Religion is a funny thing. It presents special problems to people (like me) who think religious beliefs are mostly false, because religion seems to be a human universal. It exists in all human societies, past and present. How can this be, given that false beliefs are usually a hindrance to those who have them? As a rule, false beliefs tend not to promote the proliferation of the believer’s genes or culture. Is religious belief an exception to that rule?
I want to sketch an explanation that turns the usual understanding of religion on its head. The usual understanding assumes that religious beliefs inspire religious behaviour. In other words, it assumes that our ideas about “how the world is arranged” cause and hence explain rituals, ceremonies and religious “observances” of various kinds. For example, the belief that God is omnipotent gives people a reason to pray, and that explains why religious believers pray. Note the pattern there: belief first, behaviour second. I want to suggest the very opposite is true: that religious behaviour is selected for because it serves a real biological function, and that religious belief is a sort of after-the-fact rationalization of that behaviour. Instead of the beliefs explaining the behaviour, the behaviour explains the beliefs. For example, acts of prayer tend to promote belief in an omnipotent God rather than vice versa.
Looked at that way, religious beliefs are not an exception to the rule above, but are instead a mere side effect of something else — something that really does confer a biological advantage. There is much more to religion than belief, and religious beliefs are highly “theoretical”: they tend occupy the “inner” recesses of our minds, emerging in behaviour only rarely, such as when we discuss theology (if ever). Nor do many of us have first-hand “religious experiences” by observing miracles or supernatural phenomena. So I would argue that religious beliefs survive despite being mostly false, because selective pressures do not bear on them very directly — not as directly as, for example, beliefs about the current location of the furniture in the room. When that sort of belief is false, it leads to injury: I bruise my shins on the coffee table every morning when walking across the room to let the dog out.
I submit that religious behaviour promotes social cohesion in a special way. It works within groups (but not between groups) by providing a framework for expressions of sincerity. This is how rituals work in the social lives of other animals, such as in the courtship displays of monogamous birds. Ritual has a much greater importance for our own species, because ours is easily the most cooperative species on Earth. Religions also give their adherents distinctive ways of understanding right and wrong, so that co-religionists tend to agree on moral questions. All of which probably oils the wheels of reciprocal altruism. Reciprocal altruism is the biological version of “I’ll scratch your back if you’ll scratch mine” between unrelated individuals. It requires trust and familiarity, but cannot depend on shared genes as a guarantee of good faith. The next best alternative is to check that “we’re all singing from the same hymn sheet”, in the present context almost literally.
If religion serves social functions of the above sort, then it is bound to involve “costly signalling”. This is an important idea that has entered mainstream biology in recent decades. To advertise something of value such as good genes or good faith, the most reliable indicators are those that cannot be imitated; for that, they have to cost an honest communicator something that dishonest communicators cannot afford. The classic examples involve sexual ornamentation such as the peacock’s tail. The tail handicaps its owner, but it must if it is to send a convincing signal that honestly advertises his genetic worth. The tail is not long and unwieldy despite but rather because it’s an obvious danger to its owner’s life. Costly signals have to handicap the signaller.
This sort of costly “handicapping” signal seems quite common in human social life. Parents of engaged couples throw wedding parties of enormous expense to indicate the sincerity of their commitment to the new union. Men drive cars that invite comments such as “that must’ve set you back a bit!” — By being “set back a bit” yet avoiding disaster, they advertise their solvency. And in both sexes, conspicuous self-denial can reach the level of self-destruction and beyond (as in suicide bombers).
In these cases of handicapping, hardships are not imposed by the environment. They are self-imposed in more or less arbitrary fashion by the signallers themselves. I deliberately choose the words ‘arbitrary’ and ‘fashion’ here. The peacock’s tail feathers happen to have blue “eye” features, but purple stripes would have done just as well, had that fashion taken hold, and as long as they were equally hard to arrange.
If religious observances serve the social functions described above, it wouldn’t matter much what they involve as long as they are demanding enough — by calling for a costly fastidiousness, or self-harming for example. Flagellation is as good as fasting — and more convenient if you’re short on time.
Let’s pause for a moment to reflect on the differences between my approach here and that of mainstream atheism, which tends to see religion as ridiculous, as a pathological waste of time, or as something we would be better off without. But I see it as arbitrary; and if it is time-consuming, it has to be to serve its social purpose. We can’t really hope to get rid of it, because our sociality is indispensable, apart from a few individuals here and there. As Burke remarked, “man is by his constitution a religious animal.” The few who really succeed in giving it up have to relinquish their own constitution a bit.
Despite the biological requirement of some such framework for ritual observances, we shouldn’t be surprised if we find a reasonably wide range of different frameworks, because social cohesion within a group often requires friction between such groups, especially if they have competing interests as groups. A given set of observances indicate specific commitment to this group, and not that group.
That seems to be what we find when we look around the world at different peoples, especially those whose interests conflict. Over the course of history, where group interests begin to conflict, their religions begin to diverge. For example, most of the population of the US is descended from people drawn from a smallish range of religions. If we ignore the “social cohesion” aspect of religion, we would expect to find a similar homogeneity of religious practices in the present-day US. But what we find in fact is that new religions and cults spring up as groups and their interests diverge — notably in newly-acquired territories such as Utah and California. It’s as if social cohesion within groups is promoted where differences are created between groups.
So far, I have focused on religious behaviour. But what of religious belief? Mainstream atheism tends to see religious belief as one of the “craziest” aspects of religion, the origin and cause of all the “pointless” ritualistic behaviour. But look at it from the other direction: that sort of behaviour only seems pointless from a traditional perspective that overlooks handicapping. Once we take account of handicapping, the behaviour may be arbitrary and guided by fashion, but it isn’t pointless at all. I suggest that the beliefs “come later” as the product of rationalization, and are in fact the “least crazy” aspect of the whole operation, even though most of them are false.
Recent philosophers in the tradition of American pragmatism such as Daniel Dennett and Donald Davidson have argued that the content of beliefs (and other mental states) is a matter of interpretation guided by a “principle of charity”. (Anything that has “meaning” such as mental states or language depends on interpretation like that.) An agent’s beliefs as a whole have to “make as much sense as possible” by meshing with each other, with the agent’s behaviour, and with the way the world is. That determines what they are “about”. It seems to me that ritualistic activities and the beliefs that rationalise them are indeed woven into a coherent whole like that. But this coherence of beliefs has a cost — what they gain by meshing smoothly with each other they lose in clarity, concreteness, testability, and truth. Thus a typical religious belief is likely to be “mystical”, abstract, hard to check, and false.
It is hardly surprising that there are differences between the beliefs of different religions. But they have quite a lot in common too. I would argue that they converge mostly as a result of that principle of charity, which in effect insists that they be consistent with shared human nature and with our most widely-held assumptions.
Let’s look at an example. I mentioned above that religious beliefs were highly “theoretical”. They have to be, because otherwise they would be too strongly selected against. So they exist as a hotchpotch of commitments to such things as “creation” stories about the origins of the world and of humans, to “legalistic” rules and regulations, to “cautionary” accounts of punishments and rewards, to vaguely “spiritual” ideas about the nature of the person. Consider the last of these.
We humans are unique in our use of language and in the complexity of our thoughts. When combined, these human abilities give rise to something even stranger — our use of language to describe the contents of our thoughts. This special sort of language is remarkable for the “roundabout” way it describes its subject matter. For example, when we describe an experience we’re having — an experience of red, say — we use a word whose primary purpose is not to characterise experiences but common public objects such as tomatoes and poppies. Children first learn to use the word ‘red’ to describe the colour of that sort of public object, and only later co-opt the very same word to describe the nature of their own “internal” experiences, experiences of a sort that normally occur when they look at those public objects. That is, they describe the state of their brains by using a word that originally applied to the quality of light reflected off fruit and flowers. Something similar applies to the content of beliefs (and desires, hopes, fears, etc.). We describe the way we believe things are arranged in the world by using an “embedded sentence” that describes how things would be arranged if the belief were true. For example, I might describe my belief that there is someone on the roof by uttering the sentence ‘I think there’s someone on the roof.’ That description of my mental state contains the embedded sentence ‘Someone is on the roof’, which on its own describes a state of affairs in the world rather than a mental state in my head.
To make a long story short, the language we use to describe our internal mental representations of the world “mirrors” the language we use to describe the real external world. If we’re not careful, this apparent “mirroring” can weave a sort of spell, a spell that tricks us into assuming that our ideas are ghostly versions of the physical things they are ideas of. We assume these “ghostly” entities must inhabit a “spiritual” realm, so now we think our minds are themselves “ghostly”, and maybe even entirely separated from the physical world. This is just the sort of “conjuring trick” played by language that prompted Wittgenstein to characterise philosophy as a “battle against the bewitchment of our intelligence by means of language”. As it happens, practically the entire population of the world has fallen for this trick, and subsequently most of us understand minds as non-material “ghosts”, which need not always stay inside the “machines” of their hosts’ physical bodies.
Now along comes religion, gravitating as it always does towards the theoretical and the arcane. It seizes on this almost universal “ghost” idea, and develops it by weaving a mythology around it. Practically all religions have developed the idea, each in its own distinctive way, of course, as befits any framework whose details are partially determined by the requirement that it be different from other frameworks.
This interplay of convergence and divergence is not all that unusual. We can see something like it in the development of computer software. Windows has its “Explorer” and Mac OS X has its “Finder”: their main purpose is similar, but their names and finer details differ, not by accident but by design. Or again, image editing software such as Gimp and Inkscape deliberately differ from Photoshop and Illustrator to avoid breaching copyright law.
I mentioned above that mainstream atheists see religious belief as pathological. But I would agree with Burke that it’s “natural” for humans. It’s therefore very hard to shake off. It may be easy to dismiss the idiosyncrasies of specific religious beliefs, because it’s pretty clear that they are more or less arbitrary products of fashion and random divergence. But where religious beliefs converge, as almost all do on the idea that the mind is “spiritual” or “ghostlike”, they reflect nearly universal human assumptions.
Another nearly universal assumption is the idea that the living world is filled with precariously balanced “ecosystems”, which is really belief in design or “a way things were meant to be”. Even people who explicitly reject design in nature treat “ecosystems” with a misplaced reverence. Again, we see man’s religious constitution at work.
But perhaps the most ubiquitous and unshakable aspect of our religious constitution can be found in our epistemic habits. Most of the time, most of our beliefs are acquired through adoption from others — we take beliefs on because we want to fit in with a group whose members have those beliefs already. The prevalence of that habit throughout human life — and disgustingly to me it seems to be everywhere — is the surest evidence that social cohesion plays an overwhelming role in how we think and act. It seems to me that philosophers must work hard to lose that habit, at least as far as our human nature will let us.