Good Intentions Are Not a Constraint on Power

"It is in vain to say that enlightened statesmen will be able to adjust these clashing interests, and render them all subservient to the public good. Enlightened statesmen will not always be at the helm."
— James Madison, Federalist No. 10 (1787)

Take a plastic bottle cap. Across the European Union it now stays tethered to the bottle, fixed there by a hinge of cheap plastic, because a directive decided that loose caps end up in oceans and tethered caps do not. The reasoning is not absurd. Most people support it, or notice it for a moment and forget it by the next sip. A classical liberal sees something else: a government using its monopoly on law to impose a single judgment on a billion small acts it has decided to supervise. The problem is not that the intent is bad. The intent is fine. The problem is that good intentions have never once been a reliable constraint on power.

This is what makes the state different from every other actor that can err. It is not one tool among many but a monopoly: it writes the law and enforces it, and it claims the sole right to use violence to back it up. Unlike a business or a club or a town, it cannot be left, outcompeted, or opted out of. The motive behind a grant of state power tells you nothing about what that power will eventually do.

The Asymmetry of Power

A business that misjudges its market loses customers to a competitor. A voluntary association that fails its members dissolves, and the members go elsewhere. A local government that governs badly faces residents who can pack up and leave. Every one of these mistakes is contained, because the people subject to it can adapt, compete, or exit, and the error corrects itself. The national state has no such mechanism. When it errs, everyone is subject to the error at once, with nowhere to go. And power granted for one purpose does not stay confined to it.

That is why the burden of proof on any new state power must be high, and why the question is never only whether this government will abuse it. Democracy is dynamic by design. Majorities shift, and the authority built today is inherited intact by whoever wins next. You do not grant power to a person; you grant it to an office, and the office will not always be held by people who share your values. Madison made this argument first. Federalist No. 10 built the American constitution on the assumption that enlightened men would not always hold office, and designed for the ones who would not. Build a tool for your friends and you have built it for your enemies, because the same tool changes hands the moment the vote does.

How Power Accumulates

The slippery slope earns its bad name when it claims that one rule must end in tyranny. That is not the argument. Each intervention normalizes a standard of reasoning — act wherever a problem exists and a solution can be imagined — and under that standard the scope of permitted action only grows while the threshold for the next one only falls. The PATRIOT Act was sold as temporary emergency measures. Then the powers became permanent and expanded past their original justification, until counterterrorism authority was reaching into ordinary domestic enforcement that had nothing to do with terrorism. The emergency did not end the powers. It established them, and left them available to every administration since.

The European Union needed no emergency at all. It began as a project of market integration and now reaches into product design, agriculture, labor, digital services, and the cap on the bottle. Each step was justified on its own terms. The cumulative result is a government in the habit of intervening anywhere a rationale can be built, and the rationale is never hard to construct. This is why classical liberals oppose even the well-intentioned expansions, the ones that will probably never be abused. Whether this particular government misuses this particular power is beside the point: the expansion normalizes the standard and leaves behind a capability that every later actor inherits. The question is not whether the power will be abused. The question is whether the power should exist at all.

Hate Speech Laws

Hate speech laws are the easiest case to use as an example, because the concern behind them is real. Speech can cause measurable harm, and most people who support these laws are not authoritarians but decent people responding to genuine injury. But speech is not commerce. It is the medium of democratic life itself: how citizens form their opinions, and how a despised minority argues its way into a majority. Once a government takes the authority to define which speech is allowed, it is shaping the political process from the inside, and a democracy cannot reliably claw that power back, because the power touches the very process by which it would be checked. There is a paradox the defenders rarely notice: they are using their own freedom of speech to argue that freedom away for everyone else.

The Soviet Union did not announce totalitarianism on day one. It began by defining certain speech as destabilizing and harmful to the collective good, and worked outward from there. "Politically correct" was a literal enforcement standard, not a joke. Harm in speech resists objective definition because it is context-dependent, and that is the whole problem: a future administration need not repeal a hate speech law to weaponize it. It need only reinterpret what counts as harm, because the architecture was installed the day the law passed and the only open variable is who operates it. Classical liberals do not reject the concern with harm. They reject building the machine.

The Powers We Accept

None of this is an argument against government as such. Some functions are monopolistic by their very nature, because a decentralized alternative is not merely worse but incoherent. A country cannot field competing armies. A city cannot run parallel sewage systems. National defense, law enforcement, foundational infrastructure, and a basic floor of education sufficient for citizens who must one day vote all require centralized authority, because the alternative to the central provision is not a market in the service but the absence of it. These powers are accepted not reluctantly but as a recognition of what the problem actually requires, and the test is always the same: is a decentralized alternative genuinely viable, or only imaginable?

The monopoly on violence that lets the state enforce its laws is identical in kind to the one that lets a state become tyrannical. There is no separate, gentler power held in reserve for legitimate purposes. What separates a functional democracy from an authoritarian one is not the absence of power. It is the framework around the power.

That framework is constitutional limits and the separation of powers, and it is the foundation, not the decoration. It defines the scope of authority and shields certain rights from the pressure of any transient majority, and it can be widened only by a broad and durable consensus. Some of these limits exist only because they inconvenience whoever currently holds office, and the inconvenience is the entire point. A government constrained only by its own goodwill is not constrained at all.

The Alternative

The alternative is not the absence of collective problem-solving. It is the insistence that how a problem is solved matters as much as whether it is solved. Markets, voluntary associations, and local governance are preferable not because they are flawless but because none of them demands a permanent transfer of authority to a body no one can escape. Each carries its own correction — the customer who walks, the resident who moves one town over — and none of those corrections survives once the actor is the national state. That is the whole difference.

This is the argument both the left and the right have abandoned. They no longer fight over how much power should exist; they fight over who controls it, each building tools in the confidence that their own hands will wield them. Both are right about the present and wrong about the future. The left's regulatory architecture will end up protecting the incumbents it meant to check and silencing the dissent it meant to protect. The right's expanded executive will be operated by administrations it cannot stand to look at. The classical liberal is an unpopular messenger because the argument cuts against every coalition at once.

All of this returns to the cap on the bottle. It is a trivial example of a principle that is not trivial at all. Once good intentions are accepted as sufficient justification for intervention, the principled ground for refusing the next one is gone, because the next one will have good intentions too. And there is always a next one.