindi sentence kahā jātā ha e (It is said).
Present Participle
140. The present participle is an adjective qualifying the subject of the main verb : oft 3togfrau arata=By her, living in the house of Rāvaņa. 758 Tahdo=By my departing husband. वायसु उड्डावन्तिअए पिड दिट्ठउ=I saw my husband while scaring the crow. गय कुंभई दारेन्तु देक्खु महारा कन्तु=See my husband, tearing the temples of the elephants. ad 565 चिन्तन्ताहं पच्छइ होइ विहाणु=While the fools are thus thinking, it becomes morning.
(a) Before the emergence of Hindi, the nasal element of the participle was lost and the auxiliaries like ahai began to combine with it to make new tenses.
141. Besides being used as adjectives, it had developed a peculiar use of expressing condition : लज्जेज्जन्तु वयंसिअहु जइ भग्गा of gag=I would have been put to shame, had he come back wounded. At gifstoure at the Af aff# Fifacere =If the moon could be fashioned, then only he could get the similarity of the face of the fair girl.
(a) This peculiar use of the present participle has also come down to Hindi.

Page #69

A Grammar of Apabhramsa
Past participles
142. The Apbh participles are used as main verbs in the past tense : एम्वउ सुरउ समत्तु=Thus the love ended. विहिं पयारेहि गइअ धण=The lady is lost in both ways. वइदेही पंचत्ति गय=Sita died. 5E FEHET 7991=If she loved me, then she is dead. अम्बणु लाइवि जे गया-Who went after making love.
143. But very frequently it is also used as adjective: stot go कासु न इठ्ठ=To whom the wealth is not dear. फल लिहिया मुंजन्ति =One enjoys the fruit as ordained.
144. Gradually the past participles began to lose the adjectival force. When it was reinforced by the auxiliary element huā, it was no more Apbh.
(a) In the eastern dialects, the adjectival force was retained by the reinforcement suffix.da> -la : Sutta-la (asleep), etc.
145. In Apbh, we find the past participles being used as new verbs : sukkha-i (gets dry), paittha-i (enters), etc.
146. When this past participle was extended by hae and thā, there was the origin of the Perfect and Pluperfect tenses in Hindi,
Infinitive
147. The infinitive forms show purpose : da 57778 faetat atat =It is difficult to give one's wealth. afon agfaqur FTTT tt popote paafa=Who is able to conquer the whole earth and then forego it. E TUTE T=There is a desire to enjoy happiness. FICUT 77 73 9f%878=Practising penance is not palatable. This last example is nearer to Hindi karnā.
148. There is also a peculiar use of the infinitive with the verb ja भुज्जणहं न जाइ=It cannot be enjoyed.

Page #70

An Outline of Syntax
Absolutive
149. The absolutive form expresses the action of the subject which precedes the main verb : कर चुम्बिवि जीउ रक्खइ =Maintains life by kissing the hands. at fautefa af =You go leaving my arms. HT75 EU 45€fy=It kills after entering into the heart.
(a) When geeft was reinforced by aft, it was the exit point of Apbh.

Page #71

Appendix
Post-Apabhramśa vernaculars
Though the Apbh language has been described in the preceding chapters, the description should not be taken to be comprehensive. It has touched all the important items of the Apbh language, but it has no reference to the dialectal variations. It should, however, be borne in mind that Apbh had several dialects, each going back to the respective Prakrits and each being the parent of the different vernaculars.
The dialect of Apbh which has been described by Hemacanda and which is the basis of the present grammar, is more or less the parent of Awadhi. The language of the Ram Carit Mānas of Tulsi too is said to be Awadi, but Tulsi did not write in the language he spoke. It was about hundred years older than him, which is clear by his use of the numerous duplicates. It was rather a kavi-bhāṣā or sant-bhāṣā, which was more widely understood than spoken. It was a proto-type of the Ardhamāgadhi of the Jaina canons.
The central structure of Apbh was what one is likely to see in the present grammar, but showing the dialectal variations in Apbh was not my target, While showing the evolution of some modern Hindi forms a certain dialect of Apbh spoken around Delhi and Agra has been reconstructed whenever needed. But that dialect is not known to Hemacanda. The identification of this dialect of Apbh may be a good piece of research.

Page #72

Appendix
A slight different variation, being the parent of Brajabhāṣā, may be reconstructed or identified in some more advanced research.
Though we have heard about a Māgadhi Apbh, which is the parent of the eastern dialects like Oriya, Bengali, Assamese, etc., no grammar is available. With due regard to the eastern peculiarities in grammar and vocubulary, the Apbh texts of Vidyāpati can be studied by the help of this grammar.
A Bihārī language too is supposed to have evolved from the Māgadhi Apbh, but the various natural obstacles in Bihar did not allow the growth of a common Bihārī language. Therefore, three dialects called Maithili, Magahi and Bhojpuri began to develop independently. Maithili has by this time acquired a beautiful literature, though it has given up its old script and adopted Nagari.
A Bajjikā dialect is also postulated for the Tirhut division of Bihar, But it is probably an import from the western and central Uttar Pradesh by some migrating communities, as is clear from several non-Magadhan elements in grammar and vocabularly.
In the present book our comparison has been restricted to Hindi and Awadhi. A skeleton grammar of Hindi and Awadhi may be quite worth the while.
Hindi grammar
The sounds of the language are the same as those for Apbh. But the conjunct consonants have been simplified by lengthening the preceding vowel : āj for ajj (today), fia for nidda (sleep), etc.
The typical Hindi words end with a consonant or with the long vowels ā,1,ū : TT TTT Tear at tot coat 19 a.
There are two genders, masc. and fem., and two numbers, sg. and pl.
Hindi abounds in loan words from different languages.

Page #73

62
A Grammar of Apabhramsa
There are three cases. namely : Direct, postpositional and vocative. An outline of declension is as follows:
Nouns
Cases
Direct
sg. pl. sg. pl. sg. pl. sg. pl. sg. pl. लोग लोग रात रातें बच्चा बच्चे हवा हवाएँ साथी साथी sg. pl. sg. pl. sg. pl. बच्ची बच्चियां डाकू डाकू बहू बहुएं postp.
- लोगों-रातों बच्चे बच्चों-हवाओं- साथियों—डाकुओं-बहुओं Vocative लोगो-रातो बच्चे बच्चो-हवाओ-साथियो-बच्चिओ-डाकुओ-बहुओ
Pronouns
Pronouns
Direct
मैं
हम
तू
तुम वह
वे यह
ये
कौन
जो
Postp.
मुझ - तुझ - उस
उन इस इन किस किन जिस जिन
Adjectives
Only the adjectives in a change. Other adjectives are unchangeable :
Direct
अच्छा
अच्छे

Page #74

Appendix
Postp.
अच्छे
अच्छे
Vocative
अच्छे अच्छे The conjugation of verbs is lengthy because of the origin of many periphrastic tenses. Primary tenses and moods :
A. Conditional B. Continuous C. Past D. Modal forms ___sg. pl. sg. pl. sg. pl. sg. pl. m. ता ते रहा रहे आ ए ३. ए एं f. ती ती रही रहीं ई ई २. ए ओ
Periphrastic tenses :
A,B,C+
. sg. pl. sg. pl. sg.
३. है हैं A,B,C+m. था थे D+m. गा २. है हो f. थी थीं . गी
pl. गे गी
Passive root+या+Verb जाना Causal root+वा/ला Present participle : A+हुआ Past participle : C+हुआ Infinitive करना चाहना, खाने जाना Absolutive खाकर, जाकर
Besides, there is reduplication of forms for various purposes. A number of auxiliary verbs (कर चुकना, बोल देना) and compound verbs (मालूम करना) have also developed.

Page #75

64
The sounds in Awadhi are the same as those for Apbh. The conjunct consonants have been simplified as in Hindi.
There are two genders and two numbers as in Hindi. There are innovations both in declension and conjugation. The number of cases persist to be as in Apbh, but the endings have dropped down and the postpositions are unable to modify the form of the words.
There is -u in nom. acc. sg., specially in the past participles: गयउ, भयउ, भुआलु etc.
Though the endings had been lost, we find मुखनि for मुखेन, and so on. The postposition (in), with its variants T, etc. were widely used. Usually hi or hu was used to save metre. The locative absolute was in use:
A Grammar of Apabhramsa
Awadhi grammar
Z Corresponding to # there is . The object form a atat are notable. Often the postpositions have merged with base:
मोसों, arai
The older genitive forms persist to stay:
जासु arg
कासु
But the earlier कोवि जोवि was modernised as कोई जोई.
The present endings imitate the Apbh pattern :
३.
sg.
2.
pl.
fe
इ
fer
v
करहि
करह
करहुँ
But the archaic forms विराजति, होति etc. are widely used.
The past tense in Awadhi is represented by the past participles as in Apbh, and the forms are generally extended by -u: भयउ, गयउ.
करइ
करसि
करउँ

Page #76

Appendix
Awadhi continues to form future forms by the affix -hi/baand the present endings : 3p. afzafe, Tefé, etc.; 1 sg. aftes, etc. In 3sg. the ending has contracted with the base : Aftet, oftet, etc. But there are also examples of b-future as in the eastern dialects : 957, etc. Sometimes there are reinforcements by -gā as in Hindi : 3789 #fat alfe, etc.
The modal forms seem to be more archaic in Awadhi. The optative forms like 3. sg. şê seem to be sharply distinguished from the imp. 475, etc. Similarly, Opt. 3. pl. Gori, Afife : Imp. पुरवहु 2. sg. opt. दीजै, कीजै : Imp. जानव, लाइव; and 1 sg. subjunctive कहौं, करौं shows deviation from the present कहउँ, करउँ
There are also polite forms with -ia : #frar
The passive and causal bases in Awadhi are the same with -ā (as often in Hindi : qarat - TRÈ FER qarat).
The examples of denominatives are rare. The present participles (777, a, fagca) and past participles (Fies, 3) are abundant.
The infinitives are nearer to Apbh : trei ES, af AT The absolutives never require the assistance of kar.i : 975, afe, etc.

Page #77


Page #78

INDEX
Absolutive 44, 59 Adjective, attributive 51 Aggi 22 Apabhramśa 5, 9
-Sounds of 10 Ardbamāgadhi 4 Awadhi 60
- grammar 64 āya (this) 27 Bajjikā 61 Bachché 15 bālā 23 bases, verbal 33 bhāşā 3 Bibari language 61 Brajbhāşā 61 Buddha 3 Cases 19 (in Hindi) 20 Cattāri 28 Causal 41 Compounds 48 Concord 52 Consonants, classified 13 Co-relation (of pronouns) 52 dark period 7, 38 decade (of numbers) 29 desi (word) 18 denominative 4 i dhamma 3 donni 28
eba 27 ěkka 28 ending, case 16-17 verbal 31
in Apbh. 32
in Hindi 32 forms (imp., subjunctive, opta
tive) 37 genitive 54 ghodā 22 historical survey
of vowels 11
of consonants 14 Hindi grammar 61 Hazār 19
ho- 33 Indeclinable 48 Infinitive 58, 43 Instrumental 53-54 Ja-(who) 27 Kālidāsa 6 Kanta 21 Kar-, present forms of 35 Ka-, kavana (who) 27 Kavi-bhāşā 60 Koi 28 Loc. 54 Māgadhi 3, 5 Mahārāştri 5 Middie Indo-Aryan 1

Page #79

61
Index
Mood 30-31, 56
-conditional 39 Number 51 Ordinal (number) 29 Object 53 Oba- (that) 27 Pabucca- (to reach) 34 Pali 4, 8 Pāṇini 3 Passive 53, 56, 39.40 Patañjali 6 Persian (language) 18 Participle,
Sabi (girl-friend) 23 Sāmi, Sāmia- 22 Sanskrit 4 Sauraseni 5 Suffixes
primary 47
Secondary 47 Subject 53 Ta- (he) 26 Taru- 22 Tense 56
present 30 past 35, 30
future 36, 30 Tinni 28 Tulsi 64 Vabū 23 Verb, secondary 39 Venacular, modern 7, 18 Vikramorvasiya 6 Vowels;
classified 11 distingushed 11
pronounced short 11 Word order 51
present 42, 57
past 42, 58
future 43 Phonemic value 11 Possessive 51 Postposition 20, 24 Prakrit 4, 9 Pronoun I p. 24
II p. 25 Relative52 Rām carit Mānas 60

Page #80
VIDYANIDHI PRAKASHAN D-1061, Gali No. 10 (Near Shri Mahagauri Mandir) Khajuri Khas, Delhi-110094 Phone : 2175638