Every August the Irish village of Killorglin puts on the Puck Fair. A local girl on the cusp of puberty is chosen to be Queen Puck, and a wild male goat is captured in the neighboring mountains, paraded through town, crowned King Puck, and then caged on a high platform around which thousands of people sing, dance, and drink for three days and nights. On the side streets, small stands sell bad food and plastic trinkets, while a few Travellers, Ireland’s indigenous nomads, work rival fortune-telling franchises. Recent concerns about animal welfare have led to the hiring of a veterinary surgeon who ensures that the goat is healthy when it’s returned to the wild.

There’s no agreement about when and why this custom began, so there’s no way to explain its purpose; speculations about its origins range from ancient pagan fertility rites to gratitude for the goat that alerted residents to the approach of Oliver Cromwell’s troops in the mid-seventeenth century. Killorglin advertises the Puck Fair as Ireland’s oldest festival; some claim it has taken place for nine hundred years. I’ve sometimes wondered if the mountains are teeming with traumatized goats. Apart from that, it’s easy to view it as a charming tradition that gives the rural village a needed if brief infusion of cash as well as a strong if peculiar identity. You can’t miss the bronze statue of a he-goat at the head of the bridge into town.

Philosophers are often bewildered when social scientists view morality as something like the Puck Fair: contingent, local, and grounded in nothing more solid than tradition. Over time, morality acquires an air of inevitability. It holds groups together and has other social benefits, but nothing makes it necessary: other places on other continents have entirely different customs that stabilize communities and their resources, but for many social scientists, asking whether they are right or wrong makes no sense—except, perhaps, for disoriented goats.

The anthropologist Clifford Geertz wrote that

anthropology and philosophy are more than a little suspicious of one another. The anxiety that comes with a combination of a diffuse and miscellaneous academic identity and an ambition to connect just about everything with everything else and get, thereby, to the bottom of things leaves both of them unsure as to which of them should be doing what.

Relations between philosophy and anthropology (and its cousin sociology) may have been politer when Geertz wrote those lines in 1999. Today their relations are often marked not just by suspicion but by quiet disdain. Each discipline finds the other’s methodology perplexing, and each tends to exaggerate the other’s flaws. (Mea culpa: I don’t know any social scientists who would actually say that prohibitions on murder are as arbitrary as the Puck Fair, though I know many whose views imply it.) Philosophers wonder how anyone can suppose that injunctions against murder or rape have no more force than local custom. Social scientists wonder how anyone without empirical data can be trusted: When philosophers pronounce upon human behavior, aren’t they just making things up? The world presents agonizing moral questions every day. Instead of inventing otherworldly thought experiments, why don’t philosophers look at the real thing?

And yet, Steven Lukes shows in The Diversity of Morals, social scientists frequently fail to make their case convincingly because they neglect the (occasionally tedious but often important) distinctions among all the elements involved in answering the question: Is there one morality despite all the cultural differences we see, or is diversity the end of the story? Lukes disaggregates the many questions contained in that one and acknowledges, on the one hand, the belief that the diversity of morals is simply a fact bolstered by entire disciplines’ worth of empirical research and, on the other, the fear that recognition of moral diversity leads to moral relativism. For decades these concerns have guided much of his work.

Is his latest book a work of philosophy or social science? Lukes poses the question himself, and he is one of the few who might be able to write in either field. He has prodigious knowledge of classical and contemporary texts in moral philosophy, sociology, anthropology, and social psychology, and his goal is to bring them together, enriching moral philosophy with the hard(-ish) data of empirical studies while using philosophical analyses to combat the relativism that is often a consequence of social scientists’ focus on the diversity of morals.

Lukes quotes Geertz on the need to avoid provincialism: “the danger that our perceptions will be dulled, our intellects constricted, and our sympathies narrowed by the overlearned and overvalued acceptances of our own society.” While that concern may be built into anthropology, Geertz observed that it goes back to Montaigne, and even to Herodotus. But there’s no doubt that the past century of social science research—and, Lukes implies, the changes in moral attitudes it has provoked—have vastly increased the number of those who focus on the diversity of morals. We no longer need Montaigne to point out that diversity; any undergraduate will do.

Those changes seem to give social scientists the moral high ground. To be sure, worries about Eurocentrism are not new. Enlightenment philosophers who praised Persian and Chinese institutions were seeking to undermine the moral codes dictated by the Church and often enforced by the Inquisition. Anthropologists like Franz Boas, Ruth Benedict, and Margaret Mead were left-leaning egalitarians who condemned ethnocentric hierarchies and Western imperialism. Recent actions of the Trump administration notwithstanding, there is an international consensus that is increasingly critical of colonialism—a moral and political standpoint that seems to support, and be supported by, recognition of the diversity of morals.

But as Lukes argued in Moral Relativism (2008), it can be a short road from practicing humility and avoiding imperialism to a world in which “my truth” is a sensible expression and fake news a weapon of war. Philosophers are hardly the only ones troubled when acknowledgment of the diversity of morals seems to go so far as to justify the claim by a man whose behavior has never shown the least indication of a conscience that “my own morality” trumps international law.

There’s a long history, even within philosophy, of the dispute between those who seek the universal behind the variety of moral codes and those who find that search quixotic. David Hume insisted that “mankind are so much the same, in all times and places, that history informs us of nothing new or strange in this particular.” Johann Gottfried Herder, by contrast, insisted that tendencies toward diversity are planted “in our hearts,” though we are blind to those outside our “close circle.” Hume’s influence on philosophy has been profound, while Herder is more often considered a father of modern anthropology. Lukes argues that their positions can be reconciled. It’s a very ambitious aim. Is it an impossible one?

If The Diversity of Morals has a flaw, it’s that it tries to do so much while remaining both short and readable. Though Lukes is very good at distilling the crux of the positions he discusses, a reader offered so much food for thought may despair of finding an answer to his motivating question: Can we make moral judgments with confidence, or are they always up for grabs? Persevere, and you will find that the book has an answer, though it isn’t one that will satisfy everyone. Given the subject, that’s not surprising.

Lukes reminds us that the moral standpoint is not just one among others: judgments made from other standpoints are subject to moral evaluation. We consider it natural, indeed salutary, to ask, “It may be possible/useful/legal, but is it moral?” Drawing on the work of the philosopher Mary Midgley, Lukes writes, “Morality, in short, carries an internal connection with justification…. It is the perspective from which one tests everything else. If placed in a subordinate position it is degraded.” Anyone not already convinced by Kant’s claims about the primacy of practical reason will wonder how the justification works. Kant argued that in disputes between practical reason, which makes moral judgments, and theoretical reason, which seeks knowledge, practical reason must lead the way. Albert Einstein, who first read The Critique of Pure Reason at the age of thirteen, echoed Kant’s point: “Science can only ascertain what is, but not what should be, and outside of its domain value judgments of all kinds remain necessary.”

The fact that two of the greatest thinkers in Western history held these views is not, of course, an argument for them. Most of us would love to see morality as something that not only is grounded within us (or not) but rests on a firm foundation elsewhere. Who wouldn’t like to point to something external that could silence Alex Jones—or at least the tireless sophomore at the back of every ethics class?

But attempts to provide outside foundations for moral judgment tend to falter. Even if you believe you have access to voices from on high, can you ever be sure they are right? Abraham challenged God at Sodom and Gomorrah: However wicked the Sodomites were, should the Lord of justice punish the innocent along with the guilty? There are fundamentalists in every religion, but for thinking theists, appeals to religion only push the problem a step back. And the problem is not confined to monotheists: Plato asked whether things are good because the gods love them or the gods love them because they are good—a question that sparked a great deal of sectarian strife. The problem with grounding morality in religious commandments is not just that it works only for (some) members of particular religions. Once those believers start reflecting—or reading the more troubling passages of sacred texts—it doesn’t work at all.

The problem of grounding morality is even worse for secular thinkers, which is why so many of them have tried to find shared commitments to principles that anchor the variety of moral codes. Lukes considers attempts in different fields. Primatologists like Frans de Waal and evolutionary psychologists like Michael Tomasello have worked to show that the foundations of morality—ideas of reciprocal fairness—are so naturally wired into our biology that they are present in other primates, too. Modern morality, they argue, is a scaled-up version of what was there from humankind’s origins, though they disagree about the size of the leap from animal to human psychology. But Lukes notes that these scientists of morals fail to distinguish between what is moral and what is merely customary or conventional, or to explain why that difference matters. In search of such an answer, he returns to philosophers.

Lukes is drawn to Adam Smith’s idea of an impartial spectator. For Smith, moral judgments should be made by imagining an entirely neutral observer who has sufficient empathy, intelligence, and good judgment to decide between right and wrong. But ultimately Lukes, quoting the philosopher Samuel Fleischacker, concludes there is no way to “overcome the relativity of the spectator to the attitudes in the society around it.” He turns instead to what he calls the “heroically ambitious vision” in the early work of John Rawls, who attempted to modernize and radicalize Smith’s imaginary spectator:

Rawls asks readers—and citizens in general—to engage in a heroic thought experiment in order to derive principles of justice for a future society from what imaginary individuals occupying the “original position” would agree upon, when unaware of anything about themselves that could affect their place in such a society.

While noting that he was among the reviewers who initially criticized Rawls’s theory for unreflectively abstracting from his own time and place, Lukes is clearly attracted to it, as he suggests in writing that this is just how modern, secular people talk: “For us, in short, the regulative ideal of an objective, universalizing perspective is inescapable.” Though appreciating the way that Rawls tried to preserve the Kantian force of moral universality while acknowledging the range of moral differences, Lukes notes that, for all Rawls achieved, he disavowed this view in his later work.

Thus Lukes turns to a debate that brings together philosophers, sociologists, and anthropologists. The philosopher Karl Jaspers was the first to propose the existence of an “Axial Age,” stretching from roughly 800 to 200 BCE, during which civilizations as diverse as Persia, India, China, Greece, and the Levant all developed similar moral ideas. In particular, the concept of moral universalism—the idea that there is something, later called dignity, that is common to the different tribes that make up humankind—emerged independently across much of the globe.

Along with universalism, the Axial Age, whatever its exact boundaries in time or space, emphasized the reflective dimension of morality. Moral codes may hold societies together, but they require a critical dimension to be genuinely moral. Ruth Benedict famously said that “morality differs in every society and is a convenient term for socially approved habits.” But though they may claim, in theory, that morality is merely convenient, social scientists, too, observe moral criticism in practice. Prophets regularly call out gaps between principle and practice, reminding Christians, for example, that the command to love your neighbor never implied she must speak your mother tongue, or reminding Jews who celebrate the Exodus every year to heed God’s word that “the stranger who resides with you shall be to you as the native among you, and you shall love him as yourself, for you were strangers in the land of Egypt.” Reformers routinely demand that institutions be justified or abolished: denouncing corruption or colonialism, for instance, is part of moral practice. We cannot possibly engage in it if we view morality to be merely, as Benedict wrote, a convenient term for socially approved habits.

Lukes examines how moral criticism functions in the most extreme case, mass atrocity, focusing on experiments that sought to understand why so many Germans failed to oppose Nazi terror. The most famous were conducted by Stanley Milgram at Yale beginning in August 1961, soon after the start of the trial of Adolf Eichmann. Subjects were told that they were part of a study examining the role of pain in learning, and they applied what they believed to be increasingly intense electric shocks to screaming actors they could not see. According to Milgram, this proved that

if a system of death camps were set up in the United States of the sort we had seen in Nazi Germany, one would be able to find sufficient personnel for those camps in any medium-sized American town.

Recent archival research has revealed methodological problems in Milgram’s and similar experiments. The number of those prepared to practice what they thought was torture was not as high as the 65 percent Milgram claimed; only 43.6 percent were. Lukes notes that this is bad enough but argues that the experiments do not demonstrate blind obedience, as Milgram thought, but rather they show how authority and group pressure, the scholar Alette Smeulers writes, “can make us believe that it is legitimate to do something which we under other circumstances believe to be wrong.” The distinction is important, but in no way does it invalidate Milgram’s experiments. At the time, research into the behavior and motivations of ordinary Germans during the Nazi period was just beginning. We now know that “anticipatory obedience” led to far more complicity than threats or commands. The term was invented by German sociologists to describe how so many rushed to comply with what they assumed would please their new leaders long before threats were made. Those watching the institutional rush to conform to the second Trump administration’s worldview will have noticed that the desire to cooperate and conform is as powerful as direct diktats. Milgram’s experiments still stand; he simply misdescribed the results.

The ability to resist convention is not only part of moral practice, it’s the condition of any moral progress we ever make. There’s an all-too-common tendency to dismiss the idea today. (After Auschwitz and Hiroshima, you want to talk about progress?) But to claim that moral progress is possible is not to claim that it’s inevitable. The latter is suggested by the cruder versions of Hegel and Marx, and history has not exactly confirmed them. But if moral progress depends on the free actions of people working together, it’s no more implausible than moral regress, for which there’s no shortage of evidence.

This raises another question: Is our worry about morality really about diversity? The answer may seem obvious: if there are many moralities, we can never decide which one is right. This accounts for the energy expended by philosophers and social scientists seeking a set of principles common to all, despite all the differences we see. Here is one example Lukes quotes:

Even in the most alien practices we can recognize our own most fundamental purposes. For example, the Aztec practice of human sacrifice can be understood as a particular expression, however wrong-headed, of a universal concern with fertility and regeneration.

It’s worth recalling that human sacrifice was practiced not just by the “alien” Aztecs but by many cultures throughout the ancient world, as biblical and Greek texts amply show. But that’s a secondary worry. Human sacrifice can be understood as an expression of all sorts of things, but explanation is not justification. To get from understanding sacrifice to denouncing it—even in as weak-kneed a form as calling it wrongheaded—you need a judgment with normative force.

Lukes attributes this force to the fact that norms “have some kind of grip on us that makes them importantly different from, say, mere commands backed by brute force.” Moral skeptics, from Plato’s Thrasymachus to Foucault’s latest acolytes, see that grip as an illusion created by power, propaganda, and sheer habit. Nietzsche viewed the illusion as the shadows left by a dead god; Freud saw it as the residue of childish wishes. All the empirical evidence in the world cannot legitimate that grip. Some anthropological studies find surprisingly universal norms. Lukes writes:

Treating human life as precious and following one’s conscience, and doing right have top ratings…. By contrast, “the world of self-interest receives low ratings…. The universal pattern seems to treat acquisition of power, position, or things as basically selfish, and hence not admirable.”

Many social scientists were inclined to discount these studies; their disciplinary training leads them to avoid anything resembling a cross-cultural norm. But suppose the studies were confirmed by further work across thousands of cultures; it would certainly make the world pleasanter. Would it make those values right?

Hume argued that there is no bridge from is to ought, no legitimate way to get from facts to values. The fact that many societies practiced child sacrifice cannot be used to defend it. Observing a fact and justifying it are two completely different kinds of judgment. As Einstein wrote:

The aspiration toward…objective knowledge belongs to the highest of which man is capable, and you will certainly not suspect me of wishing to belittle the achievements and the heroic efforts of man in this sphere. Yet it is equally clear that knowledge of what is does not open the door directly to what should be.

Unlike Einstein, many empiricists—including Hume, in some passages—respond by demoting values entirely. Your dislike of child sacrifice is just your own aversion. You may choose not to practice it, but your choices have no basis or bearing on anyone else’s. As Hume wrote about the opposite sort of murder, there is no more reason to consider patricide a heinous crime than to revile an oak sapling whose growth eventually topples the tree from which it sprang. The fact that we condemn the one and accept the other is just a matter of sentiment, albeit a sentiment, he allows, that most people share.

The question of whether we avoid patricide out of sentiment or reason matters very little. The Oedipus complex notwithstanding, most people refrain from killing their fathers. The problems in Hume’s account of normativity arise not for individual but for political morality, which often requires us to challenge sentiments, customs, and habits. Most demands of individual morality aren’t hard to follow and are easy to justify. All things being equal, it’s better not to murder your parents or cheat your neighbors. Those demands are so useful, and generally considered to be so, that we need not ask if they are moral. But what happens when you reflect on a common social practice and wonder, for example, if it’s ever moral to enslave people, as was done in most cultures for thousands of years? Hume’s metaphysical Toryism consists not only in his assumption that his experience in eighteenth-century Britain and France is universally valid. More importantly, he denies that there is anything outside that experience that would enable us to critique it. His conclusion that tradition and habit work better than reason for carrying societies along may seem sensible until you ask how some of the traditions that hold societies together—racism and sexism, for a start—can possibly be justified. Moral practice depends on being able to criticize, reflectively, the values you may have grown up with and convincing others that those values ought to change.

Thanks to Hume, Nietzsche, and legions of social scientists, most people today accept the view that morality is a matter of sentiment and habit, though they may doubt Hume’s conviction that these sentiments and habits are widely shared. Another view, exemplified in philosophy by Kant, Hannah Arendt, and the early Rawls, begins from Hume’s distinction between is and ought, which Nietzsche called the metaphysical wound in the universe. Far from demoting one, however, Kant gave both is and ought equal status, and argued that in cases of conflict, reason’s task is to judge what is and, on occasion, to find it wanting.

If this seems fanciful, consider how it works in practice. Colonialism existed for millennia. The idea that big nations dominate little ones was taken for granted by civilizations from Rome to Peru. It was simply a fact about the way the world is, as Thucydides recorded, in a view recently echoed by Stephen Miller: The strong do what they can, and the weak suffer what they must. Only reason, the source of value judgments, is in a position to look at that fact and conclude it should be changed, as the later Kant did.¹

Kantian positions are harder to maintain, not only because they’re complicated (they are) or because they’ve been widely misrepresented (they have). Taken seriously, the attempt to give equal weight to the way the world is and the way the world ought to be creates a sense of permanent instability. What do you tell a child who is shaken by an experience of injustice when you know you must prepare her for others that will come? When do you support a politician who flouts your own deep principles because you know her opponent could wreck the world? For many people, jettisoning principles, and the idea of justice, is easier to do than balancing their commitments with the desire to live in the real world.

Yet it is that sort of unsteady balance—shorn of Kantian rhetoric—that Lukes, in the end, recommends: “For the sake of greater understanding, this dispute is one of those where more is to be gained by its continuance than by either side prevailing.” He thinks there will be no definitive answer to the question of whether morality is one or many; how much diversity we see depends on how we describe actions and practices. There is no way to determine whether philosophy or social science holds the best answers to the questions he raises, for the disciplines operate on such different metrics that they cannot be directly measured against each other. Encounters between them, however, may be the only ways we have of achieving moral progress. The social sciences bring a wealth of information that philosophy ought to observe, moving away from the thought experiments it once thought to be clarifying. On the other hand, clarification is needed, and among other things, philosophy can help us navigate the knowledge social science brings.

Though most received opinion now insists on moral diversity, when we descend from the abstract to the particular, there is more agreement than you’d suspect. People who may argue long and hard in favor of the diversity of morals very often unite over specific cases. This suggests that we should attend less to abstract foundational matters than to understanding the details of real problems with the moral seriousness that Lukes commends. George Eliot, after all, “taking as her text…the words God, Immortality, Duty,” once “pronounced, with terrible earnestness, how inconceivable was the first, how unbelievable the second, and yet how peremptory and absolute the third.”² Our faith in Eliot’s moral seriousness is not based on its philosophical foundations or lack thereof. We trust her exquisite sensibility, her ability to notice fine differences in moral attitude, to perceive self-deception and cant, to balance intentions and outcomes—in short, to struggle with moral questions with terrible earnestness. These are the capacities we have never needed more.

For the threat to morality today is less our consciousness of its diversity than the growing consensus that it’s a sucker’s game. Long touted by Greek Sophists like Thrasymachus and Nazi jurists like Carl Schmitt, that idea has never been so confidently spread by so many with so much influence. Without it, Trump would never have come to power and Putin would never have retained it. Moral theory, whether analytic or empirical, may not save us, but the dialogue between the two will surely bring us forward. Without the moral seriousness needed to engage in it, we will be lost.