www.ssoar.

info

The hidden costs of power-sharing: reproducing

insurgent violence in Africa
Mehler, Andreas; Tull, Denis M.

Veröffentlichungsversion / Published Version

Zeitschriftenartikel / journal article

Zur Verfügung gestellt in Kooperation mit / provided in cooperation with:

GIGA German Institute of Global and Area Studies

Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG geförderten) Allianz- bzw. Nationallizenz
frei zugänglich. / This publication is with permission of the rights owner freely accessible due to an Alliance licence and
a national licence (funded by the DFG, German Research Foundation) respectively.

Empfohlene Zitierung / Suggested Citation:

Mehler, A., & Tull, D. M. (2005). The hidden costs of power-sharing: reproducing insurgent violence in Africa. African
Affairs, 104(416), 375-398. https://doi.org/10.1093/afraf/adi034

Nutzungsbedingungen: Terms of use:

Dieser Text wird unter einer Deposit-Lizenz (Keine This document is made available under Deposit Licence (No
Weiterverbreitung - keine Bearbeitung) zur Verfügung gestellt. Redistribution - no modifications). We grant a non-exclusive, non-
Gewährt wird ein nicht exklusives, nicht übertragbares, transferable, individual and limited right to using this document.
persönliches und beschränktes Recht auf Nutzung dieses This document is solely intended for your personal, non-
Dokuments. Dieses Dokument ist ausschließlich für commercial use. All of the copies of this documents must retain
den persönlichen, nicht-kommerziellen Gebrauch bestimmt. all copyright information and other information regarding legal
Auf sämtlichen Kopien dieses Dokuments müssen alle protection. You are not allowed to alter this document in any
Urheberrechtshinweise und sonstigen Hinweise auf gesetzlichen way, to copy it for public or commercial purposes, to exhibit the
Schutz beibehalten werden. Sie dürfen dieses Dokument document in public, to perform, distribute or otherwise use the
nicht in irgendeiner Weise abändern, noch dürfen Sie document in public.
dieses Dokument für öffentliche oder kommerzielle Zwecke By using this particular document, you accept the above-stated
vervielfältigen, öffentlich ausstellen, aufführen, vertreiben oder conditions of use.
anderweitig nutzen.
Mit der Verwendung dieses Dokuments erkennen Sie die
Nutzungsbedingungen an.

Diese Version ist zitierbar unter / This version is citable under:

https://nbn-resolving.org/urn:nbn:de:0168-ssoar-374503
African Affairs, 104/416, 375–398 doi:10.1093/afraf/adi034
© The Author [2005]. Published by Oxford University Press on behalf of Royal African Society. All rights reserved

THE HIDDEN COSTS OF POWER-SHARING:

REPRODUCING INSURGENT VIOLENCE
IN AFRICA
DENIS M. TULL AND ANDREAS MEHLER

ABSTRACT
This article analyzes some factors underlying the spread of insurgent viol-
ence in Africa. It focuses on the impact external factors have on power
struggles on the continent. The first of these is the unsteady support for
democracy from Western donors, which has impeded more far-reaching
domestic changes in much of Africa. Second are wider changes in the
international setting that dramatically enhanced the international standing
of armed movements in the post-1989 period. The article argues that the
interplay of both factors has induced would-be leaders to conquer state
power by violent rather than non-violent means. This becomes particu-
larly evident in regard to Western efforts to solve violent conflict through
power-sharing agreements. The hypothesis is put forward that the institu-
tionalization of this practice for the sake of ‘peace’, i.e. providing rebels
with a share of state power, has important demonstration effects across
the continent. It creates an incentive structure would-be leaders can seize
upon by embarking on the insurgent path as well. As a result, and irre-
spective of their effectiveness in any given case, power-sharing agreements
may contribute to the reproduction of insurgent violence.

SINCE THE END OF THE COLD WAR INSURGENCIES HAVE DOMINATED the range
of violent conflicts in West and central Africa and have become a critical ele-
ment in rapid social change in most of the continent’s sub-regions. Aside
from their quantitative expansion, the significance of insurrections is also
noteworthy at the qualitative level. For one thing, vast stretches of Africa have
been ‘governed’ by insurgency movements for sustained periods and, more
important for the purpose of this article, an increasing number of insurgents
eventually find themselves in the government of the state they seek to con-
quer. Over the past decade, for example, in both Liberia and the Democratic
Republic of Congo (DRC), two successive governments were either displaced
by insurgents or were forced to share power with their rebel foes.
The steady recurrence of insurrections in a number of countries, as well
as their increasingly frequent inclusion into governments, thus seems to

Denis M. Tull ([email protected]) is a research associate at the German Institute for

International Affairs and Security in Berlin. Andreas Mehler is the director of the Institute of
African Affairs is Hamburg.
375
376 AFRICAN AFFAIRS

indicate that would-be leaders have some reason to conceptualize the

organization of violence as a viable path to occupying at least parcels of
state power. Taking this assumption as a starting point, this article aims to
explore underlying factors that contribute to the proliferation of insurgen-
cies as well as their often successful outcomes — success being defined as
the taking of state power or parts thereof. Rather than analyzing the root
causes of rebellion, it probes contextual factors that bear on the rationale of
politically ambitious leaders to start insurgency warfare.
To date, careful analysis of the significance of the insurgency phenome-
non has been an exception.1 Arguably, this is to some extent due to the
deeply rooted tradition of state-centred analysis in political science and its
sub-disciplines.2 Unsurprisingly, then, the weakening of state capabilities
has been identified as a major explanatory factor of high levels of internal
conflict and insurrection in Africa.3 It is beyond doubt that the limited mil-
itary capacities of many African states lower the costs of forming successful
insurgencies. Other commentators have emphasized economic aspects,
notably the availability of natural resources as a facilitating factor helping
to set up and sustain insurrections.4 While both factors are important, we
argue that an analysis of insurgency-related conflicts in Africa is incom-
plete without taking into consideration external factors relating to shifts in
the post-1989 international environment, the way outside actors seek to
solve violent conflict in Africa and the impact these changes have on the
calculus of would-be leaders in organizing violence.
Our argument is straightforward: over the past fifteen years, power-sharing
agreements between embattled incumbents and insurgents have emerged
as the West’s preferred instrument of peace-making in Africa. In almost
every country in which insurgent leaders mustered sufficient military
power to attract the attention of foreign states, they were included in ‘gov-
ernments of national unity’. We argue that the institutionalization of this
practice demonstrates Western willingness to provide political pay-offs for
insurgent violence and thereby creates incentive structures which turn the
rebel path into an appealing option in the pursuit of otherwise blocked
political aspirations. If valid, this hypothesis should have important impli-

1. See, for example Christopher Clapham (ed.), African Guerillas (James Currey, Oxford,
1998).
2. Douglas Lemke, ‘African lessons for international relations theory’, World Politics 56
(2003), pp. 114–38.
3. Jeffrey Herbst and Greg Mills, The Future of Africa: A new order in sight? Adelphi Paper
361. (International Institute of Strategic Studies, London, 2003), p. 40; Crawford Young,
‘The end of the post-colonial state in Africa? Reflections on changing African political dynamics’,
African Affairs 103, 410 (2004), pp. 43–6.
4. Among a growing body of literature, still most frequently quoted is Paul Collier and Anke
Hoeffler, Greed and Grievance in Civil War (World Bank, Washington, DC, 2001); equally
burgeoning are the critiques, e.g., Chris Cramer, ‘Homo economicus goes to war: methodological
individualism, rational choice and the political economy of war’, World Development 30, 11
(2003), pp. 1845–64.
 THE HIDDEN COSTS OF POWER-SHARING 377

cations for the policies of Western governments towards Africa and their
stated objective of conflict prevention.
The first section of the article briefly sketches the trajectory of the big-
gest insurgency movement of the Congo war, the Rassemblement congolais
pour la démocratie (RCD), which will illustrate how the current debate on
greed and grievance neglects the extent to which the strategies of insurgen-
cies are significantly shaped by the international environment and strategies
of extraversion. Section two addresses the political inconsistencies of West-
ern political engagement in Africa. We shall argue that parts of the profound
political malaise in much of the continent stem from the ambiguous stance
that the West has adopted in regard to both democracy assistance and con-
flict resolution. We examine the extent to which these policies have contrib-
uted to ever-more violent politics on the continent. As we shall analyze in the
third section, these inconsistencies have pushed some African countries into
a vicious cycle that corroborates the reproduction of insurgency-induced
violence. We develop the hypothesis that the West’s preferred instrument of
conflict resolution — power-sharing agreements — turns the rhetoric of con-
flict prevention on its head in that it inadvertently encourages would-be lead-
ers elsewhere to embark on the insurgency path. Finally, in the conclusion,
we attempt to formulate policy prescriptions that may help to overcome the
dilemma between conflict resolution and conflict prevention.

The case of the Rassemblement congolais pour la démocratie

In spite of military support from Namibia, Angola and Zimbabwe, the
government of President Laurent Kabila proved unable to crush the rebellion
of the Rwandan-backed RCD that began in August 1998. A year later, signi-
ficant pressure from regional and Western governments resulted in the
Lusaka agreement which foresaw the holding of a national dialogue, the
envisaged outcome of which was a ‘new political order’ for the Congo. This
so-called inter-congolese dialogue was based on the principle that all the par-
ticipants in the negotiations ‘shall enjoy equal status’.5 After significant
delays, and the emergence of further rebellions, the dialogue was finally con-
cluded in late 2002. In accordance with its raison d’être, the forum resulted in
a comprehensive power-sharing formula which provided the RCD with one
of the four vice-presidencies as well as numerous other government posts. In
the remainder of this section we shall offer a broad sketch of the RCD rebellion,
its leadership and objectives and its interplay with the international arena.
For a start, it seems difficult to categorize the RCD in terms of either
ideology or greed. In regard to ideology, for example, the RCD leadership

5. Lusaka Ceasefire Agreement, Annex A, Chapter 5, Arts 5.1 and 5.2.b. available at http://
www.usip.org/library/pa/drc/drc_07101999_toc.html
378 AFRICAN AFFAIRS

was far too heterogeneous to patch together a coherent political programme.

Much like its forerunner, Kabila’s Alliance des forces démocratiques pour la libéra-
tion (AFDL), it was a ‘coalition of convenience’6 that contained some strange
bedfellows. Roughly speaking, the RCD was composed of four groups: first,
former AFDL lieutenants of Rwandophone origin from the Kivus, including
Moïse Nyarugabo, Bizima Kahara and Azarias Ruberwa. Former followers of
the late President Mobutu, notably Alexis Thambwe and Lunda Bululu, made
up a second group. A third component of the initial nucleus was constituted
by some well-known academics and professionals such as Professor Ernest
Wamba dia Wamba and former United Nations (UNESCO) official Z’ahidi
Ngoma. The inclusion of these personalities was arguably meant to provide
the RCD with internationally renowned figureheads in an effort to shield it
from the predictable accusation of being a foreign proxy. A fourth group con-
sisted of various individuals disappointed or marginalized by Kabila, including
Emile Ilunga and Joseph Mudumbi, among others.7
Given the heterogeneity of the movement’s initial leadership, and the
well-known disgust its leading intellectuals (Wamba, Ngoma) harboured
for former ‘Mobutists’, some of whom were to become RCD founding
members, it seems almost inconceivable that this motley crew of would-be
insurgents could have assembled without the impetus of a third party’s hid-
den hand, i.e. Rwanda and Uganda. It was Rwanda that provided the
RCD leaders with an opportunity to conquer state power.
The story of the RCD seems to underscore Clapham’s suggestion that
‘insurgencies derive basically from blocked political aspirations’.8 Contrary
to their foot soldiers, contemporary rebel leaders, at least in the DRC, the
Central African Republic, Chad and the Republic of Congo, are no social
outcasts. Instead, they are a manifestation of ‘elite-recycling’, a term intro-
duced to describe the limited renewal of political elites in the context of the
post-1990 democratization period. Virtually all RCD leaders had served
formerly in senior government positions under Mobutu and Kabila and
were therefore members of the political establishment.9 As such, they did
not fight to address societal grievances but in order to reintegrate them-
selves into a system from which they had been excluded. Former ministers
or prime ministers under Mobutu, Lunda Bululu and Thambwe, as well as
Kabila’s former followers Kahara, Mudumbi and Nyarugabo, among oth-
ers, all fall into this category. These personalities possess the connections
and resources to organize a rebellion as a means to enforce their (re-)inclu-
sion into a political system which they have few incentives to transform.

6. Filip Reyntjens, La guerre des Grands Lacs (L’Harmattan, Paris, 1999), p. 220.
7. Gauthier de Villers with Jean Omasombo and Erik Kennes, Guerre et politique: Les trente
derniers mois de L.D. Kabila (Institut Africain, Tervuren, 2001), p. 49.
8. Christopher Clapham, ‘Introduction’, in Clapham, African Guerillas, p. 5.
9. A point also made by William Reno, ‘The politics of insurgency in collapsing states’,
Development and Change 33, 5 (2002), pp. 841f.
 THE HIDDEN COSTS OF POWER-SHARING 379

What is more, their strategies are firmly rooted in well-established postco-

lonial political practices such as clientelism.
The RCD’s poor record of governance in the territories under its control
underscores the lack of an agenda for political renewal.10 In fact, the single
most important factor to explain this outcome was its largely externally-
induced creation and its military dependence on Rwanda. Kigali’s policy of
imposing its security interests was by and large incompatible with the more
politically-oriented approach advocated by the RCD’s first president,
Wamba. The latter’s dismissal was indeed a result of these conflicting
interests. His successors, likewise appointed by Kigali, were both unable
and unwilling to rid themselves of these outside constraints.11 Perhaps
inevitably, the resulting lack of Congolese political ownership of the insur-
rection was an obstacle to the RCD building up a measure of political
credibility, let alone fostering local constituencies. Its lack of a political
programme, however, does not invalidate the hypothesis that its leaders
were seeking to ‘unblock’ political aspirations through violence.
As for the political economy of the insurgency, the strategies of the RCD
leadership as well as its asymmetrical power relations with its Rwandan god-
fathers throughout the war sit uneasily with the current discourse that
describes insurgency movements as essentially greed-driven. While there is
no shortage of evidence that economics has played a powerful role in extend-
ing and exacerbating the war in the DRC,12 it is much more difficult to
maintain that greed was the primary motive of the RCD. To be sure, RCD
elites did seize opportunities to acquire significant resources through tax rev-
enues and the marketing of mineral riches. But this is simply stating the
obvious, i.e. that warfare generates resources for violent elites. It does not
allow a mechanistic reading to the effect, for instance, that the top RCD
leadership has sought to prolong the war for economic purposes. For one
thing, it is debatable that the RCD was the main beneficiary of resource
exploitation in eastern Congo. As the UN Panel of Experts observed with
regard to mining businesses, Rwanda ‘perennially deprived its junior part-
ner, RCD-Goma, of any significant share in resources and prerogatives’ and
‘administrators have frequently pointed out that they were unable to manage
their army without sufficient resources’.13 For another, the RCD followed
a consistent policy throughout the war intent on accessing state power in
Kinshasa, a goal which finally paid off by way of the power-sharing agreement

10. Denis M. Tull, ‘The reconfiguration of political order in Africa: a case study from North
Kivu (DR Congo)’, (Institute of African Affair, Hamburg, 2005), chap. 3–5.
11. International Crisis Group, The Kivus: The forgotten crucible of the Congo conflict (IGG,
Brussels/Nairobi, 2003), pp. 14–22; De Villers et al., Guerre et politique, pp. 65–76.
12. United Nations, Final Report of the Panel of Experts on the Illegal Exploitation of Natural
Resources and Other Forms of Wealth of the Democratic Republic of the Congo (UN, New York,
2002).
13. Ibid, para. 78.
380 AFRICAN AFFAIRS

of December 2002.14 This objective was, of course, linked to the perspective

of acquiring resources deriving from sovereignty,15 but ‘this is no more than
to say that war is very similar to politics, with the crucial addition of a high
degree of explicit violence’.16 One can therefore conclude that a regime
change was the only identifiable goal the rebellion sought to achieve.
We would suggest that the RCD leadership’s readiness to subjugate
itself to Rwandan interests, as well as its use of violence, was an entirely
rational approach in the pursuit of political power. Without both of them,
access to state power would have remained closed. Furthermore, it would
have mattered little had the RCD respected human rights or sought to gar-
ner local support in eastern Congo in a ‘Maoist fashion’. RCD leaders did
not ignore local hostility towards their movement, but they had few incen-
tives to mend this state of affairs.17 Once the military deadlock prevented
an outright victory on the battlefield, it was in the international realm that
the gathering of support and respectability mattered most; and, conse-
quently, the RCD put much emphasis on its external relations.
Thus, in addition to the use of violence as a bargaining chip, the second
factor facilitating insurgency leaders’ access to state power is international
recognition. As we shall explain in the following section, the West’s will-
ingness to deal with insurgents has dramatically increased since the end of
the Cold War. In response, insurgents have expanded their efforts to bol-
ster external ties; that is, to seek formal international recognition that
either provides diplomatic support at the expense of internationally mar-
ginalized incumbents or facilitates their inclusion in externally brokered
power-sharing governments. Frequently, the energy invested by insurgents
in warfare is only matched by diplomatic efforts, thereby pushing domestic
political agendas even further into the background.
The RCD’s preoccupation with international recognition as a legitimate
contender for Congolese state power was arguably the very reason why
intellectuals like Wamba were granted formal leadership of the movement.
Mimicking other insurgents, the RCD hired lobbying firms to represent its
interests in Washington.18 Valuable help came from President Kabila him-
self, whose erratic rule and close relations with so-called rogue states like

14. Initially, however, it had hoped for an outright military victory.

15. Pierre Englebert, Why Congo Persists: Sovereignty, globalization, and the violent reproduc-
tion of a weak state Queen Elizabeth House Working Paper No. 95 (Queen Elizabeth House,
Oxford, 2003).
16. Stephen Ellis, ‘The old roots of Africa’s new wars’, International Politics and Society 2
(2003), p. 34. [http://www.fes.de/ipg/IPG2_2003/ZEITSCHRIFTE.HTM]
17. We do not find Mkandawire’s explanation for this, e.g. the urban malaise, compelling,
not least because a good number of the most violent acts by rebel movements have taken
place in cities like Bangui, Monrovia or Brazzaville. See Thandika Mkandawire, ‘The terrible
toll of postcolonial “rebel movements” in Africa: towards an explanation of the violence
against the peasantry’, Journal of Modern African Studies 40, 2 (2002), pp. 181–215.
18. ‘Etats-Unis/Congo-K: Emile Ilunga’, La lettre du continent, 28 October 1999.
 THE HIDDEN COSTS OF POWER-SHARING 381

Libya and North Korea had compromised him in Western eyes. Conse-
quently, the RCD justifiably regarded Kabila ‘as its best asset’.19
Not only in Africa did the RCD make rapid diplomatic inroads, hailed
by Vice-President Ngoma as ‘the start of recognition’20; it also achieved de
facto recognition in the Western capitals (Paris, Brussels, Washington)
which several RCD delegations visited at an early stage of the insurrection.21
After a trip to Paris, Ngoma was reported to have:

. . . welcomed the way Western countries were reacting and the French position on the
dispute in particular. He admitted having had contacts in government circles in Paris:
‘It is my job . . . France is a country that has understood what we are about, I am
pleased to say. The fact that it is keeping out of current events is a good sign [a refer-
ence to Paris’s strained relations with the RCD’s Rwandan backers]. Of course, we
are label[l]ed rebels, which makes it difficult for the international community to adopt
a stance, but it is encouraging for us to see that, in Africa as elsewhere, we are not
alone in this battle’.22

Ngoma’s comments are remarkable for their sensitivity to the conventions

of juridical statehood and, more importantly, an understanding of its
changed operational nature in the international realm over the past decade.
In the case at hand, for example, the then US assistant secretary of state for
African affairs, Susan Rice, held talks with RCD officials in Kigali shortly
after the start of the rebellion (November 1998) — ‘the highest-level contact
yet’ between the US administration and the rebels.23 Given past diplo-
matic practices, the talks in Kigali were a spectacular act of recognition.
They underlined how shifting international conventions benefit insurgen-
cies, whereas incumbent governments are no longer the privileged, let
alone sole interlocutors of outside powers.

Democracy assistance and conflict management

The ambiguities of Western engagement in sub-Saharan Africa and even
the lack of clear political strategies help to explain the exacerbation of mili-
tarized politics over the past fifteen years.
The political inconsistencies of Western governments have been particularly
damaging in the field of democracy. Initially, most Western governments

19. ‘Entrenched and overstretched’, Africa Confidential, 9 October 1998.

20. ‘African foreign ministers meet rebel leaders in Goma’, BBC Monitoring, 20 August
1998.
21. See the following BBC Monitoring reports: ‘Rebel leader says Kabila selling off the
national heritage’, 2 October 1998; ‘Ministers, rebel counterpart speaks to Belgian paper
about health situation’, 6 October 1998; ‘Rebel leader leads delegation to Washington’,
19 October 1998.
22. ‘Rebel leader Ngoma praises French, Western stance on rebellion’, BBC Monitoring,
20 August 1998.
23. ‘US envoy Susan Rice holds talks in Rwanda with DR Congo rebels’, BBC Monitoring,
6 November 1998.
382 AFRICAN AFFAIRS

supported democratic reforms and embraced liberal political reform as a

force for change on the continent. However, as Clapham and Wiseman
correctly predicted in 1995, Western pressure for democratization was
bound to be ephemeral.24 Except for a few egregious cases of foul play,
Western countries showed little inclination to penalize reform-resistant
governments in one way or another. In the former Zaire, for example, the
period of political liberalization (1990–96) proved a frustrating exercise for
the opposition. President Mobutu created dozens of partis alimentaires to
undermine an already fragmented opposition. Although donors suspended
development aid, initial Western support for the democracy movement fal-
tered the longer the domestic power struggle endured.
Strict enforcement of political conditionality remained the exception.
Throughout the 1990s, this was evidenced by the fact that the evolution of
aid levels of individual African countries bore little relation to their system
of government.25 Some governments, such as France’s clients in Côte
d’Ivoire, Togo and Cameroon, were even rewarded for their democratic
recalcitrance with sharply increased bilateral aid in the early 1990s.26 What
is more, reform-minded states received scant reward for undertaking com-
paratively extensive democratic engineering, and the quality of democratic
governance had no measurable impact on aid levels.27 Hence, the para-
digm of political conditionality that many Western countries embraced as a
determinant factor of aid allocation was more rhetorical than actual.
Ultimately, the battleground for international respectability proved to be the
field of electoral politics that Western donors turned into the ultimate
yardstick of democratic governance. However, the heavy dominance of the
executive and the often pervasive blending of state and ruling party generally
circumscribed the overall freedom of political choice in many African coun-
tries. Western actors revealed a strong unwillingness or inability to tackle these
deeply ingrained patterns of political behaviour and the concomitant fraudu-
lent electioneering.28 As a consequence, past and present incumbents find it
fairly easy to satisfy demands for more appropriate electoral procedures, while
the defeat of the opposition is in many instances a foregone conclusion.29

24. Christopher Clapham and John A. Wiseman, ‘Conclusion: assessing the prospects for
the consolidation of democracy’, in John A. Wiseman (ed.), Democracy and Political Change in
Sub-Saharan Africa (Routledge, London, 1995), p. 228.
25. Nicolas van de Walle, African Economies and the Politics of Permanent Crisis, 1979–1999
(Cambridge University Press, Cambridge, 2001), p. 268.
26. Michael Bratton and Nicolas van de Walle, Democratic Experiments in Africa (Cambridge
University Press, Cambridge, 1997), p. 241.
27. Van de Walle, African Economies, pp. 268f.
28. See, for example, Richard Sandbrook, Closing the Circle: Democratization and development
in Africa (Zed Books, London, 2000), pp. 26–32.
29. Although useful recommendations were formulated by experts: see Timothy D. Sisk,
‘Elections and conflict managements in Africa: conclusions and recommendations’, in Timothy
D. Sisk and Andrew Reynolds (eds), Elections and Conflict Management in Africa (US Institute
for Peace, Washington, DC, 1998), pp. 145–71.
 THE HIDDEN COSTS OF POWER-SHARING 383

Keeping in mind the rise of the election monitoring industry, it is telling

that many ruling parties have shifted their efforts from influencing the out-
come of polls on election day itself (e.g. ballot stuffing) to the pre-election
period through the intimidation of opposition politicians and their sup-
posed supporters, manipulating voters’ lists or the banning of opposition
figures from participation in the election (more recently on grounds of
‘dubious nationality’). Such techniques as blocking opposition access to
the media and the massive use of state resources in favour of the incum-
bent may no longer go unnoticed, as the final statements of electoral mon-
itoring bodies routinely point to serious irregularities. But outsiders equally
routinely affirm that the elections represent a major step forward in the dir-
ection of democracy. Even in instances where severe criticism was issued,
as was recently the case in Nigeria, Rwanda, Zambia and Malawi, donors
have shied away from taking any consequential action to uphold the credi-
bility of political conditionality. Many African governments with poor
democratic credentials are thus able to maintain international respectabil-
ity and, hence, a continued aid flow.
In summary, the third wave of democratization produced few tangible
changes in the rules of the political game. Perhaps worse, the seal of inter-
national legitimacy that incumbent front-runners claim under the guise of
electoral democracy has created immense frustration among opposition
leaders across the continent. From their perspective, it is a bitter irony of
the post-1989 period that their chances to access state power were only
marginally enhanced, while ‘formerly’ authoritarian incumbents are able to
cast themselves as elected democrats. All of this is not to deny that tremen-
dous democratic progress has been achieved in some African countries, but
this pertains mainly to basic rights, a factor that may to some extent
explain why public opinion about democracy remains fairly volatile.30 In a
considerable number of African countries, one witnesses the rise of so-
called illiberal democracies, a euphemism for electoral autocracies, where
effective opportunities for electoral change are simply non-existent.31
A variety of reasons explain Western reluctance to challenge bad old
habits in electoral autocracies. One of them is the belief that added pres-
sure on incumbents may prove counterproductive. According to this view,
strict enforcement of political conditionality may further destabilize fragile
polities and potentially fuel violence by contenders. Keeping the lid on
African countries by tacitly supporting incumbent governments may not

30. In a briefing paper by Afrobarometer (Democracy and Electoral Alternation: Evolving African
attitudes, April 2004), it is argued that democratic commitments tend to decline with the pas-
sage of time, but can be reinvigorated by an electoral alternation of power. See http://
www.afrobarometer.org/AfrobriefNo9.pdf.
31. There were ten democracies in Africa in 2002. See Monty G. Marshall and Ted Gurr,
Peace and Conflict 2003: A global survey of armed conflicts, self-determination movements, and
democracy (College Park, University of Maryland, MD, 2003), p. 25.
384 AFRICAN AFFAIRS

qualify as a viable strategy, but there is little doubt that this approach is still
guiding the policies of Western states towards Africa. This attitude, how-
ever, may strongly influence the calculations of would-be leaders who see
little chance to effect a turnover of government through elections. Perceiv-
ing the path to state power as being blocked in a situation where, at least in
theory, it should be open, has created the widespread sentiment among
opposition politicians that they are fighting a futile political battle.
The ambivalent stance of Western governments toward democratization
and the perceived impossibility of opposition groups securing election to
office sends powerful signals. Not mincing his words, Jean-François Bayart
has accused the European Union and its member states of having

. . . generally blocked the political revolutions that alone would have led to the trans-
formation of the productive texture of societies. In so doing, Europe has condemned
Africa to further military turmoil in the form of civil wars and interventions of a para-
colonial or proto-colonial type by some Sub-Saharan states.32

The recent developments in Côte d’Ivoire seem to confirm this observation.33

Throughout the 1990s, the opposition parties Front populaire ivoirien (FPI)
and Rassemblement démocratique républicain (RDR) were exposed to all sorts
of electoral manipulation by the government. International reactions to the
obviously rigged elections in both 1990 and 1995 were lukewarm at best.34
France’s rhetoric of democracy bonuses and political conditionality in the
wake of President Mitterrand’s La Baule speech must have sounded partic-
ularly hypocritical in that country. When, at the end of 1999, a mutiny
turned into a coup against the ailing Bédié regime, both opposition parties
were allegedly preparing their own coups d’état, indicating that recourse to
violence was already an instrument that both parties were contemplating –
unsurprisingly.35
In early 2000 the FPI and the RDR were represented in the transitional
government of the coup leader General Guéï, certainly as a recognition of
their political weight, but perhaps also of their ‘spoiling capacities’. Subse-
quently the RDR was excluded from government and its leader, Alassane
Ouattara, was barred from standing as a candidate in the 2000 presidential
elections. Rumours circulated of coup preparations by military officers

32. Jean-François Bayart, ‘Commentary: towards a new start for Africa and Europe’, African
Affairs 103, 412 (2004), p. 456.
33. See the special issues of Afrique Contemporaine 206 (2003) ‘Dossier Côte d’Ivoire’ and
Politique Africaine 89 (2003) ‘La Côte d’Ivoire en guerre’; on French involvement see espe-
cially Stephen Smith, ‘La politique d’engagement de la France à l’épreuve de la Côte
d’Ivoire’, Politique Africaine 89 (2003), pp. 112–26.
34. One of the present authors served as an election monitor in an ill-defined observation
mission to Côte d’Ivoire in 1995 and can attest to this. The professional standard of, for
example, EU election observation missions improved in the late 1990s. However, more
important is the political will to use the findings and reports of the missions.
35. Source: interviews in Abidjan, February 2000.
 THE HIDDEN COSTS OF POWER-SHARING 385

close to the RDR, and the polarization of the country’s political system set
the stage for further political violence.
A second aspect of the involuntary involvement of external actors, or
rather their inaction, is important. The government in Côte d’Ivoire did
not have substantial defence capacities by the end of the 1990s. The
national army was always less important for the security of the regime than
the presence of French troops. This meant that necessary investments in
the national army or the gendarmerie were never made.36 The coincidence
of France’s relative disengagement — in Africa as a whole and in Côte
d’Ivoire in particular — and the deep internal crisis proved to be an import-
ant factor aggravating the conflict: the hasty recruitment of young soldiers,
sometimes informally, led to the creation of uncontrollable units and to a
less discriminate use of violence in the period between 2000 and 2002.
The September 2002 rebellion met little resistance by the security forces
and only a late and half-hearted reaction by the French military which,
however, saved the Gbagbo regime. It was only in the aftermath that the
regime expanded its military capabilities substantially — again at a highly
problematic pace and by very suspect means. On balance, the long-lasting
but declining military tutelage by an outside force proved to be detrimental
to a peaceful settlement of the crisis which had arguably been prepared by
France’s ambivalent political role in the country throughout the 1990s.
The case of Côte d’Ivoire lends support to the finding that states under-
going transition, i.e. states governed by hybrid regimes, are six times more
likely than democracies and two and a half times more likely than autocra-
cies to witness outbreaks of societal wars.37 Needless to say, this does not
establish a causal relationship between Western ambivalence towards
(flawed) democratization processes, on the one hand, and the rise of insur-
gency movements, on the other. And, of course, insurgencies often take
place in countries that do not even qualify as electoral autocracies. However,
many internal conflicts originate in failed experiments in democratization
as, for example, the cases of Sudan, Rwanda, Burundi, Congo-Brazzaville
or the DRC suggest.
As a result, most Western governments and organizations shifted their
priorities in Africa from support of democracy to the fields of conflict pre-
vention, conflict management or back to Cold war-type notions of ‘stability’
in the course of the 1990s.38 Still Africa remains the continent with the

36. An interesting account is given by the Franco-Canadian journalist Guy-André Kieffer,

who ‘disappeared’ in Côte d’Ivoire in 2004: ‘Armée ivoirienne: le refus du déclassement’,
Politique Africaine 78 (2000), pp. 26–44. See also the anonymous article (CDG), ‘L’armée
ivoirienne: de la marginalisation à la prise du pouvoir’, Afrique Contemporaine 193 (2000), pp.
9–12.
37. See Marshall and Gurr, Peace and Conflict 2003, p. 17.
38. For the case of the European Union, see Gorm Rye Olsen, ‘Promoting democracy,
preventing conflict: the European Union and Africa’, International Politics 39 (2002), pp.
311–28.
386 AFRICAN AFFAIRS

highest incidence of violent conflict. Even though the responsibility for this
disturbing fact should not be attributed solely to outside actors, it remains
nonetheless true that the foreign policies of Western countries continue to
have an appreciable impact on political processes on the continent. What
explains the high numbers of conflicts, in spite of the purported attempts
of the West to solve or even prevent them? As already noted, inconsistent
support for democracy is one possible explanation for this outcome. Fur-
thermore, the pervasiveness of conflicts in Africa may be linked to concep-
tual weaknesses underlying Western policy shifts from democracy support
to conflict management and prevention, notably the neglected relationship
between democracy and violent conflict. Finally, the failure to prevent viol-
ent conflict may be attributed to a bias for conflict management at the
expense of prevention. For all the pride of place that has been given to con-
flict prevention, reactive decision-making remains as salient as ever. All of
this seems to underscore the harsh verdict that there ‘is no discernible
project that the West somehow seeks to impose on Africa. Rather, the
Western posture is one of seeking to be engaged at low cost.’39
Western attempts at resolving ongoing conflicts are characterized by lim-
ited engagement. Confronted with increasingly effective pressure from the
media and international human rights groups to ‘do something’, standing
back and letting conflicts run their course is not an option which Western
governments can easily contemplate.40 At the same time, the West is dem-
onstrating an obvious reluctance to be sucked in to African conflicts, par-
ticularly after the disastrous experiences in Somalia and Rwanda. Crisis
diplomacy to influence the turn of events in insurgency-affected countries
has become the option of choice to find a compromise between the coun-
tervailing logics of appropriateness and consequentiality.
This reactive behaviour has typically taken the form of power-sharing
agreements between embattled governments and insurgencies.41 A specific
instrument of conflict mediation, power-sharing agreements are usually
brokered in stalled conflicts where neither side has the military clout to
decisively defeat the other. They include the negotiating of a peace settle-
ment between incumbents and rebels that provides for the partition of
power within a government of national unity. This is followed by provi-
sions for a political transition whose end-point is multiparty elections.
Peace settlements are often accompanied by the deployment of a United
Nations peacekeeping mission to support the transition. The logic of
power-sharing rests on the assumption that the accommodation of the

39. Herbst and Mills, ‘The future of Africa’, p. 40.

40. See, for example, the evolution of the ‘public pressure curve’ that rose in regard to the
Ituri conflict and the ethnic cleansing in Darfur.
41. Ian S. Spears, ‘Understanding inclusive peace agreements in Africa: the problems of
sharing power’, Third World Quarterly 21, 1 (2000), pp. 105–18.
 THE HIDDEN COSTS OF POWER-SHARING 387

demands of anti-regime movements has the potential to demilitarize the

political context. Likewise, it is believed that power-sharing institutions
‘promote moderate and cooperative behaviour among contending groups
by fostering a positive-sum perception of political interactions’.42
To date, the record of power-sharing agreements appears to be mixed at
best, not least because international mediators have assumed that their job
is all but finished with the signing ceremony of peace accords.43 This art-
icle cannot offer a critical survey of the effectiveness of internationally bro-
kered power-sharing agreements in regard to the restoration of peace.
Rather, it seeks to outline some of the fallacies and unintended conse-
quences that outsiders’ promotion of power-sharing as a general practice of
peacemaking may unfold. We shall start with an overview of the changed
international context in which contemporary insurgents are operating.

Conquest by power-sharing
The role of insurgency movements in world politics underwent dramatic
changes in the post Cold War-period, enhancing their international
standing.44 Prior to the collapse of the Soviet Union in 1989, the super-
powers regarded insurgency movements as useful instruments to wage
proxy battles in many parts of the southern hemisphere. While strategic
interests therefore lent international weight to rebel groups, the exigencies
of a bipolar international system and international conventions still put a
premium on the incumbent government of a sovereign state. Since even
Cold War competition was framed by juridical sovereignty, external support
for insurgents faced certain limits, and state rulers — those controlling the
capital – were clearly in an advantageous position. Their incumbent status
guaranteed an access to outside resources (military, economic, diplomatic)
far superior to those available to their rebel challengers.
These mechanisms have undergone a major shift since the end of the
Cold War, even as the ‘negative sovereignty’ associated with Third World
polities continues to protect the state as a juridical entity. In the process,
the state’s sovereignty appears to have been somewhat de-linked or even
separated from those claiming to represent it, i.e. incumbent governments.

42. Caroline Hartzell and Matthew Hoddie, ‘Institutionalizing peace: power sharing and
post-civil war conflict management’, American Journal of Political Science 47, 2 (2003), p. 318.
43. See Stephen John Stedman, Donald Rothchild and Elizabeth M. Cousens (eds), Ending
Civil Wars: The implementation of peace agreements (Lynne Rienner, Boulder, CO, 2002);
Timothy D. Sisk, Peacemaking in Civil Wars: Obstacles, options, and opportunities Occasional
Paper (The Joan B. Kroc Institute for International Peace Studies, South Bend, IL, 2001);
Roland Paris, At War’s End: Building peace after civil war (Cambridge University Press,
Cambridge, 2004).
44. The introduction to this section relies heavily on Christopher Clapham’s work. See his
Africa and the International System (Cambridge University Press, Cambridge, 1996), chap. 9,
and ‘Degrees of statehood’, Review of International Studies 24 (1998), pp. 143–57.
388 AFRICAN AFFAIRS

The imposition of economic conditionalities by Western governments and

the international financial institutions as well as their increasing collabora-
tion with non-state actors (primarily NGOs), indeed the sidelining of the
state, notably in the field of ‘development’, presents a vivid illustration of
this change.45
The most spectacular marker of this shift relates to international atti-
tudes towards insurgency movements since 1989. As a rule, their interna-
tional status reached even higher levels over the last decade or so, but this
rise occurred for reasons that differed greatly from those factors that
underpinned the importance of rebel movements during the Cold War era.
Since internal conflicts were no longer defined as a function of superpower
competition, the raison d’être of waging proxy wars was no longer clear.
Rather than seeking to win insurgency/civil wars (over the incumbent gov-
ernment allied with the rival superpower), outsiders, and Western states in
particular, espoused at least a rhetorical commitment to bringing internal
wars to an end.
A sharp decline of vital interests and changes at the international level,
the domestic level, or sometimes both (for example, France), was accom-
panied by the rolling back of Western involvement in the domestic affairs
of African states. Significantly, this was also the case in countries that wit-
nessed armed conflicts or the emergence of insurgencies. As a result, the
very same reasons that inclined outside powers to abjure fighting proxy
wars also meant that strong political engagement to prevent and solve viol-
ent conflict in Africa was unlikely. This development signalled an import-
ant departure from previous external attitudes, to the extent that
incumbents no longer enjoyed solid access to outside support and hence
guarantees of political survival, even on judicial grounds such as incum-
bency or sovereignty which had given them a distinct advantage in the
Cold War context. Conversely, armed non-state movements were more
readily integrated into the international relations of sovereign states as out-
siders attempted to solve violent conflict in war-torn countries. Because of
limited interests, Western actors turned to the instrument of power-sharing
to terminate the fighting, an approach that requires by definition the recog-
nition of all the warring parties, including the insurgents. Thus, it entails a
major adjustment in the domestic balance of power, since external actors
level the political playing-field in favour of insurgents at the expense of
state leaders:

Instead of regarding one party as representing the state, and the others as opposing it,
external mediators came to conceive all the parties as subsisting on a more or less
equal footing; their function in turn was no longer to protect those who could

45. Firoze Manji amd Carl O’Coill, ‘The missionary position: NGOs and development in
Africa’, International Affairs 78, 3 (2002), pp. 578–80.
 THE HIDDEN COSTS OF POWER-SHARING 389

claim . . . to represent the state, but rather to achieve a political settlement through
recognition of all the competing parties.46

This is not to say that outsiders were always willing to engage with insur-
gents on political terms and that power-sharing was in every instance a
foregone conclusion. However, it is significant that the neighbouring coun-
tries of insurgency-affected countries in Africa often showed more hostility
towards insurgencies than the Western-led international community. Such
cases include, for example, the military interventions of Senegal in Guinea-
Bissau (1998) and the joint intervention by South Africa and Botswana in
Lesotho (1998). These interferences were driven by concerns about polit-
ical and economic instability in their immediate neighbourhood (Lesotho),
the imminent displacement of friendly regimes by insurgents suspected of
supporting rebels in the intervening country (insurgent General Mane and
the Casamançais rebels in Guinea-Bissau/Senegal) and, last but not least,
the prospect of a presumed easy military victory over the insurgencies.
In another set of cases, sub-regional responses were less resolute. In
Liberia, the Economic Community of West African States peacekeeping
force ECOMOG prevented Charles Taylor’s military victory in 1990 but
showed little inclination to give President Doe decisive support. Even less
help came from Doe’s former key ally, the United States. Given its sharp
drop of interests in Liberia after 1989, it did no more than evacuate its res-
idents from Monrovia although the deployment of 2,000 Marines had cre-
ated other expectations.47 After a dozen unsuccessful attempts to broker a
peace settlement, the Abuja accords ushered in a transitional arrangement.
Very similar responses occurred in Sierra Leone. First, the international
community refused to recognize the Koroma regime that had overthrown
President Kabbah in May 1997. Likewise, ECOMOG troops dislodged the
Revolutionary United Front (RUF) insurgents from Freetown in January
1999. Even so, the Nigerian government was unwilling to extend the pres-
ence of its ECOMOG troops in the country after the 1999 Nigerian elec-
tions, although it was commonly known that the RUF was not a spent
force. Nigeria as well as the US urged Kabbah to negotiate with the rebels,
a move that resulted in the Lomé agreement.48
The insurrection that toppled President Patassé in the Central African
Republic in 2003 was condemned by the African Union but was quickly
recognized by the Communauté économique et monétaire de l’Afrique centrale
(CEMAC). The French (technical) support for the CEMAC peacekeepers
who did not intervene to save Patassé was a clear indication that the rebels

46. Clapham, ‘Degrees of statehood’, p. 153.

47. Stephen Ellis, The Mask of Anarchy: The destruction of Liberia and the religious dimension of
an African civil war (Hurst & Co, London, 1999), p. 4.
48. William Reno, ‘The failure of peacekeeping in Sierra Leone’, Current History (May
2001), pp. 220f.
390 AFRICAN AFFAIRS

of former chief of staff François Bozizé had secured the necessary degree of
outside support. France’s military intervention in Côte d’Ivoire in 2002 was
characterized by similar ambiguities. Paris prevented the military victory of
the insurgency, but it refused to take sides with the Gbagbo government.
The ‘containment approach’ to internal conflicts requires not only exter-
nal recognition of insurgents. By putting a premium on violence, it also
leaves civilian opposition parties in an uneasy position. Irrespective of their
participation in negotiations, they are either forced to take a pro-government
position (and risk being dealt with as a negligible partner) or a pro-rebellion
position (exposing themselves to all sorts of accusations).49
From the perspective of the West, the only way to overcome the conundrum
of limited interests and the urge ‘to do something’ has been a low-key engage-
ment, whereby militarily effective insurrections are conceived of as legitimate
stakeholders in domestic power struggles. Regardless of their often appalling
human rights record, their power warrants international consideration.
Perhaps the most infamous case is that of the RUF rebels in Sierra
Leone, who were incorporated into a government of national unity follow-
ing the July 1999 Lomé peace agreement. Similar processes are currently
taking place in Burundi, where Western actors continue to press for inclu-
sive peace talks at any cost. Most recently, and disturbingly so, the European
Union allegedly continued to urge the Forces nationales de libération (FNL)
to join peace talks even in the immediate wake of the assault on a refugee
camp in August 2004 which left some 160 people dead and for which the
FNL had claimed responsibility.50
Early and recurrent examples of a rather successful strategy to profit
from foreign-sponsored (but nationally-brokered) power-sharing agree-
ments are to be found in Chad. There, politico-military movements pro-
vide illustrations of violent rent-seeking, inasmuch as the change from
civilian party agent to rebel leader (and back again) is common. Both of
the country’s most prominent politicians — President Idriss Déby and the
chairman of the National Assembly, ‘General’ Abdelkader Kamougué —
were at one point ‘re-civilized’ warlords. The two rebel movements, Comité
de sursaut national pour la paix et la démocratie (under Moïse Ketté) and its
offspring, the Forces armées pour la république fédérale, were recognized as
political parties in 1994 and 1998.51 Ketté enjoyed ministerial rank before

49. In the case of Côte d’Ivoire the northern rebellion effectively succeeded where the RDR
had failed: the French-brokered Marcoussis peace accord in January 2003 addressed both the
main structural causes of the conflict and the northern grievances. But the equally French-
brokered power-sharing arrangement at the following Kléber summit (with the entry of rebel
ministers into the government) was widely interpreted as putting a premium on violence (and
discredited the French mediation in the eyes of some major players).
50. ‘UN weighs situation in Burundi following massacre’, Reuters, 15 August 2004. On the
Burundian peace process, see the reporting of the International Crisis Group.
51. However, in the case of FARF this happened only after the murder of its leader Laokein
Bardé who was probably betrayed by his own followers.
 THE HIDDEN COSTS OF POWER-SHARING 391

slipping out of the sinecure system and founding a new rebel movement
(before being murdered in 2001). This is the purest form of politico-milit-
ary entrepreneurship in that it entails blackmailing sinecures by military
means.
Strategies of shifting from peaceful to violent opposition (and back
again) can be very helpful to prove a certain nuisance capacity as the basis
for attracting rents of violence, namely inclusion in government. The com-
plement for lower ranks is the forcing of material rewards in the framework
of demobilization programmes.52 The bill is usually paid by some donor
organizations hoping — frequently in vain — to contribute to sustainable
peace. The opposite might be actually closer to the truth: rent-creating fos-
ters rent-seeking behaviour. While spoilers can hardly be ignored, it is
highly dangerous simply to reward them.
By the same token, the three mutinies (1996–97) in the Central African
Republic can be partly explained by the ‘rents of violence’ syndrome. The
crisis was temporarily ended by forming a coalition government including
all political camps and the rebels. Former heads of state and chairmen of
political parties Dacko and Kolingba received substantial state pensions —
inducements to renounce violence after threatening to use it.53 In a coun-
try where recurrent budgetary aid, mainly from France, is necessary to pay
part of the salary arrears in the civil service (at one point peaking at 29
months of unpaid salaries), this arrangement amounted to a foreign-
sponsored subvention for these two leaders, whereas the main causes of the
conflict were hardly addressed.
In summary, significant changes in Western foreign policies toward
Africa have emerged since the early 1990s. These relate first to the some-
what tacit pressure for democratization beyond the level of electoral proce-
dures and, second, to the international recognition of insurgency
movements, to the detriment of embattled governments of weak states and
of the civilian opposition. Both aspects touch upon domestic processes that
have arguably dominated political events over the last 15 years: political
reform and violent conflict. The relationship we hypothesize between
flawed democracy and violent conflict is that the road to state power in
electoral autocracies is usually closed to non-violent political actors. The
struggle for access to state power by opposition politicians, often pursued
at great personal risk, cannot pay off. At the same time, opposition politi-
cians and other would-be leaders observe radically different responses from
Western governments to violent political action in insurgency-affected

52. See, for example, Mike Crawley, ‘Fewer guns, but tensions persist in Liberia’, The Christian
Science Monitor, 28 October 2004.
53. Andreas Mehler, ‘Meuterei der Armee und Tribalisierung von Politik in der “demokra-
tisierten Neokolonie” Zentralafrikanische Republik (ZAR)’, in Heidrun Zinecker (ed.),
Unvollendete Demokratisierung in Nichtmarktökonomien (Fakultas, Amsterdam, 1999), p. 208.
392 AFRICAN AFFAIRS

countries, where strongmen embarking on the rebel path impose them-

selves by establishing political facts which the West is usually willing to
accept. Western buy-in tactics to engage militarily effective insurgencies
and to accommodate their demands by advocating their incorporation into
national governments for the sake of ‘peace’ amount to an incentive to
have recourse to violence to conquer state power, or at least to receive a
seat at the bargaining table.54 Over the past 15 years, this logic was at work
in Burundi, Rwanda, the DRC, Sierra Leone, Liberia, Côte d’Ivoire,
Sudan, Central African Republic and Chad.
As a result, the behaviour of would-be leaders turned insurgents is partly
a response to incentive structures that outside actors are unwittingly estab-
lishing. This is part of the extraversion strategies used to assert new claims
on resources and to authority which are far from being the privileged
domain of state rulers.55 This observation suggests that the hypothesized
relationship between the insurgent path and Western engagement with
violent non-state actors is far from speculative. What could provide more
compelling evidence than the fact that insurgents go to great length to set
up and maintain international relations with foreign governments, interna-
tional NGOs and transnational actors?56 Should the hypothesis concerning
the interplay between external incentives and insurgent violence be valid, it
reflects a troubling political development that bodes ill for the prevention
of conflicts in Africa.
Beyond the demonstration effects individual cases set for would-be lead-
ers in other countries, it is also questionable whether power-sharing agree-
ments are truly conducive to the establishment of peace inside war-torn
countries. First, some faction leaders taking part in negotiations seem
barely interested in achieving conclusive agreements. Rather, they particip-
ate in seemingly endless negotiations for the sake of material benefits, most
notably in Somalia, where the most powerful leaders of armed groups have
held peace talks for many years even though the resurrection of a Somalian
state is clearly not in their interest.57 Second, even where political settle-
ments are achieved, any chance to return to some degree of normality
seems to be remote. Power-sharing formulas allow many rebel leaders to
behave in office much as in wartime (i.e. Sankoh, Taylor), thus tapping the

54. Reno, ‘The failure of peacekeeping’.

55. Robert Latham, Ronald Kassimir and Thomas M. Callaghy, ‘Introduction: transboundary
formations, intervention, order, and authority’, in Thomas M. Callaghy, Ronald Kassimir,
Robert Latham (eds), Intervention and Transnationalism in Africa (Cambridge University
Press, Cambridge, 2001), p.13; Jean-François Bayart, ‘Africa in the world: a history of extra-
version’, African Affairs 99, 395 (2000), pp. 217–67.
56. See Clifford Bob, ‘Marketing rebellion: insurgent groups, international media, and
NGO support’, International Politics 38 (2001), pp. 311–34.
57. Martin Doornbos, ‘State collapse and fresh starts: some critical reflections’, Development
and Change 33, 5 (2002), p. 807.
 THE HIDDEN COSTS OF POWER-SHARING 393

combined revenues deriving from peace and sovereignty rents as well as

war economies.
Third, some agreements are so complicated that they effectively defy
implementation. Without significant and long-term external backing,
these agreements are bound to collapse and ignite a return to war. Fourth,
the notion of inclusion that underpins the concept of power-sharing runs
the risk of generating ever more insurgent groups that are not included in
any given settlement. Again, the Congo war and the trajectory of the RCD
insurgency are revealing in this regard, for the Lusaka accord conveyed a
powerful and unanimous message to some parties to the conflict. While
insurgencies may be prone to defections, it is no coincidence that as soon
as the political terms of Lusaka (power-sharing, transitional government)
had been established, the defections from the RDC and the proliferation
of smaller insurgencies started in earnest, including the RCD-National
and the RCD-ML which progressively fragmented even further into fac-
tions led by Wamba, Tibasima, Nyamwisi and Lubanga, striving to
become rebel leaders in their own right. Given the underlying logic of
power-sharing agreements according to which all armed insurgents are to
be included in negotiations, these personalities understandably expected
to be treated accordingly by the mediators; indeed, this was the very rea-
son the new rebel groups were created.58 By sticking to the principle of
inclusive negotiations for the sake of peace, the mediators were confronted
with the problem of accommodating the demands of an ever growing
number of factions. In the end, protracted discussions were required to
decide whether and how the new factions should be included. Given that
only a small minority of embattled governments is confronted by one
insurgency movement alone, the proliferation of rebel groups amidst
ongoing internal wars may well be an indirect result of looming power-
sharing accords.

Conclusion
In the early 1990s, there was much talk both inside and outside Africa
to the effect that violent takeovers of state power would no longer be per-
mitted. Bodies such as the Organization of African Unity even declared
that putsch leaders would no longer be granted outside recognition.
Intended to discourage army officers from staging coups d’état, such
laudable discourses now seem to be outdated, to the extent that violence
remains an effective instrument to access state power in Africa which is

58. A similar point is made by Kidane Mengisteab, ‘Africa’s intrastate conflicts: relevance
and limitations of diplomacy’, African Issues 32 (2004), pp. 37f.
394 AFRICAN AFFAIRS

subsequently legitimized by external actors in their hapless attempts to

limit the damage.59
It is our conclusion that external efforts to terminate internal warfare
may be as much part of the problem as they are part of the solution. Like
emergency assistance and development aid, conflict management is an
immensely complex field of political intervention in which moral ambitions
alone are not sufficient to create desired outcomes.60 This is so because the
indiscriminate use of power-sharing agreements as an instrument of peace-
making brings unintended but potentially dangerous consequences. The
institutionalization of power-sharing creates a degree of predictability for
politically ambitious entrepreneurs. Falling short of outright military vic-
tory, insurgents can reasonably expect to receive parcels of state power in
return for ‘peace’. A host of power-sharing agreements in Africa since the
early 1990s has set numerous precedents which have created an opportun-
ity structure for violent entrepreneurs elsewhere. Neglecting this contex-
tual dimension carries the risk of unnuanced reporting, whereby insurgents
are either portrayed as ‘freedom fighters’ or, more often, as ‘greed-driven
warlords’. Making sense of contemporary insurgencies therefore begs the
consideration of the nexus of internal and external factors and its bearing
on the motives, trajectory and outcome of any given insurgency. Perhaps
counter-intuitively, attempts at conflict resolution have thus undermined
the perspective of conflict prevention.
As a result, outside actors keen to resolve violent internal conflicts in
Africa face a profound dilemma. To overcome it, two rather bold solutions
come to mind: first, let conflicts run their course; second, always provide
support (diplomatically, militarily) to incumbent regimes attacked by
insurgents. The first one echoes Luttwak’s (in)famous proposal to ‘give
war a chance’, which argues that inept meddling by outsiders has often
postponed peace and perpetuated war and human suffering.61 One
observer of the Ivorian crisis — even before the events of November 2004 —
reiterated this reasoning in the following terms: ‘Interposition, in this kind
of case, is a guarantee for lasting crisis. [France] should have either
crushed the rebellion or let it go on.’62

59. In accordance with its protocol, the Economic Community of West African States
(ECOWAS) decisively intervended to reverse the coup d’état in Togo and to re-establish the
constitutional order in early 2005. Intriguingly, however, even before the results of the April
elections were announced, the chairman of the African Union, Olusegun Obasanjo, convened a
meeting between Faure Gnassingbé and opposition leader Gilchrist Olympio in Abuja in the
vain hope to broker the formation of a government of national unity in Togo irrespective of the
outcomes of the presidential polls.
60. On emergency assistance, see Mary B. Anderson, Do No Harm: How aid can support peace
or war (Lynne Rienner, Boulder, CO, 1999); on development aid, see Peter Uvin, Aiding
Violence: The development enterprise in Rwanda (Kumarian Press, West Hartford, CT, 1998).
61. Edward N. Luttwak, ‘Give war a chance’, Foreign Affairs 78 (1999), pp. 36–44.
62. Quoted in ‘Une crise test pour les relations France-Afrique’, Le Figaro, 15 October 2004.
 THE HIDDEN COSTS OF POWER-SHARING 395

Although the logic underpinning Luttwak’s argument is somewhat com-

pelling, its limits are numerous. Most notably it neglects the fact that many
‘civil wars’ are not — strictly speaking — internal, a point which accounts
for their durability and even self-sustaining character (for example, Angola,
Sudan). Letting them ‘bleed out’ hardly passes for the truly humanitarian
thing to do. Again invoking the recent case of Côte d’Ivoire, it is by no
means a foregone conclusion that an eventual capture of Abidjan by the
insurgents would have ended the war. As for providing exclusive support to
incumbent governments, it is self-evident that such a policy will almost
inevitably entrench autocratic regimes, in turn forestalling any chance of
promoting positive political change.
Outside actors need to recognize that the short-term quelling of conflicts
in the guise of power-sharing is not to be confounded with peace. In Sierra
Leone as well as Liberia, for example, power-sharing agreements (Lomé
and Abuja) resulted in each case in a disastrous and unsustainable ‘war-
lord’s peace’ which triggered further insurgent violence.63 Given the gross
and systematic human rights abuses the rebels committed during the war,
it was short-sighted to presume that they would change their attitudes once
they were occupying government offices.
Since more conflicts will surely erupt and peace settlements in some
form or another will invariably be brokered, we therefore put forward two
proposals. First, external brokers need to raise the threshold which grants
insurgents a place at the negotiating table. As such, it is imperative to think
beyond violence as the primary measure of political inclusion. Armed
groups preying on local communities and committing serious human rights
abuses should be disqualified as negotiating partners. By contrast, some
rebels provide some measure of order or even collective goods such as
security, and they should therefore receive a political premium in negotia-
tions, for they come at least close to carrying out functions that a government
is supposed to fulfil.
Needless to say, efforts to promote accountability and legitimacy in the
field of conflict resolution will not prevent violent entrepreneurs from con-
quering state power, but they are at least a step to limiting the lawlessness
and impunity that characterizes insurgency-affected countries even after
the conflict is terminated. One way to promote accountability even during
conflicts would be the institutionalization of criminal investigations to be
undertaken by internationally sanctioned juridical bodies (for example,
special courts such as in Sierra Leone; or the International Criminal Court
in the case of the DR Congo) in the wake of or even amidst every internal

63. Amos Sawyer, ‘Violent conflicts and governance challenges in West Africa: the case
of the Mano River basin area’, Journal of Modern African Studies 42, 4 (2004), p. 451;
Adekeye Adebajo, ‘Liberia: a warlord’s peace’, in Stedman et al., Ending Civil Wars, pp.
599–630.
396 AFRICAN AFFAIRS

conflict.64 The threat of internationally-backed criminal investigations

against all warring parties may help to restrain the worst abuses. It should
send a message to would-be leaders that raw power is not sufficient to gain
international recognition as stakeholders in national power struggles. For if
they are unable or unwilling to live up to certain standards, it is safe to
assume that they will be unlikely to play a constructive part in the post-
conflict period.
All of this does not release Western actors from the necessity more gen-
erally to rethink their policies towards Africa. In light of both the poor
record of conflict resolution as well as post-conflict peace-building where
significant resources and energy are ineptly used and, perhaps worse,
invested only after the fact, this should be an urgent task.65 Power-sharing
agreements are not the place to start. There is a need to bring democracy
back in or at least to promote a framework for mutually acceptable ways to
access power. Governments and civilian opposition parties should receive
credit for respecting accepted rules and not for bending them. This holds
for intrusive neighbours too. A less technical and more politically-informed
Western approach to promoting accountability in the fields of the rule of
law, election monitoring, democracy assistance and administrative reform
would be helpful in this regard. The central dilemma of outsiders is not the
degree to which they seem to exhibit too much willingness to engage with
insurgents and too little determination to promote democratic politics on
the continent. The key challenge is the development of a qualitative kind of
political engagement that espouses a long-term vision surmounting the ser-
ious inconsistencies and even mutual contradictions that have, by and
large, characterized Western approaches to conflict resolution and support
for democracy.

Bibliography of books and articles

References to other sources, including interviews, archives, newspaper articles, websites and grey
publications, are contained in relevant footnotes.

Adebajo, Adekeye, ‘Liberia: a warlord’s peace’, in Stedman et al., Ending Civil

Wars, pp. 599–630.
Anderson, Mary B., Do No Harm: How aid can support peace or war (Lynne Rienner,
Boulder, CO, 1999).
Anonymous (C.D.G.), ‘L’armée ivoirienne: de la marginalisation à la prise du
pouvoir’, Afrique Contemporaine 193 (2000), pp. 9–12.

64. Andrew Rigby, Justice and Reconciliation: After the violence (Lynne Rienner, Boulder, CO,
2001); Jacquie Cassette, ‘Towards justice in the wake of armed conflicts? The evolution of
war crimes tribunals’, African Security Review 9, 5/6 (2000), pp. 24–34.
65. Paris, At War’s End; Marina Ottaway, ‘Rebuilding state institutions in collapsed states’,
Development and Change 33, 5 (2002), pp. 1001–23; ‘Liberia: a shattered nation on a long
road to recovery’, IRIN News, 24 August 2004.
 THE HIDDEN COSTS OF POWER-SHARING 397

Bayart, Jean-François ‘Commentary: towards a new start for Africa and Europe’,
African Affairs 103, 412 (2004), pp. 453–8.
Bayart, Jean-François, ‘Africa in the world: a history of extraversion’, African Affairs
99, 395 (2000), pp. 217–67.
Bob, Clifford, ‘Marketing rebellion: insurgent groups, international media, and
NGO support’, International Politics 38 (2001), pp. 311–34.
Bratton, Michael, and Nicolas van de Walle, Democratic Experiments in Africa
(Cambridge University Press, Cambridge, 1997).
Cassette, Jacquie, ‘Towards justice in the wake of armed conflicts? The evolution
of war crimes tribunals’, African Security Review 9, 5/6 (2000).
Clapham, Christopher (ed.), African Guerillas (James Currey, Oxford, 1998).
Clapham, Christopher, ‘Introduction’, in Clapham, African Guerillas, pp. 1–18.
Clapham, Christopher, ‘Degrees of statehood’, Review of International Studies, 24
(1998), pp. 143–57.
Clapham, Christopher, Africa and the International System (Cambridge University
Press, Cambridge, 1996).
Clapham, Christopher, and John A. Wiseman, ‘Conclusion: assessing the prospects
for the consolidation of democracy’, in John A. Wiseman (ed.), Democracy
and Political Change in Sub-Saharan Africa (Routledge, London, 1995),
pp. 220–32.
Collier, Paul, and Anke Hoeffler, Greed and Grievance in Civil War (World Bank,
Washington, DC, 2001).
Cramer, Chris, ‘Homo economicus goes to war: methodological individualism,
rational choice and the political economy of war’, World Development 30,
11 (2003), pp. 1845–64.
De Villers, Gauthier, with Jean Omasombo and Erik Kennes, Guerre et politique: Les
trente derniers mois de L.D. Kabila (Institut Africain, Tervuren, 2001).
Doornbos, Martin, ‘State collapse and fresh starts: some critical reflections’, Devel-
opment and Change 33, 5 (2002), pp. 797–815.
Ellis, Stephen, ‘The old roots of Africa’s new wars’, International Politics and
Society 2 (2003), pp. 29–43.
Ellis, Stephen, The Mask of Anarchy: The destruction of Liberia and the religious
dimension of an African civil war (C. Hurst & Co., London, 1999).
Hartzell, Caroline, and Matthew Hoddie, ‘Institutionalizing peace: power sharing
and post-civil war conflict management’, American Journal of Political Sci-
ence 47, 2 (2003), pp. 318–32.
Herbst, Jeffrey, and Greg Mills, The Future of Africa: A new order in sight? Adelphi Paper
No 361 (International Institute of Strategic Studies, London, 2003).
Kieffer, Guy-André, ‘Armée ivoirienne: le refus du déclassement’, Politique Afric-
aine 78 (2000), pp. 26–44.
Latham, Robert, Ronald Kassimir and Thomas M. Callaghy, ‘Introduction: trans-
boundary formations, intervention, order, and authority’, in Callaghy,
Kassimir, Latham (eds), Intervention and Transnationalism in Africa
(Cambridge University Press, Cambridge, 2001), pp. 1–20.
Lemke, Douglas, ‘African lessons for international relations theory’ World Politics,
56 (2003), pp. 114–38.
Luttwak, Edward N., ‘Give war a chance’, Foreign Affairs 78 (1999), pp. 36–44.
Manji, Firoze, and Carl O’Coill, ‘The missionary position: NGOs and develop-
ment in Africa’, International Affairs 78, 3 (2002), pp. 567–83.
Marshall, Monty G., and Ted Gurr, Peace and Conflict 2003: A global survey of
armed conflicts, self-determination movements, and democracy (College Park,
University of Maryland, 2003).
398 AFRICAN AFFAIRS

Mehler, Andreas, ‘Meuterei der Armee und Tribalisierung von Politik in der
“demokratisierten Neokolonie” Zentralafrikanische Republik (ZAR)’, in
Heidrun Zinecker (ed.), Unvollendete Demokratisierung in Nichtmarktökon-
omien (Fakultas, Amsterdam, 1999), pp. 193–211.
Mengisteab, Kidane, ‘Africa’s intrastate conflicts: relevance and limitations of
diplomacy’, African Issues 32 (2004), pp. 25–39.
Mkandawire, Thandika, ‘The terrible toll of postcolonial “rebel movements” in
Africa: towards an explanation of the violence against the peasantry’,
Journal of Modern African Studies 40, 2 (2002), pp. 181–215.
Olsen, Gorm Rye, ‘Promoting democracy, preventing conflict: the European Union
and Africa’, International Politics 39 (2002), pp. 311–28.
Ottaway, Marina, ‘Rebuilding state institutions in collapsed states’, Development
and Change 33, 5 (2002), pp. 1001–23.
Paris, Roland, At War’s End: Building peace after civil war (Cambridge University
Press, Cambridge, 2004).
Reno, William, ‘The politics of insurgency in collapsing states’, Development and
Change 33, 5 (2002), pp. 837–58.
Reno, William, ‘The failure of peacekeeping in Sierra Leone’, Current History (May
2001), pp. 219–25.
Reyntjens, Filip, La guerre des Grands Lacs (L’Harmattan, Paris, 1999).
Rigby, Andrew, Justice and Reconciliation: After the violence (Lynne Rienner, Boulder,
CO, 2001).
Sandbrook, Richard, Closing the Circle: Democratization and development in Africa
(Zed Books, London, 2000).
Sawyer, Amos, ‘Violent conflicts and governance challenges in West Africa: the
case of the Mano River basin area’, Journal of Modern African Studies 42,
4 (2004), pp. 437–63.
Sisk, Timothy D., Peacemaking in Civil Wars: Obstacles, options, and opportunities,
Occasional Paper. (The Joan B. Kroc Institute for International Peace
Studies, South Bend, IL, 2001).
Smith, Stephen, ‘La politique d’engagement de la France à l’épreuve de la Côte
d’Ivoire’, Politique Africaine 89 (2003), pp. 112–26.
Spears, Ian S., ‘Understanding inclusive peace agreements in Africa: the problems
of sharing power’, Third World Quarterly 21, 1 (2000), pp. 105–18.
Stedman, Stephen John, Donald Rothchild, and Elizabeth M. Cousens (eds), Ending
Civil Wars: The implementation of peace agreements (Lynne Rienner, Boulder,
CO, 2002).
Stedman, John Stephen, ‘Introduction’, in Stedman, Rothchild, Cousens, Ending
Civil Wars, pp. 1–40.
Tull, Denis M, The reconfiguration of political order in Africa: a case study from
North Kivu (DR Congo) (Institute of African Affairs, Hamburg, 2005).
Uvin, Peter, Aiding Violence: The development enterprise in Rwanda (Kumarian Press,
West Hartford, CT, 1998).
Van de Walle, Nicolas, African Economies and the Politics of Permanent Crisis, 1979–
1999 (Cambridge University Press, Cambridge, 2001).
Young, Crawford, ‘The end of the post-colonial state in Africa? Reflections on
changing African political dynamics’, African Affairs 103, 410 (2004),
pp. 23–49.