|
Getting your Trinity Audio player ready...
|
Ol Kalou, Nyandarua County. Thursday, July 16, 2026. 1:58 in the afternoon.
A police vehicle sits motionless outside a polling station, surrounded not by officers but by residents. Uniformed police unzip bags, turn out pockets, lift shirts and folded clothes into the afternoon light. A woman’s voice rises above the murmuring crowd, equal parts relief and accusation: “Hakuna kitu kama karatasi. Zote ni nguo.” There is nothing like paper here. Just clothes.
The crowd is not rioting. They are not armed. They have no statutory mandate, no reflective jackets, no authority except the one they have conferred upon themselves. They call themselves vote guards, and for the better part of an hour they have been doing what they no longer trust the state to do: making sure nothing enters the polling station that should not.
A few kilometres away, outside Ol Kalou Comprehensive School, another crowd has arrived at a similar conclusion by different means. Stones have been dragged across the road; two police vehicles idle behind the makeshift barricade while residents insist that anyone wishing to reach the polling station will proceed on foot. Sammy Ngotho, the Democratic Change Party candidate, steps between the two sides and listens before speaking. The crowd answers with remarkable clarity: We are not stopping people. We are stopping vehicles. The barricades come down. Voting resumes.
Then, at 2:45 p.m., it doesn’t.
Several Toyota Prados turn into the town centre. The men who emerge wear balaclavas. Witnesses will later say they carried tear-gas launchers, and within moments canisters begin arcing over the streets. The same residents who had spent the morning policing the integrity of a public election scatter through clouds of gas. A photojournalist, Enos Teche, is assaulted and stripped of his camera. Gunshots are reported, or believed to have been fired – a distinction that matters in court, but scarcely matters to people already running.
At that moment, no one knows who the masked men are. It will take days before investigators begin assembling the convoy’s anatomy – a DCI probe, a human rights count of unmarked vehicles, journalists comparing footage and tracing registrations toward names. Slowly, a picture emerges of coordination far more complex than the panic witnessed that afternoon.
That is not where this story begins, nor where it ends. It begins with two crowds inhabiting the same town less than an hour apart – one that dispersed because someone appealed to reason, the other that dispersed only when its purpose had been fulfilled. Both entered the public record. Both became part of Kenya’s democratic memory. Yet journalism and official statement would gradually compress them into a single, deceptively ordinary noun.
This essay begins with that compression – because every political order inherits two things from the one before it: its institutions and its vocabulary. The first governs bodies. The second governs what those bodies are permitted to mean.
The name that wouldn’t change
Names are rarely innocent. They are among the first political settlements a society reaches after an event has occurred – before courts determine liability, before commissions assemble timelines, before historians reconstruct causes, a public vocabulary has already begun assigning meaning. The first word attached to an event often survives long after the evidence has outgrown it.
Ol Kalou illustrates this unsettling persistence. Within forty-eight hours, the Directorate of Criminal Investigations opened a formal probe, promising to pursue every lead through forensic analysis, digital evidence, and witness testimony. The Kenya Human Rights Commission documented at least ten unmarked vehicles moving through the constituency, carrying armed men in civilian dress, many with their faces concealed. Journalists went further still. The Nation traced a vehicle registration in the convoy back to a sitting legislator, Kiambaa MP John Njuguna Wanjiku, while a police source described senior officers coming under pressure from politically connected figures – one of whom allegedly invoked the president’s name – to act outside their mandate. NTV built an entire investigation, using footage, open-source intelligence, and digital forensics, around unmasking the men themselves.
This was not a failure of reporting, nor a shortage of evidence. Kenya’s journalists did what democratic journalism is supposed to do: they followed number plates instead of rumours, and they asked who financed the convoy, who coordinated its movement, and who stood to benefit from its arrival.
The facts became steadily more specific. The language did not. Even after the Nation named a sitting member of parliament, Tuko’s own headline on the story still read: “UDA MP’s Vehicle Linked to Goons Who Terrorized Ol Kalou.” Interior Cabinet Secretary Kipchumba Murkomen, addressing the incident publicly, vowed to punish those responsible for “goonism”, while denying that any parallel command structure existed alongside the police. Across headlines, statements, and commentary, one word continued to do all the work.
Goons. It appeared so often that it ceased to function as description and became assumption. Yet the word explains remarkably little – nothing about organization, financing, command, logistics, or political purpose. It describes conduct while concealing structure.
The distinction matters. A man who assaults a commuter during a robbery may be called a goon. A hired gang disrupting a neighbourhood dispute may attract the same label. But a coordinated convoy arriving in multiple vehicles, equipped with protective clothing and crowd-control weapons, operating according to an apparent plan and traceable to a sitting legislator’s own vehicle, belongs to a different analytical category altogether. Whether that category is ultimately established as a politically sponsored militia, an organized criminal group, or another legally recognized form of coordinated violence depends on the evidence and the findings of investigators. The point is not to substitute one slogan for another. It is to let our vocabulary evolve as the facts do.
In Ol Kalou, the evidence moved. The word remained still.
This is not merely a linguistic curiosity. Institutions often inherit vocabularies long after they have forgotten the circumstances that produced them. Certain words become habits of perception, reached for not because they are the most accurate but because they make disturbing realities appear familiar. To call every organized act of electoral violence “goonism” is to flatten the distance between opportunistic criminality and coordinated political coercion – to see a recurring pattern as a series of disconnected incidents rather than ask whether the incidents form part of a larger architecture.
A headline announcing that “goons disrupted voting” quietly concludes the inquiry before it has begun. It directs attention toward the men throwing stones or firing canisters while leaving unasked the questions democratic societies most urgently require: who assembled them, who transported them, who financed them, who expected to benefit. Those questions point beyond the individual, toward systems – which is why precision is not an academic luxury. It is a democratic obligation.
The question, then, is not whether the men who appeared in Ol Kalou behaved violently. That much is visible. The question is whether our public vocabulary has the courage to follow the evidence wherever it leads – even to a sitting legislator’s vehicle – or whether it retreats into familiar words at precisely the moment unfamiliar truths begin to emerge. We cannot investigate structures we refuse to name.
A genealogy of disposable force
The convoy that entered Ol Kalou did not invent itself. It belongs to a much older lineage – not of identical organizations, but of recurring political solutions. Across Kenya’s modern history, governments, parties and political entrepreneurs have repeatedly confronted the same problem: how to exercise coercion while avoiding responsibility for it. The answer has changed its uniform many times. The principle has not.
Under colonial rule, coercion rarely appeared as naked force alone. It was administered through emergency regulations, loyalist auxiliaries, informant networks and legally sanctioned exceptions. The State of Emergency declared in October 1952 did more than criminalize the Mau Mau uprising; it transformed ordinary social existence into potential evidence of insurgency, dissolving the distinction between civilian and combatant under the authority of emergency law. Violence was never merely physical. It was administrative. Classification became a weapon.
Independence dismantled colonial rule. It did not dismantle its repertoire. The one-party decades inherited a subtler lesson from empire: power is often exercised most effectively through actors who stand just beyond the formal boundary of the state. KANU’s Youth Wing occupied precisely that ambiguous terrain – mobilizing rallies, disrupting opponents, intimidating critics, while remaining neither ordinary civilians nor conventional security officers. Their ambiguity was not an institutional defect. It was their political value. A police officer carries a badge. A proxy carries plausible deniability.
Multiparty politics decentralized this architecture rather than eliminating it. By the 1990s, political competition increasingly relied on networks that could be assembled, financed and disbanded according to electoral need. Groups such as Jeshi la Mzee, the Taliban and Mungiki emerged within distinct contexts, but together revealed a transformation: violence had begun entering the marketplace, patronage replacing permanent loyalty, coercion becoming something contracted rather than continuously maintained.
By the twenty-first century, ideology had receded and logistics had taken its place. Today’s political contractor needs neither a revolutionary manifesto nor lifelong party allegiance – only availability: young men excluded from stable employment, vehicles that move quickly, money that leaves few fingerprints. The infrastructure has become lighter. Its political function has remained constant.
This is what makes contemporary electoral violence deceptively difficult to recognize. Kenya’s newer formations resemble temporary supply chains more than standing armies – they assemble rapidly, perform a task, disperse before institutional accountability can catch up, and leave behind little more than scattered footage and competing narratives. The same economic conditions that produce precarious work also produce precarious violence; young people abandoned by one political economy become available to another.
The essential question is no longer “Who are these men?” It is: who assembled this network, who secured the vehicles, who financed the operation, who expected a political return? The men who emerged from the Prados in Ol Kalou may or may not ultimately satisfy the legal definition of a militia – that determination belongs to investigators and, where appropriate, the courts. But the available reporting already suggests organized coordination, logistical planning, and political purpose, precisely the features that demand a vocabulary capable of distinguishing structured coercion from ordinary violence.
History, in this sense, is not repeating itself. It is refining itself. The auxiliaries of empire became party youth wings; party youth wings gave way to ethnically anchored militias; militias increasingly yielded to contracted political networks. The costume changed. The architecture endured. Power prefers instruments that can disappear before responsibility arrives – perhaps the defining political innovation of our time is not that violence has become more visible, but that it has become more disposable.
The politics of naming
Political violence does not begin with weapons. It begins with categories. Long before a society decides how to respond to violence, it has already decided what kind of violence it believes it is witnessing – and those decisions are rarely made in courtrooms. More often they are made in headlines, press briefings, and the quiet habits of everyday speech.
Classification is never merely descriptive. It is political. To name an act is to place it inside a moral and institutional framework; every word carries a theory of causation, a theory of responsibility, and a theory of what questions no longer need to be asked. This is why vocabulary matters – not because words are more powerful than institutions, but because institutions themselves often operate through language.
Empires understood this long before republics did. Colonial administrations did not simply wage wars against insurgencies; they renamed them. Anti-colonial resistance became “banditry”. Political prisoners became “detainees”. States of exception became “emergencies”. Independent states inherited more than roads, ministries and legal codes. They inherited administrative vocabularies – sometimes without noticing.
That inheritance helps explain why certain words remain resilient even after the facts around an event have changed. Once a category becomes familiar, it acquires a gravitational pull; new evidence bends toward old terminology rather than the terminology being revised to fit new evidence. Ol Kalou reveals precisely this phenomenon. Investigations grew increasingly sophisticated – vehicle registrations traced, timelines reconstructed, a legislator named – yet the dominant vocabulary remained almost entirely unchanged, still foregrounding the visible perpetrators while backgrounding the organizational structures that assembled them.
Editors must publish quickly; officials often exercise legal caution; journalists work within conventions designed to minimize error and defamation. Institutional language develops for reasons, many of them legitimate. But habits of description can still produce political consequences nobody explicitly intended. If every organized deployment of coercion is described simply as “goonism”, each episode appears isolated from the next. A pattern dissolves into incidents. Coordination becomes coincidence. Systems are mistaken for spectacles.
The philosopher Michel Foucault, writing in Discipline and Punish, argued that power does not merely prohibit or repress; it produces the very categories of knowledge through which reality becomes intelligible – that power and knowledge “directly imply one another”. Political violence is no exception. A society rarely misunderstands coercion because it lacks information. More often, it misunderstands coercion because the available vocabulary was built for a different political landscape.
That lag is not merely semantic. It is democratic. If the language available to citizens consistently individualizes what is organizational, accountability is weakened before any investigation has even begun. Language becomes part of the architecture of impunity – which is why precision should not be mistaken for pedantry. To ask whether an organized operation is more accurately understood as a criminal network, a politically sponsored militia, or a proxy force is not academic classification. It is an attempt to ensure that our vocabulary expands as our understanding does.
A society that consistently misnames violence eventually ceases to perceive its architecture at all. By then, the vocabulary has already done its work – making the extraordinary appear ordinary, the organized appear accidental, the political appear merely criminal. To name precisely is not simply to describe reality. It is to refuse its concealment.
When the law already knows
There is an irony running quietly beneath the events at Ol Kalou. Long before journalists debated whether “goon” was the appropriate word, Kenyan law had already developed a far more precise vocabulary for thinking about organized wrongdoing. The law is often criticized – sometimes rightly – for moving too slowly. In this instance, it is public language that has lagged behind.
The legal architecture already anticipates a simple truth: violence is rarely the work of isolated individuals. Someone convenes. Someone finances. Someone coordinates. Someone expects a result. Liability does not begin with the person who throws the first stone or fires the first canister. Section 21 of the Penal Code – the doctrine of common intention – holds that when two or more people form a common purpose to pursue an unlawful act together, each is deemed to have committed whatever offence results, provided it was a probable consequence of that shared purpose. Section 393, in its own dedicated chapter on conspiracies, criminalizes the planning itself – the phone calls, the vehicle hire, the coordination – independent of whether the underlying felony succeeds. And Section 117, Conspiracy to Defeat Justice, reaches the officers who stand down not by judgment but by instruction.
The visible actor is not always the principal actor. The man holding the weapon may be the final link in a much longer chain of decisions. Law understands chains. Public conversation often does not – and that difference matters, because democratic accountability depends on following responsibility upstream rather than stopping where violence becomes visible.
This is why the Prevention of Organized Crimes Act, 2010, occupies such an important place within Kenya’s legal framework. POCA recognizes that dismantling organized criminal activity requires more than arresting those encountered at the scene; it seeks to follow financial relationships and organizational structures – asset forfeiture, financial-pipeline tracing, liability for anyone who manages or assists an organized criminal group. Whether and how its provisions apply here depends on the evidence investigators assemble and, ultimately, judicial determination. Its significance is conceptual as much as procedural: it demonstrates that Kenyan law already possesses instruments designed to think structurally rather than episodically – instruments that, in the weeks after July 16, went largely uninvoked. The DCI’s public statements spoke the vocabulary of an assault investigation, not an organized-crime one.
That is the larger lesson. The law already asks better questions than our headlines: who benefited, who organized, who communicated, who financed, who expected the operation to succeed. This is also why precision in public language is not an invitation to prejudge criminal liability. Quite the opposite. Responsible democratic language should resist two equal temptations: exaggeration, assigning legal categories before evidence justifies them, and minimization, relying on vague labels long after evidence has revealed organization that demands closer scrutiny.
The law cannot compel society to think clearly. It can only provide the vocabulary. The investigators traced vehicles. The journalists reconstructed events, down to a legislator’s number plate. The law already possessed concepts capable of following responsibility beyond the men visible in the streets. The remaining question belonged to all of us: would our language be willing to follow the evidence just as far?
The vote guards and the vanishing men
The images from Ol Kalou will eventually fade. The tear gas will disperse, the barricades will be cleared, the shattered windscreens replaced. News cycles, by their nature, move on. But democracies are shaped less by what they remember than by how they remember.
The question this essay has pursued is larger than a single by-election, a single convoy, or a single afternoon in Nyandarua County. It is a question about political memory itself. History is not only an archive of events. It is an archive of names. Every generation inherits a vocabulary through which it learns to recognize authority, resistance, order and violence – inheritances that are rarely neutral. Some words illuminate structures of power. Others obscure them until organized coercion appears as little more than recurring disorder.
Ol Kalou reminds us that democracy depends on two very different forms of vigilance. The first takes place in streets and polling stations, performed by citizens who refuse to surrender elections to intimidation. The vote guards embodied that vigilance. They possessed neither constitutional office nor statutory authority, yet understood something fundamental: democratic institutions survive only when ordinary people decide they are worth protecting.
The second form of vigilance is quieter. It takes place in language – in whether our descriptions possess the courage to keep pace with our evidence, whether our public vocabulary illuminates the architecture of power or merely records its visible effects.
This is not an argument for rhetorical escalation. Democratic societies should resist the temptation to inflate every act of violence into evidence of conspiracy; legal categories should follow facts, not precede them. But restraint is not the same as imprecision. A mature democracy owes itself the discipline of naming carefully – for the health of a republic depends not only on the integrity of its elections, but on the integrity of the language through which those elections are understood.
The DCI may reconstruct timelines. Journalists may trace vehicles to a sitting legislator’s garage. Human rights organizations may document patterns. Courts may determine liability. Each performs an indispensable democratic function. Yet there remains another responsibility that belongs to every citizen: to refuse the comfort of lazy description, to ask, whenever organized violence appears, not merely who acted, but how, under whose direction, through what networks, and toward what political purpose. Accountability has always travelled in the opposite direction from spectacle – spectacle invites us to stare at the hand that strikes; accountability asks us to follow the hand until it reaches the mind that instructed it.
The masked men who entered Ol Kalou will one day become a matter for archives. The language we choose to describe them may endure far longer. The vote guards understood that democracy is not self-executing – it survives because citizens show up. The vanishing men teach a different lesson: power rarely disappears. It simply learns new disguises. For every political order is remembered not only by the violence it permitted, but by the violence it was willing to name. And perhaps that is the first duty of a democratic citizen: to see clearly, to name honestly, and, having named honestly, to refuse to look away.
