Two weeks ago, fighting began in Sudan between the regular military led by General Abdel Fattah al-Burhan and the paramilitary Rapid Support Forces led by Mohamed Hamdan Dagalo, or Hemedti. What initially looked like a violent struggle over military integration and political control is increasingly becoming something more dangerous. Repeated ceasefires have failed to stop sustained fighting, civilians are fleeing Khartoum, hospitals and basic services have been severely disrupted, and foreign governments have evacuated thousands of their citizens.
The obvious solution seems to be negotiation. Both sides are imposing enormous costs on Sudan, neither can guarantee a rapid victory, and continued fighting risks destroying much of the state they are competing to control. Yet this is precisely what makes civil wars difficult to end: the fact that peace would benefit the country does not mean peace is individually safer for the people commanding the armies.
Burhan and Hemedti face a problem economists would recognise as a credible commitment problem. Imagine they negotiate a ceasefire and Hemedti agrees to integrate the RSF into a unified military. Once his independent armed organisation disappears, what prevents Burhan from arresting or marginalising him? Now reverse the situation. If Burhan accepts an arrangement preserving substantial RSF autonomy, what prevents Hemedti from rebuilding his strength and challenging the regular military later? Each side may genuinely prefer peace to war while simultaneously believing that disarming first would leave it vulnerable.
This is why simply asking combatants to “stop fighting” misses the structure of the conflict. Peace requires more than both sides disliking war. Each side must believe the agreement will remain enforceable after it gives up some of the military power currently protecting it. Sudan lacks a neutral domestic institution strong enough to provide that guarantee because the institutions that would normally enforce an agreement are themselves divided between the combatants.
The same mechanism explains why ceasefires can collapse even when both sides publicly accept them. A ceasefire creates an opportunity for humanitarian relief, but it can also give an opponent time to resupply, reposition forces or reinforce strategically important territory. If either side suspects the other is using the pause to gain an advantage, continuing to respect the ceasefire becomes dangerous. Mutual distrust then becomes self-fulfilling: both sides prepare for violation because they expect violation, and those preparations make violation more likely.
This is a prisoner’s dilemma with artillery. Cooperation produces a better collective outcome, but each actor fears being the only one cooperating. International monitors or peacekeepers can sometimes reduce the problem by providing information and raising the cost of violations, but deploying those mechanisms requires political agreement that Sudan currently does not possess.
The fighting is also beginning to change the incentives of the organisations themselves. Before 15 April, Burhan and Hemedti were political rivals controlling armed institutions. After weeks of combat and casualties, they increasingly lead constituencies whose members have lost colleagues, relatives and resources to the other side. Every additional death creates people with stronger personal reasons to oppose compromise.
War therefore produces its own political interests.
Commanders gain battlefield authority. Armed groups recruit. Smuggling and wartime commerce become profitable. Political moderates lose influence because negotiation appears increasingly unrealistic. Civilians who previously identified mainly through political or regional interests can begin organising around whichever armed group appears capable of protecting them.
A conflict that began because institutions were fragmented can therefore create even more fragmentation as it continues.
This is one reason duration matters enormously. A war lasting two weeks and a war lasting two years may involve the same original leaders but become completely different political systems. Economic networks adapt to conflict, armed organisations decentralise, local commanders gain autonomy and foreign governments develop relationships with particular factions. Eventually the original disagreement becomes only one reason the war continues.
Sudan is not necessarily at that stage yet, and predicting an inevitable prolonged civil war would be premature. The danger is that the mechanisms producing one are already visible.
The humanitarian consequences themselves can accelerate institutional collapse. Hospitals in Khartoum have struggled to function amid fighting, electricity and water systems have been disrupted, banks and businesses have closed, and civilians have difficulty obtaining food and medicine. These are not merely side effects of military conflict. They reduce the state’s future ability to govern.
A hospital destroyed today does not automatically reopen when a ceasefire is signed tomorrow. Doctors may leave the country. Businesses may move capital abroad. Families with education and resources may emigrate. Tax revenue falls as formal economic activity contracts. Government salaries become harder to pay. Each loss weakens the civilian institutions that would eventually need to reconstruct Sudan after the fighting ends.
This creates an institutional multiplier on war damage. The immediate destruction of one asset reduces future productive capacity, which reduces the resources available for reconstruction, which slows recovery further. Economists usually discuss multiplier effects in relation to spending. Conflict can generate a multiplier in reverse.
Human capital may be particularly difficult to replace. Buildings can eventually be reconstructed with enough money. A surgeon who emigrates to another country may never return. A student whose university remains closed for several years loses education that cannot be recovered instantly. A generation growing up during conflict accumulates less schooling, poorer health and fewer economic opportunities.
This is why the true economic cost of civil war is usually much larger than the value of physically destroyed infrastructure.
Foreign evacuations this week illustrate another mechanism. Governments including the United States, Britain, France and several regional states have moved diplomats and citizens out of Sudan. From the perspective of those governments, evacuation is rational risk management. Collectively, however, every evacuation also reduces the number of international institutions physically capable of operating inside the country.
Diplomatic withdrawal can therefore become another signal of deteriorating expectations. Airlines stop flying because they expect danger. Businesses stop investing because they expect instability. Aid organisations reduce operations because staff cannot be protected. Those decisions worsen economic conditions, making the original expectation of instability more accurate.
Expectations become causal again.
The regional consequences could be equally serious. Sudan borders Egypt, Libya, Chad, the Central African Republic, South Sudan, Ethiopia and Eritrea, several of which already face their own political or security problems. Refugees, weapons and armed groups can cross borders much more easily than political instability can be contained within them. A prolonged Sudanese conflict could therefore create externalities across an already fragile region.
South Sudan is especially vulnerable because much of its oil is exported through pipelines and facilities running through Sudan. Conflict affecting that infrastructure could damage government revenue in a country whose economy depends heavily on petroleum. Sudan’s internal war can therefore produce economic consequences for a neighbouring state that has no control over Burhan or Hemedti’s decisions.
This is why civil wars are rarely purely domestic once they continue long enough. Refugee flows affect neighbours, armed groups seek foreign resources and outside governments begin supporting whichever faction best serves their interests. The conflict becomes embedded inside regional geopolitics.
External involvement can help produce peace, but it can also weaken incentives for compromise. If one faction expects weapons, money or diplomatic support from a foreign patron, the expected payoff from continuing to fight rises. A war that might otherwise end through exhaustion can continue because outside resources change the budget constraint.
The international community therefore faces its own collective-action problem. Egypt, Gulf states, the African Union, the United States and other actors broadly prefer a stable Sudan, but they do not necessarily prefer the same political settlement. Each can rationally support relationships that protect its own influence while collectively making a unified peace process harder.
Again, everyone can prefer stability while disagreeing over who should control the stable state.
The civilian movement that helped remove Omar al-Bashir in 2019 risks becoming the largest political loser. Sudan’s civilians repeatedly demonstrated that they could mobilise enormous public support, but they never acquired coercive power comparable with the military and RSF. As the conflict becomes militarised, political relevance increasingly shifts towards whoever controls weapons.
This produces a dangerous selection effect. War rewards organisations structured for war. Civil-society organisations, professional associations and democratic parties become relatively weaker precisely because their comparative advantage lies in peaceful politics. The longer violence continues, the more the political system selects for actors capable of violence.
That is why ending fighting is not enough to restore Sudan’s democratic transition. If Burhan defeats Hemedti completely, Sudan does not automatically become democratic. If Hemedti defeats Burhan, the same applies. A military monopoly can stop a civil war while creating another authoritarian regime.
Sudan therefore faces two separate institutional problems. It needs one coercive state rather than two competing armies, but that unified coercive state then needs to become subordinate to civilian political authority. Solving the first without the second could restore order without restoring democracy.
This distinction matters because stability is politically seductive after violence. Citizens exhausted by war may reasonably prioritise security over institutional reform. A victorious military leader can then present concentrated authority as the price of peace. Historically, civil conflict has repeatedly created exactly this trade-off.
The challenge is designing a settlement in which disarmament does not simply determine which general becomes dictator.
That will require civilians to regain a meaningful role in negotiations rather than allowing peace to become an agreement dividing power between the two armed factions. If the people who created the war become the only people empowered to design the post-war state, violence effectively becomes a qualification for political representation.
That creates terrible incentives for the future.
Sudan’s conflict therefore demonstrates why peace and democracy are related but not identical objectives. Peace asks whether organised violence stops. Democracy asks who governs once it stops and through which rules. A ceasefire between Burhan and Hemedti could achieve the first while leaving the second almost completely unresolved.
For now, the immediate priority has to be reducing civilian suffering and making ceasefires credible enough for humanitarian access. But any durable settlement will eventually need to address the structural problem that produced the fighting: independent armed organisations cannot remain permanent political actors with their own economic empires and command structures.
Two weeks ago, Sudan’s crisis demonstrated why a democratic transition cannot function while two armies retain independent political power. This week reveals the next problem: once those armies begin fighting, neither can safely surrender because the very weapons destroying the country are also what each believes protects it from the other.
That is why wars can continue even when almost everyone would be better off if they stopped. The relevant question is not whether peace is collectively preferable. It is whether institutions can make peace individually safer than continuing to fight.
Sudan does not have those institutions yet. Every additional week of war makes building them harder.
