What Is Liberalism?
I. The Problem It Was Trying to Solve
Liberalism begins as a refusal of arbitrary power.
This is not an abstract starting point. The thinkers who built the liberal tradition, Locke writing in the shadow of civil war and royal prerogative, Mill navigating the suffocating conformism of Victorian society, Rawls responding to the ideological catastrophes of the twentieth century, were each addressing a specific and legible injury: the capacity of power, unchecked, to override the lives of those subject to it.
The foundational intuition is that human beings possess a capacity for self-direction that arbitrary authority violates. There is something genuinely wrong about compelling a person to live according to values, beliefs, or arrangements they did not choose and cannot contest. Liberty, in this reading, is the condition under which a recognisably human life becomes possible at all.
From this intuition, the architecture follows logically. If the problem is unchecked power, the answer is constraint on power: law that binds rulers as well as ruled, rights the state cannot override, a private sphere into which public authority cannot reach. And if what is being protected is the capacity for self-direction, then the self must be the unit around which protections are organised. The individual becomes sovereign not as ideological preference but as logical consequence.
II. The Internal Logic
The architecture liberalism builds from this foundation has a genuine internal coherence. Its components are not arbitrarily assembled; each follows from the last with a logic that, once you are inside it, feels almost inevitable.
If the individual is the sovereign unit of political life, then individuals must possess rights that precede and constrain the state. Not rights granted by the state, which could be revoked, but rights that inhere in persons by virtue of their humanity. Locke’s natural rights, to life, liberty, and property, are the first expression of this move. They establish a domain the state did not create and therefore cannot legitimately destroy.
But rights without enforcement are wishes. The state must therefore be organised to protect them, which means the state must itself be subject to law. The rule of law, the principle that authority operates through publicly known, consistently applied rules rather than the discretion of the powerful, is liberalism’s answer to arbitrary government. It converts power from a personal possession into an institutional function, and in doing so makes it, in principle, contestable.
Contestable power requires a mechanism of contest. Democracy, in the liberal account, is not primarily about collective self-determination; that is a different and more demanding idea. It is about accountability. Elections are feedback mechanisms: the governed can remove those who govern badly. The franchise is the individual’s instrument for registering their judgment on the exercise of power over them.
This is where procedural neutrality enters, and it is doing more work than it first appears. If individuals are sovereign, and if they hold genuinely diverse views about how a good life should be lived, then the state faces a problem: whose conception of the good should it enforce? Liberalism’s answer is that it should enforce none of them. The state’s role is to maintain the conditions under which individuals can pursue their own conceptions of the good, not to adjudicate between them. Law concerns itself with harm and obligation, not with virtue or meaning.
This is the liberal settlement, and it has a real elegance. It permits genuine pluralism; people may live very differently within the same political order; while maintaining a shared framework of rights and rules. It separates the question of how we should live, which individuals answer for themselves, from the question of how we should govern, which the framework answers for everyone. The private and the public become distinct domains, each with its own logic.
The framework presents itself, crucially, as neutral between competing visions of the good life. It is not taking sides; it is providing a container within which sides can be taken safely. This is liberalism’s most powerful self-description, and the one most worth examining carefully.
III. The Genuine Achievements
The liberal architecture, where it has functioned as intended, produced things worth defending.
The abolition of torture as legitimate state practice. The establishment of habeas corpus and the presumption of innocence. The slow, contested, still incomplete extension of civil liberties to people initially excluded from them. The freedom to dissent from official doctrine without imprisonment. The freedom to practise religion, or none, without legal penalty. The freedom of the press to report on power without prior authorisation from power.
These are not small things. They represent centuries of struggle against arrangements that were once considered natural and permanent, and they have made material differences to the lives of real people. A political philosophy that delivered them has earned a serious hearing, whatever its limitations.
The rule of law, imperfectly realised as it remains, established the principle that authority is not identical with the person who holds it; that even the powerful are subject to publicly known rules they did not write for their own convenience. This was not inevitable. It required sustained institutional effort and it remains fragile wherever it exists.
Liberal democracy, for all its structural weaknesses, created mechanisms through which governments could be removed without violence. That is rarer in human history than it should be, and rarer in the present than comfortable assumptions suggest.
These achievements belong to liberalism. They should be held clearly before the analysis proceeds.
IV. The Stress Points
Every political philosophy contains tensions it cannot resolve from within its own terms. Liberalism’s are not incidental to its architecture; they are generated by it.
The first follows directly from the foundational move. The liberal subject, sovereign, self-directing, the unit around which all protections are organised, is imagined as prior to relation. They arrive at the social contract already formed, with a self to direct and interests to pursue. But human beings are not prior to relation. They are constituted by it. The capacity for self-direction that liberalism exists to protect is itself produced by families, communities, languages, traditions, and webs of mutual dependence that the liberal framework takes for granted without theorising.
This matters politically because it means liberalism can protect the exercise of agency without being able to account for the conditions that make agency possible. Where those conditions are stable and widely distributed, the gap is invisible. Where they erode, where communities dissolve, where the social fabric thins, where the inherited structures that formed subjects are disrupted, liberalism has no internal resources to respond. It can defend the right to self-direction; it cannot explain why self-direction has become, for so many people, increasingly difficult to exercise. The framework was not built to see that problem.
The second stress point concerns procedural neutrality, and it requires careful handling because it is liberalism’s most elegant and most problematic claim simultaneously. The liberal state, in its own account, does not take sides between competing conceptions of the good. It provides a framework within which individuals pursue their own visions of a good life, adjudicating only where those visions conflict. This is presented as an absence of position; the framework is simply the container, not a contestant.
But procedural neutrality is itself a substantive position, and a demanding one. It requires that all conceptions of the good be commensurable within a single framework of individual rights and public reason. Conceptions of the good that are irreducibly communal, or that locate meaning in obligations rather than choices, or that understand the self as constituted by its attachments rather than prior to them, these do not fit the container without being significantly reshaped in the fitting. Liberalism tolerates diverse ways of living; it is considerably less tolerant of diverse ways of thinking about what politics is for.
The framework immunises itself from this critique by classifying challenges to it as unreasonable or illiberal. To contest the procedural container is not to make a political argument within liberal democracy; it is to place yourself outside the bounds of legitimate political discourse. This is not a conspiracy or a deliberate act of suppression. It is what any framework does when it mistakes itself for neutrality. The story becomes invisible precisely because it presents itself as the absence of a story.
The third stress point is material. Liberalism requires certain conditions to function as advertised: a degree of economic independence sufficient for meaningful choice, access to education and information sufficient for genuine participation, security sufficient to make long-term autonomous planning possible. These are not luxuries; they are the preconditions for the sovereign individual the framework is built around. Without them, liberal freedoms are formal rather than substantive. The right to speak freely means something different to someone with a platform than to someone without one. The right to contract freely means something different to someone with options than to someone without them.
Liberalism has no internal account of how to produce or maintain these preconditions. Its architecture is designed to protect the individual from the state, not to equip the individual for self-direction. Where the preconditions exist independently, this gap is invisible. Where they are absent or eroding, the framework continues to operate; rights are formally intact, procedures are followed, while delivering progressively less of what it promised. The machinery runs; the output diminishes; and the framework cannot explain why, because the explanation requires concepts it was not built to hold.
The fourth stress point is where the previous three converge. The public/private distinction that liberalism places at the heart of its architecture is not neutral territory. It is where structural power lives.
Liberalism’s property rights argument does essential double duty. As a protection against the state, it is genuinely valuable; your home, your labour, your accumulated resources cannot be arbitrarily seized. But property rights also naturalise existing distributions. What you hold is pre-political; it precedes the social contract and is therefore beyond its reach. Locke’s original proviso, that appropriation is legitimate only while enough and as good remains for others, disappears from the tradition almost immediately, leaving the protection in place without the condition that justified it.
The consequence is that the domain liberalism places beyond political contestation is precisely the domain where the most consequential power relations operate. The authority of the employer over the employee, the landlord over the tenant, the creditor over the debtor; these are private relationships between formally sovereign individuals, matters of contract rather than politics, and therefore not the kind of thing the liberal framework is designed to address. The power they represent is real and often coercive, but it is structurally invisible within liberalism’s own terms. It is not tyranny; it is just the natural order of voluntary exchange.
This is not a corruption of the liberal tradition. It follows from the architecture. The private sphere was always going to be where power concentrated once the state was constrained, because power does not disappear when you limit one of its expressions. Liberalism solved the problem of arbitrary state power with considerable success. In doing so, it created the conditions under which a different kind of arbitrary power could operate beneath the threshold of political visibility, protected by the very framework built to resist domination.
V. The Cybernetic Diagnosis
Liberalism, read through a cybernetic lens, is a sophisticated feedback architecture. It is worth being precise about what that architecture does well before identifying where it fails.
At the first order, liberalism’s mechanisms are genuine and significant. Elections allow the governed to remove those who govern badly. A free press generates information about the exercise of power. The rule of law creates predictable, publicly known constraints on state behaviour. These are real feedback loops, and they work. Compared to the systems liberalism displaced, and compared to the systems that exist where it has never taken hold, they represent a substantial improvement in a state’s capacity to receive and act on signals from those it governs.
At the second order, the picture is more complicated. Liberal systems can update policy continuously; this is precisely what electoral feedback is designed to produce. What they cannot easily update is the framework itself. The procedural container is, by design, placed beyond ordinary political contestation. This is not irrational; a framework that could be dismantled by a simple majority would offer weak protection against tyranny. But it means that when the framework itself is generating the problem, the feedback loop has no address to deliver its signal to. The system can hear complaints about outcomes; it is structurally hard of hearing when the complaint is about the architecture.
This is where the third order failure becomes specific. The story liberalism’s mechanisms tell is coherent and genuinely appealing: you are a sovereign individual, your rights are protected, the procedures are neutral, your participation is meaningful. Where the material and social preconditions for that story are intact, the experience of living inside a liberal system broadly confirms it. Where those preconditions are absent or eroding, where formal rights do not translate into substantive agency, where neutrality operates between individuals but not between structural positions, where the private sphere is experienced not as freedom but as exposure to unaccountable power, the story and the experience diverge.
That divergence is the third order failure. Not that liberalism lacks a theory of legitimacy, but that its mechanisms actively produce a legitimating narrative that an increasing number of its participants cannot reconcile with their lived reality. The framework cannot process this as a signal, because doing so would require it to recognise itself as the source of the story. The neutrality commitment, liberalism’s most powerful self-description, is precisely what prevents that recognition. A framework that mistakes itself for the absence of a position cannot hear the feedback that its position is generating consequences.
A final diagnostic question concerns what might be called the exits problem. In any regulatory system, exits are the mechanisms by which divergence from the central narrative remains possible without becoming existential. They are the structural equivalent of coexistence; the capacity to tolerate challenges to the framework as information rather than threat. Systems that close off exits do not become more stable; they become more brittle. Divergence that cannot be expressed through legitimate channels does not disappear; it accumulates.
Liberalism is genuinely tolerant of diverse ways of living within its framework. It is not tolerant of diverse ways of thinking about what the framework should be. Challenges to the procedural container are not processed as political arguments; they are classified as unreasonable, illiberal, or dangerous. The system has exits for individuals; it has no exits for the framework itself. When the framework is the problem, that closure is not incidental. It is structural.
VI. The Historical Instantiation
Liberalism’s historical record is neither the story its defenders prefer nor the story its critics usually tell. It is more instructive than either.
The colonial question cannot be avoided, and it should not be softened. The liberal tradition developed in precise historical parallel with European empire, and the relationship between them is not coincidental. It is the stress points in operation.
The framework’s inability to account for the conditions that produce the sovereign individual meant it had no internal mechanism for recognising structural exclusion as a political problem. Colonised peoples did not register within liberalism’s architecture not because liberals made a conscious decision to withhold the framework’s protections, but because the framework was not built to see what was happening to them as politics at all. They fell below the threshold of visibility the architecture was capable of achieving. The private sphere’s naturalisation of existing property relations extended, without obvious strain, to the appropriation of land and labour on an industrial scale. The procedural neutrality that protected dissent at home provided no vocabulary for what was happening abroad, because what was happening abroad was prior to politics, not part of it.
When justification was demanded, dualism supplied it; the civilised and the uncivilised, the developed and the undeveloped, those ready for self-government and those who were not yet. But the justification was post-hoc. The invisibility preceded it. Which is the more structurally damning point: liberalism did not need to be consciously hypocritical to enable empire. It needed only to be itself.
The twentieth century welfare state represents a different and more ambiguous moment. Faced with the material preconditions problem, the growing visibility of the gap between formal liberal freedoms and the substantive agency they were supposed to deliver, liberal democracies in the post-war period made a series of significant concessions to social democratic pressures. Public healthcare, unemployment insurance, universal education, social housing: these were not liberal achievements in the strict philosophical sense. They were the product of organised labour, socialist parties, and the political pressure generated by two world wars and a depression. Liberalism accommodated them rather than generating them.
What is instructive is what happened next. The welfare state settlement held while the material preconditions it created produced broadly shared experiences of substantive agency. When that settlement came under sustained pressure from the 1970s onward, liberalism’s architecture offered no internal defence. The private sphere’s protection of property relations, the neutrality commitment’s resistance to structural critique, the framework’s inability to process the conditions problem as a political question; all of these meant that the dismantling of the post-war settlement could proceed within perfectly orthodox liberal terms. It was, after all, just the restoration of individual freedom from state interference.
The current moment is the third case, and it is still unfolding. Liberal democracies across the Western world are experiencing a legitimacy crisis that their own frameworks cannot adequately diagnose. Participation is declining, trust in institutions is eroding, and the gap between the story liberal mechanisms tell, sovereign individuals, neutral procedures, meaningful participation, and the experience of living inside those mechanisms is widening visibly. The liberal response has been consistent and largely ineffective: more transparency, better communication, incremental reform within the existing framework. These are first order responses to a third order problem. The framework is generating the divergence; it cannot hear the signal because the signal is about the framework.
Each of these moments traces directly to the stress points identified earlier. None of them required bad actors or corrupted institutions. They required only the architecture functioning as designed.
VII. The Verdict
Liberalism is a genuine achievement in response to a genuine problem. Where arbitrary state power is the primary threat to human agency, the liberal architecture remains among the most effective solutions yet devised. Its feedback mechanisms are real, its protections are meaningful, and the freedoms it secured, however incompletely and unevenly, represent a substantial advance on what preceded them.
The structural verdict, however, is precise. Liberalism is a solution to one problem that generates a different set of problems when it mistakes itself for a complete account of political life. Its feedback architecture functions well at the first order and poorly at the third. It can hear complaints about outcomes; it cannot easily hear complaints about the frame that produces them. It protects the exercise of agency without being able to account for the conditions that make agency possible. It tolerates diverse ways of living while remaining structurally intolerant of diverse ways of thinking about what politics is for.
Most consequentially, it naturalises the conditions it was built inside. The relational substrate it depends on, the material preconditions it requires, the property relations it inherits, these are placed beyond political contestation before the analysis begins, and presented as the neutral background against which politics occurs. This is not a deliberate mystification. It is what any framework does when it cannot see its own foundations. But the effect is the same regardless of intention: the most consequential questions about how power is distributed and what conditions human agency requires are removed from the domain where liberalism’s considerable strengths could be brought to bear on them.
What liberalism cannot do is see the limits of what liberalism can do. That is not a trivial limitation. It is the one that matters most.