For perfectly understandable reasons, much writing about contemporary Africa has focused on instances where there has been a partial or complete breakdown of central authority—as was true of Liberia and Sierra Leone in the 1990s or is the case in Somalia today—or where predatory rulers holed up in capital cities have lived off the rents derived from oil and mining enclaves: for example Chad, Congo-Brazzaville or the Democratic Republic of Congo (drc).footnote1 Accounting for the more extreme configurations is a necessary exercise, as is the effort to explain how a degree of normality is possible in the absence of a functioning state.footnote2 What is problematic is when these cases are made to stand for Africa as a whole. There has been a trend in recent scholarship to posit a pattern common to pretty much all African countries, with the possible exception of Mauritius, Botswana and South Africa: namely endemic levels of clientelism, which turn all government institutions into ciphers of particularistic interests and have an in-built dynamic towards chronic instability.footnote3

Accompanying this inclination to flatten the African landscape, there has been a movement away from investigating institutions towards a narrower concern with how networks function. Indeed, many contributions that are ostensibly about the state are really dealing with something entirely different. For example, Jean-François Bayart’s The State in Africa has a great deal to say about conceptions of power, but it tells us surprisingly little about the state in Africa—at least in an institutional sense.footnote4 If pushed, the authors in question would no doubt insist that the state is so impregnated with societal interests that it can only be studied as embedded within society rather than being artificially separated from it. In this vein, Patrick Chabal and Jean-Pascal Daloz conclude:

Hence, the notion that [African] politicians, bureaucrats or military chiefs should be the servants of the state does not make sense. Their political obligations are, first and foremost, to their kith and kin, their clients, their communities, their regions, or even to their religion. All such patrons seek ideally to constitute themselves as ‘Big Men’, controlling as many networks as they can . . . We are thus led to conclude that, in most African countries, the state is no more than a décor, a pseudo-Western façade masking the realities of deeply personalized political realities.footnote5

In a subsequent collaborative work, they reiterate that the existence of the state is not something inevitable or normal, but grew out of a very particular European history. To this day, they maintain, the formal existence of a state in somewhere like Nigeria is confounded by the reality of quite different logics at play.footnote6 Forget trying to apply Western concepts to African realities, we are told, and focus on what is actually happening.

This may sound like wise advice, but the reality is that the amount of research which actually traces how networks function—rather than merely asserting their existence—is surprisingly limited outside of the cases of the oil-rich states. The most recent work on the state has (significantly perhaps) been carried out by anthropologists rather than by political scientists, and often tends to support the view that institutions remain rather important.footnote7 Despite the realities of everyday corruption and chronic inefficiency, there is a surprisingly deep-seated attachment to bureaucratic rules and behaviour.footnote8 These may be honoured in the breach, but they remain a constant frame of reference for the actors concerned, and not just because they can be manipulated for personal gain. The most interesting set of questions is how institutional structures are moulded to deal with everyday realities, such as the scarcity of manpower and logistics (including uniforms, petrol, vehicles and telephones) and the inadequacy of official remuneration. The underlying message here is that institutions in Africa are not elaborate fictions or a cover for something else, but help to inform the behaviour of official and non-state actors alike in fundamental ways. Ticking off African states against a checklist of criteria derived from an ideal-type, and then finding them wanting, may create the appearance of instilling greater analytical rigour, but in reality it closes down an analysis of how institutions actually work.

‘Bringing the state back in’ has become a cliché, but undoubtedly there is a need to integrate an analysis of social dynamics with a closer examination of African state logics. This entails a number of overlapping agendas. Firstly, there is a need to return to a sustained study of political phenomena at different levels simultaneously—that is, local, regional, national and indeed trans-national. Although the first two used to be the preserve of a rather conventional American political science, a multi-level approach is fundamental to any attempt at understanding African politics. Secondly, the role of the state in mediating the production and reproduction of social inequalities—which has become lost in a rather mushy literature about neo-patrimonialism—should receive far greater attention than it presently does. Finally, there is a need for a comparative approach that does not conflate experiences, but opens up the possibility of understanding a range of phenomena across African countries as well as a mechanism for understanding the salient differences. In what follows, I wish to demonstrate, primarily through a comparison of four West African countries—Ghana, Togo, Senegal and the Gambia—how a return to political economy can help to create a more complex picture of how African states and peoples have engaged with one another. These examples cannot stand for Africa as a whole, but they do provide an insight into the workings of countries where the state has not collapsed or been skewed by the flow of oil rents. The article falls into four parts. In the first, I briefly sketch out the existing landscape of writing on African states in order to be able to better position my own interpretation. I then turn to a comparison of colonial states and their enduring legacies. In the third section, I examine the reasons why many of the social contracts became untenable in the last decades of the twentieth century. Finally, I deal with the potential for the emergence of new social contracts in contemporary Africa.

The central question that underlies most of the literature on African states is why they have proved to be such weak Leviathans or, phrased in more normative terms, why they have failed to generate meaningful public goods. The answer is typically sought in some combination of historical and structural factors. On the resolutely historical side of things, Jean-François Bayart seeks an overarching explanation in a history of African ‘extraversion’ dating back to the era of the slave trade, which warped African institutions in fundamental ways.footnote9 Whereas his account takes the long view, many still continue to insist on colonialism as the operative watershed. Hence it has been argued, in a stronger and a weaker version by Basil Davidson and Patrick Chabal respectively, that pre-colonial traditions of statecraft were fractured during the colonial takeover.footnote10 The result was that imported institutions lacked basic legitimacy, while the indigenous forms that were permitted to continue were stripped of their mechanisms of accountability—coming to embody what Mahmood Mamdani has described as ‘decentralized despotisms’.footnote11