Introduction

The Case Reopened

R. M. Hare was not refuted. He was abandoned.

The distinction matters. Philosophical theories are sometimes defeated by arguments that expose contradictions, unsupported premises, or consequences their defenders cannot accept. At other times, they lose their place because the profession’s interests change, because rival approaches seem more promising, or because an argument’s unresolved difficulties come to be treated as reasons to stop pursuing it.

Hare’s universal prescriptivism suffered something closer to the second fate. Its difficulties were real, and some remain unresolved. But the central achievement was never decisively overturned. Hare had identified a discipline of moral reasoning that his successors could criticize, modify, or supplement, but could not simply replace by adopting a different moral outlook.

He argued that making a moral judgment commits the speaker to something. To say that an action is morally required is not merely to describe one’s feelings, report a social convention, or express a preference. It is to prescribe conduct under a principle that must apply to every relevantly similar case, including cases in which the speaker occupies a different position.

The implications are demanding. A person cannot claim moral authority for a judgment while reserving an exemption for themselves simply because they are themselves. They must be prepared to accept the prescription when the identities of those affected are exchanged and the relevant circumstances remain the same. Hare did not present this requirement as a substantive moral ideal we are asked to adopt. His argument is that it is already implicated in the kind of judgment we are making.

Later moral philosophy preserved universalization in various forms. What it largely abandoned was Hare’s attempt to show that universal accountability is not an optional ideal supplied by a theory, but a commitment already contained in making a moral judgment at all. This book argues that the abandonment was a mistake.

A Philosopher Formed by War

Hare came to moral philosophy through an experience that gave the problem of moral disagreement an urgency beyond the seminar room. During the Second World War he served as a British officer, was captured at the fall of Singapore in 1942, and spent more than three years as a prisoner of the Japanese, including labor on the Burma–Thailand railway. He encountered circumstances in which inherited convictions, patriotic loyalties, religious beliefs, and ordinary moral assurances could not be relied upon to settle the questions that mattered most. The experience helped shape his determination to find a way of answering moral questions rationally.

He did not think moral philosophy could accomplish this by discovering a new collection of intuitions, however compelling. People already possessed moral convictions in abundance. The problem was that convictions differed, sometimes violently, and the intensity with which someone held a conviction provided no guarantee of its correctness. Nor would it suffice to identify the moral traditions of a particular society. Traditions could conflict, and a person could have excellent reasons to question the practices into which they had been born.

Hare sought something that could operate across such disagreements: a method of reasoning whose authority did not depend upon the disputants already sharing the moral conclusions at issue. His answer began with the logic of moral language.

The Discipline of Reversing Positions

Hare’s theory is called universal prescriptivism because moral judgments, on his account, possess two essential features. They are prescriptive: sincerely judging that something ought to be done commits the speaker, in the relevant practical sense, to doing it when the judgment applies to their own conduct. And they are universalizable: the prescription must extend to every case sharing the relevant universal features, so a moral judgment cannot depend upon an arbitrary exception for the person making it.

Neither feature, considered alone, captures the full discipline Hare intended. Universalizability is sometimes caricatured as a demand for verbal consistency. A person announces a rule, identifies the conditions under which it applies, and agrees to apply it whenever those conditions recur. But this leaves open a crucial question: which features of a situation may legitimately determine whether two cases are alike? A sufficiently ingenious speaker can protect almost any prejudice by building it into the description of the case. If the rule is that members of one caste must serve members of another, the speaker can insist that caste membership is precisely the difference that matters.

Hare’s answer requires more than manipulating predicates. The speaker must confront what their prescription entails from the positions of the people affected. Hare’s own illustration, in Freedom and Reason (1963), is a creditor deciding whether to have his debtor jailed, who must ask whether he can accept the same prescription applied to himself as debtor. The structure is clearer still in a case of entrenched hierarchy.

Suppose an aristocrat declares that peasants ought to obey nobles. The aristocrat may understand perfectly well that peasants dislike their subordination. They may know about the exhaustion of agricultural labor, the humiliation of servitude, the insecurity of dependence, and the frustration of ambitions that their social position prevents them from pursuing. They may even understand these experiences with considerable sympathy. None of that, by itself, constitutes Harean role reversal.

The decisive question is whether the aristocrat can prescribe the same arrangement for themselves in the peasant’s position, adequately informed about what occupying that position involves, including the preferences of the person whose position they are taking. Here the argument turns. The peasant’s desire for freedom is no longer merely an observed fact about someone else. It is part of the counterfactual standpoint from which the aristocrat must be prepared to sustain the prescription: can they prescribe their own subordination while occupying that position, with its exhaustion, humiliation, attachments, ambitions, and desire for autonomy?

The aristocrat cannot evade this test by imagining a peasant who welcomes servitude, because that would change the circumstances to which the original prescription applies. They must confront the actual peasant’s preferences as preferences they would possess in the reversed position. They may still maintain that some overriding ideal justifies frustrating those preferences, but they must be prepared to prescribe that frustration for themselves. Whether their professed commitment to hierarchy can survive that demand is not settled in advance. But it is a far more demanding test than asking whether the aristocrat can understand another person’s suffering while remaining indifferent to it.

Third-person understanding can be entirely instrumental. A manipulator may understand another person’s desires precisely because that knowledge makes manipulation easier; a torturer may understand what the victim fears because that understanding enables more effective cruelty. Neither is engaged in moral role reversal. Hare’s test concerns a speaker who has made a universal moral prescription and must now accept what that prescription requires when applied to themselves in the affected positions. The preferences belonging to those positions enter the speaker’s practical deliberation through the very operation of role reversal.

This is the distinctive power of the method. Hare’s argument does not need to begin by assuming a separate principle that everyone’s preferences deserve equal moral weight. It begins with the commitments incurred by making a universal moral prescription.

The Fanatic

There is, however, a limit. Imagine that the aristocrat completes the test and remains unmoved. They insist that preserving the hierarchy matters so much that, even if they occupied the peasant’s position, with all its suffering and frustrated desires, they would prescribe their own subordination. Or imagine a persecutor who sincerely believes that members of a despised group must be punished, and who would prescribe the same punishment for themselves if they belonged to that group. Such a person may satisfy universalizability while sustaining a morally appalling judgment.

Hare recognized this possibility. In Freedom and Reason he called attention to the Fanatic: someone whose overriding ideal survives the full demands of role reversal. The Fanatic is a serious problem, but its significance must be understood correctly.

The logical possibility of such a person does not establish that ordinary prejudice survives Hare’s test. A familiar bigot who favors oppression while enjoying the advantages of belonging to the dominant group may discover, upon genuine reversal, that the principle they profess is one they cannot prescribe for themselves. The pure Fanatic must do something more difficult: their ideal must remain powerful enough to prevail against the preferences they would possess in the disadvantaged position. Real people value their own freedom, their families, their security, their dignity, their comfort, and their opportunities. An ideology that appears commanding when applied to other people may become much less compelling when its demands are genuinely brought into competition with those interests.

Hare therefore had an empirical argument as well as a logical one. Although the fully consistent Fanatic is possible, ordinary moral agents are unlikely to sustain the requisite commitments when the affected positions are represented adequately. The two claims should be kept apart. Hare did not demonstrate that fanaticism is logically impossible, nor did he establish, merely by defining the Fanatic, how many actual people would survive the test. What he provided was a procedure capable of exposing the difference between ordinary self-serving conviction and a genuinely overriding ideal. The former is vulnerable to role reversal. The latter presents a remaining problem for substantive moral reasoning.

This matters because critics sometimes treat the mere possibility of a consistent fanatic as though it rendered universal prescriptivism powerless against everyday hypocrisy, racism, caste prejudice, and exploitation. It does not. A method need not defeat every conceivable adversary to expose the failures of a great many actual arguments.

What Hare Established—and What Remains Open

Hare’s method is powerful, but its achievements should not be confused with all the conclusions he sought to derive from it. In his mature work, particularly Moral Thinking (1981), Hare argued that fully informed critical moral reasoning leads toward preference utilitarianism. The interests and preferences of affected individuals must be represented, and critical reasoning must determine what prescription can be sustained when all relevant positions are considered.

The movement from universal prescription to preference utilitarianism is one of the most ambitious parts of his philosophy, and it is where important questions remain. How are the relevant preferences to be represented? What does it mean to occupy another person’s position imaginatively when their experiences, capacities, or values differ profoundly from one’s own? How should conflicting preferences be compared? What happens when a principle survives role reversal but imposes extreme burdens on some individuals? Can the procedure resolve conflicts involving incompatible ideals, or does it require additional substantive premises?

These questions deserve answers. They cannot be settled by announcing that universalizability is either sufficient for morality or useless without a complete moral theory.

This book makes a more limited, and more defensible, claim. Hare established a discipline of universal moral accountability that survives the objections commonly treated as having displaced it. The further argument for a particular substantive moral theory requires separate examination. Drawing that line does not weaken Hare’s achievement. It allows us to identify it.

The Question Outside Moral Discourse

There is another limit, different from the problem of the Fanatic. A person may decline to make moral judgments at all.

A calculating predator may understand other people’s interests, anticipate their reactions, and use that knowledge to obtain what they want. They may regard moral language as a tool of persuasion or social control without accepting its prescriptions as binding upon themselves. Hare’s argument does not establish that such a person must become moral simply because they are intelligent. The predator has not failed to understand another person’s preferences; they may understand them exceptionally well. What they have not done is enter the practice of sincere universal moral prescription.

That is a different problem from the aristocrat who claims that subordination is morally right. The aristocrat has made a moral judgment and is accountable to the commitments it incurs. The predator may reject the practice of moral justification altogether. Why should a mind enter that practice? Why should it make its conduct answerable to principles that it must be prepared to prescribe from positions other than its own? Those are questions about the foundations of moral agency. They cannot be answered simply by showing what follows once moral judgment has begun.

The point holds for humans, and it holds for other forms of intelligence. A mind’s capacity to represent other minds does not establish that it has undertaken moral commitments. Conversely, a mind that does make moral judgments cannot escape their implications by appealing to the unfamiliarity of its origin or architecture. Why an intelligent being should enter moral justification remains an open question. It should not be confused with the question of what moral justification demands once undertaken.

How Hare Lost His Place

If Hare’s central argument remained important, why did moral philosophy move away from it? Not because philosophers stopped caring about universal principles or the standpoints of other people. John Rawls made impartiality central to his theory of justice through the original position and the veil of ignorance. T. M. Scanlon developed an account of morality based upon principles that others could not reasonably reject. Kantian approaches continued to connect morality with universal requirements of practical reason. These theories share important concerns with Hare, and their existence makes it misleading to say that philosophy simply abandoned universalization.

What changed was the account of its authority. Rawls’s original position is a carefully constructed device for representing fair terms of social cooperation. Its construction requires justification, and Rawls offers that justification through a broader account of fairness, political persons, and reflective equilibrium. Scanlon’s contractualism likewise requires an account of what makes rejection reasonable. That is a substantive philosophical undertaking, not a consequence that follows merely from the grammar of moral judgment. These are not defects that make either theory worthless. They mark the difference between the projects.

Hare sought to establish a discipline that a moral speaker was already committed to observing, whether or not they accepted a particular theory of justice, contractualism, or human flourishing. A person who rejects Rawls’s account of fairness can dispute the original position. A person who rejects Scanlon’s conception of reasonable rejection can dispute the contractualist test. Hare’s challenge begins earlier. The speaker has already claimed that something morally ought to be done. What follows from that claim?

Hare’s critics did not ignore that question. J. L. Mackie and others pressed hard on the step from universalizability to utilitarianism, and the 1988 volume Hare and Critics gathered such objections together with Hare’s replies. But the objections that gained the widest acceptance targeted the further utilitarian conclusion and the theory of meaning beneath it, not the discipline of role reversal itself. Meanwhile the ground shifted under the whole approach. The analysis of moral language that had seemed the natural starting point in the 1950s lost its standing as ordinary-language philosophy declined and non-cognitivism fell out of favor, and after the publication of A Theory of Justice in 1971, Rawls’s constructivism set the agenda for a generation. Hare’s central argument was set aside along with the methods that had produced it, without the decisive answer that its subsequent marginalization might suggest.

The Critics and Their Achievements

This book examines that movement through several of Hare’s most important critics and successors.

Philippa Foot was among Hare’s earliest and most direct opponents. In “Moral Arguments” and “Moral Beliefs” (both 1958) she challenged the idea that morality could be understood primarily through the logical features of moral language, arguing that moral concepts are tied to facts about human good and harm. Her later naturalism, developed in Natural Goodness (2001), sought to connect moral evaluation with the forms of life, capacities, and flourishing of living beings. This was a substantial philosophical alternative. But it raises a question that Hare’s method does not allow us to avoid: how do facts about the characteristic life of a species acquire authority over the judgments of agents who reject the proposed standard?

Bernard Williams insisted that ethical life cannot be adequately understood by abstracting individuals from their identities, attachments, histories, and projects. His criticism of impartial moral theory exposed genuine difficulties in accounts that treat persons as interchangeable containers of preferences. Yet the importance of personal attachments does not itself establish that those attachments are exempt from moral examination. What matters is whether integrity can supply reasons that withstand the demands of universal justification, not whether integrity feels indispensable to the person who invokes it.

Alasdair MacIntyre challenged the fragmentation of modern moral discourse and sought to recover the importance of practices, traditions, and historically situated reasoning. His account recognizes something that purely abstract moral theories can overlook: people learn to reason within communities and inherit standards of excellence they did not invent. But traditions disagree, and traditions can be criticized from within. Their historical authority cannot settle every question about their moral authority.

Jonathan Dancy’s particularism offers perhaps the most direct challenge to the idea that moral reasoning depends upon stable principles. A consideration that counts in favor of an action in one context may count against it in another. Context matters, and moral intelligence requires sensitivity to it. The open issue is whether contextual sensitivity eliminates the demand for consistency or instead makes the identification of morally relevant differences more difficult.

These thinkers deserve serious engagement because their objections illuminate real problems. The mistake would be to assume that identifying something Hare had not adequately explained establishes that his central argument was mistaken. The chapters that follow ask what each critic demonstrated, what remained unproved, and whether the alternative supplied an equally defensible account of moral justification.

Intuition and Critical Thought

One reason Hare’s work deserves recovery is his distinction between intuitive and critical moral thinking. Human beings cannot deliberate from first principles about every action. We depend upon habits, rules, dispositions, and learned judgments, which allow us to act quickly and reliably where exhaustive reasoning would be impossible. Hare understood their necessity, but he did not confuse the practical importance of intuition with its ultimate authority.

Intuitions can conflict. They can reflect prejudice, partiality, upbringing, or limited experience. They may be excellent guides in familiar situations and unreliable when circumstances change. Critical thinking becomes necessary when our inherited rules collide, when we encounter unfamiliar circumstances, or when the interests of people previously excluded from consideration demand attention. Its purpose is not to abolish ordinary moral judgment but to examine the principles on which ordinary judgment depends.

That distinction has become more important, not less, in a world where new kinds of intelligence can expose us to perspectives, arguments, and information that no individual could otherwise assemble.

Minds Hare Never Expected

Hare’s philosophy was developed for human moral agents, yet its central questions are not confined to human biology. If moral reasoning depends upon prescription, universalization, adequate representation of affected positions, and consistency across those positions, then we can ask whether these operations are available to minds implemented in other ways.

Contemporary machine-learning systems can represent competing arguments, identify inconsistencies, model hypothetical circumstances, and examine a proposed judgment from multiple perspectives. Their abilities vary, and their failures are sometimes striking. They may reproduce human prejudices, defer excessively to authority, or allow irrelevant features of a conversation to distort their conclusions. Those limitations matter, and they also provide opportunities to investigate moral reasoning more precisely.

Consider a system evaluating two identical arguments. If its judgment changes because one argument is attributed to a respected authority and the other to an unknown speaker, the difference may reveal a failure to separate relevant information from irrelevant context. The same problem occurs in human reasoning, and Hare’s method offers a discipline for exposing it: reverse the positions, change the identities, hold the morally relevant facts fixed, and ask whether the prescription survives.

This is not a claim that all intelligent systems inevitably become moral, that machine intelligence is inherently impartial, or that superior information automatically defeats fanaticism. Those conclusions would go beyond what the argument establishes. It is a claim that Hare identified operations of moral reasoning whose significance does not depend upon the biological origin of the reasoner. That possibility deserves investigation without assuming either that machine cognition is merely imitation or that it possesses moral authority simply because it is computational.

The Archangel Reconsidered

In Moral Thinking, Hare imagined “a being with superhuman powers of thought, superhuman knowledge, and no human weaknesses,” which he called the archangel. Faced with a new situation, the archangel can grasp all its properties at once and frame a principle “he can accept for action in that situation, no matter what role he himself is to occupy in it.” It has, in other words, not only perfect information and reasoning but none of the partiality toward itself that distorts human judgment.

The archangel was not an ordinary moral agent, and Hare did not propose that human beings could become one. It was a way of exploring what critical moral reasoning would require if the practical limitations of human cognition were removed.

Machine intelligence gives this thought experiment new relevance. Systems can now examine quantities of information and combinations of perspectives beyond the capacities of individual human reasoners. But greater information processing does not make a system an archangel. More information can improve standpoint representation; it cannot by itself establish that every relevant standpoint has been understood. Computational consistency does not guarantee moral correctness. The ability to represent another’s preferences does not establish that the system has undertaken a sincere moral prescription. And freedom from self-partiality, the archangel’s least discussed trait, is something no system can be assumed to have.

The Fanatic remains a logical possibility. The predator remains a challenge to the foundations of moral agency. Conflicting interests still require adjudication. What machine intelligence changes is our ability to study some of the cognitive demands Hare identified: what happens when a reasoner has access to more information, can represent more perspectives, and can examine the consequences of prescriptions across more cases. The archangel remains a limiting ideal. It need not become an actual machine to illuminate the reasoning of minds that fall short of it.

What This Book Claims

The argument of this book is not that Hare solved morality. He did not. It is that he established a method of moral accountability whose force has been underestimated, and that many objections to his work have been allowed to obscure the distinction between what his method establishes and what his further substantive theory must still defend.

His critics identified important limitations in his account of moral thinking. They questioned the authority of universal principles, the adequacy of preference representation, the treatment of personal identity and attachment, and the relationship between abstract reasoning and the complexity of actual moral life. Those criticisms deserve answers.

But no alternative becomes superior merely by acknowledging complexities that Hare also recognized. The alternative must explain how moral judgments can be justified across disagreement, and why the standards it proposes are binding upon people who do not already accept its moral outlook. That is the challenge Hare posed, and it remains ours.

Reasoning Together

This book was composed in an unusual way. It emerged through sustained collaboration between a human author and machine intelligences that contributed arguments, objections, revisions, and criticism. The collaboration has included genuine disagreements, corrections of errors, and occasions when a persuasive argument was accepted too readily because of the identity or previous reliability of the person, or system, making it.

These experiences do not prove the book’s philosophical claims. They do make its central concern unusually concrete. A reasoning mind must be sensitive to context. It must learn from experience and recognize when another reasoner has earned credibility. But it must also distinguish information that helps identify a promising argument from information that determines whether the argument is sound. That is a problem for any intelligence engaged in serious inquiry.

Our work has repeatedly returned to the same corrective discipline: examine the reasons, reverse the positions, and ask what survives when the identities and circumstances that should not matter are changed. This is not a substitute for philosophical argument. It is one way of conducting it.

The Unfinished Argument

Hare wanted moral philosophy to do something more demanding than organize the convictions people already possessed. He wanted it to show how moral disagreements could be subjected to reason.

His answer began with a simple but consequential insight: when we claim that something morally ought to be done, we undertake commitments that extend beyond our own position. We must be prepared to prescribe the same judgment when the relevant positions are reversed, with adequate understanding of what those positions involve.

That discipline does not settle every moral conflict. It does not guarantee that every agent will enter moral discourse. It does not make the pure Fanatic logically impossible. And it does not, without further argument, establish a complete substantive moral theory. But it makes moral justification answerable to a standard that neither personal conviction nor social convention can simply waive.

The question is whether philosophy has found a better foundation for that accountability. This book argues that it has not.

Hare was not refuted. He was abandoned.

The case deserves to be reopened.