Wilson Philological Lectures on Sanskrit and the Derived Languages c8.s1: The Apabhraṁśa

Wilson Philological Lectures on Sanskrit and the Derived Languages c8.s1

Language: English

c8.s1.14 — 3 parts

The Apabhraṁśa

c8.s1.1

The dialect called the Apabhramś’a by the grammarians presents Indian speech in a further stage of decay and occupies a middle position between these Prâkrits and the modern vernaculars to some of which, especially to the old Hindî, the Brajabhâshâ, and the Gujarâtî it bears striking resemblances, as I shall hereafter show.

c8.s1.2

As mentioned before, we have the grammar of this dialect from Hemachandra, Trivikrama, and Kramadîśvara; but Vararuchi does not mention it. The Apabhramśa had a literature of its own. Hemachandra illustrates each of his rules about this dialect by quoting a verse. In the fourth Act of the Vikramorvaśî the Prâkrit speeches of the king in his madness are in this dialect. But it is a question whether they existed there originally, since in several manuscripts on this side of India they do not appear. The metres employed in these and in Hemachandra's quotations are the same as those popularly used in old and modern Hindî or Braj, viz., dohâ or chopai. Pandit Vrajalal mentions a work of the name of Muñjarâsa, written in the Apabhramśa, from which he gives a short extract, and another the hero of which is a king of the name of Prasenajita. He also quotes from another work; but the language of all these appears to be more modern than Hemachandra's Apabhramśa, and that of some of the verses makes a very near approach to the Gujarâtî. As before, I will introduce my remarks on this dialect by a specimen: —

c8.s1.3

एत्तहे नेत्तहे वारि घरि लच्छि विसंडुल धाइ ।
पिअ पब्भट्ठव गोरडी निच्चल कहिंवि न ठाइ ॥
जे मह दिण्णा दिअहडा दइं पवसन्तेण ।
ताण गणन्तिअ अङ्गुलिउ जज्जरिआउ नहेण ॥
जीविउ कासु न वल्लहउं धणु पुणु कासु न इट्टु ।
दोण्णि वि अवसरि निवडिअइं तिणसम गणइ विसिट्ठु ॥
जो गुण गोवइ अप्पणा पयडा करइ परस्स ।
तसु हउं कलिजुगि दुल्लहहो बलि किज्जउं सुअणस्स ॥
गङ्ग गमेप्पिण जो मुअओ जो सिवतित्थ गमेप्पि ।
कीलदि तिदसावासगउ सो जमलोड जिणेप्पि ॥

c8.s1.4

Sanskrit:

c8.s1.5

अत्र तत्र द्वारे गृहे लक्ष्मीर्विसंष्ठुला धावति ।
प्रियप्रभ्रष्टैव गौरडी निश्चला कापि न तिष्ठति ॥
ये मह्यं दत्ता दिवसा दयितेन प्रवसता ।
तेषां (तान्) गणयन्त्या अङ्गुलयो जर्जरिता नखेन ॥
जीवितं कस्य न वल्लभं धनं पुनः कस्य न इष्टम् ।
द्वे अप्यवसरे निपतिते तृणसमे गणयति विशिष्टः ॥

c8.s1.6

यो गुणान्गोपायत्यात्मीयान्प्रकटान्करोति परस्य । तस्याहं कलियुगे दुर्लभस्य बलिं करोमि सुजनस्य ॥ गङ्गां गत्वा यो मृतो यः शिवतीर्थं गत्वा । क्रीडति त्रिदशावासगतः स यमलोकं जित्वा ॥

c8.s1.7

"The unsteady goddess of wealth runs to this door and that, this house and that; like a fair one separated from her lover she does not remain firm anywhere."

c8.s1.8

"My fingers have worn away rubbed against by my nails, while counting, again and again, the days named by my lover [as the period of his absence], when he set out on his journey."

c8.s1.9

"To whom is life not dear and to whom is wealth not an object of desire? But when the occasion comes, a worthy man regards them as straw."

c8.s1.10

"I worship that good man so rarely to be met with in this Kali age, who conceals his own merits and gives publicity to those of others."

c8.s1.11

"He who dies after having gone to the Gaṅgâ and to Śivatîrtha, triumphs over the power(world) of death and sports in the habitation of the gods."

c8.s1.12

The vowel अ of the derivatives of the pronouns यद्, तद्, किम्, and इदम् is changed to इ or ए. In the principal Prâkṛit we have noticed a tendency to such a change, and one of our instances was एत्थ for अत्र. From the same tendency we have in the Apabhraṁśa जेत्यु for यत्र, तेत्यु for तत्र, and केत्यु for कुत्र, the उ of the latter being replaced by उ in virtue of the analogy of the other pronouns. In the same way तद्वत् "in that manner," which after dropping the final consonant becomes तद्ध, is changed to तेम, यद्वत् to जेम, and किंवत् to केम. This ए is also rendered a more close vowel and changed to इ, and thus we have तिम, जिम, and किम. Thus the bases of these pronouns came to be considered as ते or ति, जे or जि, and के or कि, and so we have तिथ for तथा, जिथ for यथा, and किथ for कथम्. The mute element being dropped these forms become तिह, जिह, and किह, and further तेहु, जेहु and केहु. For similar reasons we have तेत्तिअ or तित्तिअ for तावत्क, जेत्तिअं or जित्तिअ for यावत्क, केत्तिअ or कित्तिअ for कियत्क, &c., even in the Prâkṛits. The nom. sing. termination of nouns in अ is shortened to उ, and since a great many nouns were pronounced with this final उ and its original sense was forgotten, it was transferred by way of analogy to other words or grammatical forms that did not possess it before; and thus we have पुणु and विणु for पुनः and विना, and जेत्थु, तेत्थु, जेहु, &c., in the above. The long vowels are sometimes shortened as in the word गङ्गा in the last of the above verses. The consonants क् and ख् are sometimes softened to ग् and घ्, instead of being dropped, त् and थ् to द् and ध्, as in the S'aurasenî, and प् and फ् to ब् and भ्. In the principal Prâkṛit also this is sometimes the case. The labial, व् is changed to म् in a few more instances than in the Prâkṛits as in जाम and ताम for यावत् and तावत्, and जेम or जिम, तेम or तिम, &c., in the above for यत्, तत्, &c. The complete contact of the lips necessary for the pronunciation of म् is avoided in a great many cases, and thus we have वूँ for म् as in भँवँरु for भ्रमर, कवँलु for कमल, &c. This is the prevailing rule in most of the vernaculars, as we shall hereafter find. The conjunct म्ह is changed to म्म्ह; that is, the aspirate ह has been labialized and assimilated to the preceding म्, as in गिम्ह for Prâkṛit गिम्ह and Sanskrit ग्रीष्म. A few instances of this change we did find in the Prâkṛits and we shall find more in the vernaculars. The Sanskrit conjunct त्व is changed to प्प; for as व was often pronounced as ब, this ब instead of merging into the preceding mute, as semi-vowels often do, acquired prominence as य does when preceded by a dental; and the त being assimilated to it, the whole become प्प as त्य becomes च्च, as in प्इ for त्वया, प्पण or पण for the abstract termination त्वन, and प्पिण for त्विन, the suffix of the absolutive. In the Prâkṛits the त्म of आत्मन् similarly becomes प्प and so we have अप्पा for आत्मा. The letter र् when the latter member of a conjunct is sometimes not assimilated, as in ब्रोप्पि, प्राउ for प्रायः, &c., and sometimes it is introduced even when it does not exist in the original, as in धुं for तद्, प्रस्स for पस्स, &c. With these few exceptions, the rules about the assimilation of conjuncts, the elision of consonants, and others hold good generally as in the Prâkṛits, as you will see from the above extract.

c8.s1.13

DECLENSIONS—Nouns in अ—The decay of the case terminations is, however, a distinguishing feature of this dialect. The distinction between the nom. and acc. case forms which we observed lessening at each stage in the growth of our languages is here altogether lost. The ओ of the nom. sing. of masculine nouns in अ is shortened to उ, and used in both the cases, as मुअउ nom. for मृतकः, गउ for गतः, जमलोउ for यमलोकम्; and applied to neuter nouns as well, as कमलउ. The nasal of the neuter is preserved only in nouns which are augmented by the addition of क changed to अ; as कमलउं for कमलकम्. The nom. and acc. pl of the masculine ends in आ; as दिअहडा, पयडा &c. in the above. The neuter nouns preserve the Prâkṛit इ of the plural, as in कमलइ. Sometimes words are used in these two cases

c8.s1.14.1

without any terminations; as विसंठुल nom. sing., निचुल nom. sing., गुण acc. pl., गङ्ग acc. sing., तित्थ acc. sing., in the above extract. The principle observable in the other cases is the same as we have noticed in the older dialects, viz., a gradual reduction of all the declensions to a uniformity. The instrumental singular of nouns in अ has two forms, one in ॲ as देवेॲ, and the other the old one देवेण. The former is derived from this old form, the final अ being dropped, and the nasal assuming the form of an anusvâra. This new termination is transferred to nouns in इ or उ also, as अग्गिॲ. The instrumental pl. is the old one हिं, but the change of the ending vowel to ए is only optionally made; as गुणेहिं or गुणहिं. One ablative termination is हे which is appended to all nouns, and the other हु as in वच्छहे गुणहइ फलइ or वच्छहु गुणहइ, "gathers fruits from trees." Of course we may trace the first to हिन्तो by supposing it first to be changed to हेन्तो by the usual Prâkrit rule, and afterwards to have dropped its second syllable; and हु to something else. But it appears to me that a good many of the terminations having been reduced to ह by natural phonetic changes, the others also had this inserted in them by analogy, just as the nominative termination उ is introduced in words and forms in which it did not exist. This process we shall necessarily have to suppose when we examine the forms of the present tense.

c8.s1.14.2

Or the ह may have been introduced simply to prevent a hiatus and connect the vowel-termination with the base. And this is rendered probable by Hemachandra's rule that the ending syllables उं, हुं, हिं, and हं, are to be lightly pronounced, as the य that is introduced in the place of a dropped consonant is. The real syllables in these cases are therefore अं, उं, इं, and अ; and this is confirmed by the fact that the remnants of these Apabhramśa terminations existing in some of the modern vernaculars are destitute of this ह and are mere vowel-terminations, as will be seen in a subsequent lecture. Thus, then, हे may be considered the aspirated form of the ए of the Prâkrit feminine ablative, and हु of the उ of the masculine ablative.¹⁴ The operation of the law of false analogies is very wide. The abl. pl. हुं, as in गिरिसिइगहुं, may with Lassen be traced to सुन्तो, the म् being changed to ह् and the syllable तो dropped, as ति is in the case of the 3rd pers. pl. of the present, as we shall see. The genitive singular सु, and स्सु as in सुअणस्सु and परस्सु in verse 4 above, I trace to the स of the Prâkrits, the स being added by analogy as observed before. Lassen traces is to स, which, he says, must have been added to these nouns to form the genitive; and the अ being softened to उ, it is reduced to सु. But this does not account for the double स्त of the other form; and the addition of such a word as स to make up a case-form is altogether unexampled.

c8.s1.14.3

The suffix हो, as in दुल्लहहो in the same verse, is but another form of सु. The genitive plural णं, as in माणुसणं for मनुष्याणाम्, may with Lassen be derived from the नाम् of the Sanskrit pronominal declension. But the transference of this termination to nouns is nowhere seen in the Prâkrits; णं is optionally appended to nouns in इ or उ also; and the characteristic णं of the gen. is wanting in this dialect. The more probable explanation, therefore, is that the णं lost its cerebral element and was reduced merely to a nasal अ, or अ with an anusvâra, as is the case in the instrumental singular, and the ह is added, as observed before, simply to facilitate the pronunciation. The loc. sing. ends in ए, as in तले, which we see is the old termination, or in इ, as in तलि, which is a shortened form of ए. Kramadīśvara gives also हि which might be traced to the Pâli म्हि, the Sanskrit स्मिन्, or the स्सिं of the Prâkrit pronominal declension. The language of Jaina works has this स्सिं in the case of nouns also. But even here the Prâkrit स्मि, may, with reason, be supposed to have changed to हि, and the ह to be as before, a mere spiritus lenis. We have this locative हि in the Marâṭhî. The हि is extended to the plural also of these nouns; The vocative is formed by using हो which is evidently an interjection; as तरुण हो "O young men."

c8.s1.15

¹⁴ Kramadîs'vara in Lassen's extract gives हु instead of हुं; but this is an evident misreading.

c8.s1.16

Nouns in इ or उ.—There is no distinction between the nom. and acc. sing. and pl., the original base being used without any modification or addition. The iustrumental singular is formed by adding ए or णा, as in अग्गिएं, अग्गिण, अग्गिं. As एण becomes एं by dropping the final vowel, so does the इणा of the Prâkrit instrumental here become एं. The abl. sing. ends in हि as in गिरिहे and तरुहे; and the gen. sing. also according to Kramadīśvara, Hemachandra being silent. These have been transferred from the feminine. Kramadīśvara also gives हो and हि as the terminations of the sing. of these two cases; but there must be some misreading here in Lassen's extract from the author, since हि is a distinctive plural termination, The gen. pl. termination is णं or णं as in तरुणं, सउणिहं (शकुनीनाम्). The latter has been traced to the old णं, and the former appears to be only another form of it with the addition of the usual उ. The loc. sing. has हि, as in कलिहि; and the pl. has हिं and णं, as in तरुहिं, and उडुहं. The first must be considered to be the same as हिं derived from सिंस or मि with the anusvâra dropped; or as the इ of Prâkrit feminine nouns with the spiritus lenis ह. This explanation seems to be probable, since we have seen other feminine terminations also used for the cases of these nouns. The third is to be traced to the Prâkrit ॲं. Kramadîśvara in Lassen's extract gives ॲं for the ॲं of the abl. pl., and ॲं for that of the gen. pl.; but I have to make the same remark here as before. The instr. and abl. plurals are the same as those of nouns in अ; as तरुहिं and तरुहुं. Thus the plural of three cases ends in हुं, derived separately of course, and the plurals of two in हिं.

c8.s1.17

Feminine nouns.—The plurals of the nom. and acc. of feminine nouns preserve the old ओ, or its shortened form उ, as in अङ्गुलिउ जज्जरिआउ in the second verse, and सुन्दरसव्वङ्गाओ विलासिणीओ. The sing. is the original base, as जज्जरिआ, विलासिणी &c. The instr. sing. termination is ए, the old one, as in ससिमण्डलचन्दिमए, कन्तिए, &c.; the abl. is हे, as in बालाहे for बालायाः, which is an aspirated ए; and the gen. हे as in मुद्धहे for मुग्धायाः (ह being a nominal suffix) may be similarly explained or traced to the ए of the gen. sing. fem. of the pronoun, किम्, यद्, and तद्, in the Prâkrit. I prefer the former explanation. This हे has been transferred to masculine nouns in इ or उ as we have seen. The abl. and gen. pl. ends in हुं, as in वयसिअहु for वयस्याभ्यः or वयस्यानाम्. The anusvâra of the हुं of the other classes of nouns is here omitted; and if correctly so, the ह of the abl. may be considered an aspirated form of the अ which is the ablative termination in the Prâkrit; and that of the genitive may be traced to the ॲं of the singular of masculine nouns. The instr. pl. and the loc. sing. and pl. take the same terminations as masculine nouns in इ or उ; as महिहिं, महिहे, and महिहिं-हुं. Lassen gives other terminations; but he appears to me not to have interpreted Kramadîśvara correctly. Those I have given are all that I have been able to find.

c8.s1.18

You will thus see how by the various influences at work, the natural transformation of त to ह, the elision of some of the elements, and the aspirated pronunciation of the vowels, most of the old terminations have been reduced to syllables composed of ह and a vowel with or without an anusvâra. Terminations with such weak sounds are not adapted to serve the purposes of ordinary intercourse, since they require on the part of the speakers such care in pronunciation to render themselves intelligible to each other as we have not seen displayed in the course of our lingual history. The nominative and accusative throughout, and in certain classes of nouns the other cases also, have come to have the same forms. So that the purpose of expressing the different relations can be no longer performed by these poor remnants of the old declensional system; and a process of reconstruction must take place. It has already begun in the Apabhraṁśa; but we shall find it carried on much further in the vernaculurs, which may now be said to have completed their new declensional system.

c8.s1.19

Pronouns.—The abl. sing. of pronouns ends in हं, as जहं, &c., which is to be traced to the Sanskrit स्मात् and the Prâkṛit म्हा, and the loc. sing. in हिं which has been explained. The gen. sing. of यद्, तद्, and किम् are optionally जासु, तासु, and कासु, made up by adding the usual ङ to जास, तास, and कास, which again are other forms of जस्स, तस्स, and कस्स, with one of the consonants dropped and the preceding vowel lengthened as a compensation, a phenomenon to be noticed in going over the vernaculars. In the feminine these pronouns have जहे, तहे, and कहे optionally for the gen. sing., which are derived from the Prâkṛit जिसे, तिसे and किसे and the Pâli किस्साय, तिस्साय, and किस्साय. The base, however, in the Apabhraṁśa ends in अ and not ई, or the more probable explanation is that this हे is the aspirated form of the ष् of feminine nouns. इदम् has आय for its base; the instr. and gen. singulars, for instance, are आयेण and आयहो. This seems to be taken from the nom. sing. अअ, or अय by the rule of the substitution of य for अ. The neuter nom. and acc. sing. is इमु. एहो is nom. and acc. sing. masc. of एतद्, corresponding to एसो, एह fem. to एसा, and by the application of ङ to these we have the neuter एहु. इइ is the nom. and acc. pl. answering to एए Prâkṛit, with the latter ए shortened; and ओइ of अदस् which corresponds to the form अमुके, the अमु being changed to अमुं by a rule before mentioned, and afterwards to ओ; and के to ए shortened to इ. In other respects all these pronouns are declined like the corresponding nouns. The pronouns of the 1st and 2nd persons are thus declined:—

c8.s1.20

| | 1st pers. | | 2nd pers. | |
| | sing. | pl. | sing. | pl. |
| Nom. | हउं | अम्हे, अम्हईं | तुहं | तुम्हे-तुम्हईं |
| Acc. | मईं | — | तईं, तइं | — |
| Instr.— | अम्हेहिं | पईं, तईं | तुम्हेहिं |
| Abl. | महु-मज्झु | अम्हहं | तउ, तुज्झ, तुध | तुम्हहं |
| Gen. | — | — | — | — |
| Loc. | मईं | अम्हहं | पईं, तईं | तुम्हाहं |

c8.s1.21

Here we have not the wilderness of forms which we observed in the principal Prâkṛit. The nom. sing. of the 1st per. is हउं corresponding to अहकम्, the initial अ and the क् being elided and the usual Apabhraṃśa उ added. In the Prâkṛit अहकम् is represented by अहअं or अहयं. मइं is to be traced to the Sanskrit instr. मया and the loc. मयि both of which are reduced to the form of मइ in the Prâkṛit. Here it is extended to the acc. also. The pl. base is अम्ह which has been explained. The nom. pl. अम्हइं corresponds to such a form as अस्मके nom. pl., not loc. as Lassen says, through अम्हए; and the gen. pl. has the Apabhraṃśa termination ंह. The rest are old. तुह is a base derived, as I said in going over the Prâkṛits, from तुह and is used in the Prâkṛit also. पइं and तइं of the acc. instr. and loc. are from तया and त्वयि. The second is found in the older dialects, the first is peculiar to this. Of the corruption of त्व to त्प I have already spoken. The abl. and gen. तउ corresponds to a form त्वक which with the Apabhraṃśic उ is तउ, the क being dropped. Or it may be traced to तव with the व dissolved into उ. तुज्झ is a form found in the Prâkṛits also and explained by me as a corruption of the dat. तुभ्यं in the Pâli and Prâkṛits, derived from the Skr. तुभ्यम्. The dative forms, you will remember, are put under the gen. in the older dialects. The अ of तुअ is a peculiar Apabhraṃśic conjunct formed from ज्झ. The neuter of तद् is similarly तं and ंतं. The base of the plural is तुम्हे, the same as in the Prâkṛit; and the nom. pl. तुम्हेइं is to be traced to तुम्भके for युष्मके nom. pl. All the forms are similar to those of the first personal pronoun.

c8.s1.22

The Present tense of the Apabhraṃśa verb admits besides those of the corresponding Prâkṛit or Śaurasenî tense, the following forms:—1st pers. sing. लहउं, pl. लहहुं; 2nd pers. sing. लहहि, pl. लहहु; 3rd pers. sing. लहइ, pl. लहहिं. It appears that some of these forms have arisen from a confusion of the Present Indicative with the Imperative. The prevailing and distinguishing final of the latter is the vowel उ; and here we see it is appended to the forms of the 1st pers. sing. and 2nd pers. pl., though it does not occur in those forms in any of the older dialects. The ह is another characteristic of these paradigms. That of the second pers. sing. we get from the old सि, and that of the pl. exists in the Prâkṛit, being derived from the Skr. थ. But the 3rd pers. pl. and the 1st pers. pl. get it simply by an extension of the analogy; or it may have been introduced to pervent a hiatus and thus may, like those of the cases, have been simply a spiritus lenis. The हं of the latter, however, may be traced to म्हो, ओ being shortened to उ, just as म्ह and म्ह of the declensions become हि and हिं. Instead of the 3rd pers. pl. न्त we have हिं, in which, though the ह is due to analogy, the characteristic ह and the nasal are preserved. You will thus see that in the declensions as well as conjugations the ह prevails in this dialect. The Imperative second pers. sing. ends in इ, ए or उ. The first two may be traced to the Prâkrit and Sanskrit हि, ह being dropped as in the Vernaculars; and the last seems to be substituted for the अ of one of the forms in the original dialects by analogy; or it may be the remnant of सु changed first to ह, and then to उ. But a better and, I may say, the true explanation of these forms will be given in going over the Vernaculars. Hemachandra does not give any more forms for the Imperative; but Kramadîśvara gives उ for the second person plural which must very likely be ह, the same as in the present, and हु for the first pers. pl. which we have in the Present also. In one of the verses quoted by Hemachandra occurs पिबहु which seems to correspond to पिबन्तु, if the reading is correct. The truth seems to be that the forms of the Imperative were lost, and the sense confounded with that of the Present, but the distinctive forms were those of the second pers. sing.; and in this respect there is an analogy with the Hindî and the Gujarâtî which resemble the Apabhramśa the most. The Future has the Prâkrit and Śaurasenî forms in हि and इस्स; but one म् of this latter is dropped, and then the terminations of the Apabhramśa Present are added. The terminations of the absolutive are इ, इउ, अवि, इवि, एव, एविवु, एप्प, एप्पणु. इ is the same as the Śaurasenî इअ from the Skr. य, इउ is the same, with the usual उ added to it, or is to be traced to the तुम or इतुम् which is, as we have seen, confounded in tha Prâkrit with the absolutive, and the rest are various forms of the Vedic त्वीन with the Prâkrit augment इ or ए. This त्वीन is, by a rule before mentioned, changed to त्पिण which with उ becomes त्पिणु; and by dropping the final न we have त्पि. This, however, may be derived also from such a form as त्वी found in the Vedas. This त्पि or त्पिणु is then softened to वि or विणु, as प् is so softened in many cases. When the augment इ or ए is not prefixed, we have वि in the form of अवि the अ being the final vowel of the root. Some of the terminations of the absolutive are also used to form the infinitive, on account of the prevailing confusion between the two. In consequence of this very confusion, recourse is had to other ways of forming the latter which are similar to those existing in the Prâkrits; but these will be noticed hereafter.

c8.s1.23

The termination तव्य of the potential participle assumes the forms of इव्वउं, एस्वउं, and एव. The first two represent the form with the addition of क; and the इ of इव्वउं is the usual augment. The अ of the त of तव्य which remains after the consonantal portion is dropped, is by the influence of the preceding इ changed to ए. In those points which are not noticed here, the Apabhramśa follows chiefly the Śaurasenî, and the principal Prâkrit also to some extent. Thus in a great measure it represents those dialects in a further stage of decay; but it must be considered to have derived some words or forms independently also. Thus the तु of the second personal pronoun cannot be derived from the Prâkrit तइ, nor त्तूण of the absolutive from तूण or ऊण, or पण of abstract nouns from तण, but directly from the Sanskrit त्वया, त्वीन, and त्वन. This corruption of त्व must have existed in some of the older dialects too, since, as observed before, we have it in Aśoka's inscriptions; and the Apabhramśa derived it as well as a few such peculiarities from them.