Beyond Rwomushana’s arrest: Creeping militarization and the unmaking of the 1995 Constitution

The Constitution

The news arrived not through a formal legal dispatch or a parliamentary press release, but as a series of urgent, social-media-based whispers breaking through the celebratory rhythms of a traditional kwanjula in Jinja, which I was attending.

While families exchanged gifts and celebrated the happy moment of their children getting married, a chilling report rippled across reception tables: armed security personnel had entered the Kampala studios of Radio Simba and taken political commentator Charles Rwomushana away during a live broadcast of Gasimbagane ne Bannamawulire.

The joyful atmosphere gave way to a familiar, sobering refrain: “My friend, wetereze”—watch your words and your steps.

That quiet warning reveals something deeper than the fate of one commentator. It is the sound of self-censorship settling over a society. When citizens begin advising one another not merely to speak carefully, but to avoid speaking at all about certain people or subjects, the boundaries of constitutional freedom have already begun to shift.

The instinct becomes self-preservation: silence is golden because it guarantees your safety. As the Luganda adage puts it, asirise teyejjusa—the silent person regrets nothing.

But the constitutional alarm raised by Rwomushana’s seizure goes far beyond Rwomushana himself. The fundamental question is not whether his commentary was wise, reckless, provocative or polite. Nor is it whether he sometimes expressed political opinions with which one may disagree.

The constitutional question is more fundamental:

What happens to a republic when citizens can no longer safely evaluate, question or debate the conduct and political ambitions of those who command military power?

Beyond one arrest: the anatomy of militarised civilian governance

Reports now indicate that Rwomushana has been released after spending the night in detention. That development is welcome.

But his release does not retrospectively answer the constitutional questions raised by his seizure, the identity and authority of those who took him, the legal basis for his detention, or the message conveyed by armed intervention during a live political broadcast.

Latest reporting confirms that armed men removed him from Radio Simba while he was participating in the programme, although the circumstances and institutional responsibility for the operation have not been fully explained publicly.

The important point, therefore, is not that Rwomushana remained in detention indefinitely. It is that a political commentator was removed from a live broadcast by armed personnel and detained overnight in circumstances that initially lacked a clear public explanation.

That raises another unavoidable constitutional question: what happens when security power becomes the instrument through which politically sensitive public discourse is policed?

This requires us to look beyond the individual incident. Uganda’s contemporary trajectory is not a return to the classical military dictatorship of the 1970s. The courts have not been abolished. Parliament still sits. Elections are still held. The Constitution remains formally in force.

The danger is therefore subtler.

It is the possibility of militarised civilian governance: a system in which civilian institutions formally remain intact while military personnel, security institutions and coercive habits increasingly influence or determine the boundaries of permissible political behaviour.

There is an important distinction between three forms of government.

Classical military government occurs when the armed forces openly overthrow civilian authority, suspend constitutional government and assume direct political control.

Civilian constitutional government exists when the armed forces are subordinated to elected civilian authority, remain politically neutral and exercise coercive power within transparent constitutional and statutory limits.

Militarised civilian governance occupies the dangerous space between the two. Civilian institutions remain formally operative, but military personnel, security agencies and coercive logic progressively shape political life and determine what citizens may safely say or do.

Uganda increasingly exhibits characteristics of this third model.

The process operates through at least four dimensions.

First, personnel. Serving military officers increasingly become visible participants in politically consequential activity, thereby blurring the traditional boundary between military command and partisan politics.

Second, institutions. Security agencies increasingly become the first resort in dealing with political and civic disputes that ought ordinarily to be managed through civilian institutions, ordinary law and political debate.

Third, discourse. Political disagreement is increasingly capable of being reframed as a question of security, stability or national interest. Once that happens, criticism of political actors can be treated as something more sinister than criticism.

Fourth, civic psychology. Citizens begin internalising the consequences of coercive power. They censor themselves before anyone needs to censor them.

This fourth dimension may ultimately be the most consequential. A state does not need to arrest everyone in order to silence society. It only needs to scare or convince enough people that speaking carries an unacceptable risk.

The constitutional lesson of Uganda’s history

Uganda’s history demonstrates the danger of allowing coercive power to become a substitute for constitutional politics.

The 1966 constitutional crisis showed that constitutional disputes could be settled by force rather than constitutional procedure.

The 1971 coup demonstrated the consequences of allowing military power to displace civilian constitutional authority altogether.

The turmoil of 1980–1985 showed how electoral legitimacy could coexist with devastating armed conflict when political authority could not effectively control competing instruments of force.

The 1985 military intervention further entrenched the idea that military intervention could become an accepted mechanism for political transition.

These were not merely different episodes in Uganda’s turbulent history. Together they revealed a recurring constitutional trajectory: the inability of political institutions to establish an unquestioned principle that the gun must remain subordinate to law and civilian political authority.

The 1995 Constitution represented a historic attempt to break that cycle. This is particularly significant because it was enacted under an NRM government that had itself achieved power through armed struggle.

The 1995 constitutional settlement was therefore, in an important sense, an attempt to transform the source of political legitimacy.

The legitimacy of the gun was to give way to the legitimacy of the Constitution.

The armed forces were no longer to be the arbiters of political succession. Under Article 208, the Uganda People’s Defence Forces were constitutionally conceived as a national, non-partisan institution, subordinate to civilian authority and subject to constitutional discipline.

The deeper promise of 1995 was therefore not simply that Uganda would have another written Constitution.

It was that politics would command the gun, rather than the gun command politics.

That is the promise whose erosion is becoming more apparent each passing day.

The unmaking from within

As I have noted before, constitutional unmaking does not necessarily begin with the suspension of a Constitution.

Sometimes it begins with the gradual removal of the conditions that make constitutional government meaningful.

Article 105(2), which imposed presidential term limits, was removed in 2005.

Article 102(b), which imposed an upper age limit for presidential candidates, was removed in 2017.

Those were formal constitutional amendments. But constitutional erosion can also occur without changing a single word of the constitutional text.

It can occur through practice. It occurs when constitutional provisions and principles are repeatedly violated, ignored, stretched or rendered politically inapplicable until conduct once regarded as constitutionally impermissible becomes normal.

That is why Article 208 deserves particular attention.

The Constitution does not merely create an army. It establishes the constitutional character of the armed forces. Their national and non-partisan character is not ornamental. It is designed to prevent the coercive instruments of the State from becoming instruments of partisan political competition.

This brings us to the extraordinary constitutional stress test presented by the political role of the serving Chief of Defence Forces, Gen. Muhoozi Kainerugaba.

That constitutional tension has acquired a new and unusually explicit dimension. In an X post of 6 September 2026, Andrew Mwenda publicly stated that he was instructed by Gen. Muhoozi Kainerugaba, the chairman of PLU, to inform the public that PLU does not intend to become a political party but will remain “a civic organization, pressure group and social movement within the NRM” and pursue all its goals within the ruling party.

The significance of that declaration is not that PLU has chosen to remain outside the formal definition of a political party. It is that the serving head of Uganda’s armed forces is publicly identified as the chairman of an organisation whose stated purpose includes pursuing political goals within the governing political movement.

This makes the constitutional question harder to avoid. The issue is no longer simply whether a serving military commander has political aspirations. It is whether the constitutional requirement that the armed forces remain national and non-partisan is compatible with their serving commander simultaneously occupying a leadership position in an explicitly political mobilisation organisation operating within the ruling party.

Even if one attached legal significance to the distinction between a political party and a political movement, that does not by itself resolve the constitutional question of partisan political activity by the serving head of the armed forces.

The question, therefore, is not whether Gen. Muhoozi Kainerugaba is personally competent, popular or politically ambitious. Nor is it whether he has the right, as a citizen, to harbour political aspirations.

The constitutional question is this:

What constitutional provision or principle permits the serving head of a constitutionally mandated non-partisan armed force simultaneously to operate as an active political actor and prospective presidential candidate?

That question deserves a constitutional answer, not merely a political one.

The burden should not rest upon ordinary citizens to explain why the combination of active military command and partisan political ambition is compatible with Article 208.

The burden rests upon those responsible for the arrangement to explain how it can be reconciled with the constitutional requirement of a non-partisan armed force.

The stakes extend beyond the question of who may eventually occupy State House. The deeper constitutional danger arises when the institutions capable of using force become entangled in determining who should occupy it.

Once succession becomes a contest in which military loyalties, rather than constitutional procedures and electoral choice, can influence the outcome, the distinction between political authority and coercive authority begins to collapse.

This concern is no longer confined to Uganda’s domestic political debate. In its 3 September 2026 assessment of Uganda’s trajectory, The Economist warned that President Museveni’s declining grip on power could create either an increasingly authoritarian succession or a struggle among those surrounding him and within the security establishment.

Whether or not one accepts every aspect of that assessment, the warning illustrates the constitutional stakes of the present moment: what happens when presidential succession becomes inseparable from the loyalties and ambitions of those who command the coercive instruments of the State?

This matters because once military command and partisan political ambition become intertwined, political criticism itself becomes vulnerable to reclassification.

A citizen who questions a political ambition may suddenly appear to be questioning military authority.

A commentator who analyses succession may be portrayed as threatening stability.

A political dispute may acquire a security dimension. And once political questions are transformed into security questions, coercive institutions acquire a justification for entering spaces that properly belong to political debate.

That is how militarised civilian governance advances without the declaration of martial law.

The politics of “wetereze”

This brings us back to the most revealing word in this story:

Wetereze.

Align yourself to the new reality.

Watch your words.

Watch your steps.

Be careful what you say.

Be careful about whom you discuss.

Be careful about whom you criticise.

I have personally heard more than enough of this advice from family members, relatives, friends and even foes.

There is an important constitutional phenomenon hidden inside this apparently ordinary advice.

The Constitution does not become meaningless only when the State openly prohibits speech. It is also weakened when citizens cease exercising their constitutional freedoms because they fear the consequences of doing so.

That is the politics of wetereze.

Its power lies precisely in its informality.

No decree needs to be issued.

No newspaper needs to be formally banned.

No constitutional provision needs to be repealed.

One commentator is taken away.

Another is warned.

A journalist becomes cautious.

An editor softens a headline.

A politician chooses not to raise a difficult question.

A citizen deletes a social-media post.

Eventually, people begin censoring themselves.

The cumulative effect is profound.

One arrest does not unmake a Constitution. But one arrest can contribute to a political culture in which everyone learns the lesson that speaking carries high risks.

That is how constitutional unmaking occurs.

Not always dramatically.

Not necessarily through tanks in the streets.

Sometimes through a gradual alteration of the boundaries of permissible participation until citizens voluntarily retreat from the constitutional space that belongs to them.

The danger is not necessarily a coup

This is why it would be a mistake to understand creeping militarisation only through the vocabulary of coups and military dictatorships.

The greater danger may be precisely that there is no coup.

There may be no suspension of Parliament.

No abolition of the courts.

No dissolution of the Constitution.

No declaration that the military has assumed power.

Instead, the institutions remain.

But their practical relationship to coercive authority changes.

Civilian regulators defer to security agencies.

Political disagreements acquire security implications.

Serving military commanders become increasingly prominent in political affairs.

Citizens learn that some subjects are safer avoided.

The formal constitutional architecture survives while its substantive operating culture changes.

This is militarised civilian governance.

And it is potentially more difficult to recognise because every individual step can be defended as exceptional, necessary or unrelated to the larger transformation.

But constitutionalism is not protected merely by the survival of constitutional institutions on paper.

It requires a political culture in which those institutions actually constrain power.

The rule of law and the Rwomushana question

There is, of course, a legitimate role for the State in investigating unlawful conduct.

Freedom of expression is not a licence to commit crimes. If a citizen crosses a statutory boundary, the State is entitled to investigate and, where the law permits, arrest and prosecute.

But that power must itself be exercised according to law.

Any deprivation of liberty must have a lawful basis. The person detained must be informed of the reasons for the deprivation of liberty, and constitutional safeguards—including access to due process and judicial protection—must be respected.

The distinction is fundamental.

A constitutional State does not become weaker because it uses law against unlawful conduct. It becomes weaker when coercion is used in place of law.

That is why the circumstances surrounding Rwomushana’s detention matter even if one disagrees completely with anything he may have said.

The proper constitutional response to controversial political commentary is normally argument, evidence, rebuttal and, where necessary, lawful process.

The coercive response is different.

When armed personnel enter a media studio and remove a commentator from a live broadcast, the visual message extends far beyond the individual concerned.

It tells everyone watching:

There are subjects about which you should think twice before speaking.

That is the constitutional significance of the wetereze phenomenon.

The subsequent release of Rwomushana does not answer that question. It merely changes the question from where he was being held to why armed personnel were required to remove him from a live political broadcast in the first place.

The Constitution’s real test

The ultimate test of the 1995 Constitution is therefore not whether its text remains printed in the statute books.

The real test is whether its principles continue to regulate the exercise of power.

Can citizens question those who exercise political power?

Can they debate presidential succession?

Can they scrutinise the political conduct of serving military commanders?

Can journalists and commentators discuss such matters without fearing that political disagreement will be transformed into a security offence?

Can civilian institutions remain genuinely superior to coercive institutions?

And, above all, can the armed forces remain constitutionally subordinate to civilian political authority and genuinely non-partisan?

These are not abstract academic questions.

They go to the heart of the constitutional settlement Uganda made in 1995.

Uganda does not need another military dictatorship to lose the promise of that Constitution.

It only needs to become accustomed to the idea that military and security power may enter ordinary political life whenever political controversy becomes sufficiently inconvenient.

That is why the seizure of Charles Rwomushana should concern even those who dislike his politics or disagree with his commentary.

The issue is bigger than Rwomushana.

It is bigger than Muhoozi.

It is bigger than the current succession debate.

It is about whether Uganda’s constitutional order can maintain the boundary between political authority and coercive authority.

The danger facing Uganda may therefore not be an imminent military coup.

It may be something more gradual and, for that reason, more difficult to resist: the normalisation of military authority within civilian political life until constitutional government remains formally intact but becomes substantively subordinate to security power.

The question, therefore, is not simply why Charles Rwomushana was taken away.

It is what, precisely, the authorities considered so dangerous about what he was saying that a live radio programme required the intervention of armed men.

In a constitutional republic, ideas are answered with better ideas, allegations with evidence, and unlawful conduct with lawful process.

When open political debate is answered instead by the presence of armed power, the Constitution begins to lose something more important than its words.

It begins to lose its authority.

And when citizens respond by whispering wetereze, the unmaking of constitutional government may already have begun.

The writer is a senior advocate of the High Court of Uganda and a former Minister.

The post Beyond Rwomushana’s arrest: Creeping militarization and the unmaking of the 1995 Constitution appeared first on The Observer Media Ltd.