DEPENDENCE, UNFREEDOM AND SLAVERY IN AFRICA: TOWARDS AN INTEGRATED ANALYSIS

The three books reviewed in this article seek to provide interpretations of dependence, unfreedom and slavery in African societies. But they reach different conclusions; bring different methodological frameworks to bear on the circumstances they examine; and – when they are concerned with policy questions – propose different remedies. A comparison of these books is useful not only for understanding African dependence and unfreedom, but also for rethinking critically the approaches of some of the main contemporary strands of research on these phenomena.

THE RISE OF CASH TRANSFERS

James Ferguson's new book, Give a Man a Fish, is an original contribution by a scholar who distinguishes himself by his ability to think outside the box. Give a Man a Fish explores recent reconfigurations in policy approaches aimed at poverty alleviation. It urges readers to acknowledge the significance of the ‘cash transfer revolution’ (p. 12) and develop ‘new ways of reasoning about matters of poverty and distribution’ (p. 10).

Ferguson starts with a plea for reconsideration: those who still decry the decline of welfare approaches as a consequence of unfettered free-market capitalism should rethink their views. Recent decades have witnessed the rising popularity of public programmes that ‘directly transfer small amounts of cash to large numbers of low-income people’ (p. 1). In South Africa, the years since 1994 have seen continuous growth in the numbers of recipients of social assistance programmes (as well as in the quantity of funds transferred).

Ferguson gives South African government figures showing that social assistance programmes at the time of his writing were paying grants to nearly 15 million South Africans, or about 30 per cent of the population. Of these, 9 million received child support grants, while most of the rest received old age pensions and disability grants. In 2010–11, the government spent 3.5 per cent of the gross national product on such grants, with nearly 60 per cent of all households in some of South Africa's poorest provinces, such as Limpopo and the Eastern Cape, receiving one or more grants of this kind (p. 77). Other Southern African countries – Namibia, Botswana, Swaziland and Lesotho – have begun to introduce similar programmes (pp. 6–7).

Recent assessments suggest that these programmes ‘work’. Studies cited by Ferguson document a reduction in the experience of hunger, and positive nutritional, educational and health outcomes in areas targeted by social payments (pp. 7–8). Based upon these preliminary evaluations, in recent years some campaigners in South Africa and Namibia have become promoters of a more ambitious initiative, the ‘basic income grant’ (BIG), which – if implemented – would provide a small monthly cash payment to every individual citizen (p. 17). At the time of the initial proposal, the amount proposed was about $16 per month. BIG would be non-contributory – that is, grants would not be based on the recipients' contributions in the form of recurrent payments or deductions from their incomes. They would be issued to all citizens, with the better-off having their $16 recouped through progressive taxation.

Ferguson takes the rise of cash transfers as an alternative to older forms of aid and increasing support for BIG as an opportunity to rethink issues of unemployment, poverty and inequality in contemporary African societies, with a particular focus on Southern Africa. In spite of repeated expressions of caution (for example, p. 32, 188), he appears to be supporting cash transfers, BIG, and the philosophy that he attributes to them. He welcomes what he sees as their alleged potential to bypass logics rooted in gender and racial stereotypes. Universal cash payments of the sort foreseen by the advocates of BIG break away from prejudices underpinning what Ferguson calls the ‘ideology of familism’ (p. 71).

‘Familism’ emphasizes the centrality of a ‘male breadwinner’ for whom joblessness or dependence on social payments is supposedly much more shameful than for women, children and the elderly. By contrast, dependence is what defines the nature of the relation of these latter groups to adult men or – in the absence of a ‘male breadwinner’ – to the state. But with a large proportion of young men permanently jobless and women and elderly persons receiving cash transfers, the idea of these groups' dependence on the male breadwinner is rendered obsolete by its lack of correspondence with people's lived experience (pp. 17–18, 81). Basic income grants would avoid biasing poverty alleviation with implicit assumptions about the gender or status of recipients. They also appeal to a new political philosophy holding that ‘wealth, being the product of social labour, social suffering, and social innovation, should be in some way shared by society as a whole’ (p. 55 and Chapter 6).

Furthermore, cash transfers appear to provide practical solutions to seemingly unsolvable problems, in particular society's apparent inability to ‘create jobs’ due to the ‘massive contemporary oversupply of manual labour’ (p. 80). In earlier periods, when land was plentiful and labour was scarce in Africa, colonial capitalists exploited regions that were considered ‘labour reserves’ by employing low-wage migrant labour (for example in mining; see p. 10). Exploitation extended beyond migrant workers themselves, and also targeted their communities of origin; these communities supported the costs of the social reproduction of labour, for example by raising children and caring for the sick and aged (pp. 10–11). But today the poor in South African rural peripheries are the potential recipients of grants that supposedly could reach them even in the most remote regions. By contrast, labour migrants now often migrate to depend, rather than to work.

Today … a restructured capitalism has ever less need for the ready supply of low-wage, low-skilled laborers that the migrant labor system generated … This point was made to me most dramatically in the comment of a South African social researcher with long experience working with poor rural communities. ‘I wish it weren't true,’ he said, ‘but the fact is that there are at least ten million people out there who could drop dead tomorrow and the JSE [Johannesburg Stock Exchange] wouldn't register so much as a ripple.’ (p. 11)

But it is one thing to acknowledge that labour-saving innovations reduced demand for labour and increased forms of precarious employment (Benanav 2010: 5). It is quite another to suggest that workers deemed redundant should just accept their condition rather than struggle to obtain ever more elusive jobs. At some points, Ferguson seems to suggest that cash transfers should make it easier for marginalized groups to develop secure occupations. But as one reads on, it is often suggested that the normative expectation of near-universal employment should simply be abandoned; that having a job should not be perceived as the only way to achieve a viable livelihood; and that the centrality of ‘productive work’ should be discarded for an emphasis on ‘distributive work’. These ideas have recently gained currency in studies of labour in Africa and elsewhere (Weeks 2011; Barchiesi 2011), but here I will restrict my comments to Ferguson's book.

DEPENDENCE AND DISTRIBUTION

Ferguson's solution to the alleged unemployability of large sections of Africans is to redefine dependence and unemployment in more appealing ways. This solution is controversial, and not always convincing. He denounces Western stereotypes and the idea that ‘economic dependence is a threat to the integrity of the adult male citizen’ as ‘an idea with a long pedigree in Western thought’ (p. 44). He condemns Marx's supposed overemphasis on production, or productive labour, and his dismissal of distributive processes (pp. 44–5, 91–3). By contrast, he encourages readers to revalue what he calls ‘distributive labor’:

Even the unproductive are both engaged in labor and entitled to social membership. The woman begging on the street may not be producing a good or selling a service, but she works all day. To say that she is not productive is not to denigrate her labor. It is, instead, to underline the importance and value of distribution. (p. 100)

Ferguson provides four examples of ‘survivalist improvisatory labor’: a windshield washer, a panhandler, a pickpocket, and a mother making a family visit. What these apparently different activities have in common, he argues, is that they ‘involve people engaged in a form of labor, one that seeks to secure a transfer of resources from those who have them to those who don't’ (p. 101). Ferguson thinks that South Africa is a good place to study ‘distributive livelihood strategies’, because such strategies are deeply rooted in its political economy. Southern African rural-to-urban migrants used to redistribute in the countryside some of the benefits of their wages. Today, the prospects of finding waged jobs are substantially reduced (p. 104). People continue to migrate, but they often strive not to find a job that will enable them to share their wages with relatives in the countryside, but rather to move in with relatives in cities, who can be persuaded to share their income with visiting migrants (p. 108).

The growth of social assistance programmes introduced new forms of mobility. Ferguson quotes Jeremy Seekings' study of the strategic decisions of households when choosing how to make members move: children may be moved between households ‘to care for an elderly grandparent at the same time as benefiting from the grandparent's access to a pension income’ (Seekings 2008: 43, cited in Ferguson, p. 109). One wonders how these strategic ‘distributive’ movements are experienced by those involved in them: the migrant who cannot ‘move to work’ anymore and now ‘moves to depend’; the child who is moved to care for elderly grandparents and thereby facilitates income distribution. But rather than attending to the experiences of people, Ferguson builds a culturalist argument about Africa's ‘deep social logic’: ‘distributive livelihood strategies are not simply a product of poverty and deprivation, but instead rest upon a deep social logic that finds application at all social levels’ (p. 115).

Relabelled ‘distributive livelihood strategies’, joblessness and precarious employment can be reimagined as good for Africans, or South Africans, whose social institutions supposedly have an atavistic propensity towards distribution and dependence. The risk is that exalting the positive values of ‘distribution’ conceals realities where people simply have no alternatives to begging, stealing, having sex with ‘sugar daddies’, or depending upon more fortunate relatives or patrons. Relabelling these activities ‘distributive labour’ is a rhetorical device that makes them sound more acceptable. But are the subjects of Ferguson's research happy to beg to live just because such activity can be reimagined as ‘pressing a distributive claim’ (p. 101)?

Ferguson knows that the argument that Africans are different, that they somehow ‘like’ distribution and dependence, is essentialist. Therefore, he tries to present distribution as valuable in general – and not only as culturally attuned to South African institutions. While he concedes that not all distributive labour is commendable, he concludes that – broadly speaking – ‘distribution (especially in a spectacularly unequal society) is a necessary and valuable social function, and it should be recognized, named, and valued as such’ (p. 101).

As I understand it, Ferguson's argument can be summarized as follows: ‘spectacularly unequal societies’ should embrace distributive livelihoods that entail a re-evaluation of social dependence, because dependence makes possible resource transfers that do not follow productive criteria, but rather inhere in relations of a patron–client type. These relations can be seen as desirable because they appear to solve the intractable problem of ‘creating jobs’ that are ‘not needed’. When dependence and distribution are revalued as institutions that contribute to the public good, the state can support them with small cash transfers and a citizen's income (the BIG option). Such grants would enable the unemployed to go on living and pressing ‘distributive claims’ on those who are employed or otherwise better off than them.

But who decides which jobs are ‘needed’ and which ones aren't? This is a fundamental question. Perhaps capitalist enterprises do not need to employ ‘surplus labour’ and can achieve profits without it. But travelling in most African countries reveals inadequate public infrastructure. In urban and rural contexts alike, there is an enormous unmet need for public works that would contribute to environmental rehabilitation, improved hygiene and sanitation, and the maintenance of roads and buildings. The density of physicians (per 1,000 people) is exceedingly low in most African countries. There are many sectors and services that would benefit if more workers were employed in them. These sectors could absorb large cohorts of supposedly ‘unemployable’ Africans and have important impacts on the quality of life.

Perhaps these ideas will remind some readers of old, mostly flawed, recipes for Africa's ‘modernization’ and ‘development’. This is not what I am suggesting. Colonial efforts to develop labour-intensive public works schemes rarely provided wages – let alone decent wages – for the workers involved. Until the mid-1940s, the colonial administration frequently employed forced and unfree labour in the construction and maintenance of colonial infrastructure (Cooper 1996; Fall 1993). In many independent African nations, postcolonial governments reproduced colonial labour practices. Developmentalist discourses legitimized the mobilization of African labour under headings such as ‘human investment’, ‘participation’, ‘community development’ and ‘ownership building’. These concepts, introduced under colonialism and re-proposed with recurrent cosmetic shifts until the present day, conceal the continuing exploitation of their so-called ‘beneficiaries’. The point is not to replicate policies that never created wage labour opportunities because they never intended to. The point is that in many African countries there might be potential for actual employment creation. Whether this potential can be realized is a political question.

Ferguson's view that capitalism has ‘ever less need’ for an ‘oversupply’ of labour suggests that the demand and supply of labour are determined by overarching forces, independent of capital, that Give a Man a Fish leaves unspecified and unexplored. But the fact is that capital can and does influence the supply of labour. Capitalist businesses are sensitive to the same incentives: rises in competition requiring cost reductions if a loss of profits is to be avoided. As shown by Tom Brass (2013), one of capital's strategies to maximize profit (or avoid losses or failure) consists in employing cheaper and easily controllable workers. It does this by influencing the relative proportion of free and unfree workers. The closure of underperforming businesses results in a mass of unemployed workers who may at some point accept unfree labour conditions or force relatives to accept them. Alternatively, they may migrate to countries where more opportunities are available, but where their migrant status exposes them to the extortions of local employers and a corrupt police. In turn, a large supply of workers prepared to accept unfavourable conditions limits the capacity of free workers to bargain with employers and resist exploitation.

A well-documented example of this type of process is the implantation of liberalization and structural adjustment policies in Africa, which resulted in privatization and rises in unemployment. International policy organizations support the logic of capital; the notion of the ‘informal economy’, introduced by researchers and mobilized by international policy organizations, depoliticized this process by labelling it ‘informality’ (Rossi 2014). But ‘informality’ conceals the connection between job losses, deproletarianization, and the growth of a large pool of easily exploitable workers.

The idea that the unemployability of large numbers of people is simply inevitable implies that precarious or unfree workers should accept their destiny. But unemployability is not destiny. Specific politico-economic dynamics regulate the proportions of employable and so-called ‘unemployable’ people. Such dynamics work to the advantage of employers, who can increase profits, or cut costs, by replacing free workers with unfree ones: that is, by replacing workers who retain the capacity to sell, withdraw, and resell their labour power with workers who are declared ‘unemployable’, have lost such capacity, and have no choice but to accept conditions imposed by employers or ‘patrons’ (Brass 2013: 44–5). A patron can be a gangmaster or a wealthy host for workers who ‘migrate to depend’. Exploitative conditions (whether experienced as coercion or euphemized through patronizing idioms such as loyalty, obligation or respect) are accepted by people who lack better opportunities. We need to ask questions about who extracts labour from whom, about the nature of production and employment relations, how property relations are maintained, and who profits from existing circumstances.

IS DEPENDENCE GOOD FOR AFRICANS?

In Chapter 5, the reader is encouraged to discard yet another bias of the ‘emancipatory liberal mind’ (p. 145): the tendency to feel uncomfortable before the ‘spectacle of people openly pursuing a subordinate and dependent status’ (p. 143). Two examples are provided. The first is the apparent paradox of people from neighbouring chiefdoms who, in the 1820s, voluntarily surrendered to the expansive Ngoni polity and came ‘of their own volition from great distances specifically with the goal of being taken captive by the Ngoni’ (p. 142). The other example is a contemporary anecdote based on the experience of an American friend of Ferguson, who, having recently arrived in Johannesburg, found troubling the constant storm of young men trying to be employed by him: ‘the message sent by poor black South Africans to this white, foreign, would-be egalitarian seemed to be “Let me serve you! Be my boss! Exploit me!”’ (pp. 141–2). Yet, we are told, these circumstances will appear less puzzling when one tries to understand their underlying ‘social logic’ (p. 144).

The reward for being incorporated as dependants is that dependence is a necessary aspect of belonging, and in Africa ‘belonging’ is allegedly the main avenue to full personhood. While Ngoni captives entered a society ‘at the bottom of the pecking order … with new captives constantly entering the system, they could quickly end up founding new segments of their own and acquiring social and political influence’ (p. 144). In recent decades, rising unemployment could be seen to have introduced a break in the notion of ‘wealth in people’ characteristic of contexts where labour is in scarce supply. But instead Ferguson suggests that:

those rendered ‘independent’ of the wage labour system do not remain happily independent but rather seek (with more or less success) to build up new dependencies. This, of course, was what my American acquaintance was observing. (p. 153)

Ferguson provides many historical, ethnographic and anecdotal examples, which create the appearance of an empirically grounded analysis. But his discussion of ‘dependence’ is too abstract (cf. Bolt 2013). A plethora of mini-sketches from different contexts and periods does not provide sufficient information to assess whether the situations being compared are indeed comparable beyond the minimal condition that the term ‘dependence’ appears in all of them. At such an abstract level, dependence means everything and nothing.

‘Dependence’ means different things in different contexts: in some cases it means horrific dehumanization, as when slaves are worked to death to fulfil their owners' desires. In other contexts, dependent relations can and do provide security, protection and opportunities. Which contexts make different degrees of dependence acceptable? To whom is dependence acceptable? And to whom is it not? The idea advanced in the passage quoted above – that those who lost their jobs (or never got one) could remain ‘happily independent’ – says nothing about what such ‘happy independence’ entails: hunger, disease and the threat of death are conditions that make ‘dependence’ – however defined – a palatable option. This does not mean that if better opportunities were accessible ‘dependence’ would retain its appeal.

Captives enslaving themselves in precolonial wars, young unemployed men trying to earn money by offering their services to tourists, beggars, and mothers or grandmothers receiving child support grants from the state all partake in different relations. A critical assessment of these relations would require an in-depth analysis of the conditions in which people with different opportunities entered these forms of dependence, which supposedly improved their initial circumstances. Ferguson asks: ‘does this mean dependence is actually a good thing?’ (p. 155). Ferguson's answer that ‘such a claim makes us uneasy’ (ibid.) leads him into the trap of blaming ethnocentricity (beware the emancipatory liberal mind!) when in fact the only possible answer to this question is ‘It depends’. It depends on what forms of dependence and what types of relations.

After revaluing dependence as a source of both protection and social membership, the fear of some critics that BIG might promote dependence on the state will appear trivial. It is not only, we are told, that the poor would be dependent anyway (p. 157); it is also that, if we are prepared to shed the ‘moralizing attitudes that stigmatize those excluded from the labor market’ (p. 163), we can think of BIG not as a humiliating handout but as a ‘citizen's income’ that recognizes ‘various forms of dependence (including care giving and care receiving) as necessary building blocks of a healthy society’ (p. 163). The conclusion, then, is that – especially in (Southern) African contexts – one should hope that new approaches in social policy may open up ‘new ways of approaching the question of dependence in a more rewarding way, a way that would be able to credit and respect the vernacular aspirations to social relationality that is so puzzling to an emancipatory liberalism’ (p. 164, my italics).

Seeing dependence as the fruit of ‘vernacular aspirations to social relationality’ is a rhetorical artifice that does not clarify the actual (experiential) causes of a resilient attachment to dependence among certain (but not all) African subjects, and that fails to explain the many circumstances in which dependence is in fact rejected by real-life individuals who are in a position to do so.

Ferguson mentions the introduction by Kopytoff and Miers to their edited volume Slavery in Africa (1977), but only cursorily in relation to the concept of ‘wealth in people’ (p. 145). It is surprising that Ferguson does not engage their elaborate discussion of dependence and belonging in African societies, for it would be helpful to see him explain where he agrees, and where he disagrees, with these authors' influential ideas. Kopytoff and Miers famously contrasted a supposedly ‘Western’ valorization of freedom and independence and a supposedly African valorization of dependence qua belonging. This recalls Ferguson's opposition of a Western ‘emancipatory liberal mind’ and African ‘vernacular aspirations to social relationality’.

[T]he insider in most traditional societies in Africa was not an autonomous individual. His full citizenship derived from belonging to a kin group, usually corporate, which was the fundamental social, legal, political, and ritual protective unit. This contrasts with the modern Western ideology of ‘freedom’ (though not exactly with Western sociological reality). For in the Western conception, the antithesis of ‘slavery’ is ‘freedom’, and ‘freedom’ means autonomy and a lack of social bonds. However … [i]n most African societies, ‘freedom’ lay not in a withdrawal into a meaningless and dangerous autonomy but in attachment to a kin group, to a patron – an attachment that occurred within a well-defined hierarchical framework … Here, the antithesis of ‘slavery’ is not ‘freedom’ qua autonomy but rather ‘belonging’. (Kopytoff and Miers 1977: 17)

This particular argument in their introduction has stimulated substantial discussion over the last forty years in the historiography of slavery in Africa. The view that Africans do not desire freedom ‘in the Western sense of the term’ has been harshly criticized. Frederick Cooper argued convincingly that ‘absorptionist’ interpretations of African slavery are culturalist and ahistorical: by making absorption – and its corollaries, rights in people and dependence – the ultimate aim of ‘African’ societies, Kopytoff and Miers built a functionalist argument that fails to account for the specific aims and strategies of individual slaves and slave owners (Cooper 1979). Whenever the cutting of ties of dependence was possible and led to a foreseeable improvement in the slaves' conditions, enslaved Africans deserted their owners, resulting in massive exoduses (see, for example, Klein and Roberts 1980).

In Africa, we are told, independence is a ‘terrifying predicament given the importance … of relationality for personhood’ (Ferguson, p. 152). But just as independence is only terrifying under specific circumstances (to Africans and non-Africans alike), so, too, not all forms of ‘relationality’ are equally desirable. An administrator commenting on demands for liberation by slaves in Mellacorée on the Guinea Coast and cited by Martin Klein reported: ‘They say that they are tired of working for their master, that they have worked long enough for others, and that the liberty they seek is so that they can work for themselves and reap the fruits of their own labor’ (cited in Klein 1988: 209). Klein concludes the chapter on the Banamba exodus in his monograph on slavery and colonial rule in French West Africa by noting laconically that: ‘the exodus took place because slaves were willing to risk intimidation, hunger and hardship to return home or seek freedom elsewhere’ (Klein 1998: 177). In the conclusion of his classic manual on the history of African slavery, Paul Lovejoy sums up: ‘Many slaves knew exactly what they wanted, and it was not to “belong” to their master … The aim of slaves was freedom, not the modification of the conditions of slavery’ (Lovejoy 2000 [1983]: 253).

THEORIZING AFRICAN SLAVERY

Sean Stilwell's book, Slavery and Slaving in African History, reassesses some of the main debates in the historiography of African slavery. It starts and ends with ‘freedom’. Chapter 1 revisits Kopytoff and Miers and presents some of the main objections that were advanced against their argument by Claude Meillassoux and Paul Lovejoy (pp. 7–10). It provides a helpful overview of the main debates. In essence, Stilwell reiterates the Kopytoff–Miers position with the ‘slave’ distinguished slightly more markedly from the ‘free’ than in their model of the slavery-to-kinship continuum (p. 14). He shows that in many African societies the slaves' kinlessness and total marginality as outsiders did not last long (p. 19ff.). As slaves were progressively integrated into their masters' families and networks, they started belonging in corporate units of society and thereby acquiring some rights.

One problem with this argument is the claim that the emphasis on belonging as an avenue to full personhood (to the point that dependence appears attractive) is a peculiarly ‘African’ idea. Stilwell states repeatedly that ‘in most African social systems, belonging mattered. Africans could belong to numerous institutions or corporate groups' (p. 8). This approach to African slavery is culturalist. In Africa, as elsewhere, belonging to corporate institutions (primarily the family, but also other institutions such as religious groups, clubs or political parties) provides protection, but at the same time imposes demands on members. Hierarchical institutions impose a degree of subordination upon more powerful persons whose support is fundamental if membership – with the benefits that derive from it – is to be retained. Are slaves in continuum with kin ‘in Africa’? This is not a historical question, and conceptually it is a non-question: they are or they aren't, depending on the kinds of ‘continuities’ one wishes to bring into focus. That both slaves and kin experience ‘dependence’ to various degrees is an unfalsifiable statement, but its heuristic potential is close to zero.

The claim that this becomes an interesting statement for comparative purposes, because ‘African personhood’ (whatever this is) is based on belonging, ‘social relationality’ and the dependence that follows, is an untenable essentialization. Stilwell knows this, and indeed his book carefully avoids overgeneralizations. Slavery and Slaving begins with examples of different African enslaved individuals whose experiences are completely different. Stilwell explains that ‘these different experiences illustrate just how difficult it is to discuss slavery across a continent as large and complex as Africa. Slavery in Africa was diverse’ (p. 4). So why force it into an axiomatic definition that reduces ‘African slavery’ across time and space to the opposite of belonging, because of supposedly ‘African’ aspirations to social relationality? Due to its conceptual elegance, the Kopytoff–Miers argument is seductive. But the historiography of slavery in Africa has moved on: careful regional studies have shown that it is a reductionist essentialization of African slavery. Ferguson would have benefited from examining this literature. Stilwell, who knows the literature, is too kind to Kopytoff and Miers, even when his own approach – and that of other historians of slavery whom he cites – has the potential to overcome some of the weaknesses of their classic work.

Slavery and Slaving emphasizes the individual experience of enslavement in different African contexts. This is clear from the outset: the book's introduction and all following chapters begin with case studies that focus on individual biography. This level of detail has become possible in recent decades thanks to efforts by researchers to recover the voices of individual African slaves and to provide analyses of their experiences (Curtin 1967; Olivier de Sardan 1975; Romero 1988; Wright 1993; Lovejoy and Law 2003; Getz and Clarke 2011; Greene 2011; Hahonou and Strandsbjerg 2011; McDougall 1998; Rasmussen 1999; Bellagamba et al. 2013; 2016). This literature provides new and important insights into the experience of dependence, slavery and freedom.

By opening each chapter of his book with snapshots of individual experiences taken from recent studies of African slavery, Stilwell's textbook shares this literature's emphasis on the historical experience of enslaved persons as fully fledged individuals, not generic victims, numbers in a ship or caravan, or anonymous automata following the dictates of an ‘African’ culture of belonging. Analysis of these primary sources suggests that those who ‘chose dependence’ (with its distributive potential) did so only because they failed to access viable alternatives and become producers, traders or otherwise independent workers within specific labour relations and systems of production.

THE PRIMACY OF PRODUCTION

Gareth Austin has shown that the transition from slavery to freedom was facilitated by the development of cash crops (2009). Those who owned profitable land grew crops that earned them relatively high profits. Such profits were an incentive to hire workers and thereby increase the output of crops that had high market value: ‘In those parts of West Africa where the hiring of wage labour … became widespread in agriculture, notably in cocoa growing, this was made possible precisely by the profitability of export crops’ (Austin 2009: 36). By the same token, the disappearance of the conditions that had facilitated growth in particular cash-crop economies would potentially result in the return of unfree labour practices.

Falling prices or increased competition would require cost reductions on the part of employers unable to offer the same level of wages or hire the same number of workers. The same circumstances would also undermine the labourers' ability to bargain for higher wages. The most impoverished workers would accept lower pay and worsened employment conditions. Ultima ratio, people unable to own productive resources or find jobs at home or abroad would make a virtue of necessity and exalt the benefits of subservience. This suggests that distribution by dependence arises from dynamics located firmly in production. And production is shaped by politics: the presence or absence of a centralized state influences the structure of dependent relations in society.

Stilwell distinguishes African systems of slavery according to the relative demographic importance of slaves. In ‘high-density’ systems, slaves made up an important proportion of the population and played a central role in production, with many slaves living in separate agricultural estates. In ‘low-density’ systems, they often worked alongside their masters and shared living spaces with them. Stilwell suggests that this distinction is linked to the potential absorption of slaves in the society of the free: that is, to slaves' potential social mobility. In his view, high-density slavery tended to be more ‘closed’, and low-density slavery more ‘open’, because in the former slaves had fewer opportunities for becoming progressively integrated within their masters' families (p. 21). High-density closed-system slavery was more frequently found in conjunction with powerful states that had the military power to control large concentrations of slaves. Low-density open-system slavery was more often associated with decentralized societies, where slavery permitted the attachment of people who could not – at least temporarily – make the same claims as free kin (p. 67).

In the book's fifth chapter on ‘Slavery and African economies’, Stilwell discusses land–labour ratios in Africa (especially pp. 128–32). He avoids the deterministic fallacies of approaches that emphasize relative factor-endowment ratios as determinants of labour coercion by arguing that, in places where populations were large and land was relatively scarce, ‘the absolute density of population mattered less than the social organization of production and reproduction’ (p. 129). In other words, even when ‘Nieboer conditions’ (Austin 2008) ceased to obtain, retaining unfree labour was expedient for those actors who sought to maximize profit by exploiting the labour of vulnerable others.

Chapter 6 focuses on ‘the end of slavery’ in the twentieth century. It illustrates vividly the slaves' struggle for independence, which was primarily a struggle to gain access to productive resources and to move freely to areas where ex-slaves could access productive land and achieve higher returns for their work. For some persons and groups, access to property and paid employment was mediated through the renegotiation of dependence on masters-turned-patrons. Even in such contexts, dependence was not sought as an end in itself, but as an avenue to achieving greater control over one's life.

Stilwell reasserts what is by now one of the most consolidated findings of African slavery studies: that colonial abolitionism was conservative at best. Until the 1930s and 1940s, all European colonial administrations sought to limit the emancipatory impacts of legal abolition and introduced new forms of unfree labour that were sometimes considered harsher than precolonial slavery (p. 191). But slaves realized that the colonial legal and institutional framework undermined the power of the slave-owning classes and unfolded strategies of self-emancipation. Flight and migration were perhaps the most obvious rejections of dependence. As documented for Sierra Leone, Italian Somalia, Sudan, Niger, Northern Nigeria, Tanganyka, Kenya and Zanzibar (pp. 193–8), slave flights exemplified the slaves' refusal of their former subjection. Most of the slaves who left looked for land that they could clear and make their own. Slaves in pastoral societies struggled to control herds and areas of transhumance (Mauxion 2012).

Slaves ‘who stayed did so not because African slavery was automatically assimilative and benign, but because this specific historical period provided slaves with many more opportunities to expand their rights and to force their masters to ameliorate slavery’ (pp. 193–4). The most impoverished slaves remained in marginal social and economic positions until the present day, unable to access land or capital or to obtain secure jobs. When slaves continued to serve former masters, steep power inequalities were grounded in necessity: there were no better alternatives. Some slaves shifted hierarchies: they joined the army and Christian or Muslim religious orders, where their status was measured in terms of military valour or piety. Others used their relationships with former masters to gain access to land, brides and hospitality in their former owners' urban residences (Pelckmans 2012a; 2012b). They used what was now a patron–client relation to facilitate their transition to economic independence.

In South Africa, many of these dynamics took place at an earlier stage, in the first half of the nineteenth century. The nineteenth-century historiography of South African emancipation, some of which is cited by Stilwell (pp. 185–7), tells a story of successive transitions from slavery to a variety of similarly exploitative labour systems redefined in racialized terms that continue to shape present-day labour relations. Discrimination based on gender, as well as race, continues to affect the opportunities of women (pp. 205–7). Emancipation from slavery was a slower process for African women than for men, and pawning and trafficking in women and children survives today, as discussed in the next section.

As Stilwell shows, understanding these dynamics requires an analysis of production and reproduction: how is labour controlled, and who can extract profit from existing arrangements? A focus on ‘distribution’ cannot simply replace an analysis of production on the grounds that the primacy of production is ‘a stubborn idea’ of the productivist left (Ferguson, pp. 44–5). If Ferguson wishes to challenge Marx, to whom he attributes the view that ‘distribution … was explicitly secondary and derivative’ (p. 45), he has to show that production relations are not primary.

WHO BENEFITS FROM DEPENDENCE?

Belonging – and the dependence that belonging entails – matters in Africa. Or so we are told. But from the perspective of vulnerable youths perceived as expendable and pawnable, ‘belonging’ in their family did not avert an even worse form of ‘belonging’ as pawns. In the colonial and postcolonial periods, those who bought women and children tried to make slaves or pawns pass as wives and relatives (Lawrance and Roberts, p. 10). They hoped that authorities would automatically defer to ‘African tradition’ and would not challenge women's and children's subordination – or the profits that those who controlled their labour derived from that subordination. They probably hoped that colonial administrators would assume that ‘dependence is after all a good thing’ for (certain) Africans. This simplistic assumption has been, and continues to be, central to external interpretations of African society and labour. Perhaps interpreters such as Ferguson should be less worried about their supposedly ethnocentric propensity to feel ‘uneasy’ about the dependence of Africans, and should rather worry about their propensity to naturalize African dependence, and to accept it as a deep-seated African ‘social logic’.

The inclusivist tendencies discussed above influenced the historically attested African preference for female slaves over male slaves, which was reflected in the higher demand for, and higher prices of, female slaves (see Robertson and Klein 1997). The question of whether dependent women (and slave women in particular) were valued primarily for their sexual and reproductive functions, or if they were valued mainly as workers, generated substantial debate among historians of African slavery without reaching closure (see Klein 2014 for a recent reassessment). This question remains relevant in the light of the ‘persistence of demand’ for trafficked women and children in African societies today (Lawrance and Roberts, p. 9).

Contemporary African wars have resulted in widespread abductions and sexual enslavement of girls and women by African militias. At least in some cases, such as the case of Uganda's Lord's Resistance Army, the abduction of women was aimed at enabling the reproduction of the movement through the control of women's sexual and reproductive capabilities (Baines 2014; Apio 2016). Does this count as evidence of the continued relevance of ‘wealth in people’ in circumstances where the offspring of abducted women can be controlled more effectively than those of other women? How do women perceive their dependence?

The volume edited by Lawrance and Roberts shows that contemporary trafficking and enslavement are not ‘new’ phenomena, but the latest reconfigurations of forms of enforced dependence that have deep historical roots in African societies. They show, too, that the harshest forms of social dependence today (comparable to pre-abolition slavery) primarily affect African women and children. One section of the introduction is entitled ‘Traffic in dependents’ (pp. 8–10). It shows that those trafficked into slavery were often pawns, and sometimes the kin of those initiating their trafficking. Belonging, after all, did not always afford security to African dependants: different persons can turn ‘belonging’ into a means of protection (of family members, for example, or dependants whose loyalty one wishes to win) or, alternatively, of exploitation (of family members or dependants who are seen as expendable and/or ‘pawnable’). What matters is the specificity of the circumstances under which dependants are protected or exposed. Such circumstances are historically and socially shaped: when is it possible, profitable or necessary to use dependent and exploitable labour? Roberts emphasizes the issue of demand:

At the core of the problem of trafficking lies the issue of demand. Those at the end of the chain of acquisition of dependent and exploitable labor are interested in acquiring labor for a whole host of reasons, including enhancing patriarchy, power, and financial reward. (p. 12)

But ‘the issue of demand’ is not generically African: where there is a demand for profit, there is also a demand for exploitable labour. Such demand will be couched in a variety of cultural idioms, as those seeking dependants attempt to legitimize their acts as forms of patronage. The question, therefore, is not whether we can prove that dependence is a vital coping mechanism ‘in Africa’, but under which conditions – in specific social contexts – certain individuals can exploit the labour of dependants, and what mechanisms make dependants accept these circumstances.

Elisabeth McMahon's chapter explores the social vulnerability of women and children in nineteenth-century East Africa by looking at three cases taken from the 1880s and 1890s. She shows that women and children were the most vulnerable to being kidnapped and enslaved, often repeatedly (p. 32). This suggests that kidnappers targeted persons without reliable ‘male “protectors” (husbands, fathers, adult sons, owners)’ (p. 33). The slave's or the enslaveable's fear of being isolated (p. 34) demonstrates that belonging sometimes can indeed be essential if one is to escape the dangers inherent in isolation.

The case of Mia is telling. Mia had been captured on the mainland and put in an Arab dhow with other enslaved persons to be sold on Pemba Island, a part of the Zanzibar archipelago. A British navy vessel engaged the dhow and, following a skirmish, killed the slavers. Making it to shore, Mia and the other captives scattered and sought the help of local communities. Mia reached the town of Chake Chake and sought the help of the liwali, who gave her freedom but ‘committed her to the care’ of Ibrahim bin Madini (p. 38). Mia lived on the shamba (small farm) of Ibrahim for ten years, but eventually moved out on her own in 1893, when she rented a small plot of land from a woman called Binti Hamadi. Mia supported herself by making pots and growing vegetables. After three years, Amur bin Suleiman, Binti Hamadi's husband, surprised her in her hut, tied her up, and brought her to the shamba of a man named Masood bin Abdullah. After ten days in captivity, Mia escaped and went to the vice consul. Questioned by this official, Masood stated that he believed Mia to be one of his runaway slaves and had paid Amur to retrieve her. Amur denied kidnapping her. According to Mia, Amur knew she was a free woman, because he had made inquiries with Ibrahim when she tried to rent land from his wife. This inquiry had given him knowledge of Mia's story and potential vulnerability in Pemba, and he had taken advantage of it.

Mia had tried to live autonomously and support herself through the fruits of her small-scale business. But as an independent woman she was peculiarly vulnerable. It was not only that the sexual division of labour did not offer the same horizon of opportunities to men and women, making women less likely to employ dependent workers and achieve economic mobility. It was also that African and European gender ideologies on Pemba saw women's autonomy as an aberration. Given the economic risks implicit in autonomy and the ideological consistency of discourses that stigmatized autonomous women, the majority of women had few alternatives but to embrace dependence and concentrate their efforts on finding a benign patron.

The view that family or marriage relations were intrinsically benign and necessary for women's ‘protection’ meant that official representatives avoided adjudicating in ‘family affairs’. As shown by Marie Rodet in her chapter on Kayes in the French Soudan (today's Mali), ‘the administration, on the pretext that they did not wish to interfere in local customs, limited whenever possible their intervention in pawnship cases’ (p. 95). In what was perceived as ‘local custom’, the distinction between wife and slave was often blurred, as was that between slaves and pawns and fostered children (p. 94). These attitudes resulted in a lack of institutional support for the emancipation of many enslaved women, whose situations were deemed ‘indigenous marriage’ rather than lingering cases of slavery. Their dependence did not strike administrators as undesirable.

Rodet examines a large number of cases, most of which are based on court records from the period 1900–39. The picture that emerges is one of marginal young women and children enmeshed in layers of dependence. A case in point is that of a four-year-old girl who was entrusted to a tradeswoman from Nioro by her impoverished father. In a 1936 investigation, the tradeswoman showed the authorities a paper signed by the District Officer (Commandant de cercle) that was a declaration of marriage between the four-year-old girl and the woman's nephew, who was about fifty:

The inquiry concluded that the commandant had probably signed the paper without verifying the age of the ‘spouses.’ The administration entrusted the young girl, whose parents had died in the meantime, to a local notable, where she probably continued to work as a maid. (p. 95)

Part II of the volume includes contributions that focus on contemporary trafficking and anti-trafficking. Some of these chapters expose continuities with the historical case studies, in spite of differences in the legal framework. Susan Kreston's chapter tells the story of Elsie, a South African girl sold when she was about ten years old to a German man who exploited her sexually until she ran away. She soon entered a chain of relationships in which she was prostituted and abused by seven subsequent sex traffickers. Some may argue that at the origin of Elsie's sufferings is a deficit of ‘belonging’, as her mother did not want her and left her with her grandmother, upon whose death she went to live with her uncle and aunt who sold her to the German. To me this case demonstrates that ‘belonging’ per se does not do anything. Belonging is as belonging does.

One of the strengths of Lawrance and Roberts' volume is that it does not focus only on the ‘victim–perpetrator’ dyad, but also examines critically the position of self-fashioned ‘saviours’ and ‘freedom fighters’. In his probing analysis of the tropes of neo-abolitionist NGOs, Benjamin Lawrance discusses the deployment of parental testimony in a way that does not challenge the central assumption of NGOs that ‘African’ local communities and families collectively need rescuing (p. 174). Professionals who collect testimonies on behalf of human rights NGOs must represent African testimonies in ways that justify their mandate to salvage generic African ‘victims’ (p. 173). These discourses gloss over structural inequalities of opportunity between different categories of Africans, some of whom are more vulnerable than others to harsh forms of dependence and unfreedom. Instead of blaming ‘economic crisis’ for a generic ‘African poverty’, studies should focus on how particular institutions make certain categories of people more vulnerable than others.

Arguing that parents, children, spouses, masters, employers, employees, prostitutes or pimps take part in culturally shaped ‘aspirations to social relationality’ is a functionalist argument that does little to explain who benefits from certain types of relationships and institutions, who does not, and why. A person can be dependent in multiple fields: Mia, the anonymous four-year-old girl in colonial Kayes and Elsie were all marginalized by virtue of their age, gender and status. Dependence was not what they aspired to. It was their only option. Struggling for freedom would have been too dangerous for them; too many actors would have interpreted their struggle as the fault of a ‘rebellious’ temper unsuited to women of their status. Experience made them aware of the scarce support that any effort to seek autonomy on their part would elicit. The uncritical assumption that ‘perhaps dependence is a good thing’ in Africa risks undermining the potential of research to scrutinize critically different forms of dependence upon fathers/parents, husbands/partners, guardians, employers, patrons or state officials, as the case may be.

CONCLUSION

De facto if not de jure, unfree labour relations in Africa (and elsewhere) continue to be reproduced by several economic institutions – such as a lively regional, continental and global market in African women and children; and global labour regimes that benefit from the employment of workers (often migrants) who do not possess the protection of the law and are therefore hyper-exploitable and unfree, if not enslaved. Alternatives in the form of decent employment opportunities are narrow, and are likely to remain so if the ‘unemployability’ of Africans becomes an accepted mantra in academic and policy writings on this subject.

The idea that dependence ‘works’ for Africans is based on a misinterpretation of circumstances in which the most vulnerable actors have a limited range of realistic opportunities to avoid extreme destitution and exploitation. In these circumstances, patronage and dependence are desirable, and would be equally desirable to any representative of the human species. Neoliberal multinationals and other employers trying to maximize their profits at the cost of human dignity take advantage of the willingness of vulnerable people in order to obtain labour at the cheapest rates. Far from being more ‘humane’, so-called traditional institutions similarly take advantage of the vulnerability of women and children who are (often, but not always) denied alternatives due to sex or gender ideologies that define them as incapable of doing ‘a man's job’ or that give control over their sexual and reproductive potential to someone other than themselves.

Conditions of safety from extreme destitution and the threat of violence must exist if marginalized persons are to come forward and claim autonomy. If such persons do not, it is not because they have an intrinsic preference for dependence. It is because the institutional and cultural landscape in which they live has in-built barriers to their emancipation. There is a risk that cash transfers and BIG would facilitate the structural reproduction of exploitation and unfreedom by enabling the reproduction of a large reserve of exploitable labour declared ‘unemployable’ and forced to accept exploitative relations to secure mere survival. But it is also possible that such policies, decoupled from the ‘unemployability’ axiom, would afford protection to vulnerable groups and provide them with incentives to negotiate better working and living conditions. Assessing whether the latter scenario is at all realistic would require an inquiry into the economic and political feasibility of an expansion of social welfare in different African countries. The lesson from the historiography of African slavery and emancipation is that optimism is unwarranted. Abolitionist governments, be they colonial or independent, refrained from supporting marginal ex-slaves whenever facilitating their emancipation through redistributive policies would have been detrimental to the interests of political and economic elites.