The Sanskrit drama

The discovery of fragments of manuscripts on palm-leaf, of great antiquity, at Turfan, has through the energy of Professor Lüders revealed to us the existence of at least three Buddhist dramas. Of one of these the authorship is happily certain, for the colophon of the last act has been preserved, and it records that the drama was the Çāriputraprakaraṇa of Açvaghoṣa, son of Suvarṇākṣī; it gives also the fuller title Çāradvatīputraprakaraṇa and the number of acts as nine.

Açvaghoṣa is an author whose fame, thanks to his error in being a Buddhist long lost in India, has recently attained renewal by the discovery and publication of his Buddhacarita, a court epic in excellent style and spirit on the life of the Buddha. His Sūtrālaṁkāra is also known through the medium of a Tibetan translation, and illustrates his ability in turning the tale into an instrument for propaganda in support of the Buddhist faith. If the tradition which ascribes to him the Mahāyānaçraddhotpāda is correct, he was also the founder or expounder of a subtle system of metaphysics akin to the Vijñānavāda of the Mahāyāna school, and the Vajrasūcī seems to preserve in some measure the record of his onslaught on the caste system, which exalted the Brahmins at the expense of the Kṣatriyas, and condemned Buddhism on the score that it was unfitting that a Kṣatriya like the Buddha should give instructions to Brahmins. Certainly genuine is the Saundarananda, in the epic manner, which like all his works is devoted to the effective exposition of Buddhism in the language of polite literature, and also of the Brahmin schools. We recognize in him one who appreciated that it would never do to allow Buddhism to remain buried in a form inferior to the best that [81]Brahminism could produce, and it is curious that fate should have preserved the work of the rival of the Brahmins, while it has permitted his models to disappear. That he had abundant precedent to guide him is clear from the classical form already assumed by his dramas; the argument of Professor Konow1 to the contrary, on the ground that many of the standing formulae and characters are derived from the popular drama, and show that the artistic drama had not developed yet full independence, is unintelligible, since these features persist throughout the history of the Sanskrit drama. Nor does any weight attach to the argument that the Nāṭyaçāstra, assumed to be of about the same period as Açvaghoṣa, shows knowledge of only a limited variety of dramas. On the contrary it is amazing how much literature must have preceded to permit of the setting up of the main types of drama, some of which were evidently represented by many specimens, though others doubtless rested on a small basis of practice.

The brief fragments preserved of the drama of Açvaghoṣa give us the certainty of his authorship if any doubt could exist after the colophon, for one verse is taken bodily from the Buddhacarita, just as he twice refers in the Sūtrālaṁkāra to that important work. The story of the play is clear; it deals with the events which led up to the conversion of the young Maudgalyāyana and Çāriputra by the Buddha, and some of the incidents are certain. Çāriputra had an interview with Açvajit; then he discussed the question of the claims of the Buddha to be a teacher with his friend, the Vidūṣaka, who raised the objection that a Brahmin like his master should not accept the teaching of a Kṣatriya; Çāriputra repels the objection by reminding his friend that medicine aids the sick though given by one of inferior caste, as does water one aheat. Maudgalyāyana greets Çāriputra, inquiring of him the cause of his glad appearance, and learns his reasons. The two go to the Buddha, who receives them, and who foretells to them that they will be the highest in knowledge and magic power of his disciples. [82]In this point there is a deliberate and certainly artistic deviation from the ordinary version of the incident, followed in the Buddhacarita, in which the prophecy of the Buddha is addressed, not to the disciples themselves, but to others of the Buddha’s followers. The end of the play is marked by a philosophic dialogue between Çāriputra and the Buddha, which includes a polemic against the belief in the existence of a permanent self; it terminates in a praise of his two new disciples by the Buddha, and a formal benediction.

The most remarkable thing regarding this drama is its close correspondence to the classical type as laid down in the Nāṭyaçāstra. The piece is a Prakaraṇa, and it has nine acts, which accords perfectly with the rule of the Çāstra; the Mṛcchakaṭikā and Mālatīmādhava have ten apiece; the Acts bear no titles, but this is in accord with the normal usage, though the Mṛcchakaṭikā gives names. The hero is Çāriputra, who corresponds to the Brahmin hero of the Çāstra, and who is emphatically of the noble and calm type enjoined by that authority. Whether the heroine was a lady or a hetaera we do not know, nor does it appear how far the poet altered the subject-matter by invention, which is normally the case with later Prakaraṇas. The Buddha and his disciples, including, beside the two heroes, Kauṇḍinya and a Çramaṇa speak Sanskrit, and use both prose and verse; the Vidūṣaka speaks Prākrit. The presence of this figure is a remarkable proof of the fixed character attained by the drama, for in itself there is nothing more absurd than that a youthful ascetic seeking after truth should be encumbered by one who is a meet attendant on a wealthy merchant, Brahmin, or minister. It can, therefore, only be supposed that Açvaghoṣa was writing a type of drama in which the rôle was far too firmly embedded to permit its omission, and presumably in the story of the drama now lost to us the Vidūṣaka served to introduce comic relief. With natural good taste, he disappears from the last Act, where Çāriputra has no need as a member of the Buddha’s fraternity for encumbrances like a jester.

In one point only has it been claimed to find a clear discrepancy between Açvaghoṣa’s practice and that of the later drama. At the close the theory2 requires that the question, ‘Is [83]there anything further that you desire (ataḥ param api priyam asti)?’ be addressed to the hero by himself or another, to which he replies by uttering a benediction styled the Bharatavākya. In the drama of Açvaghoṣa the phrase is omitted, and the benediction proceeds, without prelude, with the words, ‘From now on shall these two ever increase their knowledge, restraining their senses, to gain release’, spoken by the Buddha, not by the hero. Lüders concludes hence that the regular form of close was not yet established by Açvaghoṣa’s time. The conclusion is clearly fallacious, and rests on a failure to recognize in this the readiness of Açvaghoṣa to give effect to a traditional usage, while not slavishly following it. It would obviously have been absurd to place the last words in the drama in the form of a benediction in the mouth of any one save the Buddha, and therefore he speaks the benediction. To preface it with the usual formula was needless in his case, but the opening words of the verse are ataḥ param, which is obviously not an incredible coincidence, but a deliberate reference to the ordinary phrase. Açvaghoṣa shows thus his knowledge of the rule and his power to vary it in case of need. Similarly Bhaṭṭa Nārāyaṇa in the Veṇīsaṁhāra puts the Bharatavākya in the mouth of Yudhiṣṭhira, but he makes Kṛṣṇa end the play by according the favour prayed for by Yudhiṣṭhira. He too felt that it would be absurd to leave the omnipotent one in the position of listening without response to the utterance of a benediction by one who cannot be more than an inferior, though nominally the hero.3

The same manuscript which contains portions of the Çāriputraprakaraṇa has also fragments of two other dramas. There is no evidence of their authorship, other than the fact that they appear in the same manuscript as the work of Açvaghoṣa, and that they display the same general appearance as the work of that writer. That they are Açvaghoṣa’s is much more probable than that they are the work of some unknown contemporary.4 [84]

The first of these is specially interesting as it represents a type of which we have otherwise no earlier specimen than the Prabodhacandrodaya of Kṛṣṇamiçra. We find the allegorical figures of Buddhi, wisdom, Kīrti, fame, and Dhṛti, firmness, appearing and conversing. This is followed by the advent of the Buddha himself, adorned with the halo which he borrowed from Greek art. We do not know whether he appeared later in actual conversation with the allegorical figures, but for this mixture of the real and the ideal we have to go beyond Kṛṣṇamiçra, who represents all his characters as abstract, Viṣṇu for instance by Faith in Viṣṇu, to Kavikarṇapūra’s glorification of Caitanya in the sixteenth century, in which allegorical figures are mingled with Caitanya and his followers, though they do not actually converse together.5 It must remain uncertain whether there was a train of tradition leading from Açvaghoṣa to Kṛṣṇamiçra, or whether the latter created the type of drama afresh; the former theory is the more likely. The characters all speak Sanskrit, but the fragments are too short to give us any real information on the general trend of the play.

The other drama gives us more interesting matter. It is one in which figures a hetaera named Magadhavatī, a Vidūṣaka named Komudhagandha, a hero styled only Nāyaka, but probably named Somadatta, a Duṣṭa, rogue, without further name, a certain Dhānaṁjaya, who may possibly be a prince if the term ‘king’s son’ (bhaṭṭidālaka), which is recognized in the Nāṭyaçāstra as the style of the younger princes of the blood, applies to him, a maid-servant, and Çāriputra and Maudgalyāyana. The drama was doubtless intended for purposes of religious edification, but what we have is too fragmentary to do more than show that the author was possessed of humour and that the Vidūṣaka was already a hungry soul. The drama alludes to an old garden as the place where part of the action passed, as in the Mṛcchakaṭikā, and also as in that drama the house of the hetaera served as the scene of another part of the action. The characters are often introduced as entering in vehicles (pravahaṇa), a further point of [85]similarity to that drama, while an allusion to a Samāja or festival on a hill-top accords with the frequent reference to such amusements in Buddhist literature. An obscure character is a person, obviously of lower rank, who is styled Gobaṁ°.

The drama shows close agreement with the classical model; the name of the Vidūṣaka is evidence of this, for not only is it connected with a real Brahmin family, but it obeys the rule that the name of that character should indicate a flower, the spring, &c., for it means literally ‘the offspring of the lotus-smelling’. The name of the hetaera does not observe the rule exemplified in the Cārudatta that the hetaera’s name should end in senā, siddhā, or dattā, but, apart from the fact that the authority for the rule is very late, the name was very probably given to the poet by the literary tradition. The fact that the Duṣṭa and the Nāyaka appear by these titles only has a parallel in the Cārudatta and the Buddhist drama of Harṣa, the Nāgānanda, but it is difficult to say whether or not this is a sign of antiquity.

The material available in the case of any of the three dramas is too scanty to give us any assurance as to what the practice was regarding the introduction, especially the use of the Nāndī, or verse of benediction. What is certain is that the Pāripārçvika, or assistant of the Sūtradhāra in the later literature, is found apparently as taking part in the opening of the drama, perhaps the Çāriputraprakaraṇa.

In accordance with the later rules we find the Buddha, his disciples, the hero of the hetaera play and Dhānaṁjaya speaking Sanskrit; the same is true of the allegorical characters, and this is also in accord with later practice, for in both Kṛṣṇamiçra and Kavikarṇapūra’s works some of the allegorical characters speak Sanskrit, though others, of more feminine appeal and character, speak Prākrit. One Çramaṇa speaks Sanskrit, another—conceivably an Ājīvika—a Prākrit.

The Sanskrit contains some errors, which are obvious Prākritisms, and which it would be unjust to attribute to the author, or authors. Genuine departures from the norm are scanty; the use of ārttha for artha has a precise parallel in the nearly contemporaneous dialect of Mathurā; tuṣṇīm is frequent in [86]Buddhist Sanskrit as well as etymologically correct; krimi is found also in the Buddhacarita where the reading kṛmi would spoil the metre; pratīgṛhīta has many Sanskrit parallels. In pradveṣam where the metre requires pradoṣam Buddhist influence is doubtless present, but yeva and tāva are probably merely errors of the scribe, to whom may be assigned such a monstrosity as paçyemas and Somadattassa. But bhagavāṁ has the support of the practice of the Mahāvastu where stems in mat and vat end thus, and it explains the Sandhi çṛṇvam puṣpā. These are minimal variants; in the main the Sanskrit is excellent and the fragments shows traces of the able versification and style of Açvaghoṣa.

The other characters speak Prākrit, and, by a curious variation from the normal practice, the stage directions, which are freely given as in the classical drama, are normally expressed in the language which the character concerned uses, though there are cases of mixture and apparent confusion which may be due to the scribe. Three different forms of Prākrit may be distinguished, the first spoken by the Duṣṭa, the second by the mysterious Gobaṁ°, and the third by the hetaera and Vidūṣaka.

The Duṣṭa’s speech in three important points is similar to the Māgadhī of the Prākrit grammarians; it substitutes l for r; reduces all three sibilants to ç; and has e in the nominative singular of masculine nouns in a. But it ignores the rules of the grammarians in certain matters; hard letters are not softened (e.g. bhoti), nor soft consonants elided (e.g. komudagandha), when intervocalic. There is no tendency to cerebralize n, and in kālanā the dental replaces the cerebral. Fuller forms of consonants remain in han̄gho (haṅho) and bambhaṇa (bamhaṇa). The later forms of development of consonantal combinations are unknown; thus for rj we have jj, not yy, as in ajja; cch remains in lieu of becoming çc; kṣ becomes kkh, not sk or ẖk; ṣṭ and ṣṭh give ṭṭh, not sṭ. In kiçça we have an older form than kīça, in ahakaṁ than ahake, hake, hage. In practically all these details we must see an earlier stage of what becomes Māgadhī in the grammarians. With it may be compared the metrical inscription of the Jogīmārā cave on the Rāmgarh hill which belongs to the period of Açoka.

The Prākrit of the Gobaṁ° agrees with this Old Māgadhī in having l for r and e in the nominative singular, but it reduces all [87]sibilants to s. It thus shows a certain similarity to the Ardha-Māgadhī of the grammarians, but that dialect often keeps r though it frequently alters it to l; for instance it has r for the kaleti of this Prākrit and the Old Māgadhī. Other points of similarity are the retention of the dental for cerebral in vanna; the lengthening of the vowel before the suffix ka (vannīkāhi); the accusative plural neuter in pupphā; and the infinitive bhuṁjitaye (bhuñjittae). There are points of difference, but they are probably all cases of earlier forms. Thus, as in Old Māgadhī, we have no softening or loss of intervocalic consonants; n is not cerebralised, but even introduced in palinata; ḷ appears in lieu of l; the instrumental in āhi has no nasal; the nominative of vat stems appears as in vā, as against vaṁ or vante; in the infinitive we find no doubling of the consonant in taye. The fact, however, of the regular change of r to l and the use of the form yeva after a long vowel as in Māgadhī and Pāli show that the Old Ardha-Māgadhī was more akin to Māgadhī than the later Ardha-Māgadhī, which came steadily under the influence of the western dialects as shown by the tendency to change e of the nominative to o.

There are strong points of similarity between this Old Ardha-Māgadhī and the language of Açoka’s pillar inscriptions. They agree as regards the use of l, s, and e, the dentals in palinata and vannīkāhi, yeva after long vowels, and the long vowel before the suffix ka. They disagree in the nominative and accusative plural neuter of a stems, which have āni in the inscriptions as against ā, but that is of no great importance, as these are doublets. The infinitive, however, is in tave, which cannot be equated with taye; Ardha-Māgadhī ttae may be from either.

The Açokan dialect is doubtless the court speech of his kingdom, and a descendant of the Ardha-Māgadhī of Mahāvīra, the founder of the Jain religion, and probably also of the Buddha, whose speech was clearly not akin to the Māgadhī of the grammarians, though it is called Māgadhī in the sacred texts.6

The theory of the Nāṭyaçāstra assigns Ardha-Māgadhī as the language of savants, sons of kings or Rājputs, and Çreṣṭhins, rich merchants, but, with the exception of Bhāsa’s Karṇabhāra, it does not appear in the extant dramas. Māgadhī, on the contrary, [88]is required in the case of men who live in the women’s apartments, diggers of underground passages, keepers of drink shops, watchers, the hero himself in time of danger, and the Çakāra. Into which category the Duṣṭa falls is not certain; the Daçarūpa ascribes this Prākrit to low people in general.

Çaurasenī is ascribed to the hetaera by the Çāstra which gives Prācyā or eastern dialect to the Vidūṣaka, but it is clear that the Prācyā is a mere variety of Çaurasenī, from which it differs only in the use of certain expressions. This is borne out by the dramas, in which there is no real distinction between the speech of these two characters. With the Çaurasenī of the grammarians it shows remarkable parallels. It has r in lieu of changing it to l; it reduces the sibilants to s; and for the nominative masculine it has o. Further, it changes kṣ into kkh, not cch; for chard it has chaḍḍ, for mard, madd; for saçrīkam irregularly sassirīkaṁ with the double s despite the epenthetic vowel; and in the third singular future issiti. The gerund kariya is parallel to karia in Hemacandra’s grammar; bhaṭṭā is the vocative of bhartṛ; iyaṁ is feminine as later iaṁ in Çaurasenī alone; bhavāṁ as nominative is comparable with bhavaṁ; bhaṇ is conjugated in the ninth class; viya is parallel to via for iva; and dāni with loss of i as a particle is similar to dāṇiṁ.

In other cases the forms of this Prākrit are clearly older than those of the grammarians’ Çaurasenī. As in the other Prākrits of the drama, there is no softening or omission of intervocalic consonants, and no cerebralization of n. Further, initial y is kept, not reduced to j; the interjection ai in lieu of aï is supported by the language of the Girnār and Udayagiri inscriptions; in nirussāsam we have an older form than ūsasida of Çaurasenī; jñ and ny give ññ, not the later ṇṇ; dy gives yy (written y) for jj; tuvaṁ and tava are both manifestly older than the forms tumaṁ and tuha, while karotha is a remarkable example of the preservation of the old strong base. Old also is the preservation of the long vowel in bhavāṁ. In adaṇḍāraho and the dubious arhessi we have two variants on the rule of Çaurasenī, which has i as the epenthetic vowel in arh, but this merely illustrates the uncertainty of these epentheses; duguṇa in lieu of diuṇa is not older, but a variant mode of treating dviguṇa, and there is no special difficulty in holding that dāṇi and idāṇi are forms which [89]were originally doublets of dāṇiṁ and idāṇiṁ in Çaurasenī, and later were superseded. From other Prākrit passages, presumably in the same Old Çaurasenī, we obtain old forms like vayaṁ, we, and tumhākaṁ in lieu of tumhāṇaṁ; edisa for erisa or īdisa; dissati for dīsadi; gahītaṁ for gahidaṁ; khu is kept after short vowels in lieu of being doubled; a long vowel is kept before tti and such forms as mhi. The future in gamissāma is probably old, while nikkhanta and bambhaṇa admit of this explanation against the later nikkanta and bamhaṇa.

In the words of the hetaera the word surada occurs, with softening of t to d; conceivably the passage might be verse, but in all probability we are merely faced with a sporadic instance of a change which later set in, due perhaps to a copyist’s error; to find in it an evidence of Māhārāṣṭrī would be unwise, especially as the very next word (vimadda) is not in the Māhārāṣṭrī form (vimaḍḍa). In the dialect of the Duṣṭa we have a form makkaṭaho which may be genitive, as in Apabhraṅça, but is not allowed in Māgadhī; but the sense is too uncertain to permit of any security.

The existence and literary use of these Prākrits is most interesting in the history both of the language and the literature, for they present archaic features which place them on the same plane of change as Pāli and the dialects of the older inscriptions. They may be set beside the inscriptions in the Sītābengā and Jogīmārā caves on the Rāmgarh hill, which both show lyric strophes. The influence of the Kāvya style in Sanskrit can be traced obviously in the later Nāsik inscription in Prākrit of the second century A.D., and even in the inscription of Khāravela of Kalin̄ga perhaps in the second century B.C.7 We cannot, therefore, see any plausibility in the idea of the gradual adaptation of Sanskrit, a sacred language, to belles lettres; on the contrary the dramas show that the Prākrits in literature were already under the influence of the Sanskrit Kāvya.