The big brain certainly looks like a runaway phenomenon and I am not the first to suggest a role for sexual selection in brain size. But previous theorists have not explained why sexual selection should pick on brain size (e.g. Deacon 1997; Miller 1993). My answer comes directly from the power of the memes.
The way memes can exploit the process of sexual selection is unique. Whatever is deemed ‘in’ can change as fast as the memes change – and that is much faster than genes can produce longer tails or an innate ability to build a fancy nest. If you follow the heuristic ‘mate with the man with the most memes’ you will soon find yourself mating with the one with the best hairdo or the best song (as well as the ability to imitate). If other females start going for good songs then it becomes advantageous to have
male children who can pick up a tune quickly. Or if females (for whatever reason) start going for ritualised hunting dances then it becomes advantageous to have male children who can copy dances. The selection pressures on the genes now change in the wake of changes in the memes. The process of sexual selection is exactly the same as it is in examples of biological evolution, but with the added twist that the things being selected for can spread at the speed of memetic evolution. Meme–driven sexual selection will favour mating with males who are not only good at imitating in general, but who are good at imitating whatever happen to be the favoured memes at the time. In this way the memes are, as it were, dragging the genes along. The leash has been reversed and, to mix metaphors, the dog is in the driving seat.
Please note, however, that sexual selection is not necessary for a memetic explanation of brain size, and its role must be an empirical question for the future. The first three processes alone will produce the selection pressures required to drive a runaway increase in brain size – if one further small assumption is made. That is, that being good at imitation requires a big brain. Interestingly, there has been so little attention paid to imitation that there is very little information to back this up. However, this theory suggests that the main tasks of our larger brains are first, the general ability to imitate, and second, the particular ability to imitate the kinds of memes that have proliferated in our species’ past.
Can this theory be tested? Like so many biological theories it is not easy to devise specific experimental tests. Nevertheless, some predictions can be made. For example, within any related group of species I would predict that imitation ability will correlate positively with brain size. That is, the best imitators will have the largest brains. Given the scarcity of imitation among other animals there will not be much data to choose from, and there will be problems with choosing an appropriate measure of encephalisation, but this study ought to be possible for various groups of birds and cetaceans.
Using humans, experiments could compare two people performing the same actions but with one person initiating the action while the other imitates it. Various measures could be used to determine just how much extra demand is created by imitating. For example, cognitive studies should show that imitation requires a lot of processing and that we have specialised mechanisms for doing it. Brain–scan studies should show that imitation requires a large amount of energy, and that the extra activity is found predominantly in the evolutionarily newer parts of the brain – those parts that differentiate us from other species. I would not be
surprised if specific neurons were found that carry out some of the basic tasks of imitation, such as relating observed facial expressions or actions to one’s own, but we will need to know a lot more about how imitation is done before we can guess what to look for.
If these predictions turn out to be right they will confirm the suggestion that imitation is an enormously demanding task, and that it takes a large brain to be able to do it. I would further predict that many aspects of language and thought will turn out to be best understood as by–products of our brains’ ability to select which aspects of the world to imitate. However, until more research on imitation is carried out, I can only speculate and say this – if being good at imitation requires a big brain then the processe described above can explain it. These are: selection for imitation, selection for imitating the best imitators, selection for mating with the best imitators, and (possibly) memetic sexual selection. Once early hominids achieved imitation the second replicator was born and these processes began to drive the increase in brain size. The enormous human brain has been created by the memes.
CHAPTER 7
The origins of language
Why do we talk so much?
This might not be a question you have agonised over, but once I started to think about it I found it more and more interesting. How much time and energy does an average person spend on talking every day? I doubt it has been measured but the answer must be several hours. A typical human form of entertainment is to sit over a meal or a few drinks and talk to a lot of other people – what about? Well, about football, or sex, or who has got off with whom, or what he said to her or she said to him, or the latest trouble at work, or the iniquities of the latest government proposals on health care, and so on and on and on. According to some estimates about two–thirds of all conversation is taken up with social matters (Dunbar 1996). It is rare for any group of people to sit in companionable silence.
Then there is work. Some jobs are silent but most are not. In shops and offices, on the buses and trains, in factories and restaurants, people talk. And if they do not talk they often have the radio on with voices and music coming at them from somewhere else. And then there are other forms of communication that use language – the letters, magazines and newspapers arrive on the doormat, the phone rings, the fax starts up, the e–mail messages flood in. The use of time and energy is phenomenal. What is it all for?
There are at least three issues here. One is
why
we talk at all – in other words, why human beings acquired language in the first place. The second is
how
we acquired language – how the human brain became structured the way it did. The third is why, having acquired language, we use it so much. I am going to tackle the last question first, partly because it is easier, and partly because the answer will help us with the much more controversial questions of how and why language evolved.
Why do we talk so much?
Talking all the time must cost energy – and a lot of it. Thinking uses some energy, but talking uses a lot more. Not only are several brain areas
necessarily active during speech, or when listening to and understanding speech, but the production of sound is itself expensive. If you have ever been very ill you will know how exhausting it is to speak. You may lie in a hospital bed perfectly able to think but when the nurse arrives you can barely manage a feeble ‘thank you’, while a few days later you will happily engage in friendly banter about the quality of the food, or what you will do when you get out – complete with smiles, laughs and completely superfluous chit–chat.
Perhaps you are a hi–fi freak. If so you will know how much energy is needed to drive big speakers, and how expensive the sound system gets when it needs to play loud, high–quality sound. Or if you prefer low tech you may have a clockwork radio, in which case you will know all too well, by the feeling in your arm, how much energy is needed to produce that sound, and how much winding you can save by turning down the volume.
This phenomenal use of energy presents something of a puzzle. Living creatures have to work hard for all the energy they consume, and efficient energy use is a critical factor in survival. If you can use less energy than your neighbour, you are more likely to pull through the hard times, to find scarce food, to win the competition for the best mate, and so to pass on your genes. Why, then, has evolution produced creatures that talk whenever they get the chance?
Several possible answers spring to mind. First, there may, after all, be a sound biological explanation. Perhaps talking serves an important function that I have overlooked, such as cementing social bonds or exchanging useful information. I will consider theories of this kind later on.
Second, a sociobiologist might argue that, with the evolution of language, culture has somehow got temporarily out of hand, and the cultural trait of speech has been stretching the leash. However, if talking is really wasteful of precious energy then the genes of the people who talk most will do less well and in time the genes will pull the leash in again.
Third, an evolutionary psychologist might argue that all this talking once had advantages for our ancestors and so we are stuck with it now, even though it doesn’t benefit our genes any more. On this view we ought to be able to find the function of so much talking in the lives of early hunter–gatherers.
All these suggestions have in common that they appeal to genetic advantage for an explanation. Memetics provides a totally different approach. Rather than asking what advantage talking provides to the genes, we can ask what advantage it provides to the memes. Now the
answer is obvious. Talking spreads memes. In other words, the reason we talk so much is not to benefit our genes, but to spread our memes.
There are several ways of looking at how memes exert pressure on us to keep talking, and I will consider three of them in more detail.
First, since talking is an efficient way of propagating memes, memes that can get themselves spoken will (in general) be copied more often than those that cannot. So these kinds of memes will spread in the meme pool and we will all end up talking a lot.
This argument is similar to the explanation I proposed for why we think so much – another example of the ‘weed theory’ of memes (p. 41). Silence is like a beautifully weeded flowerbed, just waiting for your favourite plants, and it does not stay that way for long. A silent person is an idle copying machine waiting to be exploited. Your brain is full of ideas, memories, thoughts to be shared, and actions to be carried out; the social world is full of new memes being created, spread about, and competing to be taken up by you and passed on again. But you cannot possibly speak them all. Competition to take charge of your voice is strong – just as competition to grow in the garden is strong. Keeping silence is as hard work as weeding.
So which memes will win in this competition to take over your voice? It may help to ask again our familiar question –
imagine a world full of brains, and far more memes than can possibly find homes. Which memes are more likely to find a safe home and get passed on again?
Certain memes are particularly easy to say, or almost force their hosts to pass them on. These include bits of juicy scandal, terrifying news, comforting ideas of various sorts, or useful instructions. Some of these have their ‘spread me’ effect for good biological and psychological reasons. Perhaps they tap into needs for sex, social cohesion, excitement, or avoiding danger. Perhaps people pass them on in order to conform, to be better liked, to enjoy the other person’s surprise or laughter. Perhaps the information will be genuinely useful to the other person. We can certainly study all these reasons (and indeed psychologists do just that) but for the memetic argument I am proposing here it does not matter what they are. The point is you are less likely to want to pass on some boring thing you heard about the health of your neighbour’s rose bushes than a rumour about what your neighbour was doing behind them. Such ‘say me’ memes will therefore spread better than other memes and many people will get infected with them.
The news of Princess Diana’s death in 1997 spread around the world at the speed of light within minutes of its first announcement. People all over the world told anyone who did not yet know. I did myself. I turned
on the radio, heard the continuous coverage instead of the weather forecast and called out to the rest of my family. Then I felt a bit silly for shouting so loud about something I would normally profess to take no interest in. But the death of Diana was just that sort of news. It spread like an extremely infectious virus and within weeks the princess’s reputation had become saintly and her following cult–like (Marsden 1997). Within a few months, millions of pounds had been given to her memorial fund and millions more made out of selling her image. Few memes can claim anything like this power, but the principle is quite general. Certain kinds of news spread more effectively than others. These are the things people get to hear about and want to pass on again. As a result, people talk more.
This does not mean that silence is impossible. It is just rare, and needs special rules to enforce it against the natural memetic tendency for endless talk. We see these rules all over the place, in libraries and schools, in lecture theatres and cinemas, and even in special train carriages – and we see people, despite their best intentions, finding themselves breaking the rules. True vows of silence are hard to make, and on religious retreats beginners find the rules of silence difficult to keep, even for a few days. Taking on a silence meme goes against the grain.
This suggests a second approach: to look at rules or social practices concerning speech. Again let’s compare two types of meme. Suppose there are instructions encouraging people to talk a lot. These might come in many forms, such as embarrassment at being silent in company, or rules about making polite conversation or entertaining people with chat. Now suppose there are other memes for keeping silent, such as the suggestion that idle chat is pointless, a rule of quiet etiquette, or a spiritual belief in the value of silence. Which will do better? I suggest the first type will. People who hold these memes will talk more; therefore, the things they say will be heard more often and have more chances of being picked up by other people.
If this conclusion does not immediately appear obvious think of it this way – imagine that one hundred people have been taught behaviour of the first type – such as ‘You should make polite conservation whenever you can’ – and another hundred people have been taught the rule ‘It’s polite only to talk when you have to’. The first group will, because they hold this meme, talk whenever they have the chance. The second lot will keep quiet. If talkers meet talkers they will all talk. If silents meet silents they will not. The interesting mixture is when talkers meet silent types. It is possible that nobody will ever change their minds or throw out old memes in favour of new ones, but if ever this does happen the imbalance
is obvious. A talker will talk, and either directly, or by implication, suggest that polite conversation is necessary, or that talking is fun, or useful. A silent type might be converted. But the reverse is extremely unlikely to happen. The silent type may occasionally say things like ‘I think it’s better to keep quiet’ or ‘Why don’t you shut up’ but will not, by definition, say much – and for that reason alone is unlikely to make converts. Although single memes of this explicitness are probably rare, there are notable examples, such as the British Telecom slogan ‘It’s good to talk’ and the proverbial ‘Silence is golden’. Memetics should help us understand not only why talking in general must spread, but also how some selective environments can encourage the rarer silence rule to succeed.