If
a storm can be described as perfect, then the war in the Democratic
Republic of Congo (formerly Zaire) in the second half of the 1990s
was the “perfect war”. Precipitated by the 1994 genocide
in Rwanda and the fall of the West’s client kleptocrat, President
Mobutu, and his rotten state, the war in DR Congo was dubbed Africa’s
First World War. It directly involved the armed forces of six neighbouring
states. It drew in factions and rebel groups from other African
wars, the remnant armies of defunct neighbouring regimes, and the
usual crowd of international profiteers, would-be peacemakers and
humanitarians. It was closely connected with armed conflicts in
several neighbouring countries, including those in Rwanda, Burundi,
Uganda, Central African Republic, Congo Brazzaville, and Angola.
According to one estimate published in 2003 the war may directly
and indirectly have caused the deaths of over 4 million people in
DR Congo since 1996. As has become increasingly common in Africa
the victims were almost all civilians.
 |
Soldiers train on the road to Musaki village in the Kivu region of the Democratic Republic of Congo, 2003.
(Photo by Alex Majoli/ Magnum Photos) |
The
war in DR Congo was but one demonstration of the emptiness of the
promises of a post Cold War political and economic renaissance for
Africa. There was, it is true, the remarkable transition from apartheid
to majority rule in South Africa in 1994. There were also instances
of handover to multiparty civilian rule in former one-party or military-ruled
states like Ghana, Tanzania, Senegal, Mali, Zambia, Malawi, and
even Nigeria. But by the end of the 1990s, any political and economic
progress since the end of the Cold War had been overshadowed by
a series of old and new wars that now engulfed many parts of the
continent and were tipping whole regions further into instability
and poverty.
In
West Africa another regional complex of conflicts, also driven by
greed and political disintegration, was in full swing. The late
1990s saw the culmination of the diamond and corruption fuelled
rebellion in Sierra Leone that had been going on for a decade. At
the start of 2000 a recently signed peace agreement in Sierra Leone
was on the brink of failure. Guinea was in danger of being dragged
into the conflict. Liberia, nominally at peace after its own war
in the first half of the decade but little more than a façade
of a state benefiting no one but its gangster-like regime, was still
fomenting conflict in all of its neighbours (Sierra Leone, Guinea
and Côte D’Ivoire) and was itself edging towards a renewed
civil war. Côte D’Ivoire too, once a beacon of prosperity
and stability, was increasingly beset by its own internal political
troubles that were to develop into armed conflict in 2002. Nigeria,
the regional hegemon, was ruled for most of the 1990s by a repressive
and corrupt military regime which thrived in part on fomenting ethnic
and religious tensions.
In
the Horn of Africa, Somalia was still without a central government
almost a decade after the fall of the last one (the Siad Barre regime
which had been backed and armed alternately by both sides in the
Cold War). The vacuum of state authority in Somalia left the country
in a state of low level conflict and chronic economic weakness,
on the one hand vulnerable to external interference and on the other
a source of regional instability. To the north of Somalia, border
skirmishes between Ethiopia and Eritrea developed into full scale
war in 1999. Meanwhile in Sudan, the second phase of the post independence
rebellion was well into its second decade and there were no signs
of resolution. One peace initiative after another had failed.
At
the other end of the continent, in Angola, another war that had
in an earlier phase been fomented by Cold War rivalry was still
raging. Now deprived of their superpower sponsorship, but aided
by international businesses which continued to buy the Angolans’
oil and diamonds and sell them weapons, the leaders of both sides
(MPLA government and UNITA rebels) were plundering the country to
support their war efforts and to fill their foreign bank accounts.
In a country fabulously rich in natural resources, including agriculture,
the majority of the peasant population were living in desperate
poverty, many of them living on food handouts from the international
humanitarian relief system.
Africa’s
wars in the 1990s were all very different in their specifics. But
they shared a number of important characteristics. First, one of
the main underlying causes of these wars was the weakness, the corruption,
the high level of militarization, and in some cases the complete
collapse, of the states involved. Secondly, they all involved multiple
belligerents fighting for a multiplicity of often shifting economic
and political motivations. Thirdly, they all had serious regional
dimensions and regional implications. And fourthly they were all
remarkable for the brutality of the tactics (ranging from mass murder
and ethnic cleansing, to amputation, starvation, forced labour,
rape and cannibalism) used by belligerents to secure their strategic
objectives.
The Failure to Respond
At
the United Nations at the end of the Cold War neither the Secretariat
nor the Security Council were able or willing to resolve these burgeoning
conflicts in Africa. The one UN success in Africa in the post cold
war period, the peace operation in Mozambique, was overshadowed
by the UN’s failure in Angola in 1992. Then the dramatic US/UN
failure in Somalia in 1993-5 rendered the international system helpless
in the face of the emerging crisis in central Africa. The permanent
members of the UN Security Council stood by and watched, with a
negligence verging on complicity as the Hutu nationalist government
went about the systematic slaughter of hundreds of thousands of
minority Tutsis in Rwanda in 1994 and then launched a belated international
humanitarian intervention in eastern Zaire that assisted the genocidaires
and helped to spark the war in Zaire. As one neighbouring state
after another (many of them recipients of large amounts of Western
development aid not to mention “defence” exports) piled
into Zaire/Congo in pursuit of its own political, economic and security
interests, the UN Security Council failed to respond with any significant
political action.
In
West Africa, peacekeeping and peacemaking for most of the 1990s
was left to regional players, notably Nigeria (then ruled by a military
dictatorship) which intervened first in Liberia, then in Sierra
Leone and Guinea Bissau, under the banner of the Economic Community
of West African States (ECOWAS). But like the UN in Somalia, ECOWAS
became as much part of the problem in those conflicts as the solution.
When an elected civilian government came to power in Nigeria in
1999, following the alleged assassination of the billionaire military
ruler Sani Abacha, the ECOWAS operation was withdrawn from Sierra
Leone and replaced by the UN. After the UN’s dismal record
in Africa in the preceding years, Sierra Leone was a do or die test
for the UN’s credibility as a force for peace on the continent.
Within months the operation was in crisis. In May 2000 the Sierra
Leonean rebels, the Revolutionary United Front (RUF), tore up the
peace deal they had signed with the government and kidnapped over
200 UN peacekeepers.
A
Watershed
In
retrospect, the British intervention to rescue the beleaguered UN
mission in Sierra Leone is now seen as something as a watershed
in the recent evolution of conflicts in Africa. Prime Minister Tony
Blair’s decision was taken at short notice in response to
a sudden and unexpected crisis. But it had far reaching consequences
not just for Sierra Leone but for West Africa, and perhaps for the
continent as a whole1. First it reversed the
common understanding that had emerged from Somalia and Rwanda that
Western military powers were not willing to take military and political
risks in Africa unless their vital national interests were at stake.
Secondly, it demonstrated that a professional Western military force
could bring conflicts in Africa under control at relatively little
cost (the UK deployed a few hundred soldiers and lost only one man
in combat operations).
Thirdly,
the British salvaged the UN’s reputation in Africa, not only
by saving it from a humiliating failure, but also by subsequently
insisting that the UN Security Council strengthen the operation
in Sierra Leone so that it should be able to start fulfilling the
mandate of bringing permanent peace to Sierra Leone. Fourthly, in
Sierra Leone the British and the UN provided a rough template for
tackling African wars that went beyond limited humanitarian goals
and tried (so far with only limited success) to get to grips with
some of the more fundamental political, humanitarian, economic and
regional dimensions of a conflict.
In
West Africa the British intervention in Sierra Leone had a direct
impact. Realising that their efforts in Sierra Leone would be brought
to nothing unless Charles Taylor, Liberia’s warlord turned
President, was contained, the UK browbeat the UN Security Council
into imposing economic sanctions on Taylor, a policy which coupled
with the growing internal pressure on the Liberian regime led eventually
in 2003 to Taylor’s removal from office, a peace agreement
between the government and the two rebel factions, and the deployment
of a UN peace mission in Liberia. When the political crisis in neighbouring
Côte D’Ivoire developed into a civil war in 2002, the
example of the British intervention in Sierra Leone encouraged France
to intervene with a military force to stabilise the situation in
its former colony, to fund a West African peace keeping force (with
contributions from the US and Britain), and to play the role of
diplomatic midwife in the Ivorian peace process. There is a real
possibility that without such a response Côte D’Ivoire
would by now have gone the way Liberia and Sierra Leone went in
the early 1990s.
 |
British paratroopers sit in a trench in Lungi Loi in Sierra Leone.
The village was the site of a frontline British post, with the Revolutionary United Front (RUF) rebels controlling territory in the nearby jungle.
(Photo by Yannis Behrakis. Reuters Pictures Archive) |
Towards
Conflict Resolution
Elsewhere
in Africa there have been other encouraging developments in the
past four years. In the Horn, Algeria led a successful US backed
peace mission under the banner of the Organisation of African Unity
(OAU) to mediate an end to the border war between Ethiopia and Eritrea.
Once a deal was signed in 2000, the UN deployed a peacekeeping mission
along the front line to monitor the ceasefire. In Sudan the situation is more complicated. Although a provisional peace agreement has been signed between the government and the southern rebels, perhaps marking an end to the war that has continued since 1983, a new conflict has broken out in the western province of Darfur. This war has been accompanied by repeated attacks against non-Arab civilians by Arab tribal militias known as Janjaweed, who have been supported by the Khartoum government and its armed forces. Outside states that have been influential in promoting the North-South peace process now face the risk that the progress they have achieved will unravel if they punish the Sudanese government for its apparent support of ethnic cleansing in Darfur.
In
the Horn’s most troubled and “failed” state, Somalia,
there has been little progress. The country remains without a central
government and current peace negotiations are unlikely to go anywhere.
However in the breakaway and largely peaceful north western “statelet”
of Somaliland there has been remarkable progress towards peace and
political stabilisation leading to was a smooth and democratic transition
of power following the death of President Mohammad Ibrahim Egal
in 2002.
In
southern Africa, Angola’s war is over. In 2002 the UNITA (National
Union for Total Independence of Angola) rebels were defeated, their
leader Jonas Savimbi was killed, and the rebel leadership accepted
a peace agreement more or less on the government’s terms.
This has been achieved with little or no formal external involvement
(though foreign military advisers helped the government deliver
the military coup de grace to UNITA). Elsewhere in the region in
2002 an incipient conflict in Madagascar after a stolen election
was nipped in the bud thanks to some intense international diplomacy
involving African and non African governments. There has been and
continues to be a serious political deterioration in Zimbabwe. But
in spite of dire warnings of civil war and an undercurrent of political
violence and assassinations this crisis has so far not developed
into a full blown armed conflict nor is it easy to see how it could
in the near future.
In
central Africa one hesitates to speak of an improvement. Certainly
in Burundi, in Eastern DR Congo, in the Central African Republic
and in northern Uganda there has been much killing (mostly of civilians)
in the past couple of years. And yet thanks to a coalition of efforts
by the UN, South Africa, the African Union, Tanzania, Botswana,
the EU, the US and others, there has been some progress in moving
forward peace processes in the main conflicts of the region, namely
in Burundi and in the DR Congo. Most of the foreign troops formally
withdrew from DR Congo according to the terms of a peace agreement
in 2002. After painstaking negotiations a transitional national
government comprising the main political groups was formed in 2003.
The role of the UN (which deployed a peacekeeping force, MONUC,
in 2000) has gradually been strengthened. When UN peacekeepers found
they were unable to deal with an escalation of violence and ethnic
massacres by rival tribal militias in Ituri (Eastern DR Congo) in
2003, the French dispatched a rapid reaction force under the banner
of the EU, with a contribution from the UK, Belgium and others (an
unprecedented arrangement that would have been unthinkable just
a couple of years ago). This force was able to secure the main town
in Ituri and parts of the surrounding region until the UN could
be given the means (mandate and troops) to do the job itself.
Elsewhere
in Central Africa, Africans themselves have sought to impose military
solutions with mixed results: in Congo Brazzaville the Angolan army
intervened successfully to end a civil war by ousting an elected
government and installing their own ally; in northern Uganda the
government’s eighteen month old military campaign to suppress
the rebel Lord’s Resistance Army, has not yet succeeded and
indeed the situation there is again a cause for real concern.
Explanations for Success
It
would be futile to search for a single explanation for what appears
for now to be a trend towards the resolution of African conflicts.
Africa’s wars are as heterogeneous as its many nations and
communities. The reasons why Angola’s conflict came to end
are quite different from the reasons why the belligerents in Sudan’s
civil war have been willing to engage seriously in peace talks.
However some of the successes of the past three years can be attributed
in part to a mixture of fatigue on the part of those fighting African
wars and to the fact that both Africans and non-Africans are learning
lessons from the many failures of the past fifteen years, are coming
up with more creative proposals and solutions to tackle the problem
of conflict, and are readier to take risks in implementing them.
Although the details vary widely from conflict to conflict, the
basic ingredients of resolution remain the same - a combination
of military, diplomatic, humanitarian, and economic action delivered
by a more or less complex coalition of local, regional and international
actors.
At
the UN, following the publication of a hard hitting report2
in 2000 on UN peacekeeping operations, there has been significant
improvement in planning and capacity at the Department for Peacekeeping
Operations. There also appears to be a new readiness on the part
of the Security Council (which prior to the Iraq crisis devoted
a majority of its time to deliberating about conflicts in Africa)
to equip UN missions in Africa and elsewhere with the mandate and
capacity to do their job (once the UN operation in Liberia is fully
deployed, the three largest UN peacekeeping operations in the world
will all be in Africa). Crucially there is also now a UN Security
Council acknowledgement of the imperative to incorporate the protection
of civilians into the mandates of UN and other peace support operations.
Other UN mechanisms have also been deployed to help nudge African
conflicts towards resolution including special panels to monitor
sanctions in Angola, Liberia and Somalia, a special panel monitoring
the links between conflict and the exploitation of resources in
DR Congo, international war crimes tribunals for Rwanda and Sierra
Leone and the establishment of a special UN office headed by a senior
official to examine the many regional dimensions of conflict in
West Africa.
In
Africa itself, there is a new recognition among leaders like South
Africa’s Thabo Mbeki, Senegal’s Abdoulaye Wade, Ghana’s
John Kuffuor, Nigeria’s Olusegun Obasanjo, and the new Chairman
of the African Union (AU), Alpha Omar Konare, of the negative impact
of Africa’s wars on the prospects for the whole continent
and a determination that Africans themselves should play a more
active and creative role in ending wars. As usual there is a lot
of rhetoric. But there is some substance too. The new mood (symbolised
by Mbeki’s ambitious blueprint for an African renaissance,
the New Partnership for African Development3)
has been translated into some successful African peace initiatives
on the ground (e.g. in Burundi and DR Congo) and into a programme
to reform Africa’s own regional security communities4
and to increase Africa’s own peacekeeping capacity.
Among
the big Western powers, particularly the US, the French and the
UK, there has been a greater engagement on Africa at a higher political
level and a greater willingness to bury past differences on Africa
and co-operate. This has been in part because of a real sense of
guilt at the failure of the West to prevent or halt the genocide
in Rwanda in 1994. But it is also because of a new post 9/11 consensus
in the West that, quite apart from humanitarian concerns, there
are compelling strategic reasons (oil, Islam and terrorism) for
preventing Africa from slipping further into poverty and conflict.
In policy terms these factors have led to greater support for Africa’s
own efforts to deal with conflict, support for beefed up UN operations
and, when all else fails, a greater preparedness to commit Western
troops in response to African crises.
The Remaining
Challenges
But,
in spite of these developments, one can hardly say with great conviction
that the tide of conflict in Africa that gathered during the Cold
War and engulfed the continent in the 1990s has finally turned.
The UN and others may now be implementing passably effective crisis
response and conflict resolution strategies, but they are still
struggling to develop effective models for conflict prevention and
for post conflict economic and political transition that can deliver
sustainable peace. Even in Sierra Leone, a relatively small country
where the British government and others have invested millions of
dollars in a large, multi-sectoral peace building effort, the situation
remains precarious and, without long term international commitment,
unsustainable. Ending hostilities and stabilisation have proved
to be the easiest tasks. Much more difficult, time consuming and
expensive is the job of rebuilding a state which has almost completely
disintegrated and creating a political, economic, judicial and social
environment in which the risk of a return to conflict is eliminated,
or at least minimised. If this is hard in tiny Sierra Leone then
it is going to be a many times more so in a DR Congo or a Sudan.
Now
that so many conflicts in Africa are moving towards some sort of
resolution, there is a real opportunity for action. The urgent task
is to regenerate states which have already collapsed into conflict
and to strengthen those which are in danger of doing so. The broad
outlines of what needs to be done are fairly clear. At a state level
the transition can only take place through locally owned processes
of consultation that tackle head on such crucial questions as political
legitimacy and the quality of citizenship and political participation;
transparency, accountability, and the management of economic resources;
the reform of the security sector and the demilitarisation of politics;
the establishment of the rule of law; and reconciliation and accountability
for war crimes.
Concurrently,
at a regional level more work and resources are required to improve
and implement the current programme to reform and expand Africa’s
regional security communities like those envisaged by the AU and
ECOWAS. It is only through regional co-operation that states emerging
from conflict can address important regional dimensions of conflict
like migration, refugees, organised crime and cross border trafficking
in illegally exploited resources and arms. More action at every
level is also needed to tackle the HIV/AIDS crisis which threatens
fatally to undermine state capacity even in stronger African states
and accentuate all the political and economic problems associated
with weak states and internal conflict. The problem is vast, immediate
and highly threatening.
 |
Two survivors of the Rwandan genocide sit in a restaurant in the town of Nyamata in 2004, ten years after the killing. Consolata
Mukantabana, 45 years old, lost her five sons; her husband became mentally disturbed and is unable to work. Tharcissa Mukenshimana, 27
years old, lost all the members of her family apart from two younger brothers.
(Photo by Paolo Pellegrin / Magnum Photos) |
The
industrialised rich countries can assist with all these tasks, problems
and processes - principally with financial and technical support
- but they should be careful to ensure that their political,
economic, humanitarian and military interventions, however well
intended, do not undermine transitional processes by imposing prescriptive
blueprints or by providing support to those forces which stand to
benefit from the weakness of the state and from war. They also need
to do much more, if they are really serious about ending African
conflicts, to deal with those negative aspects of the global economy
and global governance that tend to disadvantage Africa, undermine
African states, and directly or indirectly fuel African wars. These
include: trade barriers in rich countries that block Africa’s
access to rich countries’ markets; insufficient conditionality
on development aid; Africa’s massive debt burden to rich countries
(Africa is still paying more to the industrialised world in debt
servicing than it receives in foreign aid); the unchecked flows
of arms which continue to pour into Africa from manufacturers and
dealers in the West; and the gaping legislative loopholes which
continue to allow multinational companies and banks to benefit from
the abusive, sometimes criminal, exploitation of resources and from
high level corruption in conflict prone countries in Africa.
At
least some of these key issues are now being debated in international
policy circles thanks in part to sustained lobbying by some NGOs
and a few voices in the media. There was even a vague promise to
address some of these problems in the 2002 G8 Africa Action Plan
(though the plan is carefully drafted to avoid concrete commitments
on the most sensitive issues). Modest progress has been made in
some specific areas: for example the Kimberley Process Certification
Scheme5 to control the trade in “conflict
diamonds”. But there has been nothing significant on trade,
nothing on arms and little on controlling corporate behaviour. Much
more work is required to reconcile the stated commitment of the
leaders of the industrialised world to work to end conflict in Africa
with the political and economic interests that apparently make it
so difficult for them to end agricultural subsidies for their farmers,
to stop propping up corrupt or repressive governments with economic
and military assistance, and to rein in the arms dealers, corporations
and banks which continue to make vast profits from political instability
in Africa.
Finally,
there is now a formidable new challenge: not to let the so-called
“global war on terror”, undermine the progress that
Africa has made in the past few years towards resolving conflicts.
The huge cost of the war in Iraq and its aftermath should not be
allowed to divert attention, money and expertise away from the job
of post conflict reconstruction in Africa. Nor should the US, EU
and UK’s post 9/11 concerns about security and “failed
states” be allowed to skew Western agendas in Africa in favour
of counter-terrorism, military control and oil security, and away
from human rights, social justice, political participation, and
corporate responsibility. Only by promoting the latter can real
peace and security, both for Africa and for the West, be achieved.
Back
to Top
1.
It can be argued that the consequences went even further than Africa.
It is possible that the success of the UK military operation in
Sierra Leone strengthened Mr. Blair’s belief in the viability
of military solutions and thus encouraged him to embrace the idea
that military “humanitarian” intervention in Iraq was
both morally necessary and likely to succeed.
2.
http://www.un.org/peace/reports/peace_operations/report.htm
3.
http://www.nepad.org
4.
The African Union (AU) - which succeeded the Organisation
of African Unity (OAU) in 2000 and is soon to establish its own
a Peace and Security Council - the Economic Community of West
African States (ECOWAS), the Conference on Security, Stability,
Development and Cooperation in Africa (CSSDCA), among others.
5.
http://www.kimberleyprocess.com/
|