Creativity would have been even less noticeable in the predecessor of our species. Yet it must already have been evolving in that species, or ours would never have been the result. In fact the advantage conferred by successive mutations that gave our predecessors’ brains slightly more creativity (or, more precisely, more of the ability that we now think of as creativity) must have been quite large, for by all accounts modern humans evolved from ape-like ancestors very rapidly by gene-evolution standards. Our ancestors must have been continually out-breeding their cousins who had slightly less ability to create new knowledge. Why? What were they using this knowledge for?
If we did not know better, the natural answer would be that they were using it as we do today, for innovation and for understanding the world, in order to improve their lives. For instance, individuals who could improve stone tools would have ended up with better tools, and hence with better food and more surviving offspring. They would also have been able to make better weapons, thus denying the holders of rival genes access to food and mates – and so on. Yet if that had happened, the palaeontological record would show those improvements happening on a timescale of generations. But it does not.
Moreover, during the period when creativity was evolving, the ability to replicate memes was evolving too. It is believed that some members of the species Homo erectus living 500,000 years ago knew how to make camp fires. That knowledge was in their memes, not in their genes. And, once creativity and meme transmission are both present, they greatly enhance each other’s evolutionary value, for then anyone who improves something also has the means to bequeath the innovation to all future generations, thus multiplying the benefit to the relevant genes. And memes can be improved much faster by creativity than by random trial and error. Since there is no upper limit to the value of ideas, the conditions would have been there for a runaway co-evolution between the two adaptations: creativity and the ability to use memes.
Yet, again, there is something wrong with that scenario. The two adaptations presumably did co-evolve, but the driving force behind that evolution cannot have been that people were improving on ideas and passing the improvements on to their children, because, again, if they had been, they would have been making cumulative improvements on a timescale of generations. Before the beginning of agriculture, about 12,000 years ago, many thousands of years passed between noticeable changes. It is as though each small genetic improvement in creativity produced just one noticeable innovation and then nothing more – rather like today’s experiments in ‘artificial evolution’. But how can that be? Unlike present-day artificial-evolution and AI research, our ancestors were evolving real creativity, which is the capacity to create an endless stream of innovations.
Their ability to innovate was increasing rapidly, but they were barely innovating. This is a puzzle not because it is odd behaviour, but because, if innovation was that rare, how could there have been a differential effect on the reproduction of individuals with more or less ability to innovate? That there were thousands of years between noticeable changes presumably means that in most generations even the most creative individuals in the population would not have been making any innovations. Hence their greater ability to innovate would have caused no selection pressure in their favour. Why did tiny improvements in that ability keep spreading rapidly through the population? Our ancestors must have been using their creativity – and using it to its limits, and frequently – for something. But evidently not for innovation. What else could it have been used for?
One theory is that it did not evolve to provide any functional advantage, but merely through sexual selection: people used it to create displays to attract mates – colourful clothing, decorations, story-telling, wit and the like. A preference to mate with the individuals with the most creative displays co-evolved with the creativity to meet that preference in an evolutionary spiral – so the theory goes – just like peahens’ preferences and peacocks’ tails.
But creativity is an unlikely target for sexual selection. It is a sophisticated adaptation which, to this day, we are unable to reproduce artificially. So it is presumably much harder to evolve than attributes like coloration or the size and shape of body parts – some of which, it is thought, did indeed evolve by sexual selection in humans and many other animals. Creativity, as far as we know, evolved only once. Moreover, its most visible effects are cumulative: it would be hard to detect small differences in the creativity of potential mates on any one occasion, especially if that creativity was not being used for practical purposes. (Consider how hard it would be, today, to detect tiny genetic differences in people’s artistic abilities by means of an art competition. In practice, any such differences would be swamped by other factors.) So why did we not evolve multi-coloured hair or fingernails instead of the capacity to create new knowledge, or any one of countless other attributes that would have been far easier to evolve, and far easier to assess reliably?
A more plausible variant of the sexual-selection theory is that people chose mates according to social status, rather than favouring creativity directly. Perhaps the most creative individuals were able to gain status more effectively though intrigue or other social manipulation. This could have given them an evolutionary advantage without producing any progress of which we would see evidence. However, all such theories still face the problem of explaining why, if creativity was being used intensively for any purpose, it was not also used for functional purposes. Why would a chief who had gained power through creative intrigue not be thinking about better spears for hunting? Why wouldn’t a subordinate who invented such a thing have been favoured? Similarly, wouldn’t potential mates who were impressed by artistic displays also have been impressed by practical innovations? In any case, some practical innovations would themselves have helped the discoverers to produce better displays. And innovations sometimes have reach: a new skill of making a string of decorative beads in one generation might become the skill of making a slingshot in the next. So why were practical innovations originally so rare?
From the discussion in the previous chapter, one might guess that it was because the tribes or families in which people were living were static societies, in which any noticeable innovation would reduce one’s status and hence presumably one’s eligibility to mate. So how does one gain status, specifically by exercising more creativity than anyone else, without becoming noticeable as a taboo-violator?
I think there is only one way: it is to enact that society’s memes more faithfully than the norm. To display exceptional conformity and obedience. To refrain exceptionally well from innovation. A static society has no choice but to reward that sort of conspicuousness. So – can enhanced creativity help one to be less innovative than other people? That turns out to be a pivotal question, to which I shall return below. But first I must address a second puzzle.
How do you replicate a meaning?
Meme replication is often characterized (for example by Blackmore) as imitation. But that cannot be so. A meme is an idea, and we cannot observe ideas inside other people’s brains. Nor do we have the hardware to download them from one brain to another like computer programs, nor to replicate them like DNA molecules. So we cannot literally copy or imitate memes. The only access we have to their content is through their holders’ behaviour (including their speech, and consequences of their behaviour such as their writings).
Meme replication always follows this pattern: one observes the holders’ behaviour, directly or indirectly. Then, later – sometimes immediately, sometimes after years of such observation – memes from the holders’ brains are present in one’s own brain. How do they get there? It looks a bit like induction, does it not? But induction is impossible.
The process often seems to involve imitating the holders. For instance, we learn words by imitating their sounds; we learn how to wave by being waved to and imitating what we see. Thus, outwardly, and even to our own introspection, we appear to be copying aspects of what other people do, and remembering what
they say and write. This common-sense misconception is even corroborated by the fact that our species’ closest living relatives, the great apes, also have a (much more limited, but nevertheless striking) ability to imitate. But, as I shall explain, the truth is that imitating people’s actions and remembering their utterances could not possibly be the basis of human meme replication. In reality these play only a small – and for the most part inessential – role.
Meme acquisition comes so naturally to us that it is hard to see what a miraculous process it is, or what is really happening. It is especially hard to see where the knowledge is coming from. There is a great deal of knowledge in even the simplest of human memes. When we learn to wave, we learn not only the gesture but also which aspects of the situation made it appropriate to wave, and how, and to whom. We are not told most of this, yet we learn it anyway. Similarly, when we learn a word, we also learn its meaning, including highly inexplicit subtleties. How do we acquire that knowledge?
Not by imitating the holders. Popper used to begin his lecture course on the philosophy of science by asking the students simply to ‘observe’. Then he would wait in silence for one of them to ask what they were supposed to observe. This was his way of demonstrating one of many flaws in the empiricism that is still part of common sense today. So he would explain to them that scientific observation is impossible without pre-existing knowledge about what to look at, what to look for, how to look, and how to interpret what one sees. And he would explain that, therefore, theory has to come first. It has to be conjectured, not derived.
Popper could have made the same point by asking his audience to imitate, rather than merely to observe. The logic would have been the same: under what explanatory theory should they ‘imitate’? Whom should they imitate? Popper? In that case, should they walk to the podium, push him out of the way, and stand where he had been standing? If not, should they at least turn to face the rear of the room, to imitate where he was facing? Should they imitate his heavy Austrian accent, or should they speak in their normal voices, because he was speaking in his normal voice? Or should they do nothing special at the time, but merely include such demonstrations in their lectures when they themselves became professors of philosophy? There are infinitely many possible interpretations of ‘imitate Popper’, each defining a different behaviour for the imitator. Many of those ways would look very different from each other. Each way corresponds to a different theory of what ideas, in Popper’s mind, were causing the observed behaviour.
So there is no such thing as ‘just imitating the behaviour’ – still less, therefore, can one discover those ideas by imitating it. One needs to know the ideas before one can imitate the behaviour. So imitating behaviour cannot be how we acquire memes.
The hypothetical genes that caused meme replication by imitation would also have to specify whom to imitate. Blackmore, for instance, suggests that the criterion may be ‘imitate the best imitators’. But this is impossible for the same reason. One can only judge how well someone is imitating if one already knows, or has guessed, what (which aspect of behaviour, and whose) they are imitating, and which of the circumstances they are taking into account and how.
The same holds if the behaviour consists of stating the memes. As Popper remarked, ‘It is impossible to speak in such a way that you cannot be misunderstood.’ One can only state the explicit content, which is insufficient to define the meaning of a meme or anything else. Even the most explicit of memes – such as laws – have inexplicit content without which they cannot be enacted. For example, many laws refer to what is ‘reasonable’. But no one can define that attribute accurately enough for, say, a person from a different culture to be able to apply the definition in judging a criminal case. Hence we certainly do not learn what ‘reasonable’ means by hearing its meaning stated. But we do learn it, and the versions of it that are learned by people in the same culture are sufficiently close for laws based on it to be practicable.
In any case, as I remarked in the previous chapter, we do not explicitly know the rules by which we behave. We know the rules, meanings and patterns of speech of our native language largely inexplicitly, yet we pass its rules on with remarkable fidelity to the next generation – including the ability to apply them in situations the new holder has never experienced, and including patterns of speech that people explicitly try to prevent the next generation from replicating.
The real situation is that people need inexplicit knowledge to understand laws and other explicit statements, not vice versa. Philosophers and psychologists work hard to discover, and to make explicit, the assumptions that our culture tacitly makes about social institutions, human nature, right and wrong, time and space, intention, causality, freedom, necessity and so on. But we do not acquire those assumptions by reading the results of such research: it is entirely the other way round.
If behaviour is impossible to imitate without prior knowledge of the theory causing the behaviour, how it is that apes, famously, can ape? They have memes: they can learn a new way of opening a nut by watching another ape that already knows that way. How is it that apes are not confused by the infinite ambiguity of what it means to imitate? Even parrots, famously, parrot: they can commit to memory dozens of sounds that they have heard, and repeat them later. How do they cope with the ambiguity of which sounds to imitate, and when to repeat them?
They cope with it by knowing the relevant inexplicit theories in advance. Or, rather, their genes know them. Evolution has built into the genes of parrots an implicit definition of what ‘imitating’ means: to them, it means recording sequences of sounds that meet some inborn criterion, and later replaying them under conditions that meet some other inborn criterion. An interesting fact follows, about parrot physiology: the parrot’s brain must also contain a translation system that analyses incoming nerve signals from the ears and generates outgoing ones that will cause the parrot’s vocal cords to play the same sounds. That translation requires some quite sophisticated computation, which is encoded in genes, not memes. It is thought to be achieved in part by a system based on ‘mirror neurons’. These are neurons that fire when an animal performs a given action, and also when the animal perceives the same action being performed by another. These neurons have been identified experimentally in animals that have the capacity to imitate. Scientists who believe that human meme replication is a sophisticated form of imitation tend to believe that mirror neurons are a key to understanding all sorts of functions of the human mind. Unfortunately, that cannot possibly be so.
It is not known why parroting evolved. It is a fairly common adaptation in birds, and may play more than one role. But, whatever the reason, the important thing for present purposes is that parrots never have a choice about which sounds to imitate, or about what constitutes imitating them. A ringing doorbell and a barking dog may happen to provide conditions that meet the inborn criterion that initiates parroting behaviour, and, when they do, the parrot will always mimic exactly the same aspects of them: their sounds. So, it resolves the infinite ambiguity by making no choices. It does not occur to it to ignore the dog under those conditions, or to imitate the wagging of its tail, because it is incapable of conceiving of any other criterion for imitation than the one built into its mirror-neuron system. It is devoid of creativity and relies on its lack of creativity to replicate the sounds faithfully. This is reminiscent of humans in static societies – except for a crucial difference which I shall explain below.
Now, imagine that a parrot had been present at Popper’s lectures, and learned to parrot some of Popper’s favourite sentences. It would, in a sense, have ‘imitated’ some of Popper’s ideas: in principle, an interested student could later learn the ideas by listening to the parrot. But the parrot would merely be transmitting those memes from one place to another – which is no more than the air in the lecture theatre does. The parrot could not be said to have acquired the memes, because it would be reproducing only one of the countless behaviours that they could produce. The parrot�
��s subsequent behaviour as a result of having learned the sounds by heart – such as its responses to questions – would not resemble Popper’s. The sound of the meme would be there, but its meaning would not. And it is the meaning – the knowledge – that is the replicator.
The parrot is oblivious to the human meanings of the sounds that it parrots. Had those lectures been not about philosophy but about recipes for fried parrot, it would have been just as eager to quote from them to anyone who would listen. But it is not oblivious to the content of the sound – it is not like a mechanical recorder. Quite the contrary: parrots neither record sounds indiscriminately nor replay them randomly. Their inborn criteria do implicitly attribute meaning to sounds that they hear; it is just that the meaning is always drawn from the same, narrow set of possibilities: if the evolutionary function of parroting is, for instance, to create identifying calls, then every sound it hears is either a potential identifying call or not.
Apes are capable of recognizing a much larger set of possible meanings. Some of them are so complex that aping has often been misinterpreted as evidence of human-like understanding. For example, when an ape learns a new method of cracking nuts by hitting them with rocks, it does not then play the movements back blindly in a fixed sequence like a parrot does. The movements required to crack the nut are never the same twice: the ape has to aim the rock at the nut; it may have to chase the nut and fetch it back if it rolls away; it has to keep hitting it until it cracks, rather than a fixed number of times; and so on. During some parts of the procedure the ape’s two hands must cooperate, each performing a different sub-task. Before it can even begin, it must be able to recognize a nut as being suitable for the procedure; it must look for a rock and, again, recognize a suitable one.
The Beginning of Infinity Page 48