Post-Atheism: Religion Refactored

This is the second installment in a multi-part series on Post-Atheism. In the previous essay, I noted some parallels between atheism and the teenage years and argued for a more 'adult' conversation. Today we'll take a fresh look at religion, to correct some of the misconceptions atheists bring to the discussion. And finally, next week, we'll answer the question, "What would a fully scientific religion look like?"

Why religion matters

As readers of this blog are aware — perhaps painfully so — I have something of a fetish for religion. It's not that I'm religious myself (quite the opposite), but I find the phenomenon deeply fascinating.

I've been asked on a number of occasions why I'm so fascinated by religion. The truth is, I'm baffled why others aren't fascinated.

Religion is both fundamental and alien. It's fundamental to the human condition, and utterly alien (to myself, at least, and to fellow atheists). How can you look at religion and not be filled with fascination — with equal parts awe and curiosity?

As a social phenomenon, religion is a puzzle of the finest order. What's going on here? Why do so many people, across all times and cultures, believe such strange things? Why do their beliefs persist in the face of newer and much more powerful theories about how the world works? What benefit do they get from these beliefs? And are the beliefs even important, or are they a complete distraction — or misdirection?

How have religions endured for millennia, achieving stability in the face of so much social and political change? There are religions alive today which have outlived almost every other institution on the planet — every corporation, economic regime, and currency — every legal system, university, and (secular) school of thought — every dynasty and empire, every political state known to man -- even every civilization. Some languages have achieved the same feat of longevity, but they're the exception that proves the rule: most of these languages (Sanskrit, Latin) exist today only because they've been attached to a religious tradition.

People often cite the Great Wall as the only human artifact visible from space with the naked eye. This 'fact' is wrong, but it gets repeated because it induces a sense of awe and wonder. Seen from space, human creatures are sub-microscopic entities — but we've managed to build something HUGE, something that's made a visible impact on the shape of our planet. How grand! How marvelous!

But religion has the same quality of being visible across vast distances — in time, if not in space. When you take the long view of history, most individual humans disappear. Their effects on the world are fleeting, ephemeral, like ripples from a stone skipping across a river. Religion is like the river of history itself — or at least one of its major tributaries.

People like to visit archaeological sites to experience this sense of historical awe. When we cower in front of the Great Pyramids or the Acropolis, we marvel at human history. What must it have been like for people in the past? What were their lives and societies like? How did they manage to build such enduring structures?

But there are religions from the same era — social structures, built by ancient humans — that are still alive today. These aren't just inert buildings or lifeless scratchings on clay tablets, but living, breathing, embodied representations of how ancient people conceptualized the world. It's an archaeology of the cognitive variety. And you can visit these structures today, in your mind or even in person. There's no travel required, but you'll need the same spirit of adventure — because if you can't open yourself up, at least a little, you won't be able to see them for what they really are.


The tagline for the excellent blog Ribbonfarm is "Experiments in refactored perception." It's a great description for any material that tries to change how people see the world.

Refactoring is a concept borrowed from software engineering. To refactor a computer program is to change its structure (in code) without changing its behavior (to the user). Usually the goal is to impose a cleaner structure, or a structure better suited to whatever future purpose you have in mind for the program.

Similarly, today I'd like to propose a way of refactoring how we (as atheists) understand religion. We're not going to refactor religion itself, because religions are things that live out in the world — and we're just sitting here, reading (or writing) words on a screen. Instead we're going to refactor our understanding of religion. And the new structure that emerges will be useful, next week, when we discuss atheistic religion.

It should go without saying, but I'll say it anyway: much of what I'm about to discuss won't apply to all religions. These are huge, complex, multifaceted socio-symbolic institutions, and as I mentioned last week, there's disagreement about even the definition of religion. Given the size and scope of these things, generalizations are perilous. But I aspire to understand not what is essential or universal about religion, but merely what is broadly true. And I'm sure I will make mistakes, for which I welcome corrections. But I don't want to be afraid to make these mistakes. I just want to jump in and have fun.

So let's jump in.

Religion is not about beliefs

The first thing to address, in our new refactored understanding of religion, is the role of beliefs. And here I'm talking specifically about the crazy beliefs. Religions promote a lot of mundane beliefs as well (about how to treat others, how to live virtuously, etc.), and these are important, but as atheists we tend to obsess over the crazy beliefs: anthropomorphic gods, talking animals, angels, demonic possessions, reincarnation, and all the rest.

This obsession is understandable, a hazard of the atheist occupation (as it were). After all, beliefs are what sent us to atheism in the first place. We care foremost about what's true, and the crazy religious beliefs are so obviously false. But they're also loud, colorful, and heavily advertised, practically screaming for attention. How could we not focus on them?

And it's not just atheists who obsess over the beliefs. Christians, for example, are especially eager to highlight their beliefs, whether they're talking to atheists or followers of other religions.

This is because beliefs are ideological "fault lines" — the best way of mapping the territory between believers and non-believers, or between different types of believers. This is why we're all obsessed with them. And the obsession serves a purpose — for rallying together with the people on your side, or for drawing others across the divide — but it's a distraction when the goal is to understand religion as a social phenomenon.

In other words, if we want to get past questions of what's true about the world, and start asking what’s true about religion itself, we have to ignore the beliefs, because they only get in the way.

So I propose that we relegate beliefs to the sidelines, at least for now. Instead of treating religion as a system of beliefs (with some weird rituals, myths, and institutions tagging along for the ride), it will be more productive for us to treat religion as a set of practices — with some weird beliefs on the side.

Words deceive. Actions speak louder. So let's study what religions actually do, rather than what they merely say.

(This approach is useful, incidentally, for studying all kinds of human social behavior. Studying behavior lets us cut through the deceptive nature of words, to get at the heart of what's really going on.)

So yes, religions promote many silly beliefs. Yes, they're among the silliest things people believe. But by their very nature as (mere) beliefs, they aren't particularly interesting. In most cases, people don't act on abstract notions. We're confabulators, one and all, religious and secular alike. We act primarily out of self-interest, coercion, fear, love, etc., and only justify our actions, later, on the basis of beliefs.

We'll return to the beliefs in a minute. In the meantime, let's focus our attention on religious practices.

Tribal strategies

Now if a religion isn't a worldview (a system of beliefs), then what is it? What's the root of the thing, its raison d'être?

Here's the best answer I've been able to come up with:

Religions are comprehensive strategies for the survival and reproduction of human tribes.

It took me a while to settle on this understanding. But once I figured it out, everything that had once puzzled me about religion began to make eerie sense.

Note that I'm using the word "tribe" here very broadly. Maybe a better term would be "group," but I think there's something deeply tribal about religious behavior. The earliest religions (out of which later religions evolved) were very explicitly tribal, in the sense of pertaining to a single ethnic group. But even the modern universalizing religions have tribal natures. Islam and Christianity, in particular, can be seen as attempts to create massive, globe-spanning 'tribes', decoupled from ethnicity, but nevertheless united in terms of their survival and reproduction.

Let's look at a variety of religious practices to see how they support the overall strategic goals of survival and reproduction. We can divide these practices into three categories:

Survival. These are the practices which relate to the biological survival of the tribe. Note that religions don't care, too much, about the survival of any specific individuals in the tribe (that's what we have brains and bodies for, our evolutionary heritage), but rather for the tribe as a whole. And keep in mind that the biggest threats facing a tribe are (a) other tribes, and (b) internal decoherence (dissent, in-fighting, misalignment, etc.).

Survival practices include:

  • Group bonding rituals. E.g. rituals of sacrifice, rituals of synchrony, rituals of solidarity, and public displays of devotion. All of these rituals help large groups develop more trust among their members, making it easier for them to live and do business with each other, and also ensuring that they will act collectively when threatened by outside influences.
  • Shared mythology. Many religious myths serve the same purpose as the rituals of group bonding, i.e., uniting the tribe together into a single, coherent community. Often these take the form of stories which illustrate (make vivid and visceral) a shared history, shared present, and shared future. Where did we come from, who are we, and where are we going? Different traditions give different answers, but always with a strongly-implied subtext: Whatever this is, we're in it together.
  • Wariness toward outsiders. I.e., instructing the tribe in how to deal with other tribes, non-believers, earthly political powers (rendering unto Caesar what is Caesar's), etc.
  • Social harmony. Religions have a particular interest in getting their members to help one another. Such help may not always be in the narrow self-interest of the person offering the help, but it's almost always in the broader interest of the tribe. Religions also prescribe roles and responsibilities for different members of society, as well as rules and regulations, standards of ethical conduct, etc., all of which help large groups live together in relative peace and prosperity.

Reproduction. These are the practices which relate to the reproductive success of the tribe. Again, religions don't care about the reproductive success of any individual members, but only for the tribe as a whole. Thus, for example, the near-universal prescription for monogamy and the especially strong prohibition against extra-marital affairs. Because while it may be optimal for any individual to cheat on his or her spouse, this is by no means the best way for the tribe as a whole to create as many babies as possible, nor how best to raise them (i.e., in two-parent households, where the father is perfectly aligned with the mother in caring for the children he trusts to be his genetic offspring). See also the prohibitions on contraception.

Meta. These are the practices a religion uses in order to further itself. If you accept that a religion serves its more fundamental, 'object-level' purposes (survival and reproduction), and serves those purposes well, then it makes sense for the religion to stabilize and perpetuate itself. These 'meta-level' practices thus indirectly help the tribe survive and reproduce, by making the whole package more coherent and more enduring.

When Dawkins describes religion as a "meme complex" or "mind virus," it's the meta-level practices he has in mind. It's also where we can use our understanding of replicators (biological and/or ideological), whose three cardinal virtues are fidelity, fecundity, and longevity. Every meta-level religious practice plays to at least one of these three virtues.

Examples of meta-level practices include:

  • Maintenance of scriptures.
  • Indoctrination of children.
  • Cultivation of a professional priesthood to steward the religion. (Notably, most priesthoods adhere to a policy of celibacy. Why? So their loyalty to the religion won't be undermined by loyalty to their genetic offspring.)
  • Persecution of heretics, apostates, etc. Or more broadly, "moralistic punishment," all forms of which help to stabilize the religious beliefs and practices.
  • Evangelism. This helps the religion spread to more brains. Incidentally, it also furthers the survival goals of the tribe, owing to the network effects that religions enjoy (more followers = more benefits to each follower).


In light of this view of religion (as a tribal strategy), I'd like to share a little hack I sometimes use to make sense of religious practices. Whenever I hear someone say "God," I try substituting "society" in its place. E.g.:

  • God is great becomes society is great.
  • When someone says, praying before a meal, "We give thanks to God for this food," I hear praise of society, of civilization.
  • When a Muslim says that Islam is all about submission to Allah, I understand this as submission to society.

This hack allows me to make sense of a lot of religious beliefs and behaviors, in real-time. Utterances that would otherwise be inscrutable and/or meaningless suddenly become sensible statements about an entity I actually care about (society).

In fact, here's a wild idea: maybe this substitution ("society" in place of "God") is actually reverting a hack introduced thousands of years ago. If we follow Durkheim, the original hack was to personify society in the first place, as a literal God figure — a man-in-the-sky who's much easier to imagine than an abstract notion like "society."

For many people, it must surely help to take the symbol literally — to imagine a real anthropomorphic god watching and judging all of their behavior from on high. But for those of us who value literal truth highest of all, we can nevertheless try to give a generous interpretation of religion's symbolic truths.

At any rate, it's a thought.

Beliefs are shibboleths

Finally, now, we can return to the question of what to make of all the crazy religious beliefs. Remember that we're not investigating here whether the beliefs are true or false, but instead, assuming they're false, what purpose they might nevertheless serve.

As I wrote in an earlier essay:

Having decided that God isn't real, I was left with an even more challenging question. If the beliefs are false... what (on Earth!) is going on? How does it all work? Why do so many people believe, and why are the beliefs so utterly crazy?

When I realized what was going on here, it kind of blew my mind. The craziness of these beliefs isn't a bug; it's a feature. Religious beliefs are necessarily crazy, not just incidentally crazy — because only crazy beliefs can be used to identify members of the community.

As Robin Hanson points out, beliefs are like clothes. We sometimes adopt them as social objects, not because they tell us how to behave in real-world situations, but because they send signals to the people around us — about the kinds of people we are, what we value, and who our allies are.

Thus beliefs will often function as shibboleths: badges of group membership useful for distinguishing an in-group from an out-group. A crazy religious belief — e.g. that Jesus rose from the dead — can be used to separate believers (those willing to go along with their tribe members and religious authorities) from heretics (those who aren't 'properly' aligned with the group).

This function explains so many otherwise-puzzling facets of religious belief:

  • It explains why the beliefs are advertised so heavily: because they're being used as signals. If beliefs act as shibboleths, it does no good to hold them, privately, inside one's head. They need to be displayed, prominently and proudly, to the whole world.
  • It explains why the beliefs are mostly false. You can't use true propositions (like "the sky is blue") to identify members of your group, because everyone tends to believe these things, for reasons unrelated to their loyalties.
  • It explains why the beliefs aren't just false, but unfalsifiable: meaningless in the strictest (scientific) sense. No observation will ever prove — or more to the point, disprove — a belief in reincarnation or transubstantiation. The obvious benefit here is that the beliefs will never get exposed as false, and people won't waste (too much) time seeking evidence for or against the beliefs. Doomsday cults are an obvious counterexample, in that they make specific, falsifiable predictions about the end of the world. It's a strategy for drumming up support, but a shortsighted and unsustainable one. And quite sensibly, mainstream religious groups usually steer clear of making these kinds of specific predictions.
  • Finally, it explains why the beliefs are (more or less) harmless. A dangerous belief — e.g. that tigers are fun, friendly creatures — might serve well as a shibboleth, but it's going to cause problems for the survival of the tribe. Not surprisingly, beliefs that are unfalsifiable tend to be harmless as well.

Bottom line: No one is going to live or die, or make more or less money, or be nicer or meaner to her fellows, strictly because of her belief that Jesus is the son of God. But all of those things might fall out as second-order consequences of the belief, insofar as it creates a boundary between believers and non-believers, and allows believers to maintain higher levels of trust and solidarity among themselves.

Religions, of course, aren't the only place we find shibboleths. The political sphere also makes frequent use of beliefs-as-badges. We humans are always looking for ways to create divisions between groups, and then to take stock of who's on our side and who's against us. Religious beliefs are just a particularly efficient way to go about it.

Natural phenomena

Two other quick notes on this exercise in refactoring religion.

First let's talk about how to treat "religious experiences." A common view — one which I held for many, many years — is that these are delusions, the stuff of pure fantasy. God doesn't sneak into our brains and make us suddenly, overwhelmingly aware of His presence — because of course He doesn't exist. But this is to confuse the interpretation of the phenomenon with the phenomenon itself. Religious experiences — or spiritual or transcendental experiences (whatever you want to call them) — are very real indeed. They're a state of consciousness the human brain seems especially prone to, under the right conditions. And the force of these experiences is also much stronger than I've (historically) been willing to credit.

Second, I think it's useful to treat entire religions, not as artifices dreamt up by human schemers, but rather as natural phenomena. As Loyal Rue puts it:

Religious traditions... are natural entities, like atomic systems, molecular systems, organ systems, neural systems, and the rest, which have emerged by natural causes in the creative process of cosmic evolution.

One way to look at this, of course, is that everything we do is 'natural' — because we're creatures who have grown out of the natural world. But even if we make the distinction between man-made ('artificial') and non-man-made ('natural'), I still think it's productive to treat religions as natural phenomena. These are systems that grow, organically and spontaneously, in human communities. That they sprout up everywhere, and have similar structure in all cultures, is a testament to how natural they are.

Looking ahead

All of what we covered today will be important for our task next week: outlining a godless religion grounded firmly in science. It's especially important because we're going to have to break with traditional religions in a number of ways. Most notably, beliefs will be central to the enterprise. Plus we'll probably want to go easy on the tribalism. So it will be an open question whether the result is, in fact, something it makes sense to call a "religion." But it will be a fun exercise regardless, and we'll see how far we can get.

Understanding that religions are natural phenomena is also critical for giving us the necessary humility for our task. Whatever this new 'religion' is, it won't be something manufactured by a single human author, or even by a handful of authors. If it has any meaning or viability whatsoever, it's because this religion is already growing and taking root in our culture, even as we speak.


If you enjoyed this essay, you might also enjoy:

Originally published July 15, 2013.