BAḴŠĪ, a Buddhist lama or scholar, in particular during Mongol hegemony in Iran; subsequently, by extension, any kind of scribe or secretary. The word, which is Turkish, is derived from Chinese po-shih (man of learning) and not, as once believed, from Sanskrit bhikṣu, which itself denotes a Buddhist lama. Jovaynī, I, pp. 10, 44 (tr. Boyle, pp. 14, 59-60), employs the term tūyīn (Chinese tao-jen “man of the path”), the original of the tuin of his contemporary, the Flemish missionary William of Rubruck (Itinerarium, ed. A. van den Wyngaert, Sinica Franciscana I, Quaracchi and Florence, 1929, pp. 227-32, 294-97). The word baḵšī appears only later, in the writings of Rašīd-al-Dīn Fażl-Allāh and Waṣṣāf. The period covered by these authors, that of the il-khan Hülegü (Hūlāgū) and his successors, witnessed the brief emergence of Buddhism, for the last time, as a major religion in Iran; it was facilitated by the tolerant attitude of the Mongol rulers towards the representatives of all religious groups and sects, and by the fiscal exemptions granted to them. If he did not actually embrace the Buddhist faith, as his elder brother the Great Khan Qubilai did, Hülegü at least inclined towards it during his last years (Kirakos, pp. 237-38). His son Abaqa appears to have been partial to the baḵšīs and entrusted them with the care of his grandson, the future il-khan Ḡāzān (Rašīd-al-Dīn, III, text pp. 295, 373, tr. pp. 165-66, 209). After a temporary setback under the Muslim Aḥmad-Tegüder (Takūdār), came the heyday of the baḵšīs, coinciding with the reign of Ḡāzān’s father Arḡūn (683/1284-690/1291), who may actually have become a Buddhist and who showered favors upon them. During his final illness, which had apparently resulted from a life-prolonging drug prescribed by a baḵšī from India, only two Mongol amirs, his chief minister Saʿd-al-Dawla, and the baḵšīs were allowed into his presence (ibid., text pp. 223-24, tr. pp. 128-29). After his adoption of Islam in 694/1295, however, Ḡāzān, in contrast with Aḥmad-Tegüder, began to enforce Islam as the state religion. Although he had been reared by baḵšīs and, while governor of Khorasan for his father, had built a Buddhist temple in Ḵabūšān (Qūčān; ibid., text p. 373, tr. p. 209), Ḡāzān set about destroying Buddhist foundations, some of which were converted into mosques, and imposed Islam on the lamas (Waṣṣāf, p. 324). Once it became evident, however, that many were using Islam merely as a cloak for the practice of their old faith, they were given leave to depart from Iran and return to their original homes in Kashmir, India, and Tibet (Rašīd al-Dīn, III, text pp. 396-97, tr. p. 224).
The growth and termination of Buddhist influence in Mongol Iran, a phenomenon that doubtless extended very little outside the court, is an obscure process. What appear to be the remains of cave-temples from this period have been excavated in Azerbaijan; but we lack any Buddhist written sources and are dependent on those emanating from Muslim and Christian writers. One difficulty is that baḵšīān are sometimes linked with qāmān (i.e., shamans) and possibly, therefore, confused with them. The baḵšīs, for example, who strove to induce Ḡāzān’s brother and successor Öljeitü (Ūljāyṭū), himself likewise at one time a Buddhist, to abandon Islam in 707/1307-8 (Kāšānī, Tārīḵ-eŪljāyṭū Solṭān, ed. M. Hambly, Tehran, 1348 Š./1969, pp. 98-99) were most probably shamans rather than lamas (see Boyle in Camb. Hist. Iran V, p. 402). Nevertheless, a lama named Kamālasrī, who apparently came from Kashmir, was in Iran at the beginning of the 8th/14th century, assisting Rašīd-al-Dīn with the composition of the Indian section of his great historical encyclopedia (Die Indiengeschichte des Rašīd al-dīn, ed. and tr. K. Jahn, Vienna, 1980, introd., pp. 10-12).
After the suppression and eclipse of Buddhism in Iran, the term baḵšī came to denote a scribe who drafted Turkish or Mongol documents (e.g., Barthold, Turkestan3, p. 55 n. 4), and was thus synonymous with bitikčī. In time it was applied to any master, including quacks and sorcerers, and even, among the Anatolian Turkmen of the 15th-16th centuries, a wandering minstrel. The connection with the military officer entitled baḵšī in Mughal India, and with the Anglo-Indian term buxee derived from it (Sir Henry Yule and A. C. Burnell, Hobson-Jobson: A Glossary of Anglo-Indian Words and Phrases, new ed. W. Crooke, London, 1903, s.v.), is regarded as somewhat tenuous.
Primary sources: Kirakos Ganjakecʿi, tr. L. A. Khanlaryan, Moscow, 1976, pp. 225, 237-38.
Marco Polo, ed. and tr. A. C. Moule and P. Pelliot, The Description of the World, London, 1938, I, pp. 188-90.
Rašīd-al-Dīn, Jāmeʿ al-tawārīḵ, Baku, III, passim. Ricoldo of Monte Croce, Liber peregrinationis, ed. J. C. M. Laurent, Peregrinatores Medii Aevi Quatuor2, Leipzig, 1873, p. 117.
Waṣṣāf, Tajzīat al-amṣār wa tazjīat al-aʿṣār, lithograph ed., Bombay, 1269/1853, passim.
See also A. Bausani, “Religion under the Mongols,” in Camb. Hist. Iran V, pp. 540-43.
G. Doerfer, Türkische und mongolische Elemente im Neupersischen, Wiesbaden, 1959-75, II, pp. 271-77.
B. Karlgren, Grammata Serica Recensa, Stockholm, 1972, pp. 204 no. 771a, 256 no. 970a.
B. Laufer, “Loan-Words in Tibetan”, Ṭʿoung Pao 17, 1916, pp. 485-87.
Matthews’ Chinese-English Dictionary, nos. 5322, 5776. P. Pelliot, Notes on Marco Polo, Paris, 1959-73, I, p. 63.
E. Quatremère, ed., Histoire des Mongols de la Perse, Paris, 1836, pp. 184-99 n. 51.
B. Spuler, “Bakhshī,” in EI2. Idem, Mongolen3, pp. 178-87, 191.
For the archeological evidence, see W. Ball, “The Imamzadeh Maʿsum at Vardjovi: A Rock-cut Il-khanid Complex near Maragheh,” AMI 12, 1979, pp. 329-40.
For the relations of the lamas with the Mongol imperial family in general, see H. Franke, “Tibetans in Yüan China,” in China under Mongol Rule, ed. J. D. Langlois, Jr., Princeton, 1981, pp. 296-328.
L. Petech, “Tibetan Relations with Sung China and with the Mongols,” in China among Equals, ed. M. Rossabi, Berkeley and Los Angeles, 1983, pp. 173-203.
Originally Published: December 15, 1988
Last Updated: August 24, 2011
This article is available in print.
Vol. III, Fasc. 5, pp. 535-536