Civil Society in African Contexts: Reflections on the
Usefulness of a Concept
David Lewis
ABSTRACT
This article considers the usefulness of the concept of civil society -- both
as an analytical construct and as a policy tool -- in non-Western contexts,
drawing on a selected review of literature on Africa from anthropology and
development studies. Rejecting arguments that the concept has little meaning
outside its Western origins, but critical of the sometimes crude export of the
concept by Western development donors seeking to build `good governance',
the author examines different local meanings being created around the concept
as part of an increasingly universal negotiation between citizens, states and
markets. The article seeks to clarify different theoretical traditions in thinking
about civil society, and suggests distinguishing the use of civil society as an
analytical term from the set of actually existing groups, organizations and
processes which are active on the ground. The concept is therefore useful in
the analysis of contemporary politics, but is also important because it has a
capacity to inspire action.
INTRODUCTION
Since the end of the Cold War, there has been a global ubiquity to the
concept of civil society among researchers and activists, and a widespread
assumption among many policy-makers in different parts of the world of
its global relevance to strengthening development and democracy. The
aim of this article is to provide some reflections upon whether or not the
concept of civil society can be seen as useful -- both in terms of the analysis
of social and political processes, and in relation to policy intervention aimed
at poverty reduction and development -- in the context of non-Western
societies; and if so, how.
This article has its origins in a lunchtime discussion seminar given by the author at the LSE
Global Civil Society Working Group organized by Helmut Anheier, Meghnad Desai and
Mary Kaldor. Feedback from many of the participants was useful in revising the paper for
publication, as were the comments of three anonymous referees. Special thanks are due to
Marlies Glasius, Deborah James, Nazneen Kanji, Jeremy Kendall, Hakan Seckinelgin and
Ronan Toal for their comments on earlier drafts of the paper.
Development and Change 33(4): 569­586 (2002). # Institute of Social Studies 2002. Published
by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main St., Malden,
MA 02148. USA
The question of usefulness can be explored along two dimensions: on the
one hand, the concept of civil society may be `useful to think with' in the sense
of supporting analysis which can help to make sense of political and social
realities, while on the other it may be `useful to act with', by helping to inspire
action on the ground. The term civil society, which itself has multiple meanings,
may of course also be useful in different ways to different kinds
of people, including political activists, development workers, policy-makers
and academic researchers. Among some development policy-makers and
policy-focused academics, an idea of civil society has become influential
which tends to be deployed in ways which limit not only the diversity of local
civil society understandings and struggles, but also the essentially political
nature of the concept itself. In providing a critique of this trend, this article
seeks to contribute to research on the relationship between concepts, ideologies
and development policy and, perhaps, to aspects of development practice.
This introduction presents some brief general comments about the recent
rise to prominence of the concept of civil society, as both an analytical
construct and as a policy tool. The article then moves on to discuss several
different views about civil society in non-Western contexts, drawn from a
selected review of literature on Africa from anthropology and development
studies. It is not the purpose of the article to engage in detail with the
complex definitional debates about the meanings of civil society, which have
been summarized effectively elsewhere (Keane, 1998; van Rooy, 1998); nor
does it aim at a comprehensive review of the African politics literature. It is
not the intention to attempt to generalize about Africa, which is vast, complex
and diverse. Rather, the article aims to explore selectively some of the
different positions in relation to the relevance of civil society which have
appeared in this literature.
It is useful to begin with a brief discussion of the different versions of the
concept of civil society which are commonly drawn upon. Civil society is
generally understood as `the population of groups formed for collective
purposes primarily outside of the State and marketplace' (van Rooy, 1998:
30). Civil society is usually seen as being situated beyond the household,
and writers such as Robert Putnam (1993) -- whose work has been highly
influential among policy-makers in recent years -- argue that civil society is
composed of horizontal solidaristic groups which cross-cut vertical ties of
kinship and patronage. The idea of civil society has many different roots.
The Scottish enlightenment thinker Adam Ferguson saw civil society as a
socially desirable alternative both to the state of nature and the heightened
individualism of emergent capitalism (Ferguson, 1767/1995). The German
philosopher G. W. F. Hegel argued that self-organized civil society needed
to be balanced and ordered by the state, otherwise it would become selfinterested
and would not contribute to the common good (see Hegel, 1821/
1991). Both approaches shaped the concept's early evolution. Moving from
the social and political sphere to a narrower organizational focus, the work
of Alexis de Tocqueville has been widely influential and has been used to
570 David Lewis
support arguments `in favour of ' civil society. De Tocqueville's positive
account of nineteenth century associationalism in the United States (de
Tocqueville, 1835/1994) stressed volunteerism, community spirit and independent
associational life as protections against the domination of society
by the state, and indeed as a counterbalance which helped to keep the state
accountable and effective. This account -- and elements of those which
preceded it -- tended to stress the role of civil society as one in which some
kind of equilibrium was created in relation to the state and the market. The
neo-Tocquevillian position can now be seen in current arguments in many
Western countries that the level of associationalism within a society can be
associated with positive values of trust and co-operation.
Two other issues are frequently flagged in discussions of civil society. One
is the idea that civil society is essentially fragile, borne out by Putnam's
(2000) account of anxieties surrounding the `collapse' of community in the
United States, typified by his idea that people now go `bowling alone'
instead of in groups. The other is concern about the historical specificity
or otherwise of civil society, which is a concept which has clear roots in
Western European experience and which may therefore have only limited
relevance to non-Western contexts (Blaney and Pasha, 1993). Both concerns
are revealed in Brown's (2000) account of civil society as a precarious, timebound
construct which first emerged from a distinct `historical moment' in
certain areas of Western Europe during the late eighteenth century. The new
middle classes, along with a commercializing landed aristocracy, required
conditions for sanctioned private accumulation underwritten by a state
which maintained legal order and stability. However the state was unable
to impose religious conformity, which led to an atmosphere of relative
tolerance in which a new civil society flourished. The state became strong
enough to maintain law and order but not so strong as to become oppressive
-- a balance which Brown (2000: 8) argues was crucial:
There is very little margin for error here -- if the state is too extensive it will strangle civil
society at birth, too weak and private institutions will compete for its role as provider of
order; if people are too much involved in each other's lives then they will lose the sense of
distance needed to preserve civility, too little involved and they become part of an atomized
`mass society'.
Such ideas have been highly influential in relation to efforts by development
policy-makers during the past decade to promote democratic institutions
and market reforms in developing countries. This is the so-called `good
governance' agenda which was made popular in the early 1990s and which
suggested that a `virtuous circle' could be built between state, economy and
civil society which balanced growth, equity and stability (Archer, 1994).
Indeed, much of the recent interest in civil society is clearly linked to the
global dominance of neo-liberal ideologies during the past decade, which
envisage a reduced role for the state and privatized forms of services delivery
through flexible combinations of governmental, non-governmental and
private institutional actors.
Civil Society in African Contexts 571
There is a different strand of civil society thinking which has also been
influential in some parts of the world. This strand has been influenced by
Antonio Gramsci, writing much later than the earlier civil society theorists,
who argued that civil society is the arena, separate from state and market, in
which ideological hegemony is contested, implying that civil society contains
a wide range of different organizations and ideologies which both challenge
and uphold the existing order (see Gramsci, 1971). These ideas were influential
in the context of the analysis and enactment of resistance to totalitarian
regimes in Eastern Europe and Latin America from the 1970s onwards.
Gramscian ideas about civil society can also be linked to the research on
`social movements' which seeks to challenge and transform structures and
identities (Escobar and Alvarez, 1992; Howell and Pearce, 2001).
Two differing civil society traditions can therefore usefully be distinguished,
although it has been the organizational view of civil society exemplified by
de Tocqueville which has been most enthusiastically taken up by agencies
within development policy discourse during the past decade (Davis and
McGregor, 2000). Since the early 1990s, the `good governance' agenda has
deployed the concept of civil society within the wider initiatives of supporting
the emergence of more competitive market economies, building bettermanaged
states with the capacity to provide more responsive services and
just laws, and improving democratic institutions to deepen political participation.
Development donor support for the emergence and strengthening of
non-governmental organizations (NGOs) has formed a central part of this
agenda (Archer, 1994).
While the various concepts of civil society may be useful for the purposes
of analysis, they may have less value as a prescriptive tool in the hands of
policy-makers. There are four main reasons for this. The first is that, as we
have seen, different understandings of the term exist and this makes it difficult
to agree precise policy purposes. A second reason is that since the concept
of civil society is primarily a theoretical one, it may not lend itself in any
straightforward way to a practical policy level application. Third, if the
concept of civil society is arguably historically specific to particular time(s)
and place(s), then it may be sensitive to differences of history, culture and
economy. Finally, a key problem with the concept is the frequent lack of
clarity within the research literature or within political discourse as to
whether a discussion of `civil society' refers to an analytical concept or to an
actually existing social form. This is a difficulty which muddles much of the
policy level debate on the importance of civil society. With these problems in
mind, it is useful to hold on to van Rooy's (1998) characterization of the
usefulness of the concept of civil society to development policy-makers in
terms of an `analytical hat-stand' on which to hang a whole range of ideas
about politics, organization, citizenship, activism and self-help.
Despite the argument that the concept of civil society is historically
specific, the concept is generating interest and debate in both Western and
non-Western contexts. The view that the concept is of less relevance to non-
572 David Lewis
Western contexts can obscure the fact that, as Hann and Dunn (1996) argue,
there is little agreement about the concept even within Western contexts in
terms of its relevance and practical policy value. Kaviraj and Khilnani
(2001: 3) point out that `Some of the ambiguities in the contemporary Third
World discussions arose because of the multiple meanings of the term in the
Western tradition itself . . . it was not enough to say that those who used the
idea did not look carefully at the Western concept. There was no single or
simple Western concept to study and emulate'. If we move on to consider
non-Western contexts, differences of culture, history and politics will complicate
these questions of `usefulness' still further.
The Re-emergence of an Idea
As we have seen, civil society is an old idea which has in the past two
decades undergone a massive global revival. The Comaroffs describe this as
an elaborate mythological cycle of imagining, memory loss and ultimately
the recovery of the concept. After emerging in Enlightenment Europe and
later influencing important nineteenth and twentieth century thinkers such
as Hegel, de Tocqueville and Gramsci, the concept was largely forgotten.
In the 1980s, it was rediscovered and given new contemporary relevance
by dissident intellectuals in communist Eastern Europe engaged in antitotalitarian
struggle. The concept has ironically returned to the West where
it has been `re-remembered' along with a realization that `we [in the West]
have been living it without noticing' as `part of the unremarked fabric of
society itself ' (Comaroff and Comaroff, 1999: 5).
In fact, the concept of civil society did not lie as entirely dormant as
accounts such as this one suggest. As we have seen, in Latin America from
the 1970s onwards, the Gramscian concept of civil society was embraced by
activists fighting against authoritarian military regimes. Keane (1998) shows
how Yoshihiko Uchida and the `civil society school of Japanese Marxism'
(ibid.: 13) emerged in the 1960s. Drawing also on Gramsci's ideas, Uchida
argued that Japanese civil society was weak, because the patriarchal family
and a culture of individual deference towards power allowed a specific
Japanese capitalism to grow quickly, with very little social resistance.
The recent re-entry into political discussion has, as the Comaroffs (1999)
point out, brought a paradox: while some policy-makers have become
interested in how to build civil society in countries where it is perceived to
be weak or non-existent, such as in Africa, a level of anxiety has now been
generated in some parts of `the West' that it has somehow been lost, as
witnessed by Putnam's (2000) argument that associationalism has declined
in the United States. According to the Comaroffs, there is another irony in
the relatively sudden re-emergence of the concept of civil society -- it is
often misleadingly presented as being both new and conveniently free of any
historical baggage.
Civil Society in African Contexts 573
This apparent novelty is misleading and potentially dangerous. Firstly, as
an old term, it may represent an outdated `modernist' concept with limited
relevance for explaining contemporary patterns of change. Secondly, if
the apparent dormancy of the concept is investigated further, it is possible
to show that the `re-remembering' of civil society in the West has been
somewhat selective, focusing as it has on the version which emphasizes the
role of non-governmental organizations `balancing' the state. By contrast,
a Hegelian concept of `civil society' may be useful in understanding how
access to and exclusion from public space and citizenship rights was
organized in colonial African contexts, while Gramscian ideas about civil
society have long been relevant to understandings of organized resistance to
colonialism and post-colonial states. In addition to the breadth of different
understandings of `civil society', there are also both `old' and `new' understandings
of the term.
REFLECTIONS ON THE RELEVANCE OF THE IDEA TO AFRICA
In Western development policy circles the act of `re-remembering' has, as we
have seen, been a selectively political one, since the version of civil society
which has dominated development policy agendas has been one which plays
down or ignores the more conflictual implications of the Gramscian version
(Davis and McGregor, 2000). It is this view that has tended to dominate
much of the policy debate concerned with civil society in Africa. This section
now briefly analyses a range of positions found in the literature on the
relevance or otherwise of the concept of civil society.
Four different possible answers can be identified to the question `is the
concept of civil society relevant to Africa?' The first is a clear `yes', based on
the idea of a positive, universalist view of the desirability of civil society as
part of the political project of building and strengthening democracy around
the world. There are many organizations and activists that explicitly
embrace this view. For example, the global civil society network CIVICUS
aims to `. . . help advance regional, national and international initiatives
to strengthen the capacity of civil society'.1
The second possible answer is
a clear `no', based on the argument that a concept which emerged at a
distinctive moment in European history can have little meaning within such
different cultural and political settings. From this perspective, `civil society'
is just another in a long line of attempts at misguided policy transfer from
the West. The third is an adaptive view which suggests that while the
concept is potentially relevant to non-Western contexts it will take on local,
different meanings and should not therefore be applied too rigidly, either at
the level of analysis or in the implementation of policy, where it should not
1. See CIVICUS website: www.civicus.org.
574 David Lewis
be deployed instrumentally in search of `predictable' policy outcomes.
Fourth, there are those who imply that the `relevance question' is probably
the wrong question to ask, arguing that the idea of civil society -- whether
explicitly recognized as such or not -- has long been implicated in Africa's
colonial histories of both domination and resistance. This view takes a
broader perspective on social and political changes, and analyses these in
historical and cultural context, whether or not there is explicit reference
made to `civil society' concepts. Each of these points of view will now be
briefly considered.
Prescriptive Universalism
The idea that civil society in Africa is a `good thing' which needs to be
encouraged and `built' has been eagerly taken up in some quarters. This has
perhaps been in response to early (and of course continuing) ideas by
outsiders that Africa was hostile to the development of civil life. Adam
Ferguson wrote in 1757 that Africa had a `weakness in the genius of its
people' and did not inspire the virtues `which are connected with freedom,
and required in the conduct of civil affairs' (Ferguson, 1767/1995: 108).
More recently, Harbeson (1994: 1­2) has argued that civil society is crucial
to Africa's future political development: `Civil society is a hitherto missing
key to sustained political reform, legitimate states and governments, improved
governance, viable state­society and state­economy relationships,
and prevention of the kind of political decay that undermined new African
governments a generation ago' (my italics). In this view, the concept of civil
society fills a theoretical gap in social science approaches to economic and
social development in Africa, where both donors and African governments
have largely failed so far to develop sound policies. The conceptual framework
of civil society allows for setting out and agreeing working rules about
improvements to the workings of the state and economy.
According to Harbeson (1994), this framework now has the potential to
shift discussion away from a focus on the balance of power between state
and society to debates about the terms of the interdependence between state
and society. The state is seen as the binding, organizing principle of the
political order -- the arena within which processes for the authoritative
allocation of social values takes place. Individuals, groups and organizations
are seen as part of the political order when they participate in these
processes, but are `. . . part of civil society when they seek to define, seek
support for or promote changes in the basic rules of the game by which
social values are authoritatively allocated' (Harbeson, 1994: 4). In this view,
civil society is not just about associational life, but is also about individuals
and associations which take part in wider rule-setting activities. Support for
the view of civil society as `counterbalance' to the state can also be found in
more anthropological accounts. In his recent study of the negative effects of
Civil Society in African Contexts 575
large-scale modern state planning schemes in Africa and elsewhere, Scott
(1998) argues that a `prostrate' civil society which is unable to resist such
designs provides the `leveled social terrain' upon which an authoritarian
state can then build.
Following from such positive views of civil society has come the phenomenon
of prescription at the level of policy. Within development policy discourse,
the framework of good governance has brought support for civil
society as part of a policy package transferred to Africa and elsewhere by
official donors and NGOs. For example, it has taken the form of support
for the monitoring of elections and voter education by civil society organizations,
and to `capacity building' work in relation to local NGOs through
the provision of organizational support and training.
This perspective on civil society remains a relatively optimistic one,
although there have been concerns raised within it. One is the danger of
what Blair (1997) calls `civil society gridlock' in which so many different
interests are active that they paralyse social and political life through a
multiplicity of claims made on services and resources. This is particularly
likely to happen when the state is weak, because additional `claims' made
from within civil society on the state may simply weaken it further. Another
is the problem of prescriptive failure when planned or desired results fail
to emerge. Despite being technically part of `civil society', certain groups
generated by a policy intervention may not in the end `fit in well' with the
prevailing political or moral vision. For example, Garland (1999) analyses
the work of an international NGO working with a hunter-gatherer community
in Southern Africa and finds within it an assumption that Western
civil society ideals are lacking locally and therefore need to be transferred
and built. While members of the local community readily seized the opportunities
which were on offer for building civil society with this outside
support and resources, they were later criticized by the NGO for having
created the `wrong' kinds of civil society institutions because they were
based on clan, kin and market systems.
Western Exceptionalism
Following from the critique of this prescriptive universalism argument,
some researchers suggest instead that the concept of civil society really has
little meaning outside the contexts of Western Europe or North America.
For example, in rejecting many of the arguments for the simple policy
transfer of civil society to Africa, Maina (1998: 137) argues that `civil society
is a concept made to order for the political reality of Western society' which
has `limited explanatory power' for the complexities of African associational
life because it fails to understand the domination of African societies by a
predatory state, the informal character of many forms of organization, and
the fundamental roles played by class and ethnicity.
576 David Lewis
In a recent overview of Mozambique, Sogge (1997: 42) finds three sets of
reasons to doubt the relevance of the concept of `civil society'. The first is
historical. The colonial state has constrained associational space so tightly
that `Mozambiquan ways of associating together hardly ever resemble
institutions of civil society known to Westerners' because they were not
voluntaristic. Under Portuguese colonial rule, the interests of business,
labour and farmers were organized through state-controlled associational
forms, and when political leaders emerged after independence they did so
from colonial state structures, and not from the trade unions or the legal
profession as in some other African countries. The second reason is that
during the past fifteen years of structural adjustment, characterized by weak
public services and low wages, there has been a gradual diminishing of state
power and legitimacy alongside a reduction of citizen expectations of the
state. The result is that there are now few real incentives for citizens to
organize, since power is widely perceived to be exercised from outside the
country. Finally, there is an ambiguous distinction between the public and
the private. Those forms of associational activity that do exist tend to do so
within an `obscure realm' where informal social and political action is
preferred over formal, and where the forms of organizational activity
constantly change shape in the effort to avoid capture or co-option.
James Ferguson (1998: 3­4) points out that civil society `has become one
of those things (like development, education or the environment) that no
reasonable person can be against', but that `the current (often ahistorical
and uncritical) use of the concept of civil society in the study of African
politics obscures more than it reveals, and, indeed, . . . often serves to help
legitimate a profoundly anti-democratic transnational politics'. He shows
how the `nation building' paradigm of African politics, which envisaged the
construction of national identities and structures of hierarchy in place of
backward, traditional, primordial affiliations, has gradually given way to a
`state/civil society' paradigm which sees the importance of a dynamic local
civil society in balancing, taming and reducing the role of the state. The
irony is that both models, though different, mirror each other and therefore
share a basic false assumption of the existence of a vertical state/society
opposition. Power in Africa has long been exercised by entities other than
the state -- such as the British South Africa Company, a private corporation,
which established structures for colonial rule in Zambia, and more
recently the international financial institutions such as the World Bank,
which have determined state policies through `structural adjustment' programmes.
At the same time, any differentiation at the national level between
state and civil society in Africa is, according to Chabal and Daloz (1999: 18),
largely illusory `. . . and derives more from wishful thinking or ideological
bias than from a careful analysis of present conditions'.
The tendency among outside policy-makers is usually for African civil
society to be conceived of in terms of a set of development NGOs, most of
which are funded by bilateral or multilateral development donors or by
Civil Society in African Contexts 577
international NGOs. Some of these NGOs are large transnational organizations
which are effectively taking over some of the state's functions in
health or education. The range of organizations which may lay claim to
being part of civil society is of course much wider than just the NGOs. At
the same time, these transnational NGOs have complex horizontal links
which stretch beyond Africa. The reason the civil society concept is unhelpful
in such cases, Ferguson argues, is because these NGOs do not actually
challenge the state `from below' but are instead `horizontal contemporaries'
of wider institutions of transnational governmentality. These new forms of
transnational connection show up the idea of a national `state/civil society
dichotomy' as simplistic and inaccurate. This point is well illustrated by
Ferguson's ethnographic work on community organizing in a South African
township. In this account, one particular activist sets out his key challenge
as building class-based civil society alliances with international solidarity
groups and with certain official donors, rather than building local networks
with what he regards as certain repressive elements within local civil society.
Adaptive Prescription
The problem with the first `universalist' view is that it tends to view phenomena
through a limiting Western lens and ignores the historical legacy of
colonial civil society building. It also ignores those organizations which do
not fit with its prescriptions (for example because they are based on kin or
ethnicity or local `tradition') or may miss altogether groups which take an
unfamiliar form. The prescriptive view is also based on imagined past stages
of Western civilization around eighteenth and nineteenth century capitalist
society and manners (Hardt, 1995). Such discussions have come to see
Western-oriented intellectuals, lawyers, entrepreneurs, teachers and church
people as `vanguards' of civil society's development in Africa because of
efforts to develop an active public sphere, the role of voluntary organizations
and the emergence of media institutions. Rejecting both this perspective
and the notion of Western exceptionalism, a third `adaptive' view
tries to argue that there is a middle way between crudely imposing the concept
from outside or simply abandoning it altogether as being inappropriate.
Despite his earlier objections to the concept of civil society in relation to
Africa, Maina (1998) later reluctantly concedes that such a concept can have
value if it is adapted in various ways. In particular, he suggests that it be
amended to include activities and not just organizations, and that it needs
to move away from a `Western' preoccupation with rights and advocacy to
include self-help groups that are organized for personal, economic ends. An
emphasis on informal, self-help activity reveals the existence of an active
civil society founded on a strong mistrust of the state and the overcoming
of civic apathy, and Maina suggests that it may be from this sector that
independent political leadership can emerge. For example, hometown
578 David Lewis
associations in Nigeria have long been an important part of the livelihood
and survival strategies of millions of Nigerians; this form of associational
life has evolved to meet the challenges of living under colonial administration
and, more recently, the absence of local government under a highly
centralized Nigerian state (Honey and Okafor, 1998). In a study of civil society
in Tanzania between the 1950s and the 1990s, Gibbon (2001) identifies a
core of lineages, age-sets, elders' committees and women's credit groups
which have long co-existed independently of the state, alongside a set of
other more formal entities such as churches, branches of political parties
and primary co-operatives which have constantly shifted their allegiances
and identities over time. In Uganda, munno mu kabi women's groups --
informal local self-help associations -- have proved flexible in responding
both to local problems and crises in the provision of home care services to
HIV/AIDS patients and in negotiating effectively their preferred mode of
support from international NGOs (Brehony, 2000).
Hann and Dunn (1996) argue that locally-specific `counterpart traditions'
may interact with the Western `export' of a universal idea of civil society.
Edwards (1998) develops this line of thinking in a draft paper for the World
Bank which sets out a set of oppositions between Western and non-Western
ideas about civil society. Rather than current prescription based on Western
models, Edwards argues for a more open-ended view of support to civil
society in terms of process, negotiation and as a contested domain. Within
this domain, there are a series of tensions which need to be negotiated
around the concept of civil society. For example, the origins of Western
ideas about civil society within the growth of industrial capitalism and the
nation state will contrast with the different histories and contexts found
elsewhere. Secondly, the three sector model common in the West -- state,
market, civil society or `third sector' -- may contrast with the increased degree
of `blurring' between such sectoral boundaries in non-Western societies.
Finally, the normative view of civil society as a set of positive values may
clash with the mixture of traditional and modern institutions which are both
civil and `uncivil' in other societies.
The need to think more broadly about the organizational and the moral
basis of civil society in African contexts is supported by the Comaroffs
(1999: 22), who state that there is a `Eurocentric tendency to limit civil
society to a narrowly defined institutional arena', which runs counter to
Hegel's original insistence that the civil sphere of relatedness has its origins
in the historical particularities of capitalist production and exchange. There
may be partisan, parochial, or fundamentalist organizations each with a
claim on civil society roles and membership. Recognition of local counterpart
traditions may therefore counter the tendency to undervalue the role
of kin-based and ethnic organizations in helping to form public opinions
and political pressure groups. The changing role of `tradition' in helping
to structure different forms of African civil society is also important. In
Tunisia, Keane (1998) describes how Al-Ghannouchi has shown how
Civil Society in African Contexts 579
sections of the Islamic religious community are beginning to accommodate
aspects of modern democratic principles within a `post-secular' view of an
Islamic society in which, between nature and God, there is a space in which
human, context-bound judgments have to be made.
Widening the concept of civil society to include involuntary membership
and kinship relations opens up the concept to locally-specific institutions
and processes. Karlstrom (1999) argues that we need to go beyond simply
identifying civil society with `voluntary associations', pointing out that in
Uganda the restoration of the Ganda kingship is politically significant
despite its nonconformity with Western ideals of liberal participatory
government. This restoration arguably promises a political order founded
on co-existing clan-based organizations and royal rule, a more stable,
responsive and representative institution than a national party system, and
a long history of clans mediating between citizens and monarchy, checking
excesses of power. The possibility for kinship to take on both a private and
a public face in Uganda contradicts Western assumptions of civil society
in which kinship relations are considered to be outside civil society norms:
`African sociocultural arrangements provide their own logic of sovereign
accountability, their own public spheres, their own forms of nongovernmental
organization and association' (Karlstrom, 1999: 27). While a distinction
between state and society is maintained, the relationships between them are
being reshaped. For example, there is a greater capacity for local government
councils to delegate certain tasks and resources to elements of `society'
in the form of NGO staff, groups and associations at the local level rather
than having these resources `captured' as used to be the case.
Within such a perspective, the obvious danger is of course to move from
prescription into an equally unhelpful position of cultural relativism. There
are also dangers in moving too far from a generally agreed understanding of
what `civil society' really means. If it is widened to include kin groups, for
example, it is a long way from Gellner's (1995) argument that civil society
should not only be seen in terms of balancing the state, but is also a counterbalance
to what he terms the `tyranny of cousins'.
The Wrong Question to Ask?
This final position is the argument that the concept of civil society has
always been relevant to questions of African governance and citizenship,
since it was used as an organizing principle by colonial administrations. It
makes little sense therefore to ask how useful the concept is to African
contexts. By taking such histories and their legacy as an analytical starting
point -- rather than the current return to fashion of the term `civil society'
-- advocates of this point of view argue that the relevance of the concept of
civil society is self-evident. In his study of colonial citizenship Wilder (1999:
45) writes:
580 David Lewis
The point is not that European categories are inappropriate for non-Western cultures.
Rather, the source of the dilemma lies in the fact that because colonial government
rationalized and racialized native society, civil society could only be an impossible promise:
at once the justification for colonial intervention (civilize the natives) and the sign of native
exclusion (they are not yet civilized).
It is less relevant to argue about universal or local realities and instead more
useful to analyse the historical processes which have shaped civil society in
Africa -- in its widest sense -- and the forms which these processes have
taken. Mamdani (1996) suggests that the current `blinkered view' of civil
society prevents us from looking critically at either European or African
civil society, and particularly at their complex interrelationship dating from
colonial times. Mamdani shows us that this meeting was in reality very
`uncivil' and aimed to institutionalize difference between groups of citizens
and ethnicized subjects, and between civilized colonists governed by `constitutionalism'
and native tribes governed by `customary law'.
These enduring sets of tensions have persisted in the post-colonial era;
they have helped structure the competition for state power and they have
also fuelled community conflict. The growing obsession with civil society in
Africa may date from the mid-1980s, but the contents of these debates have
a far longer history. For example, some nineteenth century humanists
arguing against slavery used a discourse of civility which implied universal
human rights and norms of citizenship, while the `colonial humanism' of
official policies spoke of universal rights but excluded native subjects from
such rights (Wilder, 1999). Nationalist resistance has long been couched in
terms of citizen rights denied, and has led to many social movements and
voluntary associations, and civic activism against the state in Africa has
as long a history as the other struggles normally associated with the reemergence
of civil society elsewhere in the world.
Mamdani (1996) shows how much of the debate on civil society is
concerned with competing ideological notions about how to match universal
governance ideals against existing institutions. For liberal modernists the
call is for civil rights, while African communitarians criticize the discourse
on human rights or its ethnocentricity and argue instead for the reinstatement
of marginalized political cultures. Mamdani therefore suggests, like
Ferguson, that such dichotomies are ultimately unhelpful because they
simply replay the old dichotomies of universal and particular, of Western
democracies and the colonial orders which were created in their name.
The `wrong question' argument therefore signals the need to bring these
wider perspectives into the rather limited discussion of civil society which
currently takes place. There is a rich theoretical tradition of thinking about
civil society, but far less empirical work available. An exception is Masquelier's
(1999) work on an Islamic reform movement in Niger which produces an
active engagement with wider forces of social and cultural change. Women
are required to cover themselves in public spaces but are at the same time
encouraged to pursue an education and register to vote, thereby challenging
Civil Society in African Contexts 581
the stereotypical conservative ideologies of gender, family and society. Such
movements may be central to civil society in Africa because they seek to
contest legacies of Western colonialism. In addition to the need for a
broader historical and cultural perspective, it is therefore necessary to enrich
debates with a more detailed understanding of `actually existing' civil society
activity in Africa.
CONCLUSION
Of the four different positions on the usefulness of civil society briefly outlined
above, the third and fourth are the most persuasive. The `prescriptive
universalism' argument which has been deployed in support of the crude
export of outsider visions of civil society by Western aid donors is clearly
flawed, while the `exceptionalist' argument which states that the idea of civil
society has little meaning outside its Western origins, underestimates both
the analytical and inspirational power of the term. By examining the range
of local meanings being created around the concept in certain African
contexts, it becomes clear that `civil society' -- broadly defined, with an
emphasis on Gramscian notions of struggle, power and dissent -- refers to
increasingly universal negotiations between citizens, states and market (the
adaptive argument). While the prescriptive `strengthening' of civil society is
strongly associated with the ideological dominance of neo-liberalism, there
is the paradox that contestation by civil society organizations and social
movements can also constitute a means through which such orthodoxies are
challenged and resisted. Civil society is not a new term, nor is it one which
has uniquely contemporary relevance. The control of public space and the
exercise of power either to include or exclude sections of the population as
citizens or non-citizens has long been a component of colonial history.
Although such phenomena may not previously have been understood in
relation to `civil society', they may now usefully be analysed as precursors to
what is currently meant by the term. This is the strength of the `wrong
question' argument, which shows that it makes little sense to allow changes
in terminological fashions to obscure analysis of historical continuities
which may exist between different stages of struggles between state and
citizens.
The potential usefulness of the concept of civil society can therefore be
analysed across two main dimensions -- it can be `useful to think with' and
it may be `useful to act with'. An adaptive, historically contextualized view
of the concept of civil society is analytically useful because its re-emergence
is linked to wider structural changes and state transformation. Blaney and
Pasha (1993) argue that the claim of the universality of the idea of civil
society is warranted, but that it must include the analysis of both `structure'
and `process' to avoid being represented as a static, ahistorical concept. This
means linking the discussion of African civil society with the international
582 David Lewis
capitalist division of labour and avoiding the simple conflation of `society'
and civil society. Such analyses will also need to reject a crudely normative
view of civil society by accepting that civil society contains a range of diverse
values and intentions. A related approach is that taken by Engberg-Pedersen
and Webster (2002), who reject the generalized view of civil society as
ahistorical and apolitical, and instead take a Gramscian view which deploys
the concept of civil society to explore contested political space in which
different groups, organizations and individuals seek to influence public
policy. The concept of civil society allows us to connect local and global
dimensions of political struggle, while Mamdani and Ferguson also show
how the concept reveals the dangers of over-simple conceptualizations of the
opposition of state and civil society given the increasing importance of
horizontal transnational identities and linkages.
But how useful is the concept of civil society `to act with' at the level of
policy? Despite the tendency for development donors to see civil society as a
normative concept, and to seek the `building' of civil society mainly through
financial and organizational support to NGOs, the new aid industry interest
in civil society may focus attention usefully on human rights, citizen action
and institutions. Whether such efforts can move forward to embrace the
political aspects of Gramscian notions of civil society, or whether development
projects are necessarily neutralized by the forces which act to maintain
Ferguson's (1990) `anti-politics machine' remains an open question. Recent
work by Jenkins (2001), which argues that bilateral donors that support the
strengthening of civil society continue to `mistake governance for politics',
suggests that this is unlikely. Jenkins shows that aid donors such as USAID
build only an apolitical `sanitized' version of civil society which excludes
organizations and individuals engaged in struggles for political power and
influence and which may be critical of neo-liberal orthodoxy. However, the
efforts of donor-assisted civil society actors to develop more autonomous
`room for manoeuvre' should not automatically be ruled out. For example,
the struggle in Mozambique to recognize collective land rights in the
development of the 1997 Land Law, and the subsequent civic movement in
support of the rights which it enshrines, has involved a pragmatic combination
of mobilization and advocacy work by local NGOs, international
NGOs and some donors (Kanji et al., 2002).
The concept of civil society has become `useful to think with' in the sense
that it has a striking capacity to inspire. It is precisely the ambiguities which
lie at the heart of the idea of civil society that give it importance at this
moment in history. Cut loose from simplistic policy transfer from the West,
the concept of civil society has become part of the political and social
discourse of a wide range of groups and individuals in Africa and elsewhere.
It is animating discussion and action, for example, in the current
negotiations towards constructing a new Somali government after a decade
of statelessness. At the recent Djibouti peace conference for Somalia the
language of civil society was deployed in documents drawn up locally in
Civil Society in African Contexts 583
support of the peace process: `The representatives of civil society, together
with the warlords, must come together to agree on the road to peace in the
interests of all citizens, national harmony and the democratic right to choose
leaders in accordance with an accepted formula' (Horn of Africa Bulletin,
2000: 14). In this case, the term civil society is used to describe organized
sections of society opposed to the domination of society by Somali `warlords'.
Whether or not civil society can be identified in recognizable forms `on the
ground', it has taken on meanings which are providing researchers, policymakers
and ordinary people with the means to rethink politics and
citizenship under conditions of global change.
Finally, a key conclusion is the need for more micro-studies of actually
existing civil society. While there are many case studies of NGOs and
community development projects, there are fewer micro-level studies of
more broadly defined, culturally varied types of `civil society' activity. The
reworking of the concept in the light of African histories, politics and
cultures allows for the possibility of illuminating more than just the
problems of development policy interventions. There is a need to focus less
on high levels of rhetoric, abstraction and ahistorical generalization and
produce more close-up observation (Comaroff and Comaroff, 1999: 4). The
absence of a clear understanding of the distinct forms taken by civil society
actors and actions in African contexts requires research which links local
realities with emerging global changes, but which resists being too
normative in its definition. Mamdani (1996: 19) therefore argues the need
for `. . . an analysis of actually existing civil society so as to understand its
actual formation, rather than as a promised agenda for change'.
Civil society in Africa can have multiple meanings, and as an `all-purpose
placeholder' (Comaroff and Comaroff, 1999: 3) it can capture emerging
aspirations in the context of local level social struggles for well-being and
global economic and technological changes. The concept of civil society
contains within it the seeds of contradiction in being both unitary and
divisive, and prescriptive and aspirational, but it nevertheless leads us to
focus on changing structure and process.
REFERENCES
Archer, R. (1994) `Markets and Good Government', in A. Clayton (ed.) Governance,
Democracy and Conditionality: What Role for NGOs?, pp. 7­34. Oxford: International
NGO Research and Training Centre (INTRAC).
Blair, H. (1997) `Donors, Democratization and Civil Society: Relating Theory to Practice', in
D. Hulme and M. Edwards (eds) Too Close for Comfort? NGOs, States and Donors, pp. 109­
120. London: Macmillan.
Blaney, D. L. and M. K. Pasha (1993) `Civil Society and Democracy in the Third World:
Ambiguities and Historical Possibilities', Studies in Comparative International Development
28(1): 3­24.
Brehony, E. (2000) `Whose Practice Counts?: Experiences in Using Indigenous Health Practices
in Ethiopia and Uganda', Development in Practice 10(5): 650­61.
584 David Lewis
Brown, C. (2000) `Cosmopolitanism, World Citizenship and Global Civil Society', Critical
Review of International Social and Political Philosophy 3(1): 7­26.
Chabal, P. and J-P. Daloz (1999) Africa Works: Disorder as Political Instrument. Oxford: James
Currey.
Comaroff, J. L. and J. Comaroff (1999) `Introduction', in J. L. Comaroff and J. Comaroff (eds)
Civil Society and the Critical Imagination in Africa: Critical Perspectives, pp. 1­43. Chicago,
IL: University of Chicago Press.
Davis, P. R. and J. A. McGregor (2000) `Civil Society, International Donors and Poverty in
Bangladesh', Commonwealth and Comparative Politics 38(1): 47­64.
Edwards, M. (1998) `Nailing the Jelly to the Wall: NGOs, Civil Society and International
Development'. Unpublished mimeo, available at www.futurepositive.org.
Engberg-Pedersen, L. and N. Webster (2002) `Introduction', in N. Webster and L. EngbergPedersen
(eds) (2002) In the Name of the Poor: Contesting Political Space for Poverty
Reduction, pp. 1­29. London: Zed Books.
Escobar, A. and S. Alvarez (eds) (1992) The Making of Social Movements in Latin America:
Identity, Strategy and Democracy. Boulder, CO: Westview.
Ferguson, A. (1767/1995) An Essay on the History of Civil Society. Cambridge: Cambridge
University Press.
Ferguson, J. (1990) The Anti-Politics Machine: Development, `Depoliticization' and Bureaucratic
Power in Lesotho. Minneapolis, MN: University of Minnesota Press.
Ferguson, J. (1998) `Transnational Topographies of Power: Beyond `the state' and `civil society'
in the Study of African Politics'. Draft mimeo, Department of Anthropology, University of
California, Irvine.
Garland, E. (1999) `Developing Bushmen: Building Civil(ized) Society in the Kalahari and
Beyond', in J. L. Comaroff and J. Comaroff (eds) Civil Society and the Critical Imagination
in Africa: Critical Perspectives, pp. 72­103. Chicago, IL: University of Chicago Press.
Gellner, E. (1995) `The Importance of Being Modular', in J. A. Hall (ed.) Civil Society: Theory,
History, Comparison, pp. 32­55. Cambridge: Polity Press.
Gibbon, P. (2001) `Civil Society, Locality and Globalization in Rural Tanzania: A Forty Year
Perspective', Development and Change 32(5): 819­44.
Gramsci, A. (1971) Selections from the Prison Notebooks. London: Lawrence and Wishart.
Hann, C. and E. Dunn (eds) (1996) Civil Society: Challenging Western Models. London:
Routledge.
Harbeson, J. W. (1994) `Civil Society and Political Renaissance in Africa', in J. W. Harbeson,
D. Rothchild and N. Chazan (eds) Civil Society and the State in Africa, pp. 1­32. Boulder,
CO: Lynne Rienner.
Hardt, M. (1995) `The Withering of Civil Society', Social Text 45 14(4): 27­44.
Hegel, G. W. F. (1821/1991) Elements of the Philosophy of Right (ed. Allen D. Wood, tr. H. B.
Nisbet). Cambridge: Cambridge University Press.
Honey, R. and S. Okafor (eds) (1998) Hometown Associations: Indigenous Knowledge and
Development in Nigeria. London: Intermediate Technology Publications.
Horn of Africa Bulletin (2000) Number 1. Nairobi: Life and Peace Institute.
Howell, J. and J. Pearce (2001) Civil Society and Development: A Critical Exploration. London:
Lynne Rienner.
Jenkins, R. (2001) `Mistaking ``Governance'' for ``Politics'': Foreign Aid, Democracy and the
Construction of Civil Society', in S. Kaviraj and S. Khilnani (eds) Civil Society: History and
Possibilities, pp. 250­68. Cambridge: Cambridge University Press.
Kanji, N., C. Braga and W. Mitullah (2002) `Promoting Land Rights in Mozambique and Kenya:
How do NGOs Make a Difference?'. London: Department for International Development.
Karlstrom, M. (1999) `Civil Society and its Presuppositions: Lessons from Uganda', in J. L.
Comaroff and J. Comaroff (eds) Civil Society and the Critical Imagination in Africa: Critical
Perspectives, pp. 104­123. Chicago, IL: University of Chicago Press.
Kaviraj, S. and S. Khilnani (eds) (2001) Civil Society: History and Possibilities. Cambridge:
Cambridge University Press.
Civil Society in African Contexts 585
Keane, J. (1998) Civil Society: Old Images, New Visions. Cambridge: Polity Press.
Maina, W. (1998) `Kenya: the State, Donors and the Politics of Democratization', in A. van
Rooy (ed.) Civil Society and the Aid Industry, pp. 134­67. London: Earthscan.
Mamdani, M. (1996) Citizen and Subject: Contemporary Africa and the Legacy of Late
Colonialism. Princeton, NJ: Princeton University Press.
Masquelier, A. (1999) `Debating Muslims, Disputed Practices: Struggles for the Realization of
an Alternative Moral Order in Niger', in J. L. Comaroff and J. Comaroff (eds) Civil Society
and the Critical Imagination in Africa: Critical Perspectives, pp. 219­50. Chicago, IL:
University of Chicago Press.
Putnam, R. D. (1993) Making Democracy Work: Civic Traditions in Modern Italy. Princeton,
NJ: Princeton University Press.
Putnam, R. D. (2000) Bowling Alone: The Collapse and Revival of American Community. New
York: Simon and Schuster.
van Rooy, A. (1998) Civil Society and the Aid Industry. London: Earthscan.
Scott, J. C. (1998) Seeing Like A State: How Certain Schemes to Improve the Human Condition
have Failed. New Haven, CT: Yale University Press.
Sogge, D. (1997) Mozambique: Perspectives on Aid and the Civic Sector. Oestgeest, The
Netherlands: Gemeenschappelijk Overleg Medefinanciering.
de Tocqueville, A. (1835/1994) Democracy in America. London: Everyman.
Wilder, G. (1999) `Practising Citizenship in Imperial Paris', in J. L. Comaroff and J. Comaroff
(eds) Civil Society and the Critical Imagination in Africa: Critical Perspectives, pp. 44­71.
Chicago, IL: University of Chicago Press.
David Lewis is Senior Lecturer in the Centre for Civil Society, Department
of Social Policy, London School of Economics and Political Science
(Houghton St., London WC2A 2AE, UK). He specializes in research on
anthropology and development, agrarian change in South Asia, and NGOs
and development management.
586 David Lewis