Why religious thinking began

Pillars before farms

On a ridge in southeastern Turkey, near the city of Şanlıurfa, people raised rings of stone pillars more than eleven thousand years ago. The site is called Göbekli Tepe. Its builders were hunter-gatherers: it belongs to the Pre-Pottery Neolithic, with occupation dated to roughly 9500–8000 BCE, before pottery, before metal, before writing, and, as far as the evidence shows, before farming took hold in the region. Archaeologists from the German Archaeological Institute have dug there since 1995, first under Klaus Schmidt, who led the work until his death in 2014. The pillars are cut from limestone in the shape of a capital T. The two tallest, standing in the middle of the largest ring excavated so far (Enclosure D), are about five and a half meters high and are estimated at eight to ten metric tons each. Many are carved with animals: foxes, boars, snakes, and vultures among them. UNESCO made the site a World Heritage Site in 2018.

What were they for? Nobody knows. There is no writing, and what the builders said and believed in those rings is lost. Schmidt read the site as a gathering place for ritual and, in popular accounts, it became “the world’s first temple.” Other archaeologists point to evidence of ordinary life on the hill and read it as a place where people lived, feasted, and met, with ritual as one part of it. “Probably a place of gathering and ceremony, whose exact meaning we cannot recover” is as far as the evidence goes. But the site puts the question of this lesson in its sharpest form. People who had no cities, no kings, and no farms put years of labor into stone monuments to something. Why do human beings turn toward something beyond the world they can touch?

What makes thought religious?

Before asking why, we need to say what “religious” means, and scholars have never agreed. In 1871 the British anthropologist Edward Burnett Tylor proposed a minimum definition: belief in “Spiritual Beings.” In 1902 the American psychologist William James wrote about the feelings, acts, and experiences of individuals “in their solitude,” so far as they take themselves to stand in relation to “whatever they may consider the divine.” In 1912 the French sociologist Émile Durkheim defined a religion as a system of beliefs and practices relative to sacred things, “things set apart and forbidden,” which unite those who share them into one moral community. Each definition draws the border in a different place.

This course follows a working definition built from two marks. In this reading, thought is religious when it turns on a transcendent, something beyond ordinary nature and ordinary human nature, held to be real and lasting; and when it offers a remedy, a way out of a human predicament, which is why the word salvation comes up so often. The transcendent may be a personal God or the gods; it need not be a person. In the earliest Buddhist teaching, nirvana is described as “a not-born, a not-brought-to-being, a not-made, a not-conditioned,” in one English translation of a short text called the Udāna. The course says plainly that this is a working definition, and one among several.

Try the definitions on some cases. Which count as religion depends on which border you use, and that is itself a lesson.

A remedy for being small

The first family of explanations says religion answers a need. The German theologian Friedrich Schleiermacher (1768–1834), a pastor and later a professor in Berlin, wrote his Speeches on Religion to Its Cultured Despisers in 1799, addressed to educated friends who thought religion beneath them. He argued that religion is neither doctrine nor morals but a “sense and taste for the infinite,” a matter of intuition and feeling. In his great work of theology, The Christian Faith (1821–22, revised 1830–31), he narrowed this to the idea most often quoted: the heart of piety is a feeling of absolute dependence, the consciousness of being wholly dependent and so of being in relation with God. (The phrase is often credited to the 1799 book. It belongs to the later one.)

This course adds a reading of its own. We are finite beings in an unbounded universe, and every support we lean on is finite too: a partner, a friend, a job, a health plan. Sooner or later each gives way, and human beings look for something that will not. In this reading, that search for an absolute support is where religion comes from, and “salvation” is the name for being brought out of a collapse into a lasting peace. It is a reading, not a finding. It fits some traditions well and others loosely.

Karl Marx put the need in a harder light. In an essay written in December 1843 and January 1844 and published in Paris that February, he wrote that “Religion is the sigh of the oppressed creature, the heart of a heartless world,” and then: “It is the opium of the people.” Read the whole passage and it is not simply an insult. Marx says religious suffering is both the expression of real suffering and a protest against it. Opium was also a common medicine in the 1840s, a painkiller, and the image says that religion eases pain. His point was that the comfort is an illusion: he wanted the real happiness that would make illusions unnecessary.

Sigmund Freud, in The Future of an Illusion (1927), traced religion to human helplessness in front of nature, death, and fate, and to the old wish for a protecting father made large. He called religious ideas “illusions,” but he defined the word carefully: an illusion is a belief driven by a wish, and, he said, an illusion need not be false. Whether it is true is a separate question that his argument leaves open.

Taken together, these thinkers give the “remedy” half of this course’s definition a long history. Each of them found the root of religion in what people suffer and lack. Whether they were right to explain it away, or to explain it at all, is where believers and non-believers part. Here, a careful reader should hold two things apart: a theory about why people believe, and a judgment about whether the belief is true.

The group and the mind

A second family looks not at suffering but at how societies and minds work. Tylor’s Primitive Culture (1871) traced religion to a mistake of reasoning: people dream of the dead, see their own shadows and breath, and conclude that each person has a soul that can leave the body. He called belief in souls and spirits animism, from the Latin anima, “soul.” Later anthropologists rejected his ladder of stages that ran from “savage” to “civilized”, but the idea that religion starts in how minds explain the world has had a long second life.

Durkheim went the other way. In The Elementary Forms of the Religious Life (1912) he studied reports on the clan totemism of Aboriginal Australians, compiled by European observers, and argued that when people worship the sacred they are in effect worshiping their own society, which is larger than any member and makes demands on all of them. For him the sacred and the profane, the set-apart and the ordinary, was the core division. His critics note that he built a universal theory on secondhand accounts of a few peoples.

The newest family comes from the cognitive science of religion. In 1980 the anthropologist Stewart Guthrie argued that religion is a kind of anthropomorphism: we see humanlike agents where there are none, in clouds, in noises, in the pattern of leaves, and he developed the idea in Faces in the Clouds (1993). In 1994 and again in Religion Explained (2001) Pascal Boyer observed that the supernatural beings found in cultures around the world are “minimally counterintuitive”: a ghost that passes through walls but otherwise thinks and acts like any person, or a god who hears prayers from anywhere. Such ideas break expectations in one or two striking ways and keep the rest, which makes them easy to remember and pass on. In 2000 the psychologist Justin Barrett described what he called a hyperactive agency detection device, and in 2002 the anthropologist Scott Atran (In Gods We Trust) argued that religion is a by-product of minds that evolved for other purposes. Course 10 on this site, Thinking, Fast and Slow, describes the quick, automatic kind of thinking these accounts lean on.

The agency idea has a simple logic, which you can test. A rustle in the grass is usually wind. Once in a while it is a predator. Running from the wind costs a little effort; ignoring a predator can cost everything. When the two mistakes cost very different amounts, the best policy is an alarm that goes off too easily.

A fourth idea asks what religion does for large groups. In Big Gods (2013) the psychologist Ara Norenzayan argued that as societies grew past the size where everyone knows everyone, belief in watchful, moralizing gods helped strangers trust each other and cooperate. A test came in 2019, when a team using the Seshat historical databank published a paper in Nature reporting that in their data, big complex societies appeared first and moralizing gods came after. In 2021 the paper was retracted: critics (Beheim and colleagues) showed that the result depended on how gaps in the data had been handled, and the paper was withdrawn. The authors published a reanalysis in 2022. The question is open.

Why it began is not whether it is true

It is tempting to say that if we can explain why people believe, we have explained the belief away. That does not follow. A believer can say that if there is a God who made human beings, it is no surprise that they were made able to sense and seek that God; a skeptic can say the same facts show religion to be a natural product of minds and societies. The theories above are about origins and functions, and on the question of truth they are silent. This course does not try to decide it.

What the theories do show is that religion is common, and that several things in human life point the same way. By Pew Research Center’s count, published in 2025, 75.8 percent of the world’s people were affiliated with a religion in 2020, and 24.2 percent were not. The second group is a mixed one: it includes atheists and agnostics, and also people who answer “nothing in particular.”