Introduction
One summer day in 1972, three young Red Guards—Han Chinese Sun Daliang, Tibetan Trashi, and Hui Ma Jiajia—notice logs drifting away on the river, despite the fact that timber is no longer transported via the A’xia River (Figure 1). Without hesitation, they leap into the rushing waters to retrieve the logs. Concerned about further losses of state property, they proactively request permission from the Party committee to guard the river. Inspired by revolutionary tales, the boys establish camp in a small wooden cabin by the river and begin patrolling day and night. One evening, the playful Jiajia, while on sentry duty, gets distracted by a strange crowing chicken and leaves his post. Seizing the opportunity, unknown culprits once again steal logs from the river.
Hui Red Guard Ma Jiajia (left) together with his teenage comrades Han Sun Daliang (centre) and Tibetan Trashi (right).
Taking advantage of their lack of experience in class struggle, Ma Hade, the river transport team leader and Jiajia’s uncle, sows discord among the three boys, further escalating tensions. Meanwhile, in response to the military’s urgent need for timber, the Party committee decides to open a new waterway on the river. Faced with this development, the greedy and cunning Ma Hade conspires with a group of thieves to steal military rafts. Daliang and his friends infiltrate the ‘tiger’s den’, uncovering crucial evidence. When the thieves again make their move, the boys bravely intervene, working with the militia and Party leadership to dismantle the gang and expose Ma Hade’s deception (Figure 2).
Ma Jiajia’s grandfather, a skilled Hui wood rafter, leads the young Red Guards to recover the stolen logs from the Xia River.
Released at the tail end of the Cultural Revolution, this film—The Secrets of A’xia River—belongs to a didactic corpus of class-struggle films extolling the Red Guards, who were the enforcers of Mao’s campaign to ‘cut off the tails of capitalism’ (ge zibenzhuyi de weiba 割资本主义的尾巴).Footnote 1 In practice, the Cultural Revolution years saw an intensification in ‘collectivising’ the privately managed forests and timber farms belonging to individual commune members in southern Gansu.Footnote 2 Although little information is available on the audience and impact of The Secrets of A’xia River’s release, it was almost certainly a ‘campaign film’ aimed at trumpeting up the cause to eradicate private entrepreneurship in the timber trade.Footnote 3
The film stands out for its rarity in two key regards. First, while it reduces ethnic differences to a matter of mere garb, it is remarkable that it still finds it necessary to rely on this superficial distinction to validate the universal appeal and authority of Maoist class struggle beyond China’s Han metropole. Second, the film’s depiction of Hui profiteers operating on the margins of a Maoist-controlled economy inadvertently—or perhaps purposefully—suggests a chrono-political difference that marks the Hui as lagging behind the Han in their revolutionary development. These two features converge to construct an image of the Hui as doubly peripheral in the political economy of High Maoism: rather than being fully assimilated socialist citizens, the Hui of this north-western frontier region remain peripheral both in terms of their appearance and their modes of economic production.Footnote 4
Beyond what it reveals about the representational politics of ethnicity in High Maoism, which has been explored elsewhere, Footnote 5 the film’s mise-en-scène also invites the question of how timber farming related to the local natural environment—forests and rivers—as well as the social environment—Muslim entrepreneurs and labour, who at this moment of history were politically socialised into Maoist collectivism. This article focuses on the role of the Hui as extractivists and profiteers in the ethnic corridor of southern Gansu and addresses the question of when they harnessed the forests and rivers in the upper Yellow River region and if/how that niche economy contributed to the consolidation of a regional commercial and labour network that in turn underwrote a Hui ethnic identity in its early stages.
Using both audiovisual and historiographical sources, this article reveals that despite vastly different political framings of the Hui’s role in the timber industry, logging and transportation on the waterways persisted as chief economic activities for the upper–Yellow River Hui across multiple eras from the late nineteenth century through the early decades of the People’s Republic of China (PRC).Footnote 6 This remarkable longevity opens new avenues of inquiry beyond the prevailing discussions of cultural performativity in the study of ethnicity in China in general and Muslim religious affiliation in the study of the Hui in particular.Footnote 7 Similar to what Bin Chen has demonstrated in his study of the Hui Muslim merchant network in the mid–lower Yangzi River region,Footnote 8 this article disaggregates ethnic identification as a tripartite construct shaped by environmental conditions, economic territorialisation, and the ecumenical capacity to foster social solidarity.
Despite the coincidence between Chen’s case study and mine in focusing on riverine trade and industry, this article examines economic territorialisation of the Hui in the north-west alone. To further disambiguate terminologies here, it should be pointed out from the outset that a Hui in this article does not correspond to a Sinophone Hui recognised by the contemporary minzu shibie 民族识别 (ethnic classification regime) put in place since the 1950s.Footnote 9 Instead, it refers to a pre-PRC, broader identity based on shared participation in the same religion, the same regional economic networks, and even the same political project in the case of soldiering for the Muslim-dominated militarist governments in the region. A Hui figure in this case is embodied by a frequently polyglot, and biscript if they were literate, Muslim person who hails from any of the Hezhou, Dongxiang, Bao’an, or Salar districts of the upper Yellow River region or their relatives and associates further downstream as far as the Ordos Bend (Figure 3).Footnote 10
A topographical image of contemporary Jishishan Autonomous County of Linxia. The county’s northern border is demarcated by the Yellow River flowing eastward and emptying into the Liujiaxia Reservoir. To the county’s west is the upland Xunhua Salar Autonomous County of Qinghai, to its south-east is Linxia, the Hui Muslim stronghold often nicknamed the ‘Little Mecca of China’. Jishishan epitomises the cultural landscape of southern Gansu as a multilingual ethnic corridor because the autonomous county operates on the logic of religious autonomy for three different ethnic groups that practice Islam as their common religion—the Bao’an, Dongxiang, and Salars. Yet such divisions were largely irrelevant to the region’s cultural and political economy before the inauguration of the PRC’s minzu system.
The environmental factor as a constitutive part of ethnicity has been extensively examined in studies of Mongol pastoralism, Kazakh nomadism, and the reconstruction of an uncontaminated Tungusic identity for the Oroqen people by imperial Japan’s ethnologists.Footnote 11 Economic territorialisation, meanwhile, has been analysed as a form of state intervention—often through bank loans and repressive aid programmes—to dismantle regional and individual economic autonomy and intra-ethnic solidarity.Footnote 12 More recently, David Stroup’s study of the hand-pulled noodle economy of the Hualong Hui Muslims in contemporary China further demonstrates that, when sufficiently sanctioned by the state, ethnic minorities are capable of territorialising a niche economy and transforming it into an ethnic prerogative.Footnote 13
It is therefore important to underline that ethnicity, far from being a solely primordial and inherited identity, is subject to existential anxieties, including the potential for extinction or voluntary disaffiliation by its members. The pressure points leading to such anxieties are manifold. State-led efforts at liquidation represent one external threat, while internal anxieties such as concerns over large-scale interethnic miscegenation, particularly when supported by the state, also weigh heavily on ethnic communities.Footnote 14 Within this framework, the persistence of logging and log rafting among the Hui in southern Gansu provides a compelling example of how economic territorialisation reinforces ethnic identity. By inscribing a cultural imprint on the southern Gansu regional environment, wood trade and transportation sustained the Hui identity both environmentally and economically—two crucial dimensions largely overlooked in previous studies of Hui identity in this region.Footnote 15 The significance of this economic territorialisation cannot be overstated, as its disintegration has historically contributed to both the de-Mongolisation and the depopulation of Mongol communities in much of present-day Inner Mongolia during the late Qing and early Republican periods.Footnote 16 The controversy in recent years over halal labelling across Muslim communities in China, and the north-west in particular, is evidence that halal butchering and food production remains a bottom line of Muslim solidarity, with opposing sides of the issue mobilising both religious and economic arguments.Footnote 17
In addition to highlighting an overlooked dimension of ethnic identity formation, this study also sheds light on a revealing textual bias in the historical record regarding the economic territoriality of Hui ethnicity in southern Gansu. On the one hand, local official histories of Linxia systematically ambiguate any acknowledgement of ethnic economic territorialisation. On the other hand, this very subject constituted a top research concern for numerous inquisitive agents of the state, including curious travellers, imperialist expeditionists, and nationalist ethnographers. While this textual divide does not, in itself, serve as definitive proof of economic territorialisation being tied to a single ethnonational community, it nonetheless suggests that ethnicity acquires significant geopolitical value precisely because it overlaps with a tangible reality of ethnically demarcated economic sectors.
The body of audiovisual materials analysed in this article aligns with non-official texts in validating the existence of ethnic economic territoriality. One possible reason for this alignment is that outside filmic gaze, as an external perspective, remains less constrained by the overt political agenda of official histories. The latter, by contrast, are uniquely tasked with suppressing deep-seated interethnic cleavages and obfuscating the material reality of ethnically demarcated economic sectors, which could otherwise become sites of political contention due to competition or state-imposed regulatory biases. In what follows, this article weaves together audio-visual materials and ethnographic sources to interrogate this external gaze while contrasting it with the state’s concerted efforts to write ethnic economic territoriality into oblivion.
The Muslim dominance of Linxia’s wood trade since the 1870s
Since the late nineteenth century, Linxia possessed a resource highly coveted by both Muslims and non-Muslims downstream along the Yellow River—wood. As Hannah Theaker vividly documents,
The hinterlands of southern Gansu were among the few places in north China where substantial stands of old-growth trees remained. Felled trees were transported down to the Yellow, Xia or Tao rivers, to be rafted downriver. Trees that entered the water above Xunhua could be floated as far as Ningxia. Access to forests, however, was a privilege of those able to enter the upland slopes and to negotiate access with the rights holders. Woodlands could be designated communal resources of local Tibetan and Mongol communities, privately owned by a headman, or the property of monastic communities.Footnote 18
Linxia’s Hui merchants, many of whom were polyglot and connected through Sufi networks with upland Salars, were well positioned as middlemen in the wood trade. Their ability to secure logs from the plateau and sell them further downstream cemented their economic prominence. The trade’s centrality to Linxia’s economy was reflected in its physical landscape. Linxia is home to the National Taizishan Nature Reserve (Taizishan guojia ziran baohu qu 太子山国家自然保护区), formerly the largest wood farm in Gansu. A 1985 state survey counted 664 timber farms across the prefecture, with a permanent workforce of 1,412 lumberjacks.Footnote 19 Muslims’ historical dominance of the trade is reflected in historical records as well as material inheritance. Linxia City, seat of the prefectural government, for instance, hosts the well-storied Muslim quarter, Ba fang shisan xiang 八坊十三巷 (Eight mosque-affiliated residential clusters and thirteen alleyways). Now, a tourist favourite because of its fame as the very centre of the ‘Little Mecca of China’, Ba fang shisan xiang was formerly known as Timber Mill Village (Muchang cun 木场村), undoubtedly indicating the contribution of wood-generated wealth to the social reproduction of this Muslim community and their central location in the city.Footnote 20
Timber Mill Village and similar names are found across the region, even in upland Xunhua. an upland region dominated by Salar Muslims and bordering Tibetan and Mongol territories. A Timber Farm Village (also Muchang cun 木场村) appeared in Xunhua as early as the Guangxu reign (1875–1908).Footnote 21 In Linxia City itself, outside of Ba fang shisan xiang, the wood trade seems to have funded the construction of congregational spaces across sectarian divisions. This influence is most visible in the naming of two mosques directly tied to the trade: the Old Wood Farm Mosque (Muchang laosi 木场老寺) and the New Wood Farm Mosque (Muchang xinsi 木场新寺). The existence of these mosques not only indicates the expansion of the wood trade over time but also suggests sectarian competition, with the old mosque affiliated with the Gedimu tradition and the new mosque associated with the Ikhwan movement.Footnote 22
Several historical factors contributed to the rise of Muslim prosperity in Linxia through the wood trade. First, the demand for timber surged in the wake of the postbellum reconstruction following the Tongzhi Rebellions (1862–1877), which had devasted cities and towns across the north-west. Additionally, the Hezhou Ma Muslim militarists’ decision to align with the Qing forces after their decisive victory over Zuo Zongtang in 1872 further solidified their standing. In recognition of their pro-dynasty loyalty, the Qing court allowed them to settle in southern Gansu around Linxia while maintaining command over a sizeable military force. This political favour enabled Hui Muslims to entrench their economic interests, particularly in monopolising regional trade. As Theaker notes, of the 17 state-licensed wood trading companies active around 1890, five belonged to the extended family of the military strongman Ma Zhan’ao alone. These companies held a near-monopoly on felling rights in Xunhua.Footnote 23
We see a pattern of expansion over the first half of the twentieth century. While there was only one modern timber processing factory across the entire Gansu Province in 1926—located in midstream Lanzhou—by 1931, over 250 modern timber mills sprang up both upstream and midstream in towns like Lanzhou, Lintao 临洮, and Minxian 岷县.Footnote 24 One source documents that around 1931, Lintan County alone fed over 70,000 lumbers into Lanzhou’s timber mills each year.Footnote 25 Muslim official-gentry (guanshen 官绅) also purchased large patches of hillside land to raise plantation forests. In Linxia, Ma Buqing, brother of Qinghai government chairman Ma Bufang, owned two plantation forests. Other Muslim notables, including Ma Yunsheng, Ma Qishan, Ma Deshan, and Ma Yuan, all engaged in encircling hillside land to create their own plantation forests.Footnote 26
Two historical documentaries show that the importance of the wood trade and wood transportation grew along with the rising fortunes of Linxia’s Muslim military strongmen, who maintained their dominance over regional politics until the mid-century.Footnote 27 The Carter Holton’s missionary films (1920s–1940s) and Kukan: The Secret of Unconquerable China (dir. Rey Scott, 1941) offer rare visual documentation of the economic territorialisation of Hui Muslims in north-west China, particularly their dominance in the wood trade and wood rafting. The Holton films are a collage of ethnographic footage that the missionary Carter Holton and his wife Lora Holton took between Gansu and Qinghai spanning the 1920s to the 1940s. Kukan, filmed by Rey Scott and premiered in the United States in 1941, provides a more transnational perspective, chronicling Scott’s journey across China, including extended stays in Gansu and Qinghai, where he documents the economic life of Hui traders navigating the wartime landscape. Together, these films not only capture the Hui’s economic activities in the north-west but also reflect the geopolitical and ideological forces that framed their commercial presence within narratives of Western missionary interest and China’s national defense.
The Holtons were affiliated with the Christian and Missionary Alliance, which was contemporaneous with the British-led China Inland Mission and the Jesuits’ diocese in north-western China. While all three took special interest in Muslims of the north-west, they each had their own unique features. The China Inland Mission was probably the oldest among the three. Its textual record includes Mrs Howard Taylor’s (1865–1949; Ch: 金樂婷) extensive coverage of the 7.9 magnitude 1920 Haiyuan earthquake that killed over a quarter of a million people, the majority of whom were Hui Muslims in eastern Gansu.Footnote 28 The Jesuits’ diocese in north-western China was most famous for their prefect in Xining, the German Divine Word Father Hieronymus Haberstroh (1893–1969), who was a close confidant of Ma Bufang and supplied the latter with gasoline, Zeiss binoculars, radio receptors, handguns, and other military supplies from German sources, in addition to operating Qinghai’s first winery.Footnote 29
The Christian and Missionary Alliance by contrast was merely a loosely connected umbrella organisation, taking under its wings the famous Oberlin Band of which the 1900 Boxers Rebellion victim Rowena Bird was a member. Among all the missionaries active in the Muslim north-west, Carter Holton stood out not because of the institutional support behind him, but because of his expertise in Islamic theology and Muslim affairs—he studied Arabic and Moslem religion and culture at Princeton Theological Seminary under Dr Samuel Zwemer on furlough from 1934 to 1935 before resuming his work in Gansu and Qinghai.Footnote 30 The Holtons were likely also the best equipped as they carried with them a motion picture camera that documented some of their travels over a span of 30 years.
The Holton films were originally silent images without sound recording. The voice-over by Robert Carlson and Robert Shuster was added in 1995 and therefore offers little help to put the images in accurate context. The silent footage, however, fits the definition of ethnographic photography as most of the scenes recorded, including mule caravansaries traversing dangerous roads and water hands rafting logs in white rapids, seem to be highly contingent actions and bear no mark of intervention from the filmmaker (Figure 4). The only consistent subject matter is mostly male adults donning white hats, an unmistakable sign of their Muslim identity, and busy with economic activity—loading and offloading goods onto horsebacks, mulebacks, ferries, and carts, and rafting logs down the rapids.Footnote 31
A Muslim man rafting down his timber on the Yellow River near Linxia.
Aside from confirming the stereotype that Muslims in Linxia were adept at long-distance caravan trade, the Holton films also attest to the social reproduction of their labour on the key waterways of the upper Yellow River and its tributaries Xia and Tao, a phenomenon that has often been observed only in the small section passing through Lanzhou, the capital of Gansu Province. This social reproduction of labour complements Theaker’s analysis of the Muslim dominance of state-licensed logging companies as labour and capital reproduction worked in a symbiosis to strengthen Muslim economic productivity in the context of the often-strained interethnic relations in the ethnic corridor of southern Gansu and eastern Qinghai. Put in more concrete terms, the Muslim notables’ dominance over regional politics and business monopolies created an intra-ethnic social mobility complex that attracted the lower strata of the Muslim society to profitably attach themselves to its various sectors such as soldiering, civil bureaucracy, long-distance transportation overland and on water, and timbering and wood rafting.Footnote 32
These economic practices also benefited from the region’s geography and environment as they recalibrate the resources at the Muslims’ disposal: extensive transregional trade and religious networks, polyglotism, traditional strength in animal husbandry, particularly of pack animals, navigation skills in rough waterways, and by this time in history, plantation forestry. At the turn of the century, some of these resources were reinforced by the rise of militarism and transnational circulation of Islamic reform,Footnote 33 even though the impact of such was localised, which in turn increased the appeal of a distinct Muslim modernism in education and civil governance,Footnote 34 and of endogenous desires for political autonomy.Footnote 35 Other skills more pertinent to grassroots labour also found new use in transporting materials for road construction, factory building, and war as the north-west underwent much infrastructure modernisation even as it was increasingly imbricated into the Second World War’s China theatre,Footnote 36 an episode of history well documented in Scott’s 1941 award winning documentary Kukan: The Secret of Unconquerable China.
Kukan is a documentary film first premièred in the Lunt-Fontanne Theatre, formerly known as the Globe Theatre, in New York City on 23 June 1941.Footnote 37 It toured many American cities and won the best documentary award of the 1942 Oscars. Scholars also note that the film was presented to F. D. Roosevelt in the White House on 1 January 1941, allegedly influencing his decision to grant Chiang Kai-shek substantial loans and aid as part of the US–China alliance against Japan, many months before the USA formally entered the Pacific War.Footnote 38
Scott formally began his China journey after he made a brief stop in British Hong Kong, where he saw from across the border that Guangdong was already occupied by Japanese forces. He then decided to take a plane from Hong Kong to Haiphong in Vietnam. From Haiphong onwards, he would continue his trip to Guangxi’s capital city, Nanning, while documenting the trucks trudging along the Burma Road. From Nanning, he went north to reach Guiyang, before arriving in Chiang Kai-shek’s wartime capital, Chongqing. After staying a few days in Chongqing, he went westwards to Chengdu and from Chengdu he drove north to Lanzhou. He then went into Qinghai and spent the longest time of his trip in Xining, Kumbum, and the Qinghai–Tibet border town of Dongga. After that, he chartered a flight to take him from Xining back to Chongqing.
It was on the Lanzhou section of the Yellow River that Scott first introduced the role of the north-western Muslims, whom the Christian filmmaker called the ‘Mohammedan hordes’. His camera reveals to us the importance of those Muslims as rafters shipping wool and animal hides in exchange for Russian military supplies and oil. Scotts noted that, ‘The rafts are made of inflated goat skins and can carry from 15 to 50 tons.’ He also remarked that ‘Lanzhou is the China end of the Russian supply route that comes down from Xinjiang and the Soviet in the north.’ Scott’s observation is corroborated by more recent research on the importance of this northern route for Allied aid. As Mark Baker’s study shows, the newly built Northwest Highways delivered large volumes of war materiel from Soviet Kazakhstan to Lanzhou. Between December 1937 and June 1941, Soviet aid via this route averaged 1,000 tons per month, with fuel and the weight of donated trucks excluded.Footnote 39 Liu Zhi, meanwhile, documents that in 1939 alone, 4,400 machine guns, 100 million rounds of ammunition, 250 artillery pieces, 140,000 shells, as well as 1,000 transportation trucks were delivered to the Chinese Nationalists in Xingxingxia on the Xinjiang–Gansu border.Footnote 40 On his site visit to Lanzhou’s Yellow River waterway, Scott, a former journalist for the London-based Daily Telegraph, concluded that ‘The Burma Road in the south and the “red route” in the north are literally the only roads by which solely needed supplies can come in from the outside world, slow, backbreaking and expensive.’
On the role of the water hands, Scott expresses his fascination that the Muslims were not only transporting critical war supplies but also dry hay, watermelons, and other farm produce, a testament that the skill of water rafting was instrumental in local commerce and trade (Figure 5). But in the context of the deepening military struggle between the Chongqing regime and Japan, the Kuomintang (KMT) understandably wanted more international aid, and in larger volume, from the Soviet Union. Such increases, however, were entirely beholden to the improvement of transportation, or as Scott related to his audience, ‘The real development of this part of China is dependent upon the solving of the problem of transportation, slowly and painfully, this vast land is being prepared for the China of the future and plays its grim part in preserving the China of today.’
Two Muslim youth rafting their watermelons on the Yellow River near Lanzhou.
What that meant for the Muslims in the region was an unprecedented drive for road construction, both highways suitable for motorised automobiles and railways—the true king of freight transportation in that era. Predictably, both these endeavours required the exertion of huge amounts of local manpower and resources. Scott explains intuitively, ‘I was constantly reminded in this frontier country that 100 years ago American pioneers were struggling through just such a terrain, battling and subduing the same stubborn nature and laying the foundation of a great modern industrial society.’ No other groups fit the image of the American pioneers better than Ma Bufang’s Huihui army engaged in road building. As Scott observes, ‘Hundreds of workers were toiling along the road. And again and again it was plain why neither nature nor Japanese bombs can close China’s lifelines.’ A contemporary of Scott who also travelled to Qinghai around the same time, another American journalist, Arthur Doak Barnett explained in much detail that Ma Bufang’s forces were not just building the highways but also making them prettier and more permanent.
The smooth, crushed-rock highways in Chinghai are all lined with green willow and poplar saplings. An intensive programme of tree planting and reforestation has, in fact, made the valleys of east Chinghai the most verdant in China … Chinghai does not have an Arbor Day; it has an Arbor Fortnight. For two weeks each spring, everyone in the province goes into the countryside and plant trees. During the past ten years, 61 million have been planted, an average of more than six trees per year per man, woman, and child in the province.Footnote 41
The construction of highways and reforestation, however, could not satisfy fast enough the insatiable demands of Chongqing’s need for ever-increasing amounts of foreign aid coming from Soviet Turkestan. A more ambitious attempt at railway construction was also underway. On 11 November 1941, Ling Hongxun 凌鴻勛, bureau chief of the Tianshui–Chengdu Railway 天成鐵路, apparently in response to a request for information from Chiang Kai-shek himself, stated that his bureau had completed surveying the route for the 755-km railway, which would cost a staggering annual investment of 85 million yuan for three consecutive years. Ling further added that this railway would be extended westwards with two more lines, a 400-km line from Tianshui to Lanzhou 天蘭鐵路, and a further 860-km line from Lanzhou to Suzhou (today’s Jiuquan).Footnote 42 As though to convince Chiang of the urgency in bankrolling the costly project, Ling offered that the oil field in Yumen saw dramatic increases in output, justifying the need to spend 40 million yuan to jumpstart the construction of that section as soon as the surveying completed.Footnote 43
However, this KMT-initiated undertaking, in addition to its prohibitive cost, had a glaring missing gap in construction material—wood for crossties. The Hexi Corridor, of which Lanzhou was the largest metropolis, was completed denuded by the encroachment of the Gobi desert, making the proposed 860-km line from Lanzhou to Suzhou a complete pipe dream. Even the Tianshui–Lanzhou line was impractical due to the lack of large quantities of wood. Such circumstances once again made mounting demands on the upland wood in southern Gansu. Scott had no statistics to show us, but his camera faithfully documented the Muslim water hands rafting timber down from the Jishishan gorges to reach Lanzhou, the lonesome figures who must further denude the upper reaches of the Yellow River to ensure that the lifeline to the Chinese cause against imperial Japan keeps flowing.
It is also worth noting that the establishment of Gansu’s first match manufacturer, Lanzhou guangming huochai chang 蘭州光明火柴廠, also exemplifies the economic dominance of the Muslim Ma family in the region. Of the company’s 400 total shares, Ma Bufang and Ma Hongkui collectively controlled 240, securing an outright majority and cementing their influence over the enterprise. Their control was further reinforced by Ma Fuxiang, a key patriarch in the Ma family network, who held 40 shares. The remaining shares were split between Linxia’s magistrate, Deng Long, who owned 40 shares, and a broader pool of investors who collectively held 80 shares through crowd-funding.Footnote 44 This distribution of ownership illustrates how the Ma family leveraged both direct financial investment and political alliances to dominate emerging industries in Gansu. Their monopoly on upland wood, particularly the white poplar trees, was a key to the company’s success. Their majority stake ensured that decision-making power remained firmly within their network, aligning industrial development with their broader economic and political ambitions. In 1939, Ma Bufang founded Haiyang Chemicals Company (Haiyang hua xue chang 海陽化學廠). A year later, by acquiring all the equipment of Guangming Huochai Chang’s subsidiary in Xining, he turned Haiyang Chemicals into his own Qinghai Huochai Chang (Qinghai huochai chang青海火柴廠), equipped with six German-made basic match production machines.Footnote 45
Across the three films analysed so far—The Secret of A’xia River, the Holton films, and Kukan—the depiction of Hui Muslim wood traders and rafters shifts according to the political and ideological imperatives of their respective eras. In the early-twentieth-century Holton films, Hui traders and labourers appear as skilled, independent economic agents whose expertise in navigating the upper Yellow River underpins regional commerce. The films document the social reproduction of their labour, highlighting their dominance in timber transport and the broader economy of southern Gansu and eastern Qinghai. Their role is a natural extension of their historical participation in trade and transportation, unburdened by state intervention or ideological scrutiny. However, by the time of the Second World War, Kukan reframes their labour within the narrative of national survival. The documentary presents Hui rafters as critical to China’s war effort, transporting timber and supplies to sustain the military struggle against Japan. Here, their traditional skills are absorbed into a broader nationalist framework, transforming their work into an act of wartime mobilisation. The Hui presence is thus no longer merely an economic phenomenon but a strategic asset within China’s resistance.
By contrast, The Secret of A’xia River, produced during the Cultural Revolution, recasts Hui involvement in timber transport through the lens of class struggle and revolutionary vigilance. The film’s central conflict revolves around the theft of collective property, with Ma Hade, a Hui transport leader, emerging as the villain—an opportunist who conspires to exploit collective resources for personal gain. His portrayal reflects a broader Maoist suspicion of independent economic actors, particularly those with deep-rooted community networks outside direct Party control. The young protagonists, including the Hui boy Ma Jiajia, must learn to recognise and combat such ‘class enemies’, demonstrating the film’s pedagogical function in shaping socialist consciousness. Unlike the earlier films, where Hui rafters are autonomous or patriotic contributors to the state, The Secret of A’xia River subjects them to ideological judgement, framing their work as legitimate only when it aligns with Party interests. Together, these three films trace a historical trajectory in which Hui Muslim economic territorialisation centred around wood trade and wood transportation transitions from an organic, long-standing practice to a politically charged and contested site of state intervention.
The politics of economic territoriality in the historiography of Linxia’s Hui merchants
The question of Hui economic territoriality in southern Gansu also emerges as a contested issue in historical records. On the one hand, the PRC’s local official histories systematically obscure any explicit acknowledgment of Hui commercial dominance. On the other hand, inquisitive agents of the state—ranging from Republican officials to imperialist expeditionists and nationalist ethnographers—treated this very subject as a critical research concern. This disparity in representation reveals not only the entanglement of commerce and ethnicity in Linxia but also the ways in which different actors framed, emphasised, or concealed the economic power of Hui merchants.
External observations in the Nationalist era
A 1938 economic survey of Qinghai, authored by Zhou Zhenhe, for example, provides a striking example of how external observers fixated on economic sectors and ethnic dominance in the north-west.Footnote 46 The report details the region’s underdeveloped industrial base: coal mines remained dependent on coolie labour, and local ironsmiths engaged only in small-scale smelting, producing artisanal knives and daggers rather than industrial steel. While Qinghai possessed iron ore, lead, and copper, these resources were extracted and processed in strictly pre-industrial conditions.Footnote 47 It is important to note that Zhou’s report was commissioned by the KMT government, with one of its key objectives being to assess whether Qinghai’s regional hegemon, Ma Bufang, and his Muslim-dominated military force, the Huihui army, possessed the capability to independently manufacture firearms.
Ma Hetian (1887–1962), a prominent academic who accompanied the ninth Panchen Lama on his return to Kumbum from Nanjing in 1935 as part of a KMT delegation, also made detailed observations about Muslim dominance in Linxia’s commerce. In his journal, one of the earliest KMT-era texts advocating economic investment in the north-west, Ma highlighted Linxia’s distinctive system for funding public schools, which relied on revenue from waterway taxes levied on the Tao and Xia Rivers. These rivers, cascading down the Jishishan gorges before merging into the Yellow River, provided a crucial financial resource for local schools, especially Muslim elementary education.Footnote 48
A native of Shanxi, Ma Hetian was particularly interested in Linxia’s chambers of commerce. He carefully recorded that over the 17 years preceding his visit, the number of commercial firms in Linxia had plummeted from over 2,000 to just 300. However, this did not indicate an overall decline in investment. Rather, it was primarily outside firms—especially those from his home province of Shanxi and neighbouring Shaanxi—that suffered a drastic reduction, shrinking from two dozen firms to just four. Many collapsed under the burden of high interest rates on loans. By contrast, local firms not only endured but consolidated their holdings and even expanded, a clear indication of the increasing commercial monopoly of Hui Muslim capital.Footnote 49
Ma’s focus on the Muslim domination of Linxia’s waterways and the region’s general commercial competition contrasted with another dominant concern of state-sponsored studies—trade networks, particularly the lucrative wool industry. Japan, for instance, was particularly interested in Qinghai’s high-quality wool, especially a variety named sawaru さわる (meaning ‘touch’), which was considered superior to wool from all other Chinese regions.Footnote 50 Most Qinghai wool was exported via Tianjin to Japan and later through the transshipment port of Baotou in Inner Mongolia. To Japanese investigators, wool exported from the north-west was synonymous with Muslim wool. As early as 1918, Ma Qi opened branch shops of his family-owned companies De Yi Heng and De Shun Chang in Tianjin. Between 1924 and 1927, for example, a staggering 29 million jin (or close to 16,000 metric tons) of Qinghai wool left the port of Tianjin.Footnote 51 Trade statistics reflect a deep economic integration with Japan: despite the outbreak of the Second Sino-Japanese War, Japanese wool imports from Qinghai surged by 77.3 per cent.Footnote 52
This economic integration apparently persisted after the outbreak of the Second Sino-Japanese War. Reports from Republican officials note that Governor Zhu Shaoliang (1891–1963) complained to Chiang Kai-shek in 1938 that Xining’s wool was being trafficked by ‘treacherous merchants’ through Baotou and Hohhot.Footnote 53 However, given that Ma Bufang’s military regime tightly controlled the wool trade, these transactions were likely sanctioned rather than illicit. Most likely, the water hands that rafted the wool from Qinghai down to Baotou were the same Muslims recorded by Zhou Zhenhe and Ma Hetian.
Hui intellectual discourse aligned with Japan’s North China occupation
While external observers of the north-west show a more empirical interest, the Hui editors of The Muslim Weekly (Huijiao Zhoubao 回教週報), a contemporaneous newspaper affiliated with Japan’s North China occupation regime, directly engaged with the discursive question of Hui commercial identity, framing it not as an incidental aspect of regional trade but as the foundation of a distinct national economy. Through the newspaper’s editorial agenda, Hui elites in North China not only acknowledged but actively promoted a vision of Hui economic territorialisation. Their articulation of a Hui national/ethnic economy (huizu jingji 回族經濟) signified a deliberate attempt to define the boundaries of Hui commercial influence within the wartime order.
The North China regime’s standard administrative functions have been analysed in some depth by Prasenjit Duara.Footnote 54 However, what distinguishes the regime are two organisations specifically designed to cultivate grassroots political support: the New Citizens Societies (Xinmin hui 新民會) and the All-China Muslim League (Zhongguo huijiao zong lianhe hui 中國回教總聯合會). Established in 1942, the New Citizens Societies promoted Japan’s continental expansionist agenda among North China’s youth. Meanwhile, the All-China Muslim League, founded in 1940, sought to unify the region’s Muslim population into a centralised political organisation, channelling its influence to bolster Japan’s position against both the Chinese Nationalists in Chongqing and the Communists in Yan’an. The location of their headquarters underscored their significance: both organisations operated from Zhongnanhai, a section of the old imperial palace that also served as the nerve centre of the North China occupation regime and the headquarters of the North China Expedition Army.
The Muslim Weekly functioned as a central platform for articulating Hui political, economic, and social aspirations under the North China regime. While the All-China Muslim League worked to consolidate Hui political cohesion, The Muslim Weekly spearheaded an effort to frame Hui economic activity as a distinct national enterprise, elevating Hui commercial networks beyond local or religious affiliations into a broader ethnopolitical framework. By emphasising Hui economic independence, the paper sought to demarcate Hui territorialisation within the Japanese imperial order, presenting Hui commerce as an autonomous force within North China’s economic landscape.
A recurring column titled ‘Historical sources on Hui economy’ (Huizu jingji shiliao 回族經濟史料) catalogued industries historically monopolised or still dominated by the Hui, including wax-making,Footnote 55 leatherwork,Footnote 56 and the trade of fresh and dried fruits.Footnote 57 These economic activities were framed not as isolated trades but as components of a coherent Hui national economy (Huizu jingji 回族經濟). The column’s repeated emphasis on Hui economic heritage signalled an unmistakable desire to demarcate ethnic economic territorialisation, reinforcing intra-group cohesion while establishing clear boundaries between Hui and Han economic spheres.
Japanese ethnographic investigations reinforced this economic distinction. As detailed in Kelly Hammond’s study, Japanese officials identified the top six professions among Beijing’s Hui: fruit and vegetable vendors, pack animal traders, camel traders, beef and mutton butchers and vendors, duck vendors, and jade traders.Footnote 58 These industries, vital to the wartime economy, provided Hui traders with significant leverage. The Muslim Weekly framed their economic roles as a foundation for Hui self-sufficiency, arguing that their historical and contemporary dominance in these trades validated their claim to an autonomous economic identity.
The league’s propaganda chief, Jin Jitang 金吉堂,Footnote 59 further refined this vision by positioning Hui commercial success as evidence of their distinct national character.Footnote 60 The paper frequently cited Japanese ethnologists such as Teruyoshi Yoshihara 須田照義 of Tokyo University, who investigated the ‘Hui nation’ (Hui minzu 回民族) in Mengjiang and North China,Footnote 61 and Eiichiro Tameda 為田英一郎, whose work on the Hui was anticipated as a landmark study legitimising Hui identity within Japanese imperial discourse.Footnote 62 The Muslim Weekly eagerly promoted these studies, reinforcing the idea that Hui economic activity was not simply a byproduct of Chinese commercial life but an integral feature of a separate ethnonational identity.
This economic vision was not merely theoretical; it had practical implications for Hui self-governance. The newspaper chronicled the emergence of new Hui-led guilds, such as the recently established vegetable vendors’ guild chaired by a Hui leader. These organisations were framed as extensions of Hui political autonomy, with their economic organisation serving as both a reflection and a reinforcement of their distinct identity within the Japanese-controlled order. In sum, the Muslim platform within the North China regime provided the Hui with an institutional foundation to articulate and solidify their political and economic identity. While the Japanese administration viewed these developments as mechanisms of wartime control, the Hui editors of The Muslim Weekly strategically used the moment to assert a vision of Hui economic territorialisation—one that extended beyond immediate political contingencies to imagine a future in which Hui economic sovereignty would endure.
The PRC’s official historiography
The PRC state’s recognition of the Hui Muslims as a socialist nationality was an anomaly from the standpoint of Stalinist orthodoxy, which defined a nationality through four shared characteristics—a common consciousness, common language, common territory, and common economic life. When applied to Hui community clusters on the prefecture or sub-prefecture level, these four elements could arguably be identified. Yet, aside from a shared consciousness grounded in the Islamic faith, none of the other three traits hold true when the Hui are viewed from the macroscopic national perspective. As Dru Gladney and myself have argued, the Hui’s arrival as a socialist nationality can only be understood as a ‘corporate identity’—an amalgama that unified widely disparate populations under a single rubric, with limited attention to historical or cultural cohesion.Footnote 63
The socialist state’s leading historian charged with defining the newly promulgated Hui nationality, Bai Shouyi, thus contended in 1951 that the Hui possessed a distinctive economic system based on the religious tax of zakat (a mandatory charitable donation Muslims pay to the houses of worship, which is then distributed to the poor of the community) and the religio-political system of Sufi menhuan (self-governed, hierarchical religious organisations overseeing the transmission of power within a sect).Footnote 64 While both claims could be supported by historical evidence, neither applied exclusively to the Chinese-speaking Hui. The zakat system was equally practiced by Uyghurs, and the menhuan structure, though prevalent in the greater north-west—including Xinjiang, Ningxia and Gansu—excluded millions of Hui in eastern and central China. Faced with these inconsistencies, socialist ethnologists struggled to articulate a convincing argument for a common Hui economic life in the early PRC era.
Southern Gansu might have offered, at least regionally, an argument for economic commonality, yet it was politically inconvenient. Any explicit assertion of ethnonational identity or economic territorialisation there risked evoking the region’s recent history as a stronghold of the Ma militarist regime. Indeed, by the late 1950s, the socialist state began to target remnants of ethnic economic territoriality in Linxia and Ningxia as manifestations of ‘local nationalism’ (difang minzu zhuyi 地方民族主义), characterising the region as ‘a special zone of Hui–Han contradictions and conflicts’ (Hui han maodun de yige teshu quyu 回汉矛盾的一个特殊区域).Footnote 65 As the film The Secret of A’xia River illustrates, the collectivisation campaigns that followed the Great Leap Forward and the Cultural Revolution left no room for such forms of ethnic economic autonomy. Only in the 1980s and 1990s—alongside the revival of religion and the official discourse of ‘unity in diversity’ (duoyuan yiti 多元一体)—did the notion of ethnic economic territoriality reemerge, however cautiously and strained, as a legitimate historical narrative.
The 1993 Linxia Hui Autonomous Prefecture Gazetteer, produced under the editorial influence of many Hui cadres in the region, is the first official historiography in the reform era that exhibits a cautious ambiguity when describing Hui commercial dominance. It avoids direct claims about ethnic economic territorialisation. Instead, it offers an abstract and depoliticised portrayal:
They are good at commerce. Their specialisation in various craftsmanship, light processing industries, and the restaurant business are widely known and bear their own standard (chi yu gedi er du shu yi zhi 驰誉各地而独树一帜). Over a very long period of time, the Hui caravansaries of Linxia have traversed Beijing, Tianjin, Shanghai, Sichuan, Shaanxi, Yunnan–Guizhou, Qinghai–Tibet, Xinjiang and more, making themselves known as a crucial force in connecting the economies of east and west, north and south.Footnote 66
Rather than articulating specific sectors where Hui merchants historically dominated or monopolised trade, the gazetteer offers only an equivocal recognition of their role in commercial exchange. This vagueness contrasts sharply with the sustained scholarly and bureaucratic fascination with Hui economic agency. Fei Xiaotong, the architect of China’s Reform-era ethnic policy, for instance, explicitly characterised the Hui as the primary trading agents in Qinghai–Tibet commerce, emphasising their historical role in delivering goods to herdsmen and opening a ‘New Silk Road’.Footnote 67 His observations, made during a visit to Linxia in 1986, underscored an ethnic division of labour, where Hui expertise in commerce was juxtaposed with the perceived economic passivity of Tibetan herders and the unsuitability of Han traders for high-altitude trade.
This persistent interest in identifying an economic geography defined by Hui networks extended into the late twentieth century, particularly among Chinese scholars examining the historical role of Hui merchants in regional trade. By the 1990s, academic studies published in China built upon these earlier state-sponsored investigations but framed the discussion in ways that reflected contemporary sensitivities.
In 1994, Ma Zhankui and Ding Hua published a study titled ‘The history and characteristics of the development of Hui commerce in Linxia’, which systematically traced the historical trajectory of Hui commercial expansion. While acknowledging the historical prominence of Hui traders in Linxia’s markets, the study carefully avoided framing their dominance as an ethnic monopoly. Instead, it emphasised the adaptability of Hui merchants in different sectors, aligning with the cautious rhetoric of official histories.Footnote 68
A more explicit engagement with the economic territoriality of Hui networks appeared in Ma Ping’s 1996 article, ‘Hui middlemen in trade between the Tibetan borderlands of Gansu, Qinghai, Sichuan, and Inner China in modern times’. This study examined how Hui merchants in Linxia mediated commerce between inland Chinese markets and the Tibetan borderlands, revealing a well-integrated economic network that extended deep into Qinghai and Sichuan. Ma Ping’s research reaffirmed the historical depth of Hui commercial dominance, yet, like other late twentieth-century Chinese studies, it stopped short of drawing explicit political conclusions about ethnic economic autonomy.Footnote 69
A broader historical perspective was taken by Mian Weizhong in his 2005 article, ‘A brief discussion on Hui–Tibetan trade in the Hehuang region during the early Qing Dynasty’. This study pushed the timeline back to the early Qing period, demonstrating that the Hui role in facilitating trade between Han-dominated China and Tibetan pastoral regions was not merely a late imperial or modern phenomenon but had deep historical roots. By embedding Hui commercial networks within a longer trajectory of interethnic exchange, Mian’s work reinforced the notion of Hui economic distinctiveness while adhering to the state’s preference for narratives of interethnic cooperation over economic competition.Footnote 70
The historiographical contrast is clear: external observers—whether Republican officials, Japanese imperial strategists or modern ethnographers—routinely emphasised the Hui’s territorialised economic dominance, whereas local official histories systematically avoided making such claims. The late twentieth century Chinese academic literature reflects this tension, simultaneously acknowledging Hui economic networks in southern Gansu while conforming to the official historiographical trend of downplaying ethnic economic divisions. This tension reveals the political sensitivities surrounding ethnic commercial networks in the north-west. While the Hui trading presence in southern Gansu and eastern Qinghai was an open secret to outside observers, local historiography has preferred to elide explicit discussions of economic territoriality, leaving behind a record that is as notable for its silences as for its affirmations.
Conclusion
The surge of timbering and rafting activities along the upper reaches of the Yellow River, from Xunhua on the eastern edge of the Tibetan Plateau to Baotou in the Ordos Loop of Inner Mongolia in the first half of the twentieth century, is best understood through Fernand Braudel’s concept of ‘social time’.Footnote 71 The exchange of key resources, most notably timber and wool, was concomitant with the rise of Muslim political power in this region. As this article has demonstrated, the sharp increase in economic activity was both a cause for and a consequence of the profound political transformations led by Muslim militarists. A key indicator of such political transformations was the fall of Ningxia, just south of the Ordos Loop, into the firm control of Muslims whose original homes and upstream stronghold lay in Linxia.
Yet in the second half of the twentieth century, this social time came to an abrupt end. Massive dam projects, including the Liujiaxia Reservoir and numerous smaller counterparts built from the 1950s through 1970s, rendered long-distance rafting impossible, while socialist collectivisation, as reflected in the film The Secret of A’xia River, further stifled private entrepreneurial activity. Though now a relic of the past, except for a small stretch of the Yellow River crossing Lanzhou, the history of rafting and timber trade provides key insights into the formation of Hui ethnic identity in the early twentieth century and its complex relationship with the Nationalist state as well as its communist successor in the decades that followed.
The historiographical treatment of Hui economic territoriality in Linxia reflects broader tensions in how ethnic commerce is framed within different power regimes. Hui elites within Japan’s North China regime actively embraced and articulated their economic distinctiveness, framing it as the foundation of a Hui national economy. Through the All-China Muslim League and The Muslim Weekly, Hui intellectuals sought to reframe commercial dominance as an asset rather than a liability, positioning Hui economic networks as an essential, self-sufficient pillar of national identity. This self-presentation not only diverged from contemporaneous Chinese official observations but also reflected the broader political realities of wartime collaboration, where imperial patronage allowed for the expression of economic nationalism in ways that would have been untenable under Chinese Republican rule.
In stark contrast, local Chinese historiography, particularly in the post-revolutionary era, has persistently obfuscated the issue, portraying Hui economic activity in abstract, depoliticised terms. This contrast is not merely one of scholarly approach but speaks to deeper political anxieties: the overt acknowledgement of Hui economic autonomy risks unsettling official narratives of ethnic harmony and national unity. The reluctance of late twentieth-century Chinese scholars to explicitly engage with Hui economic territorialisation, despite clear historical evidence, illustrates the extent to which ethnic economic power remains a politically sensitive topic.
This article’s film analysis—analysing cinematic representations of Hui communities and their economic networks—complements these historiographical insights by providing a visual lens through which the politics of economic territorialisation are not only documented but also performed. Films that engage with the complexities of Hui identity often reflect and amplify the ethnic tensions and historical struggles faced by these communities. In the filmic portrayals, the Hui are not merely passive subjects of state narratives but active agents shaping their own economic networks and sense of territoriality. The interplay between state-sponsored ethnographies and the imagery of film highlights how visual media has been instrumental in both preserving and challenging official narratives.
Ultimately, the historiographical silence surrounding Hui economic territorialisation is as revealing as its explicit articulation under Japan’s occupation, demonstrating the fluidity of economic identity as a tool of both political negotiation and historical erasure. Through the lenses of both scholarly discourse and cinematic representation, we see that the construction of Hui commercial identity is far from static—it is continually reshaped by political agendas, whether in the context of Republican China, Japanese wartime governance, or contemporary national discourse. In this light, the filmic medium offers a unique space for understanding the multifaceted nature of ethnic economic territorialisation, allowing for the expression of Hui self-determination in ways that written histories, constrained by political pressures, might not fully capture.
Acknowledgements
I am grateful to Hannah Theaker, David Stroup, and colleagues in the Islam in China Studies Group, who first offered feedback on this paper at the AAS conference in Columbus, Ohio, in spring 2025. I also extend my sincere thanks to Professor Tomoko Shiroyama and Professor Asuka Imaizumi for the opportunity to present this work at CIRJE at the University of Tokyo, where I benefited from stimulating and thoughtful discussion. I am likewise indebted to the two anonymous reviewers for their constructive and insightful comments.
Conflicts of interest
None.
Open access
