The Beginning of Infinity: Explanations That Transform the World - David Deutsch (2011)

Chapter 15. The Evolution of Culture

Ideas that survive

culture is a set of ideas that cause their holders to behave alike in some ways. By ‘ideas’ I mean any information that can be stored in people’s brains and can affect their behaviour. Thus the shared values of a nation, the ability to communicate in a particular language, the shared knowledge of an academic discipline and the appreciation of a given musical style are all, in this sense, ‘sets of ideas’ that define cultures. Many of them are inexplicit; in fact all ideas have some inexplicit component, since even our knowledge of the meanings of words is held largely inexplicitly in our minds. Physical skills, such as the ability to ride a bicycle, have an especially high inexplicit content, as do philosophical concepts such as freedom and knowledge. The distinction between explicit and inexplicit is not always sharp. For instance, a poem or a satire may be explicitly about one subject, while the audience in a particular culture will reliably, and without being told, interpret it as being about a different one.

The world’s major cultures – including nations, languages, philosophical and artistic movements, social traditions and religions – have been created incrementally over hundreds or even thousands of years. Most of the ideas that define them, including the inexplicit ones, have a long history of being passed from one person to another. That makes these ideas memes – ideas that are replicators.

Nevertheless, cultures change. People modify cultural ideas in their minds, and sometimes they pass on the modified versions. Inevitably, there are unintentional modifications as well, partly because of straightforward error, and partly because inexplicit ideas are hard to convey accurately: there is no way to download them directly from one brain to another like computer programs. Even native speakers of a language will not give identical definitions of every word. So it can be only rarely, if ever, that two people hold precisely the same cultural idea in their minds. That is why, when the founder of a political or philosophical movement or a religion dies, or even before, schisms typically happen. The movement’s most devoted followers are often shocked to discover that they disagree about what its doctrines ‘really’ are. It is not much different when the religion has a holy book in which the doctrines are stated explicitly: then there are disputes about the meanings of the words and the interpretation of the sentences.

Thus a culture is in practice defined not by a set of strictly identical memes, but by a set of variants that cause slightly different characteristic behaviours. Some variants tend to have the effect that their holders are eager to enact or talk about them, others less so. Some are easier than others for potential recipients to replicate in their own minds. These factors and others affect how likely each variant of a meme is to be passed on faithfully. A few exceptional variants, once they appear in one mind, tend to spread throughout the culture with very little change in meaning (as expressed in the behaviours that they cause). Such memes are familiar to us because long-lived cultures are composed of them; but, nevertheless, in another sense they are a very unusual type of idea, for most ideas are short-lived. A human mind considers many ideas for every one that it ever acts upon, and only a small proportion of those cause behaviour that anyone else notices – and, of those, only a small proportion are ever replicated by anyone else. So the overwhelming majority of ideas disappear within a lifetime or less. The behaviour of people in a long-lived culture is therefore determined partly by recent ideas that will soon become extinct, and partly by long-lived memes: exceptional ideas that have been accurately replicated many times in succession.

A fundamental question in the study of cultures is: what is it about a long-lived meme that gives it this exceptional ability to resist change throughout many replications? Another – central to the theme of this book – is: when such memes do change, what are the conditions under which they can change for the better?

The idea that cultures evolve is at least as old as that of evolution in biology. But most attempts to understand how they evolve have been based on misunderstandings of evolution. For example, the communist thinker Karl Marx believed that his theory of history was evolutionary because it spoke of a progression through historical stages determined by economic ‘laws of motion’. But the real theory of evolution has nothing to do with predicting the attributes of organisms from those of their ancestors. Marx also thought that Darwin’s theory of evolution ‘provides a basis in natural science for the historical class struggle’. He was comparing his idea of inherent conflict between socio-economic classes with the supposed competition between biological species. Fascist ideologies such as Nazism likewise used garbled or inaccurate evolutionary ideas, such as ‘the survival of the fittest’, to justify violence. But in fact the competition in biological evolution is not between different species, but between variants of genes within a species – which does not resemble the supposed ‘class struggle’ at all. It can give rise to violence or other competition between species, but it can also produce cooperation (such as the symbiosis between flowers and insects) and all sorts of intricate combinations of the two.

Although Marx and the fascists assumed false theories of biological evolution, it is no accident that analogies between society and the biosphere are often associated with grim visions of society: the biosphere is a grim place. It is rife with plunder, deceit, conquest, enslavement, starvation and extermination. Hence those who think that cultural evolution is like that end up either opposing it (advocating a static society) or condoning that kind of immoral behaviour as necessary or inevitable.

Arguments by analogy are fallacies. Almost any analogy between any two things contains some grain of truth, but one cannot tell what that is until one has an independent explanation for what is analogous to what, and why. The main danger in the biosphere–culture analogy is that it encourages one to conceive of the human condition in a reductionist way that obliterates the high-level distinctions that are essential for understanding it – such as those between mindless and creative, determinism and choice, right and wrong. Such distinctions are meaningless at the level of biology. Indeed, the analogy is often drawn for the very purpose of debunking the common-sense idea of human beings as causal agents with the ability to make moral choices and to create new knowledge for themselves.

As I shall explain, although biological and cultural evolution are described by the same underlying theory, the mechanisms of transmission, variation and selection are all very different. That makes the resulting ‘natural histories’ different too. There is no close cultural analogue of a species, or of an organism, or a cell, or of sexual or asexual reproduction. Genes and memes are about as different as can be at the level of mechanisms, and of outcomes; they are similar only at the lowest level of explanation, where they are both replicators that embody knowledge and are therefore conditioned by the same fundamental principles that determine the conditions under which knowledge can or cannot be preserved, can or cannot improve.

Meme evolution

In the classic 1956 science-fiction story ‘Jokester’, by Isaac Asimov, the main character is a scientist studying jokes. He finds that, although most people do sometimes make witty remarks that are original, they never invent what he considers to be a fully fledged joke: a story with a plot and a punchline that causes listeners to laugh. Whenever they tell such a joke, they are merely repeating one that they have heard from someone else. So, where do jokes come from originally? Who creates them? The fictional answer given in ‘Jokester’ is far-fetched and need not concern us here. But the premise of the story is not so absurd: it really is plausible that some jokes were not created by anyone – that they evolved.

People tell each other amusing stories – some fictional, some factual. They are not jokes, but some become memes: they are interesting enough for the listeners to retell them to other people, and some of those people retell them in turn. But they rarely recite them word for word; nor do they preserve every detail of the content. Hence an often-retold story will come to exist in different versions. Some of those versions will be retold more often than others – in some cases because people find them amusing. When that is the main reason for retelling them, successive versions that remain in circulation will tend to be ever more amusing. So the conditions are there for evolution: repeated cycles of imperfect copying of information, alternating with selection. Eventually the story becomes amusing enough to make people laugh, and a fully fledged joke has evolved.

It is conceivable that a joke could evolve through variations that were not intended to improve upon the funniness. For example, people who hear a story can mishear or misunderstand aspects of it, or change it for pragmatic reasons, and in a small proportion of cases, by sheer luck, that will produce a funnier version of the story, which will then propagate better. If a joke has evolved in that way from a non-joke, it truly has no author. Another possibility is that most of the people who altered the amusing story on its way to becoming a joke designed their contributions, using creativity to make it funnier intentionally. In such cases, although the joke was indeed created by variation and selection, its funniness was the result of human creativity. In that case it would be misleading to say that ‘no one created it.’ It had many co-authors, each of whom contributed creative thought to the outcome. But it may still be that literally no one understands why the joke is as funny as it is, and hence that no one could create another joke of similar quality at will.

Although we do not know exactly how creativity works, we do know that it is itself an evolutionary process within individual brains. For it depends on conjecture (which is variation) and criticism (for the purpose of selecting ideas). So, somewhere inside brains, blind variations and selections are adding up to creative thought at a higher level of emergence.

The idea of memes has come in for a great deal of radical, and in my view mistaken, criticism to the effect that it is vague and pointless, or else tendentious. For example, when the ancient Greek religion was suppressed, but the stories of its gods continued to be told, though now only as fiction, were those stories still the same memes despite now causing new behaviours? When Newton’s laws were translated into English from the original Latin, they caused different words to be spoken and written. Were they the same memes? But in fact such questions cast no doubt on the existence of memes, nor on the usefulness of the concept. It is like the controversy about which objects in the solar system should be called ‘planets’. Is Pluto a ‘real’ planet even though it is smaller than some of the moons in our solar system? Is Jupiter really not a planet but an un-ignited star? It is not important. What is important is what is really there. And memes are really there, regardless of what we call them or how we classify them. Just as the basic theory of genes was developed long before the discovery of DNA, so today, without knowing how ideas are stored in brains, we do know that some ideas can be passed from one person to another and affect people’s behaviour. Memes are those ideas.

Another line of criticism is that memes, unlike genes, are not stored in identical physical forms in every holder. But, as I shall explain, that does not necessarily make it impossible for memes to be transmitted ‘faithfully’ in the sense that matters for evolution. It is indeed meaningful to think of memes as retaining their identity as they pass from one holder to the next.

Just as genes often work together in groups to achieve what we might think of as a single adaptation, so there are memeplexes consisting of several ideas which can, alternatively, be thought of as a single more complex idea, such as quantum theory or neo-Darwinism. So it does not matter if we refer to a memeplex as a meme, just as it does not matter if we refer to quantum theory as a single theory or a group of theories. However, ideas, including memes, cannot be indefinitely analysed into sub-memes, because there comes a point where replacing a meme by part of itself would result in its not being copied. So, for instance, ‘2 + 3 = 5’ is not a meme, because it does not have what it takes to cause itself reliably to be copied, except under circumstances which would also copy some theory of arithmetic with universal reach, which itself could not be transmitted without also transmitting the knowledge that 2 + 3 = 5.

Laughing at a joke and retelling it are both behaviours caused by the joke, but we often do not know why we are enacting them. That reason is objectively there in the meme, but we do not know it. We may try to guess, but our guess will not necessarily be true. For instance, we may guess that the humour in a particular joke lay in the unexpectedness of its punchline. But further experience with the same joke may reveal that it remains funny when we hear it again. In such a case, we are in the counter-intuitive (but common) position of having been mistaken about the reason for our own behaviour.

The same sort of thing happens with rules of grammar. We say, ‘I am learning to play the piano’ (in British English), but never ‘I am learning to play the baseball.’ We know how to form such sentences correctly, but, until we think about it, very few of us know that the inexplicit grammatical rule we are following even exists, let alone what it is. In American English the rule is slightly different, so the phrase ‘learning to play piano’ is acceptable. We may wonder why, and guess that the British are more fond of the definite article. But, again, that is not the explanation: in British English a patient is ‘in hospital’, and in American English ‘in the hospital’.

The same is true of memes in general: they implicitly contain information that is not known to the holders, but which nevertheless causes the holders to behave alike. Hence, just as native English speakers may be mistaken about why they have said ‘the’ in a given sentence, people enacting all sorts of other memes often give false explanations, even to themselves, of why they are behaving in that way.

Like genes, all memes contain knowledge (often inexplicit) of how to cause their own replication. This knowledge is encoded in strands of DNA or remembered by brains respectively. In both cases, the knowledge is adapted to causing itself to be replicated: it causes that more reliably than nearly all its variants do. In both cases, this adaptation is the outcome of alternating rounds of variation and selection.

However, the logic of the copying mechanism is very different for genes and memes. In organisms that reproduce by dividing, either all the genes are copied into the next generation or (if the individual fails to reproduce) none are. In sexual reproduction, a full complement of genes randomly chosen from both parents is copied, or none are. In all cases, the DNA duplication process is automatic: genes are copied indiscriminately. One consequence is that some genes can be replicated for many generations without ever being ‘expressed’ (causing any behaviour) at all. Whether your parents ever broke a bone or not, genes for repairing broken bones will (barring unlikely mutations) be passed on to you and your descendants.

The situation faced by memes is utterly different. Each meme has to be expressed as behaviour every time it is replicated. For it is that behaviour, and only that behaviour (given the environment created by all the other memes), that effects the replication. That is because a recipient cannot see the representation of the meme in the holder’s mind. A meme cannot be downloaded like a computer program. If it is not enacted, it will not be copied.

The upshot of this is that memes necessarily become embodied in two different physical forms alternately: as memories in a brain, and as behaviour:


A meme exists in a brain form and a behaviour form, and each is copied to the other.

Each of the two forms has to be copied (specifically, translated into the other form) in each meme generation. (Meme ‘generations’ are simply successive instances of copying to another individual.) Technology can add further stages to a meme’s life cycle. For instance, the behaviour may be to write something down – thus embodying the meme in a third physical form, which may later cause a person who reads it to enact other behaviour, which then causes the meme to appear in someone’s brain. But all memes must have at least two physical forms.

In contrast, for genes the replicator exists in only one physical form – the DNA strand (of a germ cell). Even though it may be copied to other locations in the organism, translated into RNA, and expressed as behaviour, none of those forms is a replicator. The idea that the behaviour might be a replicator is a form of Lamarckism, since it implies that behaviours that had been modified by circumstances would be inherited.


A gene exists in only one physical form, which is copied.

Because of the alternating physical forms of a meme, it has to survive two different, and potentially unrelated, mechanisms of selection in every generation. The brain-memory form has to cause the holder to enact the behaviour; and the behaviour form has to cause the new recipient to remember it – and to enact it.

So, for example, although religions prescribe behaviours such as educating one’s children to adopt the religion, the mere intention to transmit a meme to one’s children or anyone else is quite insufficient to make that happen. That is why the overwhelming majority of attempts to start a new religion fail, even if the founder members try hard to propagate it. In such cases, what has happened is that an idea that people have adopted has succeeded in causing them to enact various behaviours including ones intended to cause their children and others to do the same – but the behaviour has failed to cause the same idea to be stored in the minds of those recipients.

The existence of long-lived religions is sometimes explained from the premise that ‘children are gullible’, or that they are ‘easily frightened’ by tales of the supernatural. But that is not the explanation. The overwhelming majority of ideas simply do not have what it takes to persuade (or frighten or cajole or otherwise cause) children or anyone else into doing the same to other people. If establishing a faithfully replicating meme were that easy, the whole adult population in our society would be proficient at algebra, thanks to the efforts made to teach it to them when they were children. To be exact, they would all be proficient algebra teachers.

To be a meme, an idea has to contain quite sophisticated knowledge of how to cause humans to do at least two independent things: assimilate the meme faithfully, and enact it. That some memes can replicate themselves with great fidelity for many generations is a token of how much knowledge they contain.

The selfish meme

If a gene is in a genome at all, then, when suitable circumstances arise, it will definitely be expressed as an enzyme, as I described in Chapter 6, and will then cause its characteristic effects. Nor can it be left behind if the rest of its genome is successfully replicated. But merely being present in a mind does not automatically get a meme expressed as behaviour: the meme has to compete for that privilege with other ideas – memes and non-memes, about all sorts of subjects – in the same mind. And merely being expressed as behaviour does not automatically get the meme copied into a recipient along with other memes: it has to compete for the recipients’ attention and acceptance with all sorts of behaviours by other people, and with the recipient’s own ideas. All that is in addition to the analogue of the type of selection that genes face, each meme competing with rival versions of itself across the population, perhaps by containing the knowledge for some useful function.

Memes are subject to all sorts of random and intentional variation in addition to all that selection, and so they evolve. So to this extent the same logic holds as for genes: memes are ‘selfish’. They do not necessarily evolve to benefit their holders, or their society – or, again, even themselves, except in the sense of replicating better than other memes. (Though now most other memes are their rivals, not just variants of themselves.) The successful meme variant is the one that changes the behaviour of its holders in such a way as to make itself best at displacing other memes from the population. This variant may well benefit its holders, or their culture, or the species as a whole. But if it harms them, or destroys them, it will spread anyway. Memes that harm society are a familiar phenomenon. You need only consider the harm done by adherents of political views, or religions, that you especially abhor. Societies have been destroyed because some of the memes that were best at spreading through the population were bad for a society. I shall discuss one example in Chapter 17. And countless individuals have been harmed or killed by adopting memes that were bad for them – such as irrational political ideologies or dangerous fads. Fortunately, in the case of memes, that is not the whole story. To understand the rest of the story, we have to consider the basic strategies by which memes cause themselves to be faithfully replicated.

Static societies

As I have explained, a human brain – quite unlike a genome – is itself an arena of intense variation, selection and competition. Most ideas within a brain are created by it for the very purpose of trying them out in imagination, criticizing them, and varying them until they meet the person’s preferences. In other words, meme replication itself involves evolution, within individual brains. In some cases there can be thousands of cycles of variation and selection before any of the variants is ever enacted. Then, even after a meme has been copied into a new holder, it has not yet completed its life cycle. It still has to survive a further selection process, namely the holder’s choice of whether to enact it or not.

Some of the criteria that a mind uses to make such choices are themselves memes. Some are ideas that it has created for itself (by altering memes, or otherwise), and which will never exist in any other mind. Such ideas are potentially highly variable between different people, yet they can decisively affect whether any given meme does or does not survive via a given person.

Since a person can enact and transmit a meme soon after receiving it, a meme generation can be much shorter than a human generation. And many cycles of variation and selection can take place inside the minds concerned even during one meme generation. Also, memes can be passed to people other than the holders’ biological descendants. Those factors make meme evolution enormously faster than gene evolution, which partly explains how memes can contain so much knowledge. Hence the frequently cited metaphor of the history of life on Earth, in which human civilization occupies only the final ‘second’ of the ‘day’ during which life has so far existed, is misleading. In reality, a substantial proportion of all evolution on our planet to date has occurred in human brains. And it has barely begun. The whole of biological evolution was but a preface to the main story of evolution, the evolution of memes.

But, for the same reason, on the face of it meme replication is inherently less reliable than gene replication. Since the inexplicit content of memes cannot be literally copied but has to be guessed from the holders’ behaviour, and since a meme can be subjected to large intentional variations inside every holder, it could be considered something of a miracle that any meme manages to be transmitted faithfully even once. And indeed the survival strategies of all long-lived memes are dominated by this problem.

Another way of stating the problem is that people think and try to improve upon their ideas – which entails changing them. A long-lived meme is an idea that runs that gauntlet again and again, and survives. How is that possible?

The post-Enlightenment West is the only society in history that for more than a couple of lifetimes has ever undergone change rapid enough for people to notice. Short-lived rapid changes have always happened: famines, plagues and wars have begun and ended; maverick kings have attempted radical change. Occasionally empires were rapidly created or whole civilizations were rapidly destroyed. But, while a society lasted, all important areas of life seemed changeless to the participants: they could expect to die under much the same moral values, personal lifestyles, conceptual framework, technology and pattern of economic production as they were born under. And, of the changes that did occur, few were for the better. I shall call such societies ‘static societies’: societies changing on a timescale unnoticed by the inhabitants. Before we can understand our unusual, dynamic sort of society, we must understand the usual, static sort.

For a society to be static, all its memes must be unchanging or changing too slowly to be noticed. From the perspective of our rapidly changing society, such a state of affairs is hard even to imagine. For instance, consider an isolated, primitive society that has, for whatever reason, remained almost unchanged for many generations. Why? Quite possibly no one in the society even wants it to change, because they can conceive of no other way of life. Nevertheless, its members are not immune from pain, hunger, grief, fear or other forms of physical and mental suffering. They try to think of ideas to alleviate some of that suffering. Some of those ideas are original, and occasionally one of them would actually help. It need be only a small, tentative improvement: a way of hunting or growing food with slightly less effort, or of making slightly better tools; a better way of recording debts or laws; a subtle change in the relationship between husband and wife, or between parent and child; a slightly different attitude towards the society’s rulers or gods. What will happen next?

The person with that idea may well want to tell other people. Those who believe the idea will see that it could make life a little less nasty, brutish and short. They will tell their families and friends, and they theirs. This idea will be competing in people’s minds with other ideas about how to make life better, most of them presumably false. But suppose, for the sake of argument, that this particular true idea happens to be believed, and spreads through the society.

Then the society will have been changed. It may not have changed very much, but this was merely the change caused by a single person, thinking of a single idea. So multiply all that by the number of thinking minds in the society, and by a lifetime’s worth of thought in each of them, and let this continue for only a few generations, and the result is an exponentially increasing, revolutionary force transforming every aspect of the society.

But in a static society that beginning of infinity never happens. Despite the fact that I have assumed nothing other than that people try to improve their lives, and that they cannot transmit their ideas perfectly, and that information subject to variation and selection evolves, I have entirely failed to imagine a static society in this story.

For a society to be static, something else must be happening as well. One thing my story did not take into account is that static societies have customs and laws – taboos – that prevent their memes from changing. They enforce the enactment of the existing memes, forbid the enactment of variants, and suppress criticism of the status quo. However, that alone could not suppress change. First, no enactment of a meme is completely identical to that of the previous generation. It is infeasible to specify every aspect of acceptable behaviour with perfect precision. Second, it is impossible to tell in advance which small deviations from traditional behaviour would initiate further changes. Third, once a variant idea has begun to spread to even one more person – which means that people are preferring it – preventing it from being transmitted further is extremely difficult. Therefore no society could remain static solely by suppressing new ideas once they have been created.

That is why the enforcement of the status quo is only ever a secondary method of preventing change – a mopping-up operation. The primary method is always – and can only be – to disable the source of new ideas, namely human creativity. So static societies always have traditions of bringing up children in ways that disable their creativity and critical faculties. That ensures that most of the new ideas that would have been capable of changing the society are never thought of in the first place.

How is this done? The details are variable and not relevant here, but the sort of thing that happens is that people growing up in such a society acquire a set of values for judging themselves and everyone else which amounts to ridding themselves of distinctive attributes and seeking only conformity with the society’s constitutive memes. They not only enact those memes: they see themselves as existing only in order to enact them. So, not only do such societies enforce qualities such as obedience, piety and devotion to duty, their members’ sense of their own selves is invested in the same standards. People know no others. So they feel pride and shame, and form all their aspirations and opinions, by the criterion of how thoroughly they subordinate themselves to the society’s memes.

How do memes ‘know’ how to achieve all such complex, reproducible effects on the ideas and behaviour of human beings? They do not, of course, know: they are not sentient beings. They merely contain that knowledge implicitly. How did they come by that knowledge? It evolved. The memes exist, at any instant, in many variant forms, and those are subject to selection in favour of faithful replication. For every long-lived meme of a static society, millions of variants of it will have fallen by the wayside because they lacked that tiny extra piece of information, that extra degree of ruthless efficiency in preventing rivals from being thought of or acted upon, that slight advantage in psychological leverage, or whatever it took to make it spread through the population better than its rivals and, once it was prevalent, to get it copied and enacted with just that extra degree of fidelity. If ever a variant happened to be a little better at inducing behaviour with those self-replicating properties, it soon became prevalent. As soon as it did, there were again many variants of that variant, which were again subject to the same evolutionary pressure. Thus, successive versions of the meme accumulated knowledge that enabled them ever more reliably to inflict their characteristic style of damage on their human victims. Like genes, they may also confer benefits, though, even then, they are unlikely to do so optimally. Just as genes for the eye implicitly ‘know’ the laws of optics, so the long-lived memes of a static society implicitly possess knowledge of the human condition, and use it mercilessly to evade the defences and exploit the weaknesses of the human minds that they enslave.

A remark about timescales: Static societies, by this definition, are not perfectly unchanging. They are static on the timescale that humans can notice; but memes cannot prevent changes that are slower than that. So meme evolution still occurs in static societies, but too slowly for most members of the society to notice, most of the time. For instance, palaeontologists examining tools from the Old Stone Age cannot date them, by their shapes, to an accuracy better than many thousands of years, because tools at that time simply did not improve any faster than that. (Note that this is still much faster than biological evolution.) Examining a tool from the static society of ancient Rome or Egypt, one may be able to date it by its technology alone to the nearest century, say. But historians in the future examining cars and other technological artefacts of today will easily be able to date them to the nearest decade – and in the case of computer technology to the nearest year or less.

Meme evolution tends towards making memes static, but not necessarily whole societies. Like genes, memes do not evolve to benefit the group. Nevertheless, just as gene evolution can create long-lasting organisms and confer some benefits on them, so it is not surprising that meme evolution can sometimes create static societies, cooperate to keep them static, and help them to function by embodying truths. It is also not surprising that memes are often useful (though seldom optimally) to their holders. Just as organisms are the tools of genes, so individuals are used by memes to achieve their ‘purpose’ of spreading themselves through the population. And, to do this, memes sometimes confer benefits. One difference from the biological case, however, is that, while organisms are nothing but the slaves of all their genes, memes only ever control part of a person’s thinking, even in the most slavishly static of societies. That is why some people use the metaphor of memes as viruses – which control part of the functionality of cells to propagate themselves. Some viruses do just install themselves into the host’s DNA and do little else except participate in being copied from then on – but that is unlike memes, which must cause their distinctive behaviours and use knowledge to cause their own copying. Other viruses destroy their host cell – just as some memes destroy their holders: when someone commits suicide in a newsworthy way, there is often a spate of ‘copycat suicides’.

The overarching selection pressure on memes is towards being faithfully replicated. But, within that, there is also pressure to do as little damage to the holder’s mind as possible, because that mind is what the human uses to be long-lived enough to be able to enact the meme’s behaviours as much as possible. This pushes memes in the direction of causing a finely tuned compulsion in the holder’s mind: ideally, this would be just the inability to refrain from enacting that particular meme (or memeplex). Thus, for example, long-lived religions typically cause fear of specific supernatural entities, but they do not cause general fearfulness or gullibility, because that would both harm the holders in general and make them more susceptible to rival memes. So the evolutionary pressure is for the psychological damage to be confined to a relatively narrow area of the recipients’ thinking, but to be deeply entrenched, so that the recipients find themselves facing a large emotional cost if they subsequently consider deviating from the meme’s prescribed behaviours.

A static society forms when there is no escape from this effect: all significant behaviour, all relationships between people, and all thoughts are subordinated to causing faithful replication of the memes. In all areas controlled by the memes, no critical faculties are exercised. No innovation is tolerated, and almost none is attempted. This destruction of human minds makes static societies almost unimaginable from our perspective. Countless human beings, hoping throughout lifetimes, and for generations, for their suffering to be relieved, not only fail to make progress in realizing any such hope: they largely fail even to try to make any, or even to think about trying. If they do see an opportunity, they reject it. The spirit of creativity with which we are all born is systematically extinguished in them before it can ever create anything new.

A static society involves – in a sense consists of – a relentless struggle to prevent knowledge from growing. But there is more to it than that. For there is no reason to expect that a rapidly spreading idea, if one did happen to arise in a static society, would be true or useful. That is another aspect missing from my story of the static society above. I assumed that the change would be for the better. It might not have been, especially as the lack of critical sophistication in a static society would leave people vulnerable to false and harmful ideas from which their taboos did not protect them. For instance, when the Black Death plague destabilized the static societies of Europe in the fourteenth century, the new ideas for plague-prevention that spread best were extremely bad ones. Many people decided that this was the end of the world, and that therefore attempting any further earthly improvements was pointless. Many went out to kill Jews or ‘witches’. Many crowded together in churches and monasteries to pray (thus unwittingly facilitating the spread of the disease, which was carried by fleas). A cult called the Flagellants arose, whose members devoted their lives to flogging themselves, and to preaching all the above measures, in order to prove to God that his children were sorry. All these ideas were functionally harmful as well as factually false, and were eventually suppressed by the authorities in their drive to return to stasis.

Thus, ironically, there is much truth in the typical static-society fear that any change is much more likely to do harm than good. A static society is indeed in constant danger of being harmed or destroyed by a newly arising dysfunctional meme. However, in the aftermath of the Black Death a few true and functional ideas did also spread, and may well have contributed to ending that particular static society in an unusually good way (with the Renaissance).

Static societies survive by effectively eliminating the type of evolution that is unique to memes, namely creative variation intended to meet the holders’ individual preferences. In the absence of that, meme evolution resembles gene evolution more closely, and some of the grim conclusions of the naive analogies between them apply after all. Static societies do tend to settle issues by violence, and they do tend to sacrifice the welfare of individuals for the ‘good’ of (that is to say, for the prevention of changes in) society. I mentioned that people who rely on such analogies end up either advocating a static society or condoning violence and oppression. We now see that those two responses are essentially the same: oppression is what it takes to keep a society static; oppression of a given kind will not last long unless the society is static.

Since the sustained, exponential growth of knowledge has unmistakable effects, we can deduce without historical research that every society on Earth before the current Western civilization has either been static or has been destroyed within a few generations. The golden ages of Athens and Florence are examples of the latter, but there may have been many others. This directly contradicts the widely held belief that individuals in primitive societies were happy in a way that has not been possible since – that they were unconstrained by social convention and other imperatives of civilization, and hence were able to achieve self-expression and fulfilment of their needs and desires. But primitive societies (including tribes of hunter-gatherers) must all have been static societies, because if ever one ceased to be static it would soon cease to be primitive, or else destroy itself by losing its distinctive knowledge. In the latter case, the growth of knowledge would still be inhibited by the raw violence which would immediately replace the static society’s institutions. For once violence is mediating changes, they will typically not be for the better. Since static societies cannot exist without effectively extinguishing the growth of knowledge, they cannot allow their members much opportunity to pursue happiness. (Ironically, creating knowledge is itself a natural human need and desire, and static societies, however primitive, ‘unnaturally’ suppress it.) From the point of view of every individual in such a society, its creativity-suppressing mechanisms are catastrophically harmful. Every static society must leave its members chronically baulked in their attempts to achieve anything positive for themselves as people, or indeed anything at all, other than their meme-mandated behaviours. It can perpetuate itself only by suppressing its members’ self-expression and breaking their spirits, and its memes are exquisitely adapted to doing this.

Dynamic societies

But our society (the West) is not a static society. It is the only known instance of a long-lived dynamic (rapidly changing) society. It is unique in history for its ability to mediate long-term, rapid, peaceful change and improvement, including improvements in the broad consensus about values and aims, as I described in Chapter 13. This has been made possible by the emergence of a radically different class of memes which, though still ‘selfish’, are not necessarily harmful to individuals.

To explain the nature of these new memes, let me pose the question: what sort of meme can cause itself to be replicated for long periods in a rapidly changing environment? In such an environment, people are continually being faced with unpredictable problems and opportunities. Hence their needs and wishes are changing unpredictably too. How can a meme remain unchanged under such a regime? The memes of a static society remain unchanged by effectively eliminating all the individuals’ choices: people choose neither which ideas to acquire nor which to enact. Those memes also combine to make the society static, so that people’s circumstances vary as little as possible. But once the stasis has broken down, and people are choosing, they will choose, in part, according to their individual circumstances and ideas, in which case memes will face selection criteria that vary unpredictably from recipient to recipient as well as over time.

To be transferred to a single person, a meme need seem useful only to that person. To be transferred to a group of similar people under unchanging circumstances, it need be only a parochial truth. But what sort of idea is best suited to getting itself adopted many times in succession by many people who have diverse, unpredictable objectives? A true idea is a good candidate. But not just any truth will do. It must seem useful to all those people, for it is they who will be choosing whether to enact it or not. ‘Useful’ in this context does not necessarily mean functionally useful: it refers to any property that can make people want to adopt an idea and enact it, such as being interesting, funny, elegant, easily remembered, morally right and so on. And the best way to seem useful to diverse people under diverse, unpredictable circumstances is to beuseful. Such an idea is, or embodies, a truth in the broadest sense: factually true if it is an assertion of fact, beautiful if it is an artistic value or behaviour, objectively right if it is a moral value, funny if it is a joke, and so on.

The ideas with the best chance of surviving through many generations of change are truths with reach – deep truths. People are fallible; they often have preferences for false, shallow, useless or morally wrong ideas. But which false ideas they prefer differs from one person to another, and changes with time. Under changed circumstances, a specious falsehood or parochial truth can survive only by luck. But a true, deep idea has an objective reason to be considered useful by people with diverse purposes over long periods. For instance, Newton’s laws are useful for building better cathedrals, but also for building better bridges and designing better artillery. Because of this reach, they get themselves remembered and enacted by all sorts of people, many of them vehemently opposed to each other’s objectives, over many generations. This is the kind of idea that has a chance of becoming a long-lived meme in a rapidly changing society.

In fact such memes are not merely capable of surviving under rapidly changing criteria of criticism, they positively rely on such criticism for their faithful replication. Unprotected by any enforcement of the status quo or suppression of people’s critical faculties, they are criticized, but so are their rivals, and the rivals fare worse, and are not enacted. In the absence of such criticism, true ideas no longer have that advantage and can deteriorate or be superseded.

Rational and anti-rational memes

Thus, memes of this new kind, which are created by rational and critical thought, subsequently also depend on such thought to get themselves replicated faithfully. So I shall call them rational memes. Memes of the older, static-society kind, which survive by disabling their holders’ critical faculties, I shall call anti-rational memes. Rational and anti-rational memes have sharply differing properties, originating in their fundamentally different replication strategies. They are about as different from each other as they both are from genes.

If a certain type of hobgoblin has the property that, if children fear it, they will grow up to make their children fear it, then the behaviour of telling stories about that type of hobgoblin is a meme. Suppose it is a rational meme. Then criticism, over generations, will cast doubt on the story’s truth. Since in reality there are no hobgoblins, the meme might evolve away to extinction. Note that it does not ‘care’ if it goes extinct. Memes do what they have to do: they have no intentions, even about themselves. But there are also other paths that it might evolve down. It might become overtly fictional. Because rational memes must be seen as beneficial by the holders, those that evoke unpleasant emotions are at a disadvantage, so it may also evolve away from evoking terror and towards, for instance, being pleasantly thrilling – or else (if it settled on a genuine danger) exploring practicalities for the present and optimism for the future.

Now suppose it is an anti-rational meme. Evoking unpleasant emotions will then be useful in doing the harm that it needs to do – namely disabling the listener’s ability to be rid of the hobgoblin and entrenching the compulsion to think and therefore speak of it. The more accurately the hobgoblin’s attributes exploit genuine, widespread vulnerabilities of the human mind, the more faithfully the anti-rational meme will propagate. If the meme is to survive for many generations, it is essential that its implicit knowledge of these vulnerabilities be true and deep. But its overt content – the idea of the hobgoblin’s existence – need contain no truth. On the contrary, the non-existence of the hobgoblin helps to make the meme a better replicator, because the story is then unconstrained by the mundane attributes of any genuine menace, which are always finite and to some degree combatable. And that will be all the more so if the story can also manage to undermine the principle of optimism. Thus, just as rational memes evolve towards deep truths, anti-rational memes evolve away from them.

As usual, mixing the above two replication strategies does no good. If a meme contains true and beneficial knowledge for the recipient, but disables the recipient’s critical faculties in regard to itself, then the recipient will be less able to correct errors in that knowledge, and so will reduce the faithfulness of transmission. And if a meme relies on the recipients’ belief that it is beneficial, but it is not in fact beneficial, then that increases the chance that the recipient will reject it or refuse to enact it.

Similarly, a rational meme’s natural home is a dynamic society – more or less any dynamic society – because there the tradition of criticism (optimistically directed at problem-solving) will suppress variants of the meme with even slightly less truth. Moreover, the rapid progress will subject these variants to continually varying criteria of criticism, which again only deeply true memes have a chance of surviving. An anti-rational meme’s natural home is a static society – not any static society, but preferably the one in which it evolved – for all the converse reasons. And therefore each type of meme, when present in a society that is broadly of the opposite kind, is less able to cause itself to be replicated.

The Enlightenment

Our society in the West became dynamic not through the sudden failure of a static society, but through generations of static-society-type evolution. Where and when the transition began is not very well defined, but I suspect that it began with the philosophy of Galileo and perhaps became irreversible with the discoveries of Newton. In meme terms, Newton’s laws replicated themselves as rational memes, and their fidelity was very high – because they were so useful for so many purposes. This success made it increasingly difficult to ignore the philosophical implications of the fact that nature had been understood in unprecedented depth, and of the methods of science and reason by which this had been achieved.

In any case, following Newton, there was no way of missing the fact that rapid progress was under way. (Some philosophers, notably Jean-Jacques Rousseau, did try – but only by arguing that reason was harmful, civilization bad and primitive life happy.) There was such an avalanche of further improvements – scientific, philosophical and political – that the possibility of resuming stasis was swept away. Western society would become the beginning of infinity or be destroyed. Nations beyond the West today are also changing rapidly, sometimes through the exigencies of warfare with their neighbours, but more often and even more powerfully by the peaceful transmission of Western memes. Their cultures, too, cannot become static again. They must either become ‘Western’ in their mode of operation or lose all their knowledge and thus cease to exist – a dilemma which is becoming increasingly significant in world politics.

Even in the West, the Enlightenment today is nowhere near complete. It is relatively advanced in a few, vital areas: the physical sciences and Western political and economic institutions are prime examples. In those areas ideas are now fairly open to criticism and experimentation, and to choice and change. But in many other areas memes are still replicated in the old manner, by means that suppress the recipients’ critical faculties and ignore their preferences. When girls strive to be ladylike and to meet culturally defined standards of shape and appearance, and when boys do their utmost to look strong and not to cry when distressed, they are struggling to replicate ancient ‘gender-stereotyping’ memes that are still part of our culture – despite the fact that explicitly endorsing them has become a stigmatized behaviour. Those memes have the effect of preventing vast ranges of ideas about what sort of life one should lead from ever crossing the holders’ minds. If their thoughts ever wander in the forbidden directions, they feel uneasiness and embarrassment, and the same sort of fear and loss of centredness as religious people have felt since time immemorial at the thought of betraying their gods. And their world views and critical faculties are left disabled in precisely such a way that they will in due course draw the next generation into the same pattern of thought and behaviour.

That anti-rational memes are still, today, a substantial part of our culture, and of the mind of every individual, is a difficult fact for us to accept. Ironically, it is harder for us than it would have been for the profoundly closed-minded people of earlier societies. They would not have been troubled by the proposition that most of their lives were spent enacting elaborate rituals rather than making their own choices and pursuing their own goals. On the contrary, the degree to which a person’s life was controlled by duty, obedience to authority, piety, faith and so on was the very measure by which people judged themselves and others. Children who asked why they were required to enact onerous behaviours that did not seem functional would be told ‘Because I say so’, and in due course they would give their children the same reply to the same question, never realizing that they were giving the full explanation. (This is a curious type of meme whose explicit content is true though its holders do not believe it.) But today, with our eagerness for change and our unprecedented openness to new ideas and to self-criticism, it conflicts with most people’s self-image that we are still, to a significant degree, the slaves of anti-rational memes. Most of us would admit to having a hang-up or two, but in the main we consider our behaviour to be determined by our own decisions, and our decisions by our reasoned assessment of the arguments and evidence about what is in our rational self-interest. This rational self-image is itself a recent development of our society, many of whose memes explicitly promote, and implicitly give effect to, values such as reason, freedom of thought, and the inherent value of individual human beings. We naturally try to explain ourselves in terms of meeting those values.

Obviously there is truth in this; but it is not the whole story. One need look no further than our clothing styles, and the way we decorate our homes, to find evidence. Consider how you would be judged by other people if you went shopping in pyjamas, or painted your home with blue and brown stripes. That gives a hint of the narrowness of the conventions that govern even these objectively trivial and inconsequential choices about style, and the steepness of the social costs of violating them. Is the same thing true of the more momentous patterns in our lives, such as careers, relationships, education, morality, political outlook and national identity? Consider what we should expect to happen when a static society is gradually switching from anti-rational to rational memes.

Such a transition is necessarily gradual, because keeping a dynamic society stable requires a great deal of knowledge. Creating that knowledge, starting with only the means available in a static society – namely small amounts of creativity and knowledge, many misconceptions, the blind evolution of memes, and trial and error – must necessarily take time.

Moreover, the society has to continue to function throughout. But the coexistence of rational and anti-rational memes makes this transition unstable. Memes of each type cause behaviours that impede the faithful replication of the other: to replicate faithfully, anti-rational memes need people to avoid thinking critically about their choices, while rational memes need people to think as critically as possible. That means that no memes in our society replicate as reliably as the most successful memes of either a very static society or an (as yet hypothetical) fully dynamic society. This causes a number of phenomena that are peculiar to our transitional era.

One of them is that some anti-rational memes evolve against the grain, towards rationality. An example is the transition from an autocratic monarchy to a ‘constitutional monarchy’, which has played a positive role in some democratic systems. Given the instability that I have described, it is not surprising that such transitions often fail.

Another is the formation within the dynamic society of anti-rational subcultures. Recall that anti-rational memes suppress criticism selectively and cause only finely tuned damage. This makes it possible for the members of an anti-rational subculture to function normally in other respects. So such subcultures can survive for a long time, until they are destabilized by the haphazard effects of reach from other fields. For example, racism and other forms of bigotry exist nowadays almost entirely in subcultures that suppress criticism. Bigotry exists not because it benefits the bigots, but despite the harm they do to themselves by using fixed, non-functional criteria to determine their choices in life.

Present-day methods of education still have a lot in common with their static-society predecessors. Despite modern talk of encouraging critical thinking, it remains the case that teaching by rote and inculcating standard patterns of behaviour through psychological pressure are integral parts of education, even though they are now wholly or partly renounced in explicit theory. Moreover, in regard to academic knowledge, it is still taken for granted, in practice, that the main purpose of education is to transmit a standard curriculum faithfully. One consequence is that people are acquiring scientific knowledge in an anaemic and instrumental way. Without a critical, discriminating approach to what they are learning, most of them are not effectively replicating the memes of science and reason into their minds. And so we live in a society in which people can spend their days conscientiously using laser technology to count cells in blood samples, and their evenings sitting cross-legged and chanting to draw supernatural energy out of the Earth.

Living with memes

Existing accounts of memes have neglected the all-important distinction between the rational and anti-rational modes of replication. Consequently they end up missing most of what is happening, and why. Moreover, since the most obvious examples of memes are long-lived anti-rational memes and short-lived arbitrary fads, the tenor of such accounts is usually anti-meme, even when these accounts formally accept that the best and most valuable knowledge also consists of memes.

For example, the psychologist Susan Blackmore, in her book The Meme Machine, attempts to provide a fundamental explanation of the human condition in terms of meme evolution. Now, memes are indeed integral to the explanation for the existence of our species – though, as I shall explain in the next chapter, I believe that the specific mechanism she proposes would not have been possible. But, crucially, Blackmore downplays the element of creativity both in the replication of memes and in their origin. This leads her, for example, to doubt that technological progress is best explained as being due to individuals as the conventional narrative would have it. She regards it instead as meme evolution. She cites the historian George Basalla, whose book The Evolution of Technology denies ‘the myth of the heroic inventor’.

But that distinction between ‘evolution’ and ‘heroic inventors’ as being the agents of discovery makes sense only in a static society. There, most change is indeed brought about in the way that I guessed jokes might evolve, with no great creativity being exercised by any individual participant. But in a dynamic society, scientific and technological innovations are generally made creatively. That is to say, they emerge from individual minds as novel ideas, having acquired significant adaptations inside those minds. Of course, in both cases, ideas are built from previous ideas by a process of variation and selection, which constitutes evolution. But when evolution takes place largely within an individual mind, it is not meme evolution. It is creativity by a heroic inventor.

Worse, in regard to progress, Blackmore denies that there has been ‘progress towards anything in particular’ – that is to say, no progress towards anything objectively better. She recognizes only increasing complexity. Why? Because biological evolution does not have a ‘better’ or ‘worse’. This despite her own warning that memes and genes evolve differently. Again, her claim is largely true of static societies, but not of ours.

How should we understand the existence of the distinctively human emergent phenomena such as creativity and choice, in the light of the fact that part of our behaviour is caused by autonomous entities whose content we do not know? And, worse, given that we are liable to be systematically misled by those entities about the reasons for our own thoughts, opinions and behaviour?

The basic answer is that it should not come as a surprise that we can be badly mistaken in any of our ideas, even about ourselves, and even when we feel strongly that we are right. So we should respond no differently, in principle, from how we respond to the possibility of being in error for any other reason. We are fallible, but through conjecture, criticism and seeking good explanations we may correct some of our errors. Memes hide, but, just as with the optical blind spot, there is nothing to prevent our using a combination of explanation and observation to detect a meme and discover its implicit content indirectly.

For example, whenever we find ourselves enacting a complex or narrowly defined behaviour that has been accurately repeated from one holder to the next, we should be suspicious. If we find that enacting this behaviour thwarts our efforts to attain our personal objectives, or is faithfully continued when the ostensible justifications for it disappear, we should become more suspicious. If we then find ourselves explaining our own behaviour with bad explanations, we should become still more suspicious. Of course, at any given point we may fail either to notice these things or to discover the true explanation of them. But failure need not be permanent in a world in which all evils are due to lack of knowledge. We failed at first to notice the non-existence of a force of gravity. Now we understand it. Locating hang-ups is, in the last analysis, easier.

Another thing that should make us suspicious is the presence of the conditions for anti-rational meme evolution, such as deference to authority, static subcultures and so on. Anything that says ‘Because I say so’ or ‘It never did me any harm,’ anything that says ‘Let us suppress criticism of our idea because it is true,’ suggests static-society thinking. We should examine and criticize laws, customs and other institutions with an eye to whether they set up conditions for anti-rational memes to evolve. Avoiding such conditions is the essence of Popper’s criterion.

The Enlightenment is the moment at which explanatory knowledge is beginning to assume its soon-to-be-normal role as the most important determinant of physical events. At least it could be: we had better remember that what we are attempting – the sustained creation of knowledge – has never worked before. Indeed, everything that we shall ever try to achieve from now on will never have worked before. We have, so far, been transformed from the victims (and enforcers) of an eternal status quo into the mainly passive recipients of the benefits of relatively rapid innovation in a bumpy transition period. We now have to accept, and rejoice in bringing about, our next transformation: to active agents of progress in the emerging rational society – and universe.


Culture   A set of shared ideas that cause their holders to behave alike in some ways.

Rational meme   An idea that relies on the recipients’ critical faculties to cause itself to be replicated.

Anti-rational meme   An idea that relies on disabling the recipients’ critical faculties to cause itself to be replicated.

Static culture/society   One whose changes happen on a timescale longer than its members can notice. Such cultures are dominated by anti-rational memes.

Dynamic culture/society   One that is dominated by rational memes.


– Biological evolution was merely a finite preface to the main story of evolution, the unbounded evolution of memes.

– So was the evolution of anti-rational memes in static societies.


Cultures consist of memes, and they evolve. In many ways memes are analogous to genes, but there are also profound differences in the way they evolve. The most important differences are that each meme has to include its own replication mechanism, and that a meme exists alternately in two different physical forms: a mental representation and a behaviour. Hence also a meme, unlike a gene, is separately selected, at each replication, for its ability to cause behaviour and for the ability of that behaviour to cause new recipients to adopt the meme. The holders of memes typically do not know why they are enacting them: we enact the rules of grammar, for instance, much more accurately than we are able to state them. There are only two basic strategies of meme replication: to help prospective holders or to disable the holders’ critical faculties. The two types of meme – rational memes and anti-rational memes – inhibit each other’s replication and the ability of the culture as a whole to propagate itself. Western civilization is in an unstable transitional period between stable, static societies consisting of anti-rational memes and a stable dynamic society consisting of rational memes. Contrary to conventional wisdom, primitive societies are unimaginably unpleasant to live in. Either they are static, and survive only by extinguishing their members’ creativity and breaking their spirits, or they quickly lose their knowledge and disintegrate, and violence takes over. Existing accounts of memes fail to recognize the significance of the rational/anti-rational distinction and hence tend to be implicitly anti-meme. This is tantamount to mistaking Western civilization for a static society, and its citizens for the crushed, pessimistic victims of memes that the members of static societies are.