Not a Silk Road: trading networks between China and the Middle East as a dynamic interaction of competing Eurasian geographies

This work was supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme 669 132 – TRODITIES, ‘Yiwu Trust, Global Traders and Commodities in a Chinese International City’.

Middle East, running along an east-west axis. A related argument is that with the revival of this route, the Middle East now no longer looks principally west to Europe and the USA to source most of its consumer goods, but rather East to China (Simpfendorfer 2009). Several accounts of this newly invigorated transregional commerce emphasize that the emergence of transnational economic networks of Arab and Middle Eastern merchants connecting China to the Middle Eastern markets has been fostered by China's 'Muslim diplomacy' (Belguidoum and Pliez 2016;Gladney 1994). These accounts conceive of Yiwu's transnational economic networks as embedded in a shared Muslim culture, which has been specifically materialized in the creation of an Arab-Muslim ethnic enclave in the city of Yiwu Pliez 2012, 2016).
In this article, I draw attention to a contrasting dynamic of the internationalization of Chinese commerce that differs in key respects from the accounts of an emerging or re-emerging east-west axis of transregional connectivity. While the Arab traders active in Yiwu described below supply goods to markets in the Middle East along such an east-west axis, formative elements of their economic networks and of the commercial geographies they inhabit are obscured if we think of their transregional activities as the modern revival of a previously well-established east-west corridor. I propose an alternative theoretical frame, one that argues that the commercial geographies fashioned by Arab traders in Yiwu are, rather, in a dynamic relationship of competition and cooperation with a series of distinct but overlapping Eurasian political geographies which have been in process from the 1970s onwards. 1 Approaching the commercial routes fashioned by Arab traders as contingent and processual geographies, I analyse them as a feature of recent and contemporary geopolitical rivalries between Soviet, Sinocentric and Turko-Islamic state projects to imagine and construct a connected Eurasia.
If these different geopolitical visions of connected Eurasias have, over the last forty years, been fashioned by states from the top-down, the dynamic relations of competition and cooperation between them have been mediated from the bottom-up, by small economic networks of Arab traders who have fashioned trading careers between Taizz, Aleppo, Moscow, Yiwu and Istanbul. They have succeeded commercially through a pattern of sojourning in different cities, transferring capital and contacts accumulated in one Eurasian pole, to another. What kind of networks are these and how should they best be analysed? Much has been made of the Muslim and Arab ethnic nature of the transnational economic networks anchored in Yiwu, which are seen not only as incidental characteristics but as contributing to the cohesiveness of the networks in question and of the durability of Yiwu -which has attracted large numbers of Hui Chinese Muslims to act as cultural intermediaries -as a trading hub. More broadly, much literature on transnational economic networks connecting Yiwu to the rest of the world implicitly adopts a network governance approach that focuses on how shared cultures, religions or ethnicities generate trust effects within the networks in question. Yet, it is not clear that either Islam or Arabness is a decisive factor in the cohesiveness and durability of the commercial networks I describe; some merchants explicitly reject its relevance; and in many cases, the shared histories and institutions that ground mutual familiarity among traders are neither overtly Islamic nor Arab but rather the product of socialist and post-socialist geopolitical projects.
While a network governance approach implies the centrality of trust to the functioning of economic networks, ethnographic data collected in Yiwu has led me and others to question the significance of trust to the commercial strategies of Arab traders operating out of the city (see Introduction and Marsden in this issue). In this article, rather than adopting a network governance approach that sees these economic networks as embedded in a shared culture that furnishes trust between members, I adopt a structural analysis approach that sees these traders as brokers moving and mediating between different geographies, acting strategically and opportunistically to keep several in play at once in order to keep their options open when faced with an unforeseeable future and marginal citizenship rights. This has implications for how we analyse the likely durability of these commercial geographies. Drawing on a structural analysis approach, I argue that the persistence of Arab trading offices in Yiwu will be shaped by the effects of competition between alternative Eurasian geographies, notably between the Turko-Islamic and Sino-centric iterations of Eurasia. Turkey has recently been able to draw the capital of Yemeni and Syrian traders supplying the Saudi market away from Yiwu by competing on cost-of-living and on the offer of longterm residence and citizenship rights.

Commercial geographies of China-Middle East trade
Since the late 1990s, the city of Yiwu in China's southeastern Zhejiang province has become a major wholesale market supplying low-grade commodities to international traders who visit the city. It is famous among these merchants for its Futian market, which houses some 70,000 wholesale outlets for Chinese manufacturers of small household items, ranging from socks and kitchenware to toys and cosmetics. Several thousand merchants, mainly from Central and South Asia, Russia, the Arab world and Africa, are resident in the city, having established trading offices to coordinate exports and provide logistical and customs support to merchants visiting from across the globe. These visitors typically stay for a few weeks at a time on wholesale provisioning trips (Marsden 2015(Marsden , 2016a(Marsden , 2016b(Marsden , 2017. The city is therefore a node where diverse trading networks mingle and coalesce. Analysts have described the supply routes between Yiwu, a wholesale market city in China's commercially vibrant Zhejiang province, and its outpost markets in Asia, the Arabian Peninsula and North Africa, as a new 'Silk Road' (Belguidoum and Pliez 2012Broadman 2007;Simpfendorfer 2009). Belguidoum and Pliez (2016) note that the influence of Yiwu 'has grown to such an extent that the city has become the main destination for North African importers'. Their studies of Yiwu's emergence as a transregional trading hub document the trajectories of Algerian and Egyptian transnational entrepreneurs whom they see as fashioning new Silk Roads between China, the Arab world and Africa (Belguidoum and Pliez 2012. They reveal how the purchasing trips of North Africans were moving further and further east, from Marseille in the 1980s, to Istanbul, Dubai, Bangkok and Hong Kong in the 1990s, to Guangzhou and Yiwu after China entered the WTO in 2001 (Bertoncello et al. 2009). Dubai was a key forwarding point for Chinese goods (Lavergne 2002), but it was not long before Algerian entrepreneurs decided that 'China is the source of all products. Let's leave for the Gulf and go further east, to the source' (Belguidoum and Pliez 2016).
Seen from a North African perspective, a critical feature of the new Silk Roads converging on Yiwu is the east-west axis along which they are constituted. Analysts of the modern Silk Road connect China's emergence as a global trading hub for Arab and Muslim worlds to political watershed moments such as 11 September (9/11), when Muslim merchants who had previously visited markets in Europe and North America suddenly confronted new social and regulatory barriers in these societies that led them to change their orientations from West to East (Belguidoum and Pliez 2016). A decade earlier, EU and US sanctions against China following the events of Tiananmen Square in 1989 had led Beijing to seek new trading partners in the Middle East and North Africa (Belguidoum and Pliez 2016). China then embarked on an intensive period of Muslim diplomacy to foster these connections (Gladney 1994). Sensitive to these developments, local authorities in Yiwu successfully positioned their city as a Muslimoriented marketplace, building a mosque in 2003, and attracting thousands of Chinese Hui Muslims to serve as cultural intermediaries for merchants visiting from the Islamic world, and to staff an emerging industry in halal restaurants, supermarkets and hotels catering to these visitors. The sense that Yiwu has emerged as part of a commercial, cultural and political geography that turns on an east-west axis is also reflected in the way that North African entrepreneurs visiting the city narrate their own commercial and personal trajectories. Narratives of exploration deploy a trope of 'going East': individual traders present themselves as explorer-pioneers travelling further and further east in search of the 'source' of commodities, with Dubai acting as a key relay point in these ventures eastwards (Belguidoum and Pliez 2016).
Such imaginative geographies of a new Silk Road emerge from the trajectories and narratives of Algerian and Egyptian traders. Those of Syrian and Yemeni traders active in the city of Yiwu reveal other geographies, histories and accounts of self. Rather than pioneering journeys eastwards mediated by a shared Islamic culture, their trajectories reveal geographies and histories of Soviet patronage -careers built in Moscow, facilitated by Soviet and Russian higher education scholarships -and thus narratives of self-development connected not to discoveries in markets in Dubai, but to other circuits of mobility connected to Soviet and post-Soviet worlds. In exploring these, the analytical gaze of the article shifts from north Africa to traders hailing from the east of the Arab world; it also shifts from traders visiting Yiwu for short periods, to those who are resident in the city. The scholarly metaphor of 'Silk Roads' emerges mainly from the narratives and trajectories of those who visit the city on provisioning trips. Another important category of traders active in the city of Yiwu are residents -the owners of trading offices that offer logistical services. The geographies fashioned through their trajectories can be more varied than the development of an east-west axis embedded in a transregional Muslim culture.
Many Yemeni traders resident in Yiwu were able to establish themselves in commerce and enter transregional circuits of mobility thanks to Soviet, Russian and Chinese higher education scholarships. As part of its efforts to cultivate client states in West Asia, the USSR had through the 1970s and 1980s offered free places at Soviet universities to Yemenis from both north and south Yemen, as well as citizens of other Arab nations. Uday, a 50-year-old owner of a trading office catering to the Saudi market, is an example of a trader who was able to embark on a career as a transnational merchant thanks to a Soviet programme of international student mobility. In 1985 he scored well in his baccalaureate exams in Taiz and was offered a choice by the government in Sana'a of a scholarship to either a US or Russian university. Lacking the level of English needed for the USA he took the offer of Moscow. There he married a Russian woman, acquired Russian nationality, and established an import office in Moscow in the 1990s. The building in which he opened a trading office had been converted from a student dormitory: the infrastructures of student and merchant life were closely connected (Ibañez-Tirado 2019). He then moved to Yiwu in 2004, as the city's Futian market was rapidly developing in the wake of China's accession to the WTO in 2001.
Other Yemenis were drawn to China and settled in Yiwu because of Chinese higher education scholarships. An example is Talal, another trader from Taiz in Yemen, now in his late forties and supplying markets in Saudi Arabia. He had studied at the University of Shanghai in the late 1990s on a Chinese government scholarship and later opened a trading office in Yiwu. The pattern of his career was not unusual, and illustrates that Yemenis were able to establish similar connections between student mobility and international wholesale trade in post-WTO accession China as in post-Soviet Russia. Yemen was a particular focus for China's socialist development efforts and attempts to establish a diplomatic and economic presence in West Asia. While its patronage reflected socialist southern Yemen's early recognition of the People's Republic of China, the Chinese state had also offered support to north Yemen during the 1980s, and maintained its programme of higher education scholarships after Yemeni reunification in 1990. The granting of scholarships for Yemenis to study technical and engineering subjects led to significant numbers of Yemenis studying at universities on China's east coast. Many cultivated excellent Chinese language skills, and boosted their income by working as translators for Arabic-speaking merchants visiting trade fairs in Shanghai and Guangzhou after China joined the WTO. On graduating, many found that they were unable to access jobs back in Yemen, and decided to forge a trading career in China, catering to Yemeni visiting merchants who supplied the Saudi market. The majority of the graduates who opted to stay hailed from Taiz, an area which they saw as marginalized from state resources and employment by Ali Abdullah Saleh's government through the 1990s and 2000s. Yemeni trading networks in Yiwu grew through the late 1990s and early 2000s with the expansion of the city's Futian wholesale market and the increasing internationalization of Chinese higher education. According to one Syrian merchant who moved to Yiwu in 2005, the Yemeni students were the 'axis' (al-mihwar) who opened the markets and facilitated the arrival of other Arab students and merchants.
Like some Yemeni traders, many of the earliest Syrian ones came to Yiwu after having established themselves in Moscow. Syrian traders, like those of other Arab nationalities, have been in Yiwu since the late 1990s, when they started visiting the city to provision wholesale markets, particularly in Moscow and Dubai (Anderson 2018). There were around fifty Syrian trading offices in Yiwu by 2006, of which most had been established to supply Russian markets. Many of the first generation of Syrians came with Russian connections, having spent the 1990s in Moscow. One example is Ahmad Salteen, the owner of a trading office in Yiwu, now in his late forties. The son of a farmer from a village in Idlib province in northern Syria, he had moved to Moscow in 1992 at the age of 21. His older two brothers had left Idlib for Moscow in 1984 and 1990, when Syria was under Soviet patronage. He joined them after the collapse of the Soviet Union, working at first as their employee (mwazzaf), before opening an import office on his own. Through the 1990s, he imported perfumes from the UK and France, stockings from Italy, vaporizers from Poland, hibiscus flowers from Sudan, clothes and shoes from Syria, leather and clothes from Turkey, clothes from Vietnam and, latterly, perfumes and fashion accessories from China.
Ahmad established himself in an office at the Sevastopol Hotel, where Uday, the Yemeni trader introduced above, was also based. The two became neighbours and close friends. Like Uday, he acquired Russian citizenship while trading in Moscow, before moving to China. In 2001, seeing the Russian market declining, and hearing that Yiwu was becoming an important source for low-grade commodities, he opened a trading office in Yiwu. Until 2004, he remained based for much of the year in Moscow, importing cosmetics, perfumes and hair accessories from Yiwu, before moving his main residence to the city in 2004. Some fifteen years later, in 2019, Ahmad is considered by many Syrian owners of trading offices in Yiwu as a pillar of the community and one of the first to open the Yiwu market to Syrian merchants. His ability to do so was grounded in the decade he spent in Moscow. His sojourn in Russia enabled him, a farmer's son, to build mercantile capital, networks and expertise, which he later transferred to Yiwu. Through the 1990s and early 2000s, Ahmad imported goods from across Europe, the Middle East and later China. He sold them on in open containers in Moscow's huge 'popular' (sha'bi) or informal markets -the Cherkizovsy and Luzhniki container markets -to Russian and Vietnamese traders who transported them through the former USSR, supplying Russian and central Asian markets. As one Syrian office owner in Yiwu said of those who had worked in Moscow, Russia was 'where they learned to move goods across borders'.
This was also true of others who did not inherit commercial capital and networks in Syria. Rafiq was a 54-year-old merchant from Aleppo resident in Yiwu; he had left his job as a civil servant in Syria in 1992 and settled in Moscow. At first, he imported cosmetics from Britain into Russia and Ukraine. Then, in 2002, he moved to Yiwu where he established a cosmetics factory supplying both the Russian and Syrian markets. His sojourn in Russia had familiarized him with market tastes there, and this shaped his subsequent activities in China. Hamdan was another Syrian trader who supplied the Russian cosmetics market from Yiwu. Now in his early fifties, the Syrian state had sent him from Aleppo in 1989 as part of a student mission to study petrochemical engineering in Irkutsk. While there, he also started trading on a small scale, flying to Beijing in 1991 and provisioning himself with goods to sell in Irkutsk. On graduating he moved to Moscow and opened a trading office, later taking Russian nationality and opening a cosmetics factory. In the 2000s, he established a second cosmetics factory near Yiwu, exporting all its output to his trading office in Moscow. His connections with Russia endured after he established a base in Yiwu; in 2018 he continued to alternate his time between China and Russia, rotating between Moscow and Yiwu every two months.
In 2018, Ahmad, Rafiq and Hamdan were all operating from Yiwu; yet, their commercial careers had started in Moscow and were embedded in an important geopolitical relationship -Soviet and then Russian patronage of the Syrian regime. One way in which this patronage had been expressed was in a Soviet-era payments agreement. In 1973, the USSR agreed to accept payment for the military hardware it had sold to the Syrian regime in the form of textiles and fashion accessories. This agreement fostered private commercial relations between the USSR and Syria. Throughout the 1980s, Russia was the most significant market for Syrian exports, estimated at a third or more of all Syria's private exports in 1989; when the rest of the socialist bloc is included, that figure rises to 80 per cent. The legacy of these arrangements continued into the post-Soviet period. In 1991, the formal patronage ended, but the commercial networks endured, as Russian importers initially lacked commercial relations beyond the formerly socialist bloc, so continued to depend on Syrian factories. These arrangements encouraged many Syrians -such as Ahmad and Rafiq -to migrate north to Moscow and seek employment in trade and related services in the 1990s, organizing the importation of Syrian and other goods into Russia. Another feature of the Soviet and then Russian patronage of the Syrian regime was the provision of higher education to Syrian nationals. Several Syrians in Yiwu had first engaged in transregional mobility as university students in Russia on USSR scholarships. After completing their degrees, several stayed and sought to establish a career in commerce -as Hamdan's career illustrates.
Their careers were forged under the legacy of a Soviet project to create transregional connections across Eurasia (Anderson 2019). The commercial and educational patronage of Syria, which had brought them to Moscow, had been part of a broader Soviet and then Russian strategy to project influence into the Middle East and central Asia (Kirasirova 2011;Nunan 2011;Yolacan 2017). The USSR had offered scholarships to bring Syrians, Yemenis, Iraqis and Egyptians to its universities under the aegis of international socialism. Later, as they sought to carve out a new transregional role, the Russian authorities built large wholesale markets as commercial gateways to central Asia. Among the Syrian and Yemeni traders who cut their teeth within these institutions were those who then helped to build Yiwu as a transregional commercial hub. Their personal trajectories and the commercial geographies they fashioned were embedded first in a Soviet vision of a connected Eurasia, and then in Russian attempts to fashion new forms of inter-regional connectivity after the collapse of the USSR.
After leaving Moscow for Yiwu, they continued to supply post-Soviet markets in Russia. However, they soon shifted their focus to West Asia when China negotiated trade agreements during the 2000s that opened markets across Asia and Africa to consumer imports. Thus, they moved from the remnants of a Soviet vision of Eurasia to an emerging Sino-centric vision of this expansive transregional space. The local authorities in China positioned Yiwu as a gateway to Sino-centric Eurasia, expanding the Futian market until it comprised some 70,000 booths. In the following decade, the vision of Eurasia being connected by Chinese commodities and infrastructure became a hallmark of Xi Jinping's presidency under the 'Look West' (Li et al. 2016) policy and most recently the 'Belt and Road Project' (Clarke 2017;Ferdinand 2016;Ploberger 2017;Rolland 2017;Sidaway and Woon 2017). Yiwu was marketed as a key staging post, and foreign merchants continued to flock to the city. Those who moved from Russia were able to profit not simply by moving between different markets -for example from the Luzhniki container market in Moscow to the wholesale Futian market in Yiwu -but also by shifting from one Eurasian geopolitical project to another. While acting as middlemen connecting factories in one part of Eurasia with markets in another, they also layered different political geographies together. Having acquired commercial and political assets in post-Soviet Eurasia -merchant networks, capital, experience and citizenship -they deployed them within a Sino-centric vision of Eurasia, shipping Chinese commodities to the Levant.
Viewed from the top down, this Sino-centric vision of Eurasia is quite distinct from the Soviet vision of connected Eurasia. However, viewed from a perspective of 'globalization from below' (Tarrius 2000), from the vantage point of small economic networks of mobile Arab traders, Yiwu's rapid economic development is also layered into an earlier Soviet geography and its Russian legacy (Ibañez-Tirado 2018). The capital, expertise and mobility-enabling passports they deployed in opening Yiwu to Middle Eastern markets, arose out of that earlier geography. Moreover, in 2018, Syrian and Yemeni traders continued to mediate between the two Eurasian geographies. Ahmad, for example, maintained a residential property outside Moscow, which he continued to visit regularly, thus maintaining his networks with Syrian and Yemeni importers there. Hamdan alternated his residence between Yiwu and Moscow every two months. Both continued to export Yiwu goods to markets in Russia. They found ways to flourish and make profits in the transitions, interstices and overlaps between two geopolitical projects of connected Eurasia. Yiwu did not only emerge as transregional trading hub along a 'Silk Road' east-west axis, but it was also layered onto this earlier Eurasian inter-regional space. The commercial geographies fashioned by Arab traders connecting China and the Middle East are therefore more complex than suggested by the revival of an older East-West corridor; they are also imbricated in the dynamic relations between recent Eurasian geopolitical projects, and the shifts and overlaps between them.

Two traditions of network analysis
Another important feature of these accounts about the re-emergence of a 'Silk Road' corridor is the focus on Muslim culture as a medium for the interface between Chinese and Middle Eastern trade. They mention how the Chinese authorities sought to develop trade relations with countries with a majority Muslim population through the 1990s, with Chinese Muslims, both Hui and Uighur, playing an important role in fostering these relations (Belguidoum and Pliez 2016;Gladney 1994). At the urban level, Muslim diplomacy is seen as shaping the city of Yiwu by fostering a halal economy and encouraging Hui Muslims to act as cultural intermediaries between Chinese suppliers and visiting Middle Eastern merchants. Belguidoum and Pliez (2016) note in particular the role played by an Arab restaurant district in downtown Yiwu, and a surrounding area specializing in 'trade specially with the Arab and Muslim worlds'. It is a 'real landmark in the centre of the town where Muslim wholesalers meet in the later afternoon' with a 'strong Islamic feeling' that 'meets the need for Muslim wholesalers to find halal food' and 'familiar cultural reference points': the managers and staff are all Muslim and 'Arabic is their lingua franca'. This 'Arab enclave' plays an important part in Yiwu's globalization from below, since it is where the city's transnational networks materialize. The city depends on the 'geography of discrete networks woven by thousands of actors who have built real trade routes'; the networks are 'made up of solidarity among co-religionists, cultural references and the economic rationales of grassroots globalization'.
This urban quarter has undoubtedly played a significant role in the development of Yiwu as a transregional commercial hub, as a centre for short-term merchant visitors to the city from the Middle East and Africa. Yet, shifting our gaze to Arab residents and office owners in the city affords a different perspective on Yiwu's transregional networks. When the Syrian trader Ahmad and the Yemeni merchant Uday moved to Yiwu from Moscow in the early 2000s, they opened offices next to each other, in neighbouring buildings, thereby continuing their neighbourly relationship, which had started in the Sevastopol Hotel in Moscow where they had also shared adjoining premises. These tower blocks were known for their commercial prestige, and as being among the first to be built in the city, rather than for being in a distinctively Islamic area. What led them to cluster their offices next to each other was not their shared ethnicity or religion, but the fact that they had already established a relationship of neighbourliness in Moscow. This was an expression of a deeper similarity in their positions and trajectories: both hailed from marginal countries in formerly Soviet Eurasia; both had been drawn to Moscow through a legacy of Soviet patronage; and both had subsequently pivoted in the early 2000s in an emerging pole of a Sino-centric Eurasia. The coincidence of their trajectories can be attributed to the similarity of their structural positions in relation to post-Soviet and Sino-centric Eurasian political geographies.
Their shared position can be analysed as one of brokerage in relation to wider Eurasian formations, and their status of marginality in relation to these. The notion of brokerage implies a different theoretical perspective on networks from the one that underpins accounts of a new Silk Road. The 'New Silk Road' and Muslim diplomacy perspectives on trans-Asian commercial networks highlight the forms of culture and ethnicity in which transregional networks are embedded, and which provide a sense of internal cohesion and solidarity. As Belguidoum and Pliez (2016) put it, 'grasping the complexity of these networks requires identifying and disembedding the networks' content, which is made up of solidarity among co-religionists.' By contrast, the notion of brokerage focuses not on the forms of religion or culture in which networks are embedded, but on the ways in which the position of social actors within a broader network affords them possibilities for calculative and strategizing behaviour. As Grabher (2006) has argued, 'the perception of networks in economic geography is mainly informed by the network governance approach that is founded on Mark Granovetter's (1985) notion of embeddedness', thereby discarding an older tradition of studying social networks that focuses on network position and structure (Simmel 1890) and the 'calculative and strategic perceptions of networks' animating more recent work such as that of Burt (1992Burt ( , 2004. Of the two traditions that Grabher (2006) delineates, the network governance approach is concerned to understand the social and institutional forms -such as informal personal ties, the thick bonds of kinship and ethnicity, the common possession of group norms -in which social actors are embedded, and which therefore govern how their networks are initiated, coordinated and monitored. This is the analytical tradition that implicitly informs analyses of economic networks as a new Silk Road. The premise is that networks provide social cohesion, promote 'trustful ties that curb opportunism and engender cooperation' and thus lubricate commerce. Ethnic enclaves in particular can 'suffuse otherwise bare relationships with a sense of purpose, and provide enforceable trust' (Grabher 2006). By contrast, the tradition of social network analysis seeks to understand social actors' behaviour as a function of their position within a wider structure of social relations (Scott 2000;Wasserman and Faust 1994;Wellmann 1983). Burt (1992Burt ( , 2004 has drawn attention to the possibilities of action inherent in the position of a 'structural hole' that bridges two otherwise unconnected networks. The broker, or tertius gaudens ('third who benefits') occupying such as position is able to engage in arbitrage and to innovate, by 'selecting and synthesizing alternative ways of thinking' (Grabher 2006). Such a position also provides a measure of autonomy and manoeuvrability, for it enables actors to engage in 'strategic play' by 'keeping options open across unforeseeable futures'. In contrast to the network governance approach, he sees networks not as sites of sharing, trust and mutuality but as 'vehicles [through which] to pursue opportunistic behaviour' (Grabher 2006). This understanding of economic networks better reflects the agency of Arab offices in Yiwu than of ethnic enclaves suffused with a solidarity embedded in shared culture. It allows us to approach Syrian and Yemeni traders not as part of a cohesive trustful network, but as middlemen in small economic networks who survive by keeping different contexts in play, as I describe below.

Lives not livelihoods: Arab traders as marginal brokers between different Eurasian geographies
In September 2018, the Syrian trader Ahmad welcomed his Yemeni friends Uday and Talal into his office on the 23rd floor of one of the prestigious 'Times Square' twin towers in Yiwu. The two guests sat in the comfortable leather armchairs in Ahmad's executive suite, with an Arabic news station playing silently on the large flat-screen television on the wall. Ahmad handed them papers to sign that certified their ownership of flats in an apartment block in Asinyut, a northwestern suburb of Istanbul. Ahmad had recently returned from Dubai, making a one-day stopover in Istanbul where he had purchased five flats in the same apartment block -one for himself, and the other four on behalf of Uday, Talal, his cousin in Lebanon and an acquaintance in Belorussia. He had purchased the apartment in Istanbul, he said, not as an investment, but as a place to live and possibly trade in the future. Uday and Talal would likely be his future neighbours in Istanbul, just as Uday had been in Yiwu and in Moscow. All three understood their role as a temporary one, anticipating that when Chinese manufacturers had learned from them how to export, they would need to move on again. Ahmad said: When we first came they treated us like kings (mitl muluk)! They needed us, didn't know how to import or anything. Now they have learned from us. The customers didn't come [at first] because they were afraid. It was because of us that they came. Now they have the customers, they know about the goods and can deal directly. So they make everything complicated for us -housing, residence permit, business, everything. They need us now for import, to learn how to import. They have been turning to the import system now, for the last two years. … When they have got used to importing, they will push us out! Hailing from the marginal countries within the Eurasian geopolitical visions, whether Soviet, Russian or Sino-centric, they had capitalized on their marginal position by papering over the transitions between different Eurasian geopolitical projects and operating in the interstices between them. Viewed from a perspective on livelihoods, that was a successful commercial strategy, but viewed from a perspective on lives, it contributed to a sense of themselves as marginal actors -impermanent and dispensable. This sense was particularly acute in China, where their residence permits required renewal every year in most cases, and they had no avenues to permanent residence or citizenship.
Ahmad, Uday and Talal were positioned as middlemen between different Eurasian geographies, each with its key hubs or poles that offered possibilities not only to earn a livelihood but also for life. They evaluated different cities in these geographies by referring not only to their commercial and logistical infrastructures, but also to the broader possibilities of a stable life that they afforded, through long-term residence rights, accessible schooling, the cost of living and the moral-cultural environment in which to bring up children. A recurring concern was the issue of residence rights, which led some Yemeni traders in Yiwu, who had forged successful careers and achieved a 'good standard of living' (ma'isheh kwayyiseh) to claim 'we haven't benefited at all' (ma istafadna shi). Ahmad described Turkey as comparing favourably with China on rights of residence and the proximity of Turkish and Syrian culture: 'I won't invest in China: there is no stability (istiqrar) here. You get a visa for one year only. Any day they can tell me to leave. Turkey is good: it is like Syria in culture, conservative: not necessarily religious, but conservative.' He also compared Turkey favourably with Russia on grounds of moral culture. While he possessed Russian nationality, he preferred Turkey over Russia as a place in which to educate his young teenage son: [Turkey] has Islamic customs and traditions ('adat wa taqalid). … In Russia moral dissolution (al-inhilal al-khulqi) has gone much further. It is OK if you go as a single man. But to raise a family?! In Turkey, there is halal food, so you can relax. And there are mosques everywhere.
This wider focus on lives rather than livelihoods captures an important part of how they reflected on the geographies in which they operated. Reflecting on their lives rather than their livelihoods, they understood themselves as marginal rather than successful actors. This sense of marginality was understood to underpin the affinity between Yemeni and Syrian merchants, especially after 2011, when both were acutely conscious of their shared location in a global hierarchy of exposure to geopolitical instability and suspicion. Several merchants in Yiwu and Dubai pointed out to me that Syrians and Yemenis might stick together because they were both from 'countries in which there are problems' (bilad illi fiha mashakil). Ahmad said that 'we stick together because Yemen and Syria are both in the same situation: wars (hurub) and instability (istiqrar ma fi).' Both Syrian and Yemeni traders talked of suffering as a result from restrictions on bank accounts and international bank transfers, severe restrictions on global mobility, difficulties in obtaining Chinese visas for their clients, and even finding their access to hotels barred when travelling within China.
Passport 'strength', based on the number of countries it was possible to visit without visas, was a recurring topic of discussion. In 2018, Syrian and Yemeni passports ranked among the lowest globally. The possibility of enhancing one's own passport mobility was also a recurrent concern. Some Yemeni traders had acquired Dominican citizenship on the strength of a $100,000 investment in the country. Syrian traders talked of the possibility of acquiring Latin American passports, or long-term residence in countries on the outer margin of the EU. Ahmad himself, who socialized almost entirely within Arab trading and social networks, was talked of by other Syrian traders as 'Russian', thus highlighting the significance of passport mobility as a mode of social categorization. Another recurring concern through which they experienced social marginality was schooling. In Yiwu, even relatively wealthy foreign traders were thrown back on informal schooling options for their children. These were precarious, as the police could close the institutions any day, and did not mesh with a wider educational infrastructure, since the Chinese universities did not recognize the qualifications they provided. This was another factor in directing their gazes elsewhere, to other possible geographies, and explains why Ahmad and others made substantial financial investments in Moscow and Istanbul. Selim, the owner of another Yemeni trader office in the same office block, held similar views of Turkey as a congenial place in which to live in the future, and had started to incorporate Istanbul into some of his supply networks exporting Turkish cosmetics to the Saudi market using packaging sent from Yiwu.

A tale of three cities
These traders were brokers not only in the sense of connecting markets, but also insofar as they kept different geographical contexts in play, through property investment, travel, commercial supply networks, and passporting strategies. Istanbul became particularly attractive during the summer of 2018 because its plummeting currency made property investments more viable. Other factors, however, such as easier access to long-term residence rights, also featured in these decisions. This led some, such as Talal and Uday, to purchase property in a country of which they had no first-hand knowledge, but which had become salient in their imaginations because of favourable immigration and trade policies. These in turn reflected the attempts by Turkish political elites to position the country as a pole of an Islamic Eurasia -a strategy of the Justice and Development Party, which came to power in 2002 (Aras and Fidan 2009;Bilgin and Bilgiç 2011;Lin 2016). In the summer of 2018, Ahmad noted that Turkey was a pole that was drawing Muslim traders from Saudi Arabia and beyond, especially those originally hailing from 'unstable', 'problem-ridden' countries: 'four million have gone to Istanbul -Palestinians, Egyptians -especially the Muslim Brotherhood ones after the problems in Egypt, Afghans, Yemenis, Syrians. It has become a pole (istiqtab) that draws people. Many are now buying property therepeople who have problems in their own countries -Afghan, Yemen, Syria.' The triad of cities in which Ahmad and Uday reproduced their neighbourly relations -Moscow, Yiwu and Istanbul -were neither nodes in a static geography, nor governed by the teleology of a re-emerging Silk Road. Rather, by keeping a foot in different Eurasian poles, they mediated dynamic relations of competition and cooperation between contrasting state-led visions of a connected Eurasia. They operated as brokers between successive Eurasian formations -a Soviet and later Russian attempt to project power into central Asia and the Middle East through cultural, commercial and military connections; a Sino-centric vision to build a connected Eurasia around its 'one belt one road' infrastructure understood as conduits of a new 'imaginative geography' (see various works by Edward Said); and a Turkish AKP vision to position Istanbul and Ankara as poles of an Islamically connected Eurasia. These Eurasian formations overlapped and contested each other: the AKP project was viewed with suspicion by Beijing not least because of the influence it might permit them to wield among Xinjiang separatists; in the summer of 2018, Turkey pursued a policy of partial rapprochement with Russia, while also drawing trading capital and personnel away from Saudi Arabia amid intense political rivalry that also played out through the notorious Khashoqqi affair.
Viewed from below from Yiwu, too, the ongoing formation of Eurasian geographies was a contested and political process (Green 2014;Hann 2016). It is the contention of this article that Yiwu's Arab traders mediated these relations of competition and competition between different Eurasian geographies and political imaginations. As they moved from one hub to another through the course of their lives, they both connected them through the economic networks they built and sustained, and made differential imaginative and financial investments between them, as they sought to reckon with the marginality and impermanence that brokerage imprinted on their lives. While accounts of a new Silk Road connecting China and the Middle East deploy notions of ethnic enclaves and networks of solidarity to understand the durability of Arab transregional trading networks in Yiwu, I have argued that a different approach to networks, of positionality and brokerage, makes better sense of the transregional geographies they fashion and transregional contexts they seek to keep in play. This in turn implies a different approach to understanding the durability of Yiwu as a transregional commercial hub. From a network governance approach, the durability of Yiwu's role as an anchor of Arab transregional trading networks depends on the continuation of the city's policy of cultural accommodation towards a Muslim Middle East, and on the degree of political and economic turbulence within the Middle East region, where turbulence is assumed to lessen the demand for Chinese goods (Belguidoum and Pliez 2016). While these are important factors, Turkey has also been able to compete for these actors' capital and commercial investments through its offer of citizenship rights and through the lower cost of living in Istanbul compared with Yiwu. At the same time, some forms of political and economic turbulence within the Middle East region -namely antagonism between the Turkish and Saudi regimeshave served to decrease Istanbul's attraction as an anchor for Arab trading networks looking to supply the major Saudi market, and thus paradoxically benefit the Sinocentric Eurasian project and the role of Yiwu as a transregional trading hub.

Conclusion
By focusing on Syrian and Yemeni traders in Yiwu, in this article I have sought to theorize the process of network formation underpinning the 'globalization from below' that has driven the city's economic development. Rather than a process in which Arab traders 'go east' along a Silk Road, I have offered a different history and theoretical conceptualization of the formation of Arab trading networks, which has been critical to the emergence of Yiwu as a global trading hub. I have highlighted the significance of socialist and post-socialist relations of geopolitical patronage in drawing Syrian and Yemeni individuals into circuits of transregional mobility and commerce. These geopolitical projects and rivalries generated the concrete conditions and institutions -such as higher education scholarships, transregional student mobilities, the construction of student dorms-cum-trading office blocks, and the building of market complexes -that facilitated the emergence of transregional trading networks. Successive geopolitical visions of Eurasia have furnished the actual political, historical and spatial contexts in which the foundations of many of the Arab trading networks currently active in Yiwu were laid. In their geographical scope, these different iterations of Eurasia have not simply run east to west, but also north to south, for example stretching south from Moscow to Aleppo and Taizz.
Viewed from the perspective of these resident Arab traders, Yiwu's development does not constitute a break from earlier attempts to fashion webs of interregional connectivity in Eurasia, but is rather layered into them. At the same time, Yiwu's resident Arab traders also mediate shifting relations of competition and cooperation between these various iterations of Eurasia, through which they seek to negotiate and benefit from the rising power of both a Sino-centric and a Turkish-Islamic vision of this expansive transregional space. Their transregional activities are therefore not simply the modern revival of a previously well-established east-west corridor but are in a dynamic relationship with a series of distinct but overlapping Eurasian geographies. This analysis implies a different understanding of their economic networks from the one that emphasizes their embeddedness in a transregional Islamic culture. Rather than understanding networks as providing social cohesion and possibilities of trust and solidarity, I have analysed them as structures that offer opportunities for brokerage and strategic manoeuvring. This reflects the concerns and strategies that occupied the owners of Arab trading offices beyond the immediate task of forging a livelihood. As relatively wealthy but socio-politically marginalized actors in Yiwu, they understood themselves as embodying a state of impermanence and needing to keep a foot in different poles of Eurasia. Acting as brokers between different Eurasian contexts, they kept different futures in play for themselves and their families.
This approach questions models that depict trading networks as 'middlemen minorities' for powerful political elites (Bonacich 1973). While Syria and Yemen were marginal to both the USSR and the recent rise of Turkey in Eurasian politics, their traders have been able to capitalize on this marginality by brokering between different Eurasian formations.
Another implication of this is that the durability of Yiwu as a transregional trading city must be seen in part as depending on its competitiveness vis-à-vis rival Eurasian commercial hubs. The metaphor of the Silk Road and associated trope of Muslim diplomacy conveys important factors in Yiwu's economic development, such as the efforts of city authorities to foster a welcoming Islamic environment for short-term merchant visitors to the city. However,it can also imply a teleological conception of Yiwu's geography as the revival of an earlier, essentially linear, transregional axis, and obscure the competitive dimension of recent and ongoing Eurasian politics. Rather than following a Silk Road, a conception of Yiwu's Arab traders as brokering dynamic relations between competing Eurasian formations, which are always in process, better recognizes the contingency of the future -both in relation to individual trajectories, and in relation to the city as a transregional trading hub.