There is a scene that has repeated itself with enough regularity in Nepal over the past few years that it has started to feel normal. A government directive needs to be carried out. Maybe it is a demolition order, maybe an eviction, maybe the detention of a businessman or a political figure. Before courts have issued rulings, before affected parties have been notified, before any of the usual administrative processes have run their course, security forces arrive. The operation is swift. The result is visible. The process was the operation itself.
This is not how governance is supposed to work. But under Balendra Shah, first as mayor of Kathmandu Metropolitan City and now as Prime Minister, it has increasingly become how governance works. Security forces are being used not to respond to genuine threats but to execute administrative objectives that would otherwise require the slower, messier work of going through courts, councils and proper institutional channels. The force does not follow the decision. The force is the decision.
The April 2026 evictions along the Bagmati riverbank brought this pattern into sharp relief. Thousands of people, many of them from poor and marginalized communities who had lived along the river for years, were displaced through operations in which security personnel were the most visible presence. There were no adequate resettlement alternatives. The timelines did not allow for meaningful legal challenge. Families lost homes, documents and livelihoods in a process where due process was not delayed or imperfect but structurally absent. The security operation was the entire governance process.
To understand why this matters, it helps to be clear about what it is and is not. This is not a story about a government deploying security forces inappropriately in a crisis. Every government does that. This is something different: a pattern in which security deployment has become the routine mechanism for achieving governance objectives, rather than an exceptional response when everything else has failed. The difference is not semantic. One is emergency governance; the other is governance by emergency.
Nepal’s security forces are paying a price for this that rarely gets discussed. During the decade-long armed conflict (1996-2006), the Nepal Police, the Armed Police Force and the then Royal Nepali Army accumulated deep public distrust through documented brutality and rights violations against civilians. Entering the peace process gave them an opportunity to rebuild. Democratic positioning of the army along with programs like Police My Friend and community policing initiatives were deliberate efforts to change the relationship between uniformed personnel and the communities around them. Those efforts took years and produced real, measurable improvement in how Nepali people related to the institutions meant to serve them.
Much of that rebuilt trust collapsed during the GenZ protest movement of Sept 2025. The crackdown on protesters produced images that will not quickly fade: police officers abandoning their posts, weapons left behind in the chaos. The Nepali Army’s conduct raised separate and equally serious questions. When major public infrastructures were set ablaze during the protests, including Singhdurbar (the seat of the Executive), the Supreme Court and the Office of the President, the army stood conspicuously passive. Its failure to intervene to protect institutions of the state has never been satisfactorily explained to the public.
More recently, reports of the army collecting data on landless people across various districts have raised alarm about whether it is operating beyond the boundaries its constitutional mandate sets. When the institution charged with defending the constitution appears to be overreaching into civilian administrative functions, it signals a deeper confusion about what security forces are for.
The Bagmati evictions have added yet another layer. Security personnel are being publicly identified as the people who came and took away the homes of families who had nowhere else to go. Whether those officers agreed with the orders or had any say in the decision are questions most observers are not stopping to ask. The association between the force and the harm is direct and visible. The policy decision that produced the harm is several steps removed and considerably less visible. The government makes the call; the security forces wear the consequences.
Running underneath all of this is a problem that Nepal has never adequately confronted: there is no clear, publicly understood framework for when government security forces should be deployed and when they should not. Even where rules exist in law, they have not been communicated in any meaningful way to the public. This ambiguity is not a technical oversight. It is a structural vulnerability. When the boundaries of legitimate deployment are vague, those boundaries become easy to push. Leaders in government can use security forces to consolidate their hold on power, to intimidate opponents, and to bypass institutional checks, all without obviously crossing a line that has never been clearly drawn.
Left uncorrected, this is precisely the environment in which elected authoritarianism takes root. A leader arrives through the ballot box but governs through coercion, and the institutions that should resist find they have no clear ground to stand on.
None of this means that governments should not enforce laws, that informal settlements on public land are beyond regulation, or that accountability for corruption does not matter. Legitimate governance pursues legitimate objectives. But it pursues them through processes that respect rights, allow for contestation, and remain accountable to the people they affect. When security force deployment substitutes for that process, even legitimate objectives become harder to defend.
Since People’s Movement 2006, Nepal has spent 20 long years building, imperfectly and unevenly, a democratic system premised on civilian supremacy over security forces. That principle is not self-enforcing. It requires clearly defined rules about when the state's coercive apparatus can be activated, by whom, and subject to what oversight. What is needed now is not simply a change in political leadership but a serious, public reckoning with the scope and limits of security force deployment in a democracy. The habits being established today will be difficult to unlearn.
A government that routinely governs by force teaches its successors that force is how governing is done. That lesson, once normalized, does not stay contained to one administration. It becomes available to whoever comes next, with whatever objectives they happen to hold.
That is the question worth sitting with: not whether Balendra Shah's governance goals are right or wrong, but whether the way he is pursuing them is building the Nepal he says he wants, or quietly dismantling the foundations that any decent governance ultimately depends on.
The author is currently a Visiting Research Fellow at the Kroc Institute for International Peace Studies, University of Notre Dame, USA
