The Naqshbandiyya Mujaddidiyya in China

Abstract

The Naqshbandiyya Mujaddidiyya order is among the world's largest and most geographically widespread Sufi orders, but it has long been assumed to be absent among Chinese-speaking Muslims. Despite a handful of isolated references to local Chinese Mujaddidī groups in studies of particular communities, comprehensive histories of Chinese Islams make no mention of the Mujaddidiyya, and histories of the Naqshbandiyya Mujaddidiyya make no reference to Chinese-speaking Muslims. This article demonstrates that the Mujaddidiyya order has not only been present at various times and places among Chinese Muslims, but has also played a role in the development of nearly all major strains of Islam in China proper, including those commonly known as the Gedimu, Jahriyya, Khāfiyya, Qādiriyya, and Ikhwān. The article also uses new primary sources to provide an account of how a Mujaddidī order expanded into Eastern Turkistan and was transmitted from there to Muslims in China proper. It shows that adaptation to local environments created distinctive forms of Mujaddidī Sufism, highlights Hui-Uyghur connections, and argues that South Asia deserves a central place in any account of Islam in China.

Among the more notable characteristics of the Sufi order known as the Naqshbandiyya Mujaddidiyya is its wide geographical spread. By the late nineteenth century, the order had a substantial presence from Mozambique to Indonesia, stretching through the Ottoman empire, Central Asia, the Russian empire, and South Asia, leading Waleed Ziad to characterise it as ‘the most extensive Muslim revivalist network in Asia before the twentieth century’.¹ In many parts of the world, it was also the predominant Sufi order. However, the Chinese cultural area has long been assumed to lie outside of the zone of Mujaddidī expansion. Recent scholarship has shown that the Mujaddidī order also has a longstanding presence among the Uyghurs of Eastern Turkistan (Xinjiang, China), as well as Turkic and Mongolic groups living around the northern edge of the Tibetan Plateau (Qinghai and Gansu, China).² However, the roughly 11 million Sino-Muslims, known today as the Hui ethnic group and spread throughout China proper, are absent from existing depictions of the Naqshbandiyya Mujaddidiyya.³ The order also goes unmentioned in overviews and characterisations of ‘Chinese Islam’, despite the fact that a handful of exceptional studies have spotted Mujaddidī connections in specific communities.⁴ This article combines localised data points from secondary sources with new primary evidence to show that the Mujaddidī order has played a much more important role in the development of various Chinese Islams than previously recognised. Moreover, a number of Sufi orders known to date as ‘Hufuye’ (虎夫耶, Arabic: Khāfiyya) orders either directly trace their spiritual lineage to Mujaddidī silislas (chains of spiritual transmission) or give Mujaddidī thought a central place in their teachings. And the preeminent Mujaddadī text, the Maktūbāt, has found Hui readership from Shanghai to Urumqi, among denominations that include the Gedimu, the Jahriyya, the Qādiriyya, and the Ikhwān.

As this article will show, the Mujaddidī inheritance in China and its Inner Asian colonies is not only substantial, but also too large and varied to describe exhaustively in a single essay. Thus, I describe a selection of cases that are representative of the variety of engagements with Mujaddidī traditions, subject to limits of space and source availability. After providing some general context, the first half of the article treats five types of transmission among the Hui: 1) a formal order that claims a place in the silsila (chain of spiritual transmission) of the Mujaddidiyya's founder, Aḥmad Sirhindī, but whose leader did not seem to emphasise any distinctly Mujaddidī thought; 2) a formal order that both claims a place in the Mujaddidī silsila and maintains some of the practices and beliefs that Sirhindī advocated; 3) an influential gedimu (orthodox, supposedly non-Sufi) lineage of Islamic scholars that grounded its authority in the ceremonial transmission of Aḥmad Sirhindī's Maktūbāt and has recently reconnected with its Mujaddidī roots; 4) an influential Sufi order's adoption of the Maktūbāt as a central text without any claims to direct transmission; and 5) the use of the Maktūbāt as an ‘ideological weapon’ by a leading modernist reformer in the Ikhwānī movement.⁵ The second half of the article turns to the Mujaddidī movements of Eastern Turkistan, from which most Hui Mujaddidīs received their spark. After a brief overview of the main Mujaddidī branches in the region, I focus on the process of Mujaddidī transmission in one prominent branch, namely the Yarkand Daotang, both into Eastern Turkistan and onward toward China proper. Using several new primary sources, I show how this branch was transmitted and localised among the Turki people (today's Uyghurs), the Salars, and the Dongxiang, and then onward to the Hui.

Beyond demonstrating the long-overlooked significance of Mujaddidī lineages, texts, and ideas, these highly varied cases suggest several wider conclusions about both the Naqshbandiyya Mujaddidiyya and Islam in China. One is that the localisation of Mujaddidī movements, whether in Eastern Turkistan or in Northwest China, created new forms of Mujaddidī practice and affiliation that are distinct from the order's manifestations in other parts of Eurasia. In Eastern Turkistan, the Yarkand Daotang was drawn into the local hagiographical tradition, adopting some of the saintly miracles, sacred geography, and perhaps even genre expectations of the shrine-tazkira system that I have described elsewhere.⁶ In Northwest China, the inheritors of the Mujaddidī lineage adapted to the fissiparous menhuan system of Sufi lineages, taking local surnames and place names to distinguish their own sub-lineages amid the constant competition between menhuans. This localised naming convention, which drops the term ‘Mujaddidī’, explains how Mujaddidī connections have been largely overlooked in the secondary scholarship. For the history of Islams in China, the varied Mujaddidī circulations suggest that the role of South Asia has been sorely underestimated, while claims of Muslim isolation in China during the eighteenth century appear to be overstated. Finally, the Naqshbandiyya Mujaddidiyya in China highlight the importance of cross-ethnic exchange, particularly between the Uyghurs/Turki and the Hui or Huihui, whose Islamic literary and religious inheritances share far more than is commonly recognised.

Context

The Naqshbandī Mujaddidī order is an offshoot of the much older Naqshbandī order. The original Naqshbandī order traced its origin to Bahā al-Dīn Naqshband (1318–89 CE), though it is unclear exactly when the movement coalesced into a formal order. The origins of the Naqshbandiyya Mujaddidiyya are much more distinct, resulting from the self-consciously reformist activities of Aḥmad Sirhindī (1564–1624), who encouraged the notion that he was the mujaddid-i alf sānī, the ‘renewer of the second [Islamic] millennium’.⁷ Sirhindī wrote extensively on both the failings of existing Sufi practices and the flaws in anti-Sufi polemics offered by some juridical scholars. In his writings, he proposed a unifying philosophy that explained how the mystical Sufi path and the sharīʿa (divine law), the esoteric and the exoteric, were not, in fact, at odds. He developed a clear plan of spiritual development for his followers, adapting and arranging existing Sufi techniques of meditation. And he designated a number of his followers as khalifas—successors who were licensed to spread the renewed Sufi path that he had developed, soon to be known as the ‘Naqshbandī Mujaddidī’ path. Those successors went on to designate their own khalifas, and so on. Political instability of the mid-eighteenth century, including the 1763–64 sacking of the order's centre at Sirhind, accelerated the spread of the Mujaddidī network.⁸

The original Naqshbandī order was itself geographically expansive at the time of Sirhindī's intervention, with prominent branches throughout the Near East and Central Asia.⁹ The older form continued to spread after the appearance of the Mujaddidīs, reaching Chinese-speaking Muslims in the late seventeenth century. But, outside of China, the Mujaddidī variant gradually eclipsed its progenitor in most places, including its original home of Central Asia, to the extent that most of today's Naqshbandī orders are Mujaddidī.¹⁰ The Chinese Muslims described in this study were already exposed to, and in many cases followers of, the older Naqshbandī path when Mujaddidīs arrived to spread Aḥmad Sirhindī's vision. As this article will demonstrate, in China proper (roughly today's People's Republic of China (PRC), excluding Tibet, Eastern Turkistan, and Inner Mongolia), the line between the two has become blurred over last two centuries.

Since the publications of Joseph Fletcher, the interaction of Northwest Chinese Muslims with religious movements of the greater Muslim world throughout the seventeenth and eighteenth centuries has been firmly established. Fletcher demonstrated this through the examples of two (non-Mujaddidī) Naqshbandiyya Sufi lineages: Eastern Turkistan's Afāqī Khwāja lineage, which spread within China most famously via Ma Laichi (马来迟, circa 1681–1766); and the Naqshbandīs of Zabid, Yemen, among whom ʿAbd al-Khāliq (circa 1705–40), a master in the lineage of Ibrāhim b. Ḥasan al-Kūrānī, trained the Chinese pilgrim Ma Mingxin (马明心, 1719–81).¹¹ These examples demonstrated the importance of both Eastern Turkistan, which served as a node of Chinese interconnection with Central Asian Sufism, and the power of the pilgrimage to Mecca, a means by which Chinese Muslims often acquired learning that spawned new movements in China. To these examples must be added a third Naqshbandiyya branch—the Mujaddidiyya, which linked the Muslims of Northwest China not just to Central Asia and Mecca, but also to Afghanistan, India, and Indonesia. This branch first arrived in Eastern Turkistan in the eighteenth century and spread to Gansu and Qinghai in the nineteenth century, before being fortified by further Mujaddiyya arrivals from Central Asia and, later, both the Volga region and the Hijaz. In Gansu, Qinghai, and further on throughout China, the foundational Mujaddidī text, the Maktūbāt of Aḥmad Sirhindī, spread beyond the followers of the order and entered the advanced curricula of Islamic learning among other Sino-Muslim sects, including communities that identify as Gedimu, or traditionalist.

The substantial presence of the Mujaddidiyya among the Hui ethnic group, which accounts for the bulk of China's Sinophone Muslims, has not been recognised in English-language scholarship. Even individual Mujaddidī groups have largely escaped notice, with the notable exception of Matthew Erie's recent ethnographic account of the Mingde (明德) community in Linxia—a group that had begun calling itself Mujaddidī by the time of Erie's visits in the 2000s.¹² A handful of studies of individual non-Mujaddidī communities have noted the appearance of Mujaddidī texts, but the connections between these traces have yet to be drawn.¹³ The strongest recognition of the order I have seen is from Kawamoto Masatomo and Nakanishi Tatsuya, who, in a Japanese-language fieldwork summary, report that three Hui gongbei (tomb shrines), including Mingde, are linked to Mujaddidī teaching chains.¹⁴ In contrast to the limited notice of Mujaddidī traditions among the Hui, important scholarship has already identified substantial Mujaddidī traditions among Turkic and Mongolic ethnic groups within the borders of the PRC, mainly the Uyghurs, Salars, and Dongxiang. The works of Thierry Zarcone, Alexandre Papas, and Waleed Ziad have been crucial in uncovering the transmission of Mujaddidī lineages and traditions among these Inner Asian groups in Eastern Turkistan and on the northern edge the Tibetan Plateau, including parts of Gansu.¹⁵ The present article will demonstrate that Mujaddidī transmissions extended beyond Turkic and Mongolic groups to Sinophone Hui communities across China, arguing that the Mujaddidiyya have had a substantial role in the shaping of Chinese Islams, and that general histories of the Mujaddidiyya should be expanded to include China proper. The article will also introduce new sources on the Mujaddidiyya among Turkic and Mongolic groups, sources which expand our understanding of both the Mujaddidiyya among those groups and the history of transmission to the Hui and China proper.

The notions of ‘transmissions’ and ‘contributions’ to Chinese Islams require some elaboration. Used carelessly, these terms might suggest a unidirectional outside ‘influence’ on an essentialised, stable tradition of Chinese Islam. However, Muslims in China practised a wide range of continuously transforming Islams—sometimes competing, frequently overlapping, and usually influencing each other. At the distant, ‘foreign’ end of the exchange in question here, the Naqshbandiyya Mujaddidiyya did indeed emerge in a clearly delimited place and time: Northern India in the sixteenth century. Many of its early leaders were committed to spreading their teachings, authority, and institutions to new territories. But it would be a mistake to equate the origins of the Mujaddiyya with a transhistorical essence and neatly identify Mujaddidī phenomena in China with the ultimate origins of the movement in India. The Mujaddidī traditions that arrived in Eastern Turkistan in the mid-eighteenth century were not quite the same as what Aḥmad Sirhindī had developed. Subsequent transmissions onward from Eastern Turkistan to China each took their own distinctive forms, based partly on the constantly changing nature of the tradition in Eastern Turkistan and partly on the local needs of worshippers in the Chinese locales where they took root. In some cases, these communities maintained the commitment to lineage and formal organisation that had been present at the earliest manifestations of the Mujaddidī movement in northern India, even keeping the name. In other cases, the division of Mujaddidī lineages into sub-branches led to the coining of new lineage names and the forgetting of the term ‘Mujaddidī’ as a lineage label, even where lineages and teachings were explicitly traced back to Aḥmad Sirhindī, ‘mujaddid-i alf sānī’. In yet other examples, the Mujaddidī foundational text, the Maktūbāt of Aḥmad Sirhindī, gained reverence among leading Muslim intellectuals but without association to a formal order; the text was simply seen as a source of valuable teachings for people who trace their spiritual lineages to other origins. Finally, it is important to emphasise the shared agency of transmission. In some cases, the arrival of Mujaddidī traditions in new places was the work of proselytisers from abroad, while in others it was a result of Muslims from the Chinese interior travelling to seek learning (and authority) in foreign lands, whence they brought Mujaddidī associations back to China. Often it was a product of multiple journeys, re-enacted across centuries, in both directions.

It has not always been the case that Mujaddidī transmission, even when direct and well documented, was clearly reflected in the beliefs and practices of the tradition's inheritors. As we will see, one of the communities that most clearly and formally traces its beginnings to the Mujaddidī order does not show much evidence of investment in the teachings of Aḥmad Sirhindī. On the other hand, the Huasi Khāfiyya order, which does not trace any connection to Sirhindī or his order, gives Sirhindī's Maktūbāt a prominent place in its teachings. In any case, a complete account of the role of Sirhindī and other Mujaddidī thinkers’ teachings in the various communities discussed below would require long-term ethnographic fieldwork that has not been possible during the research for this article. Self-identification is also a problematic measure of Mujaddidī links because, in most cases, the term ‘Mujaddidī’ does not seem to have survived translation into Chinese. Rather than attempting to draw lines between ‘authentic’ Mujaddidī communities or restricting this account to communities that literally self-identify as ‘Mujaddidī’, I present here an account of evidence for a wide range of transmissions that are traceable to the movement begun by Aḥmad Sirhindī, be their linkages genealogical, textual, or organisational, in the hopes that this article can serve as a starting point for further research on the varying types of Mujaddidī presence among Chinese-speaking Muslims.

Mujaddidī lineages and texts among the Hui and their ancestors

There is no scholarly consensus on how to refer to Muslims whose first language is/was Chinese, or who share a great deal culturally with non-Muslim Chinese peoples. The terms Huihui, Hui, Sino-Muslims, Sinophone Muslims, and Muslim Chinese are used by different scholars, sometimes in reference to differently delimited historical groups. ‘Muslim Chinese’ communities, however defined, have changed radically over the centuries, as have the terms by which members of these communities identified themselves, such as hui, huihui, huijiao ren, and hanzu musilin. Nor do essentialist characterisations of the modern Hui, the ethnic group commonly described as ‘Muslim Chinese’, withstand synchronic scrutiny across space. Dru Gladney's ethnographic survey of the Hui showed that many do not even speak Chinese as their first language, while others do not consider themselves followers of the Islamic faith.¹⁶

Nonetheless, the modern Hui identity, though constructed, is today a powerful social reality, undergirded by state support. And the vast majority of Hui people both speak a Chinese language as their mother tongue and consider themselves to be Muslims. Most Hui people also believe that the category rests on a primordial and essential set of characteristics, most notably descent from Muslim immigrants to China. The Hui minzu, or ‘nationality’, is one of 56 ethno-national groups officially recognised by the government of the PRC, and it appears on the mandatory identity cards of those identified by the government as Hui. The social reality of the category in the PRC era offers a firm foundation upon which this article bases its categorisations.

The article therefore begins each of its examinations of Mujaddidiyya phenomena among the Hui from the period of the last 40 years, during which time the Hui ethnic identity has been supported by the state, and then traces the historical transformations that led to the current Mujaddidī presence among the Hui. For the pre-PRC era, it becomes more difficult to categorise identity groups. However, from at least the seventeenth century onward, many, though not all, of the ancestors of today's Hui called themselves Huihui. I therefore use the term Huihui for people and groups who are likely to have seen themselves or been seen by their neighbours as Huihui, with the caveat that such terminology was not consistently deployed and does not always neatly align with the modern, official PRC category of Hui, despite substantial continuity.

Formal Mujaddidī orders among the Hui

The formal Sufi lineages who trace their roots to Aḥmad Sirhindī are well documented in Chinese-language scholarship on the Hui. However, the connection between these communities and the global Mujaddidī movement has generally not been recognizeised. In most cases, the lineages are traced to ‘Yimamu Ranbani’ or ‘Yimamu Rebani’ (Sinicisations of Aḥmad Sirhindī's honorific, Imām Rabbānī) without identifying the figure as Aḥmad Sirhindī. With two exceptions, the terms ‘mujaddidiyya’ and ‘mujaddidī’ do not appear in Chinese-language scholarship on these orders, which tends to categorise them instead as sub-branches of the Khāfiyya order (described below), most notably in the influential work of Ma Tong.¹⁷ The first of the two exceptions is Tan Wutie and Fu Yu's 1986 survey of Islam among the Hui of Eastern Turkistan, which they later developed into a book.¹⁸ Through fieldwork with Hui religious practitioners, this valuable work identified links between various Hui groups and the lineage of Aḥmad Sirhindī. While they described Sirhindī's thought and placed several lineages of China and Eastern Turkistan in this category, they did not mention that Sirhindī's disciples and descendants spread the movement to other destinations beyond China, from Indonesia to Istanbul. Another article, by Chen Guoguang (陈国光), took the further step of recognising that lineages traceable to Sirhindī are links to a global Mujaddidī tradition.¹⁹ However, both studies focused on Mujaddidī orders in Eastern Turkistan and, while they listed the names of several orders from interior China that descend from Eastern Turkistan's Mujaddidī lines, they did not analyze or describe those Chinese orders in any detail. Nonetheless, their analysis of oral material among the Hui of Eastern Turkistan provided crucial data on the links between the region's Mujaddidīs and the Hui of China proper. Unfortunately, these researchers’ scholarship has not been taken up outside the study of Hui Muslims in Xinjiang itself. Thus, the various groups identified in Chinese-language scholarship as descending from ‘Yimamu Ranbani’ have not widely been recognised as historically connected to the Mujaddidī order, and they continue to be seen in studies of Islam in China proper as ‘Hufuye’ (Khāfiyya) orders.

Available primary sources from the Chinese Mujaddidī orders themselves use only the names of the sub-branches, such as Hongmen and Dingmen. Whereas suborders elsewhere in the world will often append a further name or names, for example the ‘Naqshbandī Mujaddidi Aslami Arshadi’ in the UK,²⁰ the Chinese Mujaddidī orders have for a several generations used only the narrowest sub-branch designation, dropping both ‘Naqshbandī’ and ‘Mujaddidī’ when referring to their own orders. In most sources, whether primary or secondary, the reader's only hint that a Sufi order is connected to the Mujaddidiyya is the inclusion of ‘Yimamu Ranbani’ in the spiritual lineage chain. All of these factors have combined to obscure the significance of the Mujaddidīs among the Hui.

At multiple points from roughly 1800 onward, seekers from China proper travelled to Eastern Turkistan and, in some cases, the Hijaz, bringing back Mujaddidī teachings to their home regions. A handful of formal Sufi orders, called menhuan (门宦) in Chinese, trace their origins to these transmissions (see Table 1). Among the Hui, these include the Dingmen (丁门), Hongmen (洪门), Jinggou (井沟), Lintiao (临挑), and Beizhuang (北庄) orders.²¹ Of the Hui formal Mujaddidī orders, the two best-documented are the Dingmen, centred in Gansu, and the Hongmen, which is based in Ningxia. Both trace their initiatic chains (silsila) to Aḥmad Sirhindī via shaykhs in Eastern Turkistan. Their Eastern Turkistani transmitters represent two distinct lines (Ayköl and Yarkand), each with its own history of dissemination from South Asia, described in the second half of this article. I present here a short overview of these two groups (Dingmen and Hongmen) to provide a sample of the ways that formal Mujaddidī orders have developed among the Hui.

Table 1. Mujaddidī transmissions to Eastern Turkistan

The Dingmen order is probably the smaller of the two, estimated to have ‘over 10 thousand’ followers in the 2000 edition of Ma Tong's comprehensive survey of Sufi orders in China.²² In 1980, the Dingmen community leader in Lintan (临潭) provided roughly similar numbers, estimating that about 2,000 families were spread across Gansu, Qinghai, Sichuan, and Xinjiang, plus an unknown number of additional followers in Ningxia and the far north-eastern province of Liaoning.²³ The major gongbei (tomb shrines) of the Dingmen order are located in Ningxia (Yinchuan 银川), Gansu (Lintan 临潭, Kangle 康乐, and Lanzhou 兰州), and Qinghai (Minhe 民和).

No written primary sources predating the 1980s have been documented for the Dingmen order, though scholars have pieced together a reasonably detailed history of the order from oral sources.²⁴ Despite the recentness of the sources, the Dingmen leader's claims to a chronologically deep affiliation with a Sirhindian lineage ring true, especially because they trace the lineage through the Mujaddidīs of Eastern Turkistan, which flourished in the early nineteenth century but dwindled into obscurity in the early twentieth century. The eponymous leader, Ding Xiang (丁香, 1728–1819), is said to have embarked on the pilgrimage to Mecca in 1768. He stopped in Badakhshan and studied with a Naqshbandī shaykh before completing the pilgrimage. On the return journey, he stopped for one year in Kashgar and Yarkand, where he studied with ‘the descendants of Yimamu Ranbani’—clearly a reference to the Yarkand Daotang, a Mujaddidī suborder discussed in detail further below.²⁵ There Ding earned an ijāzat (licence) to spread the teaching, which he then did in what is today Gansu, Sichuan, Ningxia, and Qinghai. Leadership of the order has passed from father to son into the twenty-first century. Over the centuries, each of the Ding leaders recognised their own master(s) from the wider Mujaddidī order: shaykhs from ‘Afghanistan’ for the first two, and shaykhs from Kashgar and Yarkand for the rest. These relationships were enabled by pilgrimages to the Yarkand Daotang, often paired with the hajj. The continuous communication with and subordination to the Yarkand Daotang supported that continuous sectarian awareness despite the loss of the ‘Mujaddidī’ name, as the order continued to see itself as an offshoot of the Yarkand Daotang.

The ninth-generation shaykh, Ding Shijun 丁士俊 (born 1935), penned a very brief summary of the history, teachings, and practices of the order, including quotations from the founder, Ding Xiang. Two of these quotations address the challenges of preserving religious tradition in a non-Muslim, Sinophone land, such as ‘understanding Chinese but not understanding Arabic or Persian, this cannot be considered having knowledge; understanding those two languages but not understanding Chinese is the equivalent of a mute person [trying to] speak’. Another purported quote from the founder discusses the value of local philosophical traditions: ‘In researching Confucianism, Buddhism, and Daoism, [though] each has its strengths and weaknesses, one cannot mix Islam together with them.’ Ding's summary goes on to outline a rigorous schedule of worship and emphasises the avoidance of sensory indulgences, including the consumption of garlic, but it is too general to pick out teachings particular to any Mujaddidī tradition.³⁶

Sources on the Hongmen order, by contrast, offer clear signs of connection to specific Mujaddidī practices and beliefs. The Hongmen order is centred near Tongxin, Ningxia, with followers numbering something over 20,000, mostly spread between the counties of Tongxin (同心), Haiyuan (海原), and Guyuan (固原).³⁷ Enormous annual pilgrimage festivals suggest that this estimate of follower numbers, by Ma Tong, refers only to those who closely identify with the order, perhaps even those who regard themselves as formal initiates into the menhuan. During the death day of the founding saint, enormous numbers of pilgrims visit the gongbei at Honggangzi (洪岗子), near Tongxin. The geographer Kentaro Takahashi participated in a pilgrimage to the gongbei in 2002, two days before the actual death day, and estimated that over 20,000 pilgrims were present, including groups from distant Xinjiang.³⁸ A Chinese online report of uncertain reliability estimated 200,000 visitors for the entire week-long pilgrimage festival.³⁹ To accommodate this event, which includes ritual distribution of celebratory meals (ermaili 尔麦里), the order has constructed one of Ningxia's largest tomb/mosque complexes. The main ritual complex covers approximately 18,000 square metres, not including the industrial-scale cooking facilities and other support buildings.⁴⁰ It is not at all unusual in Northwest China for an influential gongbei to draw pilgrims from across sects, so this festival should not be seen as a marker of the number of formal members of the Hongmen order, but it does demonstrate strong interprovincial and probably cross-sectarian influence for the order.

The suborder takes its name from Hong Shoulin (洪寿林, 1858–1937), who was designated the successor of a shaykh known as Liangzhouzhuang Taiye (凉州庄太爷, 1822–98), originally from Qinghai. After years of study with Liangzhouzhuang Taiye in Gansu, the master sent Hong to spread the message eastward. Shortly before his death, Liangzhouzhuang Taiye told his wife that a man named Hong would come to visit his grave, and that man was to be his successor. Liangzhouzhuang Taiye in turn traced his spiritual lineage to the ‘Ayköl’ order—a Mujaddidī organisation based in the oasis of Aqsu in Eastern Turkistan. Sources from outside the Hongmen order mention a slightly earlier student of Liangzhouzhuang Taiye who established a branch in Tongxin, so it may be that Hong Shoulin was already educated in the order before setting out to study with Liangzhouzhuang Taiye.⁴¹

A description of the Hongmen order's practices and beliefs, published in 1990, shows distinctive marks of Naqshbandī Mujaddidī spiritual exercises, especially in its account of dhikr—a ‘remembrance’ of God that takes the form meditative recitation, usually of short formulae such as the name of God.⁴² Dhikr is a common feature of various Sufi orders both within China and across the world, but the precise form of dhikr varies from one order to the next. In the Hongmen scheme, dhikr is divided into two categories: yisimuzanti (伊斯目咱提) and naifeiyisibati (乃非伊斯巴提).⁴³ The former is the recitation of ‘Allah’, known elsewhere in the Persianate sphere as ism-i ẕāt (name of the Essence). The latter is a method of meditation on the phrase ‘there is no god but Allah’, called in Persian nafī-is̠bāt (negation and affirmation). During the initiation of new followers, the shaykh presses his finger to three points on the initiate's body in succession to teach him where to direct his concentration during nafī-is̠bāt. ⁴⁴ These points are the laṭā’if (subtle centres), rendered in Chinese as xuewei (穴位). While versions of these techniques appear in the practices of various Sufi orders outside of China, the particular terminology and categorisation reflect the Naqshbandī Mujaddidī discipline as it is still practised in many parts of the world today.⁴⁵ The same source also presents general philosophical approaches that are hallmarks of the Mujaddidī tradition and Aḥmad Sirhindī's intellectual legacy. It says that the Hongmen order promotes the integration of dao (道: tariqa, the Sufi path) and jiao (教: sharīʿa, divine law): ‘like core and shell, jiao is the basis of dao, dao is the highest stage of jiao.’ The follower should cultivate the Sufi path without escaping from practical life—a task enabled by the more general Naqshbandī practice of khilwat dar anjuman.⁴⁶

A Hongmen religious scholar in Lanzhou preserves an Arabic manuscript silsila, which has been published in Chinese translation and confirms the link to the Ayköl lineage.⁴⁷ Elsewhere among the Hongmen, memorisation and recitation of the silsila are required of followers. The initiatic names (道号) of masters in the silsila are kept secret and chanted by initiates for their miracle-working power.⁴⁸ The Lanzhou manuscript silsila presents numerous other branches of transmission from the Ayköl lineage, suggesting that the Hongmen suborder is one among many. Further research is likely to show that numerous other formal suborders exist, overlooked by scholars due to their categorisation as branches of the Khāfiyya.⁴⁹ The Dingmen and Hongmen orders stand out because they have been described in some detail by Chinese scholars, but other ‘Khāfiyya’ suborders will probably be discovered to have Mujaddidī roots.

The fact that all of the formal Mujaddidī lineages in China proper have dropped the designations ‘Mujaddidī’ and ‘Naqshbandī’ reflects an adaptation to the Sufi landscape of Northwest China (today's Gansu, Qinghai, Shaanxi, and Ningxia). This region had already seen intense competition, sometimes violent, between Sufi suborders before the Mujaddidī lineages took root. Over the last 300 years, Sufi lineages have frequently split and then split again, often amid disputes over the transmission of leadership (and the attendant resources), usually taking new names with each split. The resulting communities have come to be known as menhuan, and the emphasis on rivalries between (sub)orders has often superseded affinities based on shared doctrine or common spiritual ancestry, leading to what Chinese scholars characterise as the menhuan system. Most commonly, the naming of new sub-branches mirrored Chinese systems of clan and hometown affiliation. In other cases, they adopted the names of the mosques or khanaqahs that served as headquarters.⁵⁰ The Mujaddidī lineages identified here follow the more common pattern, with Dingmen and Hongmen based on surnames, while Jinggou (井沟), Lintiao (临挑), and Beizhuang (北庄) are named after places. Of course, Naqshbandī Mujaddidī communities outside of China have engaged in their own rivalries, but the overwhelming sense of a shared inheritance from Sirhindī has tended to survive as focal point of group identity. The fissiparous menhuan system's co-optation of Mujaddidiyya groups in Northwest China would thus seem to represent a distinct manifestation of the Mujaddidī tradition. The weak presence of recognisably Sirhindian belief and practice among some of these groups may be a further product of the localisation of formal orders in Northwest China, though confident conclusions in this realm will have to await further ethnography and historical research.

Mujaddidī textual authority beyond the formal order

Beyond the formal Mujaddidī orders, the teachings of Aḥmad Sirhindī have spread widely via his written work. Indeed, the impact of the Mujaddidiyya among the Hui may be more substantial outside of the formal orders, as it extends even beyond groups that consider themselves part of Sufi traditions. The Maktūbāt (Collected Letters) of Aḥmad Sirhindī, which serves as a foundational textual resource for the Mujaddidiyya order around the world, is widely read and taught among various Hui groups. In each group, it plays a slightly different role. Here I outline two groups that have taken up the Maktūbāt, both of which are based in Linxia, Gansu—a town known both within and without the Hui community as ‘China's Little Mecca’, due to its centrality for several denominations and orders. The first case is a scholarly lineage that, as of the 1980s, self-identified as Gedimu (Ar: Qadīm): Hui Muslims who regard their teachings as ‘old’ orthodoxy passed down continuously and maintained in the face of ‘new teachings’ (xin jiao 新教) such as modernist reform movements that include the Salafi and Ikhwānī, and sometimes also in contradistinction to Sufi orders.⁵¹ In the Gedimu example presented below, the Maktūbāt served not just as a source of authoritative knowledge, but also as a marker of personal authority, asserted through claims to a special role in the transmission of the text. In the early 2000s, a new generation of the lineage endeavoured to reconnect with the Maktūbāt's origins, making pilgrimages to Sirhind and representing the lineage as a branch of the Mujaddidiyya. The second case is the leading branch of the Khāfiyya order, for whom the ideas conveyed in the Maktūbāt provide important justification for the sectarian markers that differentiate the Hufuye from its historical rival, the Jahriyya (Zheherenye 哲合忍耶).

Gedimu to Mujaddidi

Kamāl al-Dīn Qi Mingde (祁明德, 1894–1987), popularly known as ‘the deaf ahong’, was an ahong (religious scholar, Persian: akhund) from Linxia, known in his later years as one of ‘the three mainstays of the old teachings’.⁵² He had a reputation for defending traditionalist and Sufi communities against reformist movements that spread in the north-west during his lifetime and for promoting non-interference between different schools of Islamic thought—a position directly at odds with the reformists. Qi's renown expanded after his death thanks to an anthology of his writings published in 1996 under the title The Deaf Ahong.⁵³ Despite being published as ‘internal circulation materials’—a designation usually associated with localised distribution—these works have circulated widely throughout China.⁵⁴ A few years before his death, Qi Mingde founded a new mosque, which is named the Mingde Mosque in his honour and is listed as the publisher of The Deaf Ahong.

In his memoir cum family history in The Deaf Ahong, Qi grounds his authority not only in his education—he had studied in Mecca, among other places, and was said to be a master of Islamic scholarship in Arabic and Persian—but also in his lineage. Lineage claims remain a common source of religious legitimacy, and thus also of disputes, among both Sufi and non-Sufi Muslim scholars throughout China. In some cases, lineage-based claims are rooted in chains of teachers and students, but they are very commonly based also on blood lineage, with each link in the chain simultaneously representing both a father–son and a teacher–student relationship. Qi's lineage followed the latter pattern. Today, his family name denotes both a neighbourhood in Linxia and the Great Qi Mosque, originally established, according to The Deaf Ahong, in the seventeenth century.

Much of Qi's claim for the centrality of his lineage rests on the transmission of the Maktūbāt. In The Deaf Ahong, Qi begins the story of his lineage with the Linxia ahong Qi Xinyi (祁信一), Arabic name Ibrahīm Sunnī (1651–1742), whom the text describes as follows. Ibrahīm was a master of Islamic learning in Persian and Arabic, and a sought-after teacher across several provinces. While teaching in Sichuan, Ibrahīm learned that the (non-Mujaddidī) Naqshbandī saint Afāq Khwāja had arrived from Eastern Turkistan to preach in Xining. Ibrahīm set out to seek teaching from Afāq in Xining, only to learn that the master had already returned to Eastern Turkistan. He then set a new course for East Turkistan himself, travelling with Ma Jiajun (马家俊), father of the founder of the main Khāfiyya Naqshbandī branch in China. When they found Afāq Khwāja, the saint sent Ma Jiajun back to Gansu, but recognised a special character in Ibrahīm and asked him to stay. For seven months, Ibrahīm learned from Afāq. Upon completion of his study, he received a license (ijāzat) to transmit the teaching and a copy of the Mujaddidī foundational text, Maktūbāt. However, Afāq forbade Ibrahīm from transmitting the ijāzat to his children and, when Ibrahīm died, he took the ijāzat with him.

Qi Mingde's story of the presentation of the Maktūbāt to his ancestor is in one way consistent with what we know from older historical sources about Afāq Khwāja, but in another it is at odds with those sources. Afāq Khwāja (d. 1694), along with his father Muḥammad Yūsuf, is widely credited with transmitting (non-Mujaddidī) Naqshbandī Sufism to Gansu and Qinghai.⁵⁵ Nineteenth-century Turki and Persian-language hagiographies written in Eastern Turkistan by followers of Afāq Khwāja repeatedly depict the bestowal of a book as part of Afāq's ritual for designating khalifas—followers who are licensed to transmit his teaching and authority.⁵⁶ However, in these Afāqī manuscripts, the book presented is not the Maktūbāt, but instead Rūmī's Masnavī. There is no evidence that Afāq had any contact with Mujaddidī teachings, nor does he appear in any other source as a link in a Mujaddidī silsila, or chain of authority. It appears from this that the Qi lineage inserted the Maktūbāt into the widely known tradition of Afāq's Naqshbandī transmission at some later point. This seems all the more likely given that all documented Mujaddidī transmissions occurred from the mid-eighteenth century onward, whereas Afāq died in 1694. In any case, it is clear that, for the Qi lineage, the Maktūbāt was more than just one philosophical and religious text among many. It was the central symbol of the transmission of religious authority and thus a representation of the religious identity of the lineage.

The receipt of a book from a foreign master serves as a legitimating story in other Sinophone traditions as well. The Xianmen (鲜门) founder is also said to have received books from Afāq Khwāja.⁵⁷ Ma Laichi, regarded as the founder of the Khāfiyya order, received an Arabic text called Minshār from his teacher Shaykh ‘Aqīl’ in the Yemen, which is central to the identity and rituals of the order.⁵⁸ And the Jahriyya founder, Ma Mingxin, is said to have brought back from Yemen Mukhammas and Madāʿih—the two central texts upon which the Jahriyya have elaborated their distinctive rituals.⁵⁹

Qi Mingde's narrative in the first volume of The Deaf Ahong skips over the life stories of the succeeding seven generations of Ibrahīm's descendants, merely listing their names. Qi picks up the narrative thread again with his father Yūsuf Qi Huantang (祁焕堂1852–1933), the ninth-generation descendant of Ibrahīm. After a few lines describing Qi Huantang's deep education and pilgrimages to Mecca, the narrative turns to his study with a descendant of ‘Yimamu Ranbani’ and his purportedly central role in facilitating the translation of the Maktūbāt from Persian into Arabic. At the age of 43, Qi Huantang learned that a certain ‘Mulaji Afanji’ (穆拉吉阿凡吉, Mulla Haji Effendi?), a descendant of Yimamu Ranbani (Aḥmad Sirhindī), was visiting Eastern Turkistan (by this time part of the Qing empire) from ‘Central Asia’. As his ancestor had done, Qi Huantang walked to Eastern Turkistan and found the teacher he was seeking. Mulaji Afanji gave Qi Huantang copies of Rashāḥat ⁶⁰ and Tafsīr Husaynī, both in Persian. Qi Huantang told Mulaji Afanji the story of his ancestor receiving the Maktūbāt from Afāq Khwāja. The teacher was impressed and ‘praised the ancestor's dao [way, school, ṭarīqat], saying: he attained the Dao early, was of high quality, and successfully sought the Faith—it is a true rarity’.⁶¹

Qi Huantang then pleaded with Mulaji Afanji to translate the Maktūbāt, ‘which his [Qi's] family had kept for nine generations’ from the original Persian into Arabic. Mulaji Afanji replied that he would need the permission of his master. Qi Huantang then travelled with Mulaji Afanji to ‘Indonesia’ to meet the master, identified as Zewawei (则瓦为), who in turn said he could not permit the translation without permission from his own master in Mecca, Haimiji Afanji (海米吉 阿凡吉). The two then travelled with Zewawei to Mecca, where Haimiji Afanji gave approval for the translation. The group then visited ‘Shaykh Maʿṣūm of the Khāfiyya order’, who informed them that they would need the king's permission to publish the book. This they obtained, and the translation was published in 1316 AH (1898–99 CE).

The interaction between Qi Huantang and Mulaji Afanji follows the pattern of the earlier ancestor and Afāq Khwāja in a general way, lending authority to the lineage through pilgrimage to a master in Eastern Turkistan. However, it is not quite as momentous. Mulaji Afanji gives books to Qi Huantang, but does not give an ijāzat. There is no clear bestowal of formal authority, though the narrative clearly presents Mulaji Afanji as an authoritative shaykh and purported descendant of Sirhindī, whose praise of Qi Huantang is a credit to the lineage.

Zewawei likely refers to Abdallah al Zawāwī—a Meccan sayyid who travelled widely and resided for some time in Pontianak, Borneo during the 1890s. Before his exile from Mecca, Zawāwī, along with his father, had already transmitted the Naqshbandī Mujaddidī path to many Muslims of the Dutch East Indies. The reference to a Mujaddidī shaykh by the name of Zawāwī residing in what would come to be known as Indonesia thus indicates that the narrator had specific knowledge of Mujaddidī networks of the late nineteenth century. This level of accuracy also suggests that, despite the errors in other parts of the narrative, notably the presentation of Afāq Khwāja as a Mujaddidī, Qi Huantang probably did visit Indonesia on the way to Mecca.

The name Mulaji Afanji is not particularly helpful for identification. It may be a transliteration of Mullā Hajjī Efendi, which is a combination of honorifics common at that time in Eastern Turkistan as well as other parts of Central Asia. Even if this is accurate, it offers little help in identifying Mulaji Afanji. However, two other data points make clear to whom the text refers: the text says ‘Mulaji Afanji’ published the translation of the Maktūbāt in 1316 AH in Mecca, and he was a student of Zawāwī. These two characteristics apply only to one individual: Muḥammad Murād Ramzī al-Qazānī al-Manzilāwī (1855–1934/35),⁶² a well-known and well-travelled scholar from Tatarstan, at that time part of the Russian empire. Manzilāwī studied in Bukhara, spent many years in Mecca, where he studied under Mujaddidī masters, and spent the final 15 years of his life teaching in Chuguchak, in Eastern Turkistan. Manzilāwī's translation of the Maktūbāt was the first into Arabic and, to this day, it has not been superseded. Notably, Manzilāwī also translated Rasḥāhat ʿAyn al-Ḥayāt—one of the books he supposedly presented to Qi Huantang.

What is known of Manzilāwī's biography presents some chronological problems for the account in The Deaf Ahong. Manzilāwī's memoirs, accessible only via an extremely brief summary by the late Turkish scholar Ahmet Temir, report that Manzilāwī moved to Eastern Turkistan in 1919, some 20 years after the publication of his Maktūbāt translation.⁶³ Muḥammad Emin Bughra, one of the founders of the short-lived Islamic Republic of Eastern Turkistan (1933–34), claimed to have met Manzilāwī in Chuguchak even later, in 1930.⁶⁴ It is possible that Manzilāwī made an earlier, undocumented trip to Eastern Turkistan. It is equally likely, however, that The Deaf Ahong narrative conflates separate incidents in the biography of Qi Huantang. Qi made multiple pilgrimages to Mecca. He may have passed through the Dutch East Indies on his early hajj and met with Manzilāwī during later travels. The possibility that the meeting is an embellishment cannot be entirely excluded but, given the extensive travels of Qi Huantang and the accuracy of many specific data points (e.g. the presence of Manzilāwī in Eastern Turkistan), an actual meeting between the two seems probable. By connecting the meeting with Manzilāwī to the translation of the Maktūbāt, the narrative squeezes the historical material into the common Hui Sufi trope of an ancestor meeting and receiving authority from a foreign Sufi master, while at the same time giving the Qi lineage claim to a second special connection to the Maktūbāt.

Qi Mingde identified himself as a follower of the Gedimu, which he explained as the ‘venerable ancient denomination’ (古派) and the ‘old teaching’ (老教), ‘in terms of doctrine, part of the orthodox Sunni sect, and in terms of law, Hanafi’.⁶⁵ The Gedimu, he said, supports the (Sufi) menhuans, while the menhuans do not diverge from the Gedimu, and the two together are of one vein. He also called his father, Qi Huantang, an ahong of the ‘old teaching’. There is no indication in Qi Mingde's memoir and family history of formal affiliation with the Mujaddidiyya, nor any place in a silsila. The ijāzat that the ancestor Ibrahīm received from Afāq Khwāja was never passed on, and no ijāzat from Manzilāwī is mentioned. Qi Mingde was not reluctant to express his support for the Sufi menhuans, and he records in his memoir his concrete actions to defend the menhuans from attacks by reformists and their government supporters in the republican era. But he never claims in his writings membership in any menhuan, instead presenting his lineage's interactions with Mujaddidī figures as sources of learning and authority that strengthened his claim to be the defender of the Gedimu ‘old teaching’.

After the death of Qi Mingde, his son Qi Jiequan (祁介泉, d. 2012) assumed leadership of the Qi lineage and its community. Qi Jiequan began an effort to reconnect with the figures to whom the authority of his lineage was tied. Saying that his father had told him to seek out Shaykh Maʿṣūm, he made multiple trips to India, where he undertook pilgrimages to Sirhind.⁶⁶ Upon Qi Jiequan's death, three new volumes of Long Ahong were issued. One of them includes a silsila, naming Qi Huantang, Qi Mingde, and Qi Jiequan as the most recent links in a Mujaddidī chain. The chain passes through Shaykh Maʿṣūm via a successor named Abū Sharīf ʿAbd al-Qādir. This silsila is quite distinct from the Qi scholarly family lineage presented by the ‘Deaf Ahong’, Qi Mingde. Unlike the earlier work, which traces the Qi lineage back to the seventeenth generation, in the silsila, the Qi family only appears with Qi Huantang in the twentieth century. Under Qi Jiequan's leadership, members of the community used the term ‘Mujaddidī’ and made pilgrimages to Sirhind. As of 2017, a large, Chinese-style incense urn, inscribed with the name of the Mingde Mosque (founded by Qi Mingde), graced the courtyard outside of Ahmed Sirhindī's tomb.⁶⁷

For Qi Mingde, writing in the 1980s, the Maktūbāt and the stories of its bestowal were foundations of the Qi lineage's claim to special, inheritable authority, but the community seems to have identified primarily as Gedimu. His son, who enjoyed the benefits of an increasingly open and prosperous China, was able to seek the roots of these authoritative texts and connections. The result appears to be a more formal association with the Mujaddidī order as an international movement, with authority rooted in the silsila as much as the older Qi family lineage.

Mujaddidī textual authority among the Khāfiyya, Jahriyya, and Beyond

Another case from Linxia illustrates the role of the Mujaddidiyya at a step further removed from the formal transmission of Mujaddidī affiliation. At least two Chinese translations of Sirhindī's Maktūbāt have been published—one from the Persian original and one from Manzilāwī's Arabic translation. The latter translation's history demonstrates a strong interest in the Maktūbāt on the part of the Huasi menhuan, regarded as the leading branch of the Khāfiyya. Unlike the case of the Qi lineage, in which there is a claim to a formal transmission of the Maktūbāt, the Khāfiyya embrace of the Maktūbāt appears to be separated entirely from the initiatic chain of the Mujaddidiyya, rooted instead in the ideas of the text as a theological and ritual resource.

The Khāfiyya menhuan is the older of the two main divisions of the (non-Mujaddidī) Naqshbandiyya among the Hui. The founder was the son of Ibrahīm Qi Xinyi's travelling companion, whom Afāq Khwāja supposedly sent home while Qi Xinyi was invited to stay (or so it goes in The Deaf Ahong). According to Khāfiyya tradition, this companion, Ma Jiajun, received a blessing from Afāq that alleviated his inability to have a son. After his return to Gansu, he married a non-Muslim woman according to Afāq's instructions and she gave birth to the founder of the Khāfiyya order, Ma Laichi (马来迟). The order that emerged from Ma Laichi's teaching was, presumably, regarded simply as Naqshbandī for its first century. However, in the middle of the seventeenth century, another Chinese Muslim scholar, Ma Mingxin, returned from the hajj with new ideas about Naqshbandī ritual practice, learned, as Joseph Fletcher demonstrated, from a master in Yemen. Those ideas included the promotion of vocal or ‘loud’ dhikr—the ritual of remembrance of God practised by the Naqshbandiyya and other orders. The existing Naqshbandiyya in China, by contrast, advocated silent dhikr. The ritual difference became the symbol and justification of contention over authority in Northwest China—conflict that burst into open and deadly warfare that came to be known, misleadingly, as the ‘Muslim rebellions’ of the late eighteenth and late nineteenth centuries. It also gave the two factions their names, with the Khāfiyya order taking its name from the Arabic word for ‘silent’ and the Jahriyya from the Arabic for ‘vocal’. Like every large menhuan in China, the Khāfiyya has in turn divided into several branches, most of which coexist in competition but not conflict. The largest of these is the Huasi menhuan, which controls the most prominent mosque in Linxia and the tomb-shrine of the Khāfiyya founder, Ma Laichi. As we have seen, several of the menhuan that consider themselves sub-branches of the Khāfiyya in fact place themselves in Mujaddidī silsilas. This is not the case with the Huasi branch, whose silsila proceeds through Afāq Khwāja .

The Huasi mosque houses a substantial public-facing library consisting almost entirely of editions from the last 40 years. When I visited in 2015, I was able to photograph its contents. Some of the fundamental texts were kept in multiple copies. Among them were 20 two-volume sets of the Wiqāya—perhaps the most widely circulated and preserved Islamic text in China after the Qur'an. Sitting next to the Wiqāya were ten two-volume sets of the Maktūbāt in Chinese translation. The introduction to the translation was written in 2005, and a second edition was published in 2010. The translator is the late Ḥasan Ma Hongzhan (马鸿章), leader of the Huasi menhuan and descendant of the order's founder, Ma Laichi.

In his introduction to the translation, Ḥasan Ma Hongzhan explains part of the Maktūbāt's special value to his community: in one passage of the work, Aḥmad Sirhindī recommends that dhikr be carried out silently, rather than vocally. Thus, the core Mujaddidī text functions as an ideological asset in the dispute between the Khāfiyya and Jahriyya, rooted again in a presumed authority of Aḥmad Sirhindī, even outside the initiatic chain. The significance for Ḥasan Ma Hongzhan is broader than this narrow sectarian concern, however. In his introductory remarks, Ḥasan says that translating the Maktūbāt into Chinese is his ‘life's aspiration’ and the ‘long-cherished wish’ of his own master.

The Huasi community is not the only Sufi order in Linxia to embrace the Maktūbāt. The Qādirī order is also widespread within China and headquartered in Linxia. Formal members of the order adopt celibacy and adhere to a daily programme of self-cultivation through textual study and dhikr. Maktūbāt is one of the main texts that both shaykhs and followers focus on for their daily study. The Huasi and Qādirī orders are the most prominent Sufi communities in Linxia.⁶⁸ Of the four main Sufi branches traditionally recognised in China—Khāfiyya (with Huasi as the main branch and caretaker of the founder's tomb), Qādiriyya, Jahriyya, and Kubrawiyya—only these two have their headquarters and founders’ gongbei in Linxia. This centrality, and the pilgrimages it inspires, likely goes some way to maintaining the widely held notion that Linxia is China's ‘little Mecca’. The fact that these two influential Linxia-based orders give Maktūbāt a prominent place can only lend power to the ‘Gedimu’ Qi lineage's claims of a special relationship with the Maktūbāt. In his Linxia-based ethnography, Matthew Erie listed Maktūbāt among the texts taught at numerous Gedimu Islamic schools of Linxia.⁶⁹ The Maktūbāt is thus very much a shared text among Linxia's main communities—a source of intellectual authority that has leapt the bounds of the formal ṭarīqat that brought it to China.

From Linxia's central position as ‘China's little Mecca’, the Maktūbāt's prominence extends outward. The Chinese Encyclopaedia of Islam calls Maktūbāt ‘one of the four most important Sufi religious texts’ in China, though without mentioning the Mujaddidī order.⁷⁰ The presence of the work in libraries throughout Islamic China supports the claim. At least two volumes of the Maktūbāt are held in the library of the Banqiao Daotang, headquarters of one of the main branches of the Jahriyya, located outside the city of Wuzhong in Ningxia.⁷¹ Florian Sobieroj reports that the work is read at other Jahrī centres as well.⁷² The library of the Yangjiazhuang mosque in Xining, Qinghai contains a two-volume edition. Another copy, an 1898 Arabic translation (presumably the one mentioned above), is preserved in the Fu‘ad library, formerly associated with the Chengda Muslim Teacher's Academy in Beijing.⁷³

The quotation of Sirhindī in Jahriyya historical sources suggests that this national presence a not recent phenomenon. Sirhindī is quoted in the earliest historical text of the Jahriyya order, the Rashaḥat al-Sharīfat—a mixed Arabic and Persian text written in the early nineteenth century.⁷⁴ Later Mujaddidī biographies are quoted at length several times in the 1936 hagiography, Risālah Aqṣarayyah li-Bayān al-Silsilah al-Jahriyyah (better known today as Kitāb al-Jahrī).⁷⁵ Looking beyond the Jahriyya, the Maktūbāt also appears in an early twentieth-century publication catalogue of the Shanghai Believer's Classics Company.⁷⁶ Exactly how the Maktūbāt (along with other Mujaddidī texts) has spread through the rest of China and what the full contours of its influence have been will have to await further research.

Mujaddidī thought among the Ikhwān

Despite the strong anti-Sufi positions that China's first modernist reform movement, the Ikhwān, has often promoted, the leaders of the movement have made wide use of Maktūbāt. The two most influential Ikhwān leaders have roots in formal orders of the Mujaddidī silsila, which may account for their familiarity with Sirhindī's text. The Ikhwān's founder, Ma Wanfu (马万福, 1849–1934), was raised in the Beizhuang (北庄) menhuan—a self-consciously Mujaddidī order described in the second half of this article.⁷⁷ Hu Songshan (虎嵩山, 1880–1956), who softened some of the Ikhwānī positions and incorporated Chinese nationalism, was the son of a master in the Ayköl line of the Mujaddidī silsila. Hu's father received authorisation from the same Liangzhouzhuang Taiye who transmitted the tariqat to the founder of the Hongmen order. It is said that Hu was offered leadership of the order but rejected it and even went on to raze his own father's tomb as part of the Ikhwānī rejection of shrine veneration.

Of these two Ikhwānī figures, Hu can be most clearly tied to the Maktūbāt. He published a detailed Chinese annotation of the Arabic translation in 1940.⁷⁸ In a 2015 study of Hu Songshan's teaching and writing, Di Liangchuan writes that Hu promoted Sirhindī's reform programme and used the Maktūbāt as an ‘ideological weapon’ in his conflicts with various Sufi menhuans and Gedimu traditionalists.⁷⁹ According to Ikhwānī accounts, the founder Ma Wanfu authored a now-lost text, Buhualizande (布华里咱德), that outlined his position on a comprehensive range of religious issues. The work is described as a selection of content from eight texts, one of which was the Maktūbāt. Although Buhualizande was destroyed, another work, Huijiao Bizun (回教必遵), is said to be a condensation of the lost text.⁸⁰ The Maktūbāt itself was also taught in Ma's subsequent lineage, which maintained broad leadership of the Ikhwānī movement. A 2018 eulogy of Ma's grandson and successor, Ma Zhangqing (马长庆1936–2018), lists the Maktūbāt among the texts that the deceased had mastered.⁸¹ The Maktūbāt probably appealed to Ikhwānī scholars because of its message of reform. As Mohammed al-Sudairi has argued, the founding Ikhwānis, despite their common association with Wahhabī fundamentalism, were also inspired by an earlier range of texts that promoted renewal in the Hanafi world, of which the Maktūbāt was one.⁸²

The Mujaddidī spread among the Uyghurs, Salars, and Dongxiang

So far, I have set aside the question of how and when the Mujaddidī ṭarīqat reached the territory now controlled by the PRC. The earliest arrivals appear to have taken root in Eastern Turkistan, today's Xinjiang Uyghur Autonomous Region. Here again there was no single wave of Mujaddidī transmission. Numerous, often overlapping, transmissions resulted from more-or-less constant inter-Asian connection, which continuously transformed and diversified Mujaddidī communities in China proper, Eastern Turkistan, and Tibet (see Table 2). Of these three regions, Eastern Turkistan saw the earliest and most numerous arrivals of Mujaddidī proselytisers, and most Mujaddidī lineages in China proper and Tibet can be traced back through the orders that took root in Eastern Turkistan in the eighteenth century.

Table 2. Mujaddidī transmissions from Eastern Turkistan to and within China proper

The initial spread of Mujaddidī orders in Eastern Turkistan can be traced in all cases to arrivals from outside the region, as opposed to travels of Eastern Turkistan's inhabitants to other regions. The proselytisers came first from South Asia (especially Afghanistan) and then Central Asia. The later, Central Asian exchanges began in the early nineteenth century and continued through to at least the 1930s. Thierry Zarcone and Alexandre Papas have outlined these later branches in some detail, based on both written sources and fieldwork among members of their orders.⁸⁵ The earlier lines, with South Asia origins, seem ultimately to have had broader influence, generating substantial offshoots in China proper and supplying leaders in the 1864 revolts.

One of the two candidates for the earliest Mujaddidī arrival is the Ayköl branch, named after a village outside of Aqsu where an ‘Indian’ proselytiser named Ishān Muḥammad Qārī Akhūnd established a khanaqah. It is difficult to date this line securely because the only available sources are oral accounts recorded in secondary Chinese literature.⁸⁶ These accounts claim that Qārī Akhūnd arrived in the middle of the eighteenth century. Their assertions are plausible in light of their specificity and their alignment with the Lanzhou Arabic silsila. For example, they date the birth of one of Qārī Akhūnd's local khalifas, a Huihui man named Ma Fang马方, to 1755. Qārī Akhūnd arrived at the command of his own master to settle and spread the path at a place where the Moon shines from within a lake—a tale reflected in the place name ‘Ayköl’ (literally ‘moon lake’ in Turki). The only corroborating written source I have traced is the Arabic-language silsila preserved by the Hongmen order in Lanzhou, which provides the same names for Qārī Akhūnd's five khalifas as the oral material from Eastern Turkistan.⁸⁷

The other early arrival is the branch known today as the Yarkand Daotang. It is also the best-documented of the Eastern Turkistan lines. Waleed Ziad has brought to light numerous hagiographical texts from South Asia and Afghanistan that provide accounts of Mujaddidī khalifas and followers arriving in Yarkand in the late eighteenth and early nineteenth centuries. Eastern Turki (Chaghatay) accounts of the region's 1864 anti-Qing rebellion also mention Mujaddidī shaykhs who were by that time prominent enough to be raised as leaders of the initial uprisings in Yarkand, Kucha and, as will be argued below, Urumqi.⁸⁸ To these sources we may add three texts preserved by Mujaddidī establishments in Yarkand (Eastern Turkistan), Xunhua (Qinghai), and Beizhuang (Gansu). These have readerships among the Yarkand Daotang's followers but have not been treated in academic publications. The second half of this article uses these rich new sources to explore the spread of the Mujaddidī order from their centres in Eastern Turkistan onward to the Salar and Dongxiang peoples resident along the northern edge of the Tibetan Plateau and thence to the Chinese-speaking Muslims of China proper. There were additional vectors of transmission, for example to Huihui in Eastern Turkistan (where they were widely known as Dungans). The Yarkand Daotang sources examined here provide unusually detailed information on how Mujaddidī thought and institutions were absorbed, localised, and retransmitted. In the context of this article, they are important for showing how Mujaddidiyya traditions reached China proper, crossing multiple cultures and languages, and transforming in the process.

Fieldwork context

I undertook source collection for this section of the article before the government of the PRC accelerated its repression of Muslims, including mass internment of Muslims in Eastern Turkistan from 2017 onward. Muslims in China proper face less dire conditions, but they have also experienced increased state attacks on their religious practices. These include new restrictions that have long been familiar to Xinjiang's Uyghurs, such as the confiscation or destruction of Islamic books, as well as innovative state policies such as the removal of domes from mosques. Between December 2019 and May 2020, every domed mosque in Linxia saw its dome removed.⁸⁹ Satellite imagery shows that the Hongmen order, described above, has also been affected, with domes removed from its pilgrimage centre, Honggangzi.⁹⁰ In Xinjiang, roughly 1 million, perhaps more, Uyghurs and other members of mostly Muslim ethnic groups have been sent to internment camps, and several hundred thousand more to prisons, based on vague assessments of individuals’ levels of religious or ethnic identification.

In this environment, it has been impossible to safely carry out the kind of follow-up research visits that I would normally undertake. There is much more that could be achieved, and there are additional sources (some of which I have seen as physical objects but not had the opportunity to read) that could be accessed through further visits. The situation also compels me to withhold some details, not because they are legally incriminating, but because the law and thus criminality have become meaningless as the state disappears Muslims for the most surprising and arbitrary reasons. I have limited the identifiable geographical information about specific locations to what has already been published elsewhere.

Mujaddidī proselytisers in Eastern Turkistan

The most detailed of the new sources is also the most problematic. It is a Chinese-language history of the Yarkand Sufi order, called the Yarkand Daotang (叶尔羌道堂), published in 2013.⁹¹ While no author is given, the text makes it clear that at least some sections were written by the shaykh in Uyghur/Turki and translated into Chinese by one of his followers. Published under the title The Ninth-Ranked Transmission (Ch: 九品乘传),⁹² the book's various sections draw upon different generic traditions. The bulk of the work is a hagiographical account of Sufi shaykhs from Aḥmad Sirhindī to Shāh Awlīyāʿ, the founder of the Yarkand Daotang, full of the miracles that are often found in such works. The book also includes: a silsila; a catalogue of the historical and sacred objects preserved by the order; an introduction to Sufi thought and the Naqshbandiyya; moral exhortations; and a polemic asserting that the founder of the Jahriyya menhuan was also a follower. Poetry is interspersed between some chapters.

The author claims that the text is based on manuscript sources preserved by the order—a claim supported by several pieces of evidence. First, the book includes a photo of the manuscripts, one of which I was able to examine and date to circa 1800, though the particular manuscript I examined does not contribute much to The Ninth-Ranked Transmission. The hagiographical section also closely matches the generic and formal characteristics of an eighteenth- or nineteenth-century hagiography from the region. The details of the hagiography and silsila align with similar materials on the Maʿṣūmiyya branch of the Mujaddidiyya that have been documented by Waleed Ziad from manuscript sources.⁹³ Finally, the history of the spread of the order to Gansu and Qinghai is confirmed by nineteenth-century manuscripts in both of those locations (described below). Unfortunately for the historian, the material that is reproduced from the manuscript sources is paired and perhaps interspersed with more recent additions, and it is not always possible to distinguish between the two. The book's argument that the Jahriyya founder made a pilgrimage to the Yarkand Daotang is an example of material that looks very much like a recent addition. A more obvious addition from outside the manuscript sources is a collection of encouraging remarks from the comments sections of the Yarkand Daotang's internet posts.

The hagiographical section attributes the foundation of the Yarkand Daotang, and with it the arrival of the Mujaddidiyya in Eastern Turkistan, to a shaykh named Shāh Awlīyāʿ, a direct descendant of Aḥmad Sirhindī. According to the text, Shāh Awlīyāʿ lived from 1733 to 1837, and first arrived in Yarkand as a young man in the second quarter of the eighteenth century. He built the eponymous, physical Yarkand Daotang, which became the headquarters of his ṭarīqat, and spent 40 years travelling Eastern Turkistan, spreading his teachings and designating khalifas to further spread the ṭarīqat.

In other areas of Eastern Turkistan, he visited the shrines of local saints, such as the tomb of al-Ḥasan al-ʿAskarī in Qaraqash, where he left behind a follower to continue teaching his ṭarīqat. He also engaged in an activity commonly attributed to saints in Eastern Turkistan and especially Yarkand—the (re)discovery of the tombs of other saints. The two most prominent shrines in Yarkand, Chiltăn and Muḥammad Sharif, are both associated with tales of miraculous shrine rediscovery.⁹⁴ Shāh Awlīyāʿ's hagiography credits him with discovering Terăk Mazar near Ghulja, which is described as the tomb of a local preacher whom Tughluq Timur Khān had met in the fourteenth century.⁹⁵ Shāh Awlīyāʿ ordered a large tomb structure to be built there and appointed a follower to guard it and hold prayer sessions there.

In his travels, Shāh Awlīyāʿ promoted both silent and vocal dhikr, advocating that his followers practise different forms on different occasions. These included dhikr practices from the Qādirī, Kubravī, and Chishtī Sufi traditions. He also taught Mujaddidī doctrine familiar from manuals such as Khwāja Ṣafīallah Sirhindī's Makhzan al-Anwar, including the mapping of subtle centres to parts of the body.⁹⁶ Upon deputising khalifas, Shāh Awlīyāʿ would present them with a robe, a turban, prayer beads, and an engraved seal. No mention is made of the bestowal of a book in the manner we have seen in the case of the Qi lineage of Linxia or the Eastern Turki hagiographies of Afāq Khwāja. At the age of 102, Shāh Awlīyāʿ passed the leadership of his order to his son, Mīrzā Shāh Muḥammad Sharīf, and returned to his hometown of Badakhshan, where he died two years later.

The narrative is replete with miracle tales, including some shared with another local Yarkand hagiography, the Tazkira of the Seven Muḥammads. ⁹⁷ The miracles range from transforming eggs into chickens to bringing a dead child back to life. Particularly interesting is an episode in which local notables attempt to shame the shaykh by hosting a feast for him and serving cat meat—a forbidden food. Shāh Awlīyāʿ sees through the ruse and begins crying. He caresses the meat and it transforms back into a living cat.⁹⁸ The same event, minus the crying and petting, occurs in the Tazkira of the Seven Muḥammads, when Shāh Ṭālib Sarmast, who has rediscovered the tomb of the Seven Muḥammads, is tested by the king of Yarkand with a meal of cat kebab and revives the cat.⁹⁹ Another tale from the Seven Muḥammads finds a parallel later in The Ninth-Ranked Transmission. A Qādirī saint, whose son became a follower of Shāh Awlīyāʿ, was buried alongside his staff, and the staff grew into a tree.¹⁰⁰ In the Seven Muḥammads, the titular saints carried a staff given to them by the prophet Muḥammad and, upon reaching their fated graves, the staff turned green.¹⁰¹ The Tazkira of the Seven Muḥammads is attached to an earlier Yarkand shrine, already documented under that name by Mirza Muḥammad Ḥaydar in 1546.¹⁰² The shared miracle of the cat meat, in particular, suggests that the author(s) of Shāh Awlīyāʿ's hagiography were inspired by the local traditions and may have even attached them to their founding shaykh's biography in an effort to root his claims to sacred authority in Yarkand's prevailing notions of sainthood.

Further adaptation to local systems of sacred authority appear in The Ninth-Ranked Transmission's version of the Mujaddidī silsila. The silsila departs from standard Mujaddidī silsilas found throughout South and West Asia at several points, but two substantial deviations are clearly connected to the environment of Eastern Turkistan. The first of these is the interpolation of Twelver Shiʿi imams, ten of whom (all but Ali and the twelfth imam) are added to the silsila. At least five of the Twelver imams have shrines in Eastern Turkistan, and several other shrines of ‘Imams’ have hagiographies with genealogies that proceed through the eleventh imam. The popular hagiography of Afāq Khwāja also includes a genealogy passing through the first 11 Shiʿi imams.¹⁰³ In the case of Afāq Khwāja, a Naqshbandī of the (non-Mujaddidī) line from Makhdūm-i Aʿẓam, this Twelver interpolation is similarly a deviation from standard silsilas of his branch known elsewhere in Central Asia. A Twelver genealogy also appears in the centre of Yarkand itself, on the walls of the tomb of Muḥammad Sharīf.¹⁰⁴ In short, a spiritual (and genealogical) lineage proceeding through the eleventh Shiʿi imam was a widely claimed marker of sacred authority in Eastern Turkistan, even where it was at odds with other Sufi texts.¹⁰⁵ The Mujaddidī silsila in The Ninth-Ranked Transmission conforms to this pattern.

The second major deviation in the silsila of The Ninth-Ranked Transmission is the incorporation of foundational figures from the most prominent Sufi lineage of Eastern Turkistan, the Makhdūmzāda branch of the (non-Mujaddidī) Naqshbandiyya. The Makhdūmzādas were divided into two rival factions: the Isḥāqīyya and the Afāqiyya (the line of Afāq Khwāja), who traded political power in the region in the late seventeenth and early eighteenth centuries, and maintained significant influence through the nineteenth century. Specifically, the silsila adds the eponymous Makhdūmzāda shaykh, Makhdūm-i Aʿẓam, along with his son Isḥāq Walī, ancestor of the Isḥāqīyya. These anomalous (for the Mujaddidiyya) figures, along with a certain Muḥammad Qaḍī, all appear between two standard links in Mujaddidī silsilas. Those standard links are given in the forms Muhanmode Zaheide (穆罕默德 扎黑德, i.e. Muḥammad Zāhid) and Muhanmode Daireweishi Waili (穆罕默德 代热维什 外力, i.e. Muḥammad Darwish Walī).

The inclusion of Isḥāq Walī may explain the unusual name of one of the Yarkand order's branches in Linxia. A tomb-shrine in that town called Yarkand Gongbei (yue'erqiang gongbei 月儿墙拱北) bears a sign naming its associated order as the ‘Xinjiang Kashgar Black Mountain Sect Yarkand Daotang’ (Xinjiang Kashi Heishan Zong Ye'erqiang Daotang 新疆喀什黑山宗叶儿羌道堂).¹⁰⁶ In the eighteenth century, hagiographies began to speak of ‘Black Mountain’ and ‘White Mountain’ factions. David Brophy has shown that both originally referred to groups within the Afāqiyya.¹⁰⁷ However, in the nineteenth century, the ‘Black Mountain’ name came to be associated with the Isḥāqīyya and today, among both Uyghur and Hui scholars, the Black Mountain term is considered synonymous with the Isḥāqīyya. Thus, it is possible that the Black Mountain element of the Linxia tomb reflects the Yarkand Daotang's inclusion of Isḥāq Walī, the first Isḥāqīyya shaykh, as a link in the silsila.

Later sections of The Ninth-Ranked Transmission give brief descriptions of the numerous seekers who came from China proper and various parts of Eastern Turkistan to learn from Shāh Awlīyāʿ and his successor, Mīrzā Shāh Muḥammad Sharīf. Short but detailed biographical accounts are given for Ma Baozhen (馬葆真), a Dongxiang man from Gansu; ʿAbdallah Qādir (穆薩阿爷), a Salar man from Xunhua, Qinghai; and several ‘Hui’ men from northern Xinjiang, all of whom returned to their home regions and established offshoots of the Yarkand Daotang. The ‘Hui’ khalifas included Tuo Ming (妥明), who we know from other sources led the Urumqi revolt in 1864 (and whom Joseph Fletcher and Hodong Kim both speculated to be affiliated with the Jahriyya).¹⁰⁸ Tuo Ming is also regarded as a follower of the Yarkand Daotang in the oral material cited by Tan Wutie and Fu Yu.¹⁰⁹ Additional figures are listed with only a sentence or two of information, including several of the ‘Hui’ pilgrims noted above who are reported to have brought the Yarkand Daotang order back with them to Sichuan, Shaanxi, and Gansu.

A Mujaddidī transmission to the Salars of Qinghai

The tombs of the Dongxiang shaykh Ma Baozhen and the Salar shaykh ʿAbdallah Qādir are still active sites of veneration in Gansu and Qinghai, respectively, and the communities associated with them preserve nineteenth-century texts that confirm many of the claims of The Ninth-Ranked Transmission. At the tomb of ʿAbdallah Qādir, near Xunhua, Qinghai, the caretakers preserve a small manuscript written in Eastern Turki, the vernacular of Eastern Turkistan in the eighteenth to nineteenth centuries. Although the caretakers’ mother tongue, Salar, is also a Turkic language, it is too distant from Eastern Turki for them to understand the text. They presented it to me as an ijāzat and, while the text does not seem to give any explicit indication that it served such a purpose, the tradition of presenting books along with ijāzat among the Hui and the followers of Afāq Khwāja suggests that the manuscript may have played such a role. The caretakers told me that the manuscript was given to ʿAbdallah Qādir when he visited the shaykh at the Yarkand Daotang in the early 1800s. Because I am a non-Muslim, the caretakers did not allow me to touch or photograph the book. They were, however, willing to hold the book in front of me and turn the pages while I took notes. They grew impatient, however, and in total I was able to examine the book for only 10 or 15 minutes.

The manuscript consisted of two sections, the first calling itself Rāhnāma-ye Muwāfiq-i Imām Ja'far Ṣādiq and the second a silsila, or chain of transmission.¹¹⁰ The few precious moments I had were not sufficient to read the first text, so I focused on transcribing the silsila, which was short enough for me to complete before the caretakers returned the book to its storage place. The silsila is largely consistent with Mujaddidī silsilas throughout the world for the four generations leading to Aḥmad Sirhindī and his son Muḥammad Maʿṣūm. It then proceeds through two more shaykhs and ends at Shāh Ghiyath al-Dīn, likely the shaykh described in The Ninth-Ranked Transmission as the father of Shāh Awlīyāʿ.

Like the hagiography in The Ninth-Ranked Transmission, the Salar Rāhnāma/Silsila shows signs of adaptation to the environment of Eastern Turkistan. In addition to being written in the vernacular of nineteenth-century Eastern Turkistan, it deviates from South Asian silsilas of the Mujaddidiyya by adding figures well known in East Turkestan, and it does so in roughly the same ways as The Ninth-Ranked Transmission does: it includes most of the Twelver Shi'a imams and it includes the locally significant link of Makhdūm-i Aʿẓam. There are some small differences in these sections. The Salar silsila includes 11 rather than 10 of the Twelver imams; it omits Isḥāq Walī; and it inserts the Makhdūmzādā section one step later in the standard Mujaddidī silsila. Nonetheless, the alignment is strikingly close and provides a powerful corroboration of The Ninth-Ranked Transmission, showing that the localisations in that work's silsila are not recent accretions but were instead present in the Yarkand Daotang tradition by the mid-nineteenth century.

All told, the evidence from the tomb of ʿAbdallah Qādir in Qinghai provides convincing corroboration of The Ninth-Ranked Transmission's claims to being based on primary historical sources. And it suggests that the account of the transmission of the Mujaddidiyya to Eastern Turkistan by a branch traceable through Shāh Ghiyath al-Dīn, as well as the further transmission of that ṭarīqat from Yarkand to the Salars of Xunhua, is roughly accurate. As recently as the early 2000s, the caretakers of Abdallah Qādir's tomb still expressed allegiance to the Yarkand Daotang and said that their community's shaykh makes regular pilgrimages to the ṭarīqat's centre in Yarkand, though such pilgrimages are likely not possible under today's conditions in Xinjiang.

A Mujaddidī transmission to the Dongxiang of Gansu

Several kilometres outside of the town of Dongxiang, Gansu, in the village of Beizhuang, the tomb of Ma Baozhen (馬葆真) anchors a community of Yarkand Daotang followers of the Sarta, or Dongxiang, ethnicity. The Dongxiang speak a language that is primarily Mongolic, with substantial Turkic, Persian, and Chinese elements. The religious community surrounding the tomb of Ma Baozhen has recently splintered into two factions, abandoning the imposing mosque adjacent to the tomb. The mosque stood empty, with broken windows, when I visited in 2015. The two factions have each established their own new mosque, both of which are far humbler than the one they abandoned. The larger of the two preserves a 360-page manuscript in Persian describing the lives and teachings of saints in the Mujaddidī lineage of the Yarkand Daotang, as well as the transmission of that order to Ma Baozhen via his pilgrimage to Yarkand.¹¹¹ When I visited, the head of the order was travelling. A staff member and a student told me that the original manuscript was kept at the mosque and I was able to photograph a photocopied edition of the manuscript.¹¹²

The text is Nuzhat al-Qulūb, by Muḥammad Yūsuf, written in 1273 AH (1856–57 CE). The Persian includes grammatical errors in places and numbers are sometimes written in the word order of Arabic, with units followed by tens followed by hundreds, and so on, rather than in the opposite order typical of Persian. The third of three sections narrates the transmission of the Yarkand Daotang tradition to the Dongxiang region of Gansu via Ma Baozhen, and aligns neatly with the short description of Ma Baozhen's discipleship in The Ninth-Ranked Transmission, while also adding much further historical detail and hagiographic elaboration. The dates provided for Ma Baozhen's life, 1772–1826 CE (given in both Islamic and Chinese reckonings), also confirm the general chronological claims of The Ninth-Ranked Transmission, which gives Ma Baozhen's master, Shāh Awlīyāʿ, a lifespan of 1733–1837, easily overlapping with that of Ma Baozhen, even if it might be exaggerated in length.

The lineage it presents is, for the first four generations after Aḥmad Sirhindī, that of the Maʿṣūmiyya, a Mujaddidī sub-branch prominent in Afghanistan.¹¹³ Its record of the Maʿṣūmiyya lineage is standard from Sirhindī down to the eponymous Ghulām Maʿṣūm. At that point, it deviates from genealogies known from other Maʿṣūmiyya texts and instead continues the lineage via the pivotal transmitting figures of The Ninth-Ranked Transmission: Mīr Ghiyath al-Dīn and Shāh Awlīyāʿ. (The latter is the figure named in The Ninth-Ranked Transmission as the Yarkand Daotang's founder, and the former is named in that source as his father. Ghiyath al-Dīn is also the final figure listed in the Salar silsila.) As with the Salar silsila and The Ninth-Ranked Transmission, the first 11 imams of the Twelver Shi'a tradition are given a place in the silsila.

Nuzhat contains a tantalising mention of another work, which I have not been able to locate in archival catalogues or private collections, but which seems to be connected to the Beizhuang branch of the Yarkand Daotang order. Where the Nuzhat discusses the date of Ma Baozhen's death, it says that a book called Tazkirat al-Bīẓay, written by a certain Karāmat Khalifa Muḥammad Hisabullah al-Baschi, provides a different date. Given that the author of this missing text offers a death date for Ma Baozhen, who does not seem to appear in textual records outside the Yarkand Daotang tradition, and that the author bears the title ‘Khalifa’, it is likely that the author was a khalifa in the Beizhuang branch of the Yarkand Daotang order. This, combined with the word tazkira in the title, suggests that this text represents another hagiography of the leading lights of the order. Together, Nuzhat al-Qulūb, Tazkirat al-Bīẓay, and The Ninth-Ranked Transmission demonstrate that a robust hagiographical tradition accompanied the attention to the silsila, reflecting interest among followers of the Yarkand Mujaddidī line in their identity as a distinct community rooted in the renewal of Aḥmad Sirhindī. The texts to which we have access were written in Persian and Chaghatay (in its Eastern Turki variant), and show evidence of adaptation to the hagiographical norms of Eastern Turkistan that was then transferred onward to China proper. In Gansu and Qinghai, the offshoot of the Yarkand Daotang that was transmitted through the Dongxiang became known as the Beizhuang 北庄 order and it produced some prominent figures in the history of Chinese Islams. Most notable among them are two scholars who abandoned the order to found two of China's main reform movements: Ma Wanfang 马万福 (1849–1934), whom we have already encountered as founder of the Ikhwānī movement in China, and Ma Qixi 马启西 (1857–1914), founder of the Xidaotang 西道堂.

Conclusion

It is tempting to call the Mujaddidī transmission described in this article a new ‘tide’ of Chinese Islam, following the prevailing scholarship on the history Islam in China. The ‘three tides of Chinese Islam’ framing posits a series of discrete transmissions of various new-to-China forms of Islam over the last 1,400 years: the first transmission of Islam, supposedly represented by the Gedimu; the arrival of Sufi orders in the seventeenth century; and the modernist reform movements of the late nineteenth and early twentieth centuries.¹¹⁴ This model is convenient as a heuristic framing, but the Mujaddidī example throws into relief what is lost in simplification. There were multiple, overlapping transmissions of Mujaddidī texts, preaching, and organisations, into and within China. Given the regularity of pilgrimage to various sites of Mujaddidī significance, it might even be best to think of Mujaddidī transmission as a continuous process—one that is ongoing today in the pilgrimage of Hui followers to the tomb of the Mujaddidī founder in Sirhind, India. Moreover, the continuous transmission of the Mujaddidī movement overlapped and interacted with the Sufi and reformist ‘tides’. In any case, the Mujaddidiyya have claims to both of these categories.

For the study of the Chinese Islams, the Mujaddidī case expands the catalogue of trans-regional connections, revealing yet another dimension of religious and intellectual exchange. India and Afghanistan are often omitted from our story of Chinese Islams, but the Mujaddidī transmissions show continuous interaction, from the mid-1700s down to today, when pilgrims from China's ‘Little Mecca’ visit Aḥmad Sirhindī's tomb regularly. It reminds us of the importance of Persian, Arabic, and Turkic languages in the history of Chinese Islams, as media of connection to neighbouring Muslim societies, and it further undermines the myth of Chinese Muslim isolation.

Where direct Mujaddidī transmissions to China proper can be traced, they invariably passed through Eastern Turkistan (although the Qi lineage also travelled to the Hijaz via Indonesia). The importance of interactions between the Turki, or Musulmān, people of Eastern Turkistan, whose descendants are known today as Uyghurs, and the Huihui, or Hui, was already well established in the work of Trippner and then Fletcher, who demonstrated the role of Afāq Khwāja and his father in spreading Naqshbandī Sufism among the Huihui during the seventeenth century.¹¹⁵ Afāq continues to be venerated as the origin point not only of the Qi lineage, as described above, but also of the Khāfiyya and Xianmen orders. More broadly, the canon of Islamic texts consumed in nineteenth-century Eastern Turkistan and much of Islamic China proper show substantial overlap, particularly among legal and Sufi philosophical texts. As recently as a decade ago, Hui pilgrims regularly made the journey from China proper to the Uyghur shrines of Tuyuq Khojam (near Turpan) and Imam Jă’firi Sadiq (near Niya), not to mention the Yarkand Daotang and its associated shrine.¹¹⁶ The Mujaddadī transmissions effected by the Yarkand Daotang and Ayköl orders greatly expand the documented historical cases of circulations between Turki/Uyghur Muslims and the Huihui/Hui, widening our view of the long-term and varied exchange of Islamic thought that took place across this ethno-linguistic boundary and profoundly influenced the Islams of China proper.

The case of the Mujaddidiyya in China also tells us something about the spiritual revolution set off by Aḥmad Sirhindī. It supports the arguments of Waleed Ziad, who noted that a combination of organisational, ideological, and textual characteristics made the Mujaddidiyya particularly suited to wide dissemination. Ziad documents accommodation of Mujaddidī outposts by various political orders, and the localisation of Mujaddidī techniques in varied cultural environments. The Chinese case shows that Sirhindī's interventions, continuously and subtly reshaped by succeeding generations, had wide effects not just in spreading the formal order but in spreading ideas and texts among numerous Chinese Muslim sects. At the same time, the division and subdivision that the Ayköl lineage underwent upon reaching China proper, and the attendant loss of both the ‘Naqshbandī’ and ‘Mujaddidī’ designations, suggest an extreme form of localisation not seen elsewhere in the Mujaddidī world.

The evidence collected here demonstrates a range of Mujaddidī affiliations among Chinese Muslims that is too extensive to treat comprehensively in a single article. Nearly every major Chinese Islamic tradition is touched in some way by Mujaddidī influence, from the Gedimu to the Jahriyya to the Ikhwānī modernists (I have not yet seen evidence of Mujaddidī influence among the Salafīs of China). For scholars with access to Chinese Muslim communities, the scope for future research is vast. More menhuan will likely be discovered to be part of Mujaddidī lineages, especially among those that call themselves Hufuye. There is much more to be learned about the role of Sirhindī's Maktūbāt, along with other Mujaddidī texts, in various communities within China. In particular, the timing, routes, and mechanisms of Maktūbāt's spread deserve detailed investigation. I hope that, by identifying Mujaddidiyya ideas and identities as major components of Islams in the Chinese culture area, this article will encourage and facilitate such research. And, by focusing on the great variation in the ways that Mujaddidī connections have affected Chinese Muslims, I hope this case can serve to remind us of the contingent, continuously emergent, and heterogeneous nature of even highly formalised solidarities, such as the Mujaddidī order.