Heavy fighting erupted across Sudan yesterday between the country’s regular armed forces and the powerful paramilitary Rapid Support Forces. Gunfire and explosions have been reported across Khartoum, while fighting has occurred around airports, military installations and other strategic locations. Both sides have issued competing claims about control of important sites, making the precise military situation difficult to establish while combat continues. What is already clear is that a political struggle between Sudan’s two most powerful military leaders has become open armed conflict inside the capital.
The Sudanese Armed Forces are commanded by General Abdel Fattah al-Burhan, who has effectively led Sudan since the military takeover of October 2021. The RSF is commanded by Mohamed Hamdan Dagalo, usually known as Hemedti, who served as Burhan’s deputy within the military-led political structure. Until very recently, these men were not leaders of opposing rebel and government forces. They were partners at the centre of the same state, having cooperated in the coup that removed Sudan’s civilian-led transitional government. Their transformation from political partners into battlefield opponents therefore exposes something fundamental about how Sudan’s post-Bashir political order was constructed.
The immediate disagreement concerns the proposed integration of the RSF into Sudan’s regular armed forces. Negotiations over a new transition towards civilian government required the country eventually to possess a unified military command, but Burhan and Hemedti disagreed over how rapidly integration should occur and who would ultimately command the resulting force. Those details can sound technical until the underlying question is stated more plainly: which man would control the organised violence of the Sudanese state?
That question is foundational because modern states are supposed to possess what Max Weber famously described as a monopoly over the legitimate use of physical force within a territory. The formulation does not mean governments are the only actors capable of violence. Criminal organisations, insurgents and individuals obviously exist. It means legitimate organised coercion is ultimately supposed to operate through one sovereign institutional structure rather than several armed organisations negotiating with one another as political equals.
Sudan never fully achieved that consolidation. The regular military remained enormously powerful, but the RSF developed into a parallel armed organisation with tens of thousands of fighters, its own leadership, sources of revenue and political relationships. The state therefore contained two coercive centres whose interests overlapped without becoming identical. As long as Burhan and Hemedti cooperated, the arrangement could appear stable. Once their interests diverged, the institutional weakness became military reality.
This is a useful reminder that coalitions between armed actors are not equivalent to unified state institutions. Two generals can sign an agreement. Their organisations remain capable of coercing one another. Political cooperation therefore rests partly on the expectation that neither side will find violence more advantageous than bargaining. Once that expectation changes, there may be no higher institution capable of enforcing the agreement.
Civilian politics becomes extraordinarily vulnerable under those conditions. A civilian minister can negotiate with the army. A political party can negotiate with the RSF. Protest movements can demand democratic government. Yet if the civilians possess no independent coercive institution capable of enforcing constitutional rules, their political authority ultimately depends on armed organisations agreeing to remain constrained.
That is not civilian control.
It is civilian government by military permission.
Sudan’s revolution in 2019 demonstrated both the possibility and limitation of popular mobilisation under such conditions. Months of protests helped bring down Omar al-Bashir after nearly thirty years in power. The demonstrations were remarkable in scale and persistence, and civilian organisations constructed a powerful political movement demanding democratic government. Yet Bashir’s removal was ultimately carried out by the military, which meant the institution responsible for enforcing the dictatorship remained one of the institutions determining what came next.
This created a classic revolutionary dilemma. Removing the dictator does not necessarily remove the structure that made dictatorship possible. Bashir disappeared from the presidency, but Sudan’s military and security institutions retained enormous economic and coercive power. Civilians therefore had enough political legitimacy to demand participation in government but insufficient institutional power to exclude the military from the transition.
The resulting compromise created a transitional arrangement in which civilian and military leaders shared authority. That was understandable because the alternative might have been immediate confrontation. Compromise bought time and created a pathway towards elections. It also left unresolved the question of who possessed final authority if civilians and soldiers later disagreed.
In October 2021, the answer became clear. Burhan, supported by Hemedti and other military actors, dissolved the civilian-led government and detained Prime Minister Abdalla Hamdok. The democratic transition was interrupted because the armed institutions retained enough independent power to override the political institutions supposed eventually to control them.
That coup should have changed how Sudan’s transition was understood. The central obstacle was not simply designing the correct electoral timetable or negotiating another constitutional document. It was constructing a political order in which armed organisations could no longer veto civilian government.
This is the credible commitment problem at the centre of many democratic transitions. Military leaders are asked to surrender political power today in exchange for assurances that a future civilian government will respect their institutional interests, pensions, economic assets or personal security. Yet once civilians control the state, what prevents them from breaking those promises?
The generals therefore hesitate to surrender power completely.
Civilians face the opposite problem. They are asked to trust that military leaders will gradually withdraw from politics after transitional arrangements are completed. Yet as long as the military retains independent coercive power, what prevents the generals from cancelling the transition when civilian decisions threaten their interests?
Each side doubts the other’s future restraint.
That makes partial transitions unstable.
Democratisation therefore requires institutions capable of making commitments credible enough that neither side believes pre-emptive action is necessary. Constitutions, courts, international guarantees and negotiated immunity arrangements can sometimes help, although each carries serious moral and political costs. Ultimately, however, civilian democracy requires the military to accept subordination to institutions it cannot simply remove when they become inconvenient.
Sudan’s problem became even harder because it did not have only one military institution requiring subordination.
It had two.
The origins of the RSF help explain why. The organisation developed from the Janjaweed militias associated with the conflict in Darfur, where armed groups backed by Bashir’s government were accused of extensive atrocities against civilian populations. The RSF was formally established later and incorporated into Sudan’s security architecture, with Hemedti becoming one of the country’s most powerful figures.
From Bashir’s perspective, creating or empowering parallel armed forces had political logic. Authoritarian leaders frequently fear their own militaries because armies capable of protecting regimes are also capable of overthrowing them. Building alternative security organisations creates coup-proofing: no single armed institution possesses complete coercive dominance, and rival forces can monitor or balance one another.
The strategy can protect the ruler.
It weakens the state.
A dictator distributes coercive power because fragmentation reduces the probability that one general can seize everything. Once the dictator disappears, the country inherits several armed organisations that have learned to operate autonomously and possess little reason to surrender their power voluntarily.
Sudan may now be experiencing the long-term cost of that strategy.
This demonstrates how institutions designed for regime survival can create enormous externalities for the society that inherits them. Bashir had incentives to maximise his own political security. Sudan needed a unified professional security system. Those objectives were not identical.
Authoritarian rulers often build states around precisely this distinction.
Intelligence agencies overlap.
Military units compete.
Presidential guards receive better equipment.
Command structures become deliberately fragmented.
The ruler becomes indispensable because the ruler is the person balancing the competing institutions.
When the ruler disappears, the balance disappears too.
Personal political stability therefore produces institutional fragility.
This is another reason long authoritarian rule can be misleading. Bashir governed for nearly three decades, which superficially suggests political stability. The institutions underneath him were organised partly to prevent any alternative centre of power from becoming strong enough to remove him.
That architecture survived his removal.
Sudan’s current conflict therefore cannot be understood only through the personalities of Burhan and Hemedti. Their ambitions clearly matter. Both possess enormous incentives to avoid becoming subordinate to the other. But if the institutional system allowed two commanders to accumulate enough independent military and economic power to fight for control of the state, replacing either individual would not necessarily solve the structural problem.
This is the same distinction between personnel failure and system failure that appeared after the Greek railway disaster, although here the stakes are much larger. Political commentary naturally personalises conflict because individuals make decisions. Institutional analysis asks why those individuals possessed the capacity to make their rivalry nationally catastrophic.
The RSF’s economic power makes this particularly important. Hemedti’s network has benefited from involvement in Sudan’s gold economy and other commercial activities, while the regular military itself controls extensive businesses across multiple sectors. Sudan’s armed organisations therefore do not simply receive budgets from the state. They possess economic interests that can reduce their dependence on civilian government.
This weakens civilian control because fiscal dependence is a mechanism of political control. In an ordinary democratic system, parliament appropriates money for the military. Civilian representatives can therefore influence military priorities through budgets even when operational command remains professional and separate.
An armed organisation financing itself through independent commercial networks is harder to constrain.
Economic autonomy can become political autonomy.
This is why security-sector reform cannot simply mean changing uniforms or command titles. A genuinely unified military requires control over financing, procurement, appointments and chains of command. Parallel economic empires need to become subject to civilian fiscal institutions.
Those reforms threaten powerful interests.
That is exactly why they are difficult.
The proposed integration of the RSF therefore concerned much more than military efficiency. If Hemedti accepted rapid integration under Burhan’s command, he risked losing the independent institutional base that made him one of Sudan’s most powerful political actors. If Burhan accepted a prolonged period in which the RSF remained autonomous, he risked preserving a rival military organisation capable of challenging the army indefinitely.
Both men therefore had reasons to view the other’s preferred integration timetable as a threat to political survival.
This is a bargaining problem under conditions where failure can be settled through force.
That last feature changes everything.
Politicians in a parliament can negotiate aggressively because losing a vote does not normally threaten their physical survival. They can try again later. Armed leaders cannot necessarily make the same assumption. Losing control of the security apparatus can lead to imprisonment, exile or death.
The stakes make compromise harder.
This is why successful democratic transitions often require making political defeat survivable not only for voters and parties but for former authoritarian elites.
That creates morally uncomfortable trade-offs. Leaders implicated in repression may demand immunity or economic protections before surrendering power. Granting those concessions can undermine accountability for past abuses. Refusing them can give armed elites stronger incentives to sabotage the transition.
Transitional justice therefore operates inside a strategic constraint.
Justice asks what powerful actors deserve.
Democratisation asks what concessions may be necessary to persuade them to stop possessing the power to decide for themselves.
There is no clean solution.
Sudan’s civilians have repeatedly borne the cost of that dilemma. Protesters demanded accountability for violence committed during the transition, including the June 2019 massacre in Khartoum, when security forces attacked a protest encampment and killed large numbers of demonstrators. The RSF was widely accused of major involvement.
For civilian movements, asking the same armed leaders to oversee a democratic transition therefore required extraordinary distrust to be overcome.
That distrust was justified by history.
The difficulty is that transitions cannot operate through trust alone.
They need institutions capable of functioning even when nobody trusts anybody.
This is where international mediation can help but also where its limitations become clear. The United Nations, African Union, neighbouring states, Gulf countries, the United States and European governments all possess interests in Sudan’s stability. External actors can provide incentives, guarantees and diplomatic pressure.
They cannot manufacture a monopoly on violence from outside.
If Sudanese armed organisations believe control of Khartoum is more valuable than the diplomatic costs of fighting, statements from foreign capitals may have limited effect.
Again, sanctions and diplomatic pressure work through preferences. They do not mechanically produce compliance.
Regional powers can even complicate bargaining because Sudan’s factions possess relationships with different external actors. Egypt has historically maintained close ties with the Sudanese military. Gulf states have developed relationships with Sudanese political and military actors. Hemedti’s forces participated in the Saudi-led war in Yemen, generating international connections and resources.
Foreign support can therefore increase the ability of domestic actors to resist compromise.
This is the internationalisation problem common to civil wars.
Once local factions obtain external patrons, the resources available for continuing conflict increase and the bargaining environment becomes more complicated.
Yet describing Sudan simply as a proxy conflict would be equally misleading. Burhan and Hemedti possess their own objectives. External powers can influence those objectives without creating them.
Domestic agency remains central.
The immediate humanitarian danger is enormous because much of the fighting is occurring in Khartoum, a metropolitan area containing millions of civilians. Modern capitals are not designed as battlefields. Hospitals, airports, water systems and electricity networks can become military objectives or collateral damage. Civilians cannot simply evacuate easily when armed forces contest roads and transport infrastructure.
Urban warfare therefore converts state infrastructure into vulnerability.
The same airport that connects a capital to the world becomes strategically valuable to armed factions.
The same bridge used by commuters becomes a military chokepoint.
The same telecommunications system civilians need for information becomes part of command and control.
War changes the function of infrastructure.
This is why the destruction produced by civil conflict extends far beyond direct casualties. Hospitals stop operating because staff cannot travel. Supply chains fail. Banks close. Food prices rise. Businesses stop paying workers. Schools close. Foreign investment disappears.
A battle lasting days can create institutional damage lasting years.
Sudan can particularly ill afford that. Its economy was already under severe pressure after years of political instability, inflation and the loss of international support following the 2021 coup. A renewed transition towards civilian rule could have reopened access to financing and debt relief.
Conflict moves the country in the opposite direction.
This creates another vicious cycle. Economic deterioration reduces the resources available to civilian institutions. Armed organisations possessing independent financing become relatively stronger. Political authority becomes increasingly militarised. That militarisation further damages the economy.
Violence therefore redistributes power towards the actors best equipped for violence.
This is one reason civil wars become self-reinforcing.
The longer conflict continues, the weaker the institutions capable of ending it peacefully become.
As of tonight, it is far too early to know whether Sudan is entering a prolonged civil war. Both sides may still calculate that negotiation is preferable once the costs of fighting become clearer. Claims about territorial control remain contested, and the military situation is changing rapidly.
The correct analysis therefore should not predict the outcome.
It should identify what has already failed.
Sudan attempted to build a civilian political transition while leaving unresolved the question of who ultimately controlled organised violence.
That contradiction was manageable while the armed leaders cooperated.
It became catastrophic when they stopped.
This suggests that democracy requires something even more basic than elections, constitutions or free expression. Those institutions matter enormously, but they operate downstream from the state’s ability to enforce one legal order.
A parliament can pass legislation only if armed organisations cannot reject the legislation through force.
A court can issue judgements only if someone enforces them.
An election can remove a government only if the defeated side cannot retain office through its own military organisation.
The monopoly on legitimate violence is therefore not the opposite of democracy.
Properly constrained, it is one of democracy’s preconditions.
The crucial words are properly constrained. Concentrating all coercive power inside one state can produce dictatorship if civilians cannot control that state. Fragmenting coercive power among rival organisations can produce civil war.
The institutional objective is difficult: one security apparatus strong enough to prevent private armies, but constrained enough that it cannot become the government itself.
Sudan has struggled with both sides of that equation.
Under Bashir, security institutions helped sustain authoritarian rule.
After Bashir, fragmented security institutions repeatedly obstructed civilian rule.
Now those institutions are fighting one another.
That sequence demonstrates why security-sector reform is not a technical issue to be postponed until after democratisation.
It is part of democratisation itself.
Civilian government does not exist simply because civilians occupy ministries.
It exists when armed institutions accept that civilians possess the authority to decide who commands them, how they are financed and what political rules they must obey.
Sudan never completed that transfer.
The transition therefore rested on an assumption that Burhan and Hemedti would voluntarily surrender forms of power they had spent years accumulating.
Yesterday, that assumption collapsed.
The immediate priority is stopping the fighting and protecting civilians. But if Sudan eventually returns to political negotiations, another agreement dividing offices between armed leaders and civilians will not be enough. The institutional question avoided since 2019 will still be waiting underneath it: who possesses the guns, who controls the people possessing them, and what prevents those people from becoming politicians whenever they dislike what the politicians decide?
Sudan’s democratic transition was supposed to answer who should govern after Bashir.
The fighting in Khartoum has exposed the question it never resolved first: who has the power to make that answer stick?
