Mauryan emperor of India (ca. 272-231 B.C.).


AŚOKA, Mauryan emperor of India (ca. 272-231 B.C.).

i. The life and teaching of Aśoka.

ii. Aśoka and Iran.

iii. The legend of Aśoka in Khotanese.

iv. The legend of Aśoka and the founding of Khotan.


i. The Life and Teaching of Aśoka

Aśoka was the most powerful king and an inspired leader of ancient South Asia, who for at least fifteen years proclaimed rules of social behavior which were both humane and practicable.

Aśoka’s regnal years and the extent of his empire are known approximately. According to the tradition recorded in the Pali chronicles of Ceylon Aśoka ruled four years before and 37 years after his consecration in the year 218 after the Lord Buddha’s Nirvāṇa (ca. 486 B.C.). Aśoka’s authority extended over most of the Indo-Pakistani subcontinent, excluding its southernmost portion (and probably parts of Sindh and Bengal) but including considerable areas of present Afghanistan.

Aśoka was a son of Bindusāra (ca. 300-272 B.C.) and a grandson of Candragupta (ca. 324-300), the founder of the Maurya empire. He at first reigned ruthlessly like most of his predecessors had done, but a campaign against Kaliṅga (present Orissa) in his ninth regnal year brought about a complete change in Aśoka, as he himself admitted in his thirteenth Rock Edict four years later. On this occasion he emphasized that military victory is worthless; the only true victory is the victory of (or by) Dhamma (Pali: righteousness; Skt. dharma). It is clear that Aśoka was converted to Buddhism some time later and, from his thirteenth regnal year, began a vigorous drive to propagate his ideas on Dhamma, a true ideology in fact, by means of edicts engraved on rocks and pillars in different parts of the empire, especially along the main highways, at crossroads and other centers of economic activity. The versions of each individual edict show differences in script, language or dialect, (which proves that Aśoka’s inscriptions were intended for large sections of the population), and details of formulation, although the contents are basically the same. Four scripts are used: (1) brāhmī, (of Indian origin); (2) kharoṣṭhī, based on Aramaic but with additional letters and vowel marks; (3) Aramaic in the bilingual, Greek and Aramaic, Minor Rock Edict of Ṧār-i Kuna (Kandahar), as well as in two fragmentary inscriptions of Taxila and Pul-i Darunta (near Jalālābād, Afghanistan); (4) Greek (script and language) in the above-mentioned bilingual inscription and in fragments of Rock Edicts XII and XIII at Kandahar. The brāhmī and kharoṣṭhī inscriptions are in four different Prakrit dialects, depending on their location.

The most consistent single theme running through the edicts is that of social obligations. Children are admonished to obey their elders, subjects to be loyal to the king and his representatives, while all people owe respect to brahmans and ascetics, tolerance for different creeds or customs and non-violence (ahiṃsā) towards the creatures in general. This prohibition of killing living beings was, however, by no means absolute. It was apparently principally directed against certain Hindu rituals involving the slaughter of numerous animal. On the other hand, Aśoka clearly did not enjoin vegetarianism on his subjects.

It does not seem as though Aśoka’s ideology was meant to reform Indian society to any clearly defined pattern. Its aim was more probably that of laying down standards of social behavior which were regarded as proper by most articulate Indians of those times. The present tendency among scholars is to regard Aśoka as above all a practical statesman who propagated his Dhamma to supply a common ideology to the numerous peoples, tribes, and other groups in his empire. This alone could give a true meaning to the political unification of the subcontinent and create a lasting empire. Although Aśoka was apparently unsuccessful—the Maurya empire disintegrated soon after his death—it is likely that he contributed more than any other political figure of ancient India to the gradual formation of an Indian identity.



General: E. Lamotte, Histoire du bouddhisme indien des origines à l’ère Śaka, Louvain, 1958, pp. 244-384, 789-98.

K. A. Nilakanta Sastri, ed., Comprehensive History of India II, London, 1959, pp. 29-43.

R. Thapar, Asoka and the Decline of the Mauryas, Oxford, 1961.

R. K. Mookerji, Asoka, 3rd ed., Delhi, 1962.

Inscriptions: E. Hultzsche, The Inscriptions of Asoka, Corpus Inscriptionum Indicarum I, new ed., Oxford, 1925.

J. Bloch, Les inscriptions d’Asoka, traduites et commentées, Paris, 1950.

D. C. Sircar, Inscriptions of Aśoka, 2nd ed., Delhi, 1967.

B. M. Barua, Asoka and his Inscriptions, 3rd. ed., 2 vols., Calcutta and New Delhi, 1968-69.

U. Schneider, Die grossen Felsen-Edikte Aśokas, Wiesbaden, 1978.

F. R. Allchin and K. R. Norman, “Guide to the Aśokan Inscriptions,” South Asian Studies. Journal of the Society for South Asian Studies 1, 1985, pp. 43-50.

See also Camb. Hist. Iran III/2, pp. 950-51, 1180, 1236.

(J. G. De Casparis)

ii. Aśoka and Iran

In 305 or 304 B.C., Seleucus I Nikator ceded by treaty to Candragupta Maurya all the territories situated south of the Hindu Kush, conquered first by the Achaemenids and later by Alexander: Arachosia (the region of Kandahar), parts of Paropamisadae (the mountainous country south of the Hindu Kush; the frontier would have run through Laghman were inscriptions of Aśoka were found), a part of Aria (to the south of Herat?) and of Gedrosia (probably Balūčestān), Punjab and Sindh (A. Foucher, Vieille route de l’Inde II, Paris, 1947, pp. 208-09, 271). Henceforth, these territories, first attached to the Achaemenid empire under Cyrus the great and Darius and Achaemenid satrapies for two centuries, belonged to an Indian dynasty established in the valley of the Ganges. Both Candragupta, who began his march to the empire there, and Aśoka, who might have been viceroy in Taxila when young, knew this region very well. It is a remote area open to Iranian influences and which even included populations that were clearly Iranian, like the Kamboja, who are named in Aśoka’s Rock Edicts V and XIII and whose location should be sought somewhere between Kabul and Kandahar, for instance in the neighborhood of Ḡaznī (Foucher, op. cit., p. 271; E. Benveniste, JA, 1958, pp. 46-48).

In their Indian satrapies, the Achaemenids had established a bureaucracy using Iranian language and Aramaic script. The influence of this bureaucracy was so lasting that seventy years after the fall of the Achaemenid dynasty, Aśoka felt the need to have his edicts translated in “Arameo-Iranian” (Humbach’s “Aramäoiranisch”). In Taxila, Pol-e Darunta, and Kandahar there existed bilingual translations of Aśoka’s edicts in Aramaic script, where each group of Indian words was first translated in Arameo-Iranian and then followed by the original Indian words written in Aramaic script. At the end of each translated passage the word SHYTY is added, probably Iranian *sahyatai “it is spoken,” cf. Old Persian aθahya “it was said” (H. W. Bailey apud S. Shaked, JRAS, 1969, pp. 118-22; already G. Morgenstierne apud H. Birkeland, Acta Orientalia 16, 1937, p. 233 n. 1). In Kandahar, there also existed an Arameo-Iranian/Greek bilingual resume of the first Rock Edicts. In Laghman two short rock inscriptions, a kind of milestones, in Arameo-Iranian were discovered. One of them seems to mention Tarmid (Termeḏ) and proves the existence of a route connecting India and Bactria at this time (J. de Menasce in Israel Oriental Studies 2, Tel Aviv, 1970, pp. 290-92).

Thus, the Achaemenid bureaucracy had continued to function under the Mauryas. The early inscriptions of northwestern India, of which the oldest are the Aśokan edicts, are written in the so-called kharoṣṭhī script, clearly derived from Aramaic writing. The names of two Aśokan officials, who were natives from the northwest or from Iran, are known. In Brahmagiri, at the end of Minor Rock Edict I, we read Capaḍena likhite (in brāhmī script) lipikareṇa (in kharoṣṭhī script), “written by Capada, the scribe.” The inscription of Rudradāman I at Junāgaḍh in Surāṣṭra has preserved for us the remembrance of Tuṣāspha, yavanarāja, i.e., “Iranian king,” of Aśoka Maurya, who had brought improvements to a great dam built there under Candragupta. The Aśokan inscriptions bear traces of Iranian influences probably transmitted through that bureaucracy. In the expression dhaṃmalipi “text of the dharma,” by which Aśoka designates his edicts, lipi is a borrowing from Iranian dipi “writing,” with l- by analogy with Sanskrit LIP “to smear” and LIKH “to write;” dipi is even found in the northwestern copies, Shāhbāzgaṛhī and Mānsehrā (dhramadipi), where one can also note many instances of Iranian nipiš instead of Indian LIKH “to write” of the other copies. The formulaic parallels between Aśoka’s inscriptions and the Achaemenid ones have also been noted (E. Sénart, Inscriptions de Piyadasi, Paris, 1881-1886, II, pp. 296-97), but these are sentences too common to be significant by themselves. We should not exclude the possibility that Aśoka was consciously imitating the example of the great Achaemenid kings having his edicts engraved on rocks and mountains.

There is no architecture built of freestone in India before the Maurya epoch; its appearance is, in all probability, tied to the coming in India of Iranian architects and handworkers who were deprived of employment by the disappearance of the Achaemenid empire (M. Wheeler, Flames over Persepolis, London, 1968, pp. 127-45). This influence is very clear in Pāṭaliputra/Patna where in 1912/13 the ruins of a palace attributed to Candragupta, but which could equally well be attributed to his successors Bindusāra or Aśoka, were discovered. A large hypostyle hall has been uncovered that recalls the Achaemenid palaces (apadāna). Similarly, the so-called Aśokan columns, many of which were erected after Aśoka, present a complex melange of Iranian influences (material, polishing of the stone, elements of the capital) and of Indian elements (monolithic shafts, method of erection of the columns, recomposition of the capitals, called, for that reason, Indo-Persepolitan, symbolism of the monument; J. Irwin, Burlington Magazine 115, November 1973, pp. 706-20; 116, December 1974, pp. 712-27; 117, October 1975, pp. 631-43; 118, November 1976, pp. 734-53). Some art historians even see the hand of Hellenistic carvers in the most beautiful of these monuments (especially Sārnāth).

During the reign of Aśoka, Iran was under the sway of the Seleucid (Greek) rulers Antiochus Theos (261-247) and Seleucus II Kallinikos (246-226). Antiochus is named twice in Aśoka’s inscriptions. According to Rock Edict II, Aśoka sent Antiochus and the kings who were his neighbors medical treatments for animals and men, medicinal plants, roots and fruits. According to Rock Edict XIII, he succeeded in converting him to the politic of Dharma. It is to be noted that from his point of view that surely meant that the Seleucid king recognized Aśoka’s suzerainty (G. Fussman, Annales ESC, 1982, p. 628). In any case, we cannot doubt that the relations between the Indian and the Hellenistic worlds were very close under Aśoka. Thanks to the inscriptions of Kandahar, we moreover know that in Arachosia, an historically Iranian territory, a Greek (Aśoka yona, Sanskrit Yavana < Ionian, Greek Iáwōn “Ionian”) colony was established, important both by reason of its numbers and its politico-economic influence. It was for the sake of this colony, established there since the time of Alexander, that at least Rock Edicts XII and XIII had been translated into Greek and that a resume of the first Rock Edicts had been drawn up in Greek. The literary quality of these texts proves that the Greeks of Kandahar had remained in constant contact with the Mediterranean world; this contact could only have occurred through the intermediary of Hellenistic Iran. Neither can it be doubted that the Greek community in Kandahar played an important role in the relations between Hellenistic Iran and Maurya India. From this community Aśoka recruited his envoys to the Hellenistic world, or those who guided them and served them as interpreters (D. Schlumberger and L. Robert, JA, 1958, pp. 5 and 12-13).

It can be assumed that the contacts between Iran and India under Aśoka did not stop there, but there is no evidence of commercial relations between these two empires, although hoards of punch-marked coins dating from Maurya times have been found in Bactria (Aï Khanoum/Āy Ḵānom and the Oxus temple in Taḵt-e Sangīn): These hoards are best explained as parts of the booty made in India by the Greek rulers of Bactria and brought back home. It is not impossible that Greek or Hellenized Iranian artisans came to look for jobs in India (M. Wheeler, loc. cit.), but this can not be proved at present. The square and apparently regular plan of Śiṣupālgaṛh in Orissa is often attributed to Hellenistic influences coming through Iran, but the town has not been sufficiently excavated to allow certainty; it is not even clear that it was founded in Maurya times.

The Singhalese chronicles also report that under Aśoka, eighteen years after his coronation, the Buddhist vibhajyavādin monks of Pāṭaliputra, Aśoka’s capital, sent the arhant Mahārakkhita to convert the Yonaka-loka “the Greeks,” i.e., Arachosia and perhaps the Iranian countries situated further to the west. He is supposed to have preached the Kālakārāmasuttanta (Aṅguttara-nikāya, ed. Morris and Hardy, Pali Text Society, London, 1885-1910, II, pp. 24-26), converted 170,000 (or 137,000) people and ordained 10,000 monks. It is impossible to know whether, in this legend, everything is anachronistic and exaggerated, or whether the first Buddhist establishments in Kandahar effectively go back to the Maurya period. In any case, the only stūpa and Buddhist monastery ever found in Kandahar (G. Fussman, Arts Asiatiques 13, 1966, pp. 37-56) can not in their present state be attributed to so early a date (S. W. Helms, Afghan Studies 2, 1979, pp. 2-4; ibid., 3-4, p. 14). At the same time, according to the same sources, a Greek or an Iranian, Yonaka Dhammarakkhita, converted the Aparāntaka (western coast of India) (E. Lamotte, Histoire du bouddhisme indien I, Louvain, 1958, pp. 320-38).



On the Kambojas, see also E. Herzfeld, The Persian Empire, Wiesbaden, 1968, pp. 345-46; G. Fussman, Bulletin de l’Ecole Française d’Extrême Orient 61, 1974, p. 33; S. Chattopadhyaya, The Achaemenids and India, Calcutta, 1974, pp. 13, 57-58, 69-70 (outdated, but very complete on the Indian sources).

On the treaty between Seleucus I Nikator and Candragupta, see also P. Bernard, Fouilles d’Aï Khanoum IV: Les monnaies hors-trésors. Questions d’histoire gréco-bactrienne, MDAFA 38, Paris, 1985, pp. 85-95.

For finds of Maurya punch-marked coins in Bactria, see P. Bernard, RN 15, 1973, pp. 238-87; idem, Comptes rendus de l’Académie des inscriptions et belles lettres, Paris, 1978, p. 452 and ibid., 1980, p. 448; Cl. Rapin, Bulletin de correspondance hellénique 107, 1983, pp. 329-30, 334, 365; E. V. Zeymal’, Drevnie monety Tadzhikistana, Dushanbe, 1983, pp. 58-62.

On Aśoka’s inscriptions in Arameo-Indian and Arameo-Iranian, see also H. Humbach, “Arameo-Iranian and Pahlavi,” in Hommage universel II, Acta Iranica 2, 1974, pp. 237-43; idem, “Buddhistische Moral in aramäoiranischem und griechischem Gewande,” in Prolegomena to the Sources on the History of Pre-Islamic Central Asia, ed. J. Harmatta, Budapest, 1979, pp. 189-196; B. N. Mukherjee, Studies in the Aramaic Edicts of Aśoka, Calcutta, Indian Museum, 1984 (where all the earlier bibliography is to be found).

On Aśokan pillars and Achaemenid art, see also J. Irwin, “The True Chronology of Aśokan Pillars,” Artibus Asiae 44/4, 1983, pp. 247-65; N. R. Ray, Maurya and Śunga Art, Calcutta, 1965, is still useful; S. P. Gupta, The roots of Indian art, Delhi, 1980, specially pp. 284-313, denies any such influence.

(G. Fussman)

iii. The Legend of Aśoka in Khotanese

The name of Aśoka appears in late Khotanese texts in the form of Aśūʾ or Iśūʾ, i.e., phonemically Aźū, a further development of a Northwest Prakrit form *Aźūga- also seen in the Middle Chinese transcription *A-yuk (Mathews, nos. 1, 7687; Karlgren, nos. 1m, 1020a), where -y- has the transcription value of -ź- (Pulleyblank, pt. 1, p.115; Bailey, “Hvatanica IV,” p. 919 with n. 1). Note that in contrast with Khotanese, Sogdian has a form borrowed from Sanskrit, šwkʾ (Henning, “Murder,” pp. 138-41).

The legend of Aśoka is told in a number of avadānas (Buddhist stories, legends), of which the Aśoka-avadāna proper is only one. It is customary, however, to use this name for the entire story complex which in the original Sanskrit also includes the legend of Aśoka’s son Kuṇāla and various anecdotes about Aśoka’s minister Yaśas. The entire Aśoka-avadāna was translated into Chinese several times, two of the versions being still extant, whereas in Tibetan only the Kuṇāla-avadāna is still found. The earliest Chinese version was made about A.D. 300 by the monk Fă-qīng (Mathews, nos. 1762, 1095), a Parthian (ān, short for ān-xí, Mathews, nos. 26, 2495, from Middle Chinese *an-siək, from Iranian aršak, see Pulleyblank, pt. 2, p. 228; see also An-hsi). There is also a versified Kuṇāla-story in Chinese, which appears to contain some of the elements found in the Khotanese version but not in the Aśoka-avadāna (see below).

Of the Khotanese version there are two fragmentary copies, both of which were published by Bailey (Khotanese Buddhist Texts, pp. 40-42) and later translated by him (“A Tale of Aśoka”). No detailed comparison of the Khotanese with the other versions (Sanskrit, Tibetan, Chinese) has yet been undertaken; however, the Khotanese does not agree closely with any of them. Following is a résumé of the Khotanese text, with page references to the Aśoka-avadāna in the Divyāvadāna in Cowell and Neil’s edition.

Though not a sūtra, both the Sanskrit and the Khotanese start with the customary introductory formula of Buddhist sūtras: “ Thus I have heard” (cf. Strong, Legend, p. 173 n. 2). We then have a description of how, after the Buddha’s mahāparinirvāṇa, there ruled in the town of Pāṭaliputra a king named Aśoka, who reigned over the whole of Jambudvīpa (the world), and its kings acknowledged him. (This is similar to Divyāvadāna, pp. 368-69, where the Buddha prophesies of the boy Jaya, that a hundred years after the Buddha’s parinirvāṇa he would become the king Aśoka, etc.). He built 18,000 caityas, Buddhist shrines where relics of the Buddha are preserved. (The Sanskrit and Chinese have 84,000 stūpas.) There follows the story of the birth of Kuṇāla; Aśoka had two queens, Padmāvatī and Tiṣyarakṣitā, and a harem of 8000. Padmāvatī bore him a son whom they named Varmavardaṃ (probably for *Darmavardaṃ; the Sanskrit has Dharmavivardhana “Dharma-increasing,” Divy., p. 405.) Once Aśoka looked at the boy’s eyes and was struck by their beauty, he asked the ministers whether they had seen anything on earth like them. They responded that on Mount Gandhamādana there was a bird named kuṇāla who would be the only being on earth with eyes similar to those of the boy, which were like two chalcedony jewels upon a blue lotus leaf. The king ordered the bird brought to him and saw that its eyes were indeed like those of the prince, whom they consequently renamed Kuṇāla (Divy., pp. 405f.). After Kuṇāla had grown up (the Sanskrit and Chinese versions add that he was married to a girl named Kāñcanamālā) Aśoka once took him to a monastery (saṃghārāma, in Divy., p. 406 called Kurkuṭārāma), where the boy was a delight to the Elder Upagupta. (This is the only mention in the Khotanese version of Upagupta, who plays a prominent part in the Sanskrit and Chinese versions.)

There then follows the story of the minister Yaśas and the heads, which in the Sanskrit is placed at the beginning of the Kuṇāla-avadāna (Divy., pp. 382-84): On another occasion, Aśoka honored a Buddhist teacher (ācārya) by bowing with his forehead at the other’s feet, and was reproved for this by Yaśas, a very unbelieving minister. (The Khotanese text has tvari iṣadä “very unbelieving” for the Sanskrit paramaśrād dho, which confirms Przyluski’s proposal [p. 109 n. 1] that we should read this as param’ aśrāddho “very unbelieving,” not paramaśrāddho “of utmost faith,” as, e.g., Strong, p. 234.) Thereupon the king ordered all his ministers (7,000 in all) to bring one head each, without, however, killing a living being (this last condition is not in the Sanskrit). Yaśas was to bring a human head. When they brought the heads in the morning, the king ordered them to take the heads to the market and sell them. All were able to sell their heads except Yaśas. By explaining that the reason for this was the repulsive nature of the (dead) head, Aśoka was able to convince Yaśas that it was no bad thing for the king to bend his valueless human head to the feet of Buddhist monks or teachers, who possess immeasurable good qualities.

There follows the story of the rebellion of Takṣaśilā and the incestuous passion of queen Tiṣyarakṣitā for Kuṇāla, which in the Khotanese version are connected in a manner differing noticeably from the Sanskrit (and Chinese). In the Sanskrit we first have the attempt by the queen to seduce Kuṇāla (Divy., p. 407) and Kuṇāla’s rebuttal of her advances, after which the queen always sought to find fault with Kuṇāla. Next Aśoka, following the advice of his ministers, sent Kuṇāla to quell the rebellion of Takṣaśilā. During his absence Aśoka fell ill but was cured by Tiṣyarakṣitā, who in recompense was granted the kingship for seven days. The queen seized the opportunity to take revenge on Kuṇāla by ordering the citizens of Takṣaśilā to have him blinded. In the Khotanese this appears as follows: After Aśoka had gone to deal with the rebels, the queen made advances to Kuṇāla, who, however, withstood all her guiles. When Aśoka returned Kuṇāla mentioned nothing of all this, but the queen was bent upon revenge and schemed with Yaśas to have Aśoka send Kuṇāla to Takṣaśilā, who thus “lost his native land.” Yaśas, for his part, had reason to hate Kuṇāla because he had put him to shame publicly on one occasion by striking him six times on the head when his hat fell off. This story is not in the Sanskrit text but parallels are found in the Chinese versions, e.g., in the versified Kuṇāla-story (see also Bailey, “Tale,” p. 11).

It is a matter of regret that the Khotanese texts are incomplete, since the Khotanese version might be expected to contain some reference to the legend of the foundation of Khotan, which in Tibetan and Chinese sources is connected with Aśoka and Kuṇāla. For details and bibliography see below.

(P. O. Skjærvø)

iv. The Legend of Aśoka and the Founding of Khotan

Chinese and Tibetan sources have preserved four accounts of the founding of the kingdom of Khotan, all of which relate this to the reign of Aśoka in the third century B.C., and either to Aśoka himself or to his sons. The four accounts are summarized in Lamotte, Histoire, pp. 281-83. Of the four accounts two are in Chinese and date from the seventh century A.D.; the other two are in Tibetan and at least one of them was probably compiled about the seventh century (Lamotte, p. 282).

Both Chinese accounts are reported by the Chinese Buddhist pilgrim Xüan-Zang (Hsüen-Tsang), who traveled to India in 629-45. In his Buddhist Records of the Western World Xüan-Zang reports that, according to the legend, the cruel queen of Aśoka, Tiṣyarakṣitā, made the officials of Takṣaśilā blind his son Kuṇāla (see above on the Aśoka-avadāna). The officials were then exiled by Aśoka and crossing the Snowy Mountains (the Himalayas) went to the desert in the western part of Khotan (the Takla Makan desert). At the same time a Chinese prince, also in exile, came to the western part of Khotan. He conquered the Indians, then settled in the middle of Khotan, and later extended his rule to all of Khotan. When the king grew old without having obtained an heir, he prayed to the guardian god of Khotan, Vaiśravaṇa (or Vaiśramaṇa), who granted him a son. The boy was suckled by a breast swelling up from the earth, whence the local name of the kingdom, Kustana “earth-breast.”

In the second account, the Life of Xüan-Zang, which was composed by his followers, it is the king’s son, Kuṇāla, himself who is exiled and settles in Khotan. Being sonless, Kuṇāla prays to Vaiśravaṇa and a boy is born from the forehead of the god. The infant is suckled by a breast miraculously emerging from the earth. Hence the boy is called Kustana. See also Stein, Ancient Khotan, p. 159.)

In the first of the Tibetan accounts, the Gośṛṅgavyākaranā, which was compiled about the seventh century, it is a Chinese king who asks Vaiśravaṇa for a son, and the god gives him a boy who is none other than a son of Aśoka. In this version of the legend, too, an earth breast suckles the child; who is then called “he whose mother-breast is a breast from earth” (sa-las nu ma nu). His adoptive father names him king of Khotan and he goes to Khotan accompanied by the Grand Chancelor Yaśas and several Chinese armies. Well established in Khotan, they are joined by an Indian people with which they live in peace, with common use of the waters.

The fourth version of the foundation legend is found in the Tibetan “Prophecy of the country of Li” (Li-yul luṅ-bstan-pa). This version is somewhat more elaborate than the other three: The Indian king Dharmāśoka, after having been converted to the Buddhist faith by his minister Yaśas, once traveled to Khotan. While there, his wife was made pregnant by Vaiśravaṇa and bore a son. Aśoka, naturally fearing future competition from the boy, especially in view of a prophecy that the boy would be king during his father’s lifetime, left him, and the boy was suckled by an earth-breast and was called Sa-nu “earth-breast.” In the meantime, a Chinese king, worrying about being almost childless since he had only 999 sons though with the power in him for one more, had prayed to Vaiśravaṇa, who then gave him one more, namely Sa-nu. After growing up, the prince fell out with his father and brothers and returned to Khotan with 10,000 Chinese colonizers. Aśoka’s minister Yaśas, who on account of his relatives had become unpopular in India also came to Khotan, with 7,000 men. The two groups were shortly reconciled and lived in peace with each other.

Historical background of the legend. A connection between Aśoka, or any of his descendants, and Khotan has not been confirmed by any other historical sources, so it is doubtful whether the legend is anything but an attempt to provide Khotan with an eponymous founder hero (see on the name below). On the other hand, there is definite archeological evidence of a Sino-Indian symbiosis in Khotan. At the site of the ancient capital, Yotqan, coins have been found with Chinese legends on the obverse and Indian ones on the reverse. Moreover, the existence of an Indian colony in Khotan at an early time is confirmed by the finds at Niya and Endere of documents written in Indian language in Kharoṣṭhī script. The earliest of these documents dates to the third century A.D.

One is struck by the absence of any allusion whatsoever to an Iranian element in Khotan in the four founding legends. As a matter of fact, we know nothing about when and how the Iranians entered the area. It may have been quite early, though, since the Old Persian inscriptions of Darius I and Xerxes I (6th to 5th cents. B.C.) mention Iranian Saka tribes far up in the northeast. Actually, various Saka tribes may have been wandering about Central Asia from a very early period, probably from the first half of the second millennium B.C. (see Emmerick, Guide, p. 3).

There is no direct evidence for such an early presence of Iranians in the area, nor do we know the exact location of the Saka tribes mentioned by Darius and Xerxes; however, in one inscription they are listed after Sogdia and Gandhāra, and in another after Gandhāra and India, so they must have belonged to the extreme northeast. An argument against identifying the Sakas of Darius and Xerxes with the Iranians of Khotan is perhaps the fact that when Darius built his castle at Susa, he received building materials from the entire empire, but the Sakas are not mentioned in this connection. If the Sakas of Darius were already settled in Khotan, we would certainly expect them to have contributed with their famous jade stone. (On the Khotanese jade see, e.g., Stein, Ancient Khotan, pp. 132-33; Bailey, Culture, pp. 1f.; R. E. Emmerick, in R. E. Emmerick and P. O. Skjærvø, Studies in the Vocabulary of Khotanese II, Vienna, 1986, s.v. īra-.)

The earliest evidence of any kind whatsoever not only for Iranian presence, but also for Iranian supremacy in the region, is a document found at Endere, probably dating from the third century A.D. The document is dated in the “regnal year of the Great King of Khotan, King of Kings” (Khotana maharaya rayatiraya). The king bears the title of hinajha, which is a Northwest Prakrit spelling for *hīnāza, Khotanese written hīnāysa, “general,” the Iranian equivalent of Sanskrit senāpati, which is found in other documents of the region. Also, the word for “regnal year” is ch’una, the equivalent of Khotanese kṣuṇa. The use of this terminology shows that “already at that time [there must] have been a long-established connection between the Iranian inhabitants of Khotan and royal power” (Emmerick, Guide, p. 4).

Kustana and the name of Khotan. The story about Kustana and the “earth breast,” though now extant only in the Tibetan and Chinese sources, must have originated among the Indian-speaking population, which used the name Gostana for Khotan, of which it made Kustana the eponymous founder and about whom it spun the legend of the “earth breast” and the rest. Neither the local Khotanese form nor the attested Chinese name could have given rise to the legend: The Khotanese called their land Hvatana (pronounced Hwadana, later reduced to Hvaṃna and Hvaṃ, i.e., Hwãna and Hvã) and the Chinese used Yú-tiàn (Mathews, nos. 7592, 6374; Karlgren, nos. 97a, 375r; the character for the second element varies some) from *hwāh-den (Pulleyblank, pt. 1, p. 91). (The modern name is Hé-tiăn, lit. “peace field,” Mathews, nos. 215, 6362). The Indian and Chinese forms are also found in Khotanese documents as Gaustana (variants Gą̄stana, Gą̄stam, Gą̄stamä, i.e., Gostan or Gostã; the name is also found in Chinese texts as Ju-sa-dan-na from older *Gi̯ou-sat-tan-na) and Yūttiṃni (with or without kūhi: i.e., Chinese guó “land”) respectively.

Obviously, only the Indian form could give rise to the story of Kustana. As for the name Gostana, it is of course unlikely to contain Ind. stana “breast” (cf. Av. fštāna-, NPers. pestān), and Pelliot (Marco Polo, pp. 410f.), taking up a suggestion by Thomas (Tibetan Texts, pt. 1, p. 18 n. 1), proposes to derive it from sthāna “place,” or rather, Iranian stāna, the well-known suffix forming names of places and lands which apparently also influence the corresponding use of Indian sthāna in this area (cf. Ind. Cīnasthāna, Kharoṣṭhī Cinasthana, Iran. Čīn(e)stān in Mid. Pers. and Sogd.). The origin and meaning of the element Go- remains unknown (Pelliot, op. cit., pp. 412f.). Eventually, Khotanese Hva-, Indian Go-, and Chinese *Hwāh- may represent the same original, local, name, to which was later added -dan, which still later was replaced by the common -stan. The Tibetan name for Khotan is quite different: Li or Li-yul “country of Li.”



Khotanese text of the Aśoka-avadāna in H. W. Bailey, Khotanese Buddhist Texts, Cambridge, 1951, 2nd ed., 1981, pp. 40-42.

Translation by idem, “A Tale of Aśoka,” Bulletin of Tibetology 3, 1966, pp. 5-11.

The Sanskrit text of the Aśoka-avadāna was edited by E. B. Cowell and R. A. Neil, The Divyâvadâna, a Collection of Early Buddhist Legends, Cambridge, 1886, repr. Amsterdam, 1970, pp. 348-434.

It was reedited by S. Mukhopadhyaya, The Aśokāvadāna, Sanskrit Text Compared with Chinese Versions, New Delhi, 1963 (partly translated).

The Sanskrit text was translated by J. S. Strong, The Legend of King Aśoka, Princeton, 1983.

The Chinese version of Fă-qīng was translated by J. Przyluski, La legende de l’empereur Açoka (Açoka-avadāna) dans les textes indiens et chinois. Annales du Musée Guimet, Bibliothèque d’études 32, Paris, 1923.

On the Chinese transcription of Aśoka see H. W. Bailey, “Hvatanica IV,” BSOAS 10/4, 1942, p. 919 with n. 1.

E. G. Pulleyblank, “The Consonantal System of Old Chinese,” pt. 1, Asia Major, N.S. 9, 1962, pp. 58-144; pt. 2, ibid., pp. 206-65.

On Sogdian šwkʾ see W. B. Henning, “The Murder of the Magi,” JRAS, 1944, pp. 133-44 (Selected Papers II, Acta Iranica 15, Tehran and Liège, 1977, pp. [139-50]).

See also E. Lamotte, Histoire du Bouddhisme Indien I: Des origines à l’ère Śaka, Louvain-la-Neuve, 1976, pp. 261-72.

R. E. Emmerick, A Guide to the Literature of Khotan, Tokyo, 1979, pp. 17-18.

On the foundation legends see Sir Aurel Stein, Ancient Khotan I, Oxford, 1907, pp. 156-66. Lamotte, Histoire, pp. 281-83.

Tibetan texts in F. W. Thomas, Tibetan Literary Texts and Documents Concerning Chinese Turkestan. Part 1. Literary Texts, London, 1935.

R. E. Emmerick, Tibetan Texts Concerning Khotan, Oxford, 1967, pp. 15-25.

See also Emmerick, Guide, pp. 1-4.

On the names of Khotan, see H. W. Bailey, ibid., and “Hvatanica III,” BSOAS 9/3, 1938, p. 541.

P. Pelliot, Notes on Marco Polo, Paris, 1959, pp. 408-25.

E. G. Pulleyblank, ibid. pt. 1, p. 91.

The Chinese texts are listed in P. Demieville et. al., Répertoire du canon boudhique sino-japonais, Fascicule annexe du Hōbōgirin, Paris and Tokyo, 1978, pp. 164f. nos. 2042, 2045.

See also H. W. Bailey, The Culture of the Sakas in Ancient Iranian Khotan, Columbia Lectures on Iranian Studies 1, ed. E. Yarshater, New York, 1982, esp. p. 3, where an Iranian etymology for Hvatana- is proposed (tentatively “the (land of the) lords” [?], connected with Iran. xwata- “self”). Chinese quoted from Mathews’ Chinese-English Dictionary, rev. Amer. ed., Cambridge, Mass., 1972 (12th printing) and B. Karlgren, Grammata Serica Recensa, Stockholm, 1972.

(P. O. Skjærvø)

(J. G. De Casparis, G. Fussman, P. O. Skjærvø)

Originally Published: December 15, 1987

Last Updated: August 17, 2011

This article is available in print.
Vol. II, Fasc. 7-8, pp. 779-785