The Meme Machine (18 page)

A final way of looking at the memetic pressure to talk is to consider groups of memes or memeplexes, and the kinds of person who will nurture and spread them. Memes that thrive in the environment of a chatty person (and contribute to that person being chatty) will differ from those that thrive in the environment of a silent type. The chatty person will, by definition, talk more and so give her memes more chances of spreading. When another chatty person hears these ideas she will easily pick them up and pass them on again. The silent person will not talk much and so all the memes compatible with being a quiet type will have fewer chances to spread. Of course, chatty people can be extremely irritating, and silent types deeply fascinating, but this does not alter the basic imbalance, the inevitable result of which is that memes for talking, or memes that exist happily with memes for talking, will spread in the meme pool at the expense of memes for keeping silent.

These are several memetic arguments which all conspire to have the same effect. If they are correct, it means that the meme pool gradually fills up with memes that encourage talking. We all come across them and that is why we talk so much. We are driven to talk by our memes.

Memetics thus provides a very simple answer to the question – why do we talk so much. This talking is not for our benefit or to make us happy – though sometimes it may do that – nor is it for the benefit of our genes. It is just an inevitable consequence of having a brain that is capable of imitating speech.

This brings us straight back to our other two major issues – how and why we came to have speech in the first place.

The evolution of language

The question of language origins has been so contentious, that as long ago as 1866 the Société de Linguistique de Paris banned any more speculation about the issue. The glaring gulf between animal communication and human speech cried out for explanation but, with little evidence from palaeontology, the speculations of the time could run wild – our words originated from copying animals or natural sounds, or from making grunts of exertion or disgust. These theories, mockingly dubbed the ‘bow–wow’, ‘ding–dong’, ‘heave–ho’, and ‘pooh–pooh’ theories, did nothing to explain the origins of grammar and syntax. More than a century later the issue is far from settled and the arguments are still fierce. Our theorising is, however, constrained by a much better under standing of language itself, and by evidence on how the brain and language evolved together.

First, let us look briefly at the nature of modern human language.

Our language capacity is largely innate and not a by–product of intelligence or a general ability to learn – though this was once a hotly debated issue. The fact is that people do not learn language by being systematically corrected for their mistakes, nor by listening attentively and slavishly copying what they hear. Instead, they just seem to pick it up, using minimal input to build up richly structured grammatical speech. Note that by grammar I mean the natural structures of languages that distinguish who did what to whom or when it happened or in what order – not the sort of rule–book grammar that used to be taught at school.

Almost everyone can use language as grammatically as everyone else, regardless of educational attainment or general intelligence. All human societies ever discovered have language, and all of them have complex grammar. Although languages may vary considerably in the size of their vocabularies, they do not differ much in the complexity of their grammar. Hunter–gatherers and remote tribal groups have languages just as complex as modern industrial English or Japanese. Children all over the world can speak grammatically by the time they are three or four years old, and they can invent languages that are more systematic than the utterances that they hear. They can even use subtle grammatical principles for which there is no evidence in the speech they hear. If spoken language is denied them, as for the deaf, they will find other ways of making language. Sign languages are not just simplified or distorted versions of spoken language, but whole new languages that emerge wherever groups of deaf people come together. They are
languages in their own right with gestures and facial expressions that take on the grammatical functions of word endings, word orders or inflexion.

This ‘language instinct’ as Steven Pinker (1994) calls it, singles us out completely from every other species on the planet. As far as we know, no other species has any kind of grammatically structured language – nor are they capable of learning it. When psychologists first tried to teach language to chimpanzees they failed because chimps simply do not have the vocal apparatus to make the necessary sounds. However, they got on better when they trained their chimps with methods that exploit their natural manual dexterity. One chimpanzee, Sarah, has been trained to use a board containing various plastic shapes that represent familiar objects and actions, while Lana and Kanzi press buttons on a special keyboard. Most popular, however, has been the use of signing, building on the fact that chimps have agile hands and make many gestures in the wild. Among the many animals taught this way have been a chimpanzee called Washoe and a gorilla called Koko, both of whom were brought up with humans using American Sign Language.

At first it seemed as though Washoe, Koko and others really could sign. They were credited with ‘sentences’ three words long, like a child of two years or so. They even made up new words by putting signs together. But the excitement and wild claims soon gave way under careful criticism from psychologists, linguists, and native deaf signers who said that chimp signing was nothing like the rich and expressive human sign language. Wishful thinking probably accounts for much of the exaggeration. The consensus now seems to be that chimps and gorillas can learn single signs or symbols, and use short sequences of them appropriately – mostly to request things. Yet they do not use grammar of any kind and remain oblivious to all the subtleties of sentences that young children seem to take to without effort. Whereas young children just seem to absorb the words they hear and turn them into language, chimps have to be coerced, and rewarded to learn just a few paltry signs. Whatever they may be thinking on the inside (and we should not underestimate that), they just do not ‘get’ the idea of true language. There is no comparison. It is as though the chimps have to learn the words by the long slow route of ordinary learning – trial and error, and reward and punishment – whereas we just seem to absorb it. The human language capacity is unique.

So how did we get this unique ability? Did it appear all at once in some gigantic lucky leap of sudden evolution (Bickerton 1990)? Or did it appear gradually along with our slowly growing brains? And when did
language first appear? Did Lucy indulge in early social chit–chat? Did
Homo habilis
give names to their tools and inventions? Did
Homo erectus
tell stories round their fires?

No one knows for sure. Words do not leave fossils, and extinct languages cannot be brought back. There are, however, a few clues. Some archaeologists believe that we can deduce much about hominid language abilities from their artefacts and burial practices. Only 100000 years ago there occurred the Upper Palaeolithic Revolution, a time of sudden (in archaeological terms) diversification of hominid activity. For two million years or more the only hominid artefacts had been simple stone tools, the stone flakes probably used as choppers and scrapers by
H. habilis,
and hand–axes made by
H. erectus.
It was not until the Upper Palaeolithic that
H. sapiens
began to leave behind evidence of deliberate burial of the dead, simple painting and body adornment, trading over long distances, increasing sizes of settlements and an extension of toolmaking from stones to bone, clay, antlers and other materials. The view that this dramatic change coincides with the sudden origins of fully developed language is, according to Richard Leakey, common among archaeologists. However, it is based only on speculation. When our own thinking is so bound by the language we learned as children it is almost impossible for us to speculate accurately about what can and cannot be done in the way of art, toolmaking or trading, with what level of language ability. We need better evidence than this.

More solid clues come from anatomy. The major increase in brain size, of roughly 50 per cent, occurred during the transition from the australopithecines to
Homo.
By half a million years ago
H. erectus
had brains nearly as big as ours. Since we do not know the nature of the relationship between brain size and language this cannot tell us when language appeared, but perhaps we can find out something about the organisation of early brains. Obviously brains do not fossilise, but their shape can be deduced from the inside of a fossilised skull. One
H. habilis
skull apparently shows evidence of Broca’s area and of the asymmetry characteristic of our language–lateralised brains which led some people to conclude that
H. habilis
could speak. However, recent brain scan studies of living humans show that Broca’s area is also active during skilled hand movements and so cannot be definitive evidence for language. Its development might be connected more to the stone tools made by that species. Nicholas Toth of Indiana University has made a detailed study of early stone tools and he and colleagues spent months learning to make them – not an easily acquired skill as it turned out (Toth and Schick 1993). In the process they discovered that most of
the early stone tools were made by right–handed people. Brain lateralisation apparently began with the earliest appearance of
Homo
but is not proof of language.

The brain is not the only part of the body that has been modified for speech. Exquisite control of breathing is needed and this meant changes in the muscles of the diaphragm and chest. We have to be able to breathe automatically, as do all land mammals, but then to override the mechanism when speaking, which requires cortical control over the muscles. The larynx is also much lower in humans than in related primates, which makes possible a greater variety of sounds, and the base of the skull is a different shape.

When did these changes take place? Neither larynx nor muscles fossilise, but other clues can be used. One is the base of the skull, the shape of which affects the range of sounds that can be made. It appears flat in australopithecines, slightly flexed in
H. erectus,
and only becomes fully flexed as it is in modern humans, in archaic
H. sapiens,
suggesting that only modern humans could make the full range of sounds that we use now. Another clue comes from the thickness of the spinal cord. Modern humans have much larger thoracic spinal cords than apes or early hominids, presumably because speech requires precise cortical control over breathing. The palaeontologist Alan Walker made a detailed study of a 1.5-million–year-old
H. erectus
skeleton – the ‘Nariokotome Boy’ found near Lake Turkana in Kenya. This skeleton was well preserved in just the right parts of the spine and showed no thoracic enlargement. In this respect, the Nariokotome Boy was more ape–like than human. As Walker got to know the boy through his ancient remains he became more and more convinced that
erectus
was speechless, and the boy less like a human trapped in an ape’s body and more like an ape in a human body. ‘He may have been our ancestor, but there was no human consciousness within that human body. He was not one of us,’ concluded Walker (Walker and Shipman 1996, p. 235).

All these clues do not give a final answer. Even if we thoroughly understood the anatomical changes involved in producing speech we would not necessarily understand the psychological changes. As psychologist Merlin Donald (1991) points out, there is much more to modern symbolic cultures than language alone, and more than language separates us from our ancestors and from other living primates. Language evolution needs to be understood in relation to the rest of our cognitive evolution.

Perhaps the best we can conclude for now is that language did not appear suddenly, as some linguists have suggested. The evolutionary
changes which make modern language possible appear to be strung out over a long period of hominid history. Almost certainly Lucy was incapable of speech, and
H. erectus
could not have had much of a conversation around the fire. Finely controlled speech and fully modern language is unlikely to have appeared until at least the time of archaic
H. sapiens,
little more than 100000 years ago. That said, the bigger questions remain unanswered. We cannot tell whether the larger brain gradually made language possible, or the beginnings of language gradually forced the increase in brain size. We only know that the two evolved together.

It might help if we knew what language was for.

The answer is not obvious – though it is often portrayed that way. Introductory psychology textbooks tend to make ‘obvious’ statements like ‘The ability to engage in verbal behavior confers decided advantages on our species’ (Carlson 1993, p. 271), and leave it at that. The biologists Maynard Smith and Szathmáry (1995, p. 290) start their explanation of language evolution with ‘the presumption that natural selection is the only plausible explanation for adaptive design. What other explanation could there be?’ Linguists often assume that language ‘has an obvious selectional value’ or that ‘Language must surely confer enormous selectional advantage’ (Otero 1990), or talk about language adaptation, the significant selective advantage of communication, or selection pressures for the use of symbols (Deacon 1997).

They are surely right to think in terms of selective advantage. When we ask a ‘why’ question in biology, the kind of answer we are seeking is usually a functional one. Bats have sonar so that they can catch insects in the gloom. Spiders spin elegant webs to make near invisible, lightweight traps. Fur is for insulation and eyes are for seeing (though the answer never quite stops there!). According to modern Darwinian thinking, all these things gradually evolved because individuals who carried the genes that produced them were more successful at survival and reproduction. If the human language capacity is a biological system like the vertebrate eye or bat sonar then we must be able to say what function it served, and why individuals carrying the genes that increased language competence were more likely to survive and reproduce than their less language–competent neighbours. As we have seen, language cannot have come cheap. Not only are several areas of the brain specialised for understanding and producing speech, but the whole of our vocal apparatus had to be evolved. This meant complex changes in the neck, mouth and throat that compromised other functions; making drinking and breathing at the same time
impossible and increasing the risk of choking. Why were these costly and potentially dangerous modifications ever made? What made them worth it?