² Unless otherwise stipulated, the term refers to the Makhdūmzāda Khwājas, political and religious leaders of the Alitshahr region.

³ The following account is taken largely from Schwarz, Henry, ‘The Khwājas of Eastern Turkestan’, Central Asiatic Journal, 20/4, 1976, 266–96Google Scholar; Fletcher, Joseph, ‘The Naqshbandiyya in northwest China’, in Manz, Beatrice Forbes (ed.), Studies on Chinese and Islamic Inner Asia, (Aldershot: Variorum, 1995), ch. xi Google Scholar; Shaw, Robert B., ‘The history of the Khōjas of Eastern-Turkistān’, Journal of the Asiatic Society of Bengal, LXVI, 1897 Google Scholar, supplement; Hartmann, Martin, ‘Ein Heiligenstaat im Islam: das Ende der Caghataiden und die Herrschaft der Chogas in Kasgarien’, Der Islamische Orient, VI–X, 1905, 1–173.Google Scholar

⁴ After the reassertion of Mongol power in 1713, Schwarz finds reference to a nominal force of 15 soldiers in each major city (Schwarz, ‘The Khwajas’, 281).

⁵ A Naqshbandī religious leader (1461–1542) who was active in Central Asia and claimed descent from the Prophet in the twenty-second generation.

⁶ In the early eighteenth century, the Āfāqiyya and Isḥāqiyya became known as the Aqtāghlϊq (White Mountain faction) and Qarātāghlϊq (Black Mountain faction) respectively, after the Kirghiz who were associated with them. One group of these Kirghiz is said to have originated from the Pamirs (Qarātāgh) and the other from a mountain north of Artish (Aqtāgh). In an attempt to simplify events I have omitted reference to the Kirghiz, both here and later.

⁷ This point has been argued by Miao Pusheng. See Pusheng, Miao, ‘Guanyu Qingchao zhengfu dui Xinjiang Yisilanjiao zhengce zhong de jige wenti’ (‘Regarding several questions in Qing dynasty policy towards Islam in Xinjiang’), Xinjiang lishi yanjiu (Studies in Xinjiang history), 1, 1987.Google Scholar

⁸ Huijiang zhi [hereafter HJZ] (Gazetteer of the Muslim regions), compiled by Su-er-de, et al. , 1772, 22b. There are several slightly different versions of this work, I am here following an undated manuscript version in the Bodleian Library (MS Chin.e.6).Google Scholar

⁹ Qi-shi-yi (Chunyuan) ‘Xiyu wenjian lu’ (hereafter XYWJL) (‘Record of the Western Region’) c. 1764 (Preface 1760), Qingzhao tang congshu, (comp.) Yuanchun, Li, 1835, 106a.Google Scholar

¹⁰ XYWJL, 92a.

¹¹ For example in Aqsu in the absence of a hakim it was apparently ākhūnd 'Abdu Ghāfur who assumed authority. Qinding pingding Zhunga'er fanglüe [hereafter PDZFL] (Imperially commissioned account of the pacification of the Zunghars), (comp.) Fu-heng et al., 1772, zheng 65:9ab.

¹² Appointed as the temporary hakim of Kashgar, in 1760 Kush Kipak went to court where he was given a peerage of the sixth rank fuguo gong and detained permanently in Beijing—on account of his divided loyalty, no doubt. Huijiang tongzhi [hereafter HJTZ] (Comprehensive gazetteer of the Muslim regions), (comp.) He-ning, 1804, 6:8ab. Nor was he the only Isḥāqī beg to switch sides.

¹³ See their biographies in HJTZ, Juan 5.

¹⁴ See Tōrn, Saguchi, ‘The revival of the White Mountain Khwājas, 1760–1820’, Acta Asiatica, 14, 1968, 7–20.Google Scholar

¹⁵ Following conventional usage, Kashgar, Yarkand, Yangi Hisar and Khotan are referred to as the western cities of the Altishahr region, and Kucha, Ush-Turfan, Aqsu (and Bai) and Korla as the eastern cities.

¹⁶ PDZFL, zheng 75:33b–34a.

¹⁷ The etymology of the titles is varied: Persian, Arabic, Turkic and Mongol. Saguchi suggests that some had been used as the titles of Turkmen, Uzbek and Iranian officials as early as the sixteenth century. Tōru, Saguchi, Jūhachi-jūkyūseiki Higashi Torukisutan shakaishi kenkyū (Study of the history of eighteenth- to nineteenth-century Eastern Turkestan society) (Tokyo: Yoshikawa Kōbunkan, 1963), 123. Saguchi's chapter on the beg-system is still the best general account and I am much indebted to his pioneering work.Google Scholar

¹⁸ HJZ, 49a.

¹⁹ PDZFL, zheng 75:28b–29b.

²⁰ In the mid eighteenth century, Qing massacres compounded by internecine fighting, emigration and a small-pox epidemic served to wipe out hundreds of thousands of Zunghars, leaving a largely depopulated region north of Tianshan. The Zunghars had established a small Turkic-Muslim community in lli Valley to which the Manchu now added, moving hundreds of Turkic-Muslim households from the south.

²¹ HJTZ, 7:19a–20a.

²² Huijiang zeli [hereafter HJZL] (Substatutes of the Muslim regions), (comp.) Tuo-jin et al., 1842, 1:13b. The Dolons were probably of Mongol origin, although they were Muslims. There were some 400-odd households at Aqsu at the time of the conquest, PDZFL, xu 8:17a.

²³ Figures fluctuated and differ slightly from source to source. The figures given here are based on the HJTZ.

²⁴ There is only one indication of a possible change in function. According to the HJZ 64a, formerly, the dīvān (du-guari) had been responsible for official documents, but he now took charge of supplying foreign envoys with provisions. However, it is not clear when this ‘formerly’ refers to and already in 1759 Zhao-hui's report notes that the dīvān was responsible for courier stations (PDZFL, zheng 75:34a).

²⁵ See Jōhei, Shimada ‘Hōjiya jidai no bekutachi’ (‘The begs at the time of the Khwājas’), Tōhōgaku (Eastern Studies), 3, 1952, 71.Google Scholar

²⁶ There is some evidence, however, that the suffix beg was used with at least three of the posts prior to Qing rule, e.g. shangbeg, mingbeg, and yüzbeg. See for example, PDZFL, zheng 75:33b–34a and Shimada–s reference to the Kashgarian waqf record of 1662. As Shimada points out, in the case of the mīrāb, to add the suffix beg was clearly tautological; ‘mār’ meaning ‘amīr’ and ‘ab’, ‘water’. Shimada, ‘Hōjiya jidai no bekutachi’, 71.

²⁷ The khazānachī-beg and the shang-beg were responsible for control of the treasury and tax collection, respectively.

²⁸ These officials were responsible for Islamic law, deeds, construction works, policing, courier stations, gold-mining, tax collection from 1,000 households and Islamic education, respectively.

²⁹ Number One Historical Archives, Beijing: Junjichu lufu zoupi: minzu (Grand Council copies of palace memorials: minorities), 8072.38 (Daoguang, 14.12).

³⁰ Appointments to jinding posts were made locally and ratified by the Councillor (HJTZ, 9:8a). For a full account of the local and Manchu administrations see Qinding huangyu Xiyu tuzhi (Imperially commissioned illustrated gazetteer of the lands of the Western regions), (comp.) Fu-heng et al., 1782, juan 29 and 30.

³¹ Naqshbandi, Ahmed Shāh, ‘Route from Kashmir, via Ladakh, to Yarkand, by Ahmed Shah Nakshahbandi’ (tr. Dowson, J.), Journal of the Royal Asiatic Society of Great Britain and Ireland, 12, 1850, 383.Google Scholar

³² HJTZ, 5:3b.

³³ HJZ, 23a; XYWJL, 105b.

³⁴ Nawenyi gong zouyi [hereafter NWYGZY]. (Collection of the memorials of Na-yan-cheng), (comp.) Rong-an, , 1834, 78:23ab.Google Scholar

³⁵ For an account of this incident see Millward, James, Beyond the Pass: commerce, ethnicity and the Qing in Xinjiang, 1759–1864’ (Ph.D. thesis: Stanford University, 1993), 322–3.Google Scholar

³⁶ I refer to the revolt of 1815 led by the Isḥāqī ākhūnd Ziyā al-Dīn. Although the reasons for this revolt were never satisfactorily revealed, there can be no doubt that Ziyā al-Dīn was an influential character whose authority challenged that of the local begs—including the hakim of Kashgar. See Song-yun Xinjiang zougao (Song-yun's Xinjiang memorials), (Beijing: Zhongyang minzu xueyuan tushuguan, 1980)Google Scholar.

³⁷ Miao, ‘Guanyu Qingchao zhengfu’, 43–4.

³⁸ PDFZL, xu 1:11b.

³⁹ There is a problem here. Shaw notes that 'Othmān was the son of a religious man named Mīr Zāhidī (see Shaw, ‘History of the Khōjas’, preface by N. Elias, iii). I have not been able to trace any further reference to this hakim of Kashgar. Clearly it is not 'Othmān, son of Hudawī, who did not take office as hakim of Kashgar until 1778.

⁴⁰ Ziyang, Liu (ed.), Qingdai difang guanzhi kao (Study of local officialdom in the Qing dynasty) (Beijing: Zijincheng, 1988), 355.Google Scholar

⁴¹ For 'Abdu-llah's biography, see HJTZ, 2:6b–12b, 11:10a. This long succession was not achieved without some editing of the family's genealogy, see ‘An heir to Prince Mahomet of Hami’, China Review, XI, 1883, 334.Google Scholar

⁴² HJTZ, 4:la–11a.

⁴³ Bellew, H. W., ‘History of Kashgar’, in Forsyth, T. D., Report of a Mission to Yarkund 1873 (Calcutta: Foreign Department Press, 1875), 201–2.Google Scholar

⁴⁴ Nobility of the second rank junwang from Hami and Turfan were given 1,200 liang of silver and 15 pi of silk per annum, the same as the Mongol nobility of equivalent rank. Others, taiji, etc., were also afforded silver and silk on a par with that received by their counterparts among the Mongolian nobility (HJZL, 3:3a).

⁴⁵ For example, in Beijing the ennobled begs from Hami and Turfan were lodged in the Hami guan, as opposed to the Siyi guan where the Altishahr begs stayed. They also shared the superior seating positions enjoyed by the Mongol nobility at the imperial ceremonies and banquets (HJZL, 3:16b–21a).

⁴⁶ See for example, PDZFL, zheng 75:29a.

⁴⁷ Thirty four Muslims received office as a reward, but this presumably included those from Hami and Turfan (PDZFL, zheng 75:38a).

⁴⁸ PDZFL, zheng 85:13b–14a. See Khōjis's biography, HJTZ, 6:la–7b.

⁴⁹ Jihāngīr, the grandson of the Afāqī leader, Burhān al-Dīn, led the first major Khwāja incursion into the Altishahr in 1826. He was captured in late 1827 and taken to Beijing for quartering.

⁵⁰ Khōjis's title of nobility was also passed down through Kadir. HJTZ, 6:7b–8a; Report on the trade and resources of the countries on the north-western boundary of British India, (comp.) Davies, R. H. (Lahore: Government Press, 1862), Appendix XXIX B, cccxliv, ccxlviii.Google Scholar

⁵¹ Masami, Hamada, ‘Supplement: Islamic saints and their mausoleums’, Acta Asiatica, 34, 1978, 91–2.Google Scholar

⁵² PDZFL, zheng 76:5b–6b.

⁵³ HJTZ, 4:6b–7a.

⁵⁴ There was a relatively swift rise in the number of begs in the lli region as a result of the huge emigration programmes to that area. Conversely, at Ush-Turfan the revolt of 1765 caused a marked reduction in the total number of begs. There were 34 before the revolt, but after the suppression the population was so reduced that four sufficed.

⁵⁵ HJZL 2:3ab, 7:1a; HJTZ 7:18b.

⁵⁶ For an account of this case see Torbert, Preston, The Ch'ing Imperial Household Department (Cambridge: Harvard University Press, 1977), 136–71.CrossRefGoogle Scholar

⁵⁷ See Na-yan-cheng's attempt to tighten the system in 1828, NWYZY, 78:8b–9a. Most notably, in 1811 Yùnus succeeded his father, Iskandar, as hakim-beg of Kashgar.

⁵⁸ 1 bātmān = approx 5.3 dan of grain. Land area was calculated according to the grain harvest it could support. According to Lin Enxian, 1 mu supports 1 dou, therefore 1 bālmān = 5 dan 3 dou = 53 mu. Enxian, Lin, Qingchao zai Xinjiang de Han-Hui geli zhengce (The Qing dynasty policy of Han-Muslim segregation in Xinjiang) (Taibei: Guoli zhengzhi daxue, 1988), 78.Google Scholar

⁵⁹ See for example, HJZ, 71b–72a.

⁶⁰ HJTZ, 7:22b–23a.

⁶¹ HJZ, 75ab.

⁶² Waley-Cohen, J., Exile in mid-Qing China (New Haven and London: Yale University Press, 1991), 31; HJTZ, 7:18a.CrossRefGoogle Scholar

⁶³ See for example, NWYGZY, 78:1 lab.

⁶⁴ Payments ranged from 200 liang of silver for a third-ranking beg moving between Ili and Yarkand, Khoten, Kashgar, Yangi Hisar, Bugur or Korla, to 30 liang for a beg of fifth rank or below moving between the above six cities and Aqsu, Ush, Sairam, or Bai (HJTZ, 7:23a–24a).

⁶⁵ PDZFL, zheng 80:29b–30a.

⁶⁶ See Chia, Ning, ‘The Lifanyuan and the Inner Asian rituals in the early Qing (1644–1795)’, Late Imperial China, 14/1, 1993, 81.Google Scholar

⁶⁷ Pusheng, Miao, ‘Guanyu Qingchao zhengfu’, 43–4; HJTZ, 2: 9b–10a.Google Scholar

⁶⁸ See Saguchi, , Higashi Torukisutan, 156–9.Google Scholar

⁶⁹ HJZL, 4:1a, 4:2ab.

⁷⁰ See for example, NWYGZY, 78:4b.

⁷¹ ‘Tax registers for Kucha and Shayar’, MS Bodleian Library (MS Chin d. 69). I am grateful to Liu Zhiwei and David Faure for bringing these manuscripts to my attention.

⁷² HJZL, 4:6ab.

⁷³ HJZL, 4:7ab. In 1814, these figures were raised by 500 jin for begs of the third rank and by 2,000 jin for the highest-ranking nobility wang. The quotas for others remained the same (Chia, ‘The Lifanyuan’, 75).

⁷⁴ Number One Historical Archives, Beijing: Junjichu lufu zoupi: minzu, 8098.36 (Daoguang, 15.8) 8098.38 (Daoguang, 17.9).

⁷⁵ For details of beg corruption and proposals for reform, see NWYGZY, 77:19a–43b.

⁷⁶ HJTZ, 12:5b–6a.

⁷⁷ Torbert, , Ch'ing Imperial Household, 164.Google Scholar

⁷⁸ ibid., 149.

⁷⁹ Number One First Historical Archives, Beijing: Gongzhong zhupi: minzu (Palace memorials—nationalities), 639:1–11 (Xianfeng, 7.5–8.4).

⁸⁰ See Millward, ‘Beyond the Pas’.

⁸¹ For begs and ākhūnds lending money to Khoqandis see NWYGZY, 19:17a. Davies, , Report on the trade, Appendix XXIV, cxci.Google Scholar

⁸² HJZL, 6:17a. On occasions these fears may have been well-founded, as seems probable in the case of 'Abd al-Raḥīm, ishikāghā-beg of Yarkand. In the early 1760s, having been thwarted in his attempt to become hakim-beg, he was discovered to have entered into communication with the ruler of Khoqand with a view to instigating rebellion; he was consequently executed by the Manchu authorities (HJTZ, 5:3b).

⁸³ On this occasion all the other begs of Kashgar kept their heads and their posts. Akhmedov, B. A. (ed.), Materialy po istorii Srednei i Tsentral'noi Azii X–XIX vv. (Tashkent: Fan, 1988), 327 Google Scholar. Kumlukov, M., ‘Vzaimnootnosheniya tsinskogo Kitaya s Kokandskim Khanstvom’, in Kitai i sosedy (Moscow: Nauka, 1982), 206 Google Scholar. Yūnus's full name was Sa'īd Muḥammad Yūnus.

⁸⁴ In 1861, for example, the rebellion at Artush was led by 'Abd al-Raḥīm, keeper of Satuq Boghra Khan's mausoleum, while in Kucha, Khwāja Rāsh al-Dīn, keeper of the shrine of Arshad al-Dīn, was appointed leader of the rebellion. See Ho-Dong, Kim, ‘The Muslim rebellion and the Kashgar emirate in Chinese Central Asia, 1864–1877’ (Ph.D. thesis, Harvard University, 1986), 30, 55.Google Scholar

⁸⁵ See Yuan, Wei, Shengwu ji (Chronicle of military campaigns) 1842 (repr. Taibei: Wenhai, 1967), 188.Google Scholar

⁸⁶ Huang, Pei, Autocracy at work (Bloomington: Indiana University Press, 1974), 285–301.Google Scholar

⁸⁷ NWYGZY, 78:4b.

⁸⁸ Na-yan-cheng's claim that after Ishāq, hakim-beg of Kashgar, was rewarded with the privilege of being permitted to grow a queue for his part in the Jihāngīr rebellion, all the begs were clamouring to enjoy the same honour (NWYGZY, 78:3ab), should probably not be taken at face value. Nevertheless, the substatute that this memorial prompted decreed that the sons and grandsons of loyal begs with titles and those begs of fourth rank and above would henceforth be allowed to wear the queue as a mark of distinction. See HJZL, 8:2a.

⁸⁹ Ahmed Shah, ‘Route from Kashmir’, 384.

⁹⁰ Bi-chang, , Shoubian jiyao (An outline on border control), 1849 Google Scholar, postscript 4ab.

⁹¹ Vail, Leroy (ed.), The creation of tribalism in Southern Africa (London: James Currey, 1989), 10.Google Scholar