The Meme Machine (23 page)

I do not wish to underestimate the importance of sociobiology and evolutionary psychology. In the next chapter I will consider some of their greatest successes in explaining human sexuality. But they are looking at only part of the picture. Certainly, much of our behaviour has been selected because it effectively propagates the genes on which it depends. But behaviour is also driven by memetic selection and has been selected because it effectively propagates the memes on which it depends.

I like to look at it this way. There are two replicators driving the evolution and design of our bodies, brains and behaviour. For some aspects of our lives the genes do most of the driving, and the role of the memes can be safely ignored. In these cases the gene–based approach of sociobiology and evolutionary psychology is a good approximation
(though still only an approximation), but in other cases the full picture can be understood only by considering both replicators. I will turn now to some of those other cases.

Memetic drive and Dennett’s tower

The two examples I have already given are fundamental to understanding human behaviour. They are the big brain and the evolution of language. I have argued that both depend on memetic driving, and I want now to explain this process further and put it in context. The most important step is to show how and why memetic driving is not just another form of evolution in the service of the genes. Unless this is true, memetics can still be reduced to sociobiology.

Memetic driving works like this. Once imitation arose three new processes could begin. First, memetic selection (that is the survival of some memes at the expense of others). Second, genetic selection for the ability to imitate the new memes (the best imitators of the best imitators have higher reproductive success). Third, genetic selection for mating with the best imitators.

The first step means that new ideas and behaviour begin to spread memetically, from making tools and pots, to dancing, singing and speech. The second step means that the people who are best able to pick up the new memes have more offspring who are also able to pick up the new memes. So everyone tends more and more to imitate the successful memes. The third step means that mate choice is also driven by the memes that are prevalent at the time. The consequence of these processes operating together is that the direction taken by memetic evolution affects the direction taken by the genes. This is memetic driving.

Memetic driving may look at first sight as though it is the same as what is known as the Baldwin effect but it is not and I must explain why.

The Baldwin effect was first described by the psychologist James Baldwin who referred to it as ‘a new factor in evolution’ (Baldwin 1896). It explains how intelligent behaviour, imitation, and learning can all affect selection pressure on the genes. As we have seen, there is no Lamarckian ‘inheritance of acquired characteristics’ in the sense of passing the results of learning on to the next generation through the genes. Behaviour, however, does have effects on natural selection.

Imagine, for example, a salamander–like creature that eats flies. Individuals that can reach highest get the most flies. Now imagine that one of them begins to jump. It gets more flies and soon any of its mates
that cannot jump at all begin to starve. Therefore, genes for good jumping or strong back legs spread in the gene pool and soon all the creatures are more like frogs. Jumping improves and the selection pressure now favours even higher leaps. Behaviour has, in a perfectly Darwinian fashion, affected selection.

Now imagine that the flies vary in appearance and goodness to eat. Let us suppose that striped flies are inedible while spotty flies are excellent food. Froglings that prefer spotty flies will be at an advantage and so the mechanisms required for preferring spotty flies, such as sensitive spot detectors in the visual system, will spread. However, it might be the case that the pattern on the flies changes faster than frogling evolution can track. In this case it will pay the little froglings to be able to
learn
which flies to eat. Any frogling that cannot learn will be at a disadvantage and so genes for a general ability to learn will spread. This is the Baldwin effect.

As Baldwin himself puts it – the highest phenomena of intelligence, including consciousness, the lessons of pleasure and pain, maternal instruction and imitation, culminate in the skilful performances of human volition and invention. ‘All these instances are associated in the higher organisms, and all of them unite to
keep the creature alive
… By this means
those congenital or phylogenetic variations are kept in existence, which lend themselves to intelligent, imitative, adaptive, and mechanical modification during the lifetime of the creatures which have them.
Other congenital variations are not thus kept in existence.’ (Baldwin 1896, p. 445, italics in the original.) In more modern terms, genes for learning and imitation will be favoured by natural selection.

Baldwin thus saw that natural selection, without need of the inheritance of acquired characteristics, could account for the evolution of the capacity to learn. The Baldwin effect creates new kinds of creatures that are capable of adapting to change far more quickly than their predecessors. But this is not the only step in this direction. Dennett explains, using his metaphor of the ‘Tower of Generate and Test’, an imaginary tower in which each floor has creatures that are able to find better and smarter moves, and find them more quickly and efficiently (Dennett 1995).

On the ground floor of Dennett’s tower live the ‘Darwinian creatures’. These creatures evolve by natural selection and all their behaviour is built in by the genes. Mistakes are very costly (unsuccessful creatures have to die) and slow (new creatures have to be built each time).

On the next floor live the ‘Skinnerian creatures’, named after B. F. Skinner (1953) who explicitly saw operant conditioning (learning by trial and error) as a kind of Darwinian selection. Skinnerian creatures can
learn. So their behaviour is killed off rather than their whole body. If something they do is rewarded they can do it again, and if not they won’t. This is much faster because one creature can try many many different behaviours in a lifetime.

On the third floor are the ‘Popperian creatures’. They can evolve behaviours even faster because they can imagine the outcomes in their heads and solve problems by thinking about them. They are named after Sir Karl Popper who once explained that this ability to imagine outcomes ‘permits our hypotheses to die in our stead’ (Dennett 1995, p. 375). Many mammals and birds have reached this third floor.

Finally, on the fourth floor, are the ‘Gregorian creatures’, named after the British psychologist Richard Gregory (1981) who first pointed out that cultural artefacts not only require intelligence to produce them in the first place but also enhance their owner’s intelligence. A person with a pair of scissors can do more than one without; a person with a pen can exhibit more intelligence than one without. In other words, memes are intelligence enhancers. Among such memes are what Dennett calls ‘mind tools’ and the most important mind tools are words. Equipped with an environment full of tools that other people have made, and with a rich and expressive language, Gregorian creatures can find good moves and evolve new behaviours very much faster than without. As far as we know, we humans are alone on this top floor of the Tower of Generate and Test.

The importance of the Baldwin effect should now be clear. The Baldwin effect is like the escalator that lifts creatures from one floor to the next. If the necessary good trick is stumbled upon by evolution, and if the costs are not too high, then the creatures who have it are more likely to survive. At each step, they change the environment in which they live so that it becomes ever more important to be good at learning, or whatever. And at each step the creatures who are better at learning are, genetically, at an advantage. Although the Baldwin effect is normally discussed just in the context of learning (stepping up to the second floor), it can equally be applied to the evolution of imagination (getting to the third floor) and of imitation (getting to the fourth floor). Indeed, Baldwin himself explicitly includes imitation in his list of capacities that would help a creature to survive.

But all this is in the service of the genes because the behaviours that are learned, and the solutions that are found by imagining problems, are the ones that help with survival and reproduction. The Baldwin effect is essentially a form of Darwinian evolution acting in the interests of the survival and replication of genes. Several theories of coevolution use the Baldwin effect (such as Deacon’s, for example), but the theory of gene –
meme coevolution I am proposing here adds the further process of memetic driving.

The point is that everything changes when you arrive at the top floor. And it changes dramatically. This is because imitation creates a second replicator. None of the previous steps has created a second replicator – at least, not one that operates beyond the confines of the individual. For example, Skinnerian learning and Popperian problem–solving can be seen as selective processes, but they are all going on inside one animal’s head. The patterns of behaviour and the hypotheses about outcomes that are selected might be seen as replicators, but they are not let loose on the world unless they are copied by imitation and so become memes.

Getting to the fourth floor means letting loose a replicator that spreads from creature to creature, setting its own agenda as it goes. Of course the genes had no foresight. They could not know that selection for imitation would let loose a second replicator, but that is what it has done, and so we enter the phase of gene–meme coevolution. In this kind of coevolution things happen that serve to spread memes whether or not they spread genes – the dog is off its leash and the slaves have rebelled against their former owners. This is what makes the theory different from previous ones and provides alternative predictions. I suggest that the human brain is an example of memes forcing genes to build ever better and better meme–spreading devices. The brain was forced to grow bigger far faster and at much greater cost than would be predicted on the grounds of biological advantage alone, and this is why it stands out so obviously in any comparisons of encephalisation. Theories based only on biological advantage cannot explain why the genes were forced to pay such a high price in terms of energy consumption and the dangers of birth (see
Chapter 6
). Theories based on memetic advantage can.

You might still argue that in terms of sheer brain size the results are not so very different from the argument based on the Baldwin effect. However, the big difference between the theories should come in terms of the specific direction in which the brain evolves, not just its size. If the memes have replicator power then they should drive the genes to produce a brain that is specifically suitable for replicating them, rather than one that is designed for some specifically genetic purpose. We should be able to derive predictions based on the requirements of the new replicator to see whether the actual human brain fits the bill. This is precisely what I tried to do in the argument for the evolution of language. The brain we have is a brain designed for spreading memes with high fidelity, fecundity, and longevity.

As it turns out, the big brains have been enormously successful for
genes too, and humans have colonised almost the whole planet. But need this have been the case? Might not the memes have actually forced the genes into extinction by pushing for ever bigger and bigger brains and extracting too high a price? We cannot know – though it is an odd fact that we are the only surviving hominids. Could the others have gone extinct this way? The ill–fated Neanderthals did, after all, have somewhat larger brains than modern humans. This is wild speculation indeed, but the more serious point is that on this theory we need not take it for granted that big brains, intelligence and all that goes with them are necessarily a good thing for the genes. We could follow Richerson and Boyd (1992, p. 70) in asking, ‘What is so
wrong with
culture that it should be really conspicuous in only one species?’

Maybe the genes have only just managed to carry the burden and fight back in time to produce a species that manages a symbiotic relationship between its two replicators. Maybe we should not assume that when an intelligent, meme–using species evolves it necessarily has a long life ahead.

CHAPTER 10

‘An orgasm saved my life’

Sex sex sex sex sex sex – sex – sex.

Did you perk up? Did you pay more attention to the start of this chapter than any other? Probably not. I expect you have developed plenty of defences against sex memes. Nevertheless, if you want to sell magazines, television programmes, or books, one obvious strategy is to put the word ‘sex’ in a prominent position. A count at my local railway station revealed that of 63 magazines on the shelves 13 had the word ‘sex’ on the cover – and that is ignoring all the ones with erotic photographs, or headlines like ‘Naked couples reveal all’, ‘How would you like to bed this hungry hunk?’, and ‘An orgasm saved my life’.

According to the American author Richard Brodie (1996), memes that deal with sex, food, and power all press powerful meme ‘buttons’ because of the importance of these topics in our evolutionary past. And memes that press buttons are successful memes.

Another way of putting it is that genetic evolution has created brains that are especially concerned with sex, food, and power, and the memes we choose reflect those genetic concerns. Apart from the use of the word ‘meme’, the logic thus far is exactly that of sociobiologists or evolutionary psychologists who assume that the ideas we have, the stories we pass on, the cultural artefacts and skills that we develop, are all ultimately serving the genes. According to sociobiology, culture should reflect genetic concerns, since culture is ultimately
for
the genes.

Yet, in our own society there are many obvious anomalies. Birth rates have fallen dramatically now that many couples think that having two children is quite enough. Some people have decided that they want no children at all, and prefer to devote their lives to their careers or other occupations. Others adopt children who are not biologically related to them and yet bring them up with great care and devotion as though they were their own. Advertisements, films, television, and books encourage us to enjoy sex with multiple partners throughout our adult lives, without any intention of getting pregnant, and teenagers carry condoms around in their pockets. Contraception has not only brought about effective family planning, but also sex for pleasure and sex for spreading memes. Sexually, we do not behave in ways that maximise our genetic legacy. We
no longer have sex in order to get the greatest possible number of our genes into the next generation. We do not buy those magazines in order to have babies. We have largely divorced the act, and joy, and marketing of sex, from its reproductive function.