Earthly core of misty creations

Camilla Power addressed the CPGB's Communist University on the origins and evolution of religion. This is an edited version of her speech

I am going to start with the famous “opium of the masses” quote from Marx’s Contribution to a critique of Hegel’s philosophy of right, which gives an overview of Marx’s attitude to religion:

“The foundation of irreligious criticism is: Man makes religion; religion does not make man. Religion is, indeed, the self-consciousness and self-esteem of man, who has either not yet won through to himself or has already lost himself again. But man is no abstract being squatting outside the world. Man is the world of man - state, society. This state and this society produce religion, which is an inverted consciousness of the world, because they are an inverted world. Religion is the general theory of this world, its encyclopaedic compendium, its logic in popular form, its spiritual point d’honneur, its enthusiasm, its moral sanction, its solemn complement, and its universal basis of consolation and justification. It is the fantastic realisation of the human essence, since the human essence has not acquired any true reality. The struggle against religion is, therefore, indirectly the struggle against that world whose spiritual aroma is religion.

Religious suffering is, at one and the same time, the expression of real suffering and a protest against real suffering. Religion is the sigh of the oppressed creature, the heart of a heartless world, and the soul of soulless conditions. It is the opium of the people.”1

In this sense, religion is everything that makes us human: the fantastic realisation of the human essence. So the idea about the “opium of the people, the heart of the heartless world, the soul of soulless conditions”, etc does not contain in it anything about quashing religion as something holding back strategies of resistance for the working class, but revolves around the idea that religion enables people to get by in the horrible and exploitative world that they are born into. It is a call to give up illusions about the conditions, to give up the conditions that require illusions - fight the conditions, not the illusions themselves.

Contrast this with somebody like our friend Richard Dawkins, ultra-bourgeois liberal that he is, who thinks that you just have to stop people believing in things. As an anthropologist I am fascinated by the effects of religion, some of which are, of course, extremely exploitative, but some extremely empowering. Dawkins simply says, “Quash it all”.

I challenge Dawkins on two levels: not just on the level that he is failing to meet the sort of vision that Marx had of religion in his young writings, but also because Dawkins himself has the very theory which we can use to try to understand religion’s origins and role. Religion belongs to every human society. It is peculiar when it is not there. Our particular, peculiar capitalist society is very unusual. And, of course, religion keeps popping up at us - we cannot get rid of it. That is because it fundamentally belongs to humanity - it is what separates us from apes.

This is another quote, a footnote from Capital:

“Technology discloses man’s mode of dealing with nature, the process of production by which he sustains his life, and thereby also lays bare the mode of formation of his social relations, and of the mental conceptions that flow from them. Every history of religion, even, that fails to take account of this material basis, is uncritical. It is, in reality, much easier to discover by analysis the earthly core of the misty creations of religion than, conversely, it is to develop from the actual relations of life the corresponding celestialised forms of those relations. The latter method is the only materialistic, and therefore the only scientific, one.”2

This is what I want to discuss - it is from understanding the economic modes of production and the social conditions they throw up that we can draw and develop models for predicting what the first gods look like, what religion in its origins looks like. If we begin from this starting point, then we have also some chance of understanding in what directions religion may develop.

That is the materialist method, and the theory we need to use is Darwinian. It is Darwinian in the sense of behavioural ecology: counting the costs and benefits of individual strategies for reproductive success as the ultimate currency of evolution. The strategies that produced ritual behaviour - religion, as I understand it.

Three approaches

I divide these Darwinian approaches between, firstly, those that consider religion in terms of an adaptive strategy - something strategic that may achieve reproductive success; those that see religion arising in the first place as a kind of by-product (the spandrel idea) of the way our minds work; and the Dawkinsite view of maladaptation. The argument is that we are designed to be extremely receptive to new ideas when young, leaving us prey to be colonised and parasitised by religious “viruses”, to our cost.

I do not think the third approach stands up to examination. If it were true that religion is just a kind of parasite that does no good to humans, why have the human cultures around the world that developed a successful resistance to that parasite not been the most successful in terms of reproduction? That is, if religion is a parasite, then you would expect the host to develop a resistance to it. So I am going to leave Dawkins aside.

In my view Richard Sosis has done valuable work in applying a true behavioural ecology of costs and benefits, devising all kinds of interesting tests and finding historical and contemporary contexts where he can test out his main idea: religion involves very costly behaviour, which individuals engage in to demonstrate their commitment to particular groups. He examines such costly behaviour - the giving up of certain practices - and whether this results in a given group retaining its cohesion.

Robin Dunbar also views religion as a form of adaptation and a way to guard against free-riders and social defection. This is the same idea as Sosis - religion as commitment to a particular group - but Dunbar makes the idea more complex with his examination of “levels of intentionality” - taking into account what is in the mind of others who have a different perspective of the world.

Dunbar’s argument is that to have an understanding of god - or indeed to produce Shakespearian drama - five levels of intentionality are required. This is the maximum that humans are capable of achieving, the pinnacle of our evolutionary capacity. However, I believe that what happens with ritual behaviour is that you transform a perspective of I as an individual into a perspective of we - ie, we belong, we are part of this, we exist. This is essentially what ritual is doing.

And that is all you need. You do not need to build a schema of “I think that John knows that the elders believe that god is watching us” - it is not necessary to engage in this sort of Machiavellian series of calculations, which flows from a bourgeois, liberal and individualistic view of one-on-one transactions, instead of seeing religion for what it is. You need to look at what it is that creates we. That can provide us with the answer to the evolution of large-scale human cooperation - yet one that does not lose the “selfish gene” way of thinking, which always considers individual strategies first before moving on to group behaviour.

For 100,000 years at least humans were hunter-gatherers, so I focus above all on the religious experience of hunter-gatherers. The use of shamanic trance in hunter-gatherer societies is an example. This is a collective experience. In the case of southern African San Bushman religion, the healer goes into a kind of trance death. He or she (healers are not always male) will have gone through long periods of apprenticeship to learn to control a kind of “boiling energy” that rises up from the stomach and mounts up the spine in an incredibly painful and frightening experience.

Chris Knight talks about counter-dominance as the political mode of hunter-gatherers.3 Trance is the ultimate signal of counter-dominance. The healer has as much prestige as anyone in his society, yet he is exposing himself to a state of ultimate vulnerability, where he requires the help of all those around him to come through the trance experience and for them to come through it with him. So it is a huge signal of trust.

However, above all we should note the huge costs in terms of time and energy. There is the fear, pain and emotion. Around him mostly women sing and clap all through the day and the night and into the next day - at least two or three days would be spent on trance activity each month, and this in a subsistence culture that is quite food-limited. And in addition to that, all kinds of myth, story and rock art is woven and elaborated around it.

This is at odds with the hard-headed, ultra-materialist, “selfish gene” perspective of acquiring food or mates. There is a further cost of confusing the real world. For instance, making out that animals can speak or that the healer can go to god and admonish him - a typical San Bushman attitude quite without deference or asking for mercy! The cognitive confusion of mixing up the real world with an imaginary one should also be factored as a Darwinian cost. So the Darwinian approach is to ask: What selection pressures promoted an interest in sharing in the fantasies of other individuals? What drove human ancestors to expend increasing time and energy on “things that don”t exist” - fictions entertained by whole groups of individuals?

Costly signal

In order to understand this, let us look at what is known as the “costly signal theory” of animal communication.

The basics of natural selection can also be termed utilitarian - ie, selection for useful functional organs, such as wings and limbs or ears and eyes. We can expect that selection will proceed down a road of basic engineering efficiency - cutting and reducing the costs and making things more efficiently designed. Those organisms whose costs are cut will succeed and reproduce better in a Darwinian world.

In signal selection a different dynamic obtains: to prove their quality, animals signal with increasing costliness. The classic example is the beautiful display of the peacock showing himself off to all nearby peahens. Signalling can occur in various contexts: for instance between parents and young, or even between different species, like predator and prey. You would think that there would be no room for cooperation between a lion and its intended prey - a gazelle, say - but there is. The lion does not want to chase the very fastest gazelle, and any gazelle who is faster than the slowest wants to let the lion know that it is not worth her bother, so a certain amount of communication is in the interest of both. And this communication, in order to be reliable on both sides, has to be costly.

However, in the human case of the evolution of religion and symbolism, the most interesting area of sophisticated communicative signalling systems involves sexual selection - costly signalling between the sexes.

Across species, this arena of sexual conflict is a breeding ground for second-guessing, mind-reading and much more - particularly where, as in the case of birds, males and females share in the investment of reproducing their young. Here there is an obvious problem to be solved - if one partner or the other deserts, then the whole future of those young could really be threatened. So there has to be considerable trust established between the male and female of these species.

That is not the case for chimpanzees, where males do nothing but produce the sperm and compete for females. In the course of hominid evolution, humans engaged in working out this relationship between males and females. Significant evolutionary changes in communication occurred in the context of intense sexual selection.

Among birds, grebes are renowned for their ritualised behaviour, which offers some analogy to human ritual. What is interesting about ritualisation is that it may start off as quite normal behaviour - in the case of grebes preening feathers, for example. And then somehow that becomes formalised - a symmetrical kind of dance in which the pair of grebes mirror each other”s behaviour. An original functional behaviour becomes completely displaced, so that preening is transformed into a complex courtship dance. This mirrors a key aspect of symbolism: displaced reference. Collusion occurs between two animals, so that one thing, X (directly perceptible), stands for something else, Y (intangible or inferred) - here preening of feathers stands for courtship.

The other point about ritualisation is its formalisation. If we look at the lion and the gazelle, the lion is watching the gazelle’s behaviour closely in order to assess whether it can run very fast; it is also wondering whether the gazelle has noticed that it is being watched. If the gazelle has got its head down then the lion may assume he has not been seen, but if it lifts its neck and gazes then the lion assumes he has. So in the first place the lion is watching the gazelle’s normal behaviour.

But then the gazelle works out that if it really wants to show the lion that it has spotted her, it should exaggerate its normal movements. Then the whole signal evolution process becomes formalised and ritualised, with the gazelle raising its head and jumping up and down - demonstrating reliably to the lion that it is fit and fast, and that the lion should look elsewhere.

This signalling process consists of a two-way feedback that has evolved from fundamentally functional behaviour. As soon as the observed animal starts to exaggerate, formalise and stereotype that behaviour, that is when it becomes a signal and subject to the rules of signal evolution. Notice that the process is driven by an observer who is interested in the qualities of the observed animal, which responds by producing increasingly costly, formalised and displaced signals.

So how can such behaviour help us model human phenomena, such as the hunter-gatherer religious experiences? There are big differences, but these are also shared aspects: displaced reference, highly formalised activity. Could we model a process of ritualisation in human evolution - with observers driving the initially normal behaviours of the observed, who then exaggerate, formalise and stereotype their behaviours in order to communicate their qualities to observers?


In the human case of ritual, compared to the animal case, we need to talk about coalitions involved in such quality display, not just one-on-one individuals. Those coalitions are not just interested in preening feathers or showing off how strong they are, but in things that are out of this world. Our working signal evolution model of religion comes down to a process of ritualisation involving whole coalitions of “peacocks” engaged collectively in costly signalling that refers to “things out of this world”. But why?

Given that our starting point is sexual signalling, what are the possibilities? The basic scenario in sexual selection is nearly always one of males - like peacocks - signalling quality to females. So this would project coalitions of males showing off their qualities to females in competition with other rival: competing coalitions of males.

The alternative would be coalitions of females showing off their quality. This is the obverse of the sexual selection coin. Animal sexual selection only rarely produces this logic of female display to males, and when it does it has to be accounted for. Geoffrey Miller argues that culture basically consists of males showing off to females.

Let’s run with a male coalition model. Can any sort of religious symbolism evolve through it? You can imagine a bunch of Homo erectus showing off that they are a fantastic coalition by doing dances with beautiful, symmetrical hand axes, which could really impress the females. The problem is, though, where is the symbolic token - one thing, X, standing for another thing, Y? Could Homo erectus pick up a pebble and pretend it is a hand axe? Could he stick a twig in the ground as a symbol that his brother is coming in half an hour?

I argue that such a token will not do. That might possibly impress the group of females, but, as soon as tokens are offered in place of real muscle and bodies, a rival group of males will test for weaknesses. They are not going to be impressed by tokenism - what counts is ability of the males to fight.

And what exactly would females be looking for in these male coalition strategies? Presumably, they want strong, healthy and muscled men who are going to be good at protecting them and finding food, etc. It is possible that the men might want to show how tough they are by cutting themselves and bleeding a bit, but why should they use any cosmetics? Fake blood will not get the right message across. There is no reason to predict it. We also have no reason to predict that male coalitions would signal something like: “No sex today, please”. We would not expect sex taboos or males displaying as if females either.

Let us look at the alternative possibility of female coalitions signalling and see where it takes us.

There are good theoretical grounds for this reverse from normal. In sexual selection theory, the more that one partner invests in another to raise offspring, the choosier they should be about which partner they invest in. Usually it is the females who are choosy about who they want to father their offspring, because they are investing large amounts. That remains the case with humans - but in their case the males also invest through the hard work of hunting to get sexual access to a female. Then they must produce more meat and more support for that offspring as it grows up. The more they invest, the more choosy they should be too.

So the question then is: which females are they going to choose? The obvious Darwinian answer to that is to find a nice, fertile woman who can have healthy babies. Today we women appear well designed to confuse human males about when exactly we are fertile. This has the effect of rewarding males who hang around with us for a long time and keep having sex with us - they are more likely to become fathers. Until a woman is visibly pregnant, there is no moment at which the male can tell that he has been successful, at which point he might run off and find another female.

However, one signal does give the game away, allowing males to discriminate. A menstruating female can be made pregnant, whereas another who is obviously heavily pregnant or has just given birth and is breastfeeding can obviously not. In the Darwinian world, males cannot ignore that information.

Females who are not cycling are potentially threatened by the presence of a menstrual female who will get lots of attention from males. At that point there are two choices: either females should try to hide evidence of menstrual females, or they do the very opposite: make a huge song and dance out of it. The reason that the second option is the most productive for females is that males are hugely interested in a potentially fertile female in the neighbourhood. In this situation the best thing for the females who are pregnant and already have babies is to join up with the menstrual female, “borrowing” her signal. They all paint up together using blood, or blood substitutes - red cosmetics - and make a fantastic display. Now we have a costly display of quality by a coalition of females, showing off to the interested observer males.

Obviously, any male will be interested in the young female who is becoming ready to reproduce. But she is not alone. She has a whole alliance - her mother, aunts and sisters - who will be supportive of her and her future offspring. So the quality being advertised to potential investor males is not simply the presence of a single fertile female, but the fact that the women as a collective are a solid coalition. The deal is: invest in us, bring energy, bring meat for the girl”s coalition and together we will raise you great babies.

Collective ritualised signalling demonstrates the quality of such coalitions in an easy to see and hard to fake manner. From a male perspective, relatives of those women will defend their coalition, because it raises their own kin. But what of males who are potential mates of those women? A dominant male might try to fight his way through and grab hold of the obviously fertile, menstrual woman. But the majority of males actually have no interest in such behaviour. Instead, males who are willing investors should go along with the reproductive counter-dominance strategy. They would observe the “picket line”, the rule of law, and the effectively religious concept of “We’re all fertile, we’re all sacred, we’re all taboo”.

In resisting dominant philanderers, to say no loud and clear, we predict that females will produce a set of signals saying: “wrong sex, wrong species”. Whereas a female chimp employs oestrus signals to attract the most dominant male, the counter-dominant human female will signal in ritual song and dance that she is not even the right species, pretending to be a game animal, or pretending to be the wrong sex by dressing up as a male, and showing “wrong time” with cosmetic signals of ritual “menstruation”.

This is our testable model using Darwinian theory, reversing the usual direction of signal selection. Our predictions also concern cosmetics: that is, the earliest types of symbolic behaviour and the earliest attributes of religious ritual - what the first gods actually look like. Religious entities are being produced by counter-dominance, where the coalitions of females, plus all their male allies (ie, brothers and investor males) participate and share in these ritual signals of counterreality - things that do not exist in this world.

“Wrong” gods

The first gods will be in some sense be “wrong” - the wrong species, wrong sex, and red and bloody. There are strong reasons in theory why these collective representations will be maintained with extreme fidelity. Even through 100,000 years of hunter-gatherer economies, even when we come out of the other side in the Neolithic and in future developments of religions of state and market economies, there should still be a faithful conservation of “time-resistant” syntax. We expect this above all in rock art and myth of groups like San Bushmen hunters.

We expect red ochre cosmetics to relate strongly to menstrual taboos in hunter-gatherer cosmologies, and in addition, hunter-gatherer prohibitions about sex and menstruation should operate inside lunar/menstrual cosmologies. Ritual preparation for the hunt should occur at the dark side of the moon, while full moon should be the time when hunting is taking place.

To conclude, we have the emergence of hunter-gatherer religious experience through “wrong sex, wrong species” that incorporates both trance and initiation as ways of accessing this “other world” counterreality. It is likely that even before the Neolithic, and the onset of farming and property, there were strategies whereby religious symbolism became oppressive and exploitative. But there is no way that religion - the collective sharing and enacting of ritual fantasies - can have emerged in an exploitative environment.

Our image of the evolution of religion is of counter-dominance, where what is being signalled above all is trust and solidarity within groups. Trust is essential before the collective can share such visions of the supernatural.

And the very first symbol of the supernatural was the red flag of the women’s “sex strike” picket line.