Friendship fears and communities of convenience in Africa’s urban estuaries: Connection as measure of urban condition

Abstract

Chinese

Across the developing world, immigrants, internal migrants and long-time residents increasingly co-occupy and co-produce estuarial zones: sites loosely structured by the disciplines of state, formal employment or hegemonic cultural norms. In these hyper-diverse, often highly fluid sites, the appearance and form of friendships and solidarities are varied and revealing. Drawing on examples from rapidly transforming African cities – particularly Johannesburg and Nairobi – this article adds three facets to the emerging literature on urban friendship. First, it outlines conditions under which the localised intimacy of friendship represents a potentially frightening form of social obligation and regulation. Given many ‘southern’ urban economies’ uncertainty and migrants’ orientation to ‘multiple elsewheres’, local solidarities – including friendship – are often more frustration than facilitator. Second, it suggests that amidst these seemingly anomic, distrustful sites, residents forge shared values and socialities that eschew friendships’ potentially confining bonds. These ‘communities of convenience’ illustrate the value of solidarity in migrant-rich spaces while raising broader questions about the spatial scale and role of affective relationships in overcoming economic and physical precarity. It lastly argues that the relative strength of localised friendships provide a means of comparing urban sites while revealing rationalities – political, economic and social – at work: friendship fears reveal the distinct estuarial spaces shaped by ongoing movements of people into, out of, and through precarious cities of the south.

Keywords

ethics of difference friendship human mobility Kenya South Africa

Introduction

Human mobility – migration, urbanisation, and displacements – remains a socially corrosive and creative catalyst amidst Sub-Saharan Africa’s rapidly expanding, diversifying and mobile urban populations. Whether in novel peri-urban sites carved from agricultural land or amidst infrastructurally ill-equipped city centres, people’s interactions are decreasingly disciplined by state law, formal employment or hegemonic social norms. These engagements shape urban estuaries: gateways into, out of and through cities. Characterised by cyclical, inward and outward patterns of human, material and discursive mobility, these are emblematic of late capitalist, southern urbanism and give cause to reconsider the meaning of cityness and the boundaries of belonging (cf. Robinson, 2015; Roy, 2016).

Within the estuary, diverse and mobile populations are generating urban socialities that often deviate from the modes of belonging described in classic urban sociology or more recent debates around multiculturalism and friendship. Without the socio-economic conditions to create the urban solidarities of classic sociology, it is tempting to see these spaces as ungoverned or anomic (Clunan and Trinkunas, 2010). Indeed, they are often deeply fragmented by ethno-linguistic, class, religious and political allegiances. Hedging against such alienation and associated risks, one might expect private, quiet friendships (see Kathiravelu, 2012). While friendships and other forms of close intimate relationships exist within the estuary, many eschew friendship in favour of other socialities compatible with a fluid life of ‘multiple elsewheres’ (Levitt et al., 2011; Mbembe and Nuttall, 2004). These ‘communities of convenience’ are perhaps the most telling indicator of spaces in which rooting and local intimacy may be neither possible nor, more importantly, desirable. It is these alternates to space bound intimacy this article explores.

Understanding the spatiality of connection and distanciation nuances our understanding of urban socialities while reinforcing the necessity of multi-scalar analysis. Moreover, it enriches the literature on friendship by pointing to its peculiarity: that it is not a universal response to economic marginalisation and social distance, but one likely to emerge under particular conditions. More precisely, it considers the compatibility and incompatibility of friendships forged and simultaneously practiced at multiple spatial registers. Together this approach contests urban friendship’s practical and analytical efficacy in multiple ways: of its value to those involved, of its significance as a normative ideal, and of its heuristic utility as an indicator of novel urban formations and ‘new possibilities for social life’ (Robinson, 2013: 660). In this way, friendship fears become a mark of the estuary and a metric for developing the kind of post-colonial urban comparison Robinson (2015: 187) and others propose. Rather than take the city as our unit of analysis, it instead begins with vernacular social practices as the basis of urban comparison.

This article builds on Bunnell et al.’s (2012) suggestion that friendships and other social connections are objects in their own right and elements of broader socio-economic and political formations. Given humans’ seemingly intrinsic need for social recognition and intimacy, friendship fears speak to an ontologically unsettling social reality. As described below, friendship fears reveal a rationality of socio-economic survival and success associated with ongoing social fluidity and economic precarity. Noting such fears’ presence helps extend the study of what Watson (2008) terms ‘conflicting rationalities’. In more predictable, unified systems founded on formal labour systems – however exploitative – or place-bound family structures, friendship remains a hedge (or complement) to relatively stable and predictable, if oppressive, hegemonies. In spaces of massive precarity and unpredictability – but ones that retain possibility amidst this precarity – friendship fears become a way of negotiating, of surfing cross-cutting and dangerous waves that may carry individuals forward or, when nimbleness fails, drown them (see also Bridge, 2004). Ultimately reading city socialities will benefit from a schema clustering friendships’ forms and functions much as others have done regarding social networks and capital (Mayer, 2003; Portes, 1998; Putnam, 2007; Ryan et al., 2008). Mapping friendships and socialities against other spatialised indicators of, inter alia, rationalities of masculinity and familial success, economic and physical security, or diasporic engagement can reveal important correlations and causal connections.

The remainder of this article proceeds through four substantive sections. After briefly discussing my approach, I turn to Africa’s urbanisation and emerging urban estuaries. Their economic, social and physical precarity segues to a discussion of ‘friendship fears’ in which I outline how friendships – potentially desirable and necessary elsewhere – present potential risks to estuarial residents’ projects. The next section argues that even as people flee friendships, they nonetheless forge new socialities. These communities of convenience afford them the support, combinatorial freedom, and mutability needed to negotiate the estuary. The concluding section returns to questions of friendships’ heuristic potential in revealing underlying socio-economic necessities and the shape of contemporary urban life.

Approach and data

This article opens space for understanding social relationships – family, friendships, or other connections forging spatially bound and translocal community – amidst Africa’s urban precarity and fluidity. More specifically it explores possibilities that disconnection and social alienation are (a) something urban residents’ desire and (b) compatible with alternative socialities. While friendship fears may be normatively unappealing, it is by understanding the rationalities and necessities informing lets us better consider the potential practical and ethical implications of contemporary urban life.

While affect is friendship’s hallmark, this article largely considers its functional facets: its role as social resource and disciplinary mechanism that complements succour with sanction. Friendships are, in effect, voluntary contracts with unknown, emergent costs and benefits. Their appeal and perceived benefits are shaped by environmentally and historically conditioned rationalities and appetite for risks that rely, at least in part, on the other forms of social connections with which they intersect, are embedded or ultimately support or supplant (cf. Adams and Allan, 1998). This approach is undeniably functionalist and agent-centred. It attempts to strip the positive values often associated with friendship to reveal how individual rationalities and environment interact to generate dynamic urban socialities. My consideration of social capital is similarly illustrative. While debate continues over the concept’s utility, empirical manifestations, and measures, I use the term as an aggregated analysis of intergroup and interpersonal trust, organisational/associational membership, and geographically proximate forms of regular social engagement. This approach is imperfect but offers an empirically informed, conceptual provocation.

This article draws on an ecumenical set of data collected individually and collectively as part of broader research into the local historicity, competing rationalities and translocalism characterising lives within the urban estuary. Most of the information used here stems from migration-related research in Southern and Eastern Africa undertaken between 2002 and 2016. Each of the cities discussed are destinations and transit points for domestic and international migration including significant numbers of displaced persons. Together they represent a range of social, economic and political characteristics that allows for modest generalisations about trends and possibilities about African cities and, potentially, the rapidly expanding metropolises of the Global South. The account here intentionally speaks at the generalised level with illustrations from individual stories and accounts.

The survey data sparingly used here stems from interviews with 2211 people in three of the cities – Johannesburg, Maputo and Nairobi.¹ The surveys targeted particular groups of foreigners categorised by nationality. With the exception of Mozambicans included in the Johannesburg survey, data were collected on selected groups – Somalis, Rwandans, Sudanese and Congolese – straddling divides between economic and ‘forced’ migrants. Such sampling does not cover the cities’ full geographic footprints or social diversity but concentrated on areas on ‘estuarial zones’ (cf. Landau, 2015; Saunders, 2011; Singer et al., 2008). Although not prominent, this article relies on previously published statistical analysis (particularly Landau and Duponchel, 2011; Madhavan and Landau, 2011) and illustrations from qualitative fieldwork conducted alone or in collaboration with colleagues and doctoral students in Johannesburg and Nairobi, two of Africa’s most globally networked urban sites.

Mobility and Africa’s emerging urban estuaries

If friendships are formed in context, understanding them means gazing at the cities, neighbourhoods, suburbs in which they may be forged. Sub-Saharan African provides a panoply of sites as Africa’s urban population may well double in the next 20 years, from 294 million people in 2000 to 742 million by 2030. Twenty years on that figure may have doubled again, to 1.2 billion (UNDESA, 2013; UNFPA, 2008: 7). Given the precarity of land tenure, the constant search for economic opportunities and myriad other factors, people remain highly mobile within cities. As they move, they continually reshape their social space.

Three characteristics about African urbanisation warrant attention here. First, unlike the urbanisation of previous generations and regions – or contemporary middle-east migrations (Kathiravelu, 2012) – African cities’ expansion and social diversification is not driven or accompanied by heightened demand for labour. Even remarkable economic growth in GDP per capita over the last decade has not typically translated into widespread employment opportunities (Ancharaz, 2011). While African cities have long been characterised by high levels of informality and petty trade, there are now metropolises (or large parts of them) offering slim prospects of formal employment or a regular wage (De Boeck and Plissart, 2006). In some places, migration-driven urbanisation is accelerating as economic opportunities for formal work decline (Kihato and Muyemba, 2015).

Given global urbanisation rates, there are reasons to unsettle modernisation tropes of years past and more contemporary accounts of modernities’ imperfections and disillusionments (cf. Burdett and Sudjic, 2011; Ferguson, 1999). While valuable heuristics, the metanarratives of modernity and global capitalism are analytically flattening and fail to fully capture the range of extant and emerging socialities and subjectivities. Whether Southern cities foreshadow the future for late capitalist sociality everywhere (Comaroff and Comaroff, 2012), something is happening that deserves attention in ways that beg for novel ethics and analysis and open spaces to unsettle urban studies and friendship scholars’ weaving a single story.

Understanding the socio-spatial context informing friendships means considering this historicity of ‘incomplete’ urbanisation – one in which urban remained sites of extraction not belonging – and how the past still informs relations to place and people (see Bank, 2011). Only then can one relate these trends to presumptions about migrant integration and the potential value friendship holds for urban residents. Doing so reveals forms of urbanisation and social formation that resonate with elements of the American and European trajectories – the canonical modernisation experience – while suggesting forms of sociality that may well be analytically and normatively unfamiliar or unsettling to external observers. Friendship fears within fluid, communities of convenience suggests something important about the nature of urbanism and the scale at which we can observe generative social responses to late-stage capitalism and institutional frailty.

Second, Africa’s dramatic urban variations between and within cities and suburbs are generate striking diversities of urban forms. Eastern Africa was only 21% urbanised in 2008 while Central and Western Africa was 42% urbanised and Southern Africa 46% (UNFPA, 2008). When moving beyond national or regional trends, our focus is often on the continent’s economic capitals, places such as Accra, Nairobi, Johannesburg or even Duala (Bekker and Therborn, 2011). However, the most rapid forms of growth are in countries’ secondary cities, small towns, and on urban peripheries (UN-DESA, 2010). These often poorly resourced sites may also prove to be the most theoretically significant.

It is on the urban edge and in the churning neighbourhoods of the old cities where migration and mobility are the most common and the effects most pronounced. Trefon speaks about ‘the suburbs of the suburbs’ in Congo, spaces that just a few years ago would have all the outward appearance of villages or rural settings (Trefon, 2009: 15). For example, Ongata Rongai, one of Kenya’s fastest growing settlements. Ten years ago it was little more than a small trading post on the just beyond the edge of Nairobi’s administrative borders. Today it sports multi-storey apartment buildings, shopping malls and service providers (Landau, 2015). The continent’s wealthiest city, Johannesburg, has grown more slowly than Ekhureleni, its more affordable (if less prosperous) eastern neighbour. On Johannesburg’s northern frontier, Diepsloot was originally designated to temporarily house a few thousand people displaced by urban development programmes and shifting agricultural practices. Created in the mid-1990s, 15 years later it hosts tens of thousands of people from across the country along with neighbouring Mozambique, Zimbabwe, Kenya, Ethiopia, Somalia and Pakistan (see Harber, 2011). Such peri-urbanism not only stems from new arrivals, but also those evacuating city centres. These include countries’ elite, well-connected or upwardly mobile, abandoning inner-city neighbourhoods in favour of the space and security new peri-urban estates and gated communities provide (see Briggs and Mwamfupe, 2000; UN Habitat, 2011). Others are displaced by urban regeneration, slum clearance, violence, rezoning or myriad other reasons.

The growth and churn of urban peripheries and neglected neighbourhoods leads to a final observation on intersections of urbanisation and formal institutions. Tanzania, Mozambique, Namibia, South Africa and a slew of other countries long maintained severe pro-rural policy biases intended to discourage urbanisation (see Fox, 2014). Even at their peak, these policies failed to halt urbanisation but instead drove it underground and into informally managed spaces (see Owusu-Ansah and O’Connor, 2010; Piot, 2006). According to Kironde, by 1979 – at the height of Tanzania’s socialist experiment – 70% of Dar es Salaam’s urban residents were living outside of formally planned areas (Kironde, 2007: 115). The expansion of African cities continues to manifest in the economically poorest, least institutionalised parts of town: ‘between 1990 and 2000, slum areas grew at a rate of 4.53% whilst overall urban growth rates were 4.58% in the same period’ (UN-Habitat, 2010). The factors and trends described above create conditions of highly concentrated population growth and rapid social diversification accompanied by few opportunities to access formal employment or public services. The consequences include the estuaries and gateways now surrounding many cities: spaces of continued mobility, relatively weak formal regulatory infrastructure, and levels of physical and economic precarity (Saunders, 2011). The concern here is not to condemn planning failures or policy frameworks (cf. Huchzermeyer, 2011), but to characterise these urban forms and unlock their significance.

Many policy and scholarly responses to these trends are haunted by Malthusian spectres of anomie, violence and social fragmentation (cf. Kihato and Muyemba, 2015; Meyers, 2011; UN Habitat, 2008). Some analysts call for temperance among the alarmists (Potts, 2011; also Bocquier, 2005); others with calls for improved urban management and governance. With few exceptions, the literature on African urbanism avoids close discussions of the forms of sociality that form the sometimes invisible ‘human infrastructure’ (Simone, 2004). While a thin literature on the economic value or risks of social networks exists (see Madhavan and Landau, 2011; Myroniuk and Vearey, 2014), such discussions are often divorced from the human or affective aspects of social ties. Rather they concentrate on policy implications: fighting poverty, combatting disease, limiting violence and other social ills.

Drawing meaning from these meeting places of multiple human flows largely unregulated by states or dominant social institutions means recognising they are at once unstable and fertile grounds: possibility amidst precarity (see Piel and Opoku, 1994; Simone, 2004). Much like natural estuaries where the interaction between tides and rivers create unique and dynamic ecosystems, these urban gateways are marked by distinctive forms of sociality and solidarity taking shape amidst remarkable disparities of wealth, transforming gender roles, success narratives and geographic trajectories (cf. Bridge, 2004). Viewed from the urban lens, this globalisation (or translocalism) from below (cf. Portes, 1997) transforms social spaces as movements and imaginations link neighbourhoods, towns or streets to archipelagos of community and systems of meaning in home villages and diasporic communities spread (and spreading) across the continent and beyond (cf. Dick and Duchêne-Lacroix, 2014). While metaphoric language risk analytically excluding human agency, the ‘estuary’ and ‘archipelagos’ of meaning and exchange nonetheless help capture the distinctiveness of spaces shaped by multiple agents marked by transience and translocalism. It is in this context that discussions of friendships can begin: their value and what their forms might reveal about the nature of emerging urban spaces and social life.

This makes it all the more critical to explore and explain the social qualities of these emerging orders. Some of these resonate with patterns seen elsewhere: ethnic groups hunkering down (cf. Putnam, 2007) or neighbourhood associations. Inevitably, friendships are part of these stories albeit in ways that may be fleeting or far-flung. Yet closer examination of residents’ response to friendships suggests the very forms of space-based solidarities and connections often celebrated in studies of urban resilience may counter rationalities informed by residents’ trans-temporal, trans-local expectations and imaginations. Rather than embed themselves in geographically proximate relationships and friendships as a form of shelter and support, many urban residents intentionally self-alienate and self-deprive in service of kin, community and futures elsewhere (see Amuedo-Doantes and Pozo, 2006; Dzingirai et al., 2014). These translocal and trans-temporal connections and imaginations, fuelled by desires for onward movement or return, lead to desires to minimise local obligations; to friendship fears.

Friendship fears

Africa’s urban estuaries are unsettled spaces characterised by emergent and translocal patterns of social and economic exchange. Although people remain embedded in collectives – ethnicity, language groups, regional loyalty – estuaries can corrode varied corporate and ‘traditional’ identities. Many residents elude the surveillance and social sanctions stemming from visibility in ‘home town’ associations while resisting the localised obligations of spatially proximate socialities. Whereas Conradson and Latham (2005) demonstrate the mobility of friendship networks among New Zealanders in London, estuarial residents often shy away from such categorical affiliations. Given the insecurity of employment, land tenure, the possibility of violence or displacement, estuarians often maintain feet in multiple sites while resisting rooting anywhere (Freemantle, 2010). As one Congolese migrant in Johannesburg remarks: ‘Here I am nobody. I hide from the police, I hide from the South African government, I hide from my government at home. Sometimes I even hide from my own countrymen … you see this how I survive’ (in Kankonde, 2016).

The instabilities associated with African urbanism are exacerbated by residents’ own ambitions and trajectories. For many, the urban sites they occupy are stations in ongoing journeys rather than destinations. The primary goal is not to found a new life, but to extract resources from the cities that may subsidise a ‘real’ life back home or to prepare them movements on to any number of ‘multiple elsewheres’ (Mbembe and Nuttall, 2004). Indeed spouses, children, real estate and belongings are often elsewhere while single men and women earn money in the cities to sustain or enhance them. New arrivals may establish second urban families or surrogates, a well-documented form of friendship (Brettell, 2003; White, 1990). Yet ethnic and political ties elsewhere often work against full social integration into cities as people seek to limit being known in or tied to the city. Desires for a rural retirement or onward movement further limit people’s proximate financial and emotional investments. The results are spaces occupied by those physically on the move alongside others whose orientation can generate a kind of indefinite temporariness (see Kankonde, 2010; Kihato, 2013; Landau and Freemantle, 2016).

Allan (2008) argues that amidst great precarity, fluidity and diversity, individualised friendships become important means of generating social identities and protection. Walsh (2007) similarly describes friendships comforting transients. Yet living conditions and ambitions militate against the fixity and surveillance that come with strong localised allegiances and friendships. As Guyer (2007) notes, increased uncertainty about the future undermines an ability for long-term or patient planning. Amidst such precarity, success may seem arbitrary and unpredictable. It also depends far less on the kind of sustained connections and social safety nets that provide resilience and investment in more socially or economically stable environments.

Whereas social networks and friendships may offer psychological security, respectability and support under certain conditions, they are inherently binding. Friendship, no matter how novel, is founded on trust and reciprocity. It creates zones of acceptable and respectable behaviour: expectations for what one does and limits on how a task may be accomplished. As a young Congolese man in Johannesburg notes, ‘Often my young brothers call me to inform me of the gossip people tell about me and the family. I know that sometimes they lie. But if you hear that, you have to send to avoid that people laugh at them’ (in Kankonde, 2016: 40). Ultimately, these connections limit spontaneity and the possibilities of continual becoming; of reshaping a person’s connections, activities or location. So where those working in the kind of structured (if exploitative) labour conditions Kathiravelu (2012) describes follow prescribed, contractually bound courses through new spaces, estuarial residents must constantly read and remake the space around them.

Reciprocal obligations along with the economic precarity of the post-industrial (or perhaps more accurately, neoliberal, non-industrial) era further encourage decoupling from local and distant social relations and corollary spatial connections. In Worby’s (2010) account of Zimbabweans in Johannesburg disconnection is rooted less in the nature of the city, per se, but stems instead from the elaborate and occasionally oppressive social networks extending among vulnerable populations migrant and otherwise. These take the form of demands from family and sending communities to send money, food, durable goods, and clothing to their places of origin (see also Dzingirai et al., 2014; Kihato, 2013; Madsen, 2004). Perhaps more immediately, these demands may also manifest in requests to physically host and support relatives and friends in continual succession or even simultaneously. The result being a kind of pyramid scheme in which a migrant pioneer ends up hosting close relatives followed by ever more distant connections, often under cramped conditions: self-built houses, storage rooms, sub-let and sub-divided apartments, or even occupied offices. Rather than offering comfort or social security, such relationships introduce not only material obligations and inconvenience but an additional source of uncertainty.

To deny others support when you are even marginally more resource endowed is almost unimaginable, resulting in acute personal or familial shame. For migrants, the betrayal of expectations not only means an exit from potential reciprocal support at some indeterminate date, but may elicit explicit countermeasures ranging from reputation costs to curses, theft, or physical threat. In these cases, the strong social networks and overlapping reciprocal obligations which scholars and practitioners point to in providing protection and development opportunities (see Aguilera and Massey, 2003; Kanaiapuni et al., 2005; Landau and Duponchel, 2011; Munshi, 2003; World Bank, 2005), become burdensome and counterproductive. The failure to maintain them during periods of crisis can lead to a kind of ‘social death’: one’s social exclusion and stigmatisation.

Confronted with the likely failure to meet social obligations or the sometimes onerous costs of doing so – not only short-term material costs, but comfort and the possibility of capital accumulation – migrants hoping to make their lives in cities often seek forms of social invisibility. For many, desiring disconnection and invisibility stems from the seemingly shameful activities one undertakes to make it in a new place (see Kankonde, 2010; Kihato, 2013; Malauene, 2004). For those involved in sex work or other criminalised behaviours, the source of stigma is perhaps obvious (although many women describe a pride in being able to support their families at home regardless of the form of employment (see Oliveira and Vearey, 2015)). A single Somali woman with two children in Nairobi explains:

My community they judge me, gossip about me that I am a prostitute – a woman without a husband. That I chased my husband away. How can I go to these people for help? I am ashamed, I do not come out, I move my house many times to run away from people, it’s like I am not a person. (in Kihato, 2017)

For others the downgrading of skills – from accountant to taxi driver; from teacher to day labourer – is an embarrassment. Many migrants attempt to author their narrative by sending home glittering images of urban success and want to avoid networks and observational disciplines that might undermine their stories.

Writing about patterns of social integration and solidarity elsewhere, Bulley suggests that when everyone is guest and host – as is the case in the estuary – all become hostages: no one sets the terms of engagement and all become subject to others whims and will. Derrida proffers the term ‘hostipitality’, to connote the hostility such situations tend to produce; conditions where people must accommodate others around them even as they struggle to for their own economic and existential security. Hostility and manifestations of spatially chauvinist rhetoric is certainly one possible outcome and is undeniably common across Africa and elsewhere in the world (see Geschiere, 2009; Misago, 2011). But even friendship seems to do little to shield immigrants. As a Zimbabwean who had sought integration in Johannesburg described in 2009:

Yeah, you know, you cannot say I’m safe because you are a South African. Let’s say you have friends which are foreigners, you learn their language. You are found there talking their language, it means they won’t ask you a paper, they will just attack you. So everyone is not safe from xenophobia. (in Holaday, 2010)

Yet it is not only xenophobia that one must fear. The archipelagic character of the estuary means personal revelations travel with the potential for violent, locally realised risks. In Nairobi many spoke in terms similar to this Congolese man:

On 16 April 2016, I was beaten by people speaking my language here in Nairobi. It was at night. I saw people looking at a picture and pointing at me. It was the people I escaped in Congo, they have found me, they know where I am. Since then I can’t talk to anyone especially from my country. (in Kihato, 2017)

Yet the fragmentation and friendship fears within the estuary belie the productive or potentially beneficial social relationships. Indeed, as Simone’s (2004) work suggests, the kaleidoscope of moving people, goods, and organisational forms creates an infrastructure and opportunities for those with skills and readiness for risk.

Emerging solidarities as communities of convenience

The following paragraphs schematically outline means through which varied social relations take shape in environments with weak institutions, few dominant social groups, and populations who often wish to remain transient and invisible (i.e. largely without local friends). As available data do not describe friendships, per se, I begin by considering social capital, an indicator of context and socialities albeit an imperfect proxy for the presence or meaning of friendships. I then explore the countervailing pressures to be transient and invisible while building some social connections that can provide meaning, order and some modicum of assistance. These include examining religious affiliations which can offer surrogate friendships and intimacy while remaining highly individualised, flexible and intentionally fleeting. I then touch on other forms of membership and organisation, particularly a set of discourses and practices I have termed tactical cosmopolitanism. These are but examples of the multiple socialities now taking shape.

Religious, cultural or economic, collective participation is a potentially important mechanism for inculcating a sense of common purpose and creating the bases of for meaningful relationships. However, given the spaces and populations described, social networks and relationships are often spread thinly across many people and places leaving the estuaries relatively devoid of ‘thick’ connections. As such it comes as little surprise that the surveys show remarkably low levels of trust and regularised engagement between ethnic and national groups. What is more surprising is the limited trust and bonds within them. Even among citizens in both Johannesburg and Maputo, levels of social capital – trust of each other and their public institutions – are strikingly low (cf. Putnam, 2000). Nairobi offers a slightly more trusting environment, although here too the data reflect deep tensions. Networks of clan, neighbourhood or coreligionists undoubtedly exist (see Nzayabino, 2010), but these are often fragmented and functional, organised without an explicit recognition or sense of mutual obligation to those beyond familial boundaries (see Sommers, 2011). Instead, they are often limited to assisting others only to overcome immediate challenges that presents collective risks (e.g. police raid or mass eviction) or where people’s status in their sending communities may be at risk (e.g. in the case of returning corpses for burial) (see Andersson, 2006; Ayiera, 2008; Dzingirai et al., 2014).

Even as migrants may accept de facto integration and social relations within their residential communities, they continue to resist a kind of embedding or intimacy. Indeed, more than three-quarters of those included in the Johannesburg survey (76%) felt it important for migrants’ to retain their distinct identities and loyalties while staying in South Africa. Araia’s study of Eritreans in Johannesburg also found this sentiment among those owning businesses and otherwise objectively integrated. ‘South Africa is not the place for Eritreans’, one respondent notes, ‘look at it, there are none of them who live here permanently’ (Araia, 2005: 34). Broader research reveals a generalised tendency to see Johannesburg as a site of temporary residence: despite the expense and dangers of coming, only 40% of the non-South Africans in survey predicted that they would even be in the country after two years. When asked about future plans, only 20% of respondents thought they were likely to return to their countries or communities of origin while 25% were planning onward journeys. Similarly, when questioned about where they hoped to raise their children, just over one-quarter (26.8%) identified South Africa, while the rest were almost evenly divided between the respondents’ countries of origin (30.3%) or a third country (32.1%). Critically, journeys home or onwards often remain practically elusive for reasons of money, safety or social status, leaving almost two-thirds of Johannesburg’s non-national population marooned in the city but not taking within it.

This sense of transience is widespread among the city’s non-nationals, but does not represent capture by a consolidated, subjectively accepted exile/migrant community or friends from within it (the kind of relationships Kathiravelu (2012) describes elsewhere). Mang’ana (2004) and Misago (2005) both report, for example, that even people from the same country carefully avoid close association with other ‘exiles’. Although migrant groups occasionally forms collective (usually national or ethnic) identities, these are typically short lived and frequently fragment or become reconfigured. When they do associate, they use combinations of national, ethnic and political affiliations for strategic ends while avoiding standing obligations or relations that would facilitate (or demand) permanent settlement (Amisi and Ballard, 2005; Götz and Simone, 2003: 125; Mang’ana, 2004). They also cling to multiple loyalties and continue participating in multiple networks. These act as resources that provide them with the weak links needed to gather information and also help them hedge and shift affiliations and tactics at a moment’s notice (Granovetter, 1973). In doing so, they avoid capture by friends, relations and the state while reshaping the city’s social and political dynamics.

Religion is a notable exception to a relative absence of social organisation in the urban estuary. Throughout Europe and Asia, religious institutions have played central roles in binding population to each other and to place (and in excluding everyone else). In doing so, they provide an environment where close friendships and socially legitimate romantic relationships may blossom. Where states have faint influence, they can serve to help generate the subjectivities and notion of a common future necessary to maintain a strong, socially binding social space. However, a combination of factors, including the increasing heterogeneity of the urban population, suggests that religious institutions play a somewhat different role in contemporary Africa cities. Among the Nairobi citizenry we surveyed, for example, 65.6% were Protestant, 30.6% Catholic, 2.7% Muslim with only 0.3% claiming no religion. In Johannesburg, the sample was 59.7% Protestant, 18.8% no religion, 14.1% Catholic, and 6.8% Muslim. (The foreign-born population in Johannesburg was more evenly divided with 39% Protestants, 28.5% Catholics, 26% Muslims, and 6.3 claiming no religion.) While urban Africans are strongly religious, the denominational divisions within those affiliations – and the often fractured and conflictual relationships among them – serves more to divide than unify. As Schiller et al. (2006) note, travelling churches have increasingly created localised conduit to a de-ethnicised (or less ethnicised) global culture. Viewed locally they are often more fragmentary than binding.

Along with the sheer diversity of competing claims for religion and belonging, the liturgical content of many churches serves to further undermines the emergence of a territorially bound or state-centred subjectivity. This is perhaps most visible in the ever expanding pool of Pentecostal churches operating within Africa’s urban centres. At one level, these inclusive (often massive) institutions offer the possibility of bridging barriers between various groups and forging the basis for close relationships.

While they offer a sense of salvation in the form of ‘health and wealth’, they are both highly individualistic and distinctly post-territorial in their outlook. The ‘Community of Christ’ they promote is often directly situated against close ties to family, friends and non-believers (cf. Schiller et al., 2006). If there is a friendship here it is the dyadic one with the imagined deity, particularly its manifestation in human form (i.e. Jesus). In many cases, pastors preach against close connections or even contact with those outside the church, sometimes including family members who have yet to be saved (Kalu, 2008; Nzayabino, 2010). Many churches build on their strong connections to institutions in Nigeria, Ghana, Congo and the USA while the churches’ founders – who are themselves migrants – use the pulpit as a place where they can enter a global social universe. This may be materially productive but is not about forging close localised connections. In some churches, women use a personal relationship with Jesus as a substitute for failed or unavailable intimate relationships. By ‘marrying Jesus’, they further remove themselves from localised social or romantic connections. As noted earlier, there is also a strong emphasis on individual success. Their fragmentary and often conflictual sources of religious authority further serve to deny the state – or indeed even a single church – the possibility of naming what is good and the direction the collective should follow.

Religion, at least as described above, allows people to be in a place but not of it: to be neither host nor guest. In this it contributes to a broader approach to estuarial life which can be categorised as ‘tactical cosmopolitanism’ (Landau and Freemantle, 2010) or the search for ‘usufruct rights’ (Landau and Freemantle, 2016). Recognising ascendant forms or exclusion levied against them, migrants draw on a variegated language of belonging that makes claims to the city while positioning them in an ephemeral, superior and unrooted condition where they can escape localised social and political obligations. These are not necessarily grounded in normative ideas of ‘openness’ or intended to promote universal values of any form. Rather, migrants practically and rhetorically draw on various, often competing, systems of cosmopolitan rights and rhetorics to insinuate themselves, however shallowly, in the networks and spaces needed to achieve specific practical goals. Unlike transnationalism, which is often about belonging to multiple communities – or shuttling between them – these are more ‘decentred’ tactics that enable them to participate in many worlds without being bound by any. This is not to deny the possibilities of friendship and close connection, but rather to describe a situation in which at least localised friendship acts as a weight or a form of binding that works against their invisibility and fluidity.

Concluding notes on friendship in the estuary

The forms of belonging and sociality emerging across many African cities are often ‘side effects’ of other economic, social and political objectives. People’s social relations are typically structured by a motley collection of actions undertaken by individuals and groups fragmented by language, religion, legal status and mutual distrust. Rather than strong, intimate relations, people maintain – by choice or necessity – social connections allowing them to swiftly combine disparate elements of their surroundings according to current interests and objectives. Indeed, perhaps one of the most critical characteristics of the new forms of social mobilisation taking shape is the degree to which they are generating archipelagic belongings: socio-spatial forms of relationships and socialities not centred on their residential locations, but rather designed to maximise social value and connections elsewhere or across a diaspora.

Accepting friendship may counter emic understandings of ‘success’ frees us to treat amicability as both a subject for empirical inquiry and a potentially powerful heuristic. As scholars seek means of comparing and contrasting urbanism (Robinson, 2015) friendships’ presence, form or nature add to the tool box. Social capital is an important but nonetheless problematically incomplete means of capturing the realities of people’s social lives and connections. Clearly other, often more micro-level indicators of sociality are needed. Friendship can potentially provide one. Its analytical effectiveness means going beyond simply marking its absence or presence. Instead there is value in considering categories or continua of friendships that can provide more nuance and comparative leverage. Ultimately we may then begin to move beyond the anthropological or geographic project of noting and mapping. Seeking a means of drawing connections among emergent friendships at multiple geographic scales, individual or collective projects, and residential environments holds the promise of a broader comparative account of diversity, mobility and sociality in the complex and yet poorly understood urban spaces so many people now occupy.

In tying the emergence of spatialised friendship to broader processes of socio-economic transformation and mobility, we continue to the pioneering work of the Manchester school (e.g. Mitchell, 1969). Their interests were in modernisation, ours may be in the diverse paths of a neoliberal post-modernity. Like their work, the study of friendship is actor-centred, understanding how networks form in the midst of broader upheavals and socio-spatial transitions. The task here is neither to understand only individual attributes and subjectivities, but also, ‘the characteristics of the linkages in their relationship to one another’ (Mitchell, 1969: 4), whether those ties are proximate, distant, fleeting or enduring. In doing so, it also reinforces calls to see cities as interconnected archipelagos; an epistemology that looks below and above the city as unit of analysis. Embedding friendship in broader studies of networks and exchange may ultimately allow it to become an indicator of varied urban forms. If nothing else, it adds depth to our understanding of emergent urban socialities.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Adams

Allan

(1998) Contextualizing friendship. In: Adams

Allan

(eds) Placing Friendship in Context. Cambridge: Cambridge University Press, pp. 1–17.

Aguilera

Massey

(2003) Social capital and the wages of Mexican migrants: New hypotheses and tests. Social Forces 82(2): 671–701.

Allan

(2008) Flexibility, friendship and emotions. Sociological Research Online 16(1): 1–16.

Amisi

Ballard

(2005) In the Absence of Citizenship: Congolese Refugee Struggle and Organisation in South Africa. Forced Migration Working Paper #16. Available at: http://migration.wits.ac.za/AmisiBallardwp.pdf.

Amuedo-Doantes

Pozo

(2006) Remittances as insurance: Evidence from Mexican immigrants. Journal of Population Economics 19(2): 227–254.

Ancharaz

(2011) Trade, jobs and growth in Africa: An empirical investigation of the export-led jobless growth hypothesis. Draft paper prepared for the 3rd ICITE Regional Conference on Trade, Jobs and Inclusive Development in Africa, (22–23 September 2011), Gammarth, Tunisia.

Andersson

(2006) Informal moves, Informal markets: International migrants and traders from Mzimba District, Malawi. African Affairs 105(420): 375–397.

Araia

(2005) Routes, motivations, and duration: Explaining Eritrean forced migrants’ journeys to Johannesburg. Master’s Thesis, University of the Witwatersrand, Johannesburg.

Ayiera

(2008) Burying our dead in your city: Interpreting individual constructs of belonging in the context of burial of loved ones in exile. MA thesis, University of Witwatersrand, Johannesburg.

10.

Bank

(2011) Home Spaces, Street Styles: Contesting Power and Identity in a South African City. London: Pluto Press.

11.

Bekker

Therborn

(eds) (2011) Capital Cities in Africa: Power and Powerlessness. Pretoria: HSRC Press.

12.

Bocquier

(2005) World urbanization prospects: An alternative to the UN model of projection compatible with the mobility transition theory. Demographic Research 12: 197–236.

13.

Brettell

(2003) Anthropology and Migration: Essays on Transnationalism, Ethnicity, and Identity. Lanham, MD: Altamira Press.

14.

Bridge

(2004) Reason in the City of Difference: Pragmatism, Communicative Action and Contemporary Urbanism. London: Routledge.

15.

Briggs

Mwamfupe

(2000) Peri-urban development in an era of structural adjustment in Africa: The City of Dar es Salaam, Tanzania. Urban Studies 37(4): 797–809.

16.

Bunnell

Yea

Peake

et al . (2012) Geographies of friendships. Progress in Human Geography 36(4): 490–507.

17.

Burdett

Sudjic

(eds) (2011) Living in the Endless City. London: Phaidon.

18.

Clunan

Trinkunas

(2010) Ungoverned spaces? The need for re-evaluation. In: Clunan

Trinkunas

(eds) Ungoverned Spaces: Alternatives to State Authority in an Era of Softened Sovereignty. Stanford, CT: Stanford University Press, pp. 3–16.

19.

Comaroff

(2012) Theory from the South: Or, how Euro-America is evolving toward Africa. Anthropological Forum 22(2): 113–131.

20.

Conradson

Latham

(2005) Friendship, networks and transnationality in a world city: Antipodean trans-migrants in London. Journal of Ethnic and Migration Studies 31: 287–305.

21.

De Boeck

Plissart

M-F

(2006) Kinshasa: Tales of the Invisible City. Leuven: Leuven University Press.

22.

Dick

Duchêne-Lacroix

(2014) Multi-local living in the Global South and Global North: Differences, convergences and universality of an underestimated phenomenon. Trialog 116/117(1–2): 4–9.

23.

Dzingirai

Mutopo

Landau

(2014) Confirmations, coffins and corn: Kinship, social networks and remittances from South Africa to Zimbabwe. Migrating out of Poverty Research Consortium Working Paper 18, Sussex: University of Sussex.

24.

Ferguson

(1999) Expectations of Modernity: Myths and Meanings of Urban Life on the Zambian Copperbelt. Berkeley, CA: University of California Press.

25.

Fox

(2014) The political economy of slums: Theory and evidence from Sub-Saharan Africa. World Development 54: 191–203.

26.

Freemantle

(2010) ‘You can only claim your yard and not a country:’ Exploring contexts, discourse and practices of quotidian cosmopolitanism amongst African migrants in Johannesburg. PhD dissertation, Johannesburg: University of the Witwatersrand.

27.

Geschiere

(2009) The Perils of Belonging: Autochthony, Citizenship, and Exclusion in Africa and Europe. Chicago, IL: University of Chicago Press.

28.

Götz

Simone

(2003) On belonging and becoming in African Cities. In: Tomlinson

Beauregard

Bremner

et al (eds) Emerging Johannesburg: Perspectives on the Postapartheid City. New York: Routledge, pp. 123–147.

29.

Granovetter

(1973) The strength of weak ties. American Journal of Sociology 78(6): 1360–1380.

30.

Guyer

(2007) Prophecy and the near future: Thoughts on macroeconomic, evangelical, and punctuated time. American Anthropologist 34: 409–421.

31.

Harber

(2011) Diepsloot. Cape Town: Knopf.

32.

Holaday

(2010) The ‘right to have rights’: Citizenship, legal identity and political community in Johannesburg. Masters of Arts in Forced Migration, University of the Witwatersrand.

33.

Huchzermeyer

(2011) Cities with Slums: From Informal Settlement Eradication to a Right to the City in Africa. Pretoria: Juta.

34.

Kalu

(2008) African Pentecostalism: An Introduction. Oxford: Oxford University Press.

35.

Kanaiapuni

Donato

Thompson-Colon

et al . (2005) Counting on Lin: Social networks, social support, and child health status. Social Forces 83(3): 1137–1164.

36.

Kankonde

(2010) Transnational family ties, remittance motives, and social death among Congolese migrants: A socio-anthropological analysis. Journal of Comparative Family Studies 41(2): 225–244.

37.

Kankonde

(2016) Taking roots in the name of God?: Super diversity and migrant Pentecostal churches’ legitimation and social integration in post-apartheid South Africa. Thesis, Doctor of Philosophy, University of Göttingen.

38.

Kathiravelu

(2012) Social networks in Dubai: Informal solidarities in an uncaring state. Journal of Intercultural Studies 33(1): 103–119.

39.

Kihato

(2013) Migrant Women of Johannesburg: Everyday Life in an In-Between City. London: Palgrave.

40.

Kihato

(2017) Social networks amongst Nairobi’s displaced. Draft report for the Urban Institute. Washington D.C.

41.

Kihato

Muyemba

(2015) The challenges and prospects of African urbanisation: Forging Africa’s economic growth through sustainable urban policies. Unpublished paper for the African Centre for Cities, University of Cape Town.

42.

Kironde

JML

(2007) Race, class and housing in Dar es Salaam: The colonial impact on land use structure 1891–1961. In: Brenna

Burton

Lawi

(eds) Dar es Salaam: Histories from an Emerging African Metropolis. Dar es Salaam: Mkuki na Nyota, pp. 97–117.

43.

Landau

(2015) Being cosmo: Displacement, development and disguise in Ongata Rongai. Africa 85(1): 59–77.

44.

Landau

Duponchel

(2011) Laws, policies, or social position? Capabilities and the determinants of effective protection in four African cities. Journal of Refugee Studies 24(1): 1–22.

45.

Landau

Freemantle

(2010) Tactical cosmopolitanism and idioms of belonging: Insertion and self-exclusion in Johannesburg. Journal of Ethnic and Migration Studies 36(3): 375–390.

46.

Landau

Freemantle

(2016) Beggaring belonging in Africa’s no-man’s lands: Diversity, usufruct and the ethics of accommodation. Journal for Ethnic and Migration Studies 42(6): 933–951.

47.

Levitt

Lucken

Barnett

(2011) Beyond home and return: Negotiating religious identity across time and space through the prism of the American experience. Mobilities 6(4): 467–482.

48.

Madhavan

Landau

(2011) Bridges to nowhere: Hosts, migrants and the chimera of social capital in three African cities. Population and Development Review 37(3): 473–497.

49.

Madsen

(2004) Living for home: Policing immorality among undocumented migrants in Johannesburg. African Studies 63(2): 173–192.

50.

Malauene

(2004) The impact of the Congolese forced migrants’ ‘permanent transit’ condition on their relations with Mozambique and its people. MA Thesis, Johannesburg: University of the Witwatersrand.

51.

Mang’ana

(2004) The effects of migration on human rights consciousness among Congolese refugees in Johannesburg. Master’s Thesis, University of the Witwatersrand, Johannesburg.

52.

Mayer

(2003) The onward sweep of social capital: Causes and consequences for understanding cities, communities and urban movements. International Journal of Urban and Regional Research 27(1): 110–132.

53.

Mbembe

Nuttall

(2004) Writing the world from an African metropolis. Public Culture 16(3): 347–372.

54.

Meyers

(2011) African Cities: Alternative Visions of Urban Theory and Practice. London: Zed Books.

55.

Misago

(2005) The impact of refugee-host community interactions on refugees’ national and ethnic identities: The case of Burundian refugees in Johannesburg. Master’s Thesis, University of the Witwatersrand, Johannesburg.

56.

Misago

(2011) Disorder in a changing society. In: Landau

(ed.) Exorcising the Demons Within: Xenophobia, Violence and Statecraft in Contemporary South Africa. Johannesburg: Wits University Press, pp. 89–108.

57.

Mitchell

(ed.) (1969) Social Networks in Urban Situations: Analyses of Personal Relationships in Central African Towns. Manchester: Manchester University Press, for the Institute of Social Research, University of Zambia.

58.

Munshi

(2003) Networks in the modern economy: Mexican migrants in the U.S. labor market. The Quarterly Journal of Economics 118: 549–599.

59.

Myroniuk

Vearey

(2014) Social capital and livelihoods in Johannesburg: Differential advantages and unexpected outcomes among foreign-born migrants, internal migrants, and long-term South African residents. International Migration Review 48(1): 243–273.

60.

Nzayabino

(2010) Spiritual ecology: The role of the church in territorialising belonging and its impact on integration of migrants in South Africa. MA thesis, University of the Witwatersrand, Johannesburg.

61.

Oliveira

Vearey

(2015) Images of Place: Visuals from Migrant Women Sex Workers in South Africa. Medical Anthropology 34(4): 305–318.

62.

Owusu-Ansah

O’Connor

(2010) Housing demand in the urban fringe around Kumasi, Ghana. Journal of Housing and the Built Environment 25(1): 1–17.

63.

Piel

Opoku

(1994) The development and practice of religion in an Accra suburb. Journal of Religion in Africa 24(3): 198–227.

64.

Piot

(2006) Togolese cartographies: Re-mapping space in a post-Cold War city. In: Konings

Foeken

(eds) Crisis and Creativity: Exploring the Wealth of the African Neighbourhood. Leiden; Boston: Brill, pp. 197–210.

65.

Portes

(1997) Globalization from below: The rise of transnational communities. University of Oxford. Transnational Communities Working Paper.

66.

Portes

(1998) Social capital: Its origins and applications in modern sociology. Annual Review of Sociology 24: 1–24.

67.

Potts

(2011) Circular Migration in Zimbabwe and Contemporary Sub-Saharan Africa. Cape Town: University of Cape Town Press.

68.

Putnam

(2000) Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

69.

Putnam

(2007) E Pluribus Unum: Diversity and community in the twenty-first century. The 2006 Johan Skytte Prize lecture. Scandinavian Political Studies 30(2): 137–174.

70.

Robinson

(2013) The urban now: Theorising cities beyond the new. European Journal of Cultural Studies 16(6): 659–677.

71.

Robinson

(2015) Comparative urbanism: New geographies and cultures of theorizing the urban. International Journal of Urban and Regional Research 40(1): 187–198.

72.

Roy

(2016) What is urban about critical urban theory? Urban Geography 37(6): 810–823.

73.

Ryan

Sales

Tilki

et al . (2008) Social networks, social support and social capital: The experiences of recent Polish migrants to London. Sociology 42(4): 672–690.

74.

Saunders

(2011) Arrival City: The Final Migration and Our Next World. New York: Knopf.

75.

Schiller

Caglar

Guldbrandsen

(2006) Beyond the ethnic lens: Locality, globality and born-again incorporation. American Ethnologist 33(4): 612–633.

76.

Simone

(2004) People as infrastructure: Intersecting fragments in Johannesburg. Public Culture 16(3): 407–429.

77.

Singer

Brettell

Hardwick

(eds) (2008) Twenty-First Century Gateways: Immigrant Incorporation in Suburban America. Washington, DC: Brookings Institution Press.

78.

Sommers

(2011) Fear in Bongoland: Burundi Refugees in Urban Tanzania. New York: Berghahn Books.

79.

Trefon

(2009) Hinges and fringes: Conceptualising the peri-central Africa rrban. In: Locatelli

Nugent

(eds) African Cities: Competing Claims over Urban Spaces. Leiden: Brill, pp. 15–35.

80.

UN-DESA, Population Division (2010) World Urbanization Prospects: The 2009 Revision. New York: United Nations.

81.

UN-DESA, Population Division (2013) World Population Prospects: The 2012 Revision. Available at: http://esa.un.org/unpd/wpp/index.htm.

82.

United Nations Human Settlements Programme (UN Habitat) (2008) The State of African Cities 2008: A Framework for Addressing Urban Challenges in Africa. Nairobi: UN Habitat.

83.

United Nations Human Settlements Programme (UN Habitat) (2010) State of the World’s Cities 2008/9: Harmonious Cities. London: Earthscan.

84.

United Nations Human Settlements Programme (UN Habitat) (2011) State of the World’s Cities 2010/2011: Bridging the Urban Divide. Nairobi: UN Habitat.

85.

United Nations Population Fund (UNFPA) (2008) State of the World Population 2007. New York: UNFPA.

86.

Walsh

(2007) ‘It got very debauched, very Dubai!’ Heterosexual intimacy amongst single British expatriates. Social and Cultural Geography 8: 507–533.

87.

Watson

(2008) Seeing from the South: Refocusing urban planning on the globe’s central urban issues. Urban Studies 46(11): 2259–2275.

88.

White

(1990) The Comforts of Home: Prostitution in Colonial Nairobi. Chicago, IL: University of Chicago Press.

89.

Worby

(2010) Address unknown: The temporality of displacement and the ethics of disconnection among Zimbabwean migrants in Johannesburg. Journal of Southern African Studies 36(2): 417–431.

90.

World Bank (2005) Global Economic Prospects 2006: Economic Implications of Remittances and Migration. Washington, DC: World Bank.