Pramāṇa · Search · About · API

A History of Sanskrit Literature c24: XVII. The West and Indian Literature

A History of Sanskrit Literature c24

Language
English

Contents

XVII. The West and Indian Literature

1. The Fables and Märchen of Greece and India

c24.s1.1

THE obvious parallelisms between Indian and Greek fairy tales and fables have never been ignored, and have evoked lively controversies. Wagener¹ held that Greece was the recipient, but both Weber² and Benfey³ came to the conclusion that the Indian fables were borrowed from Greece, and for this view there could be adduced the question of chronology; the Greek fable is clearly in existence in the time of Hesiod, is hinted at in Homer, appears definitely in Archilochos and Simonides, and is developed into an important branch of literature, though the actual date of our collections is less certain. Herodotos, however, knew of Aisopos as a fable-teller, and Babrios (*c.* A.D. 200) and Phaedrus (*c.* A.D. 20), if themselves late, drew from earlier sources. Benfey complicated the position by holding that fairy tales were normally Indian in origin, thus establishing a dualism which was difficult to defend. Keller⁴ contended for the priority of India, and this view has recently been revived and insisted upon.⁵ As a chronological consideration stress has been laid on the monumental evidence in India, especially at Bharhut of the third or second century B.C., for the existence of beast fables, and some would accept the Jātaka stories as already existing in the fourth or fifth centuries B.C., although this is manifestly dubious. Various criteria have been imagined by which to decide priority; Weber preferred the test of simplicity, naturalness, or naïveté, Benfey thought that incompleteness was often a sign of greater age, while Keller laid stress on the doctrine of logical sequence and conformity to the habits of animals as revealed in nature. Thus he developed the argument that the fact of the jackal following the lion to partake of the remains of his kill is true to nature, and easily suggests to the early fabulist the conception of making him minister to the lion as king of beasts, whence, as the minister must according to Indian tradition be a miracle of cunning, the jackal is thus reputed; in Greece where the fox appears in the role of the jackal, his position is unexplained, for he is not in reality a very cunning animal. Unhappily this ignores, apart from the fact that it is fancy, not fact, that creates a world of intelligent beasts, the possibility that the fable had its origin neither in India nor in Greece, but was a product of lands intermediate between these countries. Weber justly contended that, if the relation of lion and jackal came thence to Greece, it would have to be changed to suit Greek conditions, while, if it later reached India from Greece, it would have been necessary there to reinstate the jackal. Or, more naturally, it may be held that the fable reached both west and east from the common source in the early fables connected with the name of Aisopos. We cannot ignore the possibility of Egypt having played a part in the genesis and transmission of fables, and Diels¹ has with special reference to Kallimachos claimed for Lydia a substantial share in the work of diffusion. Hertel,² again, has insisted that the idea of making use of fables to given instruction in politics is essentially Indian, and on the strength of it has claimed for India originality in respect of the best Greek fables; but the assertion is as little capable of proof as the claim that Greece excels in witty and pointed fables which in India have often suffered watering down at the hands of Buddhist and other preachers.

c24.s1.2

¹ *Les Apologues de l'Inde et les Apologues de la Grèce* (1854).

c24.s1.3

² *IS.* iii. 327–73; *SBA.* 1890, p. 916.

c24.s1.4

³ Trans. of *Pañcatantra*, I. x ff.

c24.s1.5

⁴ *Jahrbücher f. klass. Phil.*, iv. 309–418.

c24.s1.6

⁵ e.g. by Hertel, Cosquin, H. Lüders (*Buddh. Märchen*, p. xiii). Cf. G. d'Alviella, *Ce que l'Inde doit à la Grèce* (1897), pp. 138 ff.

c24.s1.7

¹ *Int. Wochenschrift*, iv. 995.

c24.s1.8

² ZDMG. lxii. 113.

c24.s1.9

Nor in any account can we omit to recognize the fact that in Märchen at least we may have old myths and that something must be allowed, as Grimm demanded, for the old common possession of the Indo-European people. In the tales of Herakles, Thorr, and Indra we have certainly some of this old mythology. More speculative is Kern's³ ingenious comparison of the ape king, who in a Jātaka makes himself a bridge for his following over the Ganges, and a similar exploit of the Irish king Bran, with which he suggests that the function of the Roman pontifex may be connected. We have accordingly a great field of possibilities; borrowing of India from Greece, of Greece from India, of both from a common source in Egypt or Asia Minor and Syria; common inheritance from Indo-European times, or from even further back if it is deemed worth while seeking to penetrate further into the past; and independent development due to the similar constitution of the human mind. In the face of these possibilities it will be found increasingly difficult to reach any clear decision in any particular case, while any general conclusion is absolutely out of the question. It must further be remembered that there must be admitted movements to and fro; a good story may be invented in Greece, pass to India, and return to Greece; Pausanias¹ already tells us before 180 A.D. of the snake who protected a child but was taken for its murderer and killed; it is difficult not to see in this the origin of the touching tale of the Brahmin who slays the ichneumon which had killed the snake attacking its master’s child, a legend which is famous in the form of Llewelyn and Gelert, a dog replacing the mongoose, and which can be traced widely over Europe.

c24.s1.10

³ *Gurupūjākaumudī*, pp. 93 f.

c24.s1.11

¹ x. 33. 9. Cf. Bloomfield, JAOS. xxxvi. 63 ff.

c24.s1.12

In many cases chronology is decisive against Indian influence on Greece being plausible. Thus a Corinthian vase² shows us the existence of the fable of the fox and the raven in the sixth century B.C., while in India we have the story of the fox and the crow only in the Jātaka and, therefore, of uncertain date. A painting by Polygnotos in the Lesche at Delphi of Oknos and his ass affords better evidence than the Jātaka tale of the ropemaker and the female jackal who undoes his work unperceived, both accusations of man’s industry and woman’s waste.³ Demokritos knew the story of the eagle who dropped the tortoise, which in India appears as the swans who let the same animal fall. The goat which swallowed a razor was the subject of a Greek proverb,⁴ and occurs in a Jātaka. The mice which eat iron in the *Pañcatantra* and a Jātaka are known already to Seneca and Herondas. The fable related of Daidalos in Sophokles’ *Kami-* *kioi*¹ is far better attested for Greece than for India in a late *Jātaka*. The claim that the account in Herodotos and Sophokles alike of a sister’s preference for a brother’s life to that of a husband, since she cannot have another brother, need certainly not be traced to a *Jātaka*, and the attempt to derive the delightful story of how Hippokleides lost his marriage by reason of his dance from the similar tale of the peacock in the *Jātaka* is curiously absurd. In these cases we have to do with ideas which would naturally enough develop themselves in men’s minds independently. Nor does there seem any conclusive ground for holding that the tale of the ass in the lion’s skin is older in either country. In the version in Greece the ass itself assumes a lion’s skin and is betrayed by the wind blowing it away; the Indian versions are more prosaic; the ass is given a skin by its owner to allow it to steal corn, and betrays itself by its cry.

c24.s1.13

² *Philologus*, lxxiv. 470. On classical fables, cf. Hausrath, Pauly-Wissowa, *Realencycl.*, vi. 1724 ff.; *Achiqar und Äsop* (1918); G. Thiele, *Neue Jahrbücher f. d. klass. Altertum*, xxi. 377 ff.

c24.s1.14

³ Pausanias, x. 29.

c24.s1.15

⁴ ZDMG. xlvii. 89 ff.; lxvi. 338.

c24.s1.16

¹ Zachariae, *Kl. Schriften*, pp. 108 ff.

c24.s1.17

The same doubt as to priority constantly occurs;² the story of the jackal which revealed its nature by its cry has a parallel in Phaedrus; so has the story of the ungrateful snake which bit its rescuer; the panther treats the goat as does the wolf the lamb in Phaedrus; the gods of Phaedrus who wish to drink up the stream have their parallel in the crows which would drain dry the sea; the motif of the bald-headed man and the fly, used with comic effect in Phaedrus, is turned to tragedy in the *Jātaka*; we find in Phaedrus the old tale of the eagle and the tortoise, and in India the swans in place of the eagle. The fable of the fox which compels the eagle to restore its young, which Archilochos knew, has been paralleled with a tale in the *Pañcatantra* of a crow and a snake, but the discrepancies are too great; nor is the parallel of the wolf, which a crane helps, in Phaedrus to the tale of the lion and the woodpecker sufficiently close to prove priority on either side.

c24.s1.18

² Günter, *Buddha*, pp. 52 ff.

c24.s1.19

Much that has been adduced definitely³ in favour of Greek priority is extremely dubious. The Trojan horse, however, is much older than the capture by an elephant of wood filled with soldiers of Udayana, but the *motif* is traced also in Egypt,⁴ and cannot be deemed too recondite to be original in India. The love of Phaidra for Hippolytos is striking, but the *motif* is found in the Jātaka¹ and belongs to human nature. The device of consoling the living for the dead by striking means is ascribed to Demokritos, is found in Lukianos, in Julian's letters, and in pseudo-Kallisthenes, but it also is attested by the Chinese version of the Tripiṭaka, which bids the mourner bring fire from a house where none has died. Androclus' grateful lion has an Indian parallel in the grateful elephant; Milo's death reminds us of the foolish ape in the *Pañcatantra*; India knows of paintings which deceive by likeness to life, as Parrhasios deceived even Zeuxis by his painted curtain. The tale of how an adulteress clears herself by a cunningly devised oath is early enough in India to have been deemed the source of Isolde's falsehood,² but we have the same idea in the oath of Ovid's Mestra.³ The effort to find in the tract *Physiologos* the proof of Indian influence on the western legend of the unicorn or the source of Caesar's tale of the elks of the Black Forest, which cannot rise if once they fall to the ground, is clearly a failure. From India may be borrowed the tale of the Charadrios, a bird which bears jaundice to the sun, but, as this idea is extremely early in India, it may be an ancient Indo-European belief.

c24.s1.20

³ e.g. Polykrates’ ring and the ring in the *Çakuntalā*; Surendranath Majumdar Sastri, *JBORS.* 1921, pp. 96 ff.; *Jāt.* 288.

c24.s1.21

⁴ v. d. Leyen, *Archiv f. d. Stud. d. neueren Sprachen*, cxv. 6.

c24.s1.22

¹ Bloomfield, tapa. liv. 145 ff.

c24.s1.23

² J. J. Meyer, *Isoldes Gottesurteil*, pp. 218 ff.

c24.s1.24

³ Rohde, *Griech. Roman*, p. 515.

c24.s1.25

In some cases more certainty of borrowing exists. The complex legend of Rhampsinitos in Herodotos, which he learned in Egypt, appears before a. d. 300 in India and can hardly be other than a borrowing there.⁴ But instances of this sort are rare and the issue of priority between India and Greece normally remains open. Little can be gained from general considerations such as the fact of belief in transmigration in India, the fondness of the Indian mind for romance, or the number of idle wanderers, religious men of various kinds, who went about India and perhaps beyond, telling and hearing tales. There seems to be no necessary connexion between beast fables and the belief in transmigration, for such fables exist among many peoples and represent a period when beast and human lives were not regarded as so distinct as they are in modern times; love of tales is recorded of others as of the people of Miletos, and wanderers of all kinds were evidently as common in the ancient as in the modern world. What presents much greater certainty is the actual translation of important Indian books and the transmission thus of much of fable and fairy tale to western lands, but that cannot be proved for an early date. It is difficult to believe we must¹ go to India for the idea of the gratitude of animals when we know that Agatharchos, a contemporary of Alexander the Great, told the tale of the dolphin, which rewarded kindness by saving during a shipwreck the life of the youth who bought him from some fishermen. On the other hand, it is not necessary to find in the Aīsopian fable of the fox which ate the heart of the deer killed by the sick lion and then denied that the beast had had a heart, the prototype of the jackal who ate the heart and ears of the ass and declared it never had them or it would not have been killed.

c24.s1.26

⁴ Frazer, *Pausanias*, v. 176 ff.; G. Paris, RHR. lv. 151 ff., 267 ff.; Huber, BEFEO. iv. 701 f.; Niebuhr, OLZ. 1914, p. 106.

c24.s1.27

¹ Cosquin, *Études folkloriques*, p. 21.

2. The Translations of the Pañcatantra

c24.s2.1

The enterprise of the physician Burzōe, who under Chosrau Anōsharwān (531-79) translated a version of the *Pañcatantra* into Pahlavi, was a work of the utmost importance for the Indian fable literature.² It is lost, but by A.D. 570 it was rendered by one Būd into Syriac, and about 750 an Arabic version was made by Abdallah ibn al-Moqaffaʻ from which the western versions are derived. The Syriac version is preserved in one manuscript and is imperfect, the Arabic is clearly expanded from the original, which seems to have consisted of five books corresponding to the *Pañcatantra*, five or eight other books taken from a different source³—whether or not the fusion was accomplished in India before Burzōe—and two books regarding his mission and his introduction. Of these fifteen chapters the Syriac has only ten, while the Arabic has twenty-two in all. The title of the work was clearly derived from Karaṭaka and Damanaka, the two jackals who figure in the first book of the *Pañcatantra*, variants of whose names occur regularly as the title of the translations, while the character of the work was somewhat altered by the inclusion of tales which were distinctly of a moral character.

c24.s2.2

² Hertel, *Das Pañcatantra* (1914); ZDMG. lxxii. 65 ff.; lxxiv. 95 ff.; lxxv. 129 ff.

c24.s2.3

³ From the *Mahābhārata*, xii. 138. 13 ff.; 139. 47 ff.; 111. 3 ff., three are taken; one is Buddhist (cf. A. Schiefner, *Bharatae Responsa* (1875) in Tibetan; Zachariae, *Kl. Schriften*, pp. 49 ff.); one the tale of the man in the well (see Nöldeke, *Burzōes Einleitung zu dem Buche Kalīla wa Dimna*, 1912); one of the lion and jackal also probably Buddhist; one of grateful beasts and ungrateful men; one of four friends, perhaps Buddhist; one of the mouse king and his minister is Indian in spirit.

c24.s2.4

From the Arabic version came in the tenth or eleventh century a fresh Syriac translation, and at the close of the eleventh century the Greek version of Simeon, son of Seth, which in its turn produced an Italian version of 1583 by Giulio Nuti, two Latin and one German versions, and various Slav reproductions. But more importance attaches to the Hebrew version of the Rabbi Joel (*c.* 1100), whence was made by John of Capua between 1263 and 1278 the *Liber Kelilae et Dimnae, Directorium vitae humanae*, of which two printed editions appeared in 1480. From a manuscript was made by Anthonius von Pforr the German translation, *Das buch der byspel der alten wysen*, which was repeatedly printed from 1483 onwards, and in addition to influencing deeply German literature was rendered into Danish, Icelandic, and Dutch. A Spanish version appeared in 1493, based on it, an Italian by Agnolo Firenzuola in 1546 which was translated into French in 1556, while a direct Italian version, that of A. Doni, came out in two parts in 1552, and the first part was translated into English by Sir Thomas North as *The Morall Philosophie of Doni* in 1570.

c24.s2.5

Another important translation was that made from the Arabic in 1142 or 1121 by Abu 'l-Maālī Naṣrallāh ibn Muhammed ibn ‘Abd al-Hamīd, for it produced the Persian *Anwāri Suhailī* by Ḥusain ibn ‘Alī al-Wā‘iẓ between 1470 and 1505, whence came numerous translations into eastern languages, and which became known in France in 1644 by the translation by David Sahid and Gaulmin; this, again, was soon rendered into English, German, and Swedish. Moreover, the Persian original was rendered into Turkish by ‘Alī bin Ṣāliḥ between 1512 and 1520, and it was rendered into French by Galland and Cardonne, the French then being translated into German, Dutch, Hungarian, and even Malay.

c24.s2.6

Other renderings from the Arabic were less fertile; the Hebrew version of Jacob ben Eleazer in the thirteenth century is only in part preserved, the old Spanish version (*c.* 1251) and John of Capua's work afforded material to Raimundus de Biterris who prepared his *Liber de Dina et Kalila* for Johanna of Navarre. The Italian Baldo in the early twelfth century used some version for his *Novus Esopus*. La Fontaine in the second edition of his *Fables* in 1678 expressly states that the greater part of his new matter is derived from the Indian sage Pilpay, in whose name we may recognize the Sanskrit Vidyāpati, lord of learning.

3. The Çukasaptati

c24.s3.1

Another case of translation which is certain is that of the *Çukasaptati*, whose existence, as we have seen, is attested by Hemacandra in the twelfth century when he cites an episode, not in our texts, in which the parrot is caught by a cat, proving probably that variant recensions were already in existence. By the beginning of the fourteenth century there already was extant a rude Persian version which displeased the refined taste of Nachshabī, a contemporary of Hāfiz and Saʻdī, who in 1329–30 produced the *Tūtīnāmeh*,¹ which a hundred years later was rendered into Turkish and in the eighteenth century evoked a fresh version by Kadiri. The *Tūtīnāmeh* rejected part of its original as unsuitable, substituting other tales partly from the *Vetālapañcaviṅçatikā*. From the Persian version many tales passed to western Europe *via* Asia, and one of the tales was made famous by Gottfried's *Tristan und Isolde*, in which occurs the account of the ordeal which was used to deceive by proving Isolde's innocence. In India the episode is old, for it occurs in a Chinese fifth-century version of an Indian tale and in a confused form is extant in the Jātaka book.²

c24.s3.2

¹ Pertsch, ZDMG. xxi. 505–21. The Persian of Kadiri was translated by C. J. L. Iken (1822), the Turkish by G. Rosen (1858).

c24.s3.3

² Chavannes, *Cinq cents contes*, i. no. 116; Jātaka, 62; Zachariae, *Kleine Schriften*, pp. 282 f.; J. J. Meyer, *Isoldes Gottesurteil*, pp. 74 ff.

4. Other Cases of Contact between East and West

c24.s4.1

Tales which cannot be traced thus definitely to Indian sources may yet readily be assumed to have reached the west from India in view of these proved facts. Nor is it difficult to imagine modes of transmission³; apart from literature, tales pass easily enough from mouth to mouth, and the Crusades resulted in prolonged intercourse between Christians and Mahomedans. Then the Arab rule in Spain served to mediate between the civilizations of west and east, and the Jews in their turn played an important part as intermediaries. The influence on the Mongols in this regard has been exaggerated by Benfey, but doubtless under-estimated by Cosquin.¹ There is no reason to doubt that the Gipsies² helped to spread tales, as their Indian origin is well established. Byzantine literature,³ again, must have been a factor in the literary diffusion of stories. But it would be absurd to assume that the borrowing was all from one side, as Benfey was inclined to do as regards fairy tales. Cosquin has, indeed, done much to defend this thesis by his efforts to prove that the better-motived tales are often Indian; Lang, with various qualifications, and Bédier have insisted instead on the independent generation of tales in different places, while Antti Aarne has endeavoured to work on the basis that every country may produce tales, but these tales wander far and wide, so that the end of research is to establish *motifs* which belong to one country or another; thus a group of ideas centring in a magic ring is Indian in origin, another dealing with three magic substances is British and French, another centring in a magic bird is Persian. In most cases it may be frankly admitted that it is extremely difficult to achieve any satisfactory result.

c24.s4.2

³ For the period to A. D. 600 see Kennedy, JRAS. 1917, pp. 226 ff.

c24.s4.3

¹ Cosquin, *Études folkloriques*, pp. 497 ff.

c24.s4.4

² Wlislocki, ZDMG. xli. 448 ff.; xlii. 113 ff.

c24.s4.5

³ E. Kuhn, *Byzant. Zeitschrift*, iv. 241.

c24.s4.6

A certain degree of assurance may be felt regarding the familiar tale of Sinbad. The Arabian historian Masūdi, who died in 956, expressly ascribes to the *Kitāb el Sindbād* an Indian origin; this work corresponds to the Persian *Sindibādnāmeh*, the Syriac *Sindban*, the Arabic ‘Seven Viziers’ which is found in manuscripts of the *Arabian Nights*, the Hebrew *Sandabar*, the Greek *Syntipas*,⁴ and a mass of European tales. The plan of the work is taken from the *Pañcatantra*; a king entrusts his son to a wise man who undertakes to teach him wisdom in six months; the Indian *motif* of telling tales to save the life of some one, here a prince condemned to death, is found, and the stories have usually Indian parallels; that of the ichneumon is taken from the *Pañcatantra*, and the others are often specimens of women’s tricks to cover their infidelities, which are common in India, forming as it were a supplement of the *Pañcatantra*. The Greek *Syntipas* contains various passages which can only be read successfully by recognizing that they are merely corruptions of a Sanskrit original, and everything supports the conclusion that we have here another case of an Arabic original rendered from a Pahlavi translation of a Sanskrit text.

c24.s4.7

⁴ H. Warren, *Het indische origineel van den Griekschen Syntipas*; Hertel, ZDMG. lxxiv. 458 ff.

c24.s4.8

It is natural to extend the doctrine and to find the original of the Arabian *Thousand and One Nights* in India,¹ and something substantial has been done in this direction by proving that the prologue and setting of the tales are a contamination of *motifs* which are quite well known in India. Thus we have the Jain legend of Kanakamañjarī, who retains for six months the undivided love of the king by the device of beginning a tale each night but not finishing it. Again, we have in a Chinese rendering of a Buddhist tale (A.D. 251), in the *Kathāsaritsāgara*, and in Hemacandra, variations of the theme of the man who is utterly depressed by finding out that his wife is unfaithful, but recovers happiness because he discovers that the king himself is equally being made a mock of. The further adventure of Shahriar and Shahzeman has a parallel in the *Kathāsaritsāgara*. There are other traces of Indian influence in the tales, and it is clear that it is impossible to ascribe them to borrowing from Persia; translations from Persian into Sanskrit are normally late, as in Çrīvara’s *Kathākautuka*² written on the theme of Yusuf and Zuleikha under Zainu-l-‘Abidīn in the fifteenth century. The only matter that can be in doubt is the extent of the influence; certainly there is nothing in this case to prove the taking over of a whole cycle of stories from an Indian work, now lost.

c24.s4.9

¹ Cosquin, *op. cit.*, pp. 265 ff.; Przyluski, JA. ccv. 101 ff., who finds in the Svayaṁvara of India a relic of the Austroasiatic festival dance at which young people were mated. Cf. Macdonald, JRAS. 1924, pp. 353 ff.

c24.s4.10

² Ed. and trans. R. Schmidt (Kiel, 1898).

c24.s4.11

In Europe, apart from the translations enumerated, traces of real Indian origin are hard to prove.³ A Carolingian poem of the ninth century tells how a hunter slew a boar, was himself killed by and caused the death of a snake, which is a feeble tale in comparison with the Indian story of the greedy jackal who was lucky enough to find a hunter who had killed a deer, and had also slain a boar which killed him, but meets death through eating first out of the spirit of thrift the bowstring. Peter Alfonsi (twelfth century) knows a tale which occurred in Barzōe's introduction to his version of the *Pañcatantra* and some other Indian narratives, but merely as handed down in Arabic. Walter Mapes' knowledge is doubtful, but Marie of France has clear parallels, and the bird of St. Martin recounted by Odo of Sheriton (*c.* 1215), which held up its limbs to keep up the sky but appealed in terror to the saint when a leaf fell on it, can be traced to the *Mahābhārata* and the *Pañcatantra*. Nigel of Canterbury's knowledge (*c.* 1180) of the tale of the ingratitude of man as contrasted with that of animals is not necessarily borrowed, nor is the *motif* of the fatal letter and its bearer in Saxo Grammaticus probably Indian, seeing that we have already the conception in Homer. James of Vitry, bishop of Ptolemais, a Crusader, in his *Exempla* tells from hearsay the stories of the Brahmin who was cheated by rogues, of the Brahmin who built castles in the air, and of the son who was going to bury his too long-lived grandfather, while his own son prepared a grave for him. In the *de diversis rebus praedicabilibus* of the Dominican Étienne of Bourbon, who died *c.* 1260, we find a version of the story of the blind and the lame, well known in Jain texts,¹ and a variant of the judgement of Solomon² in which two women dispute over a ball of wool and the issue is decided by asking what was the kernel used on which to wind the material³; the Indian tale, found in a Chinese version, in Buddhaghosa, and the *Çukasaptati*,⁴ of his stepfather's device of ridding himself of the Bodhisattva appears in Étienne as the tale of the page whose prince, suspecting him of an intrigue, sends him to the workers at his oven who have instructions to fling into it the first who comes with a royal message. Étienne also tells us of the innocent hound, transmuted into St. Guinefort and an object of worship, whose tomb he insisted on destroying. The *Gesta Romanorum* contains various stories which may be of Indian origin; one in a manuscript of 1469¹ is so elaborate as to leave no doubt of its origin, for it tells of how a knight who was taught in gratitude the language of the beasts managed to escape revealing it to his wife, a famous Jātaka tale. On the other hand it is impossible to ignore independent development; if Heinrich Seuse (*c.* 1330) illustrates the idea of eternity by telling of a bird which once in 100,000 years picks up a grain of corn from a millstone of the size of the earth (the period until the stone shall be made bare is but a moment in comparison with eternity), it is far-fetched to claim derivation from the Indian conception of a world age as longer than the period taken by a man who once in a hundred years rubs a mountain with a silk cloth to level it with the ground.

c24.s4.12

³ Günter, *Buddha*, pp. 99 ff. The famous tale of the poison maiden in Indian literature and in the west—told of Aristotle and Alexander in the *Secretum Secretorum* (cf. Hawthorne, *Rappaccini’s Daughter*), is discussed by Penzer, *Ocean of Story*, ii. 311 ff.

c24.s4.13

¹ Hertel, *Geist des Ostens*, i. 248 ff.

c24.s4.14

² Cf. Hertel, *loc. cit.*, 189 ff. on the issue of the ultimate original of the Indian versions of 1 Kings, iii. 16; *Jātaka*, 546.

c24.s4.15

³ Zachariae, *Kl. Schriften*, pp. 84 ff.

c24.s4.16

⁴ Cf. the legend of St. Elizabeth of Portugal, Cosquin, *Études folkloriques*, pp. 73 ff., who (p. 160) insists on the exchange of persons or messages as distinguishing these tales from such cases as Bellerophon.

c24.s4.17

¹ Günter, *Buddha*, pp. 122 ff.

c24.s4.18

From the late middle ages comes evidence of the borrowing of several stories of cleverness, as in the story of the man who finds out guilty servants by more or less accidental observations made at table.² The seven-league boots of fairy tale are found in the *Kathāsaritsāgara* and may be Indian, but many other *motifs* are hardly to be assigned to one nation; thus we have the hero who is vulnerable in one spot only much earlier in Greece than in India and independently probably in Germany; the tree which yields what it is asked for depends on the widespread belief in tree spirits; the man or animal which yields gold attests, though early in India, to community of ideas rather than borrowing; the burning of a skin which frees the enchanted prince seems ethnic. Various peoples know of flying birds which carry heroes on long journeys. Circe in the *Odyssey* need not be the source of the Yakṣiṇī of the tale of Niçcayadatta in the *Kathāsaritsāgara*.³

c24.s4.19

² Cf. Forke, *Die indischen Märchen*, pp. 36 f.; Zachariae, *op. cit.*, pp. 138 ff.

c24.s4.20

³ See Tawney's trans., i. 337 ff.; cf. the Sirenes of *Od.*, xii. 39 ff. and Jātakas 41, 96, 196, 439; *Mahāvaṅsa* (Geiger, p. 25).

c24.s4.21

Of interesting *motifs* due to India Cosquin⁴ offers a good example in the *Mahosadha Jātaka* tale of how a faithful wife served gallants who sought to seduce her in her husband's absence, ending up with conveying them in jars before the king, an episode which is proved ancient by a relief at Bharhut on which is depicted the opening of three coffers in each of which is a prisoner. The story is preserved in perhaps a more original fashion in the Kashmirian *Bṛhatkathā* legend of Upakoçā, who induces the gallants to take a bath and has them blackened by a sticky preparation, in which condition they are revealed to the king. It seems difficult to doubt that this is the source of the inferior version in the fable of Constant du Hamel and Isabeau in the thirteenth century. A variant of the same idea appears in the story of Devasmitā in the Kashmirian *Bṛhatkathā*, and it is probable enough that we must seek an Indian original for the form of the legend as it appears in the *Gesta Romanorum* (*c.* 1300), in the romance of Perceforest, and in the fifteenth-century English poem, *The Wright's Chaste Wife*. It is tempting no doubt to find¹ in the common idea of the ogre and the fascinating daughter who helped the lover to deceive her father, who despite his wickedness is stupid, the result of the Indian idea preserved in the Kashmirian *Bṛhatkathā* of the youth who is aided by the daughter of a Rākṣasa whose stupidness she asserts is due to his origin, to win her hand by accomplishing all the impossible feats set to him. But proof is wanting. Another tale,² which has a fair chance of being Indian in origin, is the type of the boiling cauldron and the pretended lack of skill as in the case of Vikramāditya, who is saved by the warning given by a skull from the ruse of a Yogin who bids him turn round a cauldron into which he means to fling him; the king asks to be shown how to act and slays the miscreant by his own device. The tale³ of the cat who held a candle for the king but at last lets it go at the sight of a third mouse, though he has permitted two to pass unnoticed, may be of Indian origin, but that is clearly not proved; it is, however, probable that the idea of the vigil of Solomon and Marcolphus, well known in the fourteenth century, is due to India, where the tale of Rohaka⁴ and the king of Ujjain is known in the twelfth century and that of Pradyota and a Gandhāran is found in the Kanjur, in the ninth century. Nor is the idea of the magician and his apprentice¹ who assumes all sorts of forms to emerge from different impasses unique; the legend of Mestra in Ovid² shows that tales of this sort could easily arise in independence of India, where indeed the *motif* is not specially important.

c24.s4.22

⁴ *Études folkloriques*, pp. 457 ff.

c24.s4.23

¹ Cosquin, *op. cit.*, p. 25.

c24.s4.24

² *Op. cit.*, pp. 349 ff.

c24.s4.25

³ *Op. cit.*, pp. 401 ff.

c24.s4.26

⁴ On him cf. Zachariae, *op. cit.*, pp. 66, 94 f., 190; Pullé, *Uno progenitore Indiano del Bertoldo* (1888).

c24.s4.27

¹ Cosquin, *op. cit.*, pp. 497 ff. For other suggestions see *Les contes indiens et l'occident* (1922), where *inter alia* he deals with the slipper *motif* in India.

c24.s4.28

² *Met.*, viii. 847 ff.

5. The Romance in Greece and India

c24.s5.1

It is natural that efforts should have been made to prove the derivation from Greece of the Indian romance, as it appears seemingly full-fledged in the works of Subandhu, Bāṇa, and even in some degree Daṇḍin. Peterson's³ argument for Greek influence, strictly limited in scope, was based in part on the indubitable fact of Greek influence on astronomy and astrology, and in part on the new spirit which he discerned in the romances, which clothed with flesh and blood the dry bones of the simple tale with its rapid but monotonous stream of adventures. He quoted, however, in support of his view merely passages illustrating the affection of Achilles Tatius in his tale of Kleitophon and Leukippe for minute descriptions of the beauty of the beloved, the effect of love upon man, and the love which other things have for each other, citing the story of the affection of the male palm for the female palm, which is given fruition by the grafting of a shoot into the heart of the male. To this Reich⁴ has added merely a list of similarities; thus we find both in Indian and in Greek romance the conception of love at first sight, of lovers revealed to each other in a dream, the swift change of fortune from good to evil and then back to prosperity, adventure and shipwreck at sea, heroes as well as heroines of wonderful beauty, free use of detailed description both of love and of nature. All these things may be admitted, but clearly they do not prove borrowing, though they render it possible. The tale of the loves of the palms, it is clear, suggests Syria rather than either Greece or India; it is decidedly different from the Indian wedlock of the mango and the jasmine recorded in the Kāv Yas.

c24.s5.2

³ *Kādambarī*, pp. 98 ff.

c24.s5.3

⁴ DLZ. 1915, pp. 553 ff., 594 ff.

c24.s5.4

More definite evidence is adduced by Rohde¹, and by Weber² who holds that we find the *motif* of the *Vāsavadattā*—which, it must be remembered, has no known antecedent in India—in a tale recounted by Athenaios on the authority of Chares of Mytilene, an official of Alexander the Great. This tale of Zariadres and Odatis contains the *motifs* of lovers who see each other in a dream, and are finally united through the intervention of the maiden’s marriage ceremonial in which she enjoys the right of choice. But even if we compare the awakening of Vāsavadattā at her lover’s embrace to the story of Pygmalion and Galatea, and find parallels in the Greek romance for armies which war for the possession of a maiden, we have the fact that the tale admittedly in the Greek version is not Greek, and in point of fact in Firdausī we learn that the daughter of the emperor of Rome sees her lover Gushtasp in a dream and herself claims him as husband. The choice of a husband in this way by a princess is an early Indian practice, and the Persian tale may easily have come from India in the first place.

c24.s5.5

¹ *Griech. Roman*², pp. 47 ff.

c24.s5.6

² IS. xviii. 456 ff.

c24.s5.7

A different aspect was given to the hypothesis by F. Lacôte,³ when he claimed that Guṇāḍhya himself was under Greek influence, thus departing from Peterson’s contrast between the predecessors of the romances and these works. But his opinion later⁴ changed, and he adduced evidence in favour of the borrowing of the Greek romance from India. Of his evidence, part may be at once dismissed as being irrelevant to the question of origin, as it concerns merely incidents and might therefore be borrowed without the romance as a whole being adopted by Greece from India. In any case, however, these details seem inadequate to prove their case; the plant which cures wounds in three days has been compared to the *vraṇasaṁrohaṇī* plant of India, but it belongs to the most primitive period of Greek as well as Indian medicine. The unwinking eyes and feet that touch not the ground which mark out the gods from men is Indian, but the latter detail at least is recognized by the artists of the Roman Empire, and Kalasiris shows that the *Iliad*⁵ was believed to be the authority for both the assertions.

c24.s5.8

³ *Essai sur Guṇāḍhya*, pp. 284–6.

c24.s5.9

⁴ *Mélanges Lévi*, pp. 272 ff. See Keith, JRAS. 1915, pp. 784 ff.

c24.s5.10

⁵ N. 71 f.; A. 200, which prove that in gait (cf. Vergil: et vera incessu patuit dea) and eyes gods revealed their divinity.

c24.s5.11

When Theagenes and Chariklea see each other for the first time, they seem to recognize each other as if they had known each other before; this is not merely a common feeling among modern people, but Plato had a doctrine of recollection which was far more likely to be present to a Greek author than an Indian romance *motif*. In the general purpose of the romances there is absolutely nothing un-Greek. On the contrary, the *Aithiopika* justifies the trials of its hero by the doctrine that he and his beloved had to be brought almost to death in order that the Aithiopians might cease to practice human sacrifice. The fate that elsewhere governs the progress of events is essentially Greek, more Greek than Indian, and it is most significant that nothing is said of the misfortunes which fall on the heroes being due to evil deeds done in past lives. Moreover, it is striking that in all the complex adventures recounted in the Greek romances we do not have Indian scenes or episodes, though there was abundant room for them, and the authors of the romances were largely themselves Orientals, not natives of Greece proper.

c24.s5.12

There remains, therefore, the argument from form. Lacôte contends that the Kathā form was original in India, that there alone did it develop, and that it was borrowed by the Greek romances from India. Every part of the proof is defective. The Kathā manner in its simpler forms is the most natural¹ of all, and Lacôte admits that we have it in the *Odyssey*, but he holds that it was not developed in Greece. Of this there is no proof whatever; the dialogues of Plato, which are reported conversations, he admits to be exceptions to his rule, but holds that the manner was confined to philosophy, which borrowed it from the Mimes of Sophron. This is a very implausible assumption, and is further contradicted by the evidence. We know of the love of Greece for tales, the story-tellers of Sybaris and Ephesos were famous, there is the evidence of Apuleius, who refers to his *Metamorphoses* in the words *ut ego tibi sermone isto Milesio varias fabulas conseram*.² It is a perfectly fair deduction to make from this definite statement that the Ephesian tales known to Apuleius—including doubtless Aristeides' *Ephesiaka* which were rendered by Sisenna,¹ already exhibited the form of a framework story with reports of experiences of the actors inserted. In Ovid’s *Metamorphoses* (v) Pallas’s adventures include meeting and hearing tales from the Muses, in whose account of Demeter and Proserpina are inserted two narratives by Arethusa; in xiv in Aeneas’ adventures we have Macareus’ narrative to Achaemenides, in which is inserted a tale by a maiden of Circe’s. We have, therefore, no conceivable need to seek in India for the prototype, especially as chronology is all against the suggestion. We know nothing that we can prove of the actual manner of the *Bṛhatkathā* and its date is utterly uncertain, assuredly not early enough to make dependence even possible,² while as regards the *Vāsavadattā* we know that it is later than any extant Greek romance of the period dealt with by Lacôte. A further insuperable difficulty would be the fact that Lacôte thinks of popular transmission, recognizing that no Greek could understand a real Indian romance in Sanskrit, while such transmission would certainly give only tales, not the elaborate construction which is the one point which could be used to prove derivation.

c24.s5.13

¹ It is found early in Egypt, and the emboxing of stories there is very early; Maspero, *Contes populaires de l'Égypte ancienne* (1906), pp. 23 ff.

c24.s5.14

² Teuffel-Schwabe, *Rom. Lit.*, § 367; H. Lucas, *Philologus*, 1907, pp. 29 ff.

c24.s5.15

¹ Teuffel-Schwabe, § 156.

c24.s5.16

² There is Apuleius’ own work (*c.* A. D. 160) and Lukianos’ *Λούκιος ἢ ὄνος*, as well as Petronius’ *Satirae* (Teuffel-Schwabe, § 305), and above all Ovid’s work. The *Arabian Nights* manner, as Tyrrell (*Latin Poetry*, p. 123) says, is not a great success there.

c24.s5.17

In point of fact there is no general agreement in the Greek romances as regards form; it would have been strange if there had been, for Greek writers are usually successful in achieving originality. Heliodoros sometimes relates himself the tale, sometimes brings the actors before us in conversation to tell of their doings, just as does Homer; Xenophon is a simple narrator; Achilles Tatius puts his tale into the mouth of Kleitophon, but the latter relates it as if he were an outsider, recording what happened to himself and to the heroine impartially. It is only in Antonius Diogenes that we find anything more complex. There the story opened with a letter from the author to his sister, sending her a copy of a letter from Balagros to Phila, enclosing a note drawn up by one Erasinides of a conversation between Deinias and Kymbas. Deinias’s narrative consists largely of a story told to him by Derkyllis, in which are inserted reports to Derkyllis by Astraios and Mantinias, and again by Astraios to Derkyllis and Mantinias; at the close of Derkyllis' narrative Deinias reports what he heard from Azulis, and the close of Deinias's conversation with Kymbas is followed by the final note of Erasinides. This is complex, no doubt, but a perfectly natural development, just as the Indian Kathā in *Vāsavadattā* is a natural development from simpler forms. The further parallels drawn by Lacôte are invalid; the letter from the author to Faustinus, which seems clearly to have been in an appendix, is only remotely similar to the introductions to the *Vāsavadattā* and Bāṇa's works, while the statements at the head of each book of the work as to stories parallel to the marvels he relates have no real resemblance to the introductory verses prefixed to each chapter of the *Harṣacarita* only, which, it may be added, is not in the slightest degree in form like the work of Antonius. Nor, it must be admitted, is it altogether reasonable to ignore the fact that, while the Greek romances are silent as to India, the existence of Yavanas and their cunning, especially in the fabrication of aerial ships, is referred to in the Kashmirian *Bṛhatkathā*, which knows their skill in architecture, and Budhasvāmin attests the use of Greek beds, suggesting that even the original *Bṛhatkathā* may have known of the Greeks as cunning and skilled craftsmen.¹

c24.s5.18

¹ Cf. Lacôte, *op. cit.*, p. 286. The existence of a Greek and Eurasian population in Gandhāra for a couple of centuries at least (Foucher, *L'Art Gréco-Bouddhique du Gandhāra*, ii. 448 ff.) cannot be ignored.

c24.s5.19

Denial of any relation of interdependence is also asserted by L. H. Gray,² who calls attention to many parallels, letters between lovers, long-winded lamentations, threats of suicide, the stories within stories, descriptions of nature, detailed personal descriptions, learned allusions and citations of precedents, even strained compounds, and alliterations, parisoi,³ homoioteleuta, and other figures of rhetoric which recall the Sanskrit Anuprāsa and Yamaka. But he insists that the least part of the Sanskrit romance is the thread of the story or the adventures of its characters; all the stress is laid on rhetorical embellishment, minute description of nature, detailed characterization of exploits, and of mental, moral, and physical qualities. In the Greek romance, on the other hand, the essence is the narrative of one improbable adventure after another, fine writing is practically discarded, description and appreciation of nature are essentially neglected. To the latter assertion there is of course admitted an exception in the case of the *Poimenika* of Longus, but that author derives directly from Theokritos, Bion, and Moschos, while the Sanskrit romance owes its love for nature to Indian feeling. The *Daçakumāracarita* with its affinities to the picaresque romance is without real parallel in the Greek romances, though it has affinities to the *Satirae* of Petronius.

c24.s5.20

² *Vāsavadattā*, pp. 35 ff. Cf. G. N. Banerjee, *Hellenism in Ancient India*, pp. 258 ff.

c24.s5.21

³ Cf. Aristotle, *Rhet.* iii. 10 ff.

c24.s5.22

An interesting parallel is drawn by Gray between the manner of Lyly in his *Euphues* and that of Subandhu. They agree in laying all stress on form rather than subject-matter, though Lyly has a didactic end foreign to Subandhu. Lyly employs the device familiar in India of emboxing a story within a story, as in the case of the tale of Callimachus, which itself includes the story of the hermit Cassander. Moreover, his paronomasias, his alliterations, his antitheses, and his learned allusions are in close harmony with the Indian practice. The instance is valuable as a reminder that parallels may arise without borrowing on either side.

6. The Hexameter and Indian Metre

c24.s6.1

An interesting suggestion has been made by Jacobi¹ that the Dohā metre of Apabhrañça, with which may be compared the Dodhaka metre of Classical Sanskrit poetry, in so far as both are essentially originally dactylic in structure, is to be traced back to the Greek hexameter, the Dohā being the result of combining two hexameters into a stanza and then dividing it in the usual Indian manner into four lines. The Ābhīras, he contends, were situated in Gandhāra and the neighbourhood during the period of the influence of the Greco-Bactrian kings, and they must have eventually felt the need for a rendering into an Indian speech of the Homeric poems which, as Dio² tells us, the Greeks loved so dearly, and clung to even when they had lost much else of Hellenic character. The version of Homer thus made for the educated classes would probably be in the metre of the original, and thus the Dohā would grow up as the peculiar metre of the Ābhīras and would cling to Apabhraṅça poetry. A parallel may be seen in the great influence exerted on Bengal prose literature by the missionaries of Serampore.

c24.s6.2

¹ *Festschrift Wackernagel*, pp. 127 ff.

c24.s6.3

² *Or.* liii. 6. On the amount of Greek known in India cf. Kennedy, JRAS. 1912, pp. 1012 ff.; 1913, pp. 122 ff.; 1917, pp. 228 ff.; Thomas, 1913, pp. 1014 f.

c24.s6.4

Jacobi's theory rests naturally on the validity of the assertion of Dio that the Indians had a translation of Homer, which is repeated by Aelian, who asserts the same of the Persian kings, and who may have used the same source as Dio, although it is possible that he merely copies the latter. The general view¹ that Dio's reference is really to the *Mahābhārata* as the Indian equivalent of Homer is possible, but there is no doubt that it is not proved. Jacobi strengthens his case by pointing out that from the later sculpture of India we should never be able to demonstrate Hellenistic influence, were it not for the Gandhāran art, which being permanent has survived to testify to the strength of Greek art, and it might be added that the proof of the influence of Greek painting has probably been lost through the disappearance of the frescoes which once existed in abundance in Gandhāra.² But, granting that the tale of Dio may have foundation, it must be admitted that it does not seem possible to accept as even probable the origin suggested for the Dohā; the dactylic form is easy to explain independently. It must, however, be said that the effort of Leumann³ to reconstruct an Indo-European metre with a quantitative basis, of which the Dohā would be a descendant, is clearly a mere *tour de force*, resting on utterly inconclusive evidence.

c24.s6.5

¹ Weber, IS. ii. 161 ff.

c24.s6.6

² Cf. Foucher, *L'Art Gréco-Bouddhique du Gandhāra*, ii. 402 f.

c24.s6.7

³ *Festschrift Wackernagel*, pp. 78 ff. and elsewhere. His work is vitiated by a complete failure to weigh evidence and inability to meet criticism. By his methods anything could be proved. Meillet and Weller (ZII. i. 115 ff.), whom he attacks, are far sounder.

Linked texts (16)