In the garden scene from Goethe’s Faust, the pious Margarete asks her suitor to give her some assurance that he believes in God. Faust offers the sly retort that no words can suffice to convey his faith. “Call it then what you will,” he explains. “Call it Happiness! Heart! Love! God! I have no names for it! Feeling is all; to name is but noise and smoke, beclouding heaven’s glow.” Although his answer may seem evasive or a mere ploy in his efforts to seduce Margarete, Faust expresses what has become a fairly widespread opinion about religion. Today many people, at least in the educated and urbanized parts of the Global North, report that traditional forms of institutional religion no longer provide a suitable language for their inchoate stirrings of belief. Among citizens of the European Union rates of religious identification vary widely. A Eurobarometer poll from 2021 shows that in Cyprus and Slovakia nearly 60 percent of people consider their religious identity very important; in France it’s 40 percent; in Sweden fewer than 20 percent. In the US, despite the enduring appeal of evangelical Christianity, data from the Pew Research Center indicate that nearly 30 percent of the population is religiously unaffiliated. If pressed, they might give the sort of answer that Faust gave to Margarete: “Spiritual but not religious.”
In Captive Gods, the philosopher Kwame Anthony Appiah asks us to think more carefully about what we mean when we call something a religion. As many sociologists and historians of religion will confirm, the word “religion” has become a point of controversy in recent years because it serves as a name for so many things. It can seem hard to define because it is so expansive in its meaning. Of more traditional cultures anthropologists have often said that religiosity touches upon every dimension of life, from personal experiences of mystical transport to public practices that govern rituals of food and sex, birth and death. Even when speaking of modern cultures, some social scientists insist that there is at least a religious quality to much that we do or believe. Isn’t nationalism a species of religion? If it is not, why all the rituals that surround a nation’s flag, and why do we feel it should not touch the ground? And do we not act as if art, too, were a form of religion? If not, why do operagoers conduct themselves with such solemnity, to the point of scowling in disapproval at anyone who coughs?
It is tempting to agree with sociologists who say that the term should be applied as broadly as possible, not just to the credulous humans of prehistory but to their incredulous descendants as well. If “religion” is equipped with such a capacious meaning, however, we may wonder if there is anything that is not religious. In the introduction to Captive Gods Appiah illustrates this definitional problem with a joke. “How’s the water?” the old fish asks the young fish. The young fish seem confused. “What water?”
To explore this problem, Appiah seeks to show how, since its inception in the nineteenth century, the modern discipline of sociology has found itself curiously entangled with religion. “The idea of social science,” he writes, “arose in dialogue with the idea of religion, explaining itself in the effort to explain religion.” For religion could only become an object of social-scientific inquiry when it no longer served as the all-inclusive but invisible medium that surrounded us. As it receded, sociology emerged as if from beneath the waves, and its exponents began to develop theories about religion, arguing among themselves about whether it is best defined as an institution or a practice, an inner experience or a codified doctrine.
Some readers may find this claim obvious—after all, every science needs to specify its object of inquiry. But Appiah’s argument comes with a surprising twist: the most prominent founders of sociology (Émile Durkheim, Georg Simmel, and Max Weber, to name just three) were also convinced that when we look at religion we are looking at the animating principles of our own social world. Society is a precondition for religion, but religion is a condition for society. It follows that sociology, even as it sought to define religion, was also trying to define itself. Appiah describes this circle of causation as an ouroboros, the mythical snake that eats its own tail.
Among the writers who might serve as a guide through this difficult terrain, Appiah stands out for both his lucidity in exposition and his sure-footed style. He has ecumenical interests and has published books on a wide range of topics, from cosmopolitanism to African identity to the uses of counterfactual idealization (inspired by the work of the German philosopher Hans Vaihinger). He is also distinctively well placed to understand just how hard it can be to pin down what a religion consists in. The son of a British mother and a father from Ghana, Appiah notes in the book’s introduction that he has spent much of his life shuttling between various religious traditions:
I have feasted at Eid al-Fitr with my Muslim cousins, celebrated Seders at home with my in-laws, recited a Sanskrit mantra as I meditated alone, and attended a nuptial Mass conducted by a cardinal.
In his childhood he sang in an Anglican choir, while back home in Ghana he attended Sunday school at an interdenominational church. Mixed together with this Christian inheritance were religious practices from Kumasi, in the Ashanti region of Ghana, where he learned to follow his father’s example by “pouring libations to our ancestors, who were once as real to me as the God whose presence I felt when I prayed.” This complex of traditions and identities gives Appiah a singular epistemological advantage, since he is at once insider and outsider to multiple faiths, no one of which, it would seem, has an exclusive claim to his allegiance. More poignant still, he notes, the very idea of “Asante religion” is a post-facto construct: the language of nineteenth-century Asante-Twi has no word for what we call “religion.”
Captive Gods is a more or less conventional intellectual history: it charts the emergence of social-scientific inquiry about religion by focusing on just a handful of the most consequential scholars. Appiah writes with a light touch, braiding together facts of biography with philosophical analysis that seldom belabors its claims. It’s a style, however, that can make even the most powerful insights seem like casual aperçus, and the inattentive reader may at times miss moments of criticism that he has woven into his tale. This would be unfortunate, I think, since his book pushes back, subtly but forcefully, against some of the pieties that have crept into the social-scientific disciplines. Here, too, as a philosopher he has the advantage of being an outsider who is more or less immune to the disciplinary shibboleths that hold sway among many scholars of religion.
Among those shibboleths is the claim that we should not speak of “religion” at all, since the term fuses together a cluster of cultural practices that have little in common. This argument first came to prominence in The Meaning and End of Religion (1962) by the Islamicist Wilfred Cantwell Smith, who wrote that “neither religion in general nor any one of the religions…is in itself an intelligible entity, a valid object of inquiry or of concern either for the scholar or for the man of faith.” Smith was especially alive to the Eurocentric implications of the term “religion” and wished to free scholars from its distorting effects. More recently, the cultural anthropologist Talal Asad has amplified this concern, arguing that the idea of “secularism” is not as innocent as it may seem, since it presupposes an implicitly Christian (more specifically, Protestant) notion of private religiosity that has gained political currency in the liberal West. Tomoko Masuzawa developed a similar argument in The Invention of World Religions (2005), which purports to show “how European universalism was preserved in the language of pluralism.”
Such arguments can serve as a welcome check on hidden bias. The difficulty, however, is that even while they dismantle false universals, they can also erect new barricades in our thinking. It’s instructive that Smith wished to banish the word “religion” but in the same breath referred to the “man of faith.” After all, “faith” is an umbrella term no less than “religion” is, and it is no less vulnerable to criticism. To finesse this problem, it may prove best to admit that when it comes to cultural analysis, every universal will inevitably pick out some features of the human world while ignoring others. Appiah suggests, with reference to Wittgenstein, that religions may relate to one another by “family resemblance.” Each member may share only some features with others, while no two members share all features in common.
But there are other concerns besides matters of logic. The nominalist who rejects “religion” as a false universal may be right, but we had better hope that such skepticism serves a further end. Asserting cultural difference is salutary, but it should not strand us on separate islands of meaning. The genealogical work of exposing what may be distortive or imperialistic in our language can become an arid exercise if it does not assist us in opening up better avenues of cross-cultural understanding.
Scholarly efforts to understand what religion is—and what all religions may share—go back to the origins of the modern social sciences. Appiah begins with an intellectual portrait of Sir Edward Burnett Tylor, a Victorian-era professor who held Oxford’s first chair in anthropology. Born into a Quaker family, Tylor was diagnosed with tuberculosis in his youth and was sent away from England for his health. He traveled across Mesoamerica, often on horseback, an experience that left him with an enduring appreciation for other cultures but also gnawed away at his faith, leaving him better positioned to draw comparisons across religions without exalting his Christian heritage as uniquely true. To be sure, he ranked some cultures higher than others on a developmental scale, as was commonplace at the time. Nevertheless, his most influential work, Primitive Culture (first published in 1871), laid what many would consider the foundations of modern anthropology.
In that book Tylor defined animism as “the belief in Spiritual Beings.” Although this may strike us as fairly uncontroversial, it raises a host of questions. First, what counts as a spiritual being? Tylor seemed to recognize that the reference to spirits threatened to rule out a great many cultural beliefs. The Tongans “imagined the human soul to be the finer or more aeriform part of the body, which leaves it suddenly at the moment of death.” Superficially this may sound like the Christian idea of the soul, but Tylor specified that the Tongan soul is “comparable to the perfume and essence of a flower as related to the more solid vegetable fibre.” He wished to set the parameters so wide that they could span the entire range of world belief, accommodating both animism and monotheism, whatever their apparent differences. But his definition did not prove wide enough. Buddhists (or at least some of them, since generalizations are hazardous) do not believe in what some would call a human soul, and this would imply that in Tylor’s view Buddhism might not count as a religion. That was not the only hard case. As Appiah notes, similar problems arise when we examine Shintoism or Jainism.
Then there is the question of what we mean by “belief.” Early on, some critics challenged Tylor’s definition. William Robertson Smith, a Presbyterian minister in the Scottish Free Church, criticized Tylor for emphasizing belief while ignoring the importance of practice. When we characterize all religion as belief in spiritual beings, we are committing the error of provincialism, projecting upon other religions the specific experiences of the Christian church, which has been riven for millennia by quarrels over dogma and doctrine. In his Lectures on the Religion of the Semites (1889) Smith suggested that the definition of religion as “creed” was ill-suited to peoples whose religion consists not in beliefs but in collective rituals:
Belief in a certain series of myths was neither obligatory as a part of true religion, nor was it supposed that, by believing, a man acquired religious merit and conciliated the favour of the gods. What was obligatory or meritorious was the exact performance of certain sacred acts prescribed by religious tradition.
A further concern was that by emphasizing belief, Tylor placed far too much weight on the individual’s inner life. Rituals, on the other hand, are not interior states of consciousness; they are public and performed by a community.
A fundamental feature of this definition is that it is social in character. Later generations of anthropologists were hardly inclined to speak favorably of Tylor’s work, and they continued to lament the bias of “intellectualism” in which belief outranked both emotion and cultural practice. In the 1950s, Appiah notes, “neo-Tylorian” was often a term of abuse. The irony, however, is that for all his intellectualism Tylor also fortified a universalist understanding of other cultures. Even an accomplished anthropologist like E.E. Evans-Pritchard, who dismissed Tylor’s developmental rankings, saw that he was right to recognize “the essential rationality of primitive peoples.” Appiah admits that in contemporary anthropology, universalism is often regarded with suspicion, since the discipline is “deeply penitential” about its origins in Western claims to superior knowledge. Meanwhile Tylor eventually came to share the basic view that the individual, taken as a single datum of analysis, is too inclined to self-interest and that religious customs must have emerged “for the benefit of society.” This social understanding of religion served as an important stepping-stone on the path toward more favorable views of so-called primitive cultures.
For Appiah the turn to society—which is defined as the essential variable for explaining the origins and structure of religion—marks the true beginning of sociology as a modern discipline. He explores this sociological turn in chapters on the three figures who are widely seen as its founders: Durkheim, Simmel, and Weber. Those who are well versed in the history of sociology may find much in this story familiar, though it is told with such clarity and grace that it will reward both specialists and nonspecialists.
Perhaps more than anyone it is Durkheim who deserves the greatest credit for pursuing the social theory of religion about as far as seems possible. Religion, for Durkheim, is simply society’s representation of itself. This is a puzzling claim, chiefly because it is all-inclusive. Every society arguably has at least some minimal awareness of its identity, its values, and the principles that allow it to cohere. But it follows for Durkheim that every society must have something that qualifies as its religion.
This is precisely what he argued in his most ambitious work, The Elementary Forms of the Religious Life (1912). Drawing especially on ethnographic evidence about Australian Aboriginal cultures and the totemic schemes that organize their clans, Durkheim concluded that totemism is not only one instance of religion among many. It is the very paradigm of a religious system. This is because it provides Aboriginal society with an ordering schema that is taken to be obligatory. To worship a totem is to show one’s allegiance to the social rules one considers sacred. Since all societies have social rules, it follows that all societies have something like a religion that represents their basic form. This prompted Durkheim’s well-known statement that it is “an essential postulate of sociology that a human institution cannot be based on error and lies. If it did, it could not survive…. There are, therefore, in essence, no religions that are false.”
One advantage of this claim is that it permits us to understand why even secular societies seem to exhibit a quasi-religious attitude when they celebrate their moral schemes. Nationalism, for Durkheim, is essentially religious. To pay fealty to a flag is to honor a sacred symbol that, much like the totem of an animal or plant, provides the collective with an image of what it most cherishes. This is why ostensibly nonreligious societies still find it meaningful to participate in collective rituals—such as the Fourth of July or Bastille Day—that celebrate the founding and the cohesion of the social order. Durkheim believed that such festivals were no different in essence from the Australian corroboree, the rite of “collective effervescence” by which Aboriginal clans celebrated their membership in the group.
More surprising, however, is Durkheim’s suggestion that even the political ideal of individualism can function as a collective representation. This idea may strike us as paradoxical: individualism would seem to imply a belief that individuals have rights and interests that should be ranked higher than those of the society of which they are members. Durkheim’s point, however, was that even this belief is a social one, since it is a norm that the collective must share. If it were not a social value, then there would be nothing to guarantee the individual’s claims. Durkheim first presented this idea in “Individualism and the Intellectuals,” an essay written in 1898 during the height of the Dreyfus Affair. As a secular Jew (he was descended from a family of rabbis) and an ardent believer in the ideals of the French Republic, Durkheim saw that a modern society need not share a traditional religion or any form of exclusionary nationalism. Its commitment to the rights of the individual can serve as a shared system of belief.
But the advantage of this idea also turned out to be its chief disadvantage. Durkheim’s definition of religion is so capacious that it seems to permit any social scheme to qualify as a religion. Many of us, whether we are religious or not, might object that this is far too permissive. Not all social representations enjoy the sacred status that we confer on those we conventionally call religious. The philosopher Charles Taylor, for example, argued in A Secular Age (2007) that religious experiences have the unique character of opening us up to “transcendence.” Those who are not open to such experiences must content themselves with living within what he calls the “immanent frame.” Taylor allows that for those who remain confined to immanence, certain experiences are available that seem to exhibit a quasi-sacred character. These include what he calls “neo-Durkheimian” experiences or “substitutes” for transcendence. (In one of the more questionable passages in his book, he mentions listening to Beethoven string quartets among those experiences that provide such a substitute.) As several critics have noted, one problem with Taylor’s distinction between immanence and transcendence is that it seems reverse engineered for the book’s conclusions, upholding a specific model of religion as the generic standard.
Still, Taylor may be right to insist that we should impose some constraints on our definition of religion so that we do not all subscribe to the Durkheimian faith. The deepest irony about Durkheim’s definition, however, is that no religious believer would ever assent to it. No one who worships a divine being thinks that they are merely worshiping a social representation. Durkheim’s definition secularizes religion but then turns the sociologist into a secular priest, an expert who can inform us what our religion actually means, even if we do not think it means that at all. Evans-Pritchard memorably quipped, “It was Durkheim and not the savage who made society into a god.”
Appiah’s portraits of Simmel and Weber are no less illuminating, and both contribute to his general thesis that the study of society and the study of religion arose in concert. In his Introduction to the Science of Morality, a two-volume work published in 1892 and 1893, Simmel wrote of “a profound analogy between attitudes toward society and attitudes toward God.” Just as the social collective looms as a higher force over the individual, so, too, God hovers as a higher force over the believer. Durkheim, of course, had made a similar observation. But this was the kind of sociological claim that, from the perspective of the religious believer, seemed to evade the most urgent question. For if society is analogous to God, should we infer that our belief in God is merely a disguised belief in society? And if sociology can explain religion, can it also explain it away? Such questions help us to understand why sociological inquiry into the origins or social ends of religion often appear as menacing to genuine faith.
Of the many sociologists who could have provoked this concern, Weber remains the most intriguing. Although born into a Protestant household, he admitted that he was “religiously absolutely unmusical.” The experiences that seemed to stir the hearts of some of his more credulous contemporaries left Weber, a sober rationalist in disposition, largely unmoved. Yet he was fascinated by religion, and he saw in it an indispensable source of morality and meaning—the German word is Sinn—without which he feared that modern society would harden into something lifeless and cold.
He explored this idea in many works, most famously in The Protestant Ethic and the “Spirit” of Capitalism (1904–1905), in which he argued that a disposition peculiar to Calvinism—he called it “this-worldly asceticism”—had spawned an attitude of methodical restraint that encouraged the growth of capitalism, especially in Protestant Europe and the colonies of North America. Once capitalism had advanced to a sufficient degree, however, its animating spirit dwindled away, leaving behind a world of meaningless compulsion.
Weber was also interested in non-Christian religions, and after The Protestant Ethic he plunged into writing an ambitious, multivolume comparative study of world religions, including Confucianism, Taoism, Buddhism, Hinduism, and Judaism. But what he called “Occidental rationalism” remained his consuming interest, and he assigned it a primary status in creating the bureaucratic-legal forms of social and economic organization that now prevail throughout the modern West. Our age, he declared, is characterized by rationalization and “the disenchantment of the world.”
Toward those who were still religious believers, Weber adopted a condescending tone. “To anyone who is unable to endure the fate of the age like a man,” he declared, “we must say that he should return to the welcoming and merciful embrace of the old churches.” Any religious institution would compel its members to make a “sacrifice of the intellect,” but in Weber’s view that was simply the price one had to pay for religious belief. Today few sociologists would issue such bold pronouncements, and the classical thesis of secularization that Weber helped to promote has relatively few adherents. In The Power of the Sacred (2017), the German sociologist Hans Joas warned that the idea of world disenchantment is too grand in its sweep: it belongs among those “dangerous nouns of process” that blinded Weber to the enduring vitality of religion.
Appiah sounds a somewhat different note. Weber was mistaken to forecast a time when the bonds of religion would dissolve. Today, Appiah writes, we hardly suffer from a scarcity of religion; it would be more accurate to say that we confront a surplus. But while religion can hold us together, it can also tear us apart.
English (US) · 