Introduction
Scholarship regarding Afro-Asian entanglements suffers from a presentist bias, overwhelmingly focused on the recent influx of Chinese state capital, infrastructure projects, resource extraction, and a “Global China” (Benton & Liu, 2004; Cheru & Obi, 2010; Lee, 2017; Mohan & Lampert, 2013). It frequently overlooks the deep historical stratifications of Asian presence in Africa and fails to account for the diverse modalities of migration that exist outside the logic of state-to-state diplomacy (Mathews, 2011). To understand the complex Afro-Asian relations, we look beyond the political spectacle and turn our attention to the divergent strategies of those communities that have historically acted as the connective tissue between worlds: the middlemen.
This article proposes a comparative anthropological examination of Asian mobilities within the Lusophone (Portuguese-speaking) African sphere. We juxtapose the trajectories of the Nizari Ismaili Muslims in Mozambique, which may be regarded as an “old diaspora” deeply embedded in the Indian Ocean littoral, with the New Chinese migrants, which could be seen as the “new frontiers” of entrepreneurship in Cape Verde. At first glance, these groups appear to inhabit incommensurable worlds. The Ismailis represent a community of historical depth and high institutionalization (Ho, 2006), while the Chinese shopkeepers in Cape Verde represent a recent, rapid, and often transient migration (Haugen & Carling, 2005). Yet, despite these temporal and spatial divergences, they share a critical structural position: they have functioned, in different epochs and under different logics (Zenner, 1991), as “middlemen minorities” (Bonacich, 1973). They occupy the evolving socioeconomic interstices of the Lusophone world, mediating between the local population, the colonial or post-colonial state, and global circuits of capital.
Analyzing the historical architecture of the middleman figure, beginning in colonial Mozambique and following migratory circulations in the post-Colonial period through the former Lusophone space, reveals divergent forms of Asian presence. Nizari Ismailis, who had been tradesmen and indentured laborers in East Africa since the 19th century, were mobilized into the colonial economy as cantineiros (rural shopkeepers) (Leite & Khouri, 2011). Operating deep in the hinterlands, they functioned as the essential capillary vessels of the Portuguese empire, exchanging European manufactured goods for African agricultural produce (Machado, 2014). Crucially, these tradesmen, operating between the white colonial elite and the black majority, were never isolated but included within Ismaili infrastructure and antecedents of the contemporary institutional space, the Jamatkhana (prayer space), the Jamat (community) and the guidance of the Imam (Daftary, 1998). This institutional coherence helped them negotiate collective protections, pool capital, and eventually transition from petty traders to urban industrialists and transnational elites. The Ismaili “roots” in Africa were thus secured through a vertical allegiance to their spiritual leader and a horizontal solidarity within the community structure (Penvenne, 2015).
Today, the Ismaili community numbers an estimated 15 million worldwide and constitutes a transnational diaspora. Driven by a shared identity and the global circulation of finance, industry, and international development, this community maintains a spiritual, spatial, and communal orientation toward the Lisbon-based Ismaili Imamat. The Imamat’s establishment, administration, and leadership followed the resettlement of the community from Mozambique, many of whom were expatriated to Lisbon in the years preceding Mozambique’s nationalization. The development of Ismaili infrastructure and the community’s privileged socio-economic position in Portugal and Mozambique are the result of coordinated political, institutional, and family practices developed over centuries in the diaspora.
The “Old” Chinese presence in Southern Africa (including Mozambique and South Africa) was forged in the crucible of the indentured labor system, historically pejorated as the “coolie trade” (Harris, 2010; Yap & Man, 1996). Many early Chinese arrivals, especially southern Chinese from Guangdong and Fujian, were transposed not as entrepreneurs but as contract laborers for colonial infrastructure projects or agriculture, occupying a subaltern status (Rinaldi, 2011). Over decades, these Chinese laborers and their descendants transitioned from the “coolie” barracks to the shopkeepers, utilizing the Huiguan (native place associations) to pool meager savings and exit the labor market as entrepreneurs (Snow, 1988).
The “New Chinese migrants” are not simply a continuation of earlier Chinese presences in Lusophone Africa. They constitute a distinct formation born of China’s post-1978 Reform and Opening-Up policies, drawing predominantly from Fujian and Zhejiang provinces and organized through tight kinship-based and township-based (tongxiang) networks that replicate a modular retail template across diverse national contexts with minimal institutional footprint. Their presence in the PALOP states intensified after 2000, catalyzed by the Forum on China-Africa Cooperation (FOCAC). What defines them is not recency but a specific structural logic: fast capital rotation, deliberate opacity toward state institutions, and social insularity (Haugen, 2018), features that set them apart categorically from both the colonial-era Chinese merchant communities embedded in Portuguese racial-commercial hierarchies and the institutionalized transnational elite represented by the Ismaili network.
Our comparison traces the historical mutation of a structural role at different phases of the same formation, as in Marcus’s (1995) multi-sited ethnography tracing social formations across time and space. In this paper, we specifically explore the use of space as a main resource to increase social and economic mobility, critically analyzing how spaces of economy and business concretize the governing structure that informs the evolving space of the middleman in urban sites. The materiality expresses the distinctive relationships to the broader communities and how the role of the middleman is mediated through distinct forms of space. We circumscribe the economic middleman figure in Mozambique and Cape Verde as marked by juridical and social dimensions of direct and indirect rule, leading to monopolies over segments of their respective economies resulting from European Colonialism (Bonacich, 1973; de Pina Cabral, 2010; Vale de Almeida, 2004) and its legacies (Mamdani, 1996). These middlemen occupy intermediary positions imagined by colonial administrators as an effective means of mediating capital, goods, and resources.
Theorizing the Middleman in the Lusophone Ecumene
The institutionalized discourse of Lusofonia codified through the CPLP and circulated in Portuguese cultural diplomacy since the 1990s has been subject to incisive critique (Cahen, 2010; Furtado & Sansone, 2014; Mata, 2008; Ribeiro, 2004). We deploy “Lusophone Africa” as a critical structural lens. In this respect, our analytical move is closer to Furtado’s and Cahen’s critical projects than to the conception of Lusofonia they deconstruct, where they unmask Lusofonia as ideology; we use the Lusophone formation as a historical and juridical-structural concept to trace the colonial genealogy of the middleman role across post-imperial space. Lusotopia, as a deliberately neutralized, spatially defined alternative to the culturally charged Lusofonia (de Pina Cabral, 2010), points precisely toward the kind of analytical operation we are attempting, and we adopt its spirit here.
Theoretically, “Lusophone Africa” in this paper designates the set of African territories in which the Portuguese colonial administration constituted a particular socio-legal architecture of governance, and crucially for our argument (Santos, 2002), the instrumentalization of Asian minority traders as commercial and fiscal intermediaries between the colonial state and African populations (Peberdy, 2000). This architecture was not homogeneous across the PALOP states, but it was consistent in the systematic production of a racially stratified economic hierarchy in which an intermediary minority class occupied the middle position between European administrative capital and African labor. It is this shared institutional grammar (Cooper, 2005) that constitutes “Lusophone Africa” as a meaningful analytical category, not shared language, not shared culture, and emphatically not the imagined civilizational community of CPLP discourse.
The comparative analysis of colonial formations does not require homogeneity across sites (Stoler & Cooper, 1997); it requires identifying the shared structural rules that organize differently situated colonial encounters. Cape Verde and Mozambique were selected because they represent, within the Lusophone ecumene, the clearest instantiations of the grounded middleman dynamic in its two polar forms (the institutionalized transnational elite and the atomized petty trader) without the distorting effects of resource-cursed political economies. In contrast, the Chinese economic presence in Angola is dominated by state-backed construction and resource extraction enterprises operating through government-to-government bilateral agreements, deviating from the middleman grammar we theorize. Moreover, two cases in our article were theoretically strategic cases chosen for their capacity to render a structural dynamic visible that no single site (Burawoy, 1998), and no representative sample (Gluckman, 1961), could illuminate on its own.
“Lusophone Africa,” as we use it, is a differentiated social field in Bourdieu’s (1993) sense: a space defined by a common set of structural rules within which different positions produce radically different strategies, institutional forms, and developmental outcomes. Their differential positioning within a common structural field helps explain why the middleman role takes such radically distinct institutional forms in each case (the rooted transnational broker in one, the invisible logistical node in the other) and that this divergence itself tells us the range and logic of the social space of the Lusophone formation.
Georg Simmel’s figure of the stranger, “one who comes today and stays tomorrow” (1950), provided the conceptual foundation for Bonacich’s (1973) middleman minority theory, which describes groups occupying a precarious position between elite dominance and subordinate populations. Yet to apply this framework to Africa without qualification risks importing a European lens onto societies that had long developed their own coherent logics for incorporating outsiders. Skinner (1963) shows that well before colonial contact, African communities had institutionalized the stranger’s role, particularly in trade. Shack and Skinner (1979) extended this comparison, demonstrating that the stranger was not an anomaly in African social life but a structurally necessary figure, absorbed through recognized categories that simultaneously enabled and constrained participation. This matters for our analysis because when Asian migrants entered African economic life, they were not simply inserted into a colonial racial hierarchy. They were also being read through pre-existing African social grammars that already had a place, however ambivalent, for the useful outsider.
The collapse of the Portuguese empire did not dissolve the social geography it had produced (Vale de Almeida, 2004). Within this order, racialized minorities who had been recognized by the colonial state as sufficiently assimilated retained access to privileged social space, functioning as a managerial class positioned between raw resources and finance (Yap & Man, 1996). Former middlemen drew on transnational connections to maintain their footing through the civil wars (Bastos, 2005, Trovão & Batoreu, 2013) that followed independence.
When the empire dissolved, newly independent states inherited territory but lacked the institutional, logistical, and financial infrastructure needed to govern their economies (Burawoy & Verdery, 1999). Neoliberal restructuring deepened this post-colonial vacuum (Bayart et al., 1999), and it was into these structural gaps, in investment capital, financial services, and consumer goods distribution, that the former middleman class expanded. The Nizari Ismailis leveraged transnational institutions such as the AKDN to address high-level development and finance needs, evolving into what we might call Global Brokers (Gibbon & Ponte, 2005). New Chinese migrants, operating through fast capital and integrated supply chains, filled retail and logistics gaps at the street level, functioning as Logistical Nodes (Lee, 2017). Their ascent reflects not inherent cultural traits but a structural response to gaps left by states still consolidating their capacity to govern economic space.
This post-colonial middleman position extends beyond the Africa-Asia axis. The Ismaili network’s institutional reach across the Lusophone world positions it as a potential actor in South-South development flows connecting Africa to Brazil (Miyamoto, 2009). The logistics-driven networks of New Chinese migrants, concentrated in places like Cape Verde, could similarly use Luso-African commercial infrastructure as a springboard into South American markets (Haugen, 2018). Even where these connections remain unrealized, they constitute Imagined Futures in which current middleman activity in Africa forms the logistical and financial basis for a broader triangular engagement among Africa, Asia, and Latin America. The following cases, therefore, analyze how relations among a diversity of global elites, not just the Portuguese State, expanded the role and position of the middleman to power broker in Lusophone Africa.
Methodological Reflection
This article draws on a comparative synthesis of two distinct but intersecting anthropological doctoral projects conducted by Wang (8 months) and Khan (12 months) based on long-term fieldwork in Lisbon, Portugal, and Praia, Cape Verde. Adopting a multi-sited ethnographic framework (Marcus, 1995), our research design was strategically situated across Praia (Cape Verde) and Lisbon (Portugal), with retrospective archival and oral historical linkages to Mozambique. While our fieldwork on Chinese migrants engages with the active frontier of intermediation in Praia, our research on the Ismailis in Lisbon engages with the archived and reconstructed memory of intermediation in Mozambique. This multi-sited approach allows us to trace the full lifecycle of the middleman trajectory within the Lusophone sphere, and to examine the middleman not as a static category, but as a fluid role shaped by the distinct geographies of the Atlantic and Indian Oceans. Moreover, we analyze how these distinct trajectories converge in the “Lusophone ecumene.” This comparative analysis reveals how the historical links of the colonial past (represented by the Ismaili and Old Chinese experience) actively shape, and are reshaped by, the present challenges of globalization (represented by the New Chinese migrants), weaving a complex, multi-layered tapestry of Asian agency in Global Africa.
The choice of Lisbon, rather than Mozambique, as the primary fieldwork site retrospectively traces collective memory, institutional establishment, and identity narratives that are today actively organized through Imamat and the Institute of Ismaili, where Lisbon functions as a primary node in a global Ismaili network. Europe is the site of the Imamat’s headquarters and center of knowledge production, the Ismaili Institute. The Lisbon Jamat is disproportionately shaped by the Mozambican experience and has reconstituted its social world following the orchestrated migration preceding April 25th, 1974. Lisbon is not a proxy for Mozambique but the site where the object of analysis, the memory of intermediation and the institutional architecture sustaining it, is located. This strand of research draws on participant observation, semi-structured interviews tracing individual and collective memories of migration and life in Mozambique, analysis of the Ismaili Institute’s oral history materials, and examination of the Portuguese legal frameworks structuring the community’s national integration, complemented by walking ethnography focused on the flows of capital and political legitimacy that continue to connect Lisbon to Mozambique through global Ismaili institutions.
The New Chinese migrants in Cape Verde, by contrast, are characterized by precisely the opposite social logic: opacity, informality, kinship-based insularity, and deliberate avoidance of institutional visibility. No oral history project preserves their trajectories. No headquarters organizes their narratives. Accessing this field required sustained ethnographic presence, in the shops, the back rooms, and the daily rhythms of retail life in Praia. Wang’s fieldwork, therefore, relied primarily on participant observation and semi-structured interviews.
The methodological difference between our two streams thus mirrors the structural variance between the two communities, which is itself a core element of our argument. We analyze the strategies of the highly institutionalized, transnational financial and managerial elite (as exemplified by the Aga Khan Development Network and the Azinor Group) against those of micro-level Chinese retail shops, effectively contrasting two distinct modalities along the spectrum of Asian transnational economic agency.
Mozambican Ismailis in Lisbon
Khan joined an adult education course on Islamic Architecture held at the Lisbon Ismaili Centre as part of the ITREB (Tariqa Education Board) initiative to educate the Ismaili community about historic mosque architecture and its relationship to the work of the Aga Khan Award for Architecture in October 2024. It was the second part of a four-part hybrid online and in-person educational series on the History of Mosque architecture, beginning with the mosque of earth and sky, the Prophet Mohammad used to pray in the 7th century, through historic and contemporary evolutions, and how it related to the Aga Khan Award and its development in the 21st Century. Most participants were professionally employed or trained in a field related to the built environment, ranging from architects, construction workers, urban planners, real estate executives, and a few others simply interested in Mosque architecture.
Raj Isar, a seasoned development scholar and Education Director for the Aga Khan Trust for Culture, emerged as an obvious leader within the group. He traveled in and out of Lisbon from his home in Paris, where he was a university lecturer. Khan saw many members of the Jamat, who like Raj, circulated between Lisbon and work, educational, or family commitments in other countries. English served as the lingua franca amongst the many languages spoken across at least five nationalities who had migrated to Lisbon in recent years.
The students self-selectedly divided themselves into several groups in the room marked “Sala”, the institutional gathering at one table. The groups were boisterous and playfully competitive. Hussein, one of three Husseins in the room, piped up and launched into debate as had become his routine. “Isn’t the Ismaili Centre a mosque? We have a prayer hall, a courtyard, and a room for education. “In his British accent and the authority that always seemed to be fixed to it, he voiced a thought we’d probably all realized by this point. The Ismaili Centre functioned as and was spatially organized like other mosques. Another student from his table chimed in to compare the name “Ismaili Centre” with the name “mosque”, as in the Central Mosque of Lisbon, the most visible one in the city.
Although Khan and other students were all seated around conference tables with a projector and a lecturer, they would soon be without shoes in the prayer hall, kneeling on the carpet together, surrendering to the volume of a monumentally tall room and beneath a canopy of stone.
“If it talks like a duck, and walks like a duck… it’s almost a duck,” quipped Farid, the funny one, the group laughed too hard, responding to the ambiguity. Mina, a Portuguese architect with Mozambican roots, shut the debate down that day by asking, “If someone else’s name is on your house, is it your house?” Can something be called your own if it is called something else by others, and if it calls itself a name given by another?
She led the class on the guided tour of the Ismaili Centre, a service she provided regularly for the Centre, where she explained that the stones at the entrance illustrated the 99 names of Allah and were made from a stone indigenous to the Iberian Peninsula. They were sourced from a stonemason in Sintra, materially integrating Ismaili institutional space into the territorial logics of its Lisbon locale.
Both Mina and Hussein have worked professionally with the Sana Hotel Group at various points, which is owned by the multi-national Azinor Group, whose founder is also Ismaili. The Sana Hotel Group, ranked among the country’s top hotels, is also a preferred hotel for large-scale events held at the Ismaili Centre, events like the Diamond Jubilee, which gathered over 40,000 Ismaili in 2018, or the regular diplomatic visits hosted at the Imamat. In this example, commercial space simultaneously supports faith-based activities, business operations, facilitating high-level social networks.
The question that provoked debate, “Is the Ismaili Centre a mosque?”, cuts to an issue of space and place and an ambiguity and adaptability that is reflective of Islam (Ahmed, 2016; Asad, 2003; Moumtaz, 2021). What makes any space Islamic? What shapes the abstractions of land and architecture into places recognized for the system of values embodied there? Specifically, of Ismaili spaces, which are defined and redefined through an ever-evolving context of farmans (spiritual guidance), institutional language, titles, and their corresponding acronyms. The people in the room, most of whom possessed professional or technical expertise in the development of urban space, were also debating this ambiguity. The Post-Colonial European context, which naturalizes a carefully managed secular urban space, was brought into question, given the historical context introduced by the class. We learned that mosque architecture had evolved through competing global frameworks that discipline space, yet it can only be understood within its own trajectory. As presented in the class, it became obvious that mosques were defined only by the communities that practiced their faith and acted out their lives in relationship to them (Poor, 2014). There could be no definitive prescription. Is the Ismaili Centre a mosque? Certainly, no less than the Central Mosque of Lisbon, and no more than the sidewalks in the Mouraria, where Bengali communities prayed on the sidewalk due to a lack of space in their mosques. Other spaces and places animated by the faithful lives of Muslims would continue to evolve in response to their lives.
The former colonial reality, which conditioned their role as economic middlemen, fulfilling an economic role to the State while increasing their socio-economic mobility, has expanded in the post-Colonial. We argue that the ambiguity in the Ismaili Centre’s name, and the subsequent deployment of that provocation through online education, is a reality possible in this specific post-Colonial temporality, and indicative of the community’s increased capacity to define itself through the infrastructure of the Imamat and its role in the world. Ultimately, what defined the Ismaili Centre was the community which gave it meaning with their presence and to whom it belonged.
Enabled by the speed and reach of the digital reality, the Ismaili institutional space corporately facilitates social, ethical, spiritual, and economic life, with increased capacity to manage the plurality of the diaspora through education and narrative. It collectively challenges the secular/religious binary and the definition of the subjectivity it inspires, one operating through a supra-state geography of flexible citizenship (Ong, 1999) and undefined by a secular economy. The hybrid online/in-person model of the class sparked similar reflection across a global body of students connected through the online portal, launched in 2014 with over 12,000 learners. (Khan’s Fieldnotes, July 8, 2024)
One company that exemplifies the interconnectedness between materiality of urban space and capital, and the immateriality of relations and ethics is the Azinor Group, based in Lisbon with holdings in Mozambique and other African countries, including Angola, South Africa, and Tanzania. A. began as a hotel company that accommodated migrants coming from Africa in the early 1980s. The owner comes from an Ismaili family in Mozambique and has connections in Angola due to the historical presence. The company won a public bid during the civil war in Angola and greatly expanded its business across Lusophone Africa (Trovão & Batoréu, 2013). When Mozambique’s political climate stabilized following its own ten-year civil war, Azinor Group expanded and reconstituted its networks there (Bastos, 2005).
The Azinor Group is not formally linked to the AKDN, but the company has complex ties to the Imamate. Nazir Din, Azinor Group’s founder, along with his two relatives, is the majority stakeholder in Azinor Group. He also held major institutional roles within the Ismaili organization, serving as President from 1999 to 2005. Within the urbanism of Maputo, AKDN operates numerous for-profit businesses under the Aga Khan Economic Development Network, including the Serena Hotel Chain in Maputo. With a base in Portugal, expansion to Mozambique and to thirty-seven other countries, Ismaili businesses like the Azinor Group capitalized on the maintenance of their social position and material wealth that the middleman position enabled.
The Chinese “Routes” in Cape Verde
Unlike the Nizari Ismaili from Mozambique to Portugal, whose presence was woven into the colonial fabric over geography, generations, and economic settings, the Chinese arrival in Cape Verde started from the late 1980s. It is visible, rapid, and physically transforms the urban landscape of the capital, Praia (Varela, 2020). This section explores the spatial and economic reality of this “new frontier” through the lens of the Chinese Shop based on Wang’s 8 months fieldwork and three years living experiences in the capital city of Cape Verde, Praia.
In the neighborhoods of Praia, which is from the plateau of the historic center to the sprawling unplanned settlements of Vila Nova, the Chinese Shop has become the primary unit of retail. They are distinct, standardized spaces inserted into the city. During Wang’s fieldwork in 2024, he mapped the distribution of these shops. The early findings show the role of Chinese shops acting as urban anchors, where there is a Chinese shop, there is a flow of people, but not necessarily a visible flow of social interaction. The architecture of the shop dictates a specific relationship between the migrant and the host.
I entered “Loja Felicidade” (Happiness Shop) in the Vila Nova district. The transition is abrupt. Outside, the street is loud, dusty, and bright with the Atlantic sun. Inside, it is cool, dimly lit by fluorescent strips, and smells of synthetic rubber and packing tape. The space is maximized for storage, not comfort. Floor-to-ceiling shelves create a maze of goods: plastic basins, clothes with the Cape Verdean national flag from Zhejiang, synthetic textiles from Guangdong, cheap electronics from Hebei. There is no empty space, but there is always room for a large real-time monitoring screen to supervise the shop. It doesn’t feel like a local shop, but more like a container that has been unpacked directly onto the street. At the back, on a raised platform behind a high counter protected by transparent plastic bars, sits the shopkeeper. He is watching TikTok on his phone… (Wang’s Fieldnotes, Praia, December 13, 2024)
The sensory shift described in Wang’s fieldnote, which is from the “dusty, bright” street to the “cool, synthetic” interior, marks the Loja Felicidade not as a space of social dwelling, but as a logistical enclave. This stands in stark contrast to the colonial-era Ismaili cantina, functioning as a porous contact zone where credit and conversation bound the trader to the village. The “raised platform” and “transparent plastic bars” act as a physical manifestation of the social barrier (Venancio, 2024), placing the migrant in a position of surveillance rather than engagement.
Furthermore, the shopkeeper’s absorption in TikTok shows his absent presence. While his body is managing the cash flow in Praia, his social consciousness remains tethered to the digital ecosystem of China (Wei et al., 2007). This reinforces Haugen’s (2018) argument that for many Chinese petty capitalists in Africa, the host country is viewed merely as a speculative site for rapid accumulation, not a home to be cultivated (Mathews, 2011). Unlike the Ismailis, who built community halls to stay, the Chinese migrant builds a Chinese shop ready to be packed up, embodying a strategy of pure, transient extraction.
Mr. Lin, 34 years old, is originally from a rural county beside the city of Wenzhou, Zhejiang province. Our conversations were fragmented, conducted in the interstices of commerce, often interrupted by the slap of sandals on the tiled floor or the beep of the barcode scanner.
“I am not here to drink coffee,” Lin told me in Mandarin, his eyes scanning the CCTV monitor. “I borrowed 200,000 RMB (approx.25,000 euros) from my uncle and two neighbors to buy this shop and pay the lease. Every minute the door is open, the meter is running. If I go to the beach, I lose money…” Lin’s narrative shows the internal logic that necessitates the external defensive architecture we observed. He operates under the ethos of Chi Ku (“Eating Bitterness”)—a willingness to endure extreme hardship in the present for a deferred future prosperity (Rodrigues, 2012). He is a solitary node. His obligation is not vertical (to an Imam), but horizontal (to the kin network in Wenzhou that holds his debt).
Meanwhile, there is no “Bom dia” (Good morning), no inquiry about health or family, which are rituals that are normal to Cape Verdean social life. The Cape Verdean woman places the bag on the counter. Lin points to the calculator: “800” (Escudos). The woman slides a 1000 Escudo note through the gap in the bars. Lin slides 200 changes and the bag back. The interaction lasts 14 seconds. Money, here, is the sole medium of translation. This silence is potent. It shows a shift in the nature of the middleman. In the colonial era, the Ismaili cantineiro was a broker of culture as well as goods; credit created a social bond, however unequal. In the Chinese shop, the bond is snapped. Money here does not circulate locally to build community schools or mosques; it is extracted, or it goes back to China.
This operation is underpinned by guanxi (relationships) that connect the shopkeeper to kin and suppliers in Zhejiang. This network acts as a crucial credit intermediary (Park, 2009). The cash earned is instantly repatriated through informal channels, which include couriers or mobile transfers to service debt and replenish stock (Haugen & Carling, 2005). This allows capital to effectively “fly away” without circulating through local Cape Verdean banks, maintaining intentional social detachment. Furthermore, this practice creates modular insulation: the shop is visible economically, but the community remains politically and socially invisible.
Maria, a 45-year-old resident of Fazenda who shops at Chinese shops daily. “We need them,” she told me in Cape Verdean Creole, gesturing to the rows of affordable goods that the products imported from Portugal and Spain cannot match. “But they are like ghosts. They don’t show up on the streets, they don’t get married, or they don’t die. They take the money, and the money flies away. It doesn’t stay in Praia…”. The new Chinese migrants in Praia practice a form of social suspension. They arrive as healthy, working-age adults, even younger, and when they become ill, elderly, or wealthy enough, they vanish back to China. The local population rarely witnesses the vulnerability of the Chinese body or the marriage of Chinese couples, only their labor power.
At the same time, Maria’s observation that “the money flies away” captures the essence of this new Afro-Asian entanglement. Cash earned in the shop is rarely deposited in local Cape Verdean banks, where it might circulate as credit for the local economy. Instead, through informal networks and digital remittance systems, it is instantly repatriated to China to service the goods for their shop. For the Nizari Ismailis, Lisbon and Mozambique became sites of heritage, where heritage and identity reemerge from the soil, or the stone, as seen in the Sintra marble used in the architecture of the Ismaili. For migrants like Mr. Lin, Praia is not a site of heritage, but a site of transit.
Comparative Reflection
Our dual ethnography suggests that both Nizari Ismailis and New Chinese Migrants occupy the structural position of the middleman in divergent ways. Khan analyzes the historic evolution into a “merchant/managerial class,” leverages the strength of their institutional networks to expand their urban presence from the colonial era into the present, brokering access to high-value resources and expanded networks of global capital (Leite & Khouri, 2011).
In comparison, the Chinese trajectory in Cape Verde presents a model of intermediation that is streamlined, instrumental, and insulated. Lacking the deep historical sedimentation of the Ismailis, the new Chinese migrants operate not from a position of managerial oversight or established elite status, but from a position of intense market competition and high mobility (Haugen & Carling, 2005). However, when analyzed through the role of middleman, these spatio-temporal divergences reveal not a lack of comparability, but rather two distinct responses to the same structural challenge: how to mediate between the global and the local in the Lusophone ecumene.
The following section attempts to synthesize the divergence in their strategies of spatial, economic logic, and how they reimagine the role of being a middleman. Khan describes the resilience of the Ismaili presence, anchored by what the AKDN calls a “supranational entity”, an umbrella organization under the authority of the Imamate—providing spiritual guidance, high-level geopolitical insight, and transnational legitimacy, and supported by Jamat (Daftary, 1998). As seen in the case of the Azinor Group, business interests are often inextricably woven with communal obligations and philanthropic structures (Bastos 2005, Trovão & Batoreu, 2013). This integration means that an Ismaili merchant is never just a trader; they are a member of a civic body (Mamdani, 1996; Mamdani, 2020). Their social capital is fungible: success in business is a community ethic facilitating a cyclical well of communal resources and assets.
In contrast, as Wang observed, a tight, trans-oceanic axis connects the shopkeeper in Praia directly to creditors and suppliers in Zhejiang (Wei et al., 2007). This network is highly efficient for the rapid movement of capital and goods, but socially thin. For the Chinese migrant, atomization is not a failure of culture, but a strategic adaptation to a market-driven economic model (Haugen, 2018). The resilience observed in the Ismaili community allows for long-term planning and risk absorption (Lee, 2017). The New Chinese migrant’s network is designed for logistical agility, moving goods fast and restocking quickly rather than for social integration (Mathews, 2011).
Without the managerial status or the political protection enjoyed by the Ismailis, Chinese migrants in Cape Verde tend to adopt a defensive posture. The action taking place inside the shop, an unengaged shopkeeper who refuses the minor social engagement of taking a coffee in proximity to the locals, conditions the space for sterile financial transactions. The face of the colonial outpost, the Ismailis, reproduced social hierarchies through a combination of coercion and care (Mamdani, 1996). The urban warehouses stored Portugal-bound exports supplied by cantineiros located in inaccessible hinterland (Leite, 2011, Penvenne, 2015). cantineiros additionally provided financing to farmers and maintained production through a multitude of Indian and Ismaili-run shops, ensuring colonial quotas (O’Laughlin, 2002) and keeping costs and operations efficient.
In the 1990s, when the economy stabilized after the Civil war in Angola, large-scale businesses such as the Azinor Hotel Group and AKDN institutions expanded their presence primarily in the urban centers of the former colonies, operating in tourism, media, mining, and the latter creating social programs and also for-profit businesses. A network of philanthropy and business that has been re-established through colonial networks and expanded on through thick networks. The Chinese economic logic in Cape Verde is defined by fast and transnational capital, and Chi ku (Rodrigues, 2012). As Mr. Lin shared, the goal is not to build a legacy or permanently settle down in Praia, but to make a living to support the family back in China and pay back the debt in Zhejiang.
However, both logics address the same post-colonial void. Following the withdrawal of the Portuguese state and the limitations of local state capacity, there remains a gap between international supply and local demand and its reverse, between local resources and global finance (Peberdy, 2000). The AKDN aids the local Ismaili community in bridging this gap by leveraging partnerships with global development and banking networks to infuse finance into small and large business ventures. The Chinese bridge this gap through price and logistics, which is their ability to access the production factory back in Zhejiang, deliver goods cheaper, faster, of selling and renewing the stocks than anyone else acts as the guarantor. We could say one offers development, the other offers affordability.
The Ismailis’ global broker status enables them to strategically navigate high-level sectors like finance, large-scale hospitality, and infrastructure, and these may crowd out and suppress the fragile emergence of a competing managerial and financial elite (Rinaldi, 2011). At the same time, the New Chinese migrants provide unparalleled affordability through direct transnational supply chains that systematically prohibit the organic growth of local Cape Verdean petty traders, small retailers, and nascent manufacturing. In essence, the persistent middleman role, whether through institutionalized development or atomic affordability, represents a structural surrender of critical economic space. The African state cedes control over both the high-value macro-governance of capital flow and the essential micro-logistics of mass consumption, cementing a dependency on externally organized, ethnically insulated networks.
Furthermore, the Chinese strategy in Cape Verde, as evidenced by the Loja Felicidade and most of the Chinese shops in Praia, is one of modular insulation. The materials are imported, temporary, and generic. This is an architecture more like the “station,” not necessarily the home (Mathews, 2011). This material reality serves as a specific function, which could be seen as strategic flexibility. At the same time, they are highly visible as economic actors, which locals know where the shop is but invisible as social subjects especially in the political sector (Rapport & Dawson, 1998). Moreover, deepening this strategy is the enduring Chinese old saying of Luo Ye Gui Gen (falling leaves return to their roots), which dictates that no matter how far one travels, the ultimate destination is the ancestral home (Harris, 2010). For the shopkeeper in Praia, the “station” does not need to be comfortable or permanent, because the “roots” remain in China. This actually explains why they accept the temporary nature of their existence in Cape Verde. The modular shop is not a failure to integrate, but it is a strategic adaptation for a life lived in motion, ensuring that when the “leaves fall,” they are not stuck in foreign soil, but are free to return to the source.
Likewise, Ismaili spaces connected under one Canopy foster lifestyles that lend to both permanence and mobility. Ismailis, as a group, have an extensive history of living in diaspora, and the Indian Ismailis of Mozambique have thrived economically as transnationally situated businessmen, operating through co-ethnic and close community relations. This spatial, economic, and social mobility, which permeates and circulates fluidly through societies where they live, is key to their flexible citizenship (Horner, 2016; Ong, 1999). Paradoxically, this peripheral social location also exposes them to the ambivalence afforded to perpetual outsiders (Stoler, 2002).
They navigate that tension through the interventions in the built environment, like the Ismaili Centre in Lisbon, or the Ismaili Centre in Maputo, which are monumental, highly visible, and integrate local customs and materiality. Ismaili institutional spaces cultivate community relationships across sectors, a reality driven by faith and urban space. The imamate in Lisbon facilitates high-level relations with State officials and elites, both domestically and internationally (Daftary, 1998). Finally, the institutional space is used to keep a social touch on disadvantaged classes through its philanthropy and social programs (Daftary, 1998, Khan, fieldnotes May 28, 2025). Notably, these social programs, which manage tensions to the broader publics, are not often held in the high-value urban institutional spaces, but at a distance, in the urban peripheries, and previously, in the hinterland. Thus, for these groups, the material reality of their transnational urban spaces reveals the social reality as well.
Conclusion
This comparative ethnography of the Nizari Ismailis and New Chinese migrants in Lusophone Africa has sought to infuse the grand narrative of “Global Africa” through a grounded anthropological perspective. We confirm our central hypothesis: the middleman is not a fixed structural category but a mutable, adaptive social role that has transitioned from the rigid hierarchies of Portuguese subaltern colonialism to expanded agency and mobility in the global market.
The asymmetry between the Ismailis and the New Chinese migrants, in class, in institutional density, in temporal position, in methodological accessibility, is the empirical form that the bifurcation of the middleman role takes in Lusophone Africa. And this asymmetry is a faithful rendering of the full ecological niche occupied by Asian mobilities in the Lusophone world.
The Ismaili’s institutional networks and patient capital represent a form of migration pattern enacted in the post-colonial vacuum through development. In contrast, the New Chinese migrants’ modular insulation and fast capital embody a horizontal, extractive connection built on affordability and logistics. As opposed to the connected urban complexes of Ismaili-led development, New Chinese migrants maintain a singular urban form, the Chinese shops and their shopkeepers.
These two distinct strategies, together, expose the structural gaps left by the retreat and limitations of post-colonial state capacity. Our findings move beyond simple middleman theory. We argue that the middleman role has bifurcated into the Global Broker and the logistical node, which could be seen as the former seeking enduring political security and deep roots, and this fluidity and class heterogeneity could help to understand the contemporary Lusophone world.