Play Editing and Prakrit Grammarians (2001)

Offprint from: Les sources et le temps. Sources and time. A colloquium. Pondi­cherry 11-13 January 1997. Edited by Fran9ois Grimal. Pondi­chery 2001. (Publications du departement d'indologie. 91.)

PLAY EDITING AND PRAKRIT GRAMMARIANS1

ROLAND STEINER

One of the peculiarities of classical Indiandrama is its multi-lingualism. Depending on the character, a person speaks Sanskrit, or different vernacu­lar dialects for which the general term Prakrit is used. The problern which an editor faces when dealing with the different Prakrits of a play is that, in centrast to Sanskrit, he is usually confronted with a relatively poor Prakrit textual tradition of less strictly normed languages. Moreover, our know­ledge of Prakrit is far below the Ievel attained in Sanskrit. With that the question is open whether, and to what degree, the playwrights actually fol­lowed the general rules of Prakrit, or, to what extent the original Prakrit of the plays had already been standardized. Further, phonetics clearly contributes less to the investigation of literary and artificial Prakrit of the dramas than is the case of middle Indian dialects which are orientated rather towards the spoken language, e.g., the Prakrit of the inscriptions, the Ardhamägadh1 of the Jainas, or the Pali of the Bauddhas.

Be that as it may, there is no Prakrit grammar which even approxim­ately equals that of Päi;tini and his school as regards its status and effect on the language as a whole. Therefore it is a matter of principle whether and to what extent traditional Prakrit grammarians should at all be con­sulted when publishing an Indian drama. There is by no means agreement on this question in the pertinent literature; the controversy going back to the last century is characterized by irreconcilable Oppositions. Richard Pischel, the author of the basic and highly meritorious work on Prakrit in 1900, holds the extreme view when he says "only with the help of the statements of the grammarians are we able to draw even an outline sketch of these dialects [Saurasen1, Mägadhl, etc.]: the manuscripts are often dis­appointing."2 In one context, when dealing with a particular rule, Pischel

1This paper is virtually an English extract of Steiner (1997), chapter 9.1. I thank Dr Harunaga Isaacson very much for his useful comments. I am also very obliged to Drs Jayandra Soni and Ian Astley for checking and correcting the English.

2 Pischel (1900) § 42: "Nur nach den Angaben der Grammatiker können wir uns ein ungefähres Bild von diesen Dialekten entwerfen; die Handschriften versagen meist."

64 Sources and time

states clearly that: "There can be no doubt that this, like all other rules of grammarians, has to be applied against the manuscripts", because "all grammarians from Vararuci onwards are substantially in agreement on the essential points." 3 Nevertheless, even Pischel admits: "Indeed, the teach­ings of the grammarians require examination and elaboration in detail." 4

The opinion that Prakrit editing has, above all, to follow the tenets of the grammarians has been subscribed to, among others, by Alfred Hil­lebrandt, the editor of Visakhadatta's play Mudräräk§asa. The underly­ing principle of dealing with Prakrit was formulated by him in an almost classical manner in the preface of his edition (1912, pp. iif.), and that is probably why it was quite often approvingly quoted. 5 Hillebrandt says there:

At all events, by following the rules of the grammarians we gain firm ground, while by following the manuscripts and their varying practice we are constantly troubled by the feeling of inconsistency. . . . therefore we are forced to normalize the text even where no manuscript authorizes us to do so.

Recently, Richard Salomon, following the arguments of Johannes Hertel,6

again stated the contrary position. He points out that

since Pischel's time, new forms of early dramatic Pkt.-most notably those of the Central Asian fragments of Asvaghoeya's plays-have been discovered which show that the grammarians' picture of the dialects was incomplete and therefore unauthor­itative. ... In determining the original readings of Pkt. texts, the only accurate and productive method is to give the greatest weight to the mss. themselves. .. . If the mss. lead us to unex­pected and unfamiliar results, we must try to understand these results ... 7

3 Pischel (1900) § 23: "Und doch kann es keinem Zweifel unterliegen, dass diese, wie alle andern Regeln der Grammatiker, gegen die Handschriften durchzuführen ist. Alle Grammatiker von Vr. an sind in der Hauptsache einig" .

4 Pischel (1900) § 42: "Gewiss bedürfen die Lehren der Grammatiker im einzelnen der Prüfung und Ergänzung."

5 Cf., e.g., Jackson/Ogden (1923) xciv, n. 10; Das Gupta (1962) [xl]. 6 Hertel (1924) 7f. 7Salomon (1982) 124f. and 137. Theodor Bloch (1893) 48 already arrived at the

same conclusion; cf. Konow (1894) 472 and Pischel (1900) § 42 both of whom quote and reject Bloch's argument of the passage in question . Michael Coulson (1989, xliii), the editor of Bhavabhüti's Malatimadhava, has argued in a similar, but less rigorous way than Bloch and Salomon, however without referring to the relevant literature. Herman Tieken (1983) 184 makes the following remark about the Prakrit form of the particle iva: "lt should a priori be noted that not much value should be attached to absence or presence in the grammars of a particular form of the particle, which may merely

. indicate that the grammarians' observations were restricted to certain texts only or to

Play Editing and Prakrit Grammarians 65

Now, it is well-known that the Prakrit of Asvagho~a had reached the phonetic stage of Pali and the dialects which had been used in the older inscriptions. Heinrich Lüders, the editor of the Central Asian fragments of Asvagho~a's plays, designates the three middle Indian dialects employed by Asvagho~a as "old Saurasenf', "old Mägadhf', and "old Ardhamägadhf'. He sees them as mere "forerunners" of the later Prakrits of the classical plays.8 In contrast to this Vararuci, the oldest Prakrit grammarian known to us (perhaps third or fourth century AD), teaches a middle Indian dialect which is close to the lyric poetry in Mähär~tri, especially to the language used in Häla's famous collection Sattasaf9 and which, therefore, reflects a later phonetic stage. Much points to the fact that these middle Prakrits, in contrast, for example, to the old Prakrits of Asvagho~a or the inscriptions, were grammatically fixed and came into literary use only after the second or third century AD, although the origin ofthe middle Prakrits is earlier. 10

Therefore, it is hardly surprising that the Prakrit grammarians known to us (including the "pseudo-Vararuci"), all of whom are relatively late, differ from the Prakrit used by Asvagho~a. Thus, Hertel's and Salomon's observation argues neither for the authority of the grammarians nor against it as far as the later plays of the classical period are concerned.

Salomon hirnself lays stress on the particular importance of the me­dieval dramas inscribed in stone, especially the Lalitavigraharäjanätaka and the Pärijätamaiijarfnätikä.U As he pointsout (not quite accurately): "Their special value for Pkt. studies ... lies ... in the fact that we have

certain MSS of texts." The significance of a specific grammarian, then, depends on when he lived and the scope of his Observations, or, to be more precise, on his closeness to the time and the area of the investigated text.

8 Lüders (1911) 61: "Die sprachliche Untersuchung ergibt demnach, daß in unseren Fragmenten neben dem Sanskrit drei Dialekte gebraucht werden, die auf der Lautstufe des Pali und der in den älteren Inschriften verwendeten Dialekte stelien und die sich als Vorläufer von drei der späteren Prakritdialekte, der Mä.gadhi, der Ardhamä.gadhi und der Sauraseni erweisen."

9 Luigia Nitti-Dolci has shown that the last three books (X-XII) of Vararuci's gram­mar dealing with Pai§ä.ci, Mä.gadhi, and Sauraseni are additions of later authors ("pseudo-Vararuci"). Her comparison ofthe grammatical sütras and the vocabulary of Hä.la's Sattasaihas revealed that Vararuci has apparently based his sketchy description of Mähä.r~tri on a collection of Gä.hä.s which was very similar to the transmitted Sat­tasai; cf. Nitti-Dolci (1938) 11-59; and, approvingly: Renou (1938) 162, Alsdorf (1941) col 140 [= Kleine Schriften 564); cf. also von Hinüber (1986) § 65.

10Lüders (1911) 64: "Wir müssen vielmehr annehmen, daß die Mittel-Prakrits erst nach· dem zweiten Jahrhundert n.Chr., wenn auch nicht entstanden, so doch gramma­tisch fixiert und in literarischen Gebrauch kamen." At any rate, we can only conjecture that, apart from the Mä.hä.r~tri grammar, there already were "grammars" for the dif­ferent dramatic Prakritsatthat time. According to Nitti-Dolci (1938) 61-88, the gram­matical fragments and the instructions for the theatrical use of the Prakrits as weil as the Dhruvä. stanzas in the Nätya§ästra (chapters XVII and XXXII of the Baroda edition) are to be regarded as remnants of such grammars; cf., however, the critique by Renou (1938) 162f., Alsdorf (1941) col 141 (n. 4) and 143f. (= Kleine Schriften 565f.), von Hinüber (1986) § 63.

11 Cf. Salomon (1982) 125ff.

66 Sources and time

in them original Pkt. texts, exactly as they were composed in the medi­eval period (12-13th centuries), without any suspicion of scribal changes in the process of transmission. They are, in effect, autograph mss. of the texts, and show exactly what literary S[auraseni] and M[ähär~tri] werein these times ... " 12 According to Sten Konow, however, the Lalita­vigraharäjanätaka shows only insignificant deviations from Hemacandra's Sauraseni rules. He also says that "the forms of our inscription ... irrefut­ably prove here [for Mägadhi] that the rules of the grammarians were not based on nothing." 13

In our context now, however, it can be said that not only the time of the composition of texts has to be taken into consideration but also the age of the specific textual evidence which, indeed, may turn out to be of decisive importance. From this point of view one would no longer be surprised at the exceptional position of the inscribed dramas, especially if one takes into account that the majority of play manuscripts is recent. The Polish scholar Andrzej Gawronski has concerned hirnself with this question in detail. He is

certain of the fact that in their systematizing work the gram­marians paid little attention to doing justice to the relations between the living dialects .... Younger [Prakrit grammarians], and above all the important Hemacandra, were no longer in touch with living Prakrit. For them, [the Prakrits] were writ­ten languages the knowledge of which one had to acquire from manuscripts and whose popular [linguistic] basis ... had already been further developed. 14

12Salomon (1982) 125. It seems to me, however, that one has to assume at least one intermediary step of transmission from the "autograph manuscripts" to the in­scriptions; cf. Bühler (1896) 95.25-35 (§ 39). One has to note that Bühler also says that in some cases authors claim to have themselves engraved their works in stone. However, for the two plays we are concerned with here the colophons expli­citly name the engravers, who are different from the authors of the plays (Lalita­vigraharäjanätaka: mahfpatisutena pa7114itabhäskare7J.a svayam älikhyotkfr7J.7J.äni [sie.] ak§arä!J.i; Pärijätamaiijarf: rii.pakäraprakä7114asya sfhäkasyä711gajanmanä / prasastir iyam utkfr7J.7J.ä rämadevena silpinä //); cf. Kielhorn (1901) IV and 15 (= Kleine Schriften 94 and 111), Hultzsch (1905-06) 98 and 117.

13 Konow (1894) 480f.: "Die Abweichungen von Hemacandras Regeln sind somit sehr unbedeutend .... Die Formen unserer Inschrift ... beweisen hier [für die Mägadhi], daß die Regeln der Grammatiker nicht aus der Luft gegriffen sind." Also Konow (1901) 204; and even earlier Kielhorn (1893) 553 (=Kleine Schriften 337), who says about the Prakrit of these fragments that "it agrees more with the teachings of Hemacandra than is the case with any other published play" ("es mehr zu den Lehren des H!!machandra stimmt als bei irgend einem der veröffentlichten Schauspiele der Fall ist"); cf. also Pischel (1900) § 11.

14 Gawronski (1911) 26lf.: "Ich bin nun fest überzeugt, daß sich die Grammatiker wenig darum kümmerten, in ihrer systematisierenden Arbeit den Verhältnissen zwi­schen den lebenden Dialekten gerecht zu werden. . .. Die jüngeren, vor allem aber der wichtige Hemacandra, hatten keine Fühlung für lebendes Präkrit mehr. Für sie waren das Literatursprachen, deren Kenntnis man sich aus Handschriften erwerben mußte und

Play Editing and Prakrit Grammarians 67

Even if this is correct it is certainly also true, on the other hand, that, e.g., the Sauraseni of the classical dramas is not identical with the formerly spoken language of Sürasena. According to Oskar von Hinüber the transmitted Sauraseni of the dramas has been largely retranslated from Sanskrit. 15 So one might ask to what extent the poets themselves, unlike the grammarians, could have been "in touch with living Prakrit" or what Gawronski calls the "living dialects". To put it in Gawronski's own words:

The Mägadhi of the dramas ( and the grammarians) is a literary dialect , an artificiallanguage exactly to the same degree as the Sauraseni of the dramas and the grammarians ... is a literary dialect, an artificial language.16

In ordertobe able to write a literary Prakrit different from a colloquial language the playwrights themselves needed a Prakrit grammar which fur­nished them at least with the general rules. 17 And, indeed, it can be ob­served that the classical plays from the time of Kälidäsa seem to conform more and more to the grammarians. 18 Here, it must be admitted, that the specific transmission of a drama could have played a significant role. In this case our impression that Prakrit developed in growing conformity to the rules of grammar would be misleading. That Prakrit grammar­ians intentionally wrote rules for poets (kavi) to follow is clear from the introductory stanzas 14 and 22 of Laimrdharas $arf,bhä~äcandrikä (very probably composed in the middle of the sixteenth century AD). He says explicitly: 19

tena zajmidharärye'T}-a ~arjbhä~äcandrikänaghäl vidvatkavicakorä'T}-ä1"fl pritaye kriyate 'dhunä I I 14 I I

deren volkstümliche [sprachliche] Unterlagen ... sich längst weiterentwickelt hatten." 15 Cf. von Hinüber (1986) § 85; also Pischel (1900) § 6: "So the Präkrit languages are

artificial dialects in so far as they have been considerably altered for literary purposes by poets." ("Die Präkritsprachen sind also Kunstsprachen, insofern sie von den Dichtern zu Iitterarischen Zwecken erheblich verändert worden sind .")

16Gawronski (1911) 277: "die Mägadhi der Dramen und der Grammatiker ist ein literarischer Dialekt, eine Kunstsprache ... ganz in demselben Grade, wie die 'Säuraseni der Dramen und der Grammatiker ... ein literarischer Dialekt , eine Kunstsprache ist." Cf. also Schokker (1976) 151 and esp. 155: "Thus the fact , that Mähärä?~ri and Sauraseni are used for respectively the lyric and prose styles of the drama, implies that they cannot be directly connected with the spoken languages of the countries from which they are named." Schokker's premiss, however, that the text of the Karpüramaiijarz is in Sauraseni prose and Mähärä?~ri verse , is, to say the least, indeed problematic as Salomon (1982) has shown. On the use of Prakrit dialects in Sanskrit dramas, cf. also Vaidya (1952) and, not very helpfully, also Dave (1978) .

17 Cf. A.F.R. Hoernle, Indian Antiquary 2, 210: "Those who wrote Präkrit ... must have learned the literary Präkrit and must have learned it from the Präkrit grammars"; approvingly quoted by Konow (1894) 473.

18Thus von Hinüber (1986) § 59, and Bubenik (1996), p. 15; cf. , e.g., my remarks on the comparatively late La/itavigraharäjanätaka above.

19 Quoted by Nitti-Dolci (1938) 182f.

68 Sources and time

traivikrama'T(L haimacandra'T(L guror jiiätvä ca bhämahaml kavisaukhyäya tat sarvam atra sa'T(Lk§ipyate mayä I I 22 I I

ks 14. This noble La*midhara now composes the spotless 'Moon­light of the Six Languages' for the joy of the Cakora birds, [viz.], the learned poets.

22. After having learned [the grammars] of Trivikrama, Hema­candra, and Bhämaha from my teacher, I abridge all of them here for the convenience of poets.

It is not likely, however, that in the case of Prakrit there has been a one­sided relation between the poets and grammarians to the effect that only one side drew from the other. Rather one has to suppose that both mu­tually influenced each other: the poets, especially the later ones, certainly studied the grammarians who, for their part, read and assessed the literary works.20 In this context a short remark by the Jaina author Namisädhu is very instructive. In his commentary to Rudrata's Kävyäla'T(Lkära (II.12),21

which Namisädhu wrote in 1069, he concludes his brief enumeration of some 'mixtures' or 'confusions' of the different Prakrits with the following words:

ityädy anyad api sä'T(Lkarya'T(L mahäkavilak§yäd avaseyam itil vi.Se§atas tu bhä§älak§a'T}a'T(L granthäntaräd avaseyam iti

Thus, etc. Other mixtures (or: confusions) as well have tobe learned from the example of the great poets. The characterist­ics of the languages in detail, however, have to be learned from another work.

Accordingly, Prakrit sturlies in medieval times were intended to be based on the pertinent Kävya literature, i.e., on the available manuscripts, and on special grammars. The poets as well as the grammarians equally had to observe what had been sanctioned by the mere usage of accepted authors and especially by the grammatical teachings of the schools. The situation becomes even more complicated, however, through the role of copyists who had a knowledge of Prakrit grammar. We may assume that they standardized the texts they copied, as modern editors have also done. And the grammarians who read these manuscripts got the impression that their rules were confirmed. Finally, we have to note that literary Prakrit is relatively irregular when compared to Sanskrit, so that Carl Cappeller is probably right in doubting that "for Prakrit, poets had as definite rules in mind as for Sanskrit" .22 The dearth of available material, however, which

20Cf. also Winternitz (1922) 403-406, esp. 404. 21 Kävyamälä ed ., p . 15 , I. 14-15. 22 Cappeller (1909a) XVII : "Ich glaube nicht, daß den Dichtern für das Prakrit so

bestimmte Regeln vorschwebten wie für das Sanskrit ... "

Play Editing and Prakrit Grammarians 69

comes up to our critical standards allows us to pass only a preliminary judgement.

As for the grammarians, Gawronski points to their "uncriticalness and unreliability" and says that "as can be shown, they taught incorrect, in fact meaningless, things." 23 Writing on Hemacandra Pischel remarks that he lacks "every trace of scientific criticism and every familiarity with lit­erature" , and even doubts whether Hemacandra "himself ever read Häla and the Setubandha" .24

Against Pischel's argument concerning the unanimity among all the Prakrit grammarians "on essentials" Gawronski rightly says that "the grammarians largely took over examples as well as rules from their prede­cessors. It is therefore not surprising that they are unanimous on essen­tials ."25

Gawronski draws from this the following conclusion:

The question whether all the rules of the grammarians are to be applied to the manuscripts, has decidedly to be answered in the negative. The grammarians make up an important source of our knowledge, however they are not above but next to the manuscripts. They [the grammariansJ have to be always con­sulted for the sake of comparison, but may never be looked upon as containing a priori the truth. And if the rules of the grammarians do not concur with our rules deduced from the manuscripts, but through their impreciseness testify that the practice of the manuscripts was exactly as inconsistent at the times of the medieval grammarians as it is today, then we def­initely have to give preference to our rules to the same degree as one gives preference to a critical approach rather than to an uncritical one.26

23 Gawronski (1911) 279 and 280 (on the authority of Bloch (1893] 28f.) : "(Oben] wurde mehrmals die Kritiklosigkeit und Unzuverlässigkeit der einheimischen Präkritgrammatiker betont. .. . Es muß aber betont werden , .. . daß sie nachweislich Falsches, ja Sinnloses, lehrten ... "

24 Pischel (1880) VI : "Ihm (Hemacandra] fehlte jede spur wissenschaftlicher kritik und jede belesenheit in der Iiteratur. Nach meinen erfahrungen bei bearbeitung der De<;fnämamälä zweifle ich sogar, dass er je selbst den Häla und Setubandha gelesen hat ... " Repeated in Pischel (1900) § 42 (p. 46, I. 20-21) ; for Winternitz (1922) 406 "these accusations are only partially justified" ( "sind diese VorwUrfe nur zum Teil berechtigt") .

25 Gawronski (1911) 280: "wir haben ja allen Grund, zu vermuten ... daß sie [sc. die Grammatiker] zum großen Teil 'Beispiele wie die Regeln von ihren Vorgängern' herUbernahmen . Dann ist es aber kein Wunder, wenn sie in der Hauptsache einig sind."

26 Gawronski (1911) 280f.: "Die Frage also, ob alle Regeln der Grammatiker gegen die Handschriften durchzufUhren sind, ist entschieden zu verneinen. Die Grammatiker bilden eine wichtige Quelle unserer Kenntnisse, aber sie stehen nicht Uber, sondern neben den Handschriften. Sie müssen immer zum Vergleich herangezogen werden, dür­fen aber nie als dasaprioririchtige enthaltend angesehen werden. Und wenn die Regeln

70 Sources and time

To my mind, one has to subscribe to this view as a general principle, which, however, may not be claimed to be a universal rule. The following two supplementary considerations should be taken into account. Firstly, the implied premiss according to which each poet did in fact write (or dic­tate) Prakrit in an entirely precise and consistent manner, is a problematic one. Was there ever a completely regular Prakrit and, especially, a consist­ent orthography? Is it not imaginable that the Prakrit speaking persans of a drama occasionally used different grammatical or phonetic forms of one and the same word? Secondly, certain forms used by poets in indi­vidual cases could in fact be traced back to the grammarians. Therefore, as von Hinüber has stated, "one has to search forapassable middle course between the manuscripts and the grammarians" .27 Instead offollowing an abstract universal principle one should study each work separately. Thus the time of the composition, the kind of poetry, the situation of the specific textual tradition and the peculiarities of the individual author have to be taken into consideration. Another important question would be: What do we know about the Prakrit of comparable works? Moreover, the erner­ging quality of the individual items of textual evidence assessedas a whole is of special significance here. A completely disordered textual tradition certainly needs external correction by the standard of better transmitted texts and the rules of the grammarians. Good manuscripts, on the other hand, will perhaps furnish us with principles derived internally for the constitution of the Prakrit passages.

Besides, the works of the grammarians themselves are in part very unreliably transmitted. For a critical use of their grammars it is indis­pensable to be aware of the normative character and the specific date of the rules, which, of course, need not be-and actually are not- constantly in accord with the individual usage of the poets. Cappeller is of the opin­ion that one should strive for utmost consistency and always establish only one form of a Prakrit word, namely the one which has been testified bet­ter throughout the whole text, irrespective of the variant readings in the specific passage.28 As agairrst this view, which has been shared, among others, by Hillebrandt, one should take Gawronski's advice and "accept a freer usage, doublets and the like in some cases" as weil as also to take

der Grammatiker mit unseren, aus den Handschriften deduzierten Regeln nicht Ober­einstimmen, wohl aber durch ihre Ungenauigkeit Zeugnis davon ablegen, daß die Praxis der MSS. schon zur Zeit jener mittelalterlichen Grammatiker ebenso inkonsequent war, wie sie heute ist, dann mussenwir unbedingt unseren Regeln eben in dem Grade Vorzug geben, wie man einem kritischen Verfahren vor einem kritiklosen Vorzug gibt."

27 von HinOber (1986) § 59: "Hält man sich die ungemein verwilderte Überlieferung des Pkt. in den Dramen-Handschriften vor Augen, so wird man versucht sein, zwischen den Handschriften und den Grammatikern einen gangbaren Mittelweg zu suchen."

28 Cf. Cappeller (1909b) 414: "Was das Prä k r i t betrifft, so habe ich wiederum mög­lichste Konsequenz angestrebt, so daß selbst da, wo die Grammatiker mehrere Wort­formen gestatten, immer nur eine und zwar die im Laufe des ganzen Textes besser bezeugte, ohne Rucksicht auf die Lesarten an der betreffenden Stelle, durchgefUhrt ist ."

Play Editing and Prakrit Grammarians 71

into consideration "euphony" 29 instead of "following a certain rule in every case" .30 In the case of doublets we have to consider the sentence accent, intonation, the position within the sentence, and especially, emphasis as possible reasons. 31

On the other hand, we have to assume that a great number of phenom­ena which have been observed by the grammarians in the texts, i.e., in correct and faulty manuscripts, have probably been incorporated into the grammatical works. And these rules for their part could have influenced the later poets.

A conclusive decision as to whether a peculiar form goes back to the author or a copyist, is difficult to make and probably never final. Some criteria are the situation of the tradition which depends on a general assessment of the textual evidence and recensions, the total result of a text-critical examination, the distribution of variant forms in the best manuscript, etc. Other criteria are the special character of the specific Prakrit which has to be determined first, and possible reasons for vari­ations in the Prakrit thus defined. Thereby one has to provide for the fact that a single word can be transmitted at various places in different, but nevertheless authentic forms .

Now it is remarkable that the oldest available manuscript ofHar~adeva's play Nägänanda, (seventh century AD) which is dated AD 1155 (or 1156) ,32

has transmitted some Prakrit forms which we would in fact not expect there according to Pischel's grammar and the respective editions.33 With regard to the plays of the classical period, however, it is necessary to con­sider with Lionel D. Barnett that "we have no first-hand knowledge of these Prakrits; we know only the forms that editors and scribes have im­posed upon the texts ." 34 Therefore, it is not possible simply to assume, e.g., that forms which cannot be found in Kalidasa manuscripts have ac­tually not been used by Kalidasa. Each new textual source can change the standard by which one makes one's judgement. One of the most import­ant criteria for future critical revisions of play editions is the comparative reading of accurately (or at least more accurately) transmitted texts of the

29 Cf. also Hemacandra's auto-commentary ad sütra !.231 of his Prakrit grammar where he mentions §rutisukham as a relevant criterion: etena pakärasya präptayor /opavakärayor yasmin krte §rutisukham utpadyate sa tatra käryal}. j "On that account, because [both] elision as weil as [the change to] va in case of pa is correct, one should perform that [phonetic Operation] on the performance of which pleasure of ears arises."

3°Cf. Gawronski (1911) 282: "... man wird wohl im allgemeinen besser tun, in manchen Fällen freieren Sprachgebrauch, Dubletten und ähnliches anzuerkennen, als in jedem Fall eine bestimmte Regel zu befolgen .... Wir müssen u.a. den Wo h I k I an g als einen viel gewichtigeren Faktor zulassen ... " See also Alsdorf (1933) 29 ( = Kleine Schriften 510) .

31 Cf. Berger (1955) 20, n . 17. 32 Cf. Hahn (1981) 131-137. 33 For details, see Steiner (1997), chapter 9. 34Barnett (1921) 588.

72 Sources and time

same literary genre as weil as texts and manuscripts of the same age and place of origin.

Finally, we have to take the particular character of the dramatic Prakrits into account. They were, as already stated, artificial languages for the poets. In order to use them correctly the authors were dependent on the transmitted plays and the Prakrit grammars. That presumably has led the authors to a relatively conservative treatment of the literary Prakrits. The copyists, for their part, may rather have been influenced by the phonetic development of the spoken languages. Thus the specific age of the play manuscripts, the grammars, and the manuscripts of the grammars is of basic consequence for the reconstruction of the genuine literary Prakrit. Consequently, this means that for the time being we have to desist from mutually comparing the Prakrit of individual authors for the purpose of more general questions, e.g., chronological considerations, and that, in­stead, we should direct our attention only to specific works in order to prepare critical editions based on the oldest available evidence.

REFERENCES

Alsdorf, Ludwig

1933 "Bemerkungen zu Pischels 'Materialien zur Kenntnis des Apabhrarp.~a" ', in: Festschrift Moriz Winternitz. 1863 - 23. Dezember - 1933. Hrsg. von Otto Stein und Wilhelm Gam­pert. Leipzig, pp. 29-36 ( = Kleine Schriften. Hrsg. von Al­brecht Wezler. Wiesbaden 1974. [Glasenapp-Stiftung. 10.], pp. 510-517).

1941 "Zur Kenntnis der Prakrit-Grammatiker", in Orientalistische Literaturzeitung 34, col137-145 ( = Kleine Schriften, pp. 563-567).

Barnett, L[ionel] D.

1921 "Bhasa" (sie.), in: Journal of the Royal Asiatic Society, pp. 587-589.

Berger, Hermann

1955 Zwei Probleme der mittelindischen Lautlehre. München. (Münchener Indologische Studien. Heft 1.)

Bloch, Theodor

1893 Vararuci und Hemacandra. Ein Beitrag zur Kritik und Geschichte der Prakrit-Grammatik. Leipziger Dissertation. Gütersloh.

Play Editing and Prakrit Grammarians 73

Bubenik, Vit

1996 The Structure and Development of Middle Indo-Aryan Dia­lects. Delhi.

Bühler, Georg

1896 Indische Palaeographie. Von circa 350 a. Chr.- circa 1300 p. Chr. Mit 17 Tafeln in Mappe. Straßburg. (Grundriß der indo-arischen Philologie und Altertumskunde. I.11.)

Cappeller, Carl

1909a Kälidäsa 's Sakuntalä (kürzere Textform). Mit kritischen und erklärenden Anmerkungen hrsg. von C. C. Leipzig.

1909b Ratnävalf. [Ed. by C. C. in:J Otto Böhtlingk's Sanskrit-Chres­tomathie. 3. verbesserte und vermehrte Aufl. Hrsg. von Richard Garbe. Leipzig 1909, pp. 326-369. ["Präkrit-Index", pp. 370-382; "Anmerkungen", pp. 413-416.)

Coulson, Michael

1989 A Critical Edition of the Mälatfmädhava. By M. C. Revised by Roderick Sinclair. Delhi.

Das Gupta, Sibani

1962 The Ca'T}~a-Kausika of Ärya K~em[svara. Ed. by S. D. G.

Dave, T.N.

With Introduction, full Critical Apparatus of Manuscripts, English Translation and Indices. Calcutta.

1978 "Role of Prakrit Dialects In Sanskrit Dramas", in: Proceed­ings of the Seminar on Prakrit Studies (1973). Ed. by K. R. Chandra. Ahmedabad. (L. D. Series. 70.), pp. 101-108.

Gawronski, Andrzej

1911 "Am Rande des M;-cchakatika", in: (Kuhn's) Zeitschrift für vergleichende Sprachforschung 44, pp. 224-284.

Hahn, Michael

1981 "Zu einer Neuedition von Har~adevas Nägänanda", in: Wie­ner Zeitschrift für die Kunde Südasiens 25, pp.131-137. [An English version of this paper has appeared in the Journal of the Nepal Research Centre V /VI (1981/82), pp. 125-13l.J

Hemacandra

H emacandra 's Grammatik der Prakritsprachen (Siddhahema­candram Adhyaya VIII). Mit kritischen und erläuternden Anmerkungen. Hrsg. von Richard Pischel. I. Theil. Text und Wortverzeichnis. Halle 1877 [Neudruck: Osnabrück 1969.)

74 Sources and time

Hertel , Johannes 1924 Mu'!uJaka- Upani§ad. Kritische Ausgabe mit Radarneudruck

der Erstausgabe (Text und Kommentare) und Einleitung. Hrsg. von J. H. Leipzig.

Hillebrandt , Alfred

1912 Mudräräk§asa by Visakhadatta. Ed. from MSS. and provid­ed with an Index of all Prakrit words by A. H. Part I: Text. Part II: Index. Breslau.

Hinüber , Oskar von

1986 Das ältere Mittelindisch im Überblick. Wien. (Österreichi­sche Akademie der Wissenschaften. 467. Veröffentlichungen der Kommission für Sprachen und Kulturen Südasiens. 20.)

Hultzsch , Eugen

1905-06 "Dhar prasasti of Arjunavarman: Parijatamanjari-natika by Madana", in: Epigraphia Indica 8, No. 9, pp. 96-122.

Jackson, A. V. Williams / Ogden, Charles J.

1923 Priyadarsikä. A Sanskrit Drama by Harsha. Transl. into English by G[ushtaspshah] K[aikhushro] Nariman, A[braham] V[alentine] Williams Jackson and Charles J[ones] Ogden with an introduction and notes by the two latter together with the text in transliteration. New York [Repr.: 1965] . (Columbia University Indo-Iranian Series. 10.)

Kielhorn , Franz

1893 "Bruchstücke des Lalita-Vigraharäja Nätaka", in: Göttinger Nachrichten, pp. 552-570 ( = Kleine Schriften. Mit einer Auswahl der epigraphischen Aufsätze. Hrsg. von Wilhelm Rau. Teil 1.2. Wiesbaden 1969. [Glasenapp-Stiftung. 3, 1.2.], pp. 336-354) .

1901 Bruchstücke indischer Schauspiele in Inschriften zu Ajmere.

Konow, Sten

Mit vier Tafeln. Festschrift zur Feier des hundertfünfzigjähri­gen Bestehens der Königl. Gesellschaft der Wissenschaften zu Göttingen. Abhandlungen der phil.-hist . Klasse. Berlin ( = Kleine Schriften, pp. 93-126).

1894 [Review:] "Bloch, Theodor, Vararuci und Hemacandra. Ein Beitrag zur Kritik und Geschichte der Prakrit-Grammatik. Leipziger Dissertation. Gütersloh 1893", in: Göttingisehe gelehrte Anzeigen 6, pp. 472-482.

1901 Räja-9ekhara 's Karpüra-Manjarf. A drama by the Indian poet Räja9ekhara (about 900 A .D.) . Critically ed. in the original Prakrit, with a glossarial index, and an essay on the

Play Editing and Prakrit Grammarians 75

life and writings of the poet by St. K. and transl. into English with notes by Charles RockweH Lanman. Cambridge, Mass. (Harvard Oriental Series. 4.)

Lüders, Heinrich

1911 Bruchstücke buddhistischer Dramen. Berlin. (Kleinere Sans­krittexte. Heft 1.) [Repr.: Wiesbaden 1979 (Monographien zur indischen Archäologie, Kunst und Philologie. 1.)]

Namisädhu

Nätya.Sästra

The K11vyt1lankara (a Treatise on Rhetoric) of Rudrata. With the Commentary of Namist1dhU (sie.). Ed. by Pa!fdita Dur­gäprasäda and Käs'hinätha Pä!fduranga Paraba. Bombay 1886. (Kävyamälä. 2.)

Nätyasästra (of Bharatamuni}. With the commentary (Abhi­navabhäratf) of Abhinavagupta(äcärya). Ed. [ ... ] by Mana­valli Ramakrishna Kavi (and J. S. Pade). Vol. 1-4. Baroda 1926; 1934; 1954; 1964. (Gaekwad's Oriental Series. 36. 68. 124. 145.) VoLl: Revised and critically ed. with Preface and Appendices by K. S. Ramaswami Sastri. Second Ed. Baroda 1956. (Gaekwad's Oriental Series. 36.)

Nitti-Dolci, Luigia

1938 Les grammairiens prakrits. Paris. [English translation by Prabhäkara Jhä. Delhi 1972.]

Pischel, Richard

1880 Hemacandra's Grammatik der Prakritsprachen (Siddhahema­candram Adhyäya VIII). Hrsg.(,) übersetzt und erläutert von R. P. II. Teil: Übersetzung und Erläuterungen. Halle. [Repr.: Osnabrück 1969.

1900 Grammatik der Prakrit-Sprachen. Straßburg. (Grundriß der indo-arischen Philologie und Altertumskunde. 1.8.) [Repr.:

Renou, Louis

Tokyo 1977; English translation by Subhadra Jhä. 2nd rev. ed. Delhi 1981.]

1938 [Review:] "Luigia NITTI-DOLCI. Les grammairiens prakrits. -Paris ... 1938", in: Journal Asiatique 230, pp. 159-168.

Salomon, Richard

1982 "The Original Language of the Karpüra-maiijarf', in: Zeit­schrift der deutschen morgenländischen Gesellschaft 132, pp. 119-141.

76 Sources and time

Schokker, Godard H[endrik]

1976 "The Prakrits of the Drama: their literary function as illus­trated by the Karpüramaiijarl'', in: Sambodhi 5, Nos. 2-3, pp. 148-165.

Stein er, Roland

1997 Untersuchungen zu Har~adevas Nägänanda und zum indi­schen Schauspiel. Swisttal-Odendorf. (Indica et Tibetica. 31.)

Tieken, Herman

1983

Vaidya, P. L.

1952

Häla's Sattasaf. Stemma and edition (Gathäs 1-50), with transl. and notes. Proefschrift [ ... ] door Herman Joseph Hugo Tieken. Leiden.

"On the Use of Prakrit Dialects in Sanskrit Dramas", in: Annals of the Bhandarkar Griental Research Institute 33, pp.15-25.

Winternitz, Moriz

1922 Geschichte der indischen Litteratur. 3. Bd. Leipzig. Die Kunstdichtung- Die wissenschaftliche Litteratur- Neuindi­sche Litteratur - Nachträge zu allen drei Bänden. [Repr.: Stuttgart 1968.; English Translation by Subhadra Jha. Vol. 3, Pt. 1, 2, Delhi 1963, 1967.]