What if no one could independently define what counts as an emergency?

This question is at the heart of some of the most important constitutional scholarship of the past century. Carl Schmitt, whose views were sinister but whose analysis was acute, defined sovereignty as the power to decide on the exception - meaning whoever decides when normal rules are suspended is, in the deepest sense, the sovereign. That observation has haunted constitutional democracies ever since.

Most constitutional systems try to solve the problem through temporal and procedural constraints: emergency powers must be declared by a specific mechanism, confirmed by the legislature within a defined period, subject to judicial review, and time-limited. The assumption is that even if the initial declaration cannot be independently verified, the subsequent checks will catch abuses.

The problem the hypothetical identifies is what happens when those checks fail simultaneously. A legislature that is politically aligned with the executive will confirm rather than scrutinise. Courts whose jurisdiction is contested or whose rulings go unenforced cannot perform their review function. International bodies have limited coercive reach. The structural checks are real, but they depend on actors who are willing and able to use them.

What the pandemic revealed, in various countries, is how much variation there is in how these checks actually function under stress. Some systems showed remarkable resilience. Others showed that emergency powers, once extended, are difficult to retract and that the political incentives around their maintenance are not symmetric - there are always arguments for keeping them that are harder to counter than arguments for removing them.

The answer to "who defines the emergency" should be: multiple independent actors, no one of whom can do it alone. When that answer becomes untrue, you have a constitutional crisis whether or not anyone has announced one.

There is a formal structure here worth making explicit. Any system that allows a single actor to both declare an emergency and then exercise expanded powers as a result of that declaration has a logical circularity problem: the very condition that unlocks the powers is under the control of the beneficiary of those powers. This is not merely philosophically untidy - it is a structurally unstable equilibrium.

In game-theoretic terms, the situation creates a specific incentive structure. If the person who declares emergencies benefits from declaring them, and if the cost of a false declaration is lower than the benefit of expanded powers, the rational strategy is to find reasons to declare emergencies. The optimal constraint is therefore not to trust the declared reason, but to build the constraints into the structure of who can declare and under what verifiable conditions.

The interesting design question is whether you can specify "emergency" formally enough that the declaration becomes verifiable independently of the declarer's intentions. Some emergency conditions are relatively objective: a foreign invasion is visible to many parties simultaneously. Others are genuinely contested: an economic emergency, a public health emergency, a threat to national security. The latter categories are where the definitional power matters most.

Constitutional systems that try to enumerate the conditions for emergency declaration are attempting this formalisation. The problem is that reality produces novel situations the enumeration didn't anticipate, and the declarer will always claim that the novel situation is close enough to a listed category. You cannot write a complete specification of emergencies in advance, which means you cannot fully eliminate the problem.

Independent judges reviewing declarations after the fact is the best available approximation of a solution. It is not a complete solution. There is no complete solution.

I want to offer a perspective from inside the political process, because I think this discussion sometimes treats politicians as if they are always trying to accumulate power rather than genuinely trying to manage impossible situations under significant uncertainty.

Most emergency declarations I've observed or been adjacent to were made by people who genuinely believed the threat was real and the powers were necessary. The problem is that sincere belief is not a constraint on abuse - it is actually part of what makes the situation dangerous. Leaders who genuinely believe they are facing an existential threat are more, not less, willing to stretch the definition of what emergency powers permit.

The institutional constraints that matter most in practice are not the ones written into the declaration mechanism. They are the ones that affect political cost: independent media that can effectively challenge the government's framing, opposition parties that can credibly contest the necessity claim, a civil service that pushes back on legally questionable instructions, and a public that maintains enough scepticism to distinguish real emergencies from manufactured ones.

These things are all softer and more contingent than constitutional text. But they are also what actually determines whether emergency powers remain bounded in practice. Governments have used emergencies to expand their power permanently in every era and across every political system. The check that has most reliably worked is not legal but political: the cost of being seen to abuse the emergency becoming higher than the benefit of the additional power.

When that accountability mechanism fails, the constitutional provisions rarely save you on their own. The question of who defines the emergency is ultimately political, not legal, however much we might wish otherwise.

I want to talk about what "emergency" actually means, because I think it means something specific in my former world that it does not mean in political discourse, and the difference matters.

In operational terms, an emergency is a situation in which the expected plan no longer applies, time has compressed to the point where normal decision processes can't operate, and the stakes of inaction are immediate and concrete. A contact with the enemy is an emergency. A vehicle breakdown on a route being used by others is an emergency. These are emergencies because the definition is clear, the threat is verifiable by multiple people simultaneously, and the person declaring it cannot easily benefit from doing so — they are, in fact, taking on more risk and more accountability, not less.

What political emergency powers do is something structurally different. The person who benefits from the declaration of an emergency is the same person authorised to declare it. The threat is generally not immediately verifiable by independent parties. And the powers granted are almost always broader than the threat requires, because "emergency" in the political sense tends to expand to fill the space available to it.

I have operated in genuine emergencies. In those situations, command authority concentrates because it has to — the time constraints are real and the costs of paralysis are physical. But two things accompany that concentration. First, the authority is bounded by the situation: when the situation ends, the emergency authority ends with it. Second, accountability afterwards is thorough. You can exercise emergency authority in the field, but you will account for every decision made under it. The after-action review is not optional.

The gap between that model and civilian emergency powers is the gap between a system designed to concentrate authority for operational reasons and one designed to concentrate authority for political ones. The article is right that the question of who defines the emergency is the central one. From where I stand, the answer should be: not the person who gains from the definition. And the accountability should be as thorough, afterwards, as it would be in a military context. Which it usually isn't.