Foreword to volume 4

NOT much has been left to be said by way of introduction to the fourth volume of this splendid publication. The reader is familiar with the circumstances of the Sanskrit author’s life; he knows what is necessary concerning the literary sources of the work; he has considered the origin of the stories, whether Aryan or Dravidian, in India itself, and their affinities with beliefs and practices in later India; and he has contemplated the important and difficult questions of transmission—transmission of stories and motifs from country to country, people to people, and the no less certainly attested inverse process of transmission from literary source to folk-lore. Then, again, the very march of the narrative has accustomed him to the ease of the author’s style, fitting the matter like a glove, objective, impersonal and unmoved, whether the scene is earth or heaven or one of the various hells, an unvarying style equal to the burden of the long task. And the translator, as became a ripe scholar of fine literary taste, follows with a rendering as free from display as is the original itself.

It was by no means a matter of course that the Great Tale of Guṇāḍhya should come down to us in so acceptable a form. The example of Kṣemendra’s Bṛhatkathā-mañjarī (“Great-Tale Cluster”) shows clearly that we might have had to be content with a much more restricted version by an author solicitous of poetical artifice rather than of the adequate presentation of the matter. Written in an old dialect little practised and contemned as vulgar, the work of Guṇāḍhya was not safeguarded, like the Mahā-Bhārata and Rāmāyaṇa, which Kṣemendra subjected to the like treatment, by having been composed in the sacred language, by a theme relating to the great heroes of antiquity, by ancient fame and semi-divine character attaching to the author. Even as it is, the original prose story is presented to us with curtailments and amplifications of the most wholesale character. While the result has been, no doubt, favourable to the work as an Ocean of Story and a storehouse of popular idea and folk-lore, and thus better adapted to the purposes contemplated in Mr Penzer’s monumental edition, it is still a matter of some concern to the reader to realise rather more fully the vicissitudes through which the text has passed.

Apart from the Kashmir redactions there exists a Sanskrit version of Guṇāḍhya’s work, bearing the title Bṛhatkathā: śloka-saṃgraha—i.e. the “Great Tal: Verse Epitome.” In spite of its rather unassuming title, it must in its complete form have been — for we possess only about six of the twenty-six lābhas, “emprizes”—of very considerable extent, say about 25,000 ślokas or couplets. Its discoverer and editor, M. Félix Lacôte, has published (Essai sur Guṇāḍhya et la Bṛhatkathā, Paris, 1908) along with the text an elaborate discussion of all the questions of higher criticism relating to the Kathā-sarit-sāgara and the other recensions. M. Lacôte’s conclusions, which are developed with great perspicacity, may be summarised as follows.

The Śloka-saṃgraha —“Verse Epitome”—of which the MS. came from Nepal, is the work of a Nepalese writer, by name Budhasvāmin, who is at pains to bring the poem into connection with his own country. It is of relatively early date, say the eighth to the ninth century A.D. and is based upon the Paiśācl original. In its arrangement, and still more in its contents, it differs widely from the Kashmir versions. The most significant feature of these differences, however, is that they are largely by way of defect. Great masses of the subsidiary tales in the Kathā-sarit-sāgara are wanting, and thus the main narrative stands out in much greater distinctness and amplitude. It is not that Kṣemendra and Somadeva have greatly perverted the story although there are some rather considerable dislocations, which deface its logical coherence. The chief difference is that the story is reduced to a rather slender trickle, which tends to be lost in the deluge of adventitious matter. The Paiśācī original, though including, like other Indian narratives, a quota of incidental tales and episodes, was concerned predominantly with the actual adventures of Naravāhana-datta, a hero of Guṇāḍhya’s own invention. A novel, in which the subordinate characters were largely middle-class people, it was distinguished by a great variety and abundance of incident. The author, who was a born story-teller and the real creator of the literary Kathā, appears to have travelled widely, perhaps chiefly on the great trade-route which connected Prayāga (Allahabad), Kauśāmbī and Ujjain with the ports of Western and Eastern India. He had listened to the tales of wayfaring men and of voyagers from the great seas. He had visited the cities and learned the narratives of local fame: in the Kauśāmbī country, even more than in Ujjain, the adventures of Udayana must have been the talk of the village greybeards. A portion of the matter relating to Udayana and Pradyota existed already, no doubt, in literary form, and it is preserved to us in Buddhist originals or adaptations. The composition of the poem in eighteen sections may have been imitative of the Mahā-Bhārata. Is there anything in the idea, propounded by M. Lacôte, that its content, a narrative of travels and loves, was inspired by the Greek novel? The supposition has no inherent improbability. The literary influence of Greece in the East did not end with the Seleucid, or the Parthian, empire; it has recently (by Prof. H. Jacobi in Antidoron, Festschrift Jacob Wackernagel, Göttingen, 1923, pp. 126 sqq.) been suggested that a far-off reflex of the hexameter metre (known, perhaps, in Gandhāra) is traceable in some Jain poems composed in a rather late (Apabhraṃśa) dialect.

M. Lacôte subjects to a lengthy and penetrating criticism the composition of the Kashmir Brhat-kathā. Somadeva’s claim to fidelity in the handling of his original is fully justified by a comparison with the work of his predecessor, Kṣemendra. A Bṛhat-kathā such as he reproduces, a prose work in the Paiśācī dialect, existed, therefore, in Kashmir. But it was no longer the book which Guṇāḍhya had composed. It was a huge compilation, incorporating not only many particular stories from heterogeneous sources, but even whole books such as the Pañcatantra, the “Twenty-five Tales of the Vampire” (Vetāla-pañcaviṃśati) and the story of Nala. The charge of abridging, obscuring and dislocating the main narrative is valid, not against Somadeva and Kṣemendra, but against predecessors, whose work of amplification had been completed, so far as completion can be predicated, perhaps two or three centuries earlier. The process has operated in the case of other compilations within and outside India. All the rivers run into the sea; and the rhapsodists of different particular narratives were as urgent for inclusion in the Great Tale as the latter was hospitable in admitting them. We only wonder who were those compilers possessed of competence and goodwill to adorn the Pañcatantra, Nalopākhyāna, and so forth, with a Paiśācî dress, and for what audience they laboured.

The reader will have remarked the intimate connection of the story with questions of dialect and of grammars. At the outset, in the Prologue (Kathā-pīṭha), which, however, can hardly be attributed to Guṇāḍhya himself, we are confronted with a rivalry between the old Pāṇinean grammar, which demanded twelve years for the acquisition of the Sanskrit language, and the new system of the Kātantra, professing to accomplish the same result in six. A modicum of practical reality is here, no doubt, en jeu. The Pāṇinean grammar, with its artificial system and its subtle commentaries, was doubtless better adapted for a lifelong study of the language than for practical instruction. As the classical language became more indispensable for worldly people of the middle class, previously content with dialects or the imperfect Sanskrit which we find exemplified in the early Buddhist texts, their ambition for culture might have been unequal to the difficulties presented by the venerable text of Pāṇini, itself in various points out of correspondence with the current speech of the learned. With such aspirants the newer methods may have worked miracles. On the part of Guṇāḍhya the recourse at such a period to a fresh, unheard-of literary speech may be challenged with wantonness. Such are the wilful ways of genius: have we not modem stories composed entirely in Chicago slang? But, if we modems are prepared to allow such liberty to authors generally, the compatriots of Guṇāḍhya required for sanction the stipulations of a vow. This does not excuse us from demanding why and how the actual Guṇāḍhya chose the Paiśācī. There are too many, though sporadic, indications of “Paiśācī” tendencies in various parts of India to allow the supposition of a wholly artificial form of speech. On the other hand, we have as yet no real evidence of the existence of any people or class known by the name Piśāca, which denotes a man-eating demon or spirit. What designation Guṇāḍhya would have applied to the dialect we cannot say. The name Paiśācī, though it appears in the oldest Prākrit grammar, that of Vararuci, is perhaps due to the story related of Guṇāḍhya himself. In later times there were many varieties of Paiśācī, bearing subsidiary designations of a local character. Since the earliest language of the group is described as coinciding in general with a particular local speech, the Śaurasenī of the Ganges-Jumna Doab and the adjacent regions, it would seem as if the Paiśācī, which is characterised chiefly by a few striking peculiarities of pronunciation, was properly a dialect of an inferior class, or of classes, in society. The class may have been of aboriginal origin, whether Dravidian or North-Western or otherwise— there are many such in India—and it may have been more widely than numerously represented: for instance, it may have been in one of those classes (such as couriers, ostlers and the like) with which travellers came into contact; and this might explain the choice of it by the travelled author. It would be quite in accordance with Indian ways if, in this application, the term Paiśācī were an intentional perversion of a class or tribe name: but to pursue this suggestion would be hazardous. A certain reality is lent to the Paiśācī language by the further statement, perhaps itself concocted at no very early date, that it was adopted by a sect of Buddhists for their writings.

As a grade of non-human creatures the Piśācas have already been discussed (see Vol. I, pp. 92-93). They are far from respectable, except on the ground of antiquity, wherein they may rival with the best, since the Veda recognises their existence. They are with difficulty distinguished from the Rākṣasas, or demons, and the distinction is not on the credit side, since, while smaller and less formidable, they are even more odious. If the Rakṣasa marriage is the forcible abduction of a woman after killing her kinsmen and breaking into her dwelling, the Piśāca marriage, the basest of all, is the overpowering of one asleep or tipsy, or disordered in intellect. Like the Rākṣasa, his meaner confrère was an eater of human flesh. Naturally also he was a night-walker; in the books on logic he is the standing subject of the doubt: “Is it a Piśāca or a tree stump?” This gives us his probably essential character as the ordinary malignant spirit of the dusk, or more materialistically—for he is attached to water as the salamander is to fire—as Jack-o’-Lantern, the “will-o’-the-wisp.” His name is, unfortunately, not etymologis-able with prudence; and therefore the way is still open to those who would regard this part of his equipment as derived from some aboriginal people. That a differential dialect should be ascribed to the Piśāca we may ourselves (for do not ghosts “gibber”?) find natural enough; still more obvious was it to the ancient Hindus, who in their Brāh-maṇas have, like Homer (χαλκιδα κικλήσκουσι θεοί, ἄνδζες δὲ ύμινδιν etc.), quoted for us specimens of the language of the gods.

It happens that the present volume is largely concerned also with a second class of supernatural beings, regarding whom, therefore, a few remarks may not be out of place. These are the Vidyādharas—“knowledge-holders”—usually conceived of in connection with Kuvera and having a king, Cakradharman, who resides in Kuvera’s palace. In general, however, they are spirits of the air: they scatter flowers over fighting warriors. They are devoted to music and dancing, and their females are of extraordinary beauty. They are weakly distinguished from the Gandharvas and Apsarases, who historically are their predecessors. In the Pali Tripiṭaka they are still preponderatingly, like the Gandharvas, spirits who seek to enter into women—perhaps a far-off reminiscence of a stage when, as anthropologists really seem to admit, pregnancy was not known to be a consequence of marriage.

The name of this class of divinities points to their origin. It is a constantly recurring phrase in the Brāhmaṇas that “he becomes” such and such “who knows this.” The knowledge, or secret, or upaniṣad, was a key-knowledge, which afforded access to special powers, a talisman: and in later times there were very many vidyās and mahā-vidyās in the form of mantras, which were, in fact, nothing but spells. It follows from this that the Vidyādharas were, like the Siddhas, “those who had realised a certain attainment,” not seldom recruited from the race of men. In the Harṣa-carita of Bāṇa (c. iii.) a certain celebrated saint completes his career by the performance of a nocturnal rite, whereby he acquires “the hair-lock, diadem, earring, necklace, armlet, girdle, hammer and sword” and becomes a Vidyādhara: he is then rapt away through the firmament to his appointed station. Like our wizards (wise-ards), the Vidyādharas therefore are primarily the successful penetrators of superhuman secrets: that in India they attained to a distinctive status in a divine hierarchy is in full harmony with the general tendencies of Indian thought. Perhaps Mr Penzer, who has enriched this fine work with so many valuable notes and dissertations, will consider the possibility of dealing somewhat fully with the literature of “spells,” for which India supplies an inexhaustible material.

The “hammer” (mudgara) of the Vidyādhara is not without an interest of its own. Mr A. B. Cook, in his Zeus, vol. i (Cambridge, 1914), describes and illustrates (pp. 109-110) the class of beings called Kabeiroi, who were connected with the Muses:

“they have bushy hair, a thick ring round the neck, a loin-cloth about the waist, and a heavy doubleaxe or hammer on the right shoulder.”

Since the Vidyādharas are the subjects of Kuvera (Kabeiros), and since, like the Kabeiroi, they have a special mountain home, there is a good chance that the detail of the hammer may be not devoid of historical significance. Nor does the matter end here. If the Vidyādhara duplicates the Gandharva, his consort, the Vidyādharī, who is connected with music and arts, will bear a relation to the (Gandharva’s feminine associate, the Apsaras. When we have said Apsarases, we have practically said Muses, the “mountain goddesses,” who in Greece came to be patronesses of music and literature. And the Apsaras, again. in her function of receiving the spirits of heroes falling on the field of battle, seems to have more than a plausible connection with the Northern Valkyrie.

More than one reader, perhaps, will be surprised at the honourable and leading role played by the Asura Maya in the Sūryaprabha story. In general the unorthodox classes of beings in Hindu cosmology are far from being definitely reprobate (see Hopkins. Epic Mythology. Strassburg, 1915, pp..38 sqq.) individuals in all the grades are capable of meritorious works. Was not even Rāvaṇa famous as an authority in medicine and grammar? In fact, the Hindu theory of rebirths provides no place for a final damnation. In the ease of the Asura Maya we are dealing with a personage indeed, the great architect, inspirer of the Maya-mata, who in the Mahā-Bhārata is the constructor of the splendid palaces there described.

Finally. we need not demand why in so mundane a book as the Bṛhat-kathā the chief hero's exploits should be directed to an ultimate sovereignty over a celestial realm. Even from our own mediaeval tales, even from the Greek romances, it would not be feasible to exclude a supernatural element. The art of story-telling, which begins with gods for heroes, does not quickly descend to a merely human level. Guṇādhya may have thought that he had gone far enough when he accepted men and Vidyādharas in place of heroes and gods; and centuries after his date the Kādambarī, the toucing story of Bāṇa, still finds its leading personages in the Gandharva world.

F. W. THOMAS.

July 1925