Anarchism, in its philosophical form, is not the black-flag radicalism associated with street politics. It is a precise and unsettling claim: that no state—however democratic or well-intentioned—possesses genuine moral authority over the individuals it governs. This does not mean states are always unjust. Rather, anarchists argue that the law itself does not automatically create moral obligations. You may have good reasons to act as the law requires, but those reasons exist independently of the law. If anarchists are right, then the idea of political obligation—that citizens must obey the state simply because it is the state—is less a moral principle and more a social habit sustained by fear, convenience, and tradition.

At first glance, this argument may seem extreme. But philosophical anarchism is not a fringe idea. It poses a serious challenge that political theory has struggled to answer convincingly.

The most intuitive answer—consent—collapses almost immediately under scrutiny. Locke argued that individuals who voluntarily join civil society bind themselves to its laws. The problem, as Hume observed in Of the Original Contract, is that virtually no one has consented to anything. You did not sign a contract with the United States. You were born into it. To say that remaining in the country of your birth constitutes tacit consent is, as Hume put it with characteristic precision, tantamount to saying a man tacitly consents to obey a ship's captain though he was carried aboard while asleep and whose only alternative is to leap into the ocean. Simmons, one of the most rigorous contemporary defenders of anarchism, shows that consent capable of grounding genuine political obligations is extraordinarily rare, and that stretching the concept of tacit consent to cover the ordinary citizen renders it normatively empty. The Socratic argument from Crito, that benefiting from the state's protection implies a reciprocal obligation on your part, is a more convincing version of the same idea, but it too rests on a form of tacit consent and inherits its shortcomings.

Rousseau's general will proposition is the notion that citizens, when they abstract from their private interests and deliberate as civic equals, converge on a common good that legitimates the authority of law. But is the general will an actual democratic decision, or a transcendental ideal that citizens approximate but never fully reach? If the former, Rousseau himself admits that majorities can be wrong about what it requires, which means democratic outcomes do not automatically carry authority. There are some things that can never be right in a Rousseauian sense even if the majority thinks otherwise. If the latter, then the general will provides no determinate standard by which to assess whether any actual state is legitimate. Fatally, Rousseau's civic republicanism presupposes a homogeneity of values—citizens sharing commitments to particular conceptions of the good life—that is simply absent from modern pluralistic societies. A theory of political obligation that only works when everyone broadly agrees is not a theory of political obligation at all. Rousseau’s theory hinges on citizens being able to relate to each other on a deeply personal level, and I can’t see that happening in pluralistic societies with vast disparities not just in income and status but in race, culture, and values.

Nozick, on the other hand, accepts natural rights as foundational and then tries to show that a minimal state can arise through voluntary processes without violating anyone's rights. The difficulty is that his evolutionary story—in which competing protection agencies gradually consolidate into something state-like—is not obviously stable. Why would protection agencies converge rather than fragment into competing powers, as the anarchist expects? And even if Nozick's story succeeds, it only justifies a minimal state providing protection and contract enforcement, a state that is too weak to prevent powerful private actors from dominating those who lack resources. The anarchist can accept Nozick's minimal state and still argue that it entrenches private domination rather than dismantling it.

Consequentialism seems to offer a cleaner route: states are justified because they produce better outcomes than anarchy. But this dissolves political obligation rather than vindicating it. If my reason for obeying the law is simply that doing so tends to maximise welfare, then I have no special obligation to obey as a citizen—I merely have a general obligation to do whatever produces the best consequences, which may or may not coincide with what the law requires. The state becomes a useful heuristic rather than a genuine authority. And the anarchist has a pointed rejoinder: a framework that permits punishing innocents when doing so would deter crime more effectively, or sacrificing the interests of minorities to maximise aggregate welfare, is precisely the kind of institution they are warning us about.

John Rawls's political liberalism, grounded in what he calls the natural duty of justice, is the most formidable philosophical response to anarchism available. It sidesteps the empirical failures of consent theory, the homogeneity requirements of republicanism, the minimal-state limitations of libertarianism, and the authority-dissolving tendencies of consequentialism—and it does so by going directly to the anarchist's core concern, which is autonomy.

The anarchist's deepest claim, articulated with particular force by Robert Paul Wolff, is that authority and autonomy are fundamentally incompatible. To accept authority is to surrender your right to judge for yourself whether a command should be obeyed, and that surrender, Wolff argues, is a betrayal of the self-legislating capacity that defines moral agency. Rawls's response is to turn this argument on its head: autonomy, properly understood, generates duties of justice rather than precluding them. Rational agents deliberating behind the veil of ignorance—stripped of knowledge of their particular position in society, their talents, their conception of the good—would unanimously endorse two principles: equal basic liberties for all, and socioeconomic inequalities permitted only insofar as they maximally benefit the least advantaged. Crucially, choosing these principles is itself an exercise of autonomy. The veil of ignorance ensures that no one is imposing on anyone else; the principles represent what all could reasonably accept from a position of equality. When I comply with a just legal system, I am not surrendering my autonomy. I am participating in the collective realisation of principles I myself endorse as a rational agent.

There is a further move Rawls makes that is easy to underestimate. Even if every individual scrupulously respected every other individual's rights, justice could not be secured through individual moral action alone. We live in conditions of moderate scarcity, where reasonable people disagree about the principles governing fair distribution and cooperation. Without institutions capable of authoritatively resolving these disputes and enforcing determinate rules binding on all parties, each person's unilateral interpretation of their rights collides with everyone else's, creating what Rawls calls assurance problems. The natural duty of justice, then, is not merely a duty to be personally virtuous; it is a duty to support the institutions that make justice interpersonally determinate and collectively achievable.

And yet, even granting all of this, Rawls does not fully defeat anarchism. Two problems remain.

The first is the particularity problem. Rawls's argument establishes a general duty to support just institutions but not a specific duty to obey this state rather than another. In a world with multiple states approximating justice to varying degrees, why am I bound to the United Kingdom rather than to New Zealand or Singapore? The original position gives me no answer. It tells me I should support just institutions wherever they arise but says nothing about why the accident of birth should determine which particular institution commands my allegiance. Simmons writes that the natural duty of justice is too general to generate the specific, practice-bound obligations that political obligation actually requires.

Anna Stilz attempts to fill this gap by arguing that individuals have a pre-institutional right to occupy the territory where their lives and plans are situated, and that a state which effectively protects basic rights and grants people a voice in defining them thereby acquires jurisdiction over those particular individuals. It is an elegant argument, but it does not fully close the gap. The anarchist can accept that I have an occupancy right while denying that this right generates an obligation to obey the state that happens to govern this territory. Stilz's account also quietly assumes that states have already legitimately partitioned the world into exclusive jurisdictions—which, as you know, is precisely what the anarchist contests.

The second problem is content-independence. We ordinarily believe that citizens have some obligation to obey the law because it is the law, not merely because each particular piece of legislation happens to promote justice. But if obligation flows from the natural duty of justice, then my reason for obeying must always trace back to whether this specific law serves justice. Laws that do not serve justice carry no authority; laws that do serve justice carry only the authority they derive from their justice, not from their status as law. Political obligation collapses into ordinary moral obligation, and the state loses any distinctive claim on my conduct. Rawls's framework generates something closer to a duty to promote justice than a duty to obey the state, and those are not the same thing. A legal positivist like H.L.A. Hart would find the idea of laws anchored in justice bizarre, because the fact of the matter is that the power to make laws exists independently of the content of those laws.

The bar is very high. Rawls mounts the strongest possible philosophical counter to anarchism because he is the only major political philosopher who confronts the tension between autonomy and authority directly and attempts a reconciliation rather than dismissing one side or the other. Every alternative theory—consent, republicanism, libertarianism, consequentialism—either founders on empirical grounds, presupposes what it needs to prove, or concedes too much to the anarchist at the outset.

Rawls can’t quite win because the particularity problem and the content-independence problem are structural features of any theory that grounds political obligation in general moral principles rather than in the particularistic bonds of actual political life. The conclusion I arrive at is not that political obligation has been vindicated, but that it has been defended as well as it can be defended. This means anarchism can never be dismissed entirely. At best, we can only dispel half the substantive weight of its claims.

If there is one thing to take away from this school of thought, it is the importance of questioning legitimacy whenever we can. That habit of scrutiny is fitting—and essential—for the perilous political times we live in.