Kathasaritsagara (the Ocean of Story)

by Somadeva | 1924 | 1,023,469 words | ISBN-13: 9789350501351

This is the English translation of the Kathasaritsagara written by Somadeva around 1070. The principle story line revolves around prince Naravāhanadatta and his quest to become the emperor of the Vidhyādharas (‘celestial beings’). The work is one of the adoptations of the now lost Bṛhatkathā, a great Indian epic tale said to have been composed by ...

Terminal essay

WHEN, in the summer of 1919, I first approached Mr Tawney with the suggestion of reissuing his Magnum opus, little was decided about the form the Terminal Essay was to take. At that time there were so many immediate points connected with the work to be considered that any questions relating to the final volumes were to be deferred to a later date.

My own idea was to discuss briefly the manners and customs of the Hindus as illustrated in the work, together with some account of the different religious systems introduced. I then intended to speak of the debt Western literature owes to the East, and conclude with a few paragraphs on the classification of the world’s folk-tales. If room could be found, I was also going to give extracts from Speyer’s work on the Kathā-sarit-sāgara.

At that time, however, the idea of a Foreword to each volume by some eminent scholar had not been formulated, nor had the number or length of my own notes been determined.

As the scheme of the work began to take definite shape, matters became more established, and a precedent was gradually formed in accordance with what seemed to be the best way of dealing with subjects as they arose. Thus, whenever some custom, ceremony, name or incident was thought to require a note, it seemed most practicable to give it on the same page, or, if too long, at the end of the chapter.

Following this plan, all the notes which would have been used for the Terminal Essay were given in their respective places. It also proved much better to give Speyer’s translations and suggestions in situ, and not relegate them to the present volume.

My idea of inviting a different scholar to write a Foreword to each volume has proved a great success, and my work is now enriched by nine excellent Essays, each dealing with the great collection from a different angle.

With the appearance of the present volume, and its most interesting Foreword by Sir Atul Chatterjee, which approaches the K.S.S. from the economic standpoint, I find practically every subject which I might have treated in this present Essay already dealt with in a manner which I could never have equalled.

All general questions have been dealt with by Sir Richard Temple, Sir George Grierson and Dr Thomas; the study and classification of folk-tales has received expert attention from Dr Gaster, Mr Wright, Professor Bloomfield and Professor Halliday; while Sir Denison Ross has contributed original research work on the Persian recension of the Pañcatantra. I think it will thus be agreed that, on the face of it, there seems little left to write about.

There is, however, one subject which, as yet, we have not discussed in sufficient detail—the “frame-story” of the Kathā-sarit-sāgara, the arrangement and order of its contents, the sequence of events in the history of Udayana and Naravāhanadatta, the introduction of the numerous substories, and the resemblance the whole bears to the original Bṛhat-kathā of Guṇāḍhya.

I shall, therefore, devote this Terminal Essay to a brief discussion of this subject.


The “Frame-Story” of the Kathā-sarit-sāgara

In order to determine, as far as possible, the changes any recension of a lost original text may have undergone, two distinct methods at once suggest themselves: a critical examination of the version in question; and a reconstruction of the original with the help of other versions known to be derived from that same original.

In some cases it may happen that both these methods cannot be applied, and until quite recently this has been so with Somadeva’s work. Thanks, however, to the researches of Professor Lacôte, the Nepalese recension of the Bṛhat-kathā, known as the Bṛhat-kathā-śloka-saṃgraha, supplies us with evidence which can be compared with the results obtained from a close examination of the text of the Kathā-sarit-sāgara.

If the evidence from the one source corroborates that from the other, some definite conclusions will result. It is, of course, unnecessary to discuss all the points raised by Lacôte in his Essai sur Guṇāḍhya, but I shall endeavour to lay before my readers the main arguments for his conclusions, as far as they concern the present work.

The method I have adopted throughout of affixing a number to each story has not only enabled the thread of a tale long since suspended to be picked up again with ease, but facilitates the separation of the Main Story from the mass of sub-stories introduced on every possible occasion.

Readers will have noticed to what a great extent the latter are in excess of the former. This fact alone should make us suspicious, particularly when we remember[1] how, after the adventures of Naravāhanadatta had been brought to a successful close by his coronation, the long series of Vikrama tales are introduced for no apparent reason. The final return to the Main Story[2] is purely conventional, and clearly betrays the hand of a later editor.

Although many of the shorter sub-stories justify their position and introduction sufficiently well, there is a large number that fit uneasily into the places where we find them, and display no reason whatever for being there rather than anywhere else. This, of course, specially applies to whole collections, such as the Pañcatantra. Since studying Lacôte’s Essai, I am now convinced that it could never have been included in Guṇāḍhya’s original poem. A closer examination of Somadeva’s text of the Main Story will reveal many inconsistencies and inaccuracies which are largely hidden and unnoticed with the inclusion of so many sub-stories.


Book I: Kathāpitha (Vol. I, pp. 1-91).

Let us first, then, consider the Introduction to Somadeva. It will be remembered that it consists of a strange legend in which Guṇāḍhya himself plays a part. This fact did not diminish the belief of Brockhaus, Wilson and Lassen that such a person as Guṇāḍhya never existed in reality. Since their day, however, the advance in Sanskrit literary research has proved his existence beyond a doubt.

The evidence contained in the Bṛhat-kathā-śloka-saṃgraha only strengthens this opinion. We are introduced to Śiva and Pārvatī on Mount Kailāsa. In reply to a request from his wife for a story, Śiva relates his own history in one of his former lives. This is received with scorn as an age-worn tale, and Śiva is called a fraud. As compensation he promises to tell an entirely new tale that Pārvatī could never have heard before—the history of the Vidyādharas. Thus the hackneyed tales of gods, on the one hand, with their usual accompanying laudatory eulogies, and of men, on the other hand, with their sad and commonplace happenings, would both be avoided.

Pārvatī is placated, and, we are led to conjecture, listens in silence and interest to the long tale which Śiva unfolds.

This fact is significant as showing that the author puts forward strong claims to originality. The well-known Vedic and Puranic legends are not to be given—there is something that even a goddess would get a thrill over!

Yet this high standard is hardly borne out when we see later what old tales have crept in.

Kṣemendra is more cautious, and allows Pārvatī to raise no objections to Śiva’s first tale about himself, thus at once disarming criticism if well-known tales are introduced.

But let us proceed with the story.

Puṣpadanta, one of Śiva’s Gaṇas, overhears the tale by a trick and repeats it to his wife, who in turn tells it to Pārvatī. Thus Puṣpadanta is discovered, and Pārvatī’s wrath is pitiless. Both the eavesdropper and his friend Mālyavān, who pleaded on his behalf, are cursed to fall into mortal wombs.

Puṣpadanta, now to be born in Kauśāmbī under the names of Vararuci and Kātyāyana, will obtain release from the curse only when he meets a Yakṣa named Supratīka residing in the Vindhya forest under the name of Kāṇabhūti, and tells him the Great Tale. Mālyavān is to be born in Supratiṣṭhita under the name of Guṇāḍhya, and will be freed from the curse only when he has heard the tale from Kāṇabhūti.

In course of time Puṣpadanta-Vararuci-Kātyāyana meets Supratīka-Kāṇabhūti and tells him the Great Tale; then, after also relating his life-story in detail, reaches his heavenly home once again.

It is, however, with the history of Mālyavān-Guṇāḍhya that we are mainly concerned, for the legend may contain some clue to the real Guṇāḍhya. According to the story he is of semi-divine birth, his mother being a Brāhman girl and his father a Nāga prince. Thus he takes rank with the two other semi-divine authors—Vālmīki of the Rāmāyaṇa and Vyāsa of the Mahābhārata—and he is actually mentioned in Sanskrit literature as forming the third of the Epic trio.

Kṣemendra wrote maṅjarīs (abridged versions) of them all. The Nepālamāhātmya draws a comparison between the (Nepalese) versions of the legends of Vālmīki and Guṇāḍhya, showing how both men had to visit Nepal by divine command, the former to find a sacred spot worthy to be the cradle of the Rāmāyaṇa, and the latter to fulfil certain conditions necessary for his return to his previous semi-divine state. Both men erect liṅgas before leaving Nepal.

To return to Somadeva’s version, we find that Guṇāḍhya becomes a minister of King Sātavāhana in a city named Supratiṣṭhita, capital of the Pratiṣṭhāna (Vol. I, p. 60). On one occasion the king shows his ignorance of grammar (p. 69), and Guṇāḍhya offers to teach him Sanskrit grammar in six years. Thereupon another minister, Śarvavarman, promises to do it in six months, or carry his shoes on his head for twelve years. Guṇāḍhya considers this impossible, and says that if he succeeds, he, in his turn, will renounce for ever Sanskrit, Prakrit, and his own vernacular dialect.

By the favour of the god Kārttikeya a grammar known as Kātantra and Kālāpaka (on account of its conciseness) is revealed to Śarvavarman, who, with its help, wins the bet. In accordance with his vow, Guṇāḍhya, now reduced to silence, retires to the Vindhya forest. Here he learns the language of the Piśācas, and, on meeting Vararuci, writes down the Great Tale, as it is told him, in his own blood. (p. 89). This done, he sends it to King Sātavāhana, who, however, rejects it as being written in a barbarous language. On hearing this, Guṇāḍhya is in despair, and reads out the whole work to the animals of the forest, who crowd round, lost in admiration at its beauty. As he reads, so he bums the tale page by page.

Meanwhile the king, owing to a sudden and unexplained lack of nutritive qualities in his food, has fallen sick. He is informed that the explanation of this curious state of affairs is to be found in a Brāhman who is reciting a wonderful story in the forest, to which all the animals are listening motionless. Out of curiosity he goes to see for himself, and recognises Guṇāḍhya. He is, however, too late to save the Great Tale. All has been burnt, with the exception of the Adventures of Naravāhanadatta. This Sātavāhana takes back to his palace, and, in order that these strange happenings shall not be lost to the world, himself composes “the book named Kathāpīṭha, in order to show how the tale came to be first made known in the Paiśācha language” (p. 91).

Thus the first book of the Kathā-sarit-sāgara ends. But what does it all mean? Who is this Sātavāhana, at whose Court Guṇāḍhya became a minister? And what is the point of introducing a kind of grammatical controversy on the respective qualities of Sanskrit and Prakrit?

These are some of the queries that present themselves.

Sātavāhana is the family name, in inscriptions, of the Andhra dynasty, whose home lay in the Deccan, between the rivers Godāvarī and Kistna. Their capital was Pratiṣṭhāna, the modern Paithan on the north bank of the Godāvari. Thus Guṇāḍhya’s connection of king and capital is historically correct, although (as far as we can judge from Somadeva) he omits to mention which Sātavāhana is meant.

The third of the line, Śātakarṇi, is perhaps the most important of these kings. For he it was who wrested Ujjayinī from the Śuṅga king, Puṣyamitra. The evidence for this is numismatic, but the horse-sacrifice performed by him would find justification only in some such important feat of arms. Śātakarṇi gave his name to many subsequent Andhra kings, so that altogether his pre-eminence is undoubted.

But it seems most unlikely that our author would have omitted to mention, and even to enlarge on, such great victories, or to allude to the Aśvamedha. It looks, therefore, as if we must search among other Sātavāhanas. A most important point to notice is that the Andhra kings were patrons of Prakrit, and that it was only late in the history of the dynasty that Sanskrit was finally accepted as the Court language, and Prakrit was ousted from its former place of honour. Among the Sātavāhanas there was one king who became specially famous for being the centre of a literary Court and for being himself a poet of no mean order[3]—and that was Hāla. His date, though still uncertain, is considered to have been about the second or third century A.D.[4] Whether he finally became a convert to the use of Sanskrit we do not know, but grammatical controversies could not have been unknown. If it was not Hāla himself whom the legend of Guṇāḍhya makes ignorant of Sanskrit grammar, it is one of the succeeding Sātavāhanas; but in connecting any tale about the introduction of Sanskrit in the place of Prakrit with a Sātavāhana, it is Hāla that at once would be thought of.

A change so important and far-reaching as the use of a different language at the Court, and in literature generally, would, of course, take a considerable time to effect.

As patrons of Prakrit the Sātavāhanas would be the most vigorous opposers of such an innovation, and it is only in the time of Daṇḍin (sixth century) that we find the use of Prakrit becoming rare. The fact that in subsequent centuries native opinion looks upon Hāla as the central figure of Prakrit literature is surely a sufficient explanation of why Guṇāḍhya himself is represented in the legend as a native of Pratiṣṭhāna. Such evidence as exists points to Ujjayinī, or rather Kauśāmbī, as the birthplace of the real Guṇāḍhya; but once he is connected with Hāla, the champion of Prakrit, no further excuse for the work being in Paiśāchī would be needed.

It is only after the Kātantra grammar has converted the king to Sanskrit that he regards Paiśāchī as a barbarous language. Whether the real Guṇāḍhya and Hāla, or Hāla and Śarvavarman, were contemporaries or not in no way affects the argument, but it seems highly probable that Guṇāḍhya antedates Hāla, and that the growing legend used as an introduction to his work came into being later. It was well known by the sixth century, as Daṇḍin not only refers to the Bṛhat-kathā, but to the legend of Guṇāḍhya as well.

It now remains to mention Vararuci and his strange story, which, for some reason or other, has become connected with the legend of Guṇāḍhya. The stories of the two men are quite distinct. They never meet in the tale, and Vararuci could disappear, with his complete history, without upsetting the story in the least.

But the name of Vararuci is famous in connection with both Sanskrit and Prakrit grammar, and its introduction would merely assist in bringing the most famous grammarians on the stage at once. It then needed some clever invention to link the two entirely separate tales together as a single legend. On earth Kāṇabhūti is the common point of contact. But in the realms of heaven the person of Guṇāḍhya has been divided into two. It will be remembered that it is Puṣpadanta-Vararuci who originally overhears the tale and is cursed by Pārvatī. Surely, then, it is he who should have been made to repeat it on earth. Yet not only is it not so, but he receives less punishment than his friend Mālyavān-Guṇāḍhya, whose only crime was to plead for him.

Finally, Vararuci is born at Ujjayinī, the very place where internal evidence places the birthplace of Guṇāḍhya. From all these considerations Lacôte has come to the conclusion that the form of the legend as reproduced by the Kashmirian poets is purely a Kashmirian work.

“... dans la forme originate,” says Lacôte (Essai sur Guṇāḍhya, p. 33),

“Vararuci n’y paraissait pas et un seul gaṇa était maudit, le futur Guṇāḍhya. C’est ce dernier état de la légende qui devait être courant dans l’Inde.”

All the evidence certainly seems to point to this conclusion —the compiler or editor has been at work, and has produced a composite legend which, by its inclusion of grammatical disputes on the one hand, and lively sub-stories on the other hand, would appeal to both savant and bourgeois. The legend of Guṇāḍhya, as told in the Nepalese version by Budhasvāmin, confirms the belief in a much simpler original form than we find in Somadeva. There is only one Gaṇa, and he is known as Guṇāḍhya in his mortal life. Such alterations as there are can easily be explained by remembering that one of the chief objects the Nepalese had in view was to connect the names of heroes with their holy places of pilgrimage, and allow their actions to further sanctify those places.

The important point of this evidence is that Budhasvāmin dates from the eighth or ninth century, and thus antedates the Kashmirian poets. The work had not received the attention of editors who padded out the text with other collections, and thus the form of tales in the Bṛhat-kathā-śloka-saṃgr aha is much more likely to be closer to the original of Guṇāḍhya.

The title of this first Book of Somadeva is Kathāpīṭhā, which means “Introduction” or “Preface.”

The second Book has a very similar name: in fact the two words kathāpītha and kathāmukha differ in meaning little more than our “Introduction” and “Foreword.” But why should a work contain two introductions? Lacôte suggests that if, as is probable, the legend was added to the work later, a Kathāmukha was already there. The next best thing would be to use another word with almost exactly the same meaning.

Book II: Kathāmukha (Vol. I, pp. 94-189).

In accordance with the title of this Book, we should have expected it to contain merely introductory matter, such as the name of the narrator, the scope and object of his work, with possibly some laudatory reference to King Udayana and his son Naravāhanadatta. In fact we should have expected it to have resembled other “Kathāmukhas,” such as that which introduces the Pañcatantra. We have already seen that Somadeva omitted the Kathāmukha of the Pañcatantra, probably because the tales could quite easily be put into the mouths of characters in the Main Story. In this case, however, he has retained the title which he doubtless found in the texts he followed, although in the original Bṛhat-kathā the subject-matter may have been different and more in accordance with the usually accepted contents of a Kathāmukha. More than half the Book contains sub-stories which have but little connection with the Main Story, which, in order to make room for them, has had to be very considerably condensed. Otherwise the Book would have swelled to an undue size.

Thus we find the Main Story in this second Book crowded with incidents. We are hurried through the hero’s birth and childhood, and are introduced to Caṇḍamahāsena, King of Ujjayinī, who is anxious to marry his daughter Vāsavadattā to our hero. The schemes and counter-schemes to obtain this end follow, and finally the wedding takes place at Kauśāmbī. Udayana proves a fickle husband, but we are clearly given only a very condensed form of his amours. The Book ends, then, on a dramatic note, and we naturally turn to the next one to discover how things turn out.


Book III: Lāvāṇaka (Vol. II, pp. 1-116).

We are not disappointed. The Book opens with the lamentations of Udayana’s ministers at his desultory life—spent either with women or in the hunting-field. They fear he will never enlarge his realm, and are anxious for him to begin a series of conquests. Their eyes are first fixed upon Magadha, and their knowledge of political statecraft tells them that a marriage with Padmāvatī, daughter of Pradyota, King of Magadha, would be the easiest method to employ in the winning of their object. Vāsavadattā is naturally rather in the way for such an alliance, but a plot is cleverly engineered, and finally Udayana marries his second wife.

After all is smoothed over, and everyone is conciliated, the king, now roused from his idleness, determines on conquest. Accordingly he marches east to the sea, and circles India in a clockwise rotation, finally returning to Kauśāmbī.

The Book being almost entirely devoted to the Padmāvatī incident is much more easily condensed than was the case in the former Book. Hence ample opportunity occurs for the inclusion of a large number of sub-stories. The chief feature of interest in this Book, from an historic point of view, is Udayana’s conquest. We hear very little about it really, and, with the one exception of Brahmadatta, no particulars of the conquered kings, their countries, or deeds of prowess of the conquerors are forthcoming. The first point to be considered is the names of the people he conquers. He sets out eastwards to Benares, turns south, sweeps westwards and occupies Sindh. Among the tribes defeated are the Mlecchas, Turuṣkas, Pārasīkas and Hūṇas (Vol. II, pp. 93, 94).

Now Udayana was an ancient king of legendary times, yet here we find him fighting with peoples of comparatively recent times—Mohammedans, Turks, Persians and Huns. In fact the Hūṇas did not appear till the second half of the fifth century. Surely he should have fought with such peoples as the Yavanas and Śakas. The explanation seems simple. The peoples mentioned by Somadeva are those of the western and north-western frontiers, whose names would be known and appreciated in Somadeva’s time, and which, moreover, a Kashmirian would be most likely to employ.

Lacôte points out that the places supposed to have been conquered by Udayana constitute a pradakṣiṇa: the campaign is arranged like a pilgrimage. Central India is always kept on the right; and finally he visits Alakā, the city of the god Kuvera. Not a word is said as to how he gets there. No aerial chariot, magic shoes or any similar contrivance appears. Now several of the sub-stories in this Book are concerned with spells to enable one to fly through the air, yet we are given no clue as to why especially they are found in this Book. Might it not be that ancient tradition associated together Udayana’s campaign and some story of aerial transit? This would certainly explain the journey to Alakā. So perhaps in the original Bṛhat-kathā Udayana made a kind of aerial pilgrimage. In support of such a theory we have the fact that nothing more is said of all these vast conquests.

In fact, when finally Udayana leaves the world of mortals and gives all his possessions to Gopālaka, we find (Vol. VII, p. 102) that these consist only of Kauśāmbī. Surely we should be justified in expecting a long list of conquests to be enumerated!

The Kashmirian editors seem to have been very busy with this Book.


Book IV: Naravāhanadattajanana (Vol. II, pp. 125-165).

The story continues in due chronological sequence. Vāsavadattā longs for a son, and, after her pregnant whim for aerial chariots has been satisfied, Naravāhanadatta is born.

As in previous Books, the sub-stories occupy a very large part of the text.


Book V: Caturdārikā (Vol. II, pp. 170-239).

As we have already seen, Books II, III and IV form an uninterrupted series of events in the history of Udayana, but now comes a very distinct break.

Naravāhanadatta has been proclaimed a future king of the Vidyādharas, and this fact is an excuse for Śaktivega, a Vidyādhara prince, to relate in full how he reached his present high position. The tale, with its sub-stories, occupies the whole Book, and is a unity in itself. Whether it was in Guṇāḍhya’s original work in the same form as it appears here, or whether it has been compiled out of some of the adventures which formed part of Naravāhanadatta’s own adventures, are questions it seems impossible to answer.

The only point to stress is that the contents of this Book are entirely unconnected with previous or subsequent matter, and could be removed and inserted anywhere else without upsetting the text at all.


Book VI: Madanamañcukā (Vol. Ill, pp. 1-149).

The curious thing about this Book lies in the opening lines. Here we are informed that it is Naravāhanadatta[5] himself who from this point onwards is the true narrator, and that he tells his own history on a certain occasion after his coronation.

The actual words are:

“Now hear the heavenly adventures which Naravāhanadatta, speaking of himself in the third person, told from the very beginning, after he had obtained the sovereignty of the Vidyādharas and had been questioned about the story of his life on some occasion or other by the seven Ṛṣis and their wives.”

What does it all mean? It looks like the beginning of a new tale altogether, yet it is in reality a direct continuation of the story of Naravāhanadatta when last he was mentioned. So far it has been told in the third person, yet here is a note which specially tells us that henceforward Naravāhanadatta will narrate the tale in the third person. Now if it had said, in the first person, a distinct difference would naturally have been noticed at once. The value of this curious sentence, then, is quite inexplicable. If it had not appeared at all, we should have noticed nothing, for the tale would have gone straight on—still in the third person.

Why this sudden wish to introduce Naravāhanadatta as the teller of his own story? Perhaps the author of the Kashmirian recension thought that this was in accordance with tradition, and he was anxious at least to give some indication of this well-known fact. Even if this were so, we are still in the dark as to why it is inserted at this particular place, making it look like the very beginning of the whole work.

We are told nothing as to the occasion on which Naravāhanadatta was asked questions by the Ṛṣis. In fact the whole matter would remain a mystery if we were not to look ahead and find that full details of the visit to the Ṛṣis are given in Book XVI, chapters cxi, cxii.

Here we learn (Vol. VIII, p. 103) that after Udayana’s death, Naravāhanadatta spent the rainy season at Kaśyapa’s hermitage with his uncle, Gopālaka. Here it is that the Ṛṣis are assembled, and, in answer to their questions, he begins to relate his adventures.

Yet, if we are to believe the opening lines of Book VI, it was Naravāhanadatta who had been speaking all the time!

It is obvious, then, that Book XVI must have originally stood before Book VI, and, in fact, have led up to the statement that has caused all the trouble. It is not Somadeva who is to blame. He has merely followed his texts. It is the Kashmirian compilers who have purposely changed the order of the Books. Perhaps they worked from composite and incomplete texts, or perhaps they considered that the new order was better fitted to embrace all the new matter to be incorporated. Whatever may have been the true explanation, there can be no doubt that the order of the Books in the Kashmirian recension does not agree with that originally laid down by Guṇāḍhya.

The early part of the Book is taken up with the story of the Buddhist king, Kaliṅgadatta, and his daughter, Kaliṅga-senā. Many sub-stories are introduced, several of obvious Buddhist origin. The tale now centres on Kaliṅgasenā. With the help of her Apsaras friend Somaprabhā she sees Udayana, who immediately falls in love with her.

His faithful minister, Yaugandharāyaṇa, however, considers such a marriage undesirable for reasons of state, and finally manages to make it impossible by proving Kaliṅgasenā to be unchaste. A daughter is born to her by her lover, the Vidyādhara Madanavega. This child was in reality a son, but by Śiva’s orders was replaced at birth by a girl who was an incarnation of Rati. Her name is to be Madanamañcukā, and she is the destined wife of Naravāhanadatta All this is told in detail, but the rest of the Book is greatly condensed, and the events of the next ten or twelve years—the time to allow Madanamañcukā to grow up—are all crammed into Chapter XXXIV. In the next chapter we are in Book VII, and our hero is a full-grown man!

Although by far the greater portion of the Book deals with Kaliṅgasenā, yet it takes its title from Madanamañcukā. It seems obvious that the original work must have been much longer, and that the second half of the Book as it appears in Somadeva is a mere summary.

In fact there are places where we can clearly see the ruthless hand of the Kashmirian compiler, reducing what must have been incidents of considerable length to a single sentence.

For instance, we read in Chapter XXXIV (Vol. Ill, p. 140) that one day Naravāhanadatta goes to a garden called Nāgavana. What for? Nothing happens at all, except that he worships the snakes. It surely must have been the beginning of some adventure now entirely suppressed.

Even in the first part of the Book there are signs of mischievous alterations in the work. Why is Kaliṅgadatta such a nonentity, and why does he make no effort at all to protect his daughter after her trouble with Madanavega and the childish scruples of Yaugandharāyaṇa? Numerous other examples of improbabilities in the text could be given, but I think sufficient has been said to show that Guṇāḍhya’s original must have been very different to what we find in the Kathā-sarit-sāgara.


Book VII: Ratnaprabhā (Vol. III, pp. 155-300).

The first part of this Book is taken up with Naravāhanadatta’s marriage to a Vidyādharī whose name gives the Book its title. He is taken in a magic chariot to heaven for the wedding. This is the first time we hear of Naravāhanadatta leaving the earth. There is no connection between this adventure and the end of the previous Book.

With Chapter XLII (Vol. Ill, p. 259) begins the adventures of Naravāhanadatta in search of Princess Karpūrikā. They are far more important than the affair with Ratnaprabhā, and would much more fitly have given their name to the Book. It seems likely that the two parts formed separate Books in the original Bṛhat-kathā.


Book VIII: Sūryaprabha (Vol. IV, pp. 1-121).

Like Book V, this stands alone, and could be inserted anywhere as a separate story. It exhibits the highest flights of an unbridled imagination, and can be regarded as a great hotchpotch of ancient Buddhist myths and popular Hindu beliefs.


Book IX: Alaṅkāravatī (Vol. IV, pp. 122-251).

The first part of this Book is taken up with another Vidyādharī marriage—this time to Alaṅkāravatī, who gives her name to the Book. It is in no way connected with Book VIII, and could go in anywhere. The second part of the Book, beginning with Chapter LIV (Vol. IV, p. 184), stands as a complete entity, and is of considerable interest. It deals with Naravāhanadatta’s visit to Viṣṇu, and resembles the journey of the brothers Ekata, Dvita and Trita and of Nārada to the same “white island,” as related in the Mahābhārata (xii, 138, 139). The allusion in these passages to the worship of Christian communities in the East has already been pointed out.[6] Lacôte considers that the accounts of the visit to the “White Island,” as found in the Mahābhārata and the K.S.S., agree sufficiently well to suspect a common origin. Either the latter has borrowed from the former, or the Mahābhārata has taken the episode from the Bṛhat-kathā, or possibly both versions have been independently developed from a narrative derived from some traveller who had visited the Christian communities in Bactria.


Book X: Śaktiyaśas (Vol. V, pp. 1-192).

There is no connection between this Book and the previous one. After a series of tales dealing with the favourite subject of “fickleness of women,” introduced on the slightest pretext, we once again find Naravāhanadatta marrying a Vidyādharī. The wedding cannot be arranged for a month, and so an exceptionally large number of stories, including the whole of the Pañcatantra, can be successfully introduced.


Book XI: Velā (Vol. V, pp. 196-204).

This deals with Naravāhanadatta’s visit to Vaiśākha and his subsequent marriage to Jayendrasenā. The story of the merchant and his wife, Velā, gives its name to the Book. But why it is so very short and devoid of any continuity is impossible to say.

It looks as if it had been purposely compressed out of all recognition, in order, perhaps, to make up for the very long Books that precede and follow it.


Book XII: Śaśāṅkavatī (Vol. VI, pp. 1-221, and Vol. VII, pp. 1-193).

This Book has been discussed already in Vol. VII, pp. 194-196. We saw there that it is obviously in its wrong position, because we are continually told that Naravāhanadatta has lost his beloved Madanamañcukā; yet not only do we know nothing about this, but we are definitely told at the beginning of the Book (Vol. VI, p. 9) that it is Lalitalocanā who is lost.

Our attention, however, is taken off such trifles (!) by the appearance of the hermit Piśaṅgajaṭa, who proceeds to relate the huge tale of Mṛgāṅkadatta (Vol. VI, p. 10 et seq.), which stretches to p. 192 of Vol. VII.

The Book finishes without solving the mystery in the least.


Book XIII: Madirāvatī (Vol. VIII, pp. 1-17).

This short Book is a continuation of the last, for we find Naravāhanadatta still disconsolate at the loss of his beloved, who is now definitely stated to be Madanamañcukā, and not Lalitalocanā. The latter unhappy lady also is lost, but Naravāhanadatta seems to care little about her.

He meets two Brāhmans who tell tales of how they have successfully overcome difficulties, and so encourage Naravāhanadatta in his search. The heroine of the first Brāhman’s story gives her name to the Book. When the stories are finished, lo! Naravāhanadatta’s ministers turn up, and so does Lalitalocanā (nobody knows how or whence, and nobody seems to care!), and all proceed to Kauśāmbī. We have no clue whatsoever as to the loss of Madanamañcukā.


Book XIV: Pañca (Vol. VIII, pp. 21-69).

The long-awaited explanation of the loss of Naravāhanadatta’s chief wife, Madanamañcukā, is found at the very beginning of this Book. She suddenly disappears without a trace, leaving Naravāhanadatta distracted with grief. He searches for her in vain. Vegavatī, a certain unmarried Vidyādharī, is anxious to obtain Naravāhanadatta for a husband, and, taking the form of his lost wife, manages to trick Naravāhanadatta into going through the marriage ceremony again. The fraud is soon discovered, but she is soon forgiven on promising Naravāhanadatta to help to find the real Madanamañcukā, who, it appears, has been carried off by her brother, a Vidyādhara named Mānasavega.

Accordingly Vegavatī carries him through the air to the mountain Āṣāḍhapura, whither Mānasavega has hastened to kill them both. A magical combat ensues, in which Vegavatī is victorious. For safety she places Naravāhanadatta in a dry well in the city of the Gandharvas, and there leaves him (Vol. VIII, p. 27). He is soon rescued and, by his skill of playing the lyre, wins the king’s daughter Gandharvadattā for his wife. He seems to have entirely forgotten all about Madanamañcukā, and settles down to a married life of heavenly bliss. Suddenly a Vidyādharī appears, and takes Naravāhanadatta through the air to the city of Śrāvastī, with the intention of marrying him later to her daughter Ajināvatī.

While waiting in a garden, King Prasenajit comes along and marries him to his daughter Bhagīrathayaśas. One night Naravāhanadatta hears a low voice outside his sleeping-room. It is that of a beautiful Vidyādharī named Prabhāvatī, who moans the unhappy fate of Madanamañcukā in having so fickle a husband. At last Naravāhanadatta is roused by the mention of her name, and begs to be led to her presence. Accordingly Prabhāvatī flies with him through the air, and, by cleverly flying round a fire, becomes the wife of Naravāhanadatta Although Naravāhanadatta is anxious to consummate the marriage, Prabhāvatī says he must wait, and takes him to Madanamañcukā (Vol. VIII, p. 36). General rejoicings follow; but Naravāhanadatta, who is now wearing the shape of Prabhāvatī, is soon threatened by Mānasavega, who discovers his presence as Naravāhanadatta assumes his own shape. The supreme court of the Vidyādharas judge the case, and Naravāhanadatta wins. Mānasavega is far from satisfied, and a quarrel ensues. Naravāhanadatta escapes with Prabhāvatī, but Madanamañcukā remains a prisoner with Mānasavega. While Naravāhanadatta and Prabhāvatī are living together, Ajināvatī turns up with her mother and marries Naravāhanadatta He returns to Kauśāmbī with the two wives, where he is soon joined by Vegavatī and Gandharvadattā and all the relations of his various wives. A great campaign is decided upon, before which Naravāhanadatta has to obtain certain magical sciences from Śiva. While so engaged five (pañca) Vidyādharīs vow to marry him all together. This incident gives the name to the whole Book. After another marriage a great battle is fought. More marriages follow, including that to the five Vidyādharīs. Naravāhanadatta is now informed that before overcoming his final vow it is necessary for him to become possessed of the seven jewels of the Cakravartin. He wins the magic sandalwood-tree, but his obtaining of the other “jewels” is reserved for Book XV.

It has been considered necessary to give a somewhat detailed résumé of this Book, because, with the exceptions of the brief sub-stories 164, 165 and 166, it is entirely devoted to the Main Story.

There are several important points to notice. In the first place, the Book is crowded with detail. Marriages and adventures follow one upon the other at an enormous rate. In the second place, we must remember that they are all centred round the disappearance of Madanamañcukā. The Book, then, is really a cycle of marriages, with intermediate adventures. In this cycle the incident of Naravāhanadatta’s marriage to the five Vidyādharīs is comparatively unimportant, yet it gives its name to the whole Book. This fact, added to the obvious condensing of so many incidents in order to cram them into a single Book, makes it practically certain that ‘originally each marriage must have formed the subject and title of a separate Book.

We have had several examples of this already—e.g. Books VII, IX and X. Any doubt as to the probability of this is surely removed by finding that this is exactly what has happened in the case of the Bṛhat-kathā-śloka-saṃgraha. Each marriage has a Book to itself, and is recorded with far greater detail than in the K.S.S. On the evidence given by the K.S.S. itself we can definitely state that the present Book (and also Book XV, q.v.) originally must have come before Book XII, and consequently also Book XIII, which is a continuation of Book XII.

If this were not so, the events in Books XII and XIII could never have happened, for Madanamañcukā would not have been lost, and consequently the search, leading to all the other marriages and adventures, would never have taken place.


Book XV: Mahābhiṣeka (Vol. VIII, pp. 70-93).

This is a direct continuation of the previous Book. Naravāhanadatta obtains the seven jewels, and starts on the last of his expeditions. After sundry adventures and vicissitudes he conquers his sole remaining enemy, Mandaradeva. Naravāhanadatta proceeds to consolidate his empire. He marries five Vidyādharīs (a repetition of a similar incident in the last Book), and prepares for his coronation on the Ṛṣabha mountain.

The coronation takes place, and of his two dozen odd wives, Madanamañcukā alone is crowned with Naravāhanadatta Udayana, Vāsavadattā and Padmāvatī are invited, and with a blare of trumpets and general rejoicing the Book ends. Not only the Book, we would imagine, but the entire work.[7] Yet we find three more Books still unopened.


Book XVI: Suratamañjarī (Vol. VIII, pp. 94-131).

Years have passed. One night Naravāhanadatta has an evil dream, and, on awakening, calls upon the science named Prajñapti for an explanation. He is told all the news of his family in Kauśāmbī. Udayana, his wives and ministers are dead, Gopālaka has given his kingdom to Pālaka, and has retired to the Black Mountain in company with the hermits of Kaśyapa. Naravāhanadatta hastens there to see his uncle, and remains during the rainy season.

With Chapter CXII begins the incident of Ityaka’s attempted ravishing of Suratamañjarī, who gives her name to the Book. An inquiry is started. It turns out to be a family matter, and the evidence of Pālaka, his son, and his minister are needed. They accordingly are sent for, and the court sits. Evidence is found against Ityaka, but, by the request of the hermits, his life is spared.

The next chapter, the last of this Book, deals with the history of Tārāvaloka, and has nothing whatsoever to do with what precedes or follows. At the end of it Naravāhanadatta is still on the Black Mountain among the Ṛṣis. Here, then, is the occasion on which he is among the Ṛṣis already referred to in Book VI, and on which he is requested to relate “from the beginning” all his adventures.

If, then, Chapters CXI and CXII preceded Book VI, all would be clear.


Book XVII: Padmāvatī (Vol. VIII, pp. 132-209).

The Ṛṣis now ask Naravāhanadatta how he could bear his separation from Madanamañcukā. This is merely an excuse to introduce the story of Muktāphalaketu and Padmavatī, which takes up the rest of the Book. It is supposed to have been told during the period covered by Book XIV. Thus it is not in its proper chronological order.


Book XVIII: Viṣamaśila (Vol. IX, pp. 1-86).

This last Book also is out of place, as it is merely another tale told to Naravāhanadatta while he was separated from Madanamañcukā. But it is even more extraneous, as it deals with Vikramāditya, who was much later than the period to which Udayana and Naravāhanadatta must be assigned.

Somadeva (and perhaps even the Kashmirian compilers) places this Book at the very end as a kind of Appendix, for it would at once be apparent that heroes who were supposed to date from the time of Buddha could not listen to tales about a king as recent as Vikramāditya.

As already mentioned, the final return to the Main Story is purely conventional. So tame and unconvincing is the conclusion of this work, especially after the “grand finale” at the end of Book XV, that the most casual reader must at once suspect textual commutation on a fairly large scale.



Before we compare the order of the Books as found in the Bṛhat-kathā-mañjarī and Bṛhat-kathā-śloka-saṃgraha, with a view to reconstructing as far as possible the original work of Guṇāḍhya, it will perhaps be best to arrange in tabular form the points we have noticed in the foregoing pages:

Name of Book (comments)

1. Kathāpīṭha
(Legend of Guṇāḍhya. Complete in itself.)

2. Kathāmukha
3. Lāvāṇaka
4. Naravāhanadattajanana
(Uninterrupted series of events describing period from birth of Udayana to that of his son N.)

5. Caturdārikā
(Vidyādhara tale. Quite separate. Could go in anywhere.)

6. Madanamañcukā
(Unconnected. Apparently a fresh beginning. Must originally have stood after the first part of Book XVI, because of Ṛṣis incident.)

7. Ratnaprabhā
(Two love adventures. Probably once formed two separate Books.)

8. Sūryaprabha
(Like Book V. Vidyādhara tale. Quite separate. Could go in anywhere.)

9. Alaṅkāravatī
(Two distinct divisions. Both separate and unconnected.)

10. Śaktiyaśas
(Unconnected. Another marriage. Excuse for numerous sub-tales.)

11. Velā
(Another marriage. Very, and suspiciously, short.)

12. Śaśāṅkavatī
(Text shows Book must be in its wrong place. Naravāhanadatta’s chief wife is lost.)

13. Madirāvatī
(Direct continuation. Wife is still lost.)

14. Pañca
(Mystery of loss explained. Naravāhanadatta marries several other women.)

15. Mahābhiṣeka
(Direct continuation. Leads to coronation. Finale.)

16. Suratamañjarī
(Another unconnected Book. First part helps to explain Book VI and should precede it. Second part quite separate. Could go in anywhere.)

17. Padmāvatī
(Out of place. Told during period of Book XIV.)

18. Viṣamaśila
(Out of place. Told during period of Book XIV, but is also an obvious addition, and could not have been in the original.)

We can now see the situation at a glance. Books II, III and IV form a group; V and VIII are unconnected and both Vidyādhara narratives; VI looks like a new beginning, but lacks any explanatory introduction; VII, IX, X and XI are marriages, more or less unconnected; XII and XIII are closely connected, but must come after XIV and XV (also connected), and consequently also after XVII and XVIII, because the events they relate happened during the period covered by XIV. The remaining Book, XVI, must be regarded as of two distinct divisions, the first supplying the necessary introductory matter to VI, and the second being quite unconnected.

It will thus be seen that the critical inspection of the work as presented by Somadeva shows without doubt that the work has undergone much reshuffling as far as the order of Books is concerned.

We can now turn to the Bṛhat-kathā-mañjarī and see if the order followed by Kṣemendra in any way confirms our theories.

The first five Books correspond to those of Somadeva. Then the differences begin. We notice Books V and VIII are put together. This is followed by Velā, the very short Book, but the chief interest here lies in the fact that it finishes with the loss of Madanamañcukā. In the K.S.S. this incident is found at the beginning of Book XIV, Pañca. Thus, so far, we find Kṣemendra’s order much better than that adopted by Somadeva. After Velā he has placed Books XII, XVIII, XIII, XVII, thus obtaining a correct sequence of events, which is lacking in Somadeva. Book XIV follows, but with its opening incident transferred to the end of Book XI (Velā), as has been already mentioned.

Thus we see that our complaints about the order of the Books in the K.S.S. are fully justified by what we find in Kṣemendra. The question which at once presents itself is, Why did not Somadeva copy the order in Kṣemendra instead of changing it and so introducing muddling anachronisms? The answer would appear to be that he took what he considered to be the lesser of two evils; for although Kṣemendra has followed a better order of Books dependent upon the loss of Madanamañcukā, he has had to pay dearly for it in the rest of his work. For here we find chaos, and no attempt made to remove it. Such inconsistency makes us chary of giving Kṣemendra credit for the arrangement of the first part of the work. He probably left it as he found it. Somadeva, on the other hand, saw how unconnected his material was, but preferred to put together only such chapters as were undoubtedly connected. We have seen how Book XV follows directly on to XIV; but Kṣemendra, by his placing of Pañca, has been forced to separate them by other three Books, thus introducing all kinds of improbabilities and chronological impossibilities.

The incident of Naravāhanadatta relating his adventures to the Ṛṣis in the third person must have seemed entirely upsetting to Kṣemendra, and he gets over the difficulty by omitting it altogether. As Lacôte has remarked, the above clearly shows that the Kashmirian Bṛhat-kathā was a compilation and not an original work.

I think we must attribute the unsatisfactory state of the text of the Kashmirian work very largely to the simple fact that the compilers (there may have been several at different dates) were not trying to reconstruct in their entirety the adventures of Naravāhanadatta They had a very different object in view—namely, to use the story as a frame for all the tales they could collect together. The better-known incidents would have to appear in some detail, while many of Naravāhanadatta’s love-adventures could be ignored or highly compressed. The result has its pros and cons. On the one hand we are given a jumbled and very defective version of the story of Naravāhanadatta, but on the other hand we have that huge mass of tales which sheds so much light on the manners and customs, the folklore and beliefs of a country so poor in historical documentary evidence.

True, the Pañcatantra and Vetālapañcaviṃśati are found in separate collections, but scholars are not yet agreed as to the respective values of the different versions.

That Somadeva was very conscious of the difficulties in the text or texts he was using is clear from his introductory remarks (Ocean, Vol. I, p. 2), where he says:

“... the observance of propriety and natural connection, and the joining together of the portions of the poem so as not to interfere with the spirit of the stories, are as far as possible kept in view....”

The meaning of this is not perfectly clear, and great importance should be laid on the correct translation of the passage.

Many suggestions have been made, but Lacôte alone has treated it in the light of his extensive critical examination of the whole subject, taking into consideration all debatable grammatical queries and all possible modes of construction.

His translation of the full passage is as follows:—

“Tel l’original, telle cette copie; pas d’une ligne même elle ne s’en écarte. Je comprime le volume du recueil et je traduis, voilà toute la différence. Attentif à observer, autant que je le puis, les convenances (littéraires) et l’ordre logique, en ayant soin de n’interrompre ni le récit ni le ton des sentiments, je ne le suis pas moins à disposer une portion de poème régulier. Mes efforts ne vont pas à gagner une réputation d’artiste consommé; je veux simplement qu’on puisse retenir sans peine ce vaste ensemble de contes de toute espèce.”

This clearly means that he has been accurate as far as the subject-matter is concerned, but has found it necessary to alter the order of some of the Books. Here he surely must refer to Books VI-XVIII, while the “portion de poème régulier” which he has been so careful to arrange in proper order can be none other than Books XIV-XV.

When we turn to the Bṛhat-kathā-śloka-saṃgraha[8] we at once find ample support for our theories. The order of the Books is reasonable and clear, and what in the Kashmirian versions was passed over with little more than a mere reference is now detailed in full. In fact, we not only meet with entirely new adventures, but find certain of the characters presented in quite a different light.

For the first time the improbabilities found in the Kashmirian accounts of Madanamañcukā’s marriage and the romance of Kaliṅgasenā entirely disappear. Their social standing is certainly much lower, but this only adds to the strength of the plot.

Vegavatī, being of much higher birth, has been accepted by the Kashmirians practically unaltered. Their desire to raise the social standing of the principal characters to the detriment of the tale is manifest. In some cases where they have raised merchants to the rank of princes, or mortals to the degree of Gandharvas, we are able to detect the fraud, for the same names have been retained with suffixes which violate the accepted rules of Sanskrit etymology.

So great appears to be the wish of the Kashmirian compilers to raise the social tone of the work, that tales which cannot escape their low-type settings are altogether omitted, but appear in detail in the Nepalese version.

Without giving other evidence of the accuracy of Budhasvāmin’s work as detailed by Lacôte, I would mention one point which seems to me of great importance. We have, of course, noticed that throughout the whole of the Ocean the chief deity is Śiva. Now, in the Śloka-saṃgraha it is not Śiva, but Kuvera. The name of the hero alone tells us which is correct. Naravāhanadatta means “given by Naravāhana.”

Naravāhana is one of Kuvera’s, and not Śiva’s, titles. So, when Udayana was praying for a son, it must have been Kuvera whom he worshipped, otherwise our hero’s name would have been Śivadatta or some other name compounded from one of Śiva’s many titles.

It is obvious that the Kashmirian compilers have altered the name of the deity in accordance with local contemporary beliefs.

Numerous other examples of the reliability of Budhasvāmin’s work could be quoted, but full details will be found in Lacôte’s Essai. With the help, then, of the Śloka-saṃgraha, we are able to get a fairly shrewd idea of what Guṇāḍhya’s original work must have been like. The first Book corresponded to Book XVI of the K.S.S. It contained the history and abdication of Gopāla and Pālaka, which led up to the incident of Ityaka and Suratamañjarī. The subsequent trial brought Naravāhanadatta on the scene, who later was asked to relate his history. After some hesitation (only in the B.K.S.S.) he commenced (K.S.S., Bk. VI, ch. xxvii) by relating his family history (K.S.S., Bks. II, III—with possibly another, now lost, giving further details of Udayana’s amours[9]).

The story of his own birth (K.S.S., Bk. IV) follows. Ignoring the two Vidyādhara Books (K.S.S., Bks. V and VIII), which, as we have already seen, could go in anywhere, we come to the heroine of the whole story, Madanamañcukā.

N. sees her as a child and falls in love with her (K.S.S., Bk. VI, ch. xxxiv). Various adventures follow (only in the B.K.S.S.), leading up to the marriage (K.S.S., Bk. VI, ch. xxxiv-end). Then comes the sudden disappearance of Madanamañcukā (K.S.S., Bk. XIV, ch. cv), resulting in numerous adventures, usually terminating in a fresh marriage. The order and number of Books thus formed cannot be determined for certain, but in the K.S.S. they certainly included Books XIV (chaps, cvi, cvii), VII, IX-XIII, XIV (ch. cviii) and XV.

We can also add Books XVII and XVIII, if, as Lacôte thinks is the case, they are not apocryphal.

The plan of the Bṛhat-kathā resembles that of the Rāmāyaṇa to a certain extent—the setting out of the hero to recover his lost love, acquiring others on the way, the constant help of a trusty friend, the purity of the captive wife, and the final triumph on her safe recovery.

We must not press the comparison further; but to disregard it would be a mistake, because then we would miss the due appreciation of the genius of Guṇāḍhya. Not that it is evident from the fact that he has copied the plan of the great Epic, but because, having copied it, he proceeds to treat his subject-matter in a way unheard of and absolutely original.

His heroes are not borrowed from the great national epics, the deity is not the omnipotent Śiva or Viṣṇu, and the incidents in the tale are not confined to kings, princes and gods.

In place of this usually accepted precedent we find the heroes are but petty princes who rub shoulders with merchants, artisans, sailors, adventurers and beggars. The heroine is the daughter of a prostitute, but her desire to raise the level of her caste and be worthy of her husband gives great strength to the character that Guṇāḍhya has created. The chief deity is Kuvera, the god of merchants and treasures.

All this must have struck contemporary audiences as most original and novel. But there is another point that we must not miss. The nature of the work would reach a much wider public—the kind of public, in fact, which would flock together at the annual festivals held at Kauśāmbī and Ujjayinī. Perhaps long extracts from the Bṛhat-kathā were recited at these events; anyway I notice Lacôte thinks it likely.

We can now more readily understand that the Kashmirian compilers would find much to alter and suppress. The necessity for an Introduction also becomes more apparent.

Thus at the end of our short inquiry we find that the K.S.S., as we have it to-day, is but a poor and badly arranged version of the original work. This Somadeva must have known; and though we see he has done his best to rearrange certain portions of it, he was well aware that any attempt to reconstruct it entirely would mean little less than composing a new work.

There was, I think, another factor which prevented Somadeva from making too drastic alterations—namely, his wish to retain all that mass of sub-stories added by the Kashmirians. The frame-story had been altered in order to take them in as naturally as possible. Although in many cases they are introduced in the most clumsy fashion, it is clear that considerable alterations would have to be made in Guṇāḍhya’s text before it was ready to receive so many new stories.

But we must not complain—far from it—for the result has been that in about a.d. 1070 Somadeva has presented us with one of the greatest collections of tales the world has ever seen —tales which not only mirrored contemporary customs and beliefs, and exhibited the versatile genius of the story-teller, but tales which were destined to inspire the genius of unborn giants of European literature—Boccaccio, Goethe, La Fontaine, Chaucer and Shakespeare.

As to Kṣemendra, we should have lost little if he had not lived, or at any rate had not produced a version of the Bṛhat-kathā.

But with Somadeva matters are very different. We must hail him as the Father of Fiction, and his work as one of the masterpieces of the world.

Footnotes and references:


See Vol. VIII, p. 93.


Pp. 85, 86 of this volume.


In the article on “Prakrit,” by Sir George Grierson, in the Ency. Brit., vol. xxii, p. 253, he says:

“Hāla's work is important, not only on its own account, but also as showing the existence of a large Prakrit literature at the time when it was compiled. Most of this is now lost. There are some scholars (including the present writer) who believe that Sanskrit literature owes more than is generally admitted to works in the vernacular, and that even the Mahābhārata first took its form as a folk-epic in an early Prakrit, and was subsequently translated into Sanskrit, in which language it was further manipulated, added to, and received its final shape.”


See further Wintemitz, Geschichte der Indischen Litteratur, vol. iii, pp. 102, 103.


In future I shall thus refer to the hero Naravāhanadatta.


Sir George Grierson, “Modern Hinduism and its Debt to the Nestorians,” Journ. Roy. As. Soc., 1907, p. 7 et seq.


See Vol. VIII, p.93n2.


Discussed in detail by Lacôte, Essai, pp. 146-198, and edited by him, with a French translation, the same year (1908).


It seems probable that Guṇāḍhya used only a portion of the widely known Udayana cycle of legends current at the time.

Reference should be made to Burliṅgame’s Buddhist Legends, Harvard Oriental Series, vol. xxviii, pp. 247-293; Synopsis, pp. 79-84; parallels, pp. 62-63.

Like what you read? Consider supporting this website: