Navegação – Mapa do site
Pensamento social sobre o Político

Bringing the Self Back In: Politics and Accountability in Africa

Tim Kelsall
p. 133-157


Este artigo argumenta que a democracia e a boa governação em África são actualmente enfraquecidas pela comparativa ausência de actores colectivos capazes de exigir prestações de contas ao Estado. Esta fraqueza tem a sua origem na construção social da selfhood, ou subjectividade, prevalecente no continente e que geralmente assume formas fragmentadas ou contraditórias. Por sua vez, este fenómeno é visto como ligado à natureza articulada da formação social em África. A base material do selfhood fragmentado e da fraca acção colectiva é um aspecto ao qual os estudos liberais, radicais e pós-modernos sobre África prestam uma atenção insuficiente.

Topo da página

Texto integral

«The human essence is no abstraction
inherent in each single individual.
In its reality it is the ensemble
of the social relations».
Karl Marx (1977: 157)


1In recent years the idea of accountability has risen to the top of the political agenda in Africa. Partly because of its newly presumed link with development, donors and scholars devote to it an increasing amount of concern. But it is also an idea that seems capable of drawing support not just from developmentalists, but from across the ideological spectrum: accounting for one’s actions to oneself and one’s peers, seems a part of what it is to be human; most people would agree that a good society was one in which leaders were accountable to those they led.

2Given this pre-occupation with accountability, it is perhaps surprising how little candid discussion there has been in African studies on the question of what sort of human arrangements, and what sorts of arrangements of humans, are conducive to accountability in the political and economic sphere. On the right, liberals either delude themselves into thinking that Africa is already the home of homo oeconomicus, or else they devise elaborate projects for the transformation of people; but rarely are they happy to admit as much. Much of the writing on the Left meanwhile clings to an implicit assumption about the historical progressiveness of «popular» classes, and pays inadequate attention to the exceedingly rare conditions in which ordinary people actually ever succeed in taking control of their own lives. Meanwhile an anthropologically informed and theoretically eclectic school of scholars, contents itself with writing about what is there in Africa, eschewing a more constructive role.

3The argument of this paper, then, is that serious discussion of the issue of accountability is occluded in African studies because of an inadequate attention to the way in which the self is fabricated in social relations. It will attempt to demonstrate these points by critiquing the work of a selection of scholars taken as emblems of Right and Left scholarship on Africa. It then presents an analytical argument sketching the relationship of the self to social relations, the relationship of the self to collective action, and the relationship of collective action to political accountability. It ends by suggesting some avenues for future research.

Liberal Projects

4In the corridors of contemporary African treasuries, neo-liberal economics is in the ascendancy, and for the best part of two decades African populations have been subjected to its nostrums. Neo-classical economic theory has also provided the inspiration for the «new political economy» of Robert Bates, whose analyses of agricultural policy in Africa have played their part in underwriting Structural Adjustment reforms, and who we treat as representative of one side of the liberal political spectrum. As is well known, Bates’ argument is that the irrationality of African economic policy can best be explained by its political rationality; that policies discriminatory against the mass of producers and by implication the society as a whole, work to the advantage of an urban coalition of politicians, civil servants, industrialists, workers, and their allies in the countryside—large farmers (Bates 1981). While we agree that small groups frequently have an advantage in capturing the outputs of the governmental process, there is much to object to in Bates’s thesis. For the purposes of this paper, it is enough to note that its model of the self, drawn from neo-classical economics, is fantastical, representing at best a single dimension of the complex social constructions we call human (Ashmore and Jussim 1997, Burkitt 1991, Melucci 1996a).

5Rational, self-interested, utility maximising, almost invariably oriented to narrowly economic conceptions of utility, the more astute liberal economists long ago recognised that this self was an ideal to be aimed at rather than an object to be encountered in real society (Williams 2000). Nonetheless, it forms the basis of Bates’ analyses, and where at best these are usefully sceptical of the ability of international institutions to operationalise their reforms (Bates 1989), at worst they present drastically oversimplified conceptions of politics as achievable possibilities. A striking example is when Bates argues, a) that small producers have an interest in economic reform, b) that they fail to have policies made in their interest because they are dispersed and find it difficult to combine to affect the political process, therefore c) a political process equipped to accommodate dispersion, viz. a multi-party election, will result in economic reform and the realisation of small producers’ interests (Bates 1988: 350). Logically sound but practically flawed, as numerous multi-party elections have demonstrated, the argument of this paper is that for small producers to hold governments accountable, much more than elections will be needed.

6Liberal Africanists more sociologically attuned than Bates are today committed to transforming Africans (Diamond 1994; Hyden 1997; Landell-Mills 1992; Moore 1993). Their vehicle is «Governance», and it aims at constructing the markets and those necessary appendages that neo-liberals wished only to unleash (Williams and Young 1994). In reality a recrudescence of modernisation theory, Governance seeks to re-work Africa and Africans in order that they might slot into an appropriate space in the international capitalist economy. Old-school modernisation theorists, for their part, were pre-occupied with the relations between economic and cultural structures and «personality». At the centre of social change stood the self, which formed the relay between the inputs, or hardware of modernity, and the outputs or modes of behaviour appropriate to a «modernised» society (Inkeles & Smith 1974; Lerner 1964). In this perspective, either the self changed, shaping itself into a medium for these processes, or society would not. Conveniently, large-scale survey research suggested that the self could indeed be expected to respond in fairly predictable ways to developmental inputs such as cooperative societies, factories, and organs of the mass media: modern inputs created modern selves that acted to institutionalise the modern outputs. As we now know, the model of the self here deployed was much too behaviouristic. In reality, inputs interacted with selves complexly, and could be appropriated in all manner of subversive ways, resulting in outputs quite unintended by the planners (Bendix 1973).

7The latter fact ought to have prompted a re-problematisation of the self, a special focus on the missing link in Africa’s journey to modernity. There is reason to think that it did, but curiously, the self is elided in the writing of the neo-modernisation theorists. Precisely because of its centrality, perhaps, it performs a vanishing act, being substituted by the slightly less controversial objects (or touchy subjects) of «administration», «society» or «culture» (Diamond 1993; Hydén 1997; Landell-Mills 1992; World Bank 1989, 1997). Nevertheless, the self is assuredly the target of a new microphysics of power to be organised by multilateral and bilateral donors and conducted via the medium of «capacity building» and «civil society» programmes (Abrahamsen 2000; Hearn 1998; Williams 1993; Williams & Young 1994). What these programmes represent is nothing less than a battery of disciplinary interventions designed to fix African civil servants in new dispositions congenial to «sound administration», and African citizens in new modalities of political behaviour conducive to «democracy». Since the African state rightly cannot be relied upon to discipline itself or its own population, NGOs are accorded a privileged place in inserting these new governmentalities into the African social fabric (compare Foucault 1982, 1991).

  • 1 POSTMA (1994) represents a good example of this kind of tinkering within the Governance paradigm.

8Larry Diamond, echoing Tocqueville, claims NGOs can be «large free schools, where all the members of the community go to learn the general theory of association» (Diamond 1994: 8). Göran Hydén, ideologically flexible but rigidly modernist, urges donors to support horizontal, non-exclusive NGOs, capable of building the right sort of (viz. liberal) «social capital» (Hydén 1997). John Campbell – among others – bemoans the types of actually existing civil society in Tanzania, and elsewhere, as not being civic enough (Campbell 1999). Naomi Chazan celebrates the democratic potential of grassroots movements in civil society (Chazan 1986, 1988). Ever nostalgic for what was never there, liberals wish to strip Africans of the illiberal prejudices that imprison them, restoring them to an original state that in practice entails their creative destruction. That capacity building and civil society interventions fail to have these effects, and that NGOs are enveloped by Africa’s existing social capital, rather than injecting their own, is increasingly well documented, and could have been easily predicted (Bayart et alii 1999; Cameron 2001; Chabal & Daloz 1999; Igoe 1999; Ndegwa 1996; Mercer 2002). It seems that commentators both in favour and opposed to this «liberalling» of the African continent, over-excited each other (Joseph 1992; Wiseman 1995; Young 1995). Nonetheless, «Governance», within donor organisations themselves, and for the academics with whom they consult, has acquired something of the status of a «normal science», whose major axioms have become unquestionable (Abrahamsen 2000; Postma 1994)1. This ideological obsession with liberal democracy precludes the possibility of institutionalising alternative methods of accountability.

9In sum, we have two liberal schools and two strategies: that inspired by neo-liberal economics accepts the self as a given, and seeks only to transform the structures that surround it; that inspired by modernisation theory regards the self as a problem, and adopts it as the unspoken target of its transformatory interventions. A rather similar state of affairs exists among «radical» scholars. Either the self is not problematised, attention being given to «classes» or «masses», or the self is an oblique target of political intervention (most revolutionary strategies, after all, demand a transformation in the consciousness of the oppressed) (compare O’Laughlin 2000: 8). But rarely, as we shall see, is the strategy spelt out or its target adequately apprehended.

10Popular Struggles

  • 2 A recent collection by political economists centred at the University of Leeds, ends with the limp (...)

11It is a testament to the hegemony of liberal values in contemporary global society that the majority of «Left» Africanists today embrace, albeit with significant qualifications, the process of liberal democratisation on the continent2. In a noteworthy contribution, Adebayo Olukoshi welcomes current democratic openings, because they might open processes that end in more meaningful forms of governance (Olukoshi 1998). Admittedly, Africa’s liberal democracies are not in good health, he argues, for two main reasons. Firstly, economic conditionalities have precluded the formation of the kind of «welfarist» social contract which, he avers, African struggles for democracy were really all about. As this contract has failed to materialise, support for democracy has fallen off, and interest has in cases been channelled into ethnic and religious forms of politics instead (1998: 27-28). Second, the ability of the opposition in «political society» to act as a credible mouthpiece for the welfarist elements in «civil society», has been undermined by a variety of factors, key among which has been the reluctance of African governments to create a level playing field, leaving the opposition vulnerable to frustration, disillusion and cooptation (1998: 29-30). These two factors – economic conditionality and state obstructivism – have increasingly driven a wedge between political and civil society, in much the same way as a wedge was driven between African nationalist movements and the independent governments to which they gave birth (1998: 33). In both cases, the abandonment of a broadbased developmental strategy in favour of policies that catered to a narrow elite, undermined the probability of stable and sustainable democracy. In order to rectify this state of affairs, neo-liberal economic policy needs to be abandoned in favour of a return to a notion of the developmental state, while those elements of civil society pressing for the full range of democratic freedoms need to be supported (1998: 34).

  • 3 Ironically, OLUKOSHI makes a similar criticism of arguments which doubt the sustainability of democ (...)

12Several initial criticisms can be made of this analysis. First, popular support for democracy is understood «instrumentally», as a means to the end of an economic system which guarantees social welfare (compare An’yang’Ny’ongo 1987; Ake 1993). Support for ethnic or religious anti-democratic movements appears to be understood in the same way. This, it could be argued, is an overly reductive and economistic approach to political subjectivity3. The full significance of this point will become apparent later. Second, the analysis rests on the assumption that there exists a model of the developmental state capable of upholding a «social contract» which would be compatible with democracy. Yet all content to the model has been evacuated from Olukoshi’s analysis. On the one hand, this is a defensible move, since the content of the model would have to be something decided popularly and democratically. But on the other, it risks making the model an exogenous force – the objet petit á to restore plenitude to African polities. It is well-documented, however, that developmental models are never neutral tools deployed for the benefit of the majority; rather they are sites of struggle that embed politics even as they attempt to efface it (Heyer et alii 1981; Ferguson 1990). Whether these struggles are suitable to being resolved under democratic conditions depends upon a host of other factors, not least the existence and depth of a national political culture, a desideratum consistently short-changed by Olukoshi’s analysis.

13One of the advantages of the concept of neo-patrimonialism – sneered at by Olukoshi – is that it captures, albeit imperfectly, the way in which development administration is entangled in complex and inherently ambivalent political practices, which revolve around the desire of politicians to gain access to power as well as the desire of both lesser elites and «popular» classes to irrigate a «moral economy» (compare Berman 1998: 331). The political discourses that embed these practices are typically nuanced and speak in registers resonant in local communities that have experienced and interpreted colonialism, capitalism and Independence in local, particular ways. These ongoing interpretations inscribe local cultures and identities which, because they are changing, are not «primordial», but which, because they are enduring, are not «instrumental» (Berman 1998; Grignon 1998; Lonsdale 1992, 1994). They mark real but relationally perceived differences between people; differences that find more or less virulent political expression according to context and conjuncture.

14The most sensitive rendering of this recurring feature in the African social landscape is John Lonsdale’s distinction between the dialectically related phenomena of «moral ethnicity» and «political tribalism» (to be discussed below). In the context of the present discussion, it is sufficient to note that the instrumentalism of Olukoshi’s identities combined with the political neutrality of his «developmental» state, prom ises an improbably harmonious alignment between democracy, development, and the popular classes. The point, however, is that any nationally constructed «social contract» will be filtered through local lenses, which will open opportunities for elites with local roots to use the language of moral ethnicity in service of political tribalism, as part of ongoing struggles for the accumulation of power and wealth.

15So, a key problem of Olukoshi’s argument is that his understanding of political identity and action is crude, leading to unlikely predictions about the possible trajectories of political change. It also has normative implications. While the «real» interest of the masses in some sort of social welfarism is assumed unchanging, their political identities are believed plastic, flowing into any mould that might realise that interest. The effect of this is to relieve the masses of responsibility for the forms their political action might take, making it incumbent instead on leaders in «political society» and on external donors to create conditions in which the masses might find democratic institutions congenial. In this way, the determinism implicit in the understanding of mass political subjectivity is complemented by a kind of voluntarism on the part of elites. Quite aside from the fact that this is theoretically unsatisfactory, it has the practical implication that mass interest in democracy has to be «bought» by elites through dedicated socio-economic interventions. Another way of putting this is to say that the political culture of the masses must be transformed from outside. The similarities with the liberal position should by now be evident.

16Similar problems beset the recent work of Mahmood Mamdani. In the same way that Olukoshi perceives a growing gulf between «political» and «civil society», Mamdani sees the stumbling block to African democracy in the gap that has opened between urban progressives and, in his thesis, the rural masses (Mamdani 1996: 21 et passim). The legacy of colonialism, he argues, was an urban, civic space regulated by civil law, and a rural, tribal space regulated by customary law; a dichotomy that gave to the state a «bifurcated» form, with «decentralized despotism» its local expression (1996: 16-23). The colonial chief both condensed and diffused the authoritarian potentialities in pre-colonial culture (1996: 23, et passim). After independence, the civic space was de-racialised, allowing an indigenous elite to accumulate resources and political power. By contrast, the rural space remained tribalized, administered under the authority of a despotic chief, agent of a range of extra-economic compulsions. In cases where the local state was de-tribalised, it was not democratised; rather, the chief was replaced by a secular despot (the District Commissioner or his analogues).

17Beyond this, Mamdani’s precise thesis is difficult to pin down. As his text reiterates, it is about bridging two political divides, the rural-urban and the interethnic (1996, 2000). However, the precise ways in which these divides result in authoritarian politics, or impede the process of democratisation, remain ambiguous. They assuredly concern the relation between the local despot (whether it be a chief, or an administrator with clear responsibilities to the centre), and the rural population; but the nature of the effect of that relationship is difficult to grasp. Reading between the lines of Mamdani’s dizzying argument, it seems that there are a number of possibilities. At its simplest, it is an argument about space. In Africa, either the centre or the locale has been democratised, but rarely both. Thus a genuine democratisation requires that both be democratised. But this is clearly too crude, not least because Mamdani is at pains to stress a temporal dimension: the democratisation must be a simultaneous one (1996: 217, 301). One reason for this might be that undemocratic administrators compromise the freeness and fairness of elections at a local level; but this obvious point, made by Olukoshi, is not stressed by Mamdani (Olukoshi 1998: 32).

18In light of this, one is led to think that he is positing an indirect effect of the institutional form of the state on the character of politics. But how does this effect operate? One interpretation might be that since the colonial chief inscribed a set of authoritarian potentialities in indigenous culture, Africans are used to being governed undemocratically, and consequently behave in a variety of undemocratic ways (Mamdani 1996: 37-181). Partial political liberalisations are not sufficient to alter this behaviour: a total liberalisation of the political sphere is necessary to transcend this historically produced authoritarian, conservative culture. But this seems to assume a rather too immediate and mechanical relationship between just one of the institutions Africans encounter in their everyday lives, and their political culture, or subjectivity. It is difficult to believe that with the removal of the appointed local administrator, the undemocratic chief, or the dictatorial President, Africans will automatically embrace a democratic culture.

19Mamdani’s argument must be more subtle than that. He sees a simultaneous democratisation of town and country as the prelude to a process of creative engagement between progressive urban classes and the rural masses (who already evince some democratic tendencies) which will result in a new national, progressive, democratic political culture (1996: 296-301). In this respect, where the locality is already democratised, the central state has to open itself up, exposing itself to the scrutiny of the masses. Otherwise it is easy for it to fall back on a mutually profitable alliance with narrow conservative elements in town and country, who are only too adept at manipulating a divisive politics of ethnicity. This has occurred in Uganda (1996: 209-210). Where the central state is already democratised, the local state has to be thoroughly opened up also. This involves removing the rural chief, who acts as a kind of gatekeeper to the rural areas and cements his clientele with the ideology of tribalism. Moreover, since all ethnic groups would prefer their members to be the gatekeepers to local resources and populations throughout the country, this induces an explosion of ethnic politics into the national arena. This, Mamdani observes, has occurred in Kenya (1996: 289). In South Africa, chiefs actually maintain agents in town, who can act as conveyors for the conservatism of the country, the foci for a violent urban ethnicity (1996: 280-281). In slightly different cases, such as that of Tanzania, the imperative again appears to be a thoroughgoing democratisation at local level, in order to offset the possibility that local administrators will act merely as agents for compelling peasant contributions to the central state treasury (1996: 107-108).

  • 4 O’LAUGHLIN (2000: 42), accuses him of «political reductionism», a charge he only partially rebuts ( (...)
  • 5 Lenin’s orchestration of the October Revolution is perhaps the most dramatic case in which politica (...)
  • 6 By «determine» I mean, «underdetermine», that is, set the parameters within which change must, othe (...)

20This is an optimistic argument, and it is subject to many of the same criticisms as Olukoshi’s. The institutional form of the state – in which the local despot is pivotal – is being asked to bear an insupportable weight. As noted above, the politics of ethnicity is the complex cause and effect of a historical process in which class formation was experienced locally by extant if relatively amorphous ethnic communities; communities subsequently enclosed in the institutional apparatus of Indirect Rule. Or to put it more simply, «ethnicity has been the fundamental context and idiom of class formation» (Berman 1998: 331). It traverses the economic, the institutional, the ideational. It will not evaporate when any one of these—in Mamdani’s case the institutional—is altered4. It will remain as a residue, with the potential to be mobilised in new institutional contexts. It is not impossible that it should be erased in a process of engagement with progressive urban classes. The present author would not want to forswear this happening. In certain historical circumstances, «the logic of the situation» can open spaces in which farsighted leaders can transform the institutions and structures of society (Kaviraj 1989)5. But it is as well to get our understanding of the structures right. Mamdani, by emphasising the institutional level of society, fails to give sufficient emphasis to the full range of factors that exert a determining effect on African political subjectivity6.

21Thus the institutional form of the state, I shall argue, is just one of the agencies by which the subjectivity of leaders and led is overdetermined; an overdetermination which results in the curiously contradictory forms of politics characteristic of the African continent. That is to say, rather than conceiving of African polities as site to a series of divisions between the urban and the rural, the authoritarian and the democratic, the progressive and the conservative, we should see these contradictions as condensed within and reproduced through the individual subject him or herself. And it is these contradictions which help to explain not only the volatility of politics in Africa but also the difficulties to be found in institutionalising forms of accountability, democracy included (compare Comaroff & Comaroff 1997; Geschiere & Gugler 1998).

22As noted above, the work of Bruce Berman and John Lonsdale is among the most sophisticated and fertile on the Left. It presents a nuanced and convincing account of the way in which class formation imbricates ethnic identity (the internal architecture of tribe) and, interacting with the institutional forms of the nation state (the external architecture), propels a certain sort of political subjectivity – «political tribalism» (Lonsdale 1994). Processes of class formation and moral debate compel, persuade and interpellate (call out) particular ways of feeling, thinking and action; when opened onto a national stage, new compulsions, persuasions and interpellations provide these subjectivities a further twist. In this way it avoids the tendency of both Olukoshi and Mamdani to see the «popular classes» as potentially (essentially?) progressive, yet easily perverted.

23Strong on analysis, it is, unfortunately, short on solutions. In Lonsdale’s optic, the imperative is to draw on the resources of multifarious moral ethnicities to create a national morality against whose standards national politicians might be held accountable (Lonsdale 1992). Berman, meanwhile, regards the building of social trust and universal norms as prerequisite to any national ethic of civil democracy (Berman 1998: 339). This is all well and good. The problem, however, is how to operationalise these insights. Part of the problem of enforcing political accountability, at either local or national level, is that citizens lack the resources to translate their ethical evaluations into practice. And one reason for this, is that the obstacles to a particular type of collective action in Africa, which constrain and course through the subject, are immense. Satisfyingly materialist in its account of the origins of moral discourse, the solutions extended by Lonsdale are confined to the order of the symbolic. And yet these two should be, must be, linked. A moral language will only have a material bite if the concrete impediments to collective action, which are inscribed in the articulations and contradictions of Africa’s social formation, are alleviated.

Post-structural fatalism

24The work of Patrick Chabal and Jean-Pascal Daloz offers a refreshing alternative to the programmatism of much writing on both Left and Right (Chabal & Daloz 1999). Their feel for African politics is impressive, especially in their identification of ambivalence, contradictoriness and productive ambiguity. Certain of their statements fly in the face of political correctness. In addition, they elevate accountability and the self – the phenomena with which we are most concerned – to centre stage. Unfortunately, their understanding of the latter leaves much to be desired.

  • 7 They are surely overly sanguine about the reach of these networks and the extent of people’s inclus (...)

25«Accountability» is in their work an amorphous idea. Discussion of it often slides into an excursus on the character of «legitimacy», leading to comments on the nature of «representation», their account of which is surprisingly reductive (1999: 37-39, 52-62). Underlying the richly described surface of African politics they unfailingly discover the single-minded pursuit of resources, «representation quite literally entails the active improvement of the material condition of the community represented» (1999: 55). Ultimately their analysis is banal: politics in Africa is a struggle for access to wealth, and the vehicles of this struggle are patrimonial networks (however flexibly constituted), which profit in the blurry interstices of the public and the private, the legal and the illegal, and which reach from the highest to the lowest levels of society7.

  • 8 CHABAL & DALOZ are to be applauded for stating that we cannot assume that Africans act any less, or (...)
  • 9 Of course this is much too simplistic, not least because the dichotomy between cognition and affect (...)

26Let us begin with a few points of difference between myself and Chabal and Daloz; a prolegomenon to an outline of the theoretical argument of this paper. Firstly, a fundamental problem with the discussion of the self in contemporary political science is its excessive focus on cognitive and calculating dimensions. The aim, at all costs, is to show that Africans act rationally8. However, psychoanalysis, psychology, philosophy and anthropology – disciplines in which the study of the self is much further advanced – agree that powerful non-rational elements enter into the constitution of the self and the actions that issue from it (Ashmore & Jussim 1997; Henriques et alii 1998; Shweder & Levine 1984). Mental states can be categorised in a variety of ways, but for ease of exposition, I will argue simply that the current literature pays insufficient attention to affective states and their effect on both identity formation and action (compare Heald 1998; Pels 1999)9.

  • 10 People in our own society are often prepared to admit that they «acted against [their] best interes (...)

27Chabal and Daloz, for example, state that individuals in Africa are «communal», that their calculations of legitimacy are based on how far a patron rewards a communal group, and that identities are multiple, fluid and deployed instrumentally (1999: 56-62). This is confusing, for if individuals are communal, it is hard to see how they can instrumentally pick and choose identities. Unless of course a certain set of communal identifications is held fixed, and individuals instrumentally choose additional identities which they expect will best reward this community. Chabal and Daloz come close to this position when they state that the individual has a very strong orientation to his or her extended family, which is constitutive of his or her identity, but that larger groups with which the individual identifies are add-ons, instrumentally acquired (1999: 27, 52). Unfortunately this will not do, since Africa is replete with examples of discord, exploitation and divided loyalties even within the extended family (see, for example, Moore 1986). As will become clearer below, the puzzling character of everyday action can be explained if we accept that identities are both affectively and cognitively constituted, that these constitutions can exert contradictory pulls on the individual, and that one of the effects of this can be an experience of self-division on the part of the subject10. A result of this is that individuals confront a variable geometry of identification, and can be simultaneously susceptible to the conflicting appeals of different types of group.

28Second, in normative political philosophy, a debate exists over whether identities (which incorporate such things as commitments and projects) constitute the self, or whether they are contingent additions to the self, which selves can step back from, evaluate, rank, and if they desire, discard (Sandel 1992). The debate is an important one, since it has implications for the feasibility and desirability of liberal forms of governance. A common position in African studies, as we saw above, is that identities are not contingent (selves are not liberal), but that through social engineering they can become so (Hydén 1997, Landell-Mills 1992). Chabal and Daloz’s position, by contrast, appears to be that African identities are already simultaneously both. Again, at its clearest, this consists in the observation that the individual has a very strong orientation to his or her extended family, which is constitutive of his or her identity, but that larger groups with which the individual identifies are instrumental additions to the self. However, this notion of instrumentality fails, for example, to explain the way in which some Africans engage in communal violence, the potential costs of which frequently outweigh the possible benefits. Again, as we shall see, the addition of affect into our understanding of the acquisition and maintenance of these identities helps make action of this sort intelligible. More generally, our position is that all humans have a capacity to step back from their identities, conceiving of them as contingent add-ons. However, it is only in certain historical circumstances productive of certain sorts of structured situation that individuals become disposed to do this on a regular basis. At other times, they are more or less absorbed in and defined by their identities. They identify with a group, which is libidinally cathected, and which impels action.

29Having pointed to some initial differences between our own approach and that of Chabal and Daloz, I can state the argument of this paper more schematically. I have three basic disagreements with the above authors, and with the Africanist literature in general. To begin with, «accountability», if it is to be a useful concept in Africanist political science, demands sharper definition. It needs to be distinguished from «legitimacy», which exists only as an evaluative mental state, and «representation», which is one of the mechanisms by which accountability obtains. Following from this, any analysis of accountability must be underpinned by a theory of collective action, since – theories of cyber-democracy notwithstanding – accountable government demands effective collective actors. Third, different sorts of self have different dispositions toward and different capacities for, collective action. The analysis of the self in Chabal and Daloz, (and in much of the post-structuralist literature), which seems to swing between phenomenology and economic reduction, needs to be supplemented by a set of tools for its enhanced theoretical understanding.

The Self

30It is easiest to discuss these differences in reverse order. The approach to the constitution of the self offered in this paper can be loosely termed «Marxist-symbolicinteractionist». In the opinion of this author such a theory ought ultimately to be supplemented by a psychoanalytical account, but since current theory in psychoanalysis is rooted squarely in the socio-cultural conditions of Western Europe, most crucially in its account of the oedipal stage, it will be sidelined here (compare Riesman 1986). However, a loose notion of libidinal cathexis will be retained (compare Dean 1998; Melucci 1996a, 1996b). The approach offered is symbolic interactionist in that it follows G.H Mead in arguing that infant humans acquire their reflexivity, their sense of self, their identity, through interaction with others by means of symbols (gestures and words) (Mead 1934). The stimulus to interaction is assumed to be, in the first instance, a biological drive to security and need satisfaction. The infant organism is driven to communicate its needs to others. In order to make itself understood it must imagine itself in the position of the other to which it is oriented. By imagining itself as an object to the other, when in actuality it is becoming an object to itself, it becomes a subject. In the course of interaction with an other, consciousness becomes self-consciousness (Mead 1934).

  • 11 It is important to note that identifications of a face to face variety are likely to be strongest, (...)

31The individual is progressively «thrown into» self-reflection as it considers the gestural and linguistic cues appropriate to inducing desired responses in others. Providing appropriate cues to others that are expressive of intention, implies that the self has acquired competence in a language. Languages gain meaning against a field of community practice. Significantly, a large part of the community’s practice will be dedicated to securing its own reproduction (Burkitt 1991). In this endeavour, certain modes for the production and appropriation of the means of subsistence and their surplus develop. Insofar as these modalities persist over time, composing a system, it is because they apply various kinds of pressures on people to conform to their imperatives (Kaviraj 1989). These pressures, another name for which is power, are of a coercive, persuasive and interpellative nature. They exert a powerful influence on people’s beliefs, capacities, and dispositions (their subjectivities), producing persons disposed to conform to the demands of the system, while punishing those that do not. A powerful influence is exerted also on identity, since the range of identities a person is able to take up is conditioned by the range of subject positions, the viable and permissible modes of behaviour, with which s/he can realistically identify11. The pressures are applied of course by persons who are themselves products of the system (Kelsall 2000).

32It is in this respect that our position is Marxist. This is not to say either that we attribute any primacy to the economy, or that the needs of the economy account entirely for the total way of life of a community and the forms of individuality that develop within it. All we are saying is that the social relations of production enframe a large part of the life of the community and that they exert a very significant effect on forms of individuality. In consequence, if we are to apprehend the identities, beliefs, capacities and dispositions (the subjectivities) typical of a given society, an important place to look to is its social relations. It is fitting that social scientists should do this, since the total structure of the economic system is opaque to many of its participants, and its effect on personality and behaviour often unconscious. By prematurely dismissing Marxist analyses of Africa on account of the scarcity of self-conscious classes on the continent, Chabal and Daloz overlook a powerful tool for the understanding of African political life (1999: 40-41).

  • 12 Such intertwining, vividly depicted in Michael Schatzberg’s metaphor of a triple helix, is essentia (...)
  • 13 It is important to note that fragmentation is not sui generis to Africa. The same observation has b (...)

33A striking feature of African economies is that they are mixed. Some writers have spoken of an articulated mode of production; for us the term social formation is less controversial. We desire simply to stress that economics on the continent consists, almost everywhere, in an intertwining of capitalist and pre-capitalist modes (Berry 1993; Guyer 1997; Iliffe 1995). Intertwining is also a feature typical of African religious belief, and of contemporary political institutions (see, for example, Schatzberg 1986)12. If the above account of self-formation is correct, we would expect the different «moments» of this social formation to pressure Africans to behave in quite different ways in different social contexts. This will require them, inter alia, to «assume», under impulsion, a range of different identities. In this respect we disagree with the tone of much post-structuralist writing, which argues that Africans adopt multiple identities in pursuit of «maximum instrumentality» (Mbembe, cited in Werbner 1996: 1). Although it has an air of truth, this statement is too utilitarian, voluntaristic, and individualistic. We would agree that although this scattering of identities across subject positions can allow Africans, chameleon-like, to evade totalizing ambitions of power, it also impedes confrontations with power (see below), and may be experienced as an acutely stressful sense of self-fragmentation (compare Mbembe & Roitman 1995)13.

Collective Action

34It is possible now to begin to discuss collective action. Collective action requires collective actors and collective actors need collective identities. Individuals must identify with a collectivity (Calhoun 1982; Dean 1997; Melucci 1996a, 1996b). Conceivably, this identification might take the form of an identification of interests, as in Olsonian public choice theory (Olson 1965). That is, individuals will combine when the expected benefits of doing so outweigh the costs. This occurs either when the individual realises that by not acting, she/he will scupper the success of a group of which s/he is a part; or, when benefits can be withheld from or sanctions applied to, non-participating members. Of course there are two major problems with this reading of identification. In the first place, «interests», in the Olsonian schema, are understood rather narrowly as individualistic economic interests. Second, it is doubtful that this type of accounting always enters the heads of individuals who act collectively. In our view, certain psycho-sociological circumstances encourage individuals to identify with a group to the extent that the group is construed as a part of the self, and self-interest as more or less identical to group interest. As a result, the process of deciding whether or not to act collectively is less one of cost-benefit accounting and more one of psychological impulsion (or affect) (Calhoun 1982).

  • 14 Stress on dissimulation as a political strategy is one of the strengths of the recent work by BAYAR (...)

35Durable collective action demands durable collective identities, capable of securing commitment from individuals across time. If, as in the African context, individual identities are typically parcelled out among different subject positions (since social reproduction requires a straddling of those positions) we would expect, other things being equal, the commitment of individuals to any particular collectivity to be intermittent. This is not to say that collective action cannot be mobilised – the continent is full of evidence that it can – only that such action as occurs is liable to be sporadic. Collective actors might only induce a temporary commitment from their individual members for at least four, analytically distinct, reasons. First, with a multiplicity of often contradictory subject positions to straddle, the affective claims of any single one will be necessarily limited, insofar as the libido is distributed between different objects. Second, with different subject positions demanding different and often conflicting sorts of action, the ability of an individual to commit to any specific type of action over time is attenuated. Indeed, conflicting demands of a normative nature may result in a sort of paralysis. Third: insofar as social reproduction demands access to a multiplicity of networks, the individual is conditioned not to close down options by giving offence to powerful patrons in any one of them. This encourages a culture of dissimulation rather than engagement or confrontation14. Collective actors have difficulty securing sustained commitment from their putative members, patrons find clients to be fickle. Fourth, there is a basic time constraint: the demands of subsistence require individuals to parcel out their time between a number of groups and activities, rather than investing heavily in one.

  • 15 It must be stressed that the creation of collective actors is processual-dialectical. We do not imp (...)

36The experience of collective action on the continent bears this out. First, much collective action is of an ephemeral, explosive character: riots, «ethnic» flare-ups, mob lynchings. It doesn’t depend on any organisation of which to speak. It is debatable whether it is entitled to be called collective action at all; though this is certainly not to portray it as «primitivist». Second, most organised action is strongly hierarchical, organised by elites whose very reproduction is identifiable with that of the organisation or network (though they too may parcel themselves between organisations, with predictable effects), and whose members drift in and out. Pace Chabal and Daloz, such drifting is not merely a result of instrumental calculation: it is bound up in complicated ways with the way in which collective actors call out to the identities of their followers. Third, collective actors of a relatively durable kind which secure very strong commitment do sometimes emerge, and they are often, though not exclusively, of a millenarian nature (Behrand 1999; Isaacman 1993; Ranger 1986) They are frequently related, though not reducible, to a situation of extreme physical and subsistence insecurity, which encourages members to throw in their lots with the movement, identifying with it entirely. Fourth, there exist a few examples of committed and vigorous collective action. They tend to occur where individuals have been lifted out of their everyday social relations and placed in new relations, which have become for them a total way of life. It is no coincidence that the criminal-rebel groups in Sierra Leone and Northern Uganda, for instance, have relied heavily on children15.


37I am now able to return to the opening issue of this paper: the problem of political accountability. Political accountability is not the same as legitimacy. While I will argue that all politically accountable leaders are legitimate, not all legitimate leaders are accountable. Legitimacy is a judgement bestowed upon a leader by the led; it is not a concrete, interdependent relation. Thus a benevolent dictator, whose actions accidentally conform to the wishes of his subjects, may enjoy legitimacy. It would seem strange however to call him accountable. Neither is accountability the same as being representative. In Western systems, representation is one of the mechanisms through which accountability is secured, and is thus distinct from accountability itself, just as «speaking» is distinct from «communication». According to Chabal and Daloz, in African systems representation is more a matter of sameness, or identity, between ruler and ruled. In particular it is measured according to how far leaders embody the aspirations of their followers. But it is clear that this definition suffers from the same problems as «legitimacy». This is more than a trivial criticism, since unless there exist institutionalised mechanisms that relate leaders and led, there is no guarantee that a leader who one day embodies the aspirations of his followers will not the next day abandon them for the love of his Swiss bank account or his own idiosyncrasies. General Amin, for instance, embodied the hopes and aspirations of many Ugandans when he came to power, but this embodiment rapidly narrowed . Wouldn’t it be better to say that although he was temporarily legitimate, he was never accountable?

38Broadening the discussion a little, we can see that similar insights apply to the question of responsibility. A government is responsible insofar as there are institutional mechanisms for its removal or sanction, a common mechanism being elections. However, it is possible to imagine two types of state in which these mechanisms exist but that would not be entitled to the predicate «accountable». In the first instance, institutions exist on paper but do not function in practice. Such is the case in much of Sub-Saharan Africa. In the second case, a government might be elected but fail completely to respond to societal demands, consequently losing a subsequent election. The same thing happens with the next government, and the next, and the next. It would seem strange to call these governments accountable, since they are utterly unresponsive to their supporters. This leads me to argue that political accountability consists in a conjoining of political responsibility with political responsiveness. A government is accountable when it is responsive because it is responsible; when the expectation of sanction or removal causes it to act in the way society expects or demands.

39For the purposes of the current argument, the important point to note is that if a government is to be politically responsive to a majority of the population, there must exist mechanisms through which the majority can convey to it opinions, expectations and interests. There must also exist arrangements whereby governments can be sanctioned. Crucially, if these mechanisms are not to be hijacked by a well-organised minority, there must emerge in society actors with a capacity to convey opinions and enforce sanctions on a more or less continuous basis. Since practically, governments are unable to compute the preferences of hundreds of thousands of discrete individuals, these actors must be collective. Thus the inference that political accountability depends upon the existence of durable collective actors.


40Space permits only a few very provisional and speculative remarks on the kind of social conditions in which such actors might emerge. If, as I have argued, the tendency of the social formation is to produce subjects with multiple identities, multiple identities that impede the formation of durable (and accountable) collective actors, then the attainment of political accountability on the continent demands an alteration in the social relations of production and reproduction. At present – and this is highly schematic – a key characteristic of those relations is that they encompass a sphere of production oriented largely to the internal economy, both subsistence and market; a sphere oriented to international markets; and another externally oriented sphere in which relations are more predatory than productive. It is the external links which most interest me here.

  • 16 Admittedly this remark is tendentious, but since it is not uncommon to hear psychoanalysts speak of (...)

41Typically, external intervention in Africa is bolstered by inducement and compulsion. In this way, the «injection» of other norms, institutions, and forms of production, tends to increase the number of identities and behaviours expected of Africans. Other things being equal, this reduces the probability that people will be able to commit, consistently, to any of them. Another effect, is that external interventions, especially in the cultural sphere, impel identifications of a quite contradictory type. This exerts potentially debilitating effects on the subject’s sense of self and its ability to commit to projects16. For other reasons, external interventions also have a tendency to encourage damaging «externalisations of accountability» (Allen 1995; Clapham 1996; De Waal 1997).

42In principle, however, an easy way to limit the proliferation and multiplication of the identities in which Africans are «encouraged» to invest, would be to limit the pace of externally driven change. Other things being equal, this would open the possibility of a recovery of accountability. A chance for ordinary Africans to gain an increased control over their everyday lives: the ultimate agenda, surely, of both Left and Right. But this eventuality, deductively derived, demands empirical exploration and substantiation. I conclude this article, therefore, by suggesting a number of potential avenues for further research:

43– Investigations into the impact of externally driven economic change on social identities and on the instances and types of collective action.

44– Investigations into the impact of the encouragement of externally driven political identities on extant practices of accountability.

45– Investigations into externally transmitted cultural change, its impact on identity and collective action.

46Of course, research under these three headings is already ongoing in African Studies in a variety of forms. In light of the argument of this paper, it could be usefully supplemented, however, by research into:

47– Alternative arrangements for the regulation of the international economy (broadly understood), with a view to assessing their potential impact on social change, identity, and collective action.

48The links between economic relations, social identities, collective action and the ability to make political institutions work – to make them accountable – is poorly understood. But social theory suggests that links there certainly are. Until they are better apprehended, the contribution of political science to improving the subjective experience of its objects of study, will remain, in all likelihood, inconsequential.

Topo da página


Rita ABRAHAMSEN (2000), Disciplining Democracy: Development Discourse and Good Governance in Africa, Zed Books, London.

Chinua ACHEBE (1987), Anthills of the Savannah, Picador, London.

Claude AKE (1993), «The Unique Case of African Democracy», International Affairs, 69, pp. 239-244.

Chris ALLEN (1995), «Understanding African Politics», Review of African Political Economy, 65, pp. 301-320.

Peter ANYANG’ NYONG’O [ed.] (1987), Popular Struggles for Democracy in Africa, Zed Press, London.

Richard D. ASHMORE & Lee JUSSIM [ed.] (1997), Self and Identity: Fundamental Issues, Oxford University Press, Oxford.

Robert BATES (1981), Markets and States in Tropical Africa: The Political Basis of Agricultural Policies, University of California Press, Berkeley.

Robert BATES (1988), «Governments and Agricultural Markets in Africa», in Robert BATES (ed.), Toward a Political Economy of Development: A Rational Choice Perspective, University of California Press, Berkeley, pp. 331-358.

Robert BATES (1989) «Structural Adjustment: A Sceptical Appraisal», in Simon COMMANDER (ed.), Structural Adjustment and Agriculture: Theory and Practice in Africa and Latin America, ODI/James Currey, London.

Jean-François BAYART (1993), The State in Africa: The Politics of the Belly, Longman, London.

Jean-François BAYART, Steven ELLIS & Beatrice HIBOU (1999), The Criminalization of the State in Africa, James Currey, Oxford.

Heike BEHRAND (1999), Alice Lakwena and the Holy Spirits: War in Northern Uganda, James Currey, Oxford.

Reinhard BENDIX (1973), «Tradition and Modernity Reconsidered», Comparative Studies in Society and History, IX, pp. 292-346.

Bruce BERMAN (1998), «Ethnicity, Patronage and the African State: The Politics of Uncivil Nationalism», African Affairs, 97, pp. 305-341.

Sara BERRY (1993), No Condition is Permanent: The Social Dynamics of Agrarian Change in Sub-Saharan Africa, University of Wisconsin Press, Madison.

Ian BURKITT (1991), Social Selves: Theories of the Social Formation of Personality, Sage Publications, London.

Philip BURNHAM (1996), The Politics of Cultural Difference in Northern Cameroon, IAI/Edinburgh University Press, London.

Craig CALHOUN (1982), The Question of Class Struggle: Social Foundations of Popular Radicalism during the Industrial Revolution, Basil Blackwell, Oxford.

Greg CAMERON (2001), «Taking Stock of Pastoralist NGOs in Tanzania», Review of African Political Economy, 87, pp. 55-72.

John CAMPBELL (1999), «Nationalism, Ethnicity and Religion: Fundamental Conflicts and the Politics of Identity in Tanzania», Nations and Nationalism, 5, (1), pp. 105-125.

Patrick CHABAL & Jean-Pascal DALOZ (1999), Africa Works: Disorder as Political Instrument, James Currey, Oxford.

Naomi CHAZAN (1986), «The New Politics of Participation in Tropical Africa», in Marion E. DORO & Newell Maynard STULTZ (eds.), Governing in Black Africa: Perspectives on New States, Africana, New York, pp. 55-70.

Naomi CHAZAN (1989), «Ghana: Problems of Governance and the Emergence of Civil Society», in Larry DIAMOND, Juan J. LINZ & Seymour Martin LIPSET, Democracy in Developing Countries: Volume Two, Africa, Lynne Rienner Publishers, Boulder, pp. 93-140.

Christopher CLAPHAM (1996), Africa and the International System: The Politics of State Survival, Cambridge University Press, Cambridge.

John L. COMAROFF & Jean COMAROFF (1997), Of Revelation and Revolution: The Dialectics of Modernity on a South African Frontier: Volume Two, The University of Chicago Press, Chicago.

John DANIEL, Roger SOUTHALL & Maurice SZEFTEL [eds.] (1999), Voting for Democracy: watershed elections in contemporary Anglophone Africa, Ashgate, Aldershot.

Kathryn DEAN (1997), «Introduction: politics and the ends of identity», in Kathryn DEAN (ed.), Politics and the Ends of Identity, Avebury, Aldershot.

Kathryn DEAN (1998), Capitalism and Collective Action: A Marxist Account of the Erosion of Political Commitment in Liberal Capitalist Cultures, Ph. D. dissertation, London School of Economics, University of London.

Alex DE WAAL (1997), Famine Crimes: Politics and the Disaster Relief Industry in Africa, IAI African Issues Series, James Curry, Oxford.

Larry DIAMOND [ed.] (1993), Political Culture and Democracy in Developing Countries, Lynne Rienner, Boulder.

Larry DIAMOND  (1994), «Rethinking Civil Society: Toward Democratic Consolidation», Journal of Democracy, 5, (3), pp. 4-17.

James FERGUSON (1990), The Anti-Politics Machine – Development, De-politicization and Bureaucratic Power in Lesotho, Cambridge University Press, Cambridge.

Michel FOUCAULT (1982), «The Subject and Power», in Hubert L. DREYFUS & Paul RABINOW (eds.), Michel Foucault: Beyond Structuralism and Hermeneutics, Harvester, London, pp. 208-217..

Michel FOUCAULT (1991), «Govermentality», in Colin GORDON, Peter MILLER & Graham BURCHILL (eds.), The Foucault Effect: Studies in Governmentality, Harvester Wheatsheaf, Hemel Hempstead.

Peter GESCHIERE & Josef GUGLER (1998), «Introduction. The Urban-Rural Connection: Changing Issues of Belonging and Identification», Africa, LXVIII, (3), pp. 309-319.

François GRIGNON (1998), «The kithitu oath in Ukambani: A Moral Contract in Kenyan Politics?», paper presented at the ASA-UK Bi-annual Meeting, SOAS, September 14-16th, London.

Jane GUYER (1997), An African Niche Economy: Farming to Feed Ibadan, IAI/Edinburgh University Press, London.

Suzette HEALD (1998), Controlling Anger: The Anthropology of Gisu Violence, James Currey, Oxford.

Julie HEARN (1998), «The ‘NGO-isation’ of Kenyan Society: USAID and the Restructuring of Health Care», Review of African Political Economy, 75, pp. 89-100.

Julian HENRIQUES, Wendy HOLLOWAY, Cathy URWIN, Couze VENN & Valerie WALKERDINE (1998), Changing the Subject: Psychology, Social Regulation and Subjectivity, Routledge, London & New York.

Judith HEYER, Pepe ROBERTS & Gavin WILLIAMS [eds.] (1981), Rural Development in Tropical Africa, Macmillan, London.

Göran HYDÉN (1997), «Civil Society, Social Capital, and Development: Dissection of a Complex Discourse», Studies in Comparative International Development, 32, (1), pp. 3-30.

Göran HYDÉN & Michael BRATTON [eds.] (1992), Governance and Politics in Africa, Lynne Rienner, Boulder.

James IGOE (1999), Ethnicity, Civil Society and the Tanzanian Pastoral NGO Movement: The Continuities and Discontinuities of Liberalized Development, Ph. D. dissertation, African Studies Centre, Boston University.

John ILIFFE (1995), Africans: The History of a Continent, Cambridge University Press, Cambridge.

Alex INKELES & David H. SMITH (1974), Becoming Modern: Individual Change in Six Developing Countries, Heinemann, London.

Allen ISAACMAN (1993), «Peasants and Rural Social Protest in Africa», in Frederick Cooper (ed.), Confronting Historical Paradigms: Peasants, Labour and the Capitalist World System in Africa and Latin America, University of Wisconsin Press, Madison, pp. 205-317.

Richard JOSEPH (1992), «Africa: The Rebirth of Political Freedom», Journal of Democracy, 2, pp. 11-25.

Sudipta KAVIRAJ (1989), «On Political Explanation in Marxism», in Krishna BHARADWAJ, Sudipta KAVIRAJ and Sukhamoy CHAKRAVARTY (eds.), Perspectives on Capitalism: Marx, Keynes, Schumpeter and Weber, Sage, New Delhi, pp. 132-174.

Tim KELSALL (2000), Subjectivity, Collective Action, and the Governance Agenda in Tanzania, Ph.D. dissertation, School of Oriental and African Studies, University of London.

Pierre LANDELL-MILLS (1992), «Governance, Cultural Change and Empowerment», Journal of Modern African Studies, 30, pp. 543-567.

Daniel LERNER (1964), The Passing of Traditional Society, The Free Press, New York.

John LONSDALE (1986), «Political Accountability in African History», in Patrick CHABAL (ed.), Political Domination in Africa, Cambridge University Press, Cambridge, pp. 126-174.

John LONSDALE (1992), «Wealth, Poverty and Civic Virtue in Kikuyu Political Thought», in Bruce BERMAN & John LONSDALE (eds.), Unhappy Valley: Conflict in Kenya and Africa – Book Two: Violence and Ethnicity, James Currey, London, pp. 315-504.

John LONSDALE (1994), «Moral Ethnicity and Political Tribalism», in Preben Kaarsholm & Jan Hultin (eds.), Inventions and Boundaries, Occasional Paper 11, International Development Studies, Roskilde University, Roskilde, pp. 131-150.

Mahmood MAMDANI (1996), Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism, Princeton University Press, Princeton.

Mahmood MAMDANI (2000), «Indirect Rule and the Struggle for Democracy: A Response to Bridget O’Laughlin», African Affairs, 99, pp. 43-46.

Karl MARX (1977), «Theses on Feuerbach», in David McLellan (ed.), Karl Marx – Selected Writings, Oxford University Press, Oxford, pp. 156-158.

Achille MBEMBE & Janet ROITMAN (1995), «Figures of the Subject in Times of Crisis», Public Culture, 7, pp. 323-352.

George Herbert MEAD (1934), Mind, Self and Society: From the Standpoint of a Social Behaviourist, University of Chicago Press, Chicago.

Alberto MELUCCI (1996a), The Playing Self: Person and Meaning in the Planetary Society, Cambridge University Press, Cambridge.

Alberto MELUCCI (1996b), Challenging Codes: Collective Action in the Information Age, Cambridge University Press, Cambridge.

Claire MERCER (2002), «Deconstructing Development: The Discourse of Maendeleo and the Politics of Women’s Participation on Mount Kilimanjaro», Development and Change, 33, (1), pp. 101-127.

Mick MOORE [ed.] (1993), «Good Government?», IDS Bulletin Special Issue, 24, (1).

Sally Falk MOORE (1986), Social Facts and Fabrications: Customary Law on Kilimanjaro 1880-1980, Cambridge University Press, Cambridge.

Steven N. NDEGWA (1996), The Two Faces of Civil Society: NGOs and Politics in Africa, Kumarian Press, West Hartford, Conn.

Bridget O’LAUGHLIN (2000), «Class and the Customary: The Ambiguous Legacy of the Indigenato in Mozambique», African Affairs, 99, pp. 5-42.

Mancur OLSON (1965), The Logic of Collective Action: Public Goods and the Theory of Groups, Harvard University Press, Cambridge MA.

Adebayo OLUKOSHI [ed.] (1998), The Politics of Opposition in Contemporary Africa, Nordiska Afrikainstitutet, Uppsala.

Peter PELS (1999), A Politics of Presence: Contacts between Missionaries and Waluguru in Late Colonial Tanganyika, Harwood Academic Publishers, Amsterdam.

William POSTMA (1994), «NGO Partnership and Institutional Development: Making it Real, Making it Intentional», Canadian Journal of African Studies, 28, pp. 447-471.

Terence O. RANGER (1986), «Religious Movements and Politics in Sub-Saharan Africa», African Studies Review, 29, (2), pp. 1-69.

Paul RIESMAN (1986), «The Person and the Life Cycle in African Social Life and Thought», African Studies Review, 29, (2), pp. 71-138.

Michael J. SANDEL (1992), «The Procedural Republic and the Unencumbered Self», in Tracey B. STRONG (ed.), The Self and the Political Order, Basil Blackwell, Oxford, pp. 79-94.

Michael SCHATZBERG (1988), The Dialectics of Oppression in Zaire, Indiana University Press, Bloomington.

Richard A. SHWEDER & Robert A. LEVINE [eds.] (1984), Culture Theory: Essays on Mind, Self and Emotion, Cambridge University Press, Cambridge.

Richard WERBNER (1996), «Introduction: Multiple Identities, Plural Arenas», in Richard WERBNER & Terence O. RANGER (eds.), Postcolonial Identities in Africa, Zed Books, London.

David WILLIAMS (1993), «Review Article: Liberalism and “Development Discourse”», Africa, LXIII, (3), pp. 419-429.

David WILLIAMS (2000), «Constructing the Economic Space: The World Bank and the Making of Homo Œconomicus», Millennium, 28, (1), pp. 79-99.

David WILLIAMS & Tom YOUNG (1994), «Governance, The World Bank and Liberal Theory», Political Studies, XLII, pp. 84-100.

John A. WISEMAN (1995), «Introduction: The Movement Towards Democracy. Global, Continental and State Perspectives», in John A. WISEMAN (ed.), Democracy and Political Change in Sub-Saharan Africa, Routledge, London, pp. 1-10.

WORLD BANK (1989), Sub-Saharan Africa: From Crisis to Sustainable Growth.

Tom YOUNG (1995), «”A Project to be Realised”: Global Liberalism and Contemporary Africa», Millennium, 24, pp. 527-546.

Topo da página


1 POSTMA (1994) represents a good example of this kind of tinkering within the Governance paradigm.

2 A recent collection by political economists centred at the University of Leeds, ends with the limp and unsupported assertion that, «wherever Africans voted, they voted for democracy» [DANIEL et allli (1999)].

3 Ironically, OLUKOSHI makes a similar criticism of arguments which doubt the sustainability of democracy in Africa (1998: 17).

4 O’LAUGHLIN (2000: 42), accuses him of «political reductionism», a charge he only partially rebuts (MAMDANI, 2000).

5 Lenin’s orchestration of the October Revolution is perhaps the most dramatic case in which political action bucked the structures, or the long rhythms of history. Unfortunately, structures have a habit of taking their revenge, as the Soviet case arguably exemplifies.

6 By «determine» I mean, «underdetermine», that is, set the parameters within which change must, other things being equal, take place (KAVIRAJ, 1989).

7 They are surely overly sanguine about the reach of these networks and the extent of people’s inclusion in them.

8 CHABAL & DALOZ are to be applauded for stating that we cannot assume that Africans act any less, or more rationally, than inhabitants of other continents, something with which we would broadly agree (1999, xvii). However, they fail to provide any examples at all of Africans acting «irrationally». Of course the use of this language has a stained history of which we must be mindful, but it seems unfortunate if this leads us to misapprehend actions on the continent.

9 Of course this is much too simplistic, not least because the dichotomy between cognition and affect can be critiqued, and because «physical» states also exert a powerful influence on the former. BERMAN (1998, 312) appears to acknowledge the importance of affect.

10 People in our own society are often prepared to admit that they «acted against [their] best interests», that they «weren’t quite [them]selves», or even that, «I was a different sort of person then», thereby acknowledging a kind of division in the self. At other times individuals construct post facto narratives with the aim of representing their behaviour to themselves (and others) as consistent.

11 It is important to note that identifications of a face to face variety are likely to be strongest, since they are more readily libidinally cathected.

12 Such intertwining, vividly depicted in Michael Schatzberg’s metaphor of a triple helix, is essentially syncretic; the strands remain separable. This is distinguishable from a hybrid, in which the different elements would fuse to produce a substance genuinely new.

13 It is important to note that fragmentation is not sui generis to Africa. The same observation has been made of Western societies, although their capitalist hegemony probably produces states of fragmentation that are less acute.

14 Stress on dissimulation as a political strategy is one of the strengths of the recent work by BAYART et alli (1999).

15 It must be stressed that the creation of collective actors is processual-dialectical. We do not imply a univocal causality between the nature of social relations and the formation of collective actors.

16 Admittedly this remark is tendentious, but since it is not uncommon to hear psychoanalysts speak of a «crisis» in European selfhood, it seems not unreasonable to suspect that something similar, if not more acute, is occurring in Africa. Compare MBEMBE & ROITMAN (1995).

Topo da página

Para citar este artigo

Referência do documento impresso

Tim Kelsall, « Bringing the Self Back In: Politics and Accountability in Africa », Cadernos de Estudos Africanos, 5/6 | 2004, 133-157.

Referência eletrónica

Tim Kelsall, « Bringing the Self Back In: Politics and Accountability in Africa », Cadernos de Estudos Africanos [Online], 5/6 | 2004, posto online no dia 17 setembro 2013, consultado o 24 junho 2018. URL : ; DOI : 10.4000/cea.1054

Topo da página


Tim Kelsall

University of Newcastle-upon-Tyne

Topo da página

Direitos de autor

Licença Creative Commons
O trabalho Cadernos de Estudos Africanos está licenciado com uma Licença Creative Commons - Atribuição-NãoComercial-CompartilhaIgual 4.0 Internacional.

Topo da página