Memes, minds and imagination

Draft paper for forthcoming book, edited by Ilona Roth,
based on the symposium
Imaginative Minds, British Academy April 30-May 1 2004

Draft Only – Do not quote

The human imagination is a fount of creativity.  We can imagine things that we have never seen or heard, imagine multiple possible outcomes of events that have not yet happened, and even imagine things that are impossible in the real world. Sometimes we put these imaginings into action and change the world we live in. I would like to explore both the nature and origins of this capacity for imagination and question two common assumptions: first, that creative imagination evolved because it serves some biological function; second, that imagining entails consciousness, or even that consciousness is the driving force behind imagination and creativity. I shall argue that both of these assumptions are false.

Humans appear to be unique in their capacity for creativity and imagination. This is not to say that these abilities come out of nowhere. Clearly they do not. Most animals are capable of learning, and arguably both classical and operant conditioning involve some capacity for imagination. For example, my cat deciding whether to go out has learned what lies outside, the blackbird on the lawn has learned that the cats lurk behind the hedge, and even the humble snail has a simple mental map of the garden. Whether they can all be said to “imagine” the garden is a moot point, but they certainly use information about it for making decisions.

In a big step up from ordinary learning, some species can solve problems requiring insight or imagination. For example, New Caledonian crows use and make tools in the wild, and in captivity have been shown to solve problems requiring invention. One crow, provided with straight wire, spontaneously bent it into a hook to retrieve food from a narrow container (Weir, Chappell and Kacelnik 2002 ). Whether this counts as imagination or not is a tricky question, but apparently she was able to consider the consequences of using different shaped wires for the task. Although most primates are probably incapable of such feats, some apes use tools, and chimpanzees not only use and make tools but also have simple cultures, passing on ways of getting food or water, or using tools (Whiten et al 1999).

All these skills may have been available to our hominid ancestors, but modern humans are different. Turner (this symposium) calls this the “grand difference”. Our minds are constantly busy with ideas, suppositions, and plans. We use language with its infinite capacity for creative recombination. And we have created vast cultures that have not only spread around the whole planet but are so powerful that they are changing that planet itself and interfering with every ecosystem. Essential to this extraordinary development is not only the capacity for learning, but the ability to play with ideas in the mind – to try out new combinations and imagine their consequences. This is the sense in which I shall use the term “imagination”. Human culture depends on us having imaginative minds. But why do we have them?

Most theories start with a simple assumption; that imagination must have served some biological function. They can then argue that early hominids with a better capacity for imagination would have had a reproductive advantage and so passed on genes for that ability. In this way genes underlying imaginative capacity would have spread, leading ultimately to the human imaginative mind. Wilson claims that “innovation is a concrete biological process” (Wilson, 1998 p 236), and such arguments are common in evolutionary psychology (Pinker 1997). Mithen (1996 and this symposium) presents a theory of this kind.

Similar arguments can be applied to the products of imagination. Theories of gene-culture co-evolution treat human artefacts as aspects of their phenotype, and consider their adaptive value in terms of their effect on gene frequencies. For example, Cavalli-Sforza and Feldman think of “cultural activity as an extension of Darwinian fitness” (1981, p 362) and on Wilson’s famous “leash principle”, “genetic natural selection operates in such a way as to keep culture on a leash.” (Lumsden and Wilson 1981 p 13). More recently, Miller (2000) argues that the human mind was shaped by sexual selection, and treats cultural products as sexual displays. In this symposium, Boyer asks whether cultural products are themselves adaptations. The only alternative he offers is that they are mere by-products constrained only by relevance. There is, however, another possibility. This is that the “grand difference” is not explicable entirely in biological terms because we humans are the products of two replicators not one.

Memes as replicators

A replicator is information that is copied with variation and selection. Our most familiar replicator is the gene but, according to the principles of universal Darwinism, evolution must occur whenever there is a replicator (Dawkins 1976, Dennett 1995). In 1976 Dawkins speculated about the possibility of other replicators and pointed out that, in culture, information is copied between people by imitation and teaching. He called this new replicator the meme (to rhyme with gene). Examples are stories, songs, games, financial systems, scientific theories, music and the arts.

There has been some disagreement over whether memes fit the requirements for being a replicator. For example, Sperber (2000) queries the nature of the copying process, and Gil-White (2004) argues that memes are not replicated at all (see also Aunger 2000). Following Dawkins (1976) and Dennett (1995), I conclude that memes are replicators because they are copied with variation and selection. Variation occurs both through imprecise copying and by recombination and elaboration of the information that is copied. Selection occurs because some behaviours are more frequently copied than others. In other words imitation is selective.

The underlying principle of memetics is that memes (sometimes defined as “that which is imitated” but see Aunger 2001, Blackmore 1998) compete to be copied, with the successful ones shaping human minds and culture. Memes may be copied for any reason at all, including biological value, ease of copying, ease of remembering, emotional appeal or many other factors. As Dennett puts it “The first rule of memes, as it is for genes, is that replication is not necessarily for the good of anything; replicators flourish that are good at … replicating!” (Dennett, 1991, p 203). Those which harm their carriers or their genes may be likened to viruses; email viruses and chain letters are examples, but these are just at one end of a continuum which includes neutral memes and useful or life enhancing ones. All these compete to be stored in human brains or artefacts, and then to be copied again.

Memetics provides an explanation of human evolution that is quite different from purely gene based theories. It emphasises the importance of imitation and the power of the replicator. The hypothesis is that in human evolution everything changed when our ancestors became capable of imitation. This new skill let loose a second replicator; that is, the behaviours or artefacts that were copied – the memes.

The new replicator then began evolving in its own right and for no purpose other than itself. As Dawkins put it “Once this new evolution begins, it will in no necessary sense be subservient to the old.” (Dawkins, 1976, p 193-4). So once it appeared, this new replicator evolved, using the machinery provided by the first replicator, and influencing the further evolution of that machinery. Thus humans, alone on this planet, evolved as the product of two replicators not one.

In this view, the function of imagination cannot be understood without reference to memes; cultural products are not seen as by-products or as adaptive to genes, but as replicators in their own right. They can be thought of as a parasite or symbiont living and evolving along with the human brains that copy them and forcing those brains to adapt to their presence. This transforms our view of human nature and, I believe, makes far more sense of why we are so different from other species.

Dennett (1995) describes the importance of different replicators in his metaphor of the Tower of Generate-and-Test. On the ground floor are Darwinian creatures, selected by death; on the second floor are Skinnerian creatures whose behaviour is selected by learning; on the third floor are Popperian creatures whose ideas can be selected in their imagination; and on the top floor are Gregorian creatures whose memes are selected in culture. In each case there is a replicator and a selective process going on, and evolution occurs as a result, but the last step is a dramatic one because the replicator escapes from the individual out into the social world. Humans are the only species to have truly made the leap to being Gregorian creatures. They are the product of two interacting replicators, genes and memes.

This coevolution of memes and genes is a two-way process. Genes influence memes because they build the copying machinery of the human body and brain. But memes influence genes too, in a process I have referred to as memetic drive. Once a species is capable of imitation, memes start evolving; those of high fidelity, fecundity and longevity outperform the rest, leading to increasing numbers and varieties of memes, and changes in the dominant memes. This memetic evolution, with all its various products, changes the environment in which genetic selection takes place. Depending on the direction the memetic evolution happens to take, genes may be forced to follow, and in this sense they are driven by the memes (Blackmore 1999, 2001).

This theory differs from most theories of gene-culture co-evolution because of the final step. The process works like this. Before there were many memes, the major pressures on hominid genes came from the physical environment and from other individuals. But once memetic evolution took off, status, survival and reproductive chances were affected by the memes a person acquired. This would create a need for new strategies concerning whom to copy and whom to mate with. For example, it might pay to mate with those who were capable of imitating the currently most popular memes. So, if wearing skins as clothes was frequently copied (i.e. it was a successful meme) then people who were poor at copying clothes-wearing would be at a disadvantage. If there were genetic differences between people in how good they were at this skill then these genes would spread in the gene pool, gradually increasing people’s ability.

Note that wearing clothes has a biological advantage in terms of warmth and protection, so the spread of clothing and the ability to copy clothing could be explained biologically. However, the interesting point is this. Once the ability to imitate increases in the population, then more behaviours and artefacts can be copied. So memetic evolution increases. Inevitably people will copy all sorts of memes, not just clothes, including those that are biologically useful and those that are not. Unless certain memes are positively lethal to their carriers (and even then, under some circumstances), they may thrive along with the useful ones. In most theories of gene-culture evolution this is as far as the argument goes, but memetics takes a further step.

If acquiring the latest memes provides a genetic advantage, then genes for that ability will increase. This means an inherited improvement in the ability to copy whichever memes have been successful in the memetic competition. In other words, the direction of memetic evolution drives the direction taken by the genes. Genes are forced to build brains that can copy the most successful memes. Since memes thrive for memetic, not just biological, reasons this means we can only understand the design of the human brain by considering memes as well as genes. Examples of this process may include not only the evolution of the big brain and language (Blackmore 1999) but our enjoyment of art, science and religion.

The origins of art and music

There are many theories of the origins of art and almost all rely on biological functions of some kind. For example, Ramachandran and Hirstein (1999) propose eight laws of aesthetic experience, and these derive directly from the structure of perceptual systems that evolved for biological purposes. Zeki describes one of the functions of art as “an extension of the major function of the visual brain”, that is, a search for the enduring features of objects and situations (Zeki 1999 p 79), and Solso (2003) attributes artistic appreciation to a consciousness that evolved for other purposes.

Music provides an especially interesting example. Pinker declares that “As far as biological cause and effect are concerned, music is useless.” (Pinker 1997 p 528), and Dennett (1999) says we “cannot avoid the obligation to explain how such an expensive, time-consuming activity came to flourish in this cruel world”. Miller (2000) provides an explanation, citing music as an example of a sexual display, but this, in common with other biologically based theories, ignores the ways in which music itself evolves. This is the starting point for a memetic theory of the evolution of music.

Dennett (1999) imagines how music might have begun – a just-so story about the first ever infectious sounds.

One day one of our distant hominid ancestors sitting on a fallen log happened to start banging on with a stick—boom boom boom. For no good reason at all. This was just idle diddling, a byproduct, perhaps, of a slightly out-of-balance endocrine system. This was, you might say, mere nervous fidgeting, but the repetitive sounds striking his ears just happened to feel to him like a slight improvement on silence…

Now introduce some other ancestors who happen to see and hear this drummer. They might pay no attention, or be irritated enough to make him stop or drive him away, or they might, again for no reason, find their imitator-circuits tickled into action; they might feel an urge to drum along with musical Adam.

Dennett explains that it does not matter why either the first person, or the imitators did what they did – it might have been for good biological reasons or it might have been because of some quirk in the design of their brains or the weather that day. The important point is that the drumming was copied, and so with a community of other imitators around, the sounds began to spread. Among all the different drummings, some proved more infectious than others. It didn’t matter why  – maybe because they were easier, sounded nicer, or were less harmful – the point is that once they could be copied they were copied, and so the drumming virus was born.

Dennett supposes that soon some of the hominids begin humming, and humming memes spread in the meme pool. Then the number and variety of musical memes increases, and the competition hots up. Memes have to be more catchy, easier to hum, or more likely to gain attention, in order to find themselves copied. Individual brains change too because everyone now lives in a music filled culture and they learn to hum some of the tunes, with some learning faster and learning more tunes than others.

As Dennett himself says, this is only a Just-So story. We cannot know what really happened, but the point is this: once there were creatures capable of imitation something like this must have happened. Of all the many behaviours carried out, some would have been copied more than others. If any proved especially copyable, given the oddities of the hominid brain and the specifics of the environment, then those would spread, and so memetic evolution would be up and running.

The next step, which Dennett does not consider, is memetic drive. If drumming and humming became popular, and people who were good at it acquired status, then the pressures on hominid genes would change. It would now pay to have a brain that is good at copying drumming and humming, when previously it did not. Any genes that contribute to that ability are now favoured and so, gradually, hominid brains are redesigned. The co-evolutionary process then continues indefinitely. Improvements to the copying machinery mean that more sounds can be created and copied, and that in turn means further redesign and so on.

If this is how music evolved then we can easily understand why we modern humans have the sort of brains (and ears and hands) that help us enjoy making and listening to music. We are like that, not because music serves any biological function, but simply because, at some point in the past, musical memes infected our ancestors and helped to redesign their brains. Those brains are now designed to remember, hum, sing, play, and pass on music; they are skilled at mixing up all the fragments they hear to make new ones and at using the schemes and musical tricks they come across to develop them further. This is what it means to have a musical imagination, and we have it because of the replicator power of musical memes.

Another example of memetic drive in action could be the power of religious belief. It is a curious fact that humans seem to be naturally religious creatures (Boyer 2001, Newberg and D’Aquili 2001). Even today most people in the world believe in God. This is true even in highly educated and technologically advanced societies such as the USA. In contrast, Britain has one of the lowest levels of religious belief and observance, with just under half of people claiming to believe in God. Even so, this is a large proportion, and enormous amounts of money and effort are devoted to religious observance. The resources consumed are even larger if one includes cults, New Age groups and non-theistic religions as well. All this demands explanation.

Most theories of the origins of religious belief treat it as a natural consequence of having a brain designed for other purposes. For example, Pinker (1997) describes how religious concepts arise from our evolved perceptual capacities and from limitations in our understanding of the world. Boyer (2001) argues that religion consists of by-products of normal mental functioning evolved for other purposes. In contrast, Ehrlich argues that “Organized religion thus seems to have evolved to help stabilize hierarchical social structure.” (Ehrlich 2000, p 256). There may indeed be biological value in being religious; for example, membership of certain religious groups confers social advantages, reduces fear of death, and so on. However, on a memetic theory of the evolution of religiosity this is not essential.

Religions provide one of the most powerful examples of infectious memeplexes. In his analysis of “viruses of the mind” Dawkins (1993) uses Roman Catholicism as an example, pointing out all the tricks that this highly successful and long-lived memeplex uses to get itself copied and safely stored. Hidden within the complex of stories and dogmas are powerful instructions to pass on the whole package, both to one’s children and to others. This instruction is backed up by untestable threats and promises, including heaven, hell, and eternal damnation. Doubt is to be fought against and belief admired, which helps prevent intelligent children from questioning the whole idea. Giving money to the poor is encouraged, and so is giving money to the church itself. This makes possible the fabulous buildings, wonderful music, extravagant paintings, and other glories which instil awe and delight in church-goers – so encouraging them to spend time in church, to encourage others to come too, and to spread the memes still further.

Incidentally, music plays a significant role here. Having evolved a delight in music we enjoy singing and listening to others sing. The religion can then use this evolved capacity to spread itself by using the music to carry the viral words. We sing “Praise my soul the King of heaven” “Now thank we all our God” “All things bright and beautiful …” “Jesus Christ is risen today ….”. Curiously I have no trouble thinking up countless examples of hymns I learnt in my childhood. They are certainly long-lived memes.

Note that phrases like “the religion uses x” or “religions want x” are shorthand. They do not imply that the memes have plans or intentions – obviously they do not because memes are only the behaviours, words and sounds that are copied. These phrases can be unpacked as “religions that have x are copied more often than those that do not”. In this way memetics explains how and why the great religions of the world have survived so long and infected so many people. It can also explain how their power shifts with changing cultures, and why other newer religions are taking over in some places, including secular “religions” such as Transcendental Meditation, Landmark Education, and New Age beliefs.

Our early ancestors would not have had such complex and highly evolved religions, but the same principles apply. If some people adopted rituals to help hunting or relieve the pain of famine or death, and others copied them, then variations of these rituals would compete to be copied. The same would apply to any other aspect of religious behaviour or tradition. Some of these religious memes would thrive at the expense of less fit memes, and those successful ones would very slowly drive genes to provide machinery capable of copying them. In other words the human brain would gradually be redesigned for religious behaviour just as it was for language and musical ability. There need have been no biological advantage to religious behaviour at all. Whether there was or not is beside the point. The point is that the religious memes themselves could have forced our brains to end up the way they are.

I have given here some examples of how memes might have influenced the design of modern human brains. The end result is creative brains. That is, the effects of past memetic competition have brought about human brains that are especially good at taking on new memes, mixing them up and producing more. So the answer to the question why we have imaginative minds is that meme-gene coevolution designed them.

Consciousness and creativity

This memetic approach also provides a different view of the nature of human creativity. There is a common tendency to associate creativity with consciousness, or to talk about it as though it emerges from an inner conscious self. This is found in popular works, such as the numerous books with titles including “consciousness” and “creativity”, and self-help tapes and manuals advocating enhanced creative awareness. The same association can be found in philosophy and psychology. For example, Searle (1992) claims that consciousness is caused by brains and that it serves to increase creativity and flexibility. Mithen (1996) argues that consciousness plays an integrating role, bringing separate aspects of the mind together and so allowing for greater creativity.

This emphasis on consciousness meets with several problems, not least the fact that consciousness itself is impossible to define and poorly understood. The term is mostly used in contemporary science and philosophy to refer to subjective experience, or “what it’s like to be” (Blackmore 2003). This means it is hard to understand how subjective experiences could act as a force or have any function – a serious problem for theories of the evolution of consciousness, but also for any theory that makes consciousness a cause of creativity.

Attempts to link creative imagination to consciousness run into another problem if they imply that when we imagine something it must be “displayed” in consciousness, or be consciously visible to the mind’s eye, or in some other way come “into consciousness”. These phrases all imply versions of what Dennett (1991) calls the “Cartesian theatre” – the mythical time or place in the mind or brain where things come together and consciousness happens. This cannot exist, according to Dennett, not only because there is no observer inside the brain but because there is no centre of operations, nor indeed a centre of any kind. The brain is a massively parallel system with no special inner sanctum where information comes to be turned into the contents of consciousness.

The idea that consciousness is essential for creativity conflicts with another odd and interesting fact. This is that many creative writers, thinkers, scientists and artists, claim that their best work just “comes” to them. They have no idea how they do it, and indeed often feel that “they” did not really do it at all. It is as though the poem, the solution to the scientific problem, or the painting just shaped itself without any conscious effort, or even any awareness on the part of the creator. Some describe this feeling of total immersion as a state of “flow” (Csikszentmihalyi and Csikszentmihalyi 1988) in which the self seems to disappear. This kind of selfless creativity seems at odds with the idea that consciousness is the force behind creativity.

Universal Darwinism provides a completely different way of looking at creativity. In this new view, the process of copying with variation and selection is the only creative force in the universe. This simple iterative process, along with processes of self-organisation and random change, designed all the living things we see around us. It created human beings and gave us our imaginative minds. It also creates what we think of as human inventions. These emerge from memetic evolution; from the reiterative process of recombination and selective imitation of behaviours and artefacts. Some people are more creative than others because they are better at copying memes and combining them to form new ones, but not because they possess a specially creative conscious mind.

In this view we human beings are not creative because we have specially powerful conscious minds but because we are capable of selective imitation. We are the copying machines that form part of a new evolutionary process. Thus the true creative power behind human imagination is memetic evolution.

This view may seem depressing or dehumanising, with its emphasis on selective imitation and away from the power of consciousness. Yet it provides a unifying view of creative design. As information explodes, the web expands, and human life becomes ever more complex and full of cultural creations, we can see the same process at work as that which designed the living world. Everything was, and still is, designed by the power of that familiar mindless process, the evolutionary algorithm.



Aunger, R.A. (Ed) (2000) Darwinizing Culture: The Status of Memetics as a Science, Oxford University Press

Blackmore,S.J. 1998 Imitation and the definition of a meme. Journal of Memetics – Evolutionary Models of Information Transmission, 2.

Blackmore, S.J. (1999) The Meme Machine, Oxford, Oxford University Press.

Blackmore, S. (2001) Evolution and memes: The human brain as a selective imitation device. Cybernetics and Systems, 32, 225-255

Blackmore, S.J. (2003) Consciousness: An Introduction, London, Hodder and Stoughton; New York, Oxford University Press

Cavalli-Sforza, L.L. and Feldman, M.W. (1981) Cultural Transmission and Evolution: A quantitative approach. Princeton, NJ, Princeton University Press

Csikszentmihalyi, M. and Csikszentmihalyi, I.S. (Eds) (1988) Optimal Experience: Psychological Studies of Flow in Consciousness. Cambridge, Cambridge University Press

Dawkins, R. (1976) The Selfish Gene Oxford, Oxford University Press (new edition with additional material, 1989)

Dawkins, R. (1993) Viruses of the mind. In B.Dahlbohm (ed) Dennett and his Critics: Demystifying Mind. Oxford, Blackwell

Dennett, D. (1991). Consciousness Explained. Boston, Little, Brown

Dennett, D. (1995) Darwin’s Dangerous Idea, London, Penguin

Dennett, D. (1999) The evolution of culture. Charles Simonyi Lecture, Oxford, February 17 and at

Ehrlich, P.R. (2000) Human Natures: Genes, Cultures, and the Human Prospect Washington D.C., Island Press

Gil-White, F.J. (2004) Common misunderstandings of memes (and genes): The promise and the limits of the genetic analogy to cultural transmission processes. in S. Hurley and N. Chater, Perspectives on Imitation: From Mirror Neurons to Memes,MIT Press.

Lumsden, C.J. and Wilson, E.O. (1981) Genes, Mind and Culture. Cambridge, Mass., Harvard University Press.

Miller, G. (2000) The Mating Mind: How Sexual Choice Shaped the Evolution of Human Nature, London, Heinemann

Mithen, S. (1996) The Prehistory of the Mind. London, Thames and Hudson

Newberg, A. and D’Aquili, E. (2001) Why God Won’t Go Away: Brain Science and the Biology of Belief,  New York, Ballantine

Pinker, S. (1997) How the Mind Works. Penguin

Ramachandran,V.S. and Hirstein,W. 1999 The science of art: A neurological theory of aesthetic experience. Journal of Consciousness Studies, 6-7, 15-51

Solso, R.L. (2003) The Psychology of Art and the Evolution of the Conscious Brain, Cambridge, MA, MIT Press

Whiten, A., Goodall, J., McGrew, W.C., Nishida, T., Reynolds, V., Sugiyama, Y., Tutin, C.E.G., Wrangham, R.W. and Boesch, C. (1999) Cultures in chimpanzees. Nature, 399, 682-685

Wilson, E.O. (1998) Consilience: The Unity of Knowledge. New York, Knopf

Zeki, S. (1999) Art and the brain,  Journal of Consciousness Studies, 6-7, 76-96