Grooming, Gossip, and the Evolution of Language - PDF Free Download (2024)




time to avoid charges. If you have any questions, please check with your home library.

d�Bn�UBl JO uO!lnToA'3: dlp pUB 'd!SS08 '�U!lli0018

Grooming, Gossip, and the Evolution of Language ROBIN DUNBAR


( . - 'i

K\!",�""1 t :�"

i� n ,.-\�) ;;�;f'1i(t,


qa ....r- lrl"�".§'l'\Il .

....rl. "\I�_:- i;':tl'��lr



COPYflght © 1996 by Robin Dunbar

AI! rights reser\'ed

Primed in the United Stares of America Published by arrangement with Faber and Faber Limited, London Second printing, 1997

Library of Congress Cataloging-in-Publication Data


Dunbar, R. I. M. (Robin Ian MacDonald), 1947-

Grooming, gossip, and the evolution of language / Robin Dunbar. p.


Includes bibliographical references. ISBN 0-674-36334-5 1. Human evolution . 2. Social evolution. 3. Language and languages-Origin. 4. Gossip-History. 5. Human behavior. 6. Group identity. 7. Interpersonal relations. I. Title. GN281.4.D85 573'.2-dc20


1996 96-15934

Talking Heads

2 Into the Social Whirl



3 The Importance of Being Earnest


4 Of Brains and Groups and Evolution


5 The Ghost in the Machine


6 Up Through the Mists of Time


7 First Words


8 Babel's Legacy


9 The Littie Rituals of Life


10 The Scars of Evolution








Talking Heads

This book inevitably owes a great deal to a great many people. I am grateful to all those with whom I have discussed the ideas it contains, and especially so to those who have contributed to the research on which it is based. Particular thanks go to: Leslie Aiello, Rob Barton, Dick Byrne, Richard Bentall, Hiroko Kudo, Peter Kinderman, Chris Knight, Sam Lowen, Dan Nettle, Sanjida O'Connell, Boguslaw Pawlowski and Peter Wheeler. Neil Duncan, Amanda Clark, Nicola H urst, Catherine Lowe, David Free and Anna Marriott helped with various research proj ects, and Nicola Koyama tracked down the references for the biblio­ graphy. As always, I am indebted to my editor Julian Loose for his enthusiasm and patience. RD

To be groomed by a monkey is to experience primordial emo­ tions: the initial frisson of uncertainty in an untested relationship, the gradual surrender to another's avid fingers flickering expertly across bare skin, the light pinching and picking and nibbling of flesh as hands of discovery move in surprise from one freckle to another newly discovered mole. The momentarily disconcerting pain of pinched skin gives way imperceptibly to a soothing sense of pleasure, creeping warmly outwards from the centre of atten­ tion. You begin to relax into the sheer i ntensity of the business, ceding deliciously to the ebb and flow of the neural signals that spin their fleeting way from periphery to brain, pitter-pattering their light drumming on the mind's consciousness somewhere in the deep cores of being. The experience is both physical sensation and social inter­ course. A light touch, a gentle caress, can convey all the meanings in the world: one moment it can be a word of consolation, an apology, a request to be groomed, an invitation to play; on another, an assertion of privilege, a demand that you move else­ where; on yet another, a calming influence, a declaration that intentions are friendly. Knowing which meaning to infer is the very basis of social being, depending as it does on a close reading of a nother's mind. In that brief moment of mutual understanding in a fast-moving, frenzied world, all social life is distilled in a single gesture. To recognize what this simple gesture signals in the social world of monkeys and apes, you need to know the intimate details of those involved: who is friends with whom, who dominates and who is subordinate, who owes a favour in return for one granted the week before, who has remembered a past slight. The very -1-

Grooming, Gossip and the Evolution of Language complexity of the social whirl creates those am biguities we are so familiar with from our own lives. Take, for instance, lojo, who has j ust given birth to her first off­ spring. She cradles it in her arms, at once puzzled by this strange, wet thing and unsure what she should do. Already alert, the baby struggles to turn its head, as though surprised by the unfamiliar sights and sounds that surround it. They are not alone for long. loj o's mother, Persephone, comes across. She peers down at the baby, sniffs it tentatively, and reaches out a hand to touch its rump. Persephone gives a quiet grunt and begins to groom lojo, leafing through her fur, busying herself with the rituals of social interaction. But she cannot take her mind off the baby and keeps pausing to reach down and groom its head briefly, making smack­ ing noises with her tongue and lips as she does so. loj o relaxes to the rhythm of her mother's grooming, and her eyes half close. But she j olts awake again when her baby gives a whimper. Two young j uveniles are poking at the infant, fascinated by its wriggling as they pull tentatively on a leg. loj o pulls the baby away and turns her back on them, disrupting Persephone's groom ing. Persephone stares meaningfully at the two j uveni les, her head lowered and her eyebrows raised in threat. The j uveniles scamper off to annoy someone else. loj o and Persephone happen to be ba boons, members of a troop whose life centres on a rocky outcrop in the wooded grass­ lands of eastern Africa. But they could be almost anywhere in Africa. Indeed, they could be members of any one of a bout a hun­ dred and fifty species of monkeys and apes that l ive i n the forests and woodlands of Asia, Africa and South America. Moreover, there is something eerily familiar in the process of their actions and responses - as if they might be humans too, members of a ny one of some 5000 cultural groups scattered across the globe from Alaska to Tasmania, Benin to Brazil. Here, in the minutiae of everyday life, i s a point of convergence between ourselves and our nearest relatives, the monkeys and apes. Here is behaviour with which we instantly empathize, the innuendoes and subtleties of everyday social experiences. Yet there is one difference: our world is infused through and -2-

Talking Heads through with language, while theirs goes a bout its business in wordless pageant. A human baby produces its first real words at a bout eighteen months of age. By the age of two, it has become quite vocal and has a voca bul ary of some fifty words. Over the next year it learns new words daily, and by the age of three it can use about 1000 words. It is now stringing words together in short sentences of t �o or three words, calling your attention to obj ects, req uesting thIS and that. Its command of grammar is already nearly as com­ pet � nt as an adult's, though it will still make am using yet plainly . saying 'eated' i nstea d of 'ate', 'mouses ' instead of logICal mIstakes, 'mice'. Then the fl oodgates open . By the age of six, the average child has learned to use and understand around 13,000 words; by eighteen, it will have a working vocabulary of about 60,000 words. That means it has been learning an average of ten new words a day since its first birthday, the equivalent of a new word every 9 0 m inutes of its waking life. Thi s is an extraordinary achievement. It is no wonder that the machinery which makes this possible is so expensive to maintain. Although your brain accounts for no more than 2 per cent of your body weight, it consumes 20 per cent of all the energy you eat. In other words, pound for pound, the brain burns up ten times as much energy to keep itsel f going as the rest of the body does . The situation i s even more extreme i n young chi ldren, where the brain is actively growing as opposed to j ust ticking over. D uring the last stages of pregnancy, the foetus's brain is growing very rapidly and consumes 70 per cent of the total energy the mother pumps into h �r baby via the umbil ical cord - and she, of course, has to pro­ VIde all that. Even after b irth, the brain still accounts for 60 per cent of the infant's total energy consumption during the first year of life. Lactation is an exhausting business. It will come as no surprise to discover that we humans have the largest brains relative to body size of any species that has ever existed. Our brai n is nine times larger than you would expect for a mammal of our body size. It is thirty times larger than the brain of a dinosaur of the same body size. Only the porpoises and dol­ phins come close to us i n this respect; yet even though dolphins

Grooming, Gossip and the Evolution of Language

Talking Heads

are renowned for their i ntelligence and sociability, they still do not compete with humans on the verbal scale. Complex though their natural language of whistles and clicks may be, it does not seem to be in the same league as human language.

a bad thing; who is in and who is out, and why; how to deal with a difficult social situation involving a lover, child or colleague. You may happen on a particularly intense exchange about a technical problem at work or a book just read. But listen on, and I 'll wager that, within five minutes at the most, the conversation has drifted away again, back to the natural rhythms of social life. Even were you to listen to the conversations in university com­ mon rooms or the restaurants of multinational companies, there at the very hub of our intellectual and business life the situation would not be all that different. To be sure, you will occasionally come across an intense discussion of some abstruse scientific tech­ nicality or business deal. But that will only be when a visitor is being entertained or when individuals meet for the specific pur­ pose of thrashing out some key problem of mutual concern. For the rest of the everyday conversations, it's unlikely that more than about a quarter would be concerned with matters of such intellec­ tual weight as the cultural, political, philosophical or scientific issues of the day. Here are two more statistics, this time gleaned from the world of the printed page. Of all the books published each year, it is fic­ tion that tops the list in terms of volume of sales. Take a glance around your local bookshop: university campus bookshops aside, two-thirds of the shelf space will contain fiction. Even then, it is not the rip-roaring adventure yarns that attract us, but the unfold­ ing intimacies of the main characters. It is the way they handle their experiences that fascinates us, their reactions to the vagaries of life - those 'there but for the grace of God go I' situations. And out of all this fiction, it is not the writings of the acclaimed mas­ ters that top the publishers' sales-lists, but romantic fiction. Everything else - from art history to photography and sport, from the sciences and handicrafts to car mechanics for the home enthusiast - is put together under the all-encompassing label of 'non-fiction'. Only biographies can lay any claim to a significant share of the market in their own right. Every year, a torrent of such books appears, retelling the life experiences of the rich, the famous and the also-rans. Every TV newscaster, every politician, every actress, every minor sportsman from darts to football, has

Because it seems to be unique, language appears al l the more miraculous. Other species bark and scream, grunt and wa il, but none speak. Perhaps inevitably, this has encouraged us to view the human species as special, reinforcing our habits of self-worship. Yet, when we look at our nearest relatives, the monkeys and apes, we find m uch that is familiar - the same i ntensity of social life, the petty squa bbles, the j oys and frustrations, the same whining chil­ dren irritating exasperated parents as in our own private lives. However, neither monkeys nor apes have language in any sense that we would recognize from our everyday experience of human conversation. How did it come about that we, the descendants of j ust such dumb apes, have this extraordinary power when they do not? The puzzle seems all the greater because we feel so at home with the social l ives of monkeys and apes. What makes it seem familiar to us is the time they spend in close physical contact, busily attending to each other's needs i n endless grooming sessions. They think nothing of spending hours leafi ng through each other's fur, comb­ ing, p icking, parting the hairs with the single-minded ness of a human mother attending to her child's tangled mop. The answer to this apparent puzzle lies, I suggest, in the way we actually use our capacity for language. If being human is all a bout talking, it's the tittle-tattle of life that makes the world go round, not the pearls of wisdom that fall from the lips of the Aristotles and the Einsteins. We are social beings, and our world - no less than that of the monkeys and apes - is cocooned in the interests and minutiae of everyday social life. They fascinate us beyond measure. Let me give you a few statistics to reinforce the point. Next time you are in a cafe or a bar, j ust listen for a m oment to your neigh­ bours. You will discover, as we have in our research, that around two-thirds of their conversation is taken up with matters of social i mport. Who is doing what with whom, and whether it's a good or




Grooming, Gossip and the Evolution of Language

Talking Heads

published his or her story. Long-dead novelists, generals and politicians all command their fair share of attention. And why do we buy such books? It's not to learn about the sport in question, or how to read the news on TV, but to learn about the private lives of our heroes or those who have become as familiar to us as our own families. We want the intimate details, the gossip, their innermost thoughts and feelings, not detailed technical analy­ sis of method acting or parliamentary procedure. We want to know how events affected them, how they reacted to the highs and lows of life, what they thought about their friends and relations, the indignities they suffered, the triumphs they took part in. Take another look at your daily paper. How many column­ inches are devoted to substantive news about politics and eco­ nomics? Here is the score for two of yesterday's papers, the upmarket London Times and the mass-market UK tabloid The Sun. No less than 78 per cent of the 1 063 column-inches of text in the downmarket Sun was concerned with 'human interest' stories,· stories whose sole purpose seemed to be to enable the reader to become a voyeur of the intimate lives of other individuals. That leaves just 22 per cent for news and commentary on the political and economic events of the day, for the sports results, for news of upcoming cultural events, and for everything else. Even the august Times only devoted 57 per cent of the 1 993 column-inches of text in its main news and review sections to political and technical news; 43 per cent was devoted to human interest stories (inter­ views, news stories of a more salacious kind, and so on). The number of actual column-inches devoted to 'gossip' was virtually identical in the two papers: 833 and 8 5 0 respectively. It's clear that most of us would rather hear about the doings of the great and the not-so-good than about the intricacies of eco­ nomic processes or the march of science. The trial of O. J. Simpson aroused more interest and achieved higher viewing figures than the deliberations of US congressional committees, despite the fact that the conclusions of those committees will have an impact on our future lives far beyond any conceivable impact that OJ's guilt or innocence might have. Here, then, is a curious fact. Our much-vaunted capacity for

language seems to be mainly used for exchanging information on social matters; we seem to be obsessed with gossiping about one another. Even the design of our minds seems to reinforce this. Of course, great things are possible with language: Shakespeare and T. S. Eliot are not figments of our imagination, neither are the unsung writers of instruction manuals; we really can use language for profit and pleasure. And language remains our greatest trea­ sure, for without it we are confined to a world that, while not one of social isolation, is surely one that is a great deal less rich. Language makes us members of a community, providing us with the opportunity to share knowledge and experiences in a way no other species can. So how is it that we have this extraordinary ability, yet most of the time seem to do so little with it? A century of intensive research in linguistics, psychology and speech science has taught us a great deal about language: how it is produced, what grammar does, how children learn it. Yet at the same time, we know almost nothing of why it is that we alone, of all the tens of thousands of living species, possess this extraordinary ability. We do not know for sure wh�n it evolved or what the first languages ever spoken sounded like. However, during the last ten years we have learned more about the background to human evolu­ tion and the behaviour of our nearest relatives, the monkeys and apes, than we had in the previous thousand years put together; and this new evolutionary perspective, firmly rooted in modern Darwinian biology, has focused our attention on questions about language that have hitherto been overlooked. In the process, aspects of our own past that had been buried beneath the murky waters of history for hundreds of millennia ha ve finally come to light. The approach I adopt is thus very different from the perspectives of those who study language conventionally . For the past century, the study of language has focused prinicpally on three main areas: linguistics (with its pervasivie interest in the structure of gram­ mars); socio-linguistics (with its interest in the way sex and social class influence the words we use and how we pronounce them); and the neurobiology of language (the brain structures that allow us to speak and understand) . Although there has been some interest in the archaeology and the history of language (and the




Grooming, Gossip and the Evolution of Language

processes of dialect formation), these have been considered both peripheral and trivially speculative by the mainstream interests. Even less attention has been devoted to the function of language and the reasons why we have it and other species do not. Indeed, such questions have often been deliberately eschewed. Instead, language has often been viewed as an 'epiphenomenon', some­ thing that appeared as a by-product of other biological processes (notably, our super-large brains) and for which no other kind of explanation is necessary. This curious state of affairs owes its origins largely to the claim (widely held in the social sciences) that human behaviour in gener­ al, and language in particular, are social phenomena and thus lie beyond the pale of biological explanation. Neurobiology might provide us with insights into the machinery of language produc­ tion and comprehension,. but beyond that, biology sheds little light on the nature of language. - By and large, biologists have respected this demarcation line. But the recent developments in evolutionary biology have had far-reaching implications for our understanding of human behaviour as well as that of other ani­ mals, and language has inevitably come under this new and more powerful microscope. This book is about those new discoveries, and about the origins of our capacity for language. I shall examine not only what we do with language but also the more fundamental questions of why we have it, whence it came and how long ago it appeared. The story is a magical mystery tour that will take us bouncing from one unexpected corner of our biology to another, from history to hormones, from the very public behaviour of monkeys and apes to the moments of greatest human intimacy. It will take us back through the chapters of human history to the time before we were human, when we were but apes of a not especially unusual kind. What did the earliest languages sound like? Who spoke them? And why, from these early hesitant steps, did languages evolve, changing and diversifying so much that now we have around 5 000 mutually incomprehensible tongues (and that's not counting the ones that became extinct in the millennia before anyone could write them down)? -8-


Into the Social Whirl

What characterizes the social lives of humans is the intense inter­ est we show in each other's doings. We spend literally hours in each other's company, stroking, touching, talking, murmuring, being attentive to every detail of who is doing what with whom. You might think that this marks us out as a cut above the rest of life, but you would be wrong. If we have learned anything from the last thirty years of intensive research on monkeys and apes, it is that we humans are anything but unique. Monkeys and apes are just as social as we are, just as intensely interested in the social whirl around them. So to set the scene on the human story, we need to go back in time to our primate heritage. What is it about primates that makes them so different from other animals, that in turn gives us our unique character? The answer is that primates live in a very much more sophisticated kind of social world than other animals do. The Monkey on my Back

Monkeys and apes are highly social species. Their lives revolve around a small group of individuals with whom they live, work and have their being. Without its friends and relations, a monkey would no more be able to survive than a human being could. The social life of primates is intense and all-consuming. They spend a great deal of the day engaged in social grooming with their special friends. Like Jojo and Persephone, whose story opened Chapter I, these are often matrilineal relatives, related through their mother's line in an unbroken chain of personalized mother-daughter rela­ tionships that run back through the mists of time to some ances­ tral primate pre-Eve.

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

The biologist Richard Dawkins has reminded us just how short this chain of ancestry really is. Imagine yourself, he says, standing on the Indian Ocean shore just where Kenya abuts on to the southern border of Somalia. Face south and reach out to hold your mother's left hand in your right hand. Facing you is a chim­ panzee of the same age and sex, holding its mother's hand in its left hand. Your mother is holding her mother's hand in her right hand, and the chimpanzee's mother is holding her mother's hand in her left hand. The double chain of generations snakes its paral­ lel way across the African plains westwards towards the distant peak of Mount Kenya, a faint brown smudge emerging above the clouds on the horizon. By the time the chain reaches Mount Kenya itself, a distance of no more than 300 miles, the mother­ daughter lines have converged and met in a single mother-Eve. She lived somewhere on the East African savannahs some time between 5 and 7 million years ago. The number of generations between you and this ancestral Eve is surprisingly small. Even allowing a modest yard and a half for the span of outstretched arms and twenty years as the average generation length, there are no more than 350,000 individuals separating you on the Kenyan coast from her on the slopes of Mount Kenya. That's barely a third of the people who work for the National Health Service in the UK, no more than the total population of a modest English county town or, to put it into really dramatic perspective, about half the number of babies born in England and Wales each year. Even allowing just ten years per generation (probably a better estimate of the typical age at which females give birth for the first time among chimpanzees and our earliest ancestors), the line of life would stretch no further than the western shore of Lake Victoria, some 600 miles from the coast, perhaps 700,000 individuals in all. It's a sobering thought that so few generations separate us from the common ancestor we share with the chimpanzees. Here, indeed, is not just our cousin but our sister-species. It is no wonder that some biologists have started to refer to us humans as the third chimpanzee (in addition, that is, to the common chimpanzee and its closely related sister species the bonobo or pygmy chimpanzee).

But let's pursue Dawkins' graphic metaphor a little further back in time. How much further will we need to go to reach the com­ mon ancestor of the Old World monkeys and apes? At most 8 5 miles further on, a week's easy walk beyond Mount Kenya, we come to the common ancestor of the gorilla and the chimp-human lineage. That's something like 1 00,000 generations if females give birth for the first time at about ten years of age, as most great apes do. On the same scale, some 700 miles further on we come to the common ancestor of the human-chimp-gorilla family and the orang-utan, the endearing red ape of the Asian forests. We are now just on the Uganda-Zaire border, a mere stone's throw to the north of the Virunga Volcanoes where Dian Fossey lived and died watching her beloved mountain gorillas. Generation lengths get shorter as we go further back among the smaller-bodied species, perhaps 5-6 years on average once we are past the common ancestor of the living species of great ape - but at the same time, the length of an outstretched arm is now no more than 1 8 inches, a yard between adjacent noses. It's just another 400 miles to the common ancestor of the great apes and the gibbons, the lesser apes now found only in south-eastern Asia. From there to the common ancestor of all the monkeys and apes of Africa and Asia will be another 1 1 00 miles. By now, we are somewhere in the middle of Congo-Brazzaville, with still another 5 00 miles to go to the Atlantic Ocean. We have not even traversed the African continent at its narrower part. We have travelled back in time through 30 million years, and there have been just four million females in the unbroken chain that links you and that long-dead pre-pre-Eve that gambolled through the tree-tops in an ancient African forest. That's less than half the population of London or Paris, barely a quarter of the population of modern­ day Rio de Janeiro. We are almost exactly half-way back in time towards the Age of Dinosaurs. We are still a long way from the origins of the first, primitive pre-primates in the dying days of these great reptiles' empire. Most people are surprised to find that, so far from being a new and advanced product of evolution, the primates are in fact one of the oldest lineages of mammals, a close relative of the

- 10 -

- 11 -

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

insectivores, the bats and the flying foxes. Their and our early ancestors dodged the same heavy-footed lizards in the twilight years of the long reign of the dinosaurs. The ancestral primates were small, squirrel-like animals with long pointed noses that scuttled among the bushes and trees of the dense tropical forests that existed during the closing millennia of the Age of the Dinosaurs. In the new freedom brought by the post-dinosaur years, they diversified into myriad new species in hundreds of different niches, mostly in the northern hemisphere in what is now Europe and North America. These species were all prosimians, whose living descendants include the lemurs of Madagascar and the galagos and lorises of Africa and Asia. For 3 0 million years they dominated the woodlands and forests of the northern hemisphere. Then the earth's climate cooled rapidly over a period of around two million years. Water surface temperatures in the tropical Pacific dropped from a sultry 23°C to around 17°C. I The tropical zones shifted southward towards their present equatorial position. As these climatic changes developed, one of these prosimian groups began to evolve in an entirely new direction. Brain size increased, faces became more rounded. It was the beginning of a major break with the past that ultimately gave rise to the so-called anthropoid primates (the monkeys and apes as we know them

today). By this time, primates were confined pretty much to their present distribution in the equatorial regions of Africa, Asia and South America. Soon afterwards, contact between Africa and South America was lost. The South American monkey populations went their own way, evolving into species that are still reminiscent of the ancestral anthropoid primates of 3 5 million years ago. In Africa and Asia, meanwhile, evolution proceeded apace. Around 30 million years ago, this branch of the primates split into two major families, the Old World monkeys (now represented by colo bus and langur monkeys, baboons and macaques) and the apes. It was to be the apes that dominated the forests of the Old World for the next 1 5-20 million years, however. The monkeys remained relatively insignificant. Some time around 10 million years ago, the forests of the Old World began to retreat as the climate started to dry out and tem­ peratures once more plummeted. Surface temperatures on the earth's oceans dropped by another 10°C. Within a few million years the apes had succumbed, and the monkeys, better adapted to a terrestrial way of life and able to out-compete the apes by eat­ ing poorer-quality diets, came into their own. Part of the problem seems to have been that, unlike the mon­ keys, apes lack the ability to neutralize the tannins in unripe fruit. Tannins are poisons that plants produce to make their parts inedi­ ble. Mature leaves often contain high concentrations of tannins, apparently to prevent herbivores from stripping the leaves off a tree and thereby killing it. But sometimes animals can be useful to plants. Rooted to the spot, plants have a problem about dispers­ ing their seed. It doesn't pay to have all your children growing under your feet, because they simply compete with you and each other for sunlight and the nutrients in the soil. It does pay to dis­ perse your seeds widely, where the young plants will compete with (and, you hope, out-produce) other plants' offspring. The problem that plants face is how to achieve this dispersal. Their saviours, if you like, were animals. Mammals such as monkeys can travel several miles in a day. By harnessing their energies, a plant can disperse its seeds over a very wide area indeed. Species like figs, plums and apples provide their seeds with

I. We can determine temperatures in the remote past thanks to a curious quirk of

physics. There happen to be two isotopes (or forms) of oxygen, one being fraction­ ally heavier than the other. A molecule that happens to have been formed with

atoms of the heavier isotope, oxygen-I 8, will not evaporate from the oceans as eas­

ily as its counterparts made from the lighter oxygen-I6 atoms; but once evaporat­ ed, it condenses more rapidly to form snow. Hence, by measuring the ratio of the two forms of oxygen in ice or snow, we can tell whether one time period was cold­ er or warmer than another: colder periods will have a higher proportion of the lighter oxygen isotope. Ice cores drilled from the Greenland or Antarctic ice caps are examined inch by inch for changes in the ratio of the two oxygen isotopes. With their ratio in samples of today's snow as a standard, the series can be anchored to current global temperatures, and past temperatures can be read off. Another way of doing it is to examine the composition of the calcium carbonate in the shells of extinct oceanic plankton. The plankton take up oxygen from the seas in which they live to make their skeletons. In cooler times, there will be a higher proportion of the heavier oxygen isotope in sea water because more of the lighter ones will have evaporated.

- 12 -

G rooming, Gossip and the Evolution of Language

a luxurious coating of energy-rich flesh to entice animals to eat the seed. The seed passes slowly through the animal's gut (a pas­ sage that can take two or three days, by which time the animal may be several miles away), and gets dumped to germinate far from its parent. Plants that adopt this strategy have a problem, however. The seed has to develop to its final form and size to be able to germi­ nate under its own steam. All the nutrients required for the early stages of germination have to be provided by the mother plant. To prevent fruit-eating animals from destroying the immature seed before it can fend for itself, plants that produce these kinds of seed protect their fruits with tannins and other compounds. These poi­ sons gradually break down as the fruit ripens, so that by the time the seed at the centre of the fruit is ready to cope with its great journey through life, the flesh that surrounds it has shed its chem­ ical defences. It is the tannins that give unripe fruit . that bitter, mouth-drying taste. We share with the great apes an inability to digest unripe fruit. Lacking the enzymes that break down the tannins, we suffer a . stomach-ache or, in the worst cases, diarrhoea if we eat too.much of it. However, some time during their evolutionary history the Old World monkeys acquired the enzymes and other mechanisms that allow them to beat the chemical defences of plants. Being able to eat unripe fruit may well have given monkeys like the baboons and macaques a distinct edge over the ape lineages as things began to get tough in the forests seven million years ago. With significant quantities of fruit being eaten before it ripened, there was much less available for the apes. Gradually but inexorably, the apes went into decline and the monkeys took over as the dominant primates of the forest. Those few species of apes that survived were forced into increasingly marginal habitats like the forest floor and the for­ est edge, where monkeys seldom ventured. Today, the remnants of that once-successful ape lineage cling to survival by their finger­ nails, confined to small pockets here and there in Africa and Asia, their populations dwindling by the decade. Meanwhile, whole new groups of monkey species appear for

Into the Social Whirl

the first time in the fossil record and soon begin to dominate the scene. The macaques, now confined to Asia but once widespread throughout Europe and north Africa, appear some 10 million years ago. The baboons appear a few million years later. The guenons are more recent still: the earliest recognizable guenons are barely two million years old - younger than the first truly rec­ ognizable members of our own genus, hom*o, who inhabited the lake and river margins of eastern Africa some two and a half mil­ lion years ago. Most of us naturally assume that the monkeys of Africa and Asia represent the ancestral condition through which we all passed in this long story. Traditional conceptions of the evolution of the primates envisaged a natural progression by which the com­ mon monkey of the Old World transformed itself first into apes and then into modern humans. Most people are surprised to dis­ cover that this is incorrect. The new science of molecular genetics, combined with a better understanding of anatomy and a newly discovered wealth of fossil material, has revealed that the familiar monkeys of Africa and Asia, like the baboons, guenons and macaques, are in fact upstart newcomers by comparison with the apes, the lineage to which we humans belong. The disaster that hit the luckless apes 10 million years ago was not, of course, complete and final. One lineage of apes did survive the depredations of the climatic changes: the lineage that eventual­ ly led to us. Some time around 7 million years ago, one popula­ tion seems to have begun to make increasing use of the more open savannahs that bordered the forest to which the apes were (and still are) so well adapted. They were probably forced to do so by their failure to compete successfully with the other ape species that were vying with increasing desperation for the ancestral for­ est-home. But as so often happens in evolutionary history, the challenge of exploiting a marginal habitat forced the pace of evo­ lutionary change. Mortality would have been desperately high, but those that survived did so because they were able to exploit the new conditions. In that crucial moment, a mere blink of an eye on the geological time-scale, our history hung in the balance between extinction and survival. It must have been touch and go.

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

Despite their very different and at times turbulent histories, all these lineages have faced the same agonizing problem: how to sur­ vive the depredations of a seemingly endless catalogue of preda­

larger than itself. Only a handful of specialized, group-hunting predators can do this successfully. Hence by increasing your body size, you considerably reduce your vulnerability to predators. Terrestrial species are more vulnerable to predators than arboreal species because they have less opportunity to escape into the dense foliage or fine outer branches of trees where pursuit becomes too risky for predators. As a result, all the terrestrial species are larger than their arboreal counterparts. The other way of reducing the risk of predation is to live in large groups. Groups reduce the risk in a number of ways. One is simply by providing more eyes to detect stalking predators. Most predators have to get within a certain distance of their prey unde­ tected in order to have any chance of catching it. That's why your cat inches its way, belly-down in the grass, towards the birds pecking at the breadcrumbs you so kindly threw on to your lawn. Using every blade of grass and every molehill as cover, the cat moves only when it is sure the birds are not looking, freezing in its tracks the moment it senses the birds might be suspicious. Each predator has its own attack distance, depending on its speed and its style of attack. For a cheetah, capable of hitting 70 miles an hour within seconds from a standing start, the attack dis­ tance is 70 yards; for the slower and more bulky lion it is 3 0 yards, while for the lighter leopard i t i s just 1 0 yards, often less. If prey can detect a stalking predator before it can get within its attack distance, they will always be able to outrun the predator. Most predators are well enough aware of that, if only by virtue of past experience, and rarely bother to chase a prey that has already seen it. That's one reason why you will occasionally see a lion walking through a herd of wildebeest with the herd simply parting like the biblical Red Sea around the advancing predator. The wildebeest know that so long as they stay outside the lion's attack distance they are relatively safe and need do no more than keep a wary eye on it. Larger groups are also an advantage as a deterrent. Most preda­ tors will be less enthusiastic about attacking a prey animal if they know that several others will come to the victim's aid. Although it is virtually unknown for species like wildebeest to go to the aid of

tors ranging from sabre-tooth cats to lions and leopards, from hyenas and hunting dogs to monkey-eating eagles and occasional­ ly even other primates. Finding food is, of course, a perennial problem for any animal, but given enough time it can almost always scratch a living from the natural world. The problem is that, as it does so, it inevitably exposes itself to the risk of being taken unawares by a predator. In order to gain the advantage of surprise over their prey, most predators rely on stealth before throwing themselves into the final chase. So every minute spent travelling from one feeding site to another, every minute spent with its attention focused on removing a fruit or a leaf from the branch in front of it, exposes an animal to the risk of being caught by a predator lying in wait for an-inattentive prey.. The intensity of predation depends on your size. For species as large as chimpanzees and gorillas, the risks of predation are much reduced (though by no means altogether absent). But for smaller species it can be an ever-present threat. It has been estimated that about a quarter of all vervet monkey deaths are due to predation

(mostly by leopards), while as many as 20 per cent of all red colobus monkeys at the Gombe National Park in Tanzania fall prey to hunting by Jane Goodall's famous chimpanzees. Predation is a significant evolutionary problem, because ani­ mals that find themselves on the inside of a predator no longer have the opportunity to breed and reproduce. Since animals that fail to reproduce do not contribute any of their genes to future generations, there is intense pressure to find ways of circumvent­ ing this unhappy fate. Evolution is the outcome of successful solu­ tions to the problem. Indeed, the very fact that you and I are alive today is a consequence of the remarkable circ*mstance that every one of our ancestors - back to, and far beyond, that ancestral pre­ pre-Eve - successfully solved the problem of survival, at least for long enough to reproduce. They did so by exploiting two main facts about predators. One is that a predator cannot easily handle a prey animal significantly

- 16-

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

a fellow herd-member brought down by a lion or hunting dogs, group defence is more common among primates. Baboons have been known to drive leopards up a tree, and have occasionally even killed them. Red colo bus monkeys are significantly less likely to be attacked by chimpanzees if an adult male of their group is nearby; even chimps seem unwilling to risk a mass counter-attack launched by an animal that's barely a quarter of their own weight. Think of it in human terms. Being handbagged by one granny would not put off the average mugger, but even the most deter­ mined thug will balk at being hand bagged by twenty grannies simultaneously. Last but not least, a group creates confusion in a predator. Predators succeed by locking on to a target animal and running it down. When the prey runs into a group, there are animals running every which way and the predator becomes momentarily con­ fused. That moment of lost attention is often just enough for the

commuters and the inhabitants of inner-city high-rise tenements, the frustrations of chronically overcrowded housing and large fa milies. They are the centrifugal forces that drive us all apart in sea rch of the peace and serenity of being alone. Social animals hang in perpetual balance between two forces: the centripetal forces, driven by fear of predation, which have produced the feelings of sociableness that make us seek out com­ pany; and the centrifugal forces, generated by overcrowding, that send us scurrying for the sanity of a solitary life. When predators become common (and for humans, those predators can just as easily be neighbouring human groups rampaging through your territory on raids), we hanker for the close proximity of friends and tolerate all kinds of overcrowding. When predators are rare, the stresses of crowding overwhelm us and we disperse. Group size is the product of this balancing act. Primates have evolved a distinctive response to this problem. Large, tightly bonded groups are their solution to the risks of pre­ dation; but in order to be able to increase group size, it was first necessary to develop a mechanism that enabled them to keep fel­ low group members just far enough away so that they don't become too much of a nuisance, while at the same time not dri­ ving them away altogether. The fine balance is brought about by coalitions between small numbers of animals. A mother and her daughters, or two sisters perhaps, form an alliance in which they provide mutual support against everyone else. It's an 'if you'll help me, I'll help you' arrangement. It seems to be unique to the higher primates. Although male lions form coalitions to take over prides of females, these are often temporary affairs, formed on the spur of the moment with a specific end in view. The coalitions of monkeys and apes are long-term commitments, often formed months ahead of their being needed. They are a promise of future action in circ*mstances as yet unimagined. Of the thousands of hours I have spent watching monkeys in Africa, perhaps the most unashamedly enjoyable were those spent studying an obscure but unusually attractive species of baboon found only on the high mountains of the Ethiopian plateau. This is the gelada, once also known by the evocative name of the bleed-

prey to make good its escape. So primates live in groups as a mutual defence against preda­ tion. Indeed, sociality is at the very core of primate existence; it is their principal evolutionary strategy, the thing that marks them out as different from all other species. It is a very special kind of sociality, for it is based on intense bonds between group members, with kinship often providing a platform for these relationships. Primate groups have a continuity through time, a history built on kinship (usually mother-daughter relationships, but very occa­ sionally father-son ones too). A Friend i n Need

Living in groups creates its own tensions, as any member of a close-knit community knows only too well. There is that once-in­ a-while thoughtless transgression on your personal space, the time when someone treads on your tail in the melee around a particu­ larly attractive food source. Or worse still, the time when some other scoundrel unashamedly steals the very food from your mouth just as you settle down to feed. These are the everyday trials of social life, the hassles of crowding so familiar to city -1 8-

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

ing heart baboon because of the hourglass-shaped patch of bare red skin on its chest. The males are truly magnificent, sporting lion-like capes of long russet and black hair that hang from their shoulders and float on the breeze as they run along the cliff-tops. Gelada live in harems that typically consist of four or five females, their dependent offspring and a breeding male (the harem male). While sons usually leave to join an all-male bachelor group soon after puberty, daughters mature into the group to join their moth­ ers, older sisters, aunts and female cousins in a coalition of great intensity and loyalty. In effect, these alliances are formed at birth, the product of being born to a particular mother. These deeply rooted alliances have important implications for the harem males. They are perpetually at risk of being displaced by younger males anxious to get a foot on the ladder of immortality by finding females with whom they can breed. Because there are four or five females in each harem but only one breeding male,. many of the males in the population are excluded from breeding. These form the core of the all-male groups, biding their time in anticipation of a suitable opportunity. Sooner or later, desperation forces their hand and they make a bid to take over a harem of females. Needless to say, the incumbent harem males are less than enthusiastic about the prospect of losing their harems - for once the harem is lost, they have no further opportunity to mate and breed. In an effort to forestall the inevitable, harem males devote a great deal of their time to trying to scare off the opposition on the sidelines with impressive displays. At the same time, they try to ensure that their females do not stray too far from them lest they find an opportunity for illicit dallyings with these males. Whenever a female wanders too far away, or inadvertently finds herself too close to another male (even another harem male), her male will warn her with raised eyebrows and panted threats. Occasionally, these may escalate into a charge, culminating in a vigorous display of threats over the cowering female. The harem male's attempts to ride herd on his females when they stray too far from him often backfire at this point. The luck­ less victim's grooming partners invariably come to her aid. Standing shoulder to shoulder, they outface the male with out-

raged threats and furious barks of their own. The male will usual­ ly back off and walk huffily away, endeavouring to maintain an air of ruffled dignity. Occasionally, however, the male will persist, feeling perhaps unusually sensitive about his honour or security. This only leads to more of the group's females racing in to sup­ port their embattled sisters. The male invariably ends up being chased round the mountainside by his irate females in an impres­ sive display of sisterly solidarity. These alliances are established and maintained by grooming, the most social activity in which monkeys and apes engage. In some species, as much as a fifth of the entire day may be spent grooming, or being groomed by, other group members. A mother will spend hours devotedly grooming her offspring, carefully leaf­ ing through its fur in search of dead skin, matted hair, bits of leaf and burrs that have become entangled in its hair as the animal brushes its way through the vegetation during the day's foraging. She will also groom her friends and relations, in what seems to be selfless devotion to their hygienic interests. Keeping the fur clean and the skin healthy is obviously an important factor in the life of any animal. But there is more to grooming than just hygiene, at least in the monkeys and apes. For them, it is an expression of friendship and loyalty. Robert Seyfarth and Dorothy Cheney of the University of Pennsylvania spent the better part of a decade during the late 1970S and early 1980s studying the vervet monkeys that live in Kenya's Amboseli National Park. In one of their studies, they recorded the screams uttered by individual vervets that were being attacked by another member of the group. Then, when the screaming animal was physically out of sight, they played these calls back over loudspeakers hidden in bushes, at the same time recording on video the responses of the target animals sitting immediately in front of the loudspeaker. When they played the calls for help back to most of the animals in the group, they elicited little response other than a cursory glance in the direction of the hidden speaker. But when they played the calls back to an animal that the caller had groomed with during the previous two hours, that animal immediately

- 20 -

- 21 -

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

looked up and stared into the bushes. It was as though it was try­ ing to make up its mind whether to go and investigate more close­ ly, in order to fulfil its obligation to a grooming partner. Did the situation warrant help, or was it a minor spat that would quickly blow over? The vervets clearly differentiated between the animals they groomed regularly and those they didn't. A grooming partner is something special, someone who deserves particular attention, who should be supported in moments of need, on whose behalf the taking of risks is warranted. Gelada operate in the same way. Even on the small scale of the harem unit, the females are very selective about those they groom and those whom they support in altercations. The frequencies with which females are supported (both when they are involved in squabbles within their own unit and when they are attacked by members of another harem fOI accidentally transgressing into their space) correlate with the frequency with which they are groomed. They are clearly well aware of whom they owe loyalty to, and they don't have to have groomed with them half an hour beforehand to know it. Not all primate societies exhibit these characteristics, it must be said. Prosimians like the lemurs of Madagascar and the galagos of Africa rarely exhibit coalitionary behaviour, even when they live in groups. And while not unknown, it is by no means common among South American monkeys and some lineages of Old World monkeys like the colo bus. Those species that show coalitionary behaviour at its most highly developed tend to be those like the baboons, macaques and vervet monkeys, and the chimpanzee, that live in relatively large groups.

every potential ally fighting against everyone else in turn. Instead, the monkeys seem to weigh up the possibility that if Peter can beat Jim and Jim can defeat Edward, then it's very likely that Peter would defeat Edward should they ever come to blows. This kind of inference about social relations, combined with an acute sense of others' reliability as allies, seems to be the founda­ tion on which primate alliances are built. At the cognitive level, they are really quite sophisticated kinds of social inference to make. But given that you can make them, a new possibility pre­ sents itself. Knowledge can be put to bad use, as in propaganda, as well as good. Monkeys and apes use their social skills to exploit each other. A classic case of this was observed by Andrew Whiten and Dick Byrne during their studies of chacma baboons in southern Africa. A young adult female named Mel was digging a succulent tuber out of the ground. It was a particularly tough job, and one far beyond the strength of all but an adult animal. But the prize of a nutritious tuber in the impoverished habitat these animals inhabit is well worth the effort. Meanwhile, a young juvenile named Paul had been quietly watching Mel at work. Just at the crucial moment when Mel managed to wrench the tuber clear of the ground, Paul let out an ear-splitting scream, of the kind common­ ly uttered by juveniles who are being attacked by someone much bigger and stronger than themselves. Paul's mother, who had been busy feeding out of sight on the far side of some bushes, immedi­ ately came racing through. She took in the situation at a glance, added two and two and made five: Mel had obviously threatened her darling little boy. She fell on the unsuspecting Mel with the kind of enthusiasm uniquely characteristic of mothers whose chil­ dren are being molested. Needless to say, the startled Mel dropped her tuber and fled, with the outraged mum in hot pur­ suit. Paul nonchalantly picked up the abandoned tuber and settled in to enjoy lunch. Observations of this kind are not uncommon. The Swiss zoologist Hans Kummer described a case in which a young female hamadryas baboon spent twentY minutes edging her way inch by inch across a distance of just two yards to get to a rock behind which lay the

Enter Machiavelli

Coalitionary behaviour seems to be possible only because the ani­ mals understand how other individuals tick and how they rate as allies against possible opponents in the group. These are the kinds of knowledge that cannot always be acquired firsthand. Fights are not so common in a primate group that you would be able to see - 22 -

. - 23 -

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

group's young male follower. Once there, she began to groom with

But every time Nikkie forgot himself in his jealousy, Yeroen would remind him by withdrawing his support against Luit.

him, sitting upright with the top of her head in full view of her harem male a few yards away. Hamadryas have a similar social structure to. the gelada, with small harems of two or three females monopolized by a single breeding male. One key difference between the two species, however, is that hamadryas males are completely intolerant of their females grooming (or even being near) other males. It seemed as though the female had engineered herself into a position where her male would be left with the impression that she

These manipulations are only possible because monkeys and apes are able to calculate the effect their actions are likely to have. This does not of course mean that Yeroen was working out the odds on his pocket calculator; indeed, it is not even clear in this case how much of Yeroen's success came from deliberate schem­ ing and how much from luck. We can stand outside the events and, with hindsight, read a cunning plan into the story as it

was engaged in some completely innocent activity.

unfolds. But when we ourselves are embroiled in the action, we

In his book Chimpanzee Politics, the Dutch zoologist Frans de Waal describes a classic example of the subtle balancing act that

act more by instinct, sensing an opportunity to be gained on the

the more advanced primates can sometimes engage in. In the cap­ tive group of chimpanzees housed at Arnhem Zoo in the Netherlands, the young male Luit had just overthrown the old male Yeroen. Yeroen had been .dominant male for some years and during that time he had enjoyed more or less exclusive access to

rarely think about it in quite such clinical terms. Like Yeroen, we hoof. Still, there was a consistency to Yeroen's behaviour that implies some kind of forethought, even if only at the superficial level of recognizing the opportunities that circ*mstances pushed his way. The fact that similar behaviour has been observed in wild chimpanzees by the Japanese primatologist Toshisada Nishida lends credence to the view that chimps at least are capable of see­

the females when they were ready to mate. But his fall to second rank meant that he lost his privileges to Luit. Worse was to follow

ing the implications of their actions and factoring these into their

a few months later, when the young male Nikkie reached the point where he was able to defeat Yeroen too. Yeroen sank to

'alliance fickleness' to Yeroen's kind of manipulation of Nikkie.

third rank and lost all privileges. Then came the stroke of genius: rather than bemoaning his bad luck and getting depressed, the wily old male formed an alliance with the young Nikkie. Being younger than Luit, Nikkie was no match for the new dominant male on his own. But with Yeroen's support, he was able to defeat Luit. The result was a new ranking, with Nikkie at the top and Yeroen second, and Luit squeezed back into third place. Then came the coup de grace. Yeroen proceeded to make use of his position to mate with the females. Nikkie, of course, took instant umbrage and set about chastising the presumptuous Yeroen. Yeroen patiently bided his time. On the next occasion when Nikkie and Luit got into a squabble, Yeroen simply sat on the sidelines and refused to go to Nikkie's help. So Nikkie lost the battle, and would have lost the war had he not quickly settled his difference with Yeroen. So long as he tolerated the old male mat­ ing with at least some of the females, things worked out just fine.

plans for the future. Nishida gave the rather evocative name There is considerable evidence to suggest that monkeys and apes are sensitive to the risks they run in these kinds of situation and adjust their behaviour accordingly. Saroj Datta has shown that juvenile female rhesus monkeys are significantly less likely to rush into supporting an ally against a higher-ranking opponent when the opponent's mother is nearby than when she is our of sight. It is as if they know that the opponent's mother is unlikely to sit by while they flatten her precious offspring - and, worse still, that high-rank­ ing mothers tend to have large numbers of relatives who are easily drawn into a squabble in defence of their collective status. There is no point in helping a friend if by so doing you merely exacerbate the situation and cause both of you to end up being flattened. I have seen gelada behave in a similar way. One day, a young female was attacked by her harem male for straying too far away from the rest of the group. He stood over her, threatening and grinding his teeth in high dudgeon. The female's mother was feed-

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

ing about five yards away at the time. She looked up the moment the commotion started, but made no effort to intervene. Eventually, his point made, the male turned and stalked off to start feeding a few yards away. As the victim walked disconsolate­ ly back towards the rest of the group, her mother called to her with a soft grunt. The victim at once turned and walked over to her mother, who then began to groom her. I had the distinct impression that the mother did not want to become involved in a squabble with the male (perhaps because she sensed that doing so would simply escalate the fight), but at the same time she realized that failing to do so had weakened her relationship with her daughter. The grunt and the grooming seemed to say, 'I'm sorry!' Frans de Waal has described similar behaviour in chimpanzees and macaques, and has termed it 'reconciliation'. You might think of it in terms of an apology designed to restore the status quo in an alliance that has been damaged by the thoughtless behaviour of one of its members. In most cases, reconciliation involves groom­ ing, touching or other physical actions. Chimpanzees kiss each oth­ er on the lips; macaques groom or hold another's rl,lIllp in a half­ mount; male baboons will reach through to touch another male's penis. However, Joan Silk, Dorothy Cheney and Robert Seyfarth have recently reported vocal forms of reconciliation, similar to those just described in the gelada, in chacma baboons inhabiting the Okavango swamps of Botswana. They found that dominant females will give conciliatory grunts when approaching lower­ ranking females with whom they want to interact. More impor­ tantly, they are more likely to grunt before approaching a female they have threatened earlier than one they haven't. It's as though they want to say, 'Don't worry, my intentions are strictly friendly.' Reconciliations may also occur between males at times when alliances form an important part of their social strategies. When a male gelada acquires a breeding group of females by defeating an incumbent harem male, his position is far from secure. He was only able to wrest the harem away from the former incumbent because the females' loyalty to that male was weak. The two males will have fought a veritable Battle of the Titans, and probably have sustained serious injuries from each other's two-inch-Iong canine

teeth in the process. But despite the intensity and bravado of the fight, the decision on who wins and who loses lies in fact in the hands of the females; it is they who ultimately decide whether or not to desert their current male in favour of the rival, and they may decide in his favour even though he has been losing the long-run­ ning battle with the rival. The problem for the rival if he wins is that, since the females are clearly willing to desert one male in favour of another, they may be just as willing to desert the second, should he prove no more to their taste than the first. There are always plenty of onlookers at a take-over battle who would be willing to give it a go at the slightest hint that the loyalty of the females might be in doubt. And it does sometimes happen. Caught in this awful bind, victorious males work fast to estab­ lish an alliance with their defeated predecessor. The two males may have spent a whole day, sometimes two, locked in intermit­ tent and often bloody combat, but once the decision has been made and the defeated male has accepted the verdict, the winner sets about building a new relationship with him. This involves making a number of tentative overtures. The new male approach­ es the defeated male in a non-aggressive, almost submissive way. The defeated male is, of course, suspicious at first. He has just received the beating of his life at the hands of this thug. He is sore, exhausted, and nervous of another unprovoked attack. But he wants to stay on in the unit because the current batch of infants represents his last throw of the reproductive dice, and he would like to see them through at least to the point where they can sur­ vive on their own.

- 26 -

So the two males have interests in common: the new male would like to have the old male's support against further take­ over attempts (at least over the initial period while his position is still uncertain), and the old male would like to stay on in order to protect his offspring. After one or two false starts, the deal is struck surprisingly quickly; there is a simple ritual of reconcilia­ tion in which the old male reaches through to touch the new male's penis as the latter presents his rear. Then the two males groom each other with the kind of enthusiasm reserved for the aftermath of patched-up quarrels. And after that, the two are as

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

inseparable as twins when it comes to defending the females against outsiders. It is the subtleties and complexities of these interactions that give the societies of monkeys and apes their special quality. We can watch the soap opera of daily life unfolding before us, and empathize with the stratagems and counter-stratagems as they evolve. It all looks so familiar, so reminiscent of everyday life in our own societies. This is our primate heritage, our common evolu.;. tionary experience. It has important implications for the way our minds are designed, and so in turn for the design of our brains.

fused, it's small wonder that (according to the latest surveys) 4 8 per cent of the population of the USA still believes that the biblical story of Genesis is literally true. This all sits very incongruously with the fact that Darwinian evolutionary theory3 is widely recognized to be the second most successful theory in the history of science (after modern quantum physics). Not only has it been extraordinarily effective in explain­ ing why the biological world is as it is, but it has also proved remarkable in its ability to continue generating new questions to stimulate and guide empirical research. Yet, a century and a half after Darwin first proposed his theory, most people's views of the biological world are still heavily coloured by opinions that were current during the eighteenth century, long before Darwin was even born. Darwinian evolutionary theory is so fundamental to our under­ standing of the events I describe in this book that I am obliged to pause in my story to ensure that we all start with the same clear understanding of what the Darwinian account entails. The popu­ lar literature is so full of misconceptions and, in some cases, plain fictions that misunderstandings can easily arise. (Readers who feel thoroughly at home with the modern Darwinian perspective may wish to skip the rest of this chapter and go straight to the next.) One trivial example is the fact that most people are surprised to discover that Darwin did not invent the theory of evolution. In fact, biologists had come to accept the idea of evolution long before Darwin published his seminal work. The second half of the eighteenth century had witnessed a number of significant chal­ lenges to the overpowering influence of the biblical world view. One was the realization that the diversity of life on the planet

A Darwinian Detour

When Charles Darwin published his landmark book On the Origin of Species in 1 8 5 9 , he set in motion a revolution that has radically altered our understanding . of the living world. So it seems odd, nearly one hundred and fifty years later, to be remind­ ing you that the natural world is a Darwinian world. Yet the lessons of Darwin's revolution in the biological sciences remain widely misunderstood, not just by the layman but also by many scientists outside organismic biology. 2 With even scientists con2.

The biological sciences can be divided into three rather broad levels. Organismic

(or whole organism) biologists study the emergent properties of an animal's behav­ iour: this includes such topics as as ecology, animal behaviour, population biology and evolutionary processes. Infra-organismic biologists study the processes that make an animal tick: among these can be numbered physiology, cell biology, anatomy and embryology.

Finally, molecular

biologists study the chemical

processes that produce organisms: this newest and in some ways most successful of the sub-disciplines of biology focusses on the way DNA and the other components of the genetic system build cells in the great miracle of life. Although all three lay­ ers of biological science are united into a single structure by Darwin's theory of evolution by natural selection, they do not all need to worry about its details to the same extent. While organismic biologists would find it very difficult (if not impos­ sible) to do their research in the absence of Darwin's theory, infra-organism and molecular biologists are much less dependent on it (and indeed, a handful of them even manage to adopt an overtly anti-Darwinian position without compromising their science). A theory of evolution is just not needed to study how a cell works, even though it is all but imposssible to understand the behaviour of the organism of which the cell is a part without the benefit of evolutionary theory. Laymen are sometimes confused by this, assuming that the fact that a cell biologist can do without Darwin must mean that Darwin is wrong. Cell biologists undoubtedly do

- 28 -

better biology with the benefit of Darwinian theory as a working framework, but they can get by quite adequately without it, at least for the time being. Whether, as our knowledge of cell biology grows, they will always be able to get by without it remains to be seen.

3 · The modetn theory is Darwinian but it is nOt, strictly speaking, Darwin's. It

contains many elements that Darwin would not have recognized and could not

have known about. Over the last ISO years, biologists have built on Darwin's original insights to produce what, by any standards, is one of the most remarkable and comprehensive theories in science.

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

could be more easily explained as a consequence of evolution than by the conventional biblical creation story. The first concerted attempt to provide a general theory of evolution was made as ear­ ly as 1 809 by the renowned French biologist Jean Baptiste de Monet, Chevalier de Lamarck - more commonly known to suc­ ceeding generations of biologists as plain Lamarck. Darwin's con­ tribution to this debate was not to prove the theory of evolution, but to provide a mechanism - natural selection - that could explain why evolution took place. Lamarck's views were premised on the Aristotelian scala natu­ ra, the 'scale of nature', sometimes known as the Great Chain of Being. This curious hangover from the ancient Greeks supposed that all life formed a natural hierarchy which began with things like insects and worms - the original Greek version actually start­ ed with earth and water on the lowest rungs of all - proceeded through more advanced forms like fishes, reptiles and birds, to end with mammals, humans and, at the pinnacle of all, the gods. Adopted more or less wholesale by the early Christian Church with the angels and, on the final rung, God himself substituting for the ancient deities - the Great Chain of Being coloured the way everyone in post-medieval Europe thought about the biologi­ cal world. Lamarck and his contemporaries built these ideas into their theories of evolution, supposing that each species begins life on the lowest rungs of the ladder and, over long periods of time, gradually progresses up through the hierarchy of life in response to a natural unfolding of some inner force. Darwin turned all this on its head by insisting that there was no natural progression up an evolutionary ladder. In fact, no species - not even humans - could be considered as better or worse than any other. There is only one biological standard by which a species can be judged, namely its success at reproducing itself. We are all, bacterium and human alike, equally 'good' because we are each sufficiently well adapted to our particular circ*mstances to thrive and reproduce. The fate of all species is eventually either to become extinct or to be transformed into new species. But in either case, it is natural selection - reflected in the individual organism'S ability to survive and, more importantly, reproduce -

that drives these changes, not some internal biological principle or 'life force' as Lamarck assumed. Darwin's theory has two important lessons for us when we come to think about behaviour. One is that evolutionary change is driven by animals' need to adapt to changing circ*mstances. The geological sciences have revealed that the earth's climate has been in a constant state of flux, oscillating between the overheated and the frigid with almost monotonous regularity. In the 65 million years since the dinosaurs became extinct, for example, average global temperatures have dropped by an astonishing 1 8°C. Even the Antarctic was once clothed in dense forest. Associated with these climatic changes have been dramatic changes in vegetation and fauna. Most of these shifts in climate have been triggered by the chang­ ing shape and distribution of the continental masses as they have slithered around on the surface of the planet's soft inner core. Other factors have also played a role, including long-term changes in the distance between the earth and the sun and in the tilt of the earth's axis. Interspersed throughout the 4 5 0 million-year history of life on earth there have been five (possibly six) episodes of mass extinction when virtually all existing life forms were wiped out.4 The handful of survivors provided the seed corn for a whole new series of evolutionary developments that took life on earth off in an entirely new and often quite arbitrary direction. The way in which climatic changes drive evolution is nicely illus­ trated by the fact that deep-sea creatures like sharks (whose origins predate the dinosaurs) have remained virtually unchanged for hun­ dreds of millions of years, whereas species like antelopes and humans (both of whom are of very recent origin) have changed dramatically in appearance over a very short period of time. Unlike land habitats, the deep-sea environment tends to be much less 4·

There is growing (but still controversial) evidence that these mass extinctions

were associated with the impact of comets or large asteroids. The dust and water vapour thrown up into the atmosphere by these collisions is presumed to have blocked out the sunlight and created what amounts to a 'nuclear winter'. The last of these mass extinctions occurred 65 million years ago and saw the demise of the dinosaurs.

Grooming, Gossip and the Evolution of Language

Into the Social Whirl

affected by global temperature changes; consequently, deep-sea creatures face much the same conditions now as their ancestors did 200 million years ago. In contrast, the environments faced by land animals have changed dramatically as a result of sometimes cata­ strophic changes in the world's climate and vegetation. Darwin's theory can account for changes of this kind because, unlike Lamarck's species-based theory, it assumes that the indi­ vidual is the basic unit of evolution. It is the individual that repro- . duces or doesn't reproduce, and the individual that passes on its particular traits. 5 Where earlier biologists viewed species as ideal types (clones, you might say), Darwin and his colleagues began to see the species as simply a collection of sometimes quite variable individuals who shared a number of key traits. That variation was the potential that allowed species to evolve, though evolution would only occur if natural selection made change advantageous. The other key lesson of the - Darwinian approach is that in real life nothing comes for free. Evolutionary change is not automatic; it always occurs against a gradient of stability imposed by the organism's natural biological coherence. Change along one dimension of an organism's biology (let'S say growing taller) always incurs costs. One reason is that the change may throw oth� er aspects of the system out of kilter: a taller individual tends to be more gangly, and so less fast at escaping predators. A second problem is that any change (such as growing taller or developing a bigger brain) costs energy: bigger individuals need more food to fuel their bigger bodies. For evolution to occur, the benefits to be gained from changing a character have to exceed the costs. When there is no advantage to change, the costs of making changes act as a stabilizing force, selecting for constancy. To understand evo­ lutionary change, we have to understand both the costs and the benefits of any particular course of action. The one exception to this arises in those cases where the charac5. Strictly speaking, as Richard Dawkins has reminded us in his book


The Selfish

it is the gene that is the fundamental unit of evolution: evolution occurs

because certain genes are passed on to the next generation more successfully than other genes. However, speaking of individuals rather than genes is a convenient shorthand.

ters concerned are protected from the immediate impact of the forces of selection. This happens when different versions of a gene produce the same effect. When there is no selection pressure for or against a change, genetic traits can drift as a result of chance events that affect which individuals happen to reproduce. Consequently, a population that splits into two halves which remain reproduc­ tively isolated for a long enough period will accumulate minor genetic differences. The theory of neutral selection, originally pro­ mulgated by the Japanese geneticist Motoo Kimura in the I970s, has proved a valuable tool for evolutionary biologists because it allows us to determine the length of time since any two species had a common ancestor. It is simply a matter of determining the num­ ber of mutations by which their DNA differs and multiplying this by the average rate at which spontaneous mutations occur. This is how we know that humans and chimps shared a common ancestor who lived some time around 5 to 7 million years ago. One final point needs to be emphasized. Many people find Darwinism disturbing because they confuse it either with Social Darwinism, with its associations with the eugenics movement of the early I 900s, or with genetic determinism. The first of these confusions is frankly bizarre, because despite its name Social Darwinism had very little to do with Darwinism; it was largely the brainchild of the social philosopher Herbert Spencer, aided and abetted by the distinctly anti-Darwinian founder of genetics Francis Galton (ironically a cousin of Darwin's). Irrespective of whether Darwin was himself a Social Darwinist, the movement's basic philosophy - maintaining the purity of the species - was dis­ tinctly Lamarckian in conception. Indeed, by the I920S it became apparent that Darwinism had pulled the intellectual rug well and truly from under it. The Social Darwinists (and their curious after­ image, the Nazis in the I9 3 os) were motivated by fears that the excessive fertility of the socially unfit underclasses was diluting the viability of the human species. In fact, the underclasses were behaving in a respectably Darwinian fashion: reproducing as fast as they could to ensure that their genes made it into the next gen­ eration, despite the high mortality rates their children suffered thanks to their grinding poverty. If anything, they were dutifully

Grooming, Gossip and the Evolution of Language

increasing the range of diversity on which natural selection has to work, thereby reducing the likelihood of our species' extinction in the long term. Heaven forbid that we should all have ended up as clones of the upper classes! The bugaboo of genetic determinism is for many a modern-day version of Social Darwinism. But once again, the problem is large­ ly one of misinformation - sometimes exacerbated by a refusal to listen. Evolutionary biology has no preconceptions about the genetic determinism of behaviour, even though at some point genes must be involved. Many features of an animal's behaviour can be explained in terms of strategies that maximise genetic fit­ ness, even though the behavioural rules they use are learned or culturally inherited. Learning is just another example of a Darwinian process: it is differential survival of traits (behavioural rules in this case) as a result of selection. When animals make decisions about how to behave, they do so on the basis of past experience, and in the light of the costs and benefits of particular courses of action. Their decisions may well be guided by genetical­ ly instilled intuitions about how fitness can be maximized, but they do not act blindly in response to inner drives beyond their control; advanced organisms can act or hold back, according to circ*mstances. Every day of their lives animals make decisions about whether the risks entailed in behaving in a particular way are worth what they will gain by it. So much for our Darwinian detour. Now back to monkeys.



The Irnportance of Being Earnest

What makes groups of primates different from groups of other species is their 'busy-ness'. Every waking moment has something of significance going on. Here is a grooming, there a squabble that is sorted out by an ally, elsewhere a subtle deception - the whole welded together by a constant watchfulness, taking in who-is­ doing-what-with-whom. At the root of it all, however, are the long sessions of grooming so peculiarly characteristic of primate societies. Here, in some imperfectly understood way, lies the key to the processes that give primate societies their cohesion and sense of belonging. A Sense of Touch

Grooming takes up a great deal of a monkey's time. In most of the more social species, the grooming . of other individuals accounts for about 1 0 per cent of an animal's day. But in some species, it can take up as much as 20 per cent of the animal's time. That is an enormous commitment, given that finding food is a time-consum­ ing activity at the best. We know that grooming is intimately related to an animal's will­ ingness to act as an ally of another individual. At least among the Old World monkeys and apes, the time devoted to grooming during the day correlates roughly with the size of the group. This makes a certain amount of sense. If grooming is the cement that holds alliances together, then the more time you devote to grooming your ally, the more effective that alliance will be. And since alliances will be proportionately more important to you the larger the group gets, it makes sense to invest even more time in grooming your allies. But why it should be so effective in this respect remains unclear.

Grooming, Gossip and the Evolution of Language

The Importance of Being Earnest

Prosimian primates (lemurs and galagos) also spend a great deal of time grooming, and Rob Barton of Durham University has shown quite convincingly that, when lemurs groom, it is mainly for reasons of hygiene. Grooming by other individuals tends to concentrate on those parts of the body (the scalp and back) that an animal cannot reach for itself. Social grooming is in these cir­ c*mstances a helpful tit-for-tat arrangement - quite literally a case of 'you scratch my back and I'll scratch yours'. At one level, grooming is simply a pleasurable experience. Studies of captive monkeys have shown that grooming makes them more relaxed, reducing their heart rate as well as other external signs of stress. They sometimes become so relaxed that they fall asleep. In fact, we now know that grooming stimulates the production of the body's natural opiates, the endorphins; in effect, being groomed produces mildly narcotic effects. The chemicals known as enkephalins and endorphins (collec­ tively known as endogenous opiates) are produced in a region deep within the brain called the hypothalamus. They play an important role in our everyday lives as the brain's own painkillers. Their chemical signature is virtually identical to that of the more familiar opiate drugs such as opium and its derivative morphine, and they behave in rather the same way, by dampening down the pathways in the nervous system that produce pain signals. It is because morphine and other opiates are so similar in chemical structure to the endorphins that we so easily become addicted to opiate drugs. The various sites scattered throughout the brain that act as receptors for endorphins readily take up artificial opiates. However, we don't become addicted to endogenous opiates in quite the way we do to opium and morphine because the brain only produces its natural opiates in relatively small quantities. Tens of millions of years of evolution have ensured that the sys­ tem only produces what it needs. Unfortunately, of course, we can flood the body with artificial opiates, thereby inducing the hyper­ narcotic effects associated with these drugs. The opiate system probably evolved in order to dampen the effects of pain caused by injury, and as such it forms part of the body's carefully balanced system for handling pain. Pain is impor-

tant because it warns us that something fairly drastic is happening (or is about to happen) . In evolutionary terms, it has the essential function of giving us sufficient warning to allow us to remove the offending object (or ourselves) before serious damage is done. The fast pain channels of the nervous system relay the information to the brain that the skin has been broken, so that evasive action can be taken. But once action has been taken and the danger averted, the bro­ ken skin remains broken and painful. This is where the endor­ phins come into play. One of their functions seems to be to damp­ en down the pain system to enable you to get on with the more important business of living once you have removed yourself from the dangerous situation. Were it not for the brain's opiate system, you would spend your time rolling around in agony to no great purpose. Because the opiates are released into the bloodstream, they act slowly. They take time to build up and time to dissipate, unlike the nervous system's fast-acting pain channels. It seems to be yet another example of the body's finely tuned balancing act between conflicting systems. The endorphin system seems to respond best to the monoto­ nous repetition of low-level stimulation. The steady pounding of jogging is just the kind of stimulus that seems to be most effective at producing endorphins. Indeed, regular joggers actually become addicted to their sport because it produces mild opiate highs for them. When they are prevented from having their daily fix, they suffer all the symptoms of opiate withdrawal: tenseness, irritabili­ ty, sometimes even a mild form of the shakes. Opiate highs of this kind are easy to induce; any kind of monot­ onous stress on the body produces them. Animals in captivity have long been known to pace up and down endlessly, and it has recently been shown that this stimulates opiate production - as good a way as any, I suppose, of alleviating the boredom of being caged. Workaholics probably produce the same effect, since psy­ chological stress seems to be just as effective as physical stress. In this case, the intense concentration and high levels of brain-cell activity seem to act in much the same way as the jogger's pound­ ing along the streets. Like joggers, workaholics suffer the classic . - 37 -

Grooming, Gossip and the Evolution of Language

The Importance of Being Earnest

withdrawal symptoms when prevented from working. The body even seems to anticipate the need for opiates. Marathon runners show a marked increase in the level of opiate production a day or so before a big race. Women generate partic­ ularly high levels of opiates during the last three months of preg­ nancy, an invaluable preparation for the final moments of labour . Opiates, then, play a very important role in the body's chem­ istry. The surprise for us is to find them turning up in the context of grooming. Research has shown that monkeys who have been groomed have higher endogenous opiate levels than those who have not. Moreover, minute doses of morphine were sufficient to suppress grooming behaviour; when the brain's opiate receptors were flooded, the monkeys were no longer interested in grooming. And when the natural production of endogenous opiates was blocked by giving the monkeys small doses of naloxone, a drug that neutralizes morphine, they were · more irritable than normal and kept asking to be groomed by their cage-mates. The mechanisms that make grooming an attractive activity seem directly related to its ability to induce a state of relaxation and mild euphoria. This, if you like, is the reinforcer that makes monkeys willing to spend so much time in what would otherwise seem a pointless activity. Even though grooming ensures that the fur is cleaned and the skin kept free of debris and scabs, the time devoted to it by species like baboons, macaques and chimpanzees far exceeds that actually needed for these simple purposes. However, inducing opiate highs is unlikely to be the evolution­ ary reason why monkeys groom so much. It may be fun to space out on mutual mauling, but in a world full of predators it can be a dangerous thing to do. Opiate highs are surely the mechanism that encourages animals to spend so much time grooming, but something more useful is needed as the evolutionary selection pressure to drive it along. That selection pressure seems to have been something to do with cementing bonds of friendship. It is quite common for natural selection to hijack entire motiva­ tional and behavioural systems in this way in order to use them for other purposes. One example is the way feeding and diving behaviour has become incorporated into the courtship rituals of

ducks and grebes. The most remarkable example, however, is the way the three rearmost bones of the reptilian jaw were used to form the three little bones in the middle ear in the earliest mam­ mals as they evolved out of their reptile ancestors. The reptile's lower jaw consists of five bones on either side. When the earliest mammals first evolved from reptilian ancestors, the front two bones on each side fused to form the jaw as we now see it in our­ selves and all other modern mammals. The other three gradually decreased in size and became part of the hearing system. They now form a chain of tiny bones that transmits sound waves from the eardrum at the junction of the outer and middle ear to the cochlea, the organ in the inner ear that registers and encodes sounds as neural messages to the brain. In fact, this shift in use is not as surprising as it might seem, because the reptile's jaw is part of its hearing system, helping to transmit vibrations from the ground to its hearing organs. When you think about it, borrowing the jaw bones to form the ossicles was a very natural thing to do. The capture of grooming's motivational system seems to corre­ late in each case with the evolution of relatively large groups and the invasion of a more terrestrial, open habitat. The increase in group size seems to be a direct response to the invasion of habitats where the animals are more exposed to the risk of predation. Arboreal forest monkeys, like the Old World colobines and all the New World monkeys, tend to live in relatively small groups. But baboons, macaques and chimpanzees are more terrestrial and pre­ fer the more open habitats on the forest edge. Here the risk from predators is much higher, partly because it is easier for predators to approach prey undetected and partly because there are fewer trees into which the animals can escape. These species solve this problem by being larger than the average primate and, more importantly perhaps, by living in unusually large groups. The need to live in large groups raises a host of problems, how­ ever. There are many direct costs, not the least being the need to cover a proportionately larger area each day in order to provide the same quantity of food per group member. That, of course, means you have to travel further, which in turn will expose you to greater risk of predation out on the open plains. With all that

Grooming, Gossip and the Evolution of Language

The Importance of Being Earnest

extra travel, you will burn up more energy, which in turn means you have to eat more to provide all that extra energy, which in turn means you have to travel further . . . The whole process soon becomes a vicious circle. Eventually, of course, this particular circle does grind to a halt, but in the meantime you have added quite a bit extra on to the average working day for every extra body you want to have in your group. The most serious costs are, however, the indirect ones. These come in the form of heightened levels of competition for food and sleeping sites and increased levels of harassment and stress. With so many more bodies milling around in the same fig tree, it's inevitable that there will be more competition for the best figs. The bigger thugs will get their way, and the lesser thugs will be marginalized into the less enticing patches on the outer edge of the tree. There the figs may be fewer in number, while those that are available may be of poorer quality (parasitized by wasps, nibbled at by squirrels); in addition, any position on the outer edge of the tree is more vulnerable to birds of prey such as monkey-eating eagles. The edge of the group is never a good place to be. The press of bodies around the best and safest feeding or sleep­ ing sites may lead to frequent trampling underfoot, both metaphorically and literally. The constant hassle of being moved on, of being harassed by those anxious to reinforce an ambivalent dominance relationship - all these add up throughout the day to a considerable strain on the nervous system of lower-ranking ani­ mals. And, of course, the lower your rank and the larger the group, the more individuals there are to harass you. Even being harassed just once a day by each member of the group can amount to a great deal of stress for the lowest-ranking animal in a group of 3 0-4 0 adult baboons. Harassment of this kind takes its toll on the animal's bodily systems. It seems that psychological stress is every bit as effective as physical pain in stimulating opiate production. Persistent harassment can have a debilitating effect on the immune system, producing clinical depression as well as increased susceptibility to disease. But it also has an unexpected side-effect on the repro­ ductive system, leading to temporary infertility.

It turns out that the endogenous opiates we met earlier are also implicated in the control of puberty and the menstrual cycle, though why they should be thus involved in reproduction remains obscure. Barry Keverne and his colleagues at Cambridge University have demonstrated that the stress of being low-ranking produces high levels of endogenous opiates in females, and this in turn produces infertility. It seems that the harassment meted out by other group members is as stressful as anything else and precip­ itates the production of endogenous opiates, which then wreak havoc on the reproductive system. The endocrinology of this process is now quite well understood. The opiates released from the brain block the production of the hormone GNRH (gonadotrophic releasing hormone) in the hypo­ thalamus. In the absence of the chemical kick provided by GNRH, the pituitary gland near the base of the brain does not produce leutenizing hormone (LH). LH is the hormone that stim­ ulates the ovaries to switch over from the production of proges­ terone to the production of oestrogen, so triggering ovulation. Hence, with endorphins blocking the production of GNRH, the cascade of hormonal events that produces ovulation fails to occur. The result is an anovulatory menstrual cycle - a cycle that appears to be quite normal (though perhaps slightly longer than usual), but does not involve the release of an egg from the ovaries. Just how dramatic the implications of this can be is illustrated by what happens in gelada baboons. Our field studies of wild populations of this species have shown that low-ranking females experience rather low levels of harassment: on average, about two mild threats a day from each female in the group, with only about one of these escalating into a serious attack in an entire week. Even then, these could not be considered as much more than spats on a street corner, certainly nothing remotely as traumatizing as a serious mugging. Yet these seemingly trivial levels of harassment are sufficient to result in about half an offspring less over the course of a lifetime for every rank place that a female drops. This may not sound too much, but when you realize that the maximum from which you start is only about five offspring, this represents a loss of about 1 0 per cent of your lifetime reproductive output for

Grooming, Gossip and the Evolution of Language

The Importance of Being Earnest

each drop in rank. By the time she is in a group of 1 0 females, the lowest-ranking female can expect to be functionally infertile. Because the low-ranking females among our gelada had slightly longer menstrual cycles than high-ranking ones, we suspected that opiate-driven suppression might be the cause. At the time, we had no more than circ*mstantial evidence to support this hypothesis. One obvious hint was the fact that low-ranking females suffered more harassment than higher-ranking ones. However, our suspi­ cions were later confirmed by a study of captive gelada at New York's Bronx Zoo carried out by Colleen McCann; she was able to show that low-ranking females do indeed have higher levels of circulating endogenous opiates, as well as higher frequencies of anovulatory cycles, than high-ranking females. An even more extreme case has been studied by David Abbott at the Wisconsin Primate Center in the USA. The marmosets and tamarins are tiny South Ameri when they have acquired a theory of mind; then they can manip­ ulate you something rotten. Psychologists have developed a crucial test for theory of mind, known as the ' false belief test' . It asks the key question: is the child aware that someone else can hold a false belief (or at least a belief

Grooming, Gossip and the Evolution of Language that the child supposes to be false ) ? The now-classic example of this is the so-called 'Sally and Ann' test. Sally and Ann are two dolls who are presented to the child and formally introduced. The child is shown that Sally has some sweets, which she then place s under a cushion on a chair. Having done this ( perhaps with the child's help ) , Sally leaves the room. Then, while Sally is out of the room, Ann takes the sweets from under the pillow and puts them in the pocket of her dress. When Sally comes back into the room, the child is asked, 'Where does Sally think the sweets are ? ' Up to the age of four, children invariably answer, 'In Ann's pocket. ' But after about four-and-a-half, they invariably say, 'Under the pillow,' adding with conspiratorial glee, 'but they aren't there ! ' I n another classic test, a child i s shown the cardboard tube in which the sweets known in Britai n as Smarties and in the USA as M & M s are sold. When the child is asked what he or she thinks is inside, the answer is of course ' Smarties'. The top is then removed, and the child is shown that inside the tube are some pencils rather than the sweets he or she expected. Finally, the child is told that its best friend is j ust about to be brought into the room and shown the same tube; what does the child think its friend will say when asked what is in the tube ? Up to the age of four, children invari­ ably say, 'Pencils'. But after four or so, they are aware that others can hold beliefs different from their own, and so they will say, 'Smarties' . Although the appearance of ToM strikes us as a very sudden p rocess in children, in fact it is the outcome of a long process o f intellectual experimentation . From a very early age, children become aware that other obj ects in the world are capable of doing thi ngs for you. You can ask them to give you things you want; sometimes you can even blackmail them into doing so by whining and grizzling until you eventually break down their resilience. Experience with the different kinds of obj ects they encounter leads children to conclude that some of these objects are animate and others inanimate. At first they fail to distinguish clearly between a person and a doll, apparently believing that dolls share with humans all the qualities of volition they see in humans. But with experience, they come to separate these categories out. - 86 -

The Ghost in the Machine By a bout the age of three, children are into what has been te rmed 'belief/desire psychology' . They can recognize that other individuals have wants and desires that are similar to those they experience. Over the course of the following year, they use this knowledge to build up a picture of how other individuals tick. It's a very complex task for children, and of course they make many mistakes. It is now clear that understanding the social world is a far more difficult task for children to master than understanding the physical worl d. Th is realization has turned the influential psychologist Jean Piaget's theories of development on their head. Piaget's theories have dominated our thinking on child intellectual development for the better part of half a century. He was much exercised by the desire to explain children's growing understanding of their world . Like all his contemporaries, Piaget assumed that our brains are there to process information a bout the world, and hence that coming to an understanding of the underlying features of that world was the most difficult task a child has to achieve. The rest, he supposed, was plain sailing. Piaget, it seems, was deceived by the extraordinary skills of chil­ dren in the social domain into thinking that these skills were not problems of any great consequence. Children obviously acquired them rather effortlessly early on. Piaget's mistake was understand­ able: few at that time appreciated j ust how complex our social world actually is. Social skills are a matter of great urgency for a child: its very survival depends on them . So long as the minor matters of seeing and hearing develop early on, the more complex tasks like understanding the conservation of quantities and vol­ umes - the problems Piaget made so m uch of - can wait until after the child has managed to negotiate its way through the social maze into which it has been born. Is There Anybody Else Out There ?

Of course, Piaget was not all wrong. He noted, for example, that children are initially self-centred and only gradually come to break the mould of their egocentric universe to take on another

Grooming, Gossip and the Evolution of Language

The Ghost in the Machine

individual's perspective. Although the language he used is differ­ ent, Piaget's understanding of at least this part of the story seems to have been basically right. We are born without theory of mind; with time, we gradually acquire ToM and are able to understand how others feel and think, and so to use this knowledge in our social interactions . So if we set the standard for human-ness by what we see in adult humans, it follows that, strictly speaking, human babies are not fully human, and do not become so until the age of about four. Of more interest still is the fact that some people never develop ToM. We now refer to these individuals as a utistic, a medical syn­ drome first identified as recently as the I 940s, although it has sure­ ly been around for very much longer than that. Autistic people the great maj ority of whom are males, indicating a clear genetic component to the condition - vary in the degree of their impair­ ment. Some are very severely handicapped, never develop language and show no ability to interact socially with others. Others develop language, but remain social isolates. Sufferers from its mild form, known as Asperger's Syndrome, can seem quite normal aside from their social gaffes and occasionally bizarre behaviour. Only the most subtle psychological tests allow us to recognize that these individuals may not have full theory of mind. Autistic people are characterized by two key deficits. O ne is a consistent failure to pass false belief tests . The other is an appar­ ent inability to engage in pretend play. The psychologist Allan Leslie has argued that these two characteristics are intimately related to each other. Because they do not realize that other peo­ ple can hold beliefs that are false ( or at least beliefs that they sup­ pose to be false ) , they are unable to imagine other worlds or that the world could be other than as it now is. Consequently, they cannot engage in fictive play. They will not, for example, run through the motions of a dollies' tea party; dollies are not living organisms, so how could they possibly do things that real people do ? Nor will they pretend to be asleep in order to play a j oke on someone else. Nor will they tell a deliberate lie, because lying requires you to be aware that the other person might not know all you know. An autistic person simply assumes that the world is

transparent, that he and his a udience share the same information. In effect, autistic people take the world exactly as it comes. One consequence is that they fail to recognize the richness of meaning often buried in our use of language. Here is a classic example reported by the mother of an a utistic teenager. Before leaving the house to visit a neighbour across the road, she told her autistic son that if he wanted to come on over later, he was to be sure to pull the door behind him. An hour or so later he did exactly that, having first wrenched the front door off its hinges. This story should remind us of j ust how much of the meaning in our conversations depends on the listener reconstructing the speaker's mind-state. Autistic people simply cannot do that, because they do not realize that what someone else has in mind may in fact be different from the normal meaning of the words they are using. In fact, almost all of our conversational exchanges are metaphorical or require interpretation by the hearer. We com­ monly speak in telegraphic fashion, providing j ust the key points and assumi ng that the listener can fill in the bits and pieces to make sense of what we say. Here is a classic example of the deep layers of interpretation with which our conversations are often imbued:

- 88 -


I'm leaving you ! Who i s she ?

Clearly, his statement could have a dozen different and equally legitimate interpretations that hinge on the context and the past history of the individuals concerned. Yet even with so brief a script, we have no difficulty in identifying instantly the correct interpretation from her elliptical reply. We can at once fill in all the details and provide a great deal of probable background infor­ mation. Autistic people could not do that. Those suffering from the milder Asperger's condition often manage to cope with social sit­ uations, so passing false belief tests. But they do not do so by mind-reading in the way that we do. They cope by what the psy­ chologist Francesca Happe refers to as ' hacking it'. In other words, they are smart enough - being of normal intelligence - to

Grooming, Gossip and the Evolution of Language work out rules of thumb that allow them to make the right deci­ sion in a social situation nine times out ten. But they appear to have no idea why these rules of thumb work, merely that they do. The following musical analogy will perhaps give you a feel for their problem. Much though I like music, I happen to be all but tone-deaf. I can recograize Mozart's Serenade in E flat when I hear it, though I couldn't in all honesty tell you whether it's in B flat, A minor or for that matter G sharp. But a musician will know at once which key it's in, even though he's hearing it for the first time. I've learned to recognize the tune as a specific series of notes, but I have no real idea why the key is called E flat and it probably wouldn't matter if I did: even if I tried, I couldn't generalize the rules of recognition I've learned to other pieces. So it is with As perger' s. I remember the mother of an Asperger's child observing that her son, then about twelve years old, knew that people had friends . and that he did not, yet he had no idea how you went about obtaining a friend. Did you buy them at shops, or did you simply say to someone - anyone - 'You are my friend ! ' It is heartbreak­ ing to encounter cases like this, because there is no way to get through to them the deep emotional bases of normal human rela­ tionships. They simply cannot comprehend what they are or how they work. Indeed, they would find my use of the word 'observe' in the first sentence of this paragraph very puzzling: the mother wasn't seeing anything at all! Yet it is worth reminding ourselves that these individuals are otherwise completely normal, and may even be of above-average intelligence. Asperger's people are often very good at mathematics, for example, probably because they can think clearly in the abstract and not become confused by emotional and other irrelevant associations. The obvious question at this juncture is how other animal species compare with us on the intensionality scale. Are we alone in the universe of the mind or is there somebody else out there ? The most likely place to find other species that have the same theory of mind abilities as we do is among our closest relatives, the apes. The realization that theory of mind is in some sense a key to understanding the human mind inevitably provoked considerable

The Ghost in the Machine interest in how monkeys and apes compare on the intensionality scale. The problem lies in identifying a critical test we can use as an infallible criterion. The earliest attempt at this problem was by the American psychologist Gordon Gallup. He argued that the ke y factor which characterizes us is that we can recognize our­ sel ves as a separate entity from the other individuals with whom we live ( in effect, Descartes' I think, therefore I am ) . Being self­ aware provides us with the capacity to reflect on our internal mental states, and then to relate these internal states to the observ­ able behaviour of other individuals. From this comparison, we can work out that other individuals also have mental worlds. Gallup designed a n ingenious test which he claimed allowed us to decide quite unequivocally whether or not another animal is self-aware. The test involves training an animal to use mirrors. Later, the animal is anaesthetized and a small dot of dye put on an area of bare skin on its face. The key question is: once it has recovered from the anaesthetic, will the animal realize there is something different about its face ? Can it demonstrate this by touching or picking at the spot of dye ? Over the past decade, extensive experiments of this kind have been carried out on chimpanzees, gorillas, orang-utans and sever­ al species of Old World monkeys, as well as on porpoises and ele­ phants. Despite a few contradictory outcomes, the consensus on these results is fairly uncontroversial. Chimpanzees readily solve mirror problems of this kind; orang-utans and gorillas seem to be reasonably competent (though many fewer of them have been tested ) ; b ut no monkey has yet passed a mirror test. The obvious conclusion everyone has reached is that great apes are self-aware, but that other members of the Primate Order ( and that appears to include the gibbons, the so-called lesser apes) are not. When simi­ lar experiments were tried with elephants, the animals apparently failed even to recognize the wall-size mirrors as mirrors, but tried to walk through what they thought was an open door. Although many have been tempted to conclude that the great apes have theory of mind and monkeys do not, some lingering doubts a bout Gallup's experimental design niggle at the back of one's mind. Why on earth should the ability to use a mirror be a

Grooming, Gossip and the Evolution of Language

The Ghost in the Machine

convincing test of self-awareness ? After all, monkeys and apes don't have mirrors in the wild, so why should the ability to recog­ nize oneself in a mirror demonstrate the ability to appreciate that you have an independent mental life ? Just what, in fact, do the results of the mirror tests tell us ? One obvious answer is that they merely tell us whether a species is smart enough to understand the physics of mirrors. It may be, for example, that technical problem-solving skills are in some way spun off social intelligence: one way this might come a bout is if advanced mentaji*zing skills require a lot of computing power, and this computing power allows you to solve the rather trivial physi­ cal problems associated with mirrors. Gallup's mirror test is certainly telling us something, but it j ust isn't clear what. We need something that is more diagnostic of mentalizing a bilities. One idea proposed by the psychologists Dick Byrne and Andrew Whiten is tactical deception . This is the name . given to occasions when one individual tries to exploit another by manipulating its knowledge of the situation. One example would be the j uvenile baboon Paul manipulating its mother in order to take Mel's tu ber; another would be Hans Kummer's young female hamadryas baboon inching her way to sit grooming with the young male of her choice behind a rock without arousing her harem male's suspicions (see page 2 3 ) . Kummer noted another form o f deception when studying gela­ da baboons in captivity. Gelada also form tightly bonded harems, and male harem-holders are as unhappy as hamadryas males a bout their females straying too far from them. One day, the harem male was removed from the group and put in a cage out of sight of the compound where the group lived, but not out of audi­ tory contact; the harem male could still hear everything that hap­ pened in the compound, and the rest of the group could still hear and interact vocally with him. With the old male out of the way, the group's young follower male made the most of his opportuni­ ties and began to mate with one of the females. Kummer noticed that when they did so, the male and the female both suppressed the raucous calls that gelada normally give at the climax of mating - calls that can normally be heard at distances of 1 00 yards or

more. Kummer referred to this as 'acoustic hiding' . S imi lar behaviour has been reported from captive chimpanzees. Fran s de Waal once watched a female who was mating surrepti­ ti o usly behind some bushes with a low-ranking male place her h an d over the male's mouth to prevent him giving the loud cop­ ulation calls that would have been heard by the dominant males on the other side of their large compound. In both cases, the offending couples seemed to be trying not to give the game away by vocalisations that would be clearly audible over the fence. This is tactical deception: the animals are appar­ ently trying to suppress clues about what is happening, so that other animals remain in ignorance. Another form of tactical deception was described by Sue Savage-Rumbaugh. She has been engaged in a long study of the two chimpanzees, Austin and Sherman, who were taught a n artifi­ cial keyboard language. Sherman was inclined to bully Austin, much to Austin's distress. One day, Austin discovered that Sherman was afraid of noises from outside their sleeping quarters, especially at night. Thereafter, whenever Sherman's bullying got too much to bear, Austin would race into the outdoor part of their accommodation, bang vigorously on doors and other objects, then rush back in whimpering and doing his best to look terrified . Sherman invariably responded with panic and would ask to cuddle Austin for comfort. Tactical deception provides something of a bench mark for advanced mentalizing abilities because it requires at least second­ order intensionality. In order to engage in tactical deception, an animal has to be capable of appreciating that its opponent believes something to be the case. By altering the information available to the opponent, the deceiver attempts to influence the beliefs of the opponent: I have to understand that, by behaving in a certain way, you will believe that I am doing something innocu­ ous. And that obviously involves holding a false belief of the kind we discussed earlier. Whiten and Byrne collected a large database of examples of tacti­ cal deception from the primate literature and from odd instances observed by colleagues while studying various species. Their most

Grooming, Gossip and the Evolution of Language

The Ghost in the Machine

interesting finding for us is that instances of tactical deception are virtually absent from the Prosimians (lemurs, galagos, etc . ) and rare among New World monkeys. They are common among the socially advanced Old World monkeys ( baboons, macaques), but most of the instances reported to them come from chimpanzees (with a handful from the other, much less intensely studied great apes ) . In fact, Dick Byrne later compared an index of the frequency with which tactical deception was reported among the species on their database with my index of relative neocortex size and found a very good fit. Species with large neocortices and complex soci­ eties, such as chimpanzees and baboons, were much more likely to engage in tactical deception than species such as African colobus monkeys or South American howler monkeys, which had much smaller neocortices and rather small social groups. It seems that a minimum of computing power is necessary to think through the complications involved in tactical deception. Following the logic of Byrne's analysis, it occurred to my Polish colleague Boguslaw Pawlowski and I that if the Machiavellian Intelligence hypothesis really did work, we should be able to show similar correlations between neocortex size and features like the sta­ bility of the male dominance hierarchy. We reasoned that, in species with large neocortices, low-ranking males would be able to take advantage of more subtle social strategies, such as tactical decep­ tion, to circumvent the dominance of the higher-ranking males. In the normal course of events, high-ranking males are able to monop­ olize females during the mating season, so preventing lower-ranking males from mating. The result is commonly a straightforward rela­ tionship between a male's rank in the hierarchy and the frequency with which he is able to mate or sire offspring. Pawlowski and I reasoned that, if low-ranking males with big brains could exploit loopholes in the system, we should find that the relationship between male rank and reproductive s uccess becomes less strict as neocortex size increases. This is exactly what happened. Since one of the strategies males are using in these contexts resembles tactical deception, our results reinforce Byrne's findings and confirm that the use of tactical deception actually does have functional consequences, in this case by affecting the

reproductive success of males and so ultimately their genetic fit­ n ess (their contribution to future generations ) . Byrne's findings o n tactical deception and ours o n male mating strategies provide strong behavioural evidence to support the M achiavellian Intelligence hypothesis. It suggests that the ability to use subtle social strategies and to exploit loopholes in the social context depends on how much computing power you have avail­ able in your brain. These findings do not tell us, however, how the differences we see between species relate to differences in levels of i ntensionality. All we know is that chimps can do better than baboons, and baboons can do better than howler monkeys. But where do they all stand in terms of their a bility to think reflexively about the contents of other individuals' minds ? Is a ba boon better than a howler monkey because it can imagine what it is like to be anoth­ er baboon, or simply because it can carry out more complex cal­ culations about the consequences of a social act? At this point, we reach the limits of our current state of knowl­ edge. No one has so far been able to relate levels of intensionality to particular behaviour patterns in any detail. We do, however, have some clues that at least point us in the direction of the likely outcome. They are mostly anecdotal observations, but they are interesting for all that. Vicki, the chimpanzee who was raised by the Hayes family al ongside their own child during the 1 9 5 0S, was once observed to walk along pulling a piece of string behind her. At first sight, this seems innocent enough. But when the end of the string reached a step between two floor levels, she stopped and exhibited signs of consternation. Her behaviour was exactly what you might expect from a child who was pulling a toy car on a piece of string when it got j ammed. She went back to the end of the string and lifted it carefully over the step, as though freeing it from the obstruction. That done, she went on her way again . This has all the hallmarks of fictional play, of the kind Alan Leslie has identified as a key feature associated with theory of mind, and which is absent in autistic children. Another example of high-level intensionality was observed by

Grooming, Gossip and the Evolution of Language the D utch ethologist Frans Plooij while studying Jane Goodall's chimpanzees in the Gombe National Park in Tanzania. At that time, the researchers were using bananas and other foods to attract the chimps into the camp area so that they could be stud­ ied more easily. As time went by, the chimpanzees became pro­ gressively more demanding. Worse still, having discovered where the bananas were stored, they began raiding the huts and store­ houses at the camp. To prevent things getting completely out of hand , the researchers built a concrete box half-buried in the ground. The box had a lid which the researchers could release from a short distance away by means of a cable. In this way, the researchers hoped to ensure that low-ranking animals gained a fair share of bananas, and wouldn't be discouraged from coming to the camp by the fact that more dominant animals prevented them from getting any of the food on offer. One day, one of the lower-ranking males arrived alone at the . feeding point. The catch was released with an audible click to allow him to open the lid of the box and feed on the bananas inside . Just as he was a bout to do so, however, one of the domi­ nant males appeared. The first male at once pretended to show no interest in the food box. This was a reasonable ploy: because the catch was only released when a specific individual was present, it often remained locked even when chimpanzees were in the feeding area. Presumably, the male in this particular instance wanted to give the impression that the box was still locked and hence that there was little point in the other male hanging around. Tactical deception of this kind is well within the range of a chimp's a bili­ ties: both de Waal and the American psychologist Emil Menzel have observed chimps behaving in j ust this way in their respective captive colonies in Holland and the USA. But the fascinating point about this incident was the behaviour of the dominant male. Rather than investigating the banana box for himself ( which would have been a pointless task a nyway), he turned and walked away again; but when he came to the edge of the clearing, he slipped behind a tree and peered back to see whether the male at the feeding box tried to lift the lid a fter he had gone. If my interpretation of the behaviour of these chimps is correct,

The Ghost in the Machine


the do minant male was behaving in a way that clearly implies at least third-order intensionality. Something like this must have been going through his mind: I think [ I ] that Jim is trying to deceive [2] me into believing [ 3 ] that the lid is locked. It's j ust con­ ceivable he might even have been indulging in fourth-order inten­ sional thinking: I think [ I ] that Jim is trying to deceive [2] me into believing [ 3 ] that Jim thinks [4] the lid is locked. The trou ble with a necdotes of this kind is that it is a lways possible to provide an alternative explanation in terms of coinci­ dences or simpler learned behaviours. Did Austin really under­ stand that Sherman was scared of noises in the dark ? Or had he simply learned that if he made a lot of noise outside, Sherman would cuddle him instead of bullying him - even though he didn't really know why Sherman preferred to cuddle rather than bully in these particular circ*mstances ? Did the male at the food box really intend to deceive his rival - in effect saying to himself, 'If I behave in a nonchalant way, I think [ I ] this male will believe [2] that I think [ 3 ] the box is still locked' - or had he merely learned that by behaving in this way, rivals would eventually go away, for some reason completely beyond his powers to fathom ? Perhaps the dominant male in this last story stopped as an afterthought because he couldn't quite tear himself away from the food box in case the catch was released - as sooner or later it would eventual­ ly be - and happened to be behind a tree when he did so ? After all, the chimps may not have worked out why the lid would open on some occasions but not on others; rather, they may have learned that patience was eventually rewarded by food if you returned to the box often enough. We would feel surer of our ground if we could point to a long series of examples all of which showed the same kind of behav­ iour. The more examples we had, the less likely it would be that they were all the result of coincidence; although it would still be difficult to exclude simpler explanations in terms of straightfor­ ward pattern learning. It is nonetheless interesting that observations of this kind have come only from chimpanzees. Despite the hundreds of thousands of hours scientists have spent studying Old and New World

Grooming, Gossip and the Evolution of Language monkeys in the wild and in captivity, events that can be interpreted at high levels of intensionality have never been reported ( although this could, of course, si mply mean that observers have not noticed such instances because they have not expected to see them) . I n the last few years there have been several studies which attempted to circumvent this difficulty by focusing more clearly on experimental designs that mirror the kinds of tests used to esta blish whether children ha ve ToM. The earliest tests were carried out on a language-trained chimp named Sarah by the psychologist David Premack in the early 1 9 80s. Premack and his colleague Guy Woodruff showed Sarah clips of film of someone trying unsuccessfully to do something: for example, reach a banana suspended from the ceiling. They then offered her photographs of appropriate and inappropriate solu­ tions to the problem. An appropriate solution might be a set of boxes piled up one on top of the other below the banana, while an inappropriate solution might be the same boxes lying scattered on the floor. Sarah displayed considerable competence at understand­ ing the person's intentions, as shown by her picking the appropri­ ate solution more often than not. From experiments like these, Premack and Woodruff concluded that Sarah's ability to under­ stand another individual's intentions demonstrated that she had, in some sense at least, a theory of mind. Two more recent: series of tests have tried to compare apes with monkeys to see whether there are any differences between these closely related members of the Primate Order. In the first of these, the American psychologist Danny Povinelli put chimps and rhesus macaques (a representative advanced Old World monkey) through a series of tests designed to discover whether they under­ stand another individual's intentions or knowledge of a situation. Among the tests he gave chimpanzees, for example, was one in which the animal had to choose between two humans in order to get a reward it could not reach: a glass of juice to drink. The chimp was shown pictures of two assistants, and had to choose between them by pushing over the appropriate holder in which the photographs were held . The difference between the two - 98 -

The Ghost in the Machine h u mans was that one of them always deliberately poured the j uice on to the floor, whereas the other one did so accidentally, for exa mple by dropping the cup when picking it up or by tripping when handing it to the chimp. Could the chimp distinguish between deliberate and accidental behaviour? The answer seemed to be a clear yes: the chimp soon learned to choose the human that a ccidentally spilled its j uice. In another series of experiments, the chimp was given the opportunity to obtain a food reward from a baited box that was out of its reach. In order to get the reward, the chimp had to choose a human assistant to open the box and hand it the food. Two assistants pointed at different boxes, and the chimp had to decide on which assistant it thought was most likely to be right. The choice was between an assistant who was in the room and watched the box being baited, and an assistant who conspicuous­ ly left the room while the box was baited. The first assistant knew where the reward was, but the second obviously did not. So the correct answer was to choose the box which the knowledgeable assistant pointed to. Most ( but not all) of Povinelli's chimpanzees managed to solve this problem reasonably competently, but none of the monkeys did. It seems that apes, or at least chimps, can dis­ tinguish between knowledge and ignorance in other individuals, but monkeys cannot. However, even though they outshone the monkeys, the chimps do not seem to be as competent at these tasks as human children are. This was borne out by another series of tests, carried out by Sanj ida O'Connell, one of my students. She designed a mechanical version of a false belief test which aimed to meet the standards of the Sally-and-Ann test used on children. Once again, the animal was presented with a choice of four boxes. The experimenter placed a peg a bove one of the boxes; then she placed a morsel of food in the box she had previously marked with a peg. The chimp was then free to open the box of its choice and collect the reward if it chose the right box. Once the chimp had learned to respond correctly to the basic procedure, a glitch was introduced. The boxes were so designed that O 'Connell could not see the front of them when she was baiting the selected box from the back.

Grooming, Gossip and the Evolution of Language

The Ghost in the Machine

Having first put the peg in place from the chimpanzee's side of the apparatus, she then went round the back to bait the box, j ust as she had done dozens of times before. Only this time, while she was on the other side, the peg moved - apparently of its own accord, but really by means of a lever she surreptitiously operated and came to rest above another box. The critical question was: would the chimp recognize that the box which had been baited was the one the experimenter had originally identified with the peg - presumably the box the experimenter thought was the right box - or would it assume that the experimenter had the same knowledge as it did, and so would bait the box above which the peg now rested}. This is about as close to the Sally-ami-Ann test as

cognitive levels reached by three and four-year-old children j ust before they finally develop ToM . D orothy Cheney a n d Robert Seyfarth have commented that monkeys are good ethologists but bad psychologists: they are good at reading another individual's behaviour but bad at reading its mind. They give a delightful illustration of this from their stud­ ies of the vervet monkeys in Amboseli, Kenya . One day, a strange male appeared in a grove of trees not far from the troop they were studying. Lone males of this kind are invariably intent on j oining a group, and are usually able to displace the group's dominant male when they do so. For the incumbent male this is not a happy occasion, because he stands to lose his monopoly over the females. Naturally, they resist intruders by every means they can. In this case, the incumbent was positively inspired. As the intruder stepped down from his tree to cross the open ground to the trees the group was feeding in, the troop male gave a leopard alarm call . The intruder shot back into the safety of his tree. Later, satis­ fied that all was well, he tried again; once again, the troop male gave his leopard call. So far, so good: the ruse seemed to be work­ ing. Unfortunately, the troop male eventually gave the game away by giving his leopard call while he h imself was walking across the open ground. The intruder was smart enough to realize that no one in his right mind gives alarm calls while wandering noncha­ lantly across open ground where he is at risk of being caught.


it is possible to get. Although the chimps did better than a utistic adults at this task, they were nowhere near as good at it as five and six-year-old nor­ mal children (who have ToM) . The chimpanzees certainly learned to solve the problem, but they weren't as competent as one might have expected had they had full ToM . There is, however, one final caveat. Negative answers are never very satisfactory when working with animals, especially chim­ panzees; you can never be sure whether failure to perform to a cri­ terion really does indicate an inability to solve the problem we set for them, or simply a lack of interest. Not surprisingly perhaps, chimps often seem to find tests of these kinds boring, and some­ times they cannot be persuaded to take part at all. There is a love­ ly sequence in the BBC Horizon film Chimp Talk in which Sue Savage-Rumbaugh asks Kanzi to carry out a series of instructions. When it came to 'Put the bunch of keys in the refrigerator', you can sense in his momentary hesitation the puzzlement that must have been in his mind: ' What is she up to now ? O h well, I suppose I had better humour her, poor thing ! ' Our tentative conclusion must b e : there i s sufficient evidence to suggest that chimpanzees, if not all great apes, have ToM in some form - even though it may be a less advanced form than in humans - but that monkeys do not. Even though monkeys are more advanced in these respects than other animals, they certainly do not have full ToM. Rather, they probably have the kind of - 1 00 -

Into the Mind and Beyond

Theory of mind is, beyond question, our most important asset. It is a remarkable skill. Yet even ToM pales into insignificance by comparison with where this skill has taken us. ToM has given us the crucial ability to step back from ourselves and look at the rest of the world with an element of disinterest. The starting point of all this was probably our ability to reflect back on the contents of our own minds. Why do I feel the way I do ? Why am I a ngry now ? Why do I feel sadness or happiness ? Understanding our own feelings is crucial to understanding those o f other people. Without recognizing what we are seemg m - 101 -

Grooming, Gossip and the Evolution of Language

The Ghost in the Machine

others, we have no hope of delving into their minds sufficiently far to appreciate their mental reactions to the things they experience. The real breakthrough is where fully developed third-order ToM allows us to imagine how someone who does not actually exist might respond to particular situations. In other words, we can begin to create literature, to write stories that go beyond a simple description of events as they occurred to delve more and more deeply into why the hero should behave in the way he does, into the feelings that drive him ever onwards in his quest. I think I'm on safe ground in arguing that no living species will ever aspire to producing literature as we have it. This is not simply because no other species has a language capacity that would enable it to do this, but because no other species has a sufficiently well-developed theory of mind to be able to explore the mental worlds of others. To write fiction is to conceive of imaginary worlds that do not exist. I am not convinced that even the ToM abilities of chimpanzees are good enough to be a ble to do that. Chimps seem to be limited at most to third-order intensionality ( ' I believe that y o u want m e t o think that the food b o x i s locked' ) . Humans seem t o b e capable o f following arguments through to fourth-order intensionality without too much difficulty, though they do not very often run to such lengths in everyday contexts. But they do become necessary when writing stories whose plots involve both the writer and the reader understanding [ I ] what one character thinks [ 2 ] another character wants [ 3 ] the first character to believe [4 ] . Since both writer and reader become part of the chain of intensionality, they must be able to go one order beyond what the characters actually do. To keep track of that through the sequence of events in a novel is obviously very demanding. The writer has to be able to assume that his readers can achieve the same levels of intensionality as he can; if the reader was incapable of that, there would be little point in trying to sell the novel to a publisher. The a bility to detach oneself from the immediacy of one's expe­ riences is also a prerequisite for two other unique features of human behaviour, the phenomena we know as religion and sci­ ence. Some of my scientific colleagues ( notably the embryologist

Lewis Wolpert) are, however, inclined to suffer apoplexy when an yone suggests that science and religion are similar phenomena . I n o n e sense, of course, these colleagues are right: science and reli gion use radically different methods for making their claims about the world. One is a matter of belief, in which revealed truth holds centre stage as the final arbiter of all disputes, whereas to the other, individual scepticism and the rigorous testing of hypotheses based on logical deduction and reference to empirical evidence are all-important. But at another level, their apoplexy is premature, for they over­ look an important respect in which the two phenomena are virtu­ ally identical. Both are attempts to explain the world in which we live. Both serve to give the phenomenal world as we experience it sufficient coherence to enable us to steer a reasonably sensible path through the vagaries of everyday life. The radical difference in the way these two activities work should not obscure the com­ mon purpose they serve. Religions the world over provide security and comfort, a crutch that helps us through the difficult and often dangerous business of daily life. They give us a sense that all is not completely beyond our frail control, that via prayer and ritual we have recourse to mechanisms that will allow us to ensure that life will proceed in a tolerably benign way. In traditional societies, where flood, famine and marauding animals and humans are a constant threat to life and peace, resort to the supernatural may make the difference between sanity and insanity. Having carried out all the necessary rituals, we at least have sufficient sense of certainty to be a ble to proceed; religion may not entirely prevent the worst from befalling, but it probably provides us with enough confidence and courage to brush off the lesser inconveniences of life that might otherwise have overwhelmed us. In this sense religion is, as Marx famously observed, the opium of the people: it acts j ust like endogenous opiates, dulling the minor irritations of daily exis­ tence j ust enough to allow life to proceed. Science too provides us with a framework for existence and allows us to control the world. But the way it does so is, of course, completely different. Science's dramatic success rests not (as some

- I02 -

- 1 03 -

Grooming, Gossip and the Evolution of Language

The Ghost in the Machine

wishfully hope) on arbitrary constructions of reality, but on the carefu l deduction of hypotheses and their rigorous testing again st the events of the real world. Science allows us to have more confi­ dence in its findings because they have to work in the real world. Short of a grand conspiracy theory of science - which would be hard to sustain - it is difficult to imagine how anyone could force the world to produce results that happen to suit the convenience of scientific theories. The real world is simply not that kind of place: it is unyielding, and very unforgiving of incompetents. The common origins of science and religion lie in a hesitant questioning of why the world is as we find it. The answers they provide may be as different as chalk and cheese, but their function remains the same. And they both depend on the same questioning attitude towards the world. Why is the world as it is? Even to ask that question requires you to be willing to imagine that the world could be other than as it seems. It requires ToM. As such, it is spun off the deep reflexivity of our social behaviour, our ability to understand how an individual's mind can influence his or her actions and how I, in turn, can influence that individual's mind. It requires third-order intensionality at the very least, and quite possibly something beyond that. If science and religion require fourth-order intensionality, then it is clear why only humans have produced them. Since no non­ human animals other than great apes aspire to more than second­ order intensionality, none of these species will ever produce sci­ ence and religion as we know it. But a question mark remains over the great apes. If fourth-order intensionality is essential, then they almost certainly cannot aspire to science or religion. But if third-order suffices, then it is j ust conceivable that great apes do have science and/or religion. However, if the apes do have some form of science or religion, it cannot be very sophisticated. Nor will it be a unifying force in their social lives. This is because they do not have language. Language allows us to communicate ideas to each other with an efficiency that is otherwise impossible to match. Without lan­ guage, each individual has to reinvent the intellectual equivalent of the wheel for itself. We can see and copy tools or someone

else's wheel, but religion and science belong to the world of ideas, and we cannot see and copy ideas or concepts in quite the same sense. Without language, we each live in our own separate mental world. With language, we can share the worlds inhabited by oth­ ers. We can discover that other peoples' worlds are not quite the same as ours; that in turn will prompt us to realize that the world can be other than we suppose it to be. The psychologist David Premack concluded that his star chim­ panzee Sarah's mind had been 'upgraded' by being taught a sym­ bolic language. Premack's view seems to be based on the common claim among social linguists and anthropologists that language determines how we think, that without language we cannot have thoughts. In fact, this flies in the face of a great deal of evidence showing that animals do think, that they can develop concepts and all the phenomena we associate with language. It seems more likely that language is parasitic on thought, that it has the kind of grammatical structure we give it ( the subject-verb-object form) because that is how we naturally think. I am not so convinced that Sarah's mind was upgraded merely by the learning of a language: the language did not suddenly create concepts or knowledge that her mind did not previously possess. Rather, Sarah's mind was upgraded by language because language provided her with access to Premack's mind. He was able to pass on to her concepts and wa ys of looking at things that she might never have thought of on her own. And the emphasis here is very much on the 'might' rather than the 'never'. Language is thus a crucial factor in the history of ideas. It allows us to build on the knowledge of earlier generations. But it also allows us to exchange knowledge amongst ourselves so that the whole community becomes wrapped up in the same set of beliefs . If chimpanzees have religion, they must have as many religions as there are individual chimpanzees.

- 1 04 -

- 105 -

Up Through the Mists of Time C HA P T E R


Up Through the Mists of Time

Picture the scene some five million years ago. The sunlight dapples the floor of the ancient forest, while the monkeys chatter as they tumble through the treetops on the way from one tree loaded with wild figs to another. On the forest floor there are several species of great ape, not too dissimilar from the chimpanzees and gorillas of today. They travel mostly along the ground, climbing into the trees to forage for fruits and other delicacies. These apes are the remnants of a family of species that dominat­ ed the forests of Africa and Asia for the better part of 10 million years. But times are hard for them. The forests of Africa are con­ tracting under the steady cooling and drying of the world's cli­ mate. More and more species are being crammed into a smaller and smaller space. To make matters worse, the monkeys have stolen a march on the apes and are able to outperform them in the ecological race ( see Chapter 2 ) . The apes, once the most abundant of the forest primates, are now in decline. One ape lineage, it seems, eventually began to make more use of the forest edge, venturing further and further out from the safety of the forest to search for food trees that had not already been cleaned out by monkeys. In the woodlands that lie beyond the forest edge, the distance between food trees is greater and the canopy less contin­ uous. You cannot travel, monkey-fashion, from one tree to another along intersecting branches. Instead, it is necessary to descend to the ground and travel overland from one tree to the next. Stand Tall to Stay Cool

In the less heavily forested woodlands, animals travelling between trees are exposed to more heat from the sun. Peter Wheeler, an - 1 06 -

eCOlogical physiologist from Liverpool's John Moores University, has studied the heat stress these ancestral apes would have experi­ enced as they moved through the wooded savannahs of Africa. His calculations show that an animal which walks upright receives up to a third less radiant heat from the sun, especially during the middle of the day when the sun is at its hottest. This is si m ply because less of the body s urface is exposed to the direct rays of the sun when standing upright than when walking on all fours. It is a point intuitively obvious to sunbathers: they always lie down to expose as much of the body surface as possible. You'll never get brown quickly standing up. Moreover, on two legs you benefit from the slight increase in wind speed that occurs above the surface of the earth. Friction from the vegetation and even the ground itself slows the wind down close to the earth's s urface in much the same way that a brake acts on a wheel. The increase in wind speed has a significant cooling effect from about three feet a bove the ground. Large ani­ mals of course benefit from this, but smaller animals can benefit too if they stand on their hind legs. Animals about the size of chimpanzees are in the narrow range of budy size where standing upright is worthwhile. Smaller species like baboons are not tall enough for standing on two legs to make any difference. So by standing tall these apes kept cool, which enabled them to travel further into more open habitats in search of food. In the process, another device came into play. With less body surface exposed to the direct rays of the sun, there is less need of the fur that normally serves to keep animals' skin cool. Fur is inert and a good insulator, so the outer tips of the hair can heat up dramati­ cally without the heat being transferred to the body underneath. Wheeler has argued that nakedness evolved early in the ape lin­ eage that led to modern humans as a way of adding in the extra cooling properties of sweating through bare skin. The upright ape's body was protected from the worst of the sun's rays by its vertical position; consequently, the cooling effect of the wind above the ground-layer vegetation, combined with the cooling effect of evaporation brought on by sweating, made hairlessness a distinct a dvantage . So we lost our fur coats, retaining them only - 1 07 -

Grooming, Gossip and the Evolution of Language

Up Through the Mists of Time

on those surfaces still exposed to the sun at midday, namely the top of the head and shoulders. Peter Wh eeler's careful calculatio ns from the well-established equations for thermal physiology sug­ gest that a hairless, bipedal, sweating hominid could have doubled the distance it travelled on a pint of water compared to a furred quadrupedal one, a saving that would have had enormous advan­ tages for a semi-nomadic hominid out on the open savannahs. We do not, of course, know exactly when our ancestors lost their fur, because soft tissues and fur are almost never preserved in the fossil record. But we do know that they began to walk bipedally at a very early stage. Two sources of information con­ firm that for us. One is the shape of the hips and leg-bones of the earliest fossil hominids. The half-skeleton named Lucy that Do"n Johanson dug out of the Ethiopian Afar desert in 1 9 7 6 has a well­ preserved pelvis and associated leg-bones which clearly prove that this diminutive early hominid from 3 . 3 million years ago already " walked upright. We can tell this from the shape of the pelvis and from the way the knee and hip j oints articulate; modern humans have a bowl-shaped pelvis which provides a more stable platform to brace the legs against during walking, whereas apes have a long thin pelvis designed to give more support when climbing. It is clear from the shape of her bones that Lucy's style of walking was not yet fully human: she would have waddled somewhat, rather than walking with the balanced stride so characteristic of modern humans. Moreover, her fingers were longer and more curved than ours and her chest and arms were more strongly built, suggesting that she was still well-adapted to clambering about in trees. But when on the ground, it seems she almost certainly walked upright. The clinching evidence for this comes from a set of seventy or so footprints preserved under a layer of lava spewed out by a vol­ cano some 3 . 5 million years ago near a place in northern Tanzania called Laetoli. Here, three sets of tracks closely follow each other, crossing and being crossed by the tracks of antelope and other mammals, across a thirty-yard stret(:h of what had once been open plain. Walking in the soft lava ash being spewed out by

the lava i nto concrete. Hidden beneath more layers of ash, what m ay have been the last actions of this small group were preserved un til they were uncovered by the palaeontologist Mary Leakey almost four million years later, in 1 9 7 8 . There c a n b e no doubt that these footprints were made by a s mall, bipedal apelike creature about the same height as Lucy. There are no hand-prints such as would be seen when a baboon or chimpanzee walks or runs. Moreover, the big toe is tucked in beside the other toes at the front of the foot j ust like ours is, rather than being set off at right angles nearer the heel as it is in apes . This was a truly bipedal species that habitually and comfortably walked upright. Further light has been thrown on the story by new fossil finds in southern Africa. These include a foot on which the big toe is not quite parallel with the other toes, as it is in the modern human foot, although it doesn't stick out quite as much at right angles to the foot as do the big toes of living apes. Dating from around the same time as the Laetoli footprints, this fossil foot suggests an ani­ mal that could walk upright but was still at home in the trees. So it seems certain that an upright stance, and hence probably hairlessness, evolved at a very early stage in our ancestry, while 'we' were still very much apes. It was to be another two million years or more before our brain size would begin to expand signifi­ cantly beyond that typical of living apes. Crisis on the Forest Edge

a nearby volcano, the individuals in question left their dramatic imprint on history because a shower of rain shortly afterwards set

Launching out from the forest into the woodlands that lay beyond provided these early hominids with the advantage of food sources that were less heavily competed for. Most of the species that inhabit the grasslands and woodlands of eastern and southern Africa are grazers or browsers, eating mainly grasses or the leaves of herbs and small bushes. Few species compete for the fruits and seeds that grow on the trees and taller bushes. But nothing comes free in life, and in order to reap the benefits from foraging in this new environment, these ancient apes had to contend with signifi­ cantly higher predation risks.

- 108 -

- 1 09 -

G rooming, Gossip and the Evolution of Language

Up Through the Mists of Time

As we saw, primates in general exhibit two responses to increased predation: they grow physically bigger and they increase the size of their groups. O ur ancestors appear to have done both. Sta ture increases steadily through time in the fossil record. Little Lucy was barely four feet high when she roamed the savanna h woodlands of the Horn of Africa some three million years ago. By I . 7 5 million yea rs ago, the so-called Narikitome Boy, whose skull and partia l skeleton were unearthed by Mary Leakey's son Richard on the shores of Lake Turkana in northern Kenya in 1 9 84 , was a lready pushing five foot three when he died at the age of eleven. Had he lived, he would have been a willowy six foot something as an adult. If these pre-humans were following the general primate pattern, then it is likely that their group sizes were also increasing steadily through this period in response to the same pressures. How on earth are we to know what groups they went around in ? Groups do not leave evidence in the fossil record, and no reputable palaeontologist has ever made any serious claims about group sizes (at least prior to the appearance of permanent camp-sites within the last 1 00,000 years ) . It seemed as though we would never know. Yet our discovery that group size in primates is closely related to neocortex size raised the possibility of being able to estimate, even if with a degree of error, group sizes for fossil species. Being able to do so has other tantalizing prizes to offer. The relationship between group s ize and grooming time in primates might allow us to solve another elusive q uestion: when did language evolve ? Conventional wisdom offers two main suggestions, both based on very indirect evidence. The archaeologists favoured a relatively recent date around 50,000 years ago when the archaeological record undergoes a dramatic and sudden change known as the Upper Palaeolithic Revolution. At this point there is a marked change in the quality and variety of stone tools. A wider range appears over the succeeding m illennia, including awls and punches, and then nee­ dles, buttons and clasps. Art objects, such as the exquisite Venus fig­ ures and the cave paintings, appear around 3 0,000 years ago. Burials appear to be organized, with the body placed in a carefully prepared position and often accompanied by objects that might be

useful in the afterlife. It all suggests that the people concerned are expla ining things to each other, and are able to discuss sophisticated meta physical concepts such as death and the afterlife. In contrast, the anatomists favoured an earlier date, perhaps aro und 2 5 0,000 years ago at the latest, associated with the appearance of the first members of our own species, hom*o sapi­ ens. Their evidence was based mainly on the fact that an asymme­ try in the two halves of the brain could be detected at around this point. In modern humans the left hemisphere of the brain, the half where the language centres are located, is larger than the right ( for more on this, see the following chapter) . They suggested this as clear evidence for the appearance of language. The disagreement between the archaeologists and the anatomists seemed impossible to resolve, because each side had evidence from its own field to support its views. Leslie Aiello and I thought we might be able to settle this dispute if we could only solve the problem of group sizes in our early ancestors. O ur argument went like this. We knew that no primate spent more than 20 per cent of its day engaged in social grooming. Since modern-day monkeys obviously coped well enough with this ( and could probably manage a bit more at a push ) , the threshold that triggered language evolution must lie somewhere above this. At the same time, we knew it must lie well below the 40 per cent of social time that our equations predicted for modern human groups ( see page 7 7 ) . Somewhere in between, at around 3 0 per cent of social time perhaps, lay the

- 1 10 -

great Rubicon that precipitated language. But it seemed that we would never know, because we had no way of determining either of the two key variables needed to predict grooming time, neocor­ tex size or group size. While we were mulling this problem over one evening, we noticed that the neocortex ratio - the proportion of the brain made up o f neocortex - in primates ( including modern humans and, it later transpired, carnivores too) is directly related to total brain size. While we obviously know nothing at all about the neo­ cortices of fossil hominids, we do have many complete or near­ complete fossil skulls from which we can estimate total brain size. Given this, it would be a simple matter to estimate the relative - III -

Up Through the Mists of Time

Grooming, Gossip and the Evolution of Language volume of neocortex from the internal dimensions of each skull, and then use this to predict group sizes from the equation we had found between neocortex ratio and group size in primates ( see Figure 2, page 63 ) . The next day was spent i n a flurry of calculation, double-check­ ing figures and re-examining the logic of our argument. Then Figure 3 rolled triumphantly off the computer screen onto the printer. This suggests that group size increases rather slowly at first. Moreover, it remains well within the range of group sizes observed in living great apes (especially chimpanzees) until around two million years ago. At this point, a new genus appears in the fossil record, the genus hom*o to which we modern humans belong. Now, for the first time, group size begins to edge a bove the upper limits seen in modern primates. From this point on, group size rises exponentially, reaching the 1 5 0 that we found in modern humans (see Chapter 4) some time around 1 00,000 years ago. The burning issue is: when did group size pass through the crit­ ical threshold where language would have become necessary ? Looking at the figures for the corresponding grooming times (shown in Figure 4 ) , we concluded that the evidence came down strongly in favour of the earlier of the two dates. By 2 5 0,000 years ago, group sizes were already in the region of 1 20 to 1 3 0, and grooming time would have been running at 3 3 to 3 5 per cent (well above the critical 3 0 per cent threshold ) . But a closer look a t the figures suggests that we might even have to push the date back earlier still. The earliest members of our species appear around 5 00,000 years ago, and the equations would predict group sizes of I I 5 to 1 20 for them, with grooming times of around 3 0 to 3 3 per cent. The conclusion seems inescapa ble: the appearance of our own species, hom*o sapiens, was marked by the appearance of language. Our immediate predecessors, the late members of the genus hom*o erectus - who had relatively smaller brains - may already have begun to run into the same pro blem. In their case, group sizes were sometimes large enough to nibble at the 30 per cent grooming time threshold. However, most of their groups hovered in the range 1 00 to 1 20, with grooming time requirements of

about 25 to 30 per cent. Leslie Aiello and I took the view that hom*o erectus as a species did not have language, though this is probably arguable for the very late members of the species.

- 112 -

- 1 13 -

200 1 80 w



Modern humans

1 60 1 40

f, Archaic humans

c �'

1 20 � 0 a: 1 00 " c w 80 I0 60 £5 w a: 40 Il. 20 Il.

� .;I. -



H. erectus

- -

.- H. habilis




1 .5







Figure 3. Predicted group size for individual populations of fossil hominids, plotted against their age. The group size is that predicted by the relationship between group size and neocortex size for primates in general ( see Figure 2 ) . Five main groups of fossil hominids are represent­ ed: the australopithecines ( the earliest hominids) ; hom*o habilis ( the earliest member of our own genus); hom*o erectus ( the first hominid to migrate out of Africa into Europe and Asia ); archaic hom*o sapiens ( the earliest members of our own species, including the Neanderthal peoples of Europe and the Near East) ; and fossil modern hom*o sapiens ( principally ero-Magnon peoples of Europe and their African relatives) . Relative neocortex size was estimated from total brain volume. Each point is the mean for one population ( defined as all fossil specimens obtained from the same site dating from within 5 0,000 years of each other) . The horizontal line indicates the group size of 1 5 0 predicted for contemporary humans.

Grooming, Gossip and the Evolution of Language

40 en w ::E i= o z


o o a:: o

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - � - .. .. ..


- - - - - - - -




H. erectus

- :- .. � -

- -

.. .- .. - .. - ..:- '-

Up Through the Mists of Time

.. .. - .. .. �

- -

- - - - - - - - - - - - - - - - - -

_ H . habilis

- - - - - - - - - - - - - - - - - - - - - - - - -





- -

0 +-�--�--�-+--��---r--+--,--1---.--.-� 3.5 3 2.5 2.0 1 .5 0.5 o MILLIONS OF YEARS AGO

Figure 4 . Grooming time predicted for individual fossil hominid popula­ tions, plotted against time. Grooming time is that predicted by the observed relationship between grooming time and group size in living Old World monkeys and apes, with group size for fossil hominid popu­ lations predicted by neocortex size ( see Figure 3 ) . The upper horizontal line indicates the grooming time requirement of 40 per cent predicted for contemporary humans with their group sizes of 1 5 0; the lower line indi­ cates the maximum grooming time observed in any living primate population ( 20 per cent in gelada baboons ) .

However, something else struck us when w e looked closely at the patterns in Figures 3 and 4: there is no obvious j ump in grooming time that would suggest the crossing of a maj or R ubicon. Had the critical point coincided with the appearance of the first members of the genus hom*o at around 2 . 5 million years ago, we might have been able to conclude that language appeared rather suddenly. But there is no such 'big bang' around half a mil­ lion years ago, which suggests that, rather than evolving as the

re s ult of some dramatic new mutation, language emerged slowly over a long period of time. It seems that language as we know it evolved in at least three stages, becoming progressively more complex as the demands of large r group sizes became more pressing. We suggest that it has its ori gins in the conventional contact calls so characteristic of the advanced Old World monkeys and apes. In these species, as we noted in Chapter 3 , contact calling functions as a kind of groom­ ing-at-a-distance. As time-budgets became increasingly squeezed, the animals would have kept up a steady flow of vocal chatter. Its content would have been zero, rather along the lines of those for­ mulaic greetings so common in our own conversations. Remember the hackneyed 'Do you come here often ? ' This is not a question requiring a carefully thought-out answer. It's an opening gambit, a tentative searching for a reply along the lines of, 'Yes, I'd be very happy to spend the next half-hour in your company, thanks . . . ' We suggested that, as group sizes began to drift upwards from the numbers to which apes are currently limited, vocal grooming began increasingly to supplement physical grooming. This process would have begun around two million years ago with the appear­ ance of hom*o erectus. More and more emphasis was being placed on vocal as opposed to physical grooming as a bonding mecha­ nIsm. Eventually, even this form of communication would have exhausted its capacity to bond groups. A more efficient mecha­ nism for bonding was needed to allow group size to continue its upward dr ift. At this point, the vocalizations began to acquire meaning. But the content was largely social: gossip had arrived. This need not have involved any dramatic change, for as the studies by Seyfarth and Cheney have shown, primate vocaliza­ tions are already capable of conveying a great deal of social infor­ mation and commentary. The pieces of the jigsaw were already there: all they needed was to be organized into a coherent system. The continuing drive for ever-larger groups provided the neces­ sary kick at the right moment. In effect, humans were now exploiting the greater efficiency of language as a bonding mechanism to allow themselves to live in

- 1 14 - 115 -

Grooming, Gossip and the Evolution of Language

Up Through the Mists of Time

larger groups for the same investment in social time. This sugges­ tion is borne out by the fact that modern hunter-gatherers seem to devote a bout the same proportion of their day to social interac­ tion as the modern gelada ( the species that holds the record for time s pe nt grooming among non-human primates). A study of t he Kapanora tribe in New Guinea, for example, found that men spent an average of 3 . 5 hours and women around 2 . 7 hours in social interaction d uring a typica l 1 2-hour day. In other words, they spend on average a quarter of the day socializing, compared to a figure of 20 per cent for gelada. Not until very much later, perhaps as much as 400,000 years lat­ er, do we see any evidence for the appearance of symbolic lan­ guage: language capable of making reference to abstract concepts. At this point in the fossil record, we see a sudden change in the style, quality and variety of stone tools. The previous two million years had seen a lmost no development in the types of stone tool in . common use. They were limited both in the nature and complexity of their design, and hence in the skills required to fashion them; most are crude choppers and handaxes of little artistic merit. Then quite suddenly the picture changes. Tools become more delicate and more finely prepared. Substances such as red ochre appear in the fossil record for southern Africa with evidence that they have been ground a nd rubbed, suggesting they may have been used in tanning skins or manufacturing body paints. Here then are the first hints of ritual. The cultural revolution had arrived. This seems to resolve another controversy that has rumbled on over the years: did the Neanderthals have language ? The Neanderthals occupied Europe throughout the Ice Ages, from around 1 20,000 years ago. They suddenly disappeared about 3 0,000 years ago after a period of coexistence with the ancestors of modern humans, the Cro- Magnon people, who arrived from Africa around 5 0,000 years ago. The anatomist Philip Lieberman has always insisted that the Neanderthals did not have language, on the grounds that their l arynx ( the top of the tube from the lungs ) was too high in the throat to produce human vowel sounds. The Neanderthal larynx, he suggested, was in much the same position as the chimpanzee's, and they certainly cannot produce

su ch sounds . Unable to communicate with each other in anything more than grunts and screams, the dumb Neanderthals were over­ ru n b y the tall, slim modern humans from Africa with their sophi sticated culture and their language. B ut Lieberman's claims have been thrown into doubt by the dis­ covery of an almost complete Neanderthal skeleton in Israel with its hyoid - the tiny bone that supports the larynx and the base of the tongue - still in place. The Neanderthal's larynx appears to have been roughly where ours is, low enough in the throat to pro­ duce the full range of speech sounds. Anatomically, it looks as though they did have language. Our analyses would agree with this. With brains that were if anything slightly larger than those of modern humans ( the Neanderthals were larger and much stronger than we are ) , they must have had groups of the same size as other modern humans. Grooming times would have been well beyond the sustainable limits. So if they were not using their brains to maintain human-sized groups, what on earth were they doing with them ? It must have been something quite different from all other living primates, including modern humans. If the Neanderthals became extinct at the hands of the invaders from Africa, it was not because they lacked language but because they lacked the sophisticated culture and social behaviour of our African ancestors. Not only were the stone tools and artefacts of the Cro-Magnons very much more delicate and complex than those of the Neanderthals, but there is also evidence that the Cro­ Magnon people traded shells as well as flint and other stones over very wide areas. They clearly had much more sophisticated and widespread social networks than the Neanderthals. Recent research on the many cave sites in Israel suggests that the Neanderthals may have been more sedentary in their habits, whereas the modern humans who lived in the area during the same period were more nomadic and followed the changing distri­ bution of resources more closely. The moderns thus appear to have been ecologically more flexible. The fate of the Neanderthals bears an uncanny resemblance to the fate of the American Indians and the A ustralian Aborigines at the hands of the later Europeans, invaders who could draw on a

- 1 16 -

- 1 17 -

Up Through the Mists of Time

Grooming, Gossip and the Evolution of Language larger and more widely distributed political and military power base. O ld habits, it seems, die hard - though in the latter case it is worth reminding ourselves that all the parties involved ( Europeans, Americans and Australians alike) were the direct descendants of those ero-Magnon invaders from Africa. We are left with one last puzzling question: what drove the increase in group size ? The short answer is, we don't know. But we can hazard some guesses. The conventiona l wisdom on pri­ mates is that there are only two likely pressures that select for large group sizes: one is predation risk, the otqer the need to defend food sources. But if baboons can cope with groups of around 5 0, then it is difficult to see why the later humans and their immediate predecessors should have needed groups that were nearly three times bigger. The fact that they were larger than baboons ( and carried passable defensive weapons in their hands) should have meant that they could get by with smaller groups in the same habitats. And indeed, this i s pretty much what you see. Baboons typically live in groups of 50 to 60 in woodland habitats, while hunter-gather ers in eastern and southern Africa typically live in temporary camps of around 30 to 3 5 . Of course, our ancestors may have invaded even more open habitats than those currently occupied by the woodland-based ba boons and the forest-loving chimps, and so might have needed larger groups to offset the greater predation risk. Some evidence to support this suggestion is provided by the modern gelada. This species lives in very open habitats where they have few trees in which to escape from predators; they live in the largest naturally occurring groups of all primates ( typically 1 00 to 2 5 0 animals) . Moreover, the size o f their groups correlates with the predator risk in the ha bitat, being larger in those habitats that provide them with fewer safe refuges. A second possibility is that it was other human groups rather than conventional predators that posed the threat. This could have been in the form either of raiding ( for example, for women, as happens among some hunter-gatherers today) or of competition for food and/or water resources. Increases in both group size and individual stature could j ust as easily be interpreted as a response - 118 -


to these kinds of problems as to predation risk. It is j ust the kind of situation in which an arms race occurs: the raiders form bigger groups to raid more successfully, so you need to form even bigger groups to protect yourself, so the raiders need to form bigger groups still, and so on until ecological constraints impose a limit (perhaps in terms of the number of individuals that can be fed ) . A third possi bility may stem from the fact that early i n the sec­ ond phase of human evolution ( following the appearance of hom*o erectus two million years ago ) , a dramatic change in eco­ logical behaviour occurred: our ancestors became nomadic. They crossed the Ara bian land bridge into Asia for the first time and within a few hundred thousand years had reached China and the islands off the south-eastern coast of Asia. Nomadism on this scale suggests that the groups foraged over large ranges, but were not afraid to step into the unknown beyond their everyday boundaries in search of new food sources. Animals face two problems in these situations. One is that they are unfamil­ iar with the lie of the land and don't know where the safe refuges or the good feeding places are. The Swiss biologists Hans Sigg and the late Alex Stol ba showed that when the occasionally nomadic hamadryas baboons are travelling outside their normal territory they are significantly less likely to find water-holes and food trees than the groups resident in the areas concerned. The other is that resident groups may actively exclude them from access to such essential resources ( and water-holes may be absolutely crucial in hot savannah habitats ) . Migrants are always at a disadvantage. Establishing reciprocal alliances with neighbouring groups may be the only way to solve this problem. In effect, a set of neigh­ bouring groups would start to act co-operativel y to share their water-holes or other key resources . The result would be an alliance of loosely federated groups that could come and go, merge and split up, as the mood dictated. And this is exactly what one sees in modern hunter-gatherers. The ! Kung Sanl of the Kalahari, for example, live in communiI.

The ! K ung San speak a so-called 'click' language in which click-like sounds made

by the tongue and lips form part of the range of vowels and consonants in the lan­

guage. The ' ! ' is how one such sound is represented.

- 1 19 -

Grooming, Gossip and the Evolution of Language

Up Through the Mists of Time

ties of 1 00 to 200 individuals, each centred on a set of permanent water-holes that they 'own' or have rights over. The community itself, however, rarely appears as a single group. Rather, its mem­ bers forage in small groups of 25 to 40 ( typically four to six fami­ lies). The permanent water-holes are the community's life-blood, providing a safe retreat in times of drought when temporary water-holes dry up. This is known as a fission-fusion social system, because its members are constantly coming and going. Chimpanzees share this characteristic with us, except that their groups, are smaller. Chimpanzee communities are around 5 5 , while the foraging par-

action with friends. We wanted to avoid formal situations, where the rules of conversation are often deliberately constrained. We sampled conversations in university cafeterias, at public receptions, during fire practices ( while everyone was waiting for the all-clear to go back inside the evacuated building), in trains and in bars. The first thing we found was that conversation groups are not infinitely large. In fact, there appears to be a decisive upper limit of about four on the number of individuals who can be involved in a conversation. The next time you are at a social gathering such as a reception or a party, take a look around you. You will see that conversations begin when two or three individuals start talking to each other. In due course, other individuals will j oin them one by one. As each does so, the speaker and the listeners try to involve them in the conversation, directing comments to them or simply moving to allow them to j oin the circle. However, when the group reaches five people, things start to go wrong. The group becomes unstable: despite all efforts ( and groups often do try), it proves impossible to retain the attention of all the members. Instead, two individuals will start talking to each other, setting up a rival conversation within the group. Eventually, they will break away to start a new conversation group. This is a remarkably robust feature of human conversational behaviour, and I guarantee that you will see it if you spend a few minutes watching people in social settings. Since there is only ever one speaker at a time ( aside from momentary overlaps or attempts to butt in ) , this limit of four on the size of conversation groups means there are three listeners. That is a particularly interesting number, because it is three times the number of parties involved in a conventional grooming inter­ action - which always consist of j ust one groomer and one 'groomee'. The largest mean group size observed in any primate species is about 5 5 , in chimpanzees ( remember that these are mean group sizes, not the largest groups ever observed in a given species). This may be an extraordinary coincidence, but the pre­ dicted ( and observed) size of modern human groups, 1 5 0 individ­ uals, is almost exactly three times larger. In other words, the ratio between group sizes is exactly the same as the ratio between the

ties in the forested habitats they occupy often number only three to five individuals. But the fact that we share this characteristic with chimpanzees does suggest that the precursors for modern human societies of this kind were a lready established during the earliest phases of our history. Each of these three possibilities has evidence in its favour. If I had to guess which one was most likely to be right, I think it would have to be the last, in part because it fits modern hunter­ gatherer behaviour patterns and in part because we seem to share this pattern with the chimpanzees. Testing the Hypothesis

If language evolved to facilitate the bonding of larger groups, then we should be able to show that it has design features that would achieve this. One is that conversation groups should be propor­ tionately larger than conventional primate grooming cliques. Another is that conversation time should be predominantly devot­ ed to the exchange of social information. In one sense at least, the latter would be a strong test of the hypothesis, because conven­ tional wisdom has it that language exists to facilitate the exchange of information about the world in which we live - the view that we spend our time discussing the bison down by the lake. In order to test these predictions, my students and I sampled con­ versation groups in various locations. Our only prerequisite was that the people concerned should be engaged in relaxed social inter- 1 20 -

- 1 21 -

Grooming, Gossip and the Evolution of Language

Up Through the Mists of Time

number of individuals you can interact with at any one time: one for grooming monkeys, three for talking humans. My view is that human groups are three times larger than those of chimpanzees precisely because humans can reach three times as many social contacts as chimps for a given amount of social effort. The story became even more intriguing when I discovered that the size limit on human conversation groups seems not to be an accident of social rules. Rather, it turns out to be a consequence of limitations in our auditory machinery. Our ability to hear what is being said is j ust good enough to allow us to operate in conversa­ tion groups of this size. During the 1 9 5 0S and 1 9 60s the effect that noisy environments had on the detectability of speech was studied in considerable depth, and two facts of particular interest to us emerged. One is that, when the distance between speaker and listener is more than a bout two feet, the l istener finds it increasingly diffi­ cult to make out what the speaker is saying. Calculations based on the rate at which speech discrimination falls off with i ncreas­ ing distance between speaker and listener suggests that, under minimal noise conditions, there is an absolute limit of a bout five feet beyond which the listener simply cannot hear enough of what the speaker is saying. Beyond this distance, the speaker has to shout. In fact, Dan Nettle, one o f my students, has shown that the loudness with which cultures habitually speak is inversely relalted to the number of vowels in the language, and hence to the ease with which the language's sounds can be discriminated. Cultures which a bhor excessively close contact tend to shout at each other and have vowel sounds that are easily distinguished from one another. Even allowing a minimal shoulder-to-shoulder distance of six inches, a circle five foot in diameter would place an upper limit of about seven on the number of people who can hear what a speak­ er is saying. However, as background noise increases, so the mini­ mum tolerable distance between speaker and hearer shrinks and the number of people you can get into a circle decreases propor­ tionately. With background noise levels typical of a city street or a busy office, the limit on the group size is a bout five. At very noisy

co*cktail parties, it sinks to two ( and even then it can be an effort) . The second prediction made b y the hypothesis of the social lan­ guage was that people's conversations should be dominated by social topics. We have listened to conversations up and down the length of England, sampling individuals of different ages and social backgrounds. The technique was very simple. Every thirty seconds we simply asked, 'What is he/she talking a bout now?' Ou r results consistently yield th e same pattern: about two-thirds of conversation time is devoted to social topics. These include dis­ cussions o f personal relationships, personal likes and dislikes, per­ sonal experiences, the behaviour of other people, and similar top­ ics. No other individual topic accounted for more than 10 per cent of total conversation time, and most rated only two or three per cent. These included all the topics you might consider to be of great moment in our intellectual lives, namely politics, religion, ethics, culture and work. Even sport and leisure barely managed to rustle up a score of 10 per cent between them. After we had completed these studies, we discovered that other scientists had been sampling conversations. Nicholas Emler, a psychologist at Oxford University, has also been particularly interested in gossip and its uses. After listening in to conversations in Scotland while he was based at Dundee University, he too came up with a figure of a bout 60 to 70 per cent devoted to social topics. He concluded that one of the most important things gossip allows you to do is to keep track of ( and of course influence ) other people's reputations as well as your own . Gossip, in his view, is all about the management of reputation. Taken together, these observations provide strong support for the suggestion that language evolved to facilitate the bonding of social groups, and that it mainly achieves this aim by permitting the exchange of socially relevant information.

The evolution of large brains in the ancestral hom in ids, and the language facility that emerged from this, raises some fundamental questions. Brain tissue is by far the most 'expensive' tissue in the

- 1 22 -

- 1 23 -

The Expensive Tissue Hypothesis

Grooming, Gossip and the Evolution of Language

Up Through the Mists of Time

body. It differs from other body tissues in the extraordinarily high levels of activity it maintains even during the resting state. Nerv e ce lls work by maintaining an electric potential across the cell membrane by forcing sodium ions ( free atoms) out and potassium ions in. Allowing the floodgates to open creates the electrica l surge that we identify with a nerve firing. Since the ions naturally want to move from one side of the membrane where their concen­ tration is high to the side where it is low, it is necessary to have a mechanism that pumps ions in the opposite direction in order to keep the nerve ready to fire. A great deal of energy has to be expended by the ionic pumps to keep the electrical potential in the axon at the right level in readiness for action. In addition, the neurotransmitters that facilitate the transmis­ sion of electrical discharges between one neuron and the next are very expensive to manufacture, and considerable energy has to be expended in repl acing these every time an axon fires. Neural tis­ sue is, in fact, about ten times more expensive to maintain per kilogram than the rest of the body's tissues, consuming about 20 per cent of the body's total output of energy while accounting for only 2 per cent of its weight. This creates something of a problem. As we saw in Chapter 4, modern humans have brains about nine times larger than you would expect for a mammal of our body weight, and about six times larger than for a primate of our body size. Yet gross energy production is directly related to a body's size. So even though we have a brain that is much larger (and hence more expensive ) than is typical of other primates, the total energy output of our bodies is j ust what would be expected for a typical mammal of our size. How then do we ma nage to maintain a larger brain when we don't have any extra energy to fuel it? The answer, as Leslie Aiello and Peter Wheeler have pointed out recently, is that we have a much smaller gut than would be expected for a mammal of our size. When they looked at the ener­ gy consumption of different parts of the body, they discovered that, rather surprisingly, the brain, the heart, the kidneys, the liver and the gut between them account for 8 5 to 9 0 per cent of the body's total energy consumption in mammals. Hence to increase

the size of the brain, the extra energy required to fuel it must come fr om one of these organs. The problem is that the heart, kidneys and liver are related ra ther closely to body size in mammals, for the obvious reason that they are responsible for maintaining healthy tissue activity. A smaller heart means that less blood is pumped round the system, and therefore the muscles cannot work so hard; a smaller kidney or liver means that the blood cannot be purified so effectively, and thus carries less energy to the muscles ( an d an increased risk of the organism being poisoned through failure to extract toxins from the blood). In short, if you want a l arge brain, the only place you can really afford to take the extra energy from is the gut. The catch, of course, is that you cannot reduce the size of the gut without reducing the rate at which energy is absorbed into the bloodstream from the food you ingest. Catch 22. The rate at which the gut extracts energy from food is directly related to its area - which is, of course, in turn related to its overall volume. It seems that gut size ultimately places a constraint on brain size. If you want to have a bigger brain, you have to grow bigger in order to accommodate a bigger gut. The lesson here is that small mon­ keys can never be very smart - never mind evolve language because they do not have guts large enough to support the extra neural activity of a bigger brain. They simply don 't have enough scope for savings of scale. But there is one way out of this bind, and it is precisely the way the ancestral hominids found. You can reduce your gut size with­ out compromising your energy intake by eating foods that are either richer in nutrients or have their nutrients in a more easily assimilated form. That way, the gut has to work less hard to extract the same quantity of n utrients. Most primates are either leaf-eaters or fruit-eaters, although many of the small noctu rnal species are insectivorous. Although insects are a very rich source of energy, only very small animals can support themselves on such a diet, because insects are hard to catch and even the largest of them come in rather small packets. Of the two diets available to the medium to large primates, the less nutrient-rich - or at least, the one from which it is harder to

- 1 24 -

- 1 25 -

Grooming, Gossip and the Evolution of Language

Up Through the Mists of Time

extract nutrients - is the leaf-based diet. The cell walls of most leaves mainly consist of various kinds of cellulose which mammals cannot easily digest. Species with a leaf-based diet cope with the problem by using bacteria to ferment the leafy matter; the host then consumes the bacteria further on in the gut, so absorbing the nutrients from the cell indirectly. This is basically what cows, antelope and othe r ruminant mammals do: rumination is a period of fermentatio n during which the bacteria digest the leaf cells. Some primates like the colobus and langur monkeys of the Old World and the howler monkeys of the New World behave in a similar way (though with­ out ruminating or chewing the cud ) . There are severe costs to this strategy, though . In order to ferment the leaves, you need a big vat: all leaf-eaters amongst the mammals have very big guts large stomachs in which to ferment the leaves, and large intestines to provide absorbent surfaces for extracting the nutrients from the . bacteria ) . Leaf-eating is basically incompatible with a small gut. Fermenters face other problems that have important implica­ tions for our theory. A fermenting strategy requires time for the bacteria to do their work. It is also a very hot strategy, because the fermentation process generates a great deal of heat, raising the animal's core temperature. But if the body temperature rises, the bacteria cannot ferment effectively. So once they have filled up their stomachs, fermenters have to rest for a while to digest the food particles properly and create some space before they can start feeding again . V you watch a herd of cows in a field for any length of time, you will see this neatly demonstrated. After feeding for a couple of hours, they will all go and settle down in a corner to chew the cud . A couple of hours later, having cleared some space in their guts, they all set about feeding again. And so on throughout the day and night, alternating between feeding and ruminating on a roughly four-hour cycle. Among primates too there is a positive relationship between the proportion of leaf in the diet and the amount of time devoted to resting. Leaf-eating colobine and howler monkeys can spend as much as 70 to 8 0 per cent of their available daytime resting; species such as ba boons and chimpanzees that feed on more

energy-rich fruit-based diets spend as little as 1 0 to 20 per cent of their time resting. In consequence, the amount of time leaf-eaters have available for social activity is severely curtailed ( two to five per cent of their time ) compared with fruit-eaters, who may spend as much as 1 5 per cent of their time grooming. So even if their brains were big enough to support large groups, colobines and howlers would not have sufficient time to spare for grooming to be able to keep groups of any size together. All the large social primates are fruit-eaters in one form or another. Fruits, seeds and tubers (the underground storage organs of certain plants ) are the most energy-rich of all vegetable foods, and their energy is in the form most accessible to primates. As fruit­ eating apes, the ancestral hom in ids could not have significantly improved their diet as a way to reduce their gut size. Only one source of food available to them was more nutrient-rich, and that was meat. Flesh is energy-rich, and the energy is in a form particu­ larly easy to absorb during digestion. As a result, carnivores have rather small guts for their body size . By switching to a meat diet, the ancestral hominids were a ble to make significant savings in gut volume without sacrificing any of their energy intake. The initial increase in brain size about two million years ago seems to correlate with a shift from a predominantly vegetation­ based diet in the australopithecines (the early members of our lin­ eage ) to a diet with a significantly larger meat component in hom*o. Most of this meat probably came from scavenging the kills of other carnivores and opportunistic killing of birds, reptiles and baby mammals rather than from full-scale hunting. However, the sudden increase in the rate of brain expansion from around half a million years ago seems to coincide with the beginnings of orga­ nized hunting. From here on, meat becomes increasingly impor­ tant in the diet. The living floors of fossil sites in Africa and Eurasia start to contain the bones of large num bers of mammals, including other primates. The bones often bear cut-marks, s ug­ gesting deliberate dismemberment. Some species, like the now­ extinct giant gelada at the Olorgasaillie site in southern Kenya, appear in such numbers as to suggest that hominid hunting may have been a contributory factor in their subsequent extinction.

- 1 26 -

- 1 27 -

Grooming, Gossip and the Evolution of Language In short, Aiello and Wheeler argue that our super-large brai ns were possible only because we went in for meat-eating in a big way. That allowed us to live in larger groups, and that in turn allowed us to range more widely, colonizing the rest of the world in a series of migrations that fanned outwards from the ancestral homelands in eastern and southern Africa like the ripples on a pond. The Baby Needs its Bathwater

There were, however, hidden costs to developing large brains. If we plot the brain sizes of different species against their gestation periods in what is usually known as a 'mouse-to-el � phan � ' curv� , we find that the two variables are closely related. MIce, wIth theIr tiny brains weighing only tens of grams, have a three-week preg­ nancy, whereas elephants, with their human-sized brains, have a 2 I -month pregnancy. The fact that all species of mammals lie close to the line between these two extremes suggests that brain tissue is laid down at a constant rate during pregnancy, at a rate determined by the mother's a bility to channel spare energy into her foetus. In effect, it is the size of the baby's brain that deter­ mines the length of pregnancy, and all species give birth when brain growth is more or less complete. Baby primates, for exam­ ple, are born when their brains reach full size, with relatively little growth occurring after birth. Humans, however, are an exception. A baby human is born when its brain is less than one-third its final size. The rest of its brain development continues over the first year of life. In fact, if we calculate the equivalent gestation period for a conventional mammal of our brain size, we arrive at a mind-boggling 2 I -month pregnancy. This corresponds exactly with the time it takes the human baby's brain to complete its growth: nine months gestation in the womb plus an additional 1 2 months after birth. One consequence of this is that human babies are born prema­ ture and incapable of fending for themselves. A baby monkey or ape is capable of walking within hours of its birth. Within a mat­ ter of weeks it is a competent member of its social group. The - 1 28 -

Up Through the Mists of Time average human baby, by contrast, can barely muster even a gurgle to keep its long-suffering parents happy. By its first birthday, however, its brain is sufficiently developed to allow it to learn to walk and begin the serious business of living. Our premature births became necessary because the human brain size took off at a time when we were actually getting small­ er. We needed larger brains to cope with our bigger groups, but we were getting slimmer and shorter in response to other ecologi­ cal pressures. This wouldn't have been a problem were it not for the fact that the birth canal through which the infant passes as it is being born was only increasing as the square of body length whereas the brain size was increasing as the cube. Trying to squeeze an ever bigger head through what was becoming an ever smaller hole creates inevitable problems. Something had to give. What gave was the timing of birth. Instead of giving birth to infants whose brains had completed their growth (as is typical of most other mammals ) , we compromised and gave birth at the earliest possible moment at which the baby could survive, and allowed it to complete its brain growth outside the womb. We give birth to appallingly premature babies. And this is why true premature babies, those born after only six or seven months of pregnancy, have such a difficult time; they really are teetering on the brink of survival - because even when they are on time, human babies are premature. The costs of growing a large brain were thus considera ble for our ancestors. The whole business of successful reproduction, of investment in one's offspring, became drawn out and magnified. The human child takes proportionately longer than the 5 to 1 0 years typical o f monkeys and apes to absorb all the necessary information and experiences into that little brain to allow it to cope with the social world into which it has been born. In humans, the learning process extends over a 1 5 to 20-year period. In effect, as brain size increases, so everything from pregnancy and the age at weaning to sexual maturation and the business of reproduction gets slowed down and strung out over a progressive­ ly longer period. An increased parental role for males may well have become - 1 29 -

Grooming, Gossip and the Evolution of Language essential, because the females ( can we now call them women? ) could not bear the full costs o f child care while foraging o n their own. In other words, it must have been at this point that the unusually intense pair-bonding that occurs between human males and females first evolved. This is certainly implied by the reduced sexual dimorphism in body size. Throughout most of our early history, during the aus­ tralopithecine phase, males were substantially larger - in some cases as much as 50 per cent larger - than females. In mammals, striking sexual dimorphism is invariably associated with harem­ like mating systems, where a handful of powerful males share all the females between them. The reduced sexual dimorphism in the later hominids, where males are only 1 0 to 20 per cent heavier than females, suggests that females were shared more evenly among the males ( though some males still did better than the rest ) . One particularly reliable index o f the mating system i n primates is the relative size of the male canine teeth compared to those of the fema les . Species in which the males defend large harems, or compete in open arena to mate with individual females on a promiscuous basis, tend to show marked differences between the size of male and female canines, since these are the main weapons used in fights between males over access to females. In contrast, sexual dimorphism in the canine teeth is negligible (or even favours larger female canines ) in lifelong monogamists like the gibbons. Among the australopithecines and early hom*o the dimorphism is considerable, with males having canines that, rela­ tive to body size, are about 25 per cent larger than those of females , about the same relative difference as in the highly promiscuous chimpanzees. Canine dimorphism, however, appears to have declined steadily d uring the last two million years, reach­ ing a low point 5 0,000 years ago when male canines were only about 1 0 per cent larger than those of females. This implies a shift over time from strongly polygamous mating systems towards mild polygamy . This doesn't by any stretch of the imagination mean that monogamy had evo lved : despite the intensity of human pair­ bonding, there is no anatomical evidence for monogamy. But it - 1 30 -

Up Through the Mists of Time does suggest that harem groups were very much smaller, with per­ haps only two females attached to a given male. Many males wo uld have had only a single female. This is what we would expect if females were demanding increased provisions from the males whose children they were bearing. If the experience of mod­ ern hunter-gatherers is anything to go by, it is unlikely that a male could have provided sufficient meat for more than two females and their offspring. This would at least explain the incongruity between the anatomical evidence for polygamy and the intensity of pair-bonding (with its implications of monogamy) in modern humans.

First Words C HA P T E R


First Words

The evolution of language raises two fundamental and related questions. First, what form did language ( speech ? ) take when it first evolved ? Second, why, and when, did what was presumably once a single language diversify to give us the 5 000 or so lan­ guages we now have ? I am going to try to a nswer the first question in this chapter; I will address the second question in the following chapter. Since we are dealing with events in the distant past that leave no . fossil record, my answers will necessarily be fragmentary. However, as I shall try to show, we may at last be in a better position than before to provide some serious answers to these questions. Irrespective of when language first evolved, the puzzle is how non-linguistic forms of communication gave rise to a linguistic form. How did non-language transform itself into language? What did those first words sound like ? Would we recognize them today as a human language, complete with grammar and all the other accoutrements of modern conversation ? And last, but per­ haps not least, who spoke them ? Opinions differ about what the earliest forms of language might have been. One school of thought says that it arose from gestures, while another argues that it came from monkey-like vocalizations; yet another suggests that it arose from song. Each hypothesis is supported by enough evidence to make it seem plausible. It is convenient to start by considering the gestural theories of language origins, partly because they are still widely held and partly because they inadvertently raise issues that will turn out to be important to our story. I will deal with the song-based theories in the following section. - I32 -

Gestures on the Wind

One version of the gestural theory is based on the observation that the fine motor control used in both speech and aimed throwing tends to be located in the same half of the brain, the left hemi­ sphere in most people. Speaking requires very fine motor control of the lips, tongue, vocal cords and chest, all of which have to be integrated in j ust the right sequence to produce a particular sound. J ust try saying the sound a: as in hay - normally produced with the corners of the mouth pulled right back so that the l ips are narrowly parted - but with the lips rounded and pouted as they would be when producing the sound oh:. The result is recogniz­ able as an a: sound, but only j ust; it comes out more like a stran­ gled oi: than anything else. In addition to fine motor control, speech also requires precise control of breathing so that air is released from the lungs in j ust the right quantity and with j ust the right amount of force ( think about the difference between explo­ sive sounds like b or p and softer ones like e or c. ) O ne important anatomical change was necessary for this: the shift from the dog-like chest shape typical of all monkeys to the flattened chest characteristic of apes. The way a monkey's shoul­ der girdle is attached to its rib cage constricts the frequency with which it can breathe. The scapula (the flat plate of bone that pro­ vides the anchor and pivot point for each arm) lies on the side of the rib cage in most mammals, including all the monkeys. This allows the arms to move backwards and forwards when walking or running. The problem is that when the body's weight is on the arms, it restricts the chest's a bility to expand and contract during breathing: as a result, monkeys can breathe once, and once only, with each stride. When the apes adopted a climbing life-style - in which the arms are used a bove the head to pull the body vertically upwards, while the feet are braced against the tree trunk - the monkey-like rib cage of their ancestors was a serious impediment. Monkeys can­ not swing their arms in a circle; the position of the shoulder bones - I33 -

Grooming, Gossip and the Evolution of Language

First Words

blocks the movement of the arms. To allow them to reach up above their heads during vertical climbing, the ape's chest had to be flattened. With the scapula moved around to the back of the rib cage, the arm j oint could be positioned on the outer edge of the chest. With this arrangement, apes could now swing their arms in a complete circle at the shoulder. This is why gibbons (the 'lesser' apes) can brachiate (that is, swing from tree to tree using the body as a pendulum ) , the great apes can haul themselves up tree trunks and we can play baseball, while the monkeys can do none of these things. This flattened chest, in addition to preparing the way for the evolution of bipedal walking in our ancestors ( it helps to keep our centre of gravity over the feet when we are standing and walking), also freed the breathing apparatus from the constrictions suffered by the monkeys. We now breathe as often as we like, irrespective of what our arms are doing. We can now speak without interrup­ tion even when active. This small but important change in anatomy prepared the way for something else: aimed throwing. Although other apes and monkeys throw things, their accuracy is not always impressive. Only modern humans can throw a cricket ball from the outfield with the intention of hitting the wicket ( or at least getting it into the wicket-keeper's hands, given a modicum of gymnastic ability on the latter's part ) . Yet in terms of sheer arm power, the average chimpanzee could easily out-throw any Olympic field sports champion. Fortunately for all the budding Tessa Sandersons, no ape will ever stand much of a chance on the sports field because they lack the fine motor control required to throw a j avelin any distance. Aimed throwing is clearly important for hunting, so one obvi­ ous conclusion is that language evolved on the back of throwing. The fine motor control needed for aimed throwing, so the argu­ ment runs, provided us with the neural machinery for fine motor control of the organs of speech . The sensory and motor control nerves from one side of the body cross over to the other side of the brain ( the right side of the body is controlled by the left hemi­ sphere of the brain, and the left side by the right hemisphere ) , and

most people throw right-handed; it follows that motor control for throwing is located in the brain's left hemisphere, and hence that speech control will be located there too (as indeed in most people it is) . There are, however, a number of problems with this suggestion. For a start, language involves conceptual thinking of a very differ­ ent order to that needed for throwing. Secondly, it is difficult to see how a gestural language of any complexity could get going. We use gestures very effectively to give commands ( beckoning to indicate 'come here' ) , to draw someone 's attention to some­ thing ( pointing ), or to emphasize a point ( stabbing the air ) . We use them to express anger ( fist-shaking) or submission ( hands clasped in petition or held up in surrender), or to indicate friend­ ship ( waving or handshaking ) . But we never use gestures to express a bstract concepts, to indicate place or time other than the present, or to make plans for the future. If we could, charades would be a pointless game . ( OK, it is a pointless game, but the difficulty of expressing concepts in gestural form taxes our com­ munication skills to the point of a b surdity, so making it amus­ ing as a parlour game . ) More importantly perhaps, we don't use gestures to discuss other people's behaviour ( other than through simple comments such as raised eyebrows ) . In other words, we do not use gestural forms of communication to express much more than the kind of information about emotional states that monkeys and apes competently express using vocalizations ( a nd sometimes gestures too, of course ! ) . But the real problem with a gestural theory for the origins of language is simply its impracticality: you have to be in close visual contact with the person you're talking to. As the children of deaf parents quickly learn, you can cuss and swear all you like at your parents so long as their backs are turned. More sig­ nificantly perhaps, darkness occupies exactly half the day in the tropics. Those early linguistic humans would have had to sit in self-imposed 'silence' from dusk until dawn. Unable to tell sto­ ries of the old times, let alone argue about where to hunt the next day, the only possible evening pastimes they could have enj oyed were grooming and sex.

- 134 -

- 135 -

Grooming, Gossip and the Evolution of Language

First Words

Speech, on the other hand, frees us from these constraints. We can reminisce and tell stories over the dying embers of the fire; we can shout instructions or make enquiries over distances of half a kilometre or more, even when we cannot actually see the person with whom we are conversing. There's another question no one ever thinks of asking: why should the control of throwing have centred on the left side of the brain ? Why not the right ( with left-handed throwing) ? The only explanation is the rather unsatisfactory one that it was an acci­ dent of history. Now accidents of history do occur in biology; which genetic sex has the 'female' ( or egg-producing) role in reproduction is a classic case. I Brain asymmetry is more tricky, because it is not obvious why it had to be one hemisphere rather than the other: why couldn't humans have evolved into ambidex­ trous throwers, like all other primates ? The answer I want to propose ( and it is really no more than a speculation ) is that the right hemisphere was already fully preoc­ cupied doing something else much more important. Speech, when it evolved, localized in the left hemisphere because there was more free space there. And once speech had evolved, the fine motor control for throwing also centred there, either for the same reason or because the left hemisphere was already beginning to specialize in conscious thought, which is needed for accurate aimed throw­ ing. In other words, the sequence of events was precisely the reverse of that proposed by all gestural theories. I suggest this for a very simple reason: we now know that the right hemisphere is specialized for the processing of emotional information. There is evidence to show that emotional cues are detected faster when they are on the left side of the visual field (transmitted across to the right side of the brain2) than on the right side. This trait is widespread in the animal kingdom, and

appears to originate in a very ancient tendency for greater sensi­ tivity to visual cues to have evolved on one side. Fossil trilobites from 2 5 0 million years ago, for example, tend to have more scars on the right side, suggesting that pursuing predators attacked them more frequently from the ( predator's ) left-hand side. Fossil dolphin skeletons from 20 million years ago show the same pat­ tern of shark-tooth damage on the right-hand side, again suggest­


It is the XX-chromosomed sex i n mammals, but the XY-chromosomed sex in

birds and butterflies - the result of an entirely random decision on three quite sepa­ rate occasions as to which sex should have the eggs and which the sperm. 2.

Unlike all other systems i n the body, vision is split: the nerves from the left half

of each eye are fed across to the right side of the bra in, and those from the right half are fed across to the left side.

- 136 -

ing that the predator kept the prey in its left visual field as it chased it. Evidence from living species suggests that this tendency became elaborated into a generally greater sensitivity to visual and emo­ tional cues on the left side of the body. Julia Casperd and I have shown that gelada males tend to keep opponents on the left side of the visual field during fights. D uring the early phases of these encounters, the animals exchange threats using facial signals; if neither of them will submit, the confrontation will eventually escalate into a physical attack. It is obviously important for the males to monitor their opponent to pick up any inadvertent hint it gives a bout its true intentions. Is it bluffing when it makes heavy threats ? Are the eyes that momentarily flicker away betraying a rel uctance to press home an attack if pushed to the brink? The greater sensitivity of the right hemisphere means that keeping the opponent in the left visual field ( so its image falls on the right half of each retina3 ) ensures that the subtler cues are picked up. We humans do this too. Have you ever noticed that when peo­ ple are photogra phed, they tend to turn the head so that the left side of their face is directed at the camera ? Take a look at your family photo album. Only in very formally posed shots such as team photographs do people stare straight at the camera; in infor­ mally posed photos, taken when people know they are being pho­ tographed, they usually turn the head slightly to the right so as to watch the camera out of the left side of their visual field. By using an instrument called a tachistoscope to flash pictures 3 . An object on the left side of your body ( that is, in your left visual field) is pro­

jected on to the right side of the retinal field at the back of the eyeba l l. When an image passes through the lens of the eye, i t is inverted: we actually see the world reversed and upside-down.

- 137 -

Grooming, Gossip and the Evolution of Language

First Words

of actors' faces on to particular parts of the retina, Jim Denman and John Manning have shown that people are much more likely to identify correctly the emotion being expressed by the actor when the picture is flashed on to the right half of the retina than on to the left half ( or vice versa for left-handed people ) . In anoth­ er study, one of my students, Catherine Lowe, has shown that mothers carrying young babies are much more likely to detect a soundless grimace by the baby if it is being cradled on the left side of her body than on the right. (This explains why most people, and mothers in particular, cradle their babies on the left side of the body. The alternative suggestion - that it is to comfort the baby by allowing it to hear the mother's heartbeat and so remind it of its time in the womb - cannot be right: the heart lies in the centre of the chest, not on the left side as is often popularly assumed) . This asymmetry i n emotional cueing i s clearly o f very ancient origin because it is already present in other monkeys and apes. Mark Hauser has shown that the left side of the face tends to respond sooner and more intensely than the right side when mon­ keys give facial signals such as grimaces. The left side of the face is, of course, under the control of the right hemisphere of the brain. This asymmetry in sensitivity thus long predates the appearance of the hominids, never mind language-toting humans; it may even predate the origin of the primates. All this suggests that the right side of the brain was a lready heavily preoccupied with the business of monitoring and controlling emotional responses. Since it is the emotional behaviour of another animal that tells us what its intentions are, this is hardly surprising: being able to read these signals correctly, and then to respond to them with an equally emotion-driven response, is what primate social life is all a bout. Herein lies the beginnings of the hierarchy of intensionality we met in Chapter 5 . Given this, it seems natural that language should localize i n the left hemisphere. There was simply more free space there for set­ ting up the specialized neural control centres it required. I suggest that the reason we are so predominantly right-handed is because language allowed a special kind of thoughtful consciousness to

develop in the left hemisphere. This in turn made it possible to exert greater control over right-handed throwing than left-ha n ded throwing, so favouring the right side. The psychologist Julian Jaynes argued along somewhat similar lines in his seminal book The Origin of Consciousness in the Breakdown of the Bicameral Mind. For example, he used literary evidence from the Middle East to suggest that the person who wrote the Homeric poems of ancient Greece in around 1 200 BC was not fully conscious. Jaynes cites the striking lack of introspec­ tion in all the writings from this period: they do not refer to emo­ tions, but instead give straightforward narrative descriptions. Consciousness, he suggested, developed around the beginning of the first millennium BC as the left (linguistic ) hemisphere gradually succeeded in exerting control over the more unruly (emotional) right hemisphere. My own feeling is that Jaynes is on the right lines, but that his chronology confounds two separate events: the growing dominance of the conscious left hemisphere (which must have occurred when language evolved ) and people's abilities to give expression to their in ner emotional states. This contrast between the control centres for language and emotional behaviour has an interesting and rather unexpected consequence . It turns out that although language is localized on the left side of the brain, music (and poetry) is localized on the right side. In a neat study carried out some years ago, Thomas Bever and Robert Chiarello showed that untrained musicians ( people who have had less than three years of music lessons dur­ ing their lives) recognize tunes more quickly if they are played to the left ear through a set of earphones ( hence monitored by the right hemisphere of the brain ) than to the right ear. Trained musi­ cians did not show this to quite such a marked extent, as might be expected from the fact that they had been taught consciously to analyse musical excerpts into their component parts rather than hearing the tune as a unified whole. This suggests that while the conscious manipulation of music goes on in the left hemisphere (along with speech and other 'con­ scious' activities), the emotional response to the tunefulness and rhythm of music goes on in the right hemisphere. Similarly, there

- 138 -

- 139 -

Grooming, Gossip and the Evolution of Language is evidence to suggest that the music of poetry is handled in the right hemisphere, whereas the linguistic content - the words - is dealt with on the left side. This is why, when stroke patients lose their speech following a left-sided stroke, nursery rhymes are used to encourage them to speak again; these simple forms of poetry and song are more likely to be stored in the undamaged right hemisphere. M oreover, the fact that music is located in the right hemisphere is one good reason why the alternative suggestion that language evolved from song cannot be wholly right. Words are certainly used in song, but words are dealt with in the left hemisphere. Song ( and music generally) certainly arouses our emotions in a way that words alone find it difficult to do; and song can be used in a very powerful way to express a group's collective emotional arousal. But it's hard to see how something localized in the right hemi­ sphere can produce something localized in the left hemisphere. A more plausible suggestion is that when language evolved it was hijacked by the emotionally powerful music centres to produce song, because music and song are a very potent means of express­ ing a group's emotional response. All in all, gestural theories of language origin seem pretty implausible. In any case, we can point to precursors for almost all the features of human verbal communication in the vocalizations of Old World monkeys and apes. This makes vocal theories of language evolution more plausible right from the outset. Recall the vervet monkeys that we met earlier. Dorothy Cheney and Robert Seyfarth have been a ble to show that vervet vocaliza­ tions do convey meaning. They are not simply expressions of emotional state. The alarm calls refer to specific types of predator, and the hearer knows from the auditory information alone which type of predator the caller is describing; contact grunts specify a great deal about the ongoing situation. To this we can add the conversational patterns of the gelada. Everyone who has ever seen gelada has commented on the extra­ ordinary calling sessions they engage in. Bruce Richman has shown that the calls in these sequences are timed very closely to fit between the calls of others. Richman concluded that the animals

First Words must be timing their calls by anticipation rather than simply by response to the previous caller. This is one of the characteristic features of human conversation: two speakers intersperse their segments of speech and their interjected comments ( 'Oh ! ', 'Well, I never! ' ) so that (most of the time) only one person is talking. The flo� of � onversation is almost continuous. They achieve this by , antICIpatmg the end of the other speaker's phrase or sentence often using cues that the speaker provides. These include a sligh rise in pitch towards the end of a sentence, and the tendency for the speaker to glance at the listener a few words short of the point when he or she is ready to stop talking. Another bastion of the uniqueness of human language is the production of vowel sounds. Without them, language would be impossible; vowels allow us to divide up the flow of sounds into easily discriminated segments ( syllables) and hence to form words. It has long been held that monkeys and apes cannot make these sounds because their mouths and throat cavities are the wrong shape. However, several recent studies of the way monkeys such as the gelada and the macaques produce sounds have shown that in fact, they do make these vowel-like sounds. This implies tha the machinery to produce the sounds of human speech was in place long before humans were even a twinkle in the eye of evolu­ tion. The problem is not the mechanics of producing the sounds so much as the machinery for co-ordinating sound production and the cognitive mechanisms for attaching meanings to sounds of this kind (although given the vervet studies, this last point must remain in some doubt ) . In other words, w e c a n already see many hallmarks of human speech in the Old World monkeys. In the vervet's calls we have an archetypal proto-l anguage. Quite arbitrary sounds are used to refer to specific objects, to convey information a bout who is doing what (or about to do what) . In addition, these calls can be over­ laid with varying degrees of emotional overtone, much as our own verbal statements are. There is no need for a gestural phase. It can all be done by voice. It seems but a small step from here to for­ malizing sound patterns so that they can carry more information. And from there to producing language is but another small step.

- 141 -

Grooming, Gossip and the Evolution of Language

First Words

All this suggests that the evolution of the capacity for language was the result of the gradual coming-together of several originally unrelated anatomical and neurological components over a long period of time. No one of them was in itself the trigger for the evolution of language, but each was essential. Had any of them failed to evolve, humans would not be speaking to each other today, and you would not be reading this book.

a tradition even now maintained by the All Blacks, New Zealand's national rugby team. Scottish regiments were always preceded by their pipers as they marched into battle, a tradition observed as recently as the 1 944 D-Day landings in Normandy. We sing our national and club anthems on the sports field with an intensity that we reserve for almost nothing else. But not all song and dance has this kind of function in human society. We sing in church and around the camp-fire, we sing in bars and in theatres, under circ*mstances that have little to do with nationalism, battles or the mating game. So why do we do it? Paradoxically perhaps, in answering this question we will find an explanation for the one feature of h uman behaviour that is really dif­ ficult to explain: our extraordinary willingness to subject ourselves to someone else's will. The crowd effect is at the same time the most bizarre and the most frightening aspect of human behaviour. Back in the 1 9 60s psychologists identified a phenomenon that became known as 'risky shift': if you ask someone to express an opinion or do something that is slightly extreme (such as support capital punishment) , they will typically express quite moderate views. But if you let them discuss it as a group first, the result will invariably be a much more extreme opinion. We see the same effect in religion: left to their own devices, people will be moder­ ate and tolerant, but in groups their attitudes towards deviance or those of different views become more extreme. The result as often as not is a jihad, a holy war against the infidels. To this extraordi­ nary phenomenon we owe the Crusades, Northern Ireland, Rwanda, Yugoslavia and the host of ethnic wars, racial feuds and nationalist vendettas that have sullied the history of our species since time immemorial, not to mention the bizarre business of the fatwa against the writer Salman Rushdie. In an indirect way, the explanation lies, I think, in a little­ noticed aspect of song and dance: they are both very expensive activities to perform. Of course, we are all intuitively aware of this. How often have we staggered off the dance floor unable to perform another step ? Nor are the singers and musicians spared; they drip with sweat at the end of the performance. Singing, as every aspiring opera diva knows, is hard work: to do it well

Ritual and Song

Although we can accept that language began as conventional ape­ like vocalizations, there are alternative views about the next step. One is that language evolved as a form of song, reinforcing dance­ like rituals that were designed to co-ordinate the emotional states of all the members of the group. The other is that it evolved to exchange information about other individuals in the group. I have so far tended to assume that the second view is the whole story: that language evolved to exchange social information. But it is clear that song plays an especially important role in our lives, so we should consider the merits of the idea. One of the more intriguing features of human behaviour is the extent to which song and dance feature in our social life. No known society lacks these two phenomena. But when you stop to think about them, they are both very odd activities. Singing is of course something we associate with many animal species, includ­ ing birds and prima�es such as the gibbon, whose morning vocal displays we happily refer to as song. For the most part, however, the ' songs' of birds and primates are mechanisms either for defending territory or for advertising for a mate . To be sure, human song is used for these purpose too. Maasai warriors sing and dance to display their personal prowess before the assembled maidens. And the songs of males in these contexts invariably extol the manly virtues of the singers. We do not often advertise our ownership of land by song, but it is not uncommon for men to sing when going into battle. The Maoris of New Zealand, like many of the Polynesian peoples, used ritual song and dance as a means of intimidating their enemies before battle, - 142 -

- 143 -

First Words

Grooming, Gossip and the Evolution of Language requires great control of breathing and articulation, and that needs considerable practice. It is, I suspect, no accident that the most evocative and stirring singing is often in the lowest registers of the voice: for example, the bass chanting of the Orthodox Christian liturgy. Deep sounds are difficult to produce, and typi­ cally require large body-spaces to serve as resonators. There are well-studied instances in the animal kingdom of deep sounds being used to frighten off opponents by signalling that the caller is a big animal; they involve species as different as toads and red deer. 'Deep croak', as the phenomenon is known to biologists, works because it is a difficult signal to forge : only the largest animals can successfully produce the deepest sounds and sustain the energetic costs of doing so. A deep voice indicates a large and powerful body. Even we humans give way without a moment's hesitation to those we believe to be larger. We give way to bigger people on the street while barging in front of smaller people. This is a particular problem for women, because of their smaller size; some have mis­ takenly attributed it to men 's 'natural' aggressiveness towards women - but big men do it j ust as much to little men as to women, and big women are j ust as likely to do it to little women. These unwritten rules of behaviour make curious and unexpect­ ed appearances in our lives: for example, in no US presidential election since the war has the shorter candidate won. In fact, when D ukakis stood against George Bush in 1 9 8 8 , his managers insisted that he and Bush should not speak at the same lectern when they held a televised debate; a lectern set at normal chest height for the six-foot-something Bush would have practically come up to Dukakis's chin, and the managers were afraid that this alone would lose him the election. Instead, the lectern was low­ ered after each candidate had spoken, then raised again to the cor­ rect height for the next speaker, so that it always appeared at the same relative height on each candidate. Campaign teams are now very careful to ensure that candidates never have to stand side by side on these occasions. We are often surprised by the shortness of our idols when we actually meet them. The most usual comment people make after - 1 44 -

meeting the Q ueen is, 'Isn't she smaJi ! I'd expected her to be much taller ! ' We expect the successful and the powerful to be tall. When asked to choose adj ectives to descri be successful and unsuccessful peop le, or leaders and followers, people are much more likely to , conSIder words like 'tall ', 'intelligent' and 'attractive' as being true of successful people. And as it turns out, they are not entirely wrong. Research has shown that successful people do tend to be tal ler, on average. A study in Germany by A. Schumacher showed th at, when weighted for social class and educational achievement, , senIor nurses are taller than j unior nurses, skilled carpenters taller than unskilled ones, successful lawyers taller than less successful ones, and senior managers in industry taller than middle man­ agers. W et er this is beca use taller people a re more intelligent or whet er It IS beca use we are more likely to give way to them remaInS to be decided. The fact is that tall people seem to have to work less hard to be successful. If you're shorter than the average, you often have to be bloody-minded to achieve the same level of success - the Napoleon syndrome. Resorting to deep-bellied sounds is almost universal across uman cul tures in situations where we want to create a lasting , ImpreSSIOn. Maori and Maasai warriors hum and grunt in the lowest registers in their war-songs. Successful pu blic speakers do not squeak and warble in falsetto, but lower the pitch of their voic � . Adolf Hitler growled his way through his extraordinarily rousmg speeches . And it was not for nothing that Margaret Th atcher was trained by her image-makers to lower her speaking , VOIce by nearly half an octave below its natural pitch when she became leader of the British Conservative Party in 1 9 7 5 . Had she not done as they asked, it is quite conceivable that she would not have won such a landslide victory in 1 9 7 9 . In effect, Mrs Thatcher's minders wanted her t o sound more like a man. And this should remind us of the fact that boys ' voices ' break' at puberty to produce a deeper, richer tone than women's voices. Quite why this should happen has always been something of a mystery. After all, boys and girls do well enough with their lighter treble voices, and women seem to cope q uite adequately throughout their adult lives. 'Deep croak' provides us with a pos-

� �

- 145 -

Grooming, Gossip and the Evolution of Language sible answer: human males have been under intense selection pres­ sure to evolve deeper voices during their sexually active years. They have to compete with each other for access to females for mating, and yelling matches have been ( and still are) part of the process whereby rivalries are settled. Females have no need to develop deep voices because they do not compete with men or with each other in quite the same way. But they have almost cer­ tainly added fuel to the fire of sexual selection by being especially sensitive to deep male voices, thus adding the processes of female choice to the processes of inter-male rivalry ( I shall have more to say about sexual selection in Chapter 9 ) · But there i s more t o song and dance than simply ' deep croak'. It makes us feel good to sing and dance. It generates euphoric highs, as well as feelings of happiness and warmth. Both activities are hard work; and both are ideal activities to generate surges of opi­ ates from the brain - which is almost certainly why we feel so good after performing them. So why should we have latched on to this curious effect and taken up these activities with so much enthusiasm? The answer, I suggest, lies in the fact that humans live in very large groups, and large groups are difficult to keep together for any length of time. They are perpetually at risk of fragmenting because of the conflicting interests of so many differ­ ent individuals, not to mention exploitation by free-riders. As the group's size increases, factions with opposing views develop, and we begin to take sides. Trying to hold together the large groups which the emerging humans needed for their survival must have been a trying busi­ ness . We still find it difficult even now . Imagine trying to co­ ordinate the lives of I 50 people a quarter of a million years ago out in the woodlands of Africa. Words alone are not enough . No one pays attention to carefully reasoned arguments. It is rousing speeches that get us going, that work us up to the fever pitch where we will take on the world at the drop of a hat, oblivious of the personal costs. Here, song and dance play an important part: they rouse the emotions and stimulate like nothing else the production of opiates to bring about states of elation and euphoria .

First Words The anthropologist Chris Knight has argued that the use of rit­ ual to co-ordinate human groups by synchronizing everyone's emotional states is a very ancient feature of human behaviour and coincides with the rise of human culture and language. H argues that to co-ordinate behaviour within a group in the form of ritual requires language, and that this may have been the final stimulus for the evolution of language. To be a ble to say, as southern African Bushmen do, 'Now let's pretend to be eland in order to do the Eland Dance' (a dance used to celebrate a young girl 's first menstruation, her badge of membership of the women's group ) , it is necessary to have language. Knight is surely right about the way language is used to formal­ ize and manage ritual. But I am less convinced that language evolved specifically to make ritual possi ble. My own view is that it evolved to facilitate bonding through the exchange of social infor­ mation, and was later hijacked for use in these semi-religious con­ texts in such a way as to formalize what may well have been pre­ existing practices. The eland dances of the late Pleistocene would not have been formally construed or tied to specific events. Rather, they were more likely to have been informal, spontaneous affairs, more like Saturday afternoons on the terraces of the local football club. These dances seem to me to be very ancient rituals indeed. Language allowed us to formalize their spontaneity, so giving them more coherence by providing them with a metaphysi ­ cal or religious significance. But this, I think, must have happened during the Upper Palaeolithic RevQlution, when we see the first evidence for religious beliefs and symbolic thought. And this unexpectedly brings us back to one of the more curi­ ous aspects of language, namely its complete inadequacy at the emotional level. It is a most wondrous invention for conveying bald information, but fails most of us totally when we want to express the deepest reaches of our innermost souls. We are so often 'lost for words' in such circ*mstances. Language is a won­ derful introduction to a prospective relationship: we can find out a great deal about a person with whom we might be thinking of forming a relationship or an alliance. But when the relationship reaches the point of greatest intensity, we abandon language and

- I 47 -

Grooming, Gossip and the Evolution of Language return to the age-old rituals of mutual mauling and direct stimula­ tion . At this crucial point in our lives, grooming - of all the things . . we mhent from our primate ancestry - resurfaces as the way We reinforce our bonds. We use it because physical contact is deep ly . movmg and reassuring in a way that language cannot be. And it achieves that because monotonous stroking and rubbing stimulate opiate production much more effectively than words can ever do. Paradoxically, it seems that j ust as we were reaping all the advantages of language, we had to back-pedal on a bandoning the more primitive processes. Just as we were acquiring the a bility to argue and rationalize, we needed a more primitive emotional mechanism to bond our large groups and make them effective. Language allowed us to find out about each other, to ask and answer questions a bout who was doing what with whom. But of itself, it does not bond groups together. Something deeper and more emotional was needed to overpower the cold logic of verbal arguments. It seems that we needed music and physical touch to do that. We have the most remarkable computing machines in the natural world, the most articulate communication system, the most sophisticated minds, yet we are ultimately dependent on crude hormonal tricks to ensure that our groups remain coherent and focused on the common goal that we so urgently need in order to survive and reproduce effectively. Maiden Speech

All this raises an interesting question: if language evolved to facilitate group cohesion, who spoke first ? The bison-down-at­ the-lake view of language naturally assumes that it was the males trying to co-ordinate their hunting activities. However, in most pr imate species it is the females that form the core of society, cre­ . atmg the group and giving it coherence through time. The males, by contrast, are less constant in their social a ffinities, often wan­ ering from group to group in search of better mating opportuni­ tIes . In most species (chimpanzees appear to be one of a handful of exceptions), females remain in the groups into which they were born, whereas males typically leave at puberty and transfer

- 148 -

First Words to another group. Some males even continue to move from group to group throughout most of their lives. If females formed the core of these earliest human groups, and language evolved to bond these groups, it naturally follows that the early human females were the first to speak. This reinforces the s uggestion that language was first used to create a sense of emotional solidarity between allies. Chris Knight has argued a passionate case for the idea that language first evolved to allow the females in these early groups to band together to force males to invest in them and their offspring, principally by hunting for meat. This would be consistent with the fact that, among modern humans, women are generally better at verbal skills than men, as well a s being more skilful in the social domain. However, the current consensus amongst evolutionary a nthro­ pologists is that humans do not have fema le-bonded societies in this sense . The evidence for this is that patrilocality ( brides mov­ ing to the husband's village ) is characteristic of most, even if not all, traditional societies. However, this may be a consequence of the fact that most of these cultures live in conditions where males are a ble to control the resources (land, hunting grounds ) that women need for successful reproduction . Several lines of evidence suggest that in more equitable economies ( such as those of hunter­ gatherers and modern industrial societies) female kinship and alliances are much stronger, and males may as often as not be forced to move to their wives' villages. Several lines of evidence support this suggestion . Among central African pygmies, for example, Y -chromosome genes are more widely distributed than X-chromosome genes, which are much more clustered in their distribution. This suggests that the women have tended to remain close to their female kin groups, whereas men have moved across wide areas in order to mate (so distribut­ ing the genes on their Y-chromosomes much more widely) . A sim­ ilar finding emerges from sociological studies carried out in the slums of east London during the 1 9 5 0s : not only were close female kin (mothers, daughters, sisters, aunts, nieces ) crucially important in enabling a woman to live and reproduce successfully in this impoverished environment, but married women lived - 1 49 -

Grooming, Gossip and the Evolution of Language

First Words

significantly cl oser to their own parental homes than to thei r husband's parents. We have detected elements of this in our own research. In a study of social networks, Matt Spoors and I found that wome n not only had slightly larger networks of regular contacts (people they contacted at least once a month ) than men, but these also included a higher proportion of their i mmediate same-sex kin (the limit being set at cousins ) . Even more impressive evidence comes from an experimental study we carried out in collaboration with Henry Plotkin, Jean-Marie Richards and George Fieldman ( the details of which are given on page 1 64 ) . We found that women were prepared to incur almost as much physical pain to provide a financial reward for a female best friend as for themselves, but men were not nearly so altruistic towards their male best friend. Taken together, these findings suggest that female bonding may have been a more powerful force in human evolution than is sometimes supposed. If so, then the pressure to evolve language ma y well have come through the need to form and service female alliances, as Chris Knight suggests, rather than through either male bonding or male hunting activities, as conventional wisdom has always assumed. It still remains unclear whether the initial stimulus to language was provi ded by the emotional uplift of the Greek chorus or by the need to exchange information a bout other members of the alliance or group. The fact that monkeys like the gelada common­ ly use contact calls in choruses that often rise to extraordinary heights of emotional crescendo might be taken as some evidence to favour the song hypothesis. When gelada reproductive units contact call, usually only one or two animals are involved. But every now a nd then, the whole group will come together in a rousing chorus of grunts and moans, ca pped at the end by the harem male's dramatic vibrato yawn which acts like a full stop at the end of a collective sentence. But it is the females, counter-call­ ing to their grooming partners within the unit, that start the process off a nd for whom it seems to be most important. This move from simple contact ca lling of the kind seen in savannah ba boons to the emotionally charged choruses of the

gel ada would seem to be an equally natural development for our own a ncestors when grooming time requirements slid inexorably beyond the time actually available as group sizes increased through the Pleistocene ( see page 1 1 4 ) . I f so, then it seems that it was the females who provided the real impetus along this route. Although male monkeys and apes commonly form coalitions, they seem to be much less tightly bonded than those of females. Frans de Waal's studies of the Arnhem Zoo colony of chim­ panzees show this rather clearly. He found that females' alliances were more long term and based on kinship, whereas males' alliances were shorter-lived and based on expediency and the demands of the moment. Alliances of the latter kind require only that the parties recognize the situation and what their ally is really trying to do. Long-term alliances, on the other hand, depend on deeper bonding processes: that almost certainly means a more emotional basis. But at some point this purely emotional behaviour must have given way to true language and the exchange of information . It is difficult to see how the very large groups that began to emerge towards the end of the Pleistocene could have been bonded suc­ cessfully without the ability to exchange social information. Important as emtional bonding is to social meshing in our small inner circle of intimate friends, there appears to be a limit on the number of people we can bond with in this way (the so-called 'sympathy-group' of 10 to 1 5 : see page 7 6 ) . Managing our rela­ tionships with the outer circle of acquaintances depends much more heavily on social knowledge than on emotional empathy.

- 1 50 -

- 1 51 -

Babers Legacy C HA P T E R

scattered a broad upon the face of the whole earth. And the Lord came down to see the city and the tower, which the chil­ dren of men builded. And the Lord said, Behold, the people is one, and they have all one language; and this they begin to do: and now nothing will be restrained from them, which they have imagined to do. Go to, let us go down, and there confound their language, that they may not u nderstand one another's speech. So the Lord scattered them a broad from thence upon the face of all the earth; and they left off to build the city. Therefore is the name of it called Babel; because the Lord did there confound the language of all the earth.


Ba bel's Legacy

The oddest feature of human language is its ha bit of producing mutually incomprehensible tongues with astonishing speed. Barely two millennia separate French and Italian from their mutu­ al ancestor Latin, yet most native speakers of these two closely related languages could no more understand each other than they could Latin. Danish and Swedish descend from different local dialects of Scandinavian, yet after barely a thousand years they are now all but mutually incomprehensible. Read Chaucer's Canterbury Tales in the original, and you will be only too aware of the extent to which English has changed in 600 years; half the words are unrecognizable. Even Shakespeare, at a mere 400 years' remove, can prove disconcertingly opaque at times. Our task in this chapter is to try to account for this phenomenon. The Descent into Babel

The mythologies of many peoples around the world share a belief in a common origin for mankind. Most of these stories imply (though they seldom say so explicitly ) that at some remote point in history everyone spoke a common language. The biblical story of the Tower of Babel actually makes a point of telling us that all humans once spoke the same language. Chapter XI of the book of Genesis gives the story: And the whole earth was of one language, and of one speech. And it came to pass, as they j ourneyed from the east, that they found a plain in the land of Shinar; and they dwelt there . . . And they said, Go to, let us build us a city and a tower whose top may reach unto heaven; and let us make us a name, lest we be - 1 52 -

The Tower of Babel was no myth: it really did exist. Its real name was Etemenanki (meaning 'the temple of the platform between heaven and earth ' ) , and it was built some time in the sixth or sev­ enth century BC during the second great flowering of Babylonian power. It was a seven-stage ziggurat, or stepped pyramid, topped by a brilliant blue-glazed temple dedicated to the god Marduk, by then the most powerful of the local Assyrian pantheon. A century or so later, in a bout 4 5 0 BC, the Greek historian Herodotus strug­ gled up the steep stairways and ramps in the hope of seeing an idol at the top. Alas, there was nothing but an empty throne. However, the myth-makers of ancient Israel seem to have been on to something. Linguists now believe that the world's languages do in fact have a common origin, though it is only recently that they have been a ble to do more than speculate about the history of language . However, the period of this common language long predates the building of the Tower of Babel, and it is difficult to see why real events in sixth-century Be Babylon should have become associated in the minds of the later writers of the book of Genesis with what might be genuine folk memories of a time when everyone spoke a common tongue. I At the time when the Babylonians built the tower of Babel. By then, most of the world's I.

That tribal fol k memories ca n reta in knowledge o f events that ha ppened in the

d i sta nt past is w e l l i l l ustrated b y the fact that, according to the archaeologist Josephine Flood, the origin stories of some A ustra l i a n Aboriginal tri bes conta i n su rpr i s i ngly accu rate descripti o n s of the l an d surfaces b e l o w the Ta s m a n S e a o ff

- 1 53 -

G rooming, Gossip and the Evolution of Language

Babel's Legacy

major language groups were a lready well established. The reconstruction of the history of languages dates back to Sir William Jones, a scholarly and enquiring man who was appointe d a judge in Calcutta towards the end of the eighteenth century . Convinced that he ought to study Hindu legal authorities in the original if he was to give proper legal judgements, he set about learning Sanskrit, the a ncient ( but by then dead) language of northern India. As his learning progressed, Jones became increas­ ingly struck by the similarities between Sanskrit and the ancient European languages, Greek and Latin , that he knew so well. Making due allowance for sound changes, he was able to identify sufficient similarities between the words of each of these lan­ guages to claim that they derived from a common mother tongue. Cla ssic examples include words like ' brother' - which has clear similarities to the Greek phrater, the Latin (rater, the Old Irish brathir, the Old Slavic bratre and the Sanskrit bhrater - and the parts of the verb 'to be', which include is in English and Old Irish, esti in Greek, est in Latin, yeste in O ld Slav and asti in Sanskrit. In most cases, the slightly different forms of these common words are the result of consistent shifts in pronunciation, a point first noticed by the folklorist brothers Jakob and Wilhelm Grimm. Examples of these shifts were the Latin ( and ph sound changing into the b of English and other Germanic languages, and the p and t sounds of the ancestral languages ( a s in Sanskrit piter, Latin pater) changing to the ( ( or v) and th sounds of the l ater Germanic languages ( German vater, English (ather) . The scholars of the nineteenth century were inspired by this suggestion, and spent much time trying to reconstruct common

ancestors for various language groups. So much so, in fact, that by the end of the century the academicians of the Societe de Linguistique in Paris were exasperated enough to ban speculations a bout the history of languages from their meetings. But by the time historical linguistics fell out of fashion, it had been possible to reconstruct the relationships between most of the world's better-known European and Asian languages. It was soon recog­ nized, for instance, that the languages of most of Europe ( except Basque and the Uralic-Yukaghir language group that includes Hungarian, Finnish and Estonia n ) and those of southern Asia as far east as the Indian plains (including Persian and the modern derivatives of Sanskrit) all belonged to the same group, now known as Indo-European . This a ncient language i s thought t o have originated somewhere north of the Danube basin in a bout 5 000 to 6000 Be. Indo­ European languages share the same words for winter, horse and domestic animals such as sheep, pigs and cattle, as well as words associated with leatherworking, ploughing and planting grain. Taken together, these suggest that the ancestral Indo-Europeans led a semi-nomadic existence in which agriculture played a promi­ nent role . Their gods were clearly the ancestors of the Indian, Mediterranean and Celtic deities; and they had a well-esta blished socia l structure, since their descendant languages share many of the words for kinship and family structure. One thing they do lack, however, is a common word for 'sea': it's clear that the Indo­ Europeans did not live near any major ocean or lake coasts. D uring the last few decades, there has been a renewed interest in reconstructing language trees . The consensus now seems to be that Indo-European and the other language groups of Europe ( Basque, Uralic, etc . ) and Asia (the Semitic languages of North Africa and the Near East, the Altaic languages that include Turkic and Mongolian, and the Elemo-Dravidian languages of southern India ) represent the descendants of a superfamily known as Nostratic which probably originated around 1 3 ,000 Be. Thanks to the hard work of a group of Russian linguists, attempts to reconstruct Nostratic have been quite successful, at least in the sense of being a ble to produce a vocabulary of several

the coast of southern Australia, as well as accounts of how the seas rushed in to sever the land li nks between the mainland and many of the islands off the northern and southern coasts. During previous ice ages, the sea bed in these areas would have been exposed as dry land. The last time the A boriginals' ancestors could have walked on them was around


years ago, j ust before they sank beneath the

waves for the last time as the ocean levels rose with the melting of the ice-sheets at the close of the last Ice Age. Moreover, it seems that the Australian Aboriginals are not a lone in this respect. The Scandinavian Ragnarok legend gives an account of a time known as the fimbulvinter when severe winters followed each other in succes­ sion without intervening summers; it has been suggested that this is a fol k memory of the ' little ice age' that afflicted northern Europe a round

- 1 54 -

1 000 Be.

- 155 -

Grooming, Gossip and the Evolution of Language

Babel's Legacy

thousand words that appear to be plausible common ancestors for the appropriate words in the modern Eurasian languages of this family. These include words such as tik meaning finger ( or one), from which the modern English word digit, the French doigt and the Latin digitus, as well as the Hindi ek ' ( one), are all said to derive. The proto-Indo-European melg, 'to milk', has similarities with the proto-Uralic malge, 'breast', and the modern Arabic mIg, 'to suckle', again suggesting a common origin. Of particular inter­ est is the fact that these languages share words for certain kinds of things but not others. There are words for dog kujna, from which derive German Hund and English hound and for things connect­ ed with the outdoors, such as marja, ' berry', but there appear to be no words associated with farming. This suggests that the people who spoke Nostratic were hunter-gatherers who lived in the period prior to the discovery of agriculture around 10,000 years ago. In addition to Nostratic, four other non-African language superfamilies are now recognized: Dene-Caucasian ( the languages of the Arctic and sub-Arctic regions of Asia and North America ) , Amerind ( most o f the native languages o f the Americas south of the Canadian border) , proto-Australian ( the languages of the Australian continent) and Austro-Asiatic ( the languages of south­ east Asia ) . The languages of Africa south of the Sahara are more confusing, and their position with respect to these five superfami­ lies remains unclear. But three main families have been identified: Khoisan ( the languages of the southern African Bushman and related peoples), Niger-Kordofan (the Bantu languages that domi­ nate most of western, central and eastern Africa ) and Nilo­ Saharan (the languages of the largely nomadic peoples on the southern rim of the Sahara desert ) . There have even been attempts t o reconstruct the common orig­ inal of these eight superfamilies, the so-called 'proto-World' tongue. Linguists like the Russian Vitaly Shevoroshkin, now at the University of Michigan, claim to have been able to identify some 200 words of proto-World. These include nigi ( or gini) , 'tooth' , with its derivates i n the Congo-Sa haran nigi, the Austro-Asiatic gini, the Sino-Caucasian gin and the Nostratic nigi ( from which may derive the modern English words nag and gnaw) . Similarly,

the English word tell has been related to the proto-World tal (or da/), 'tongue' - telling being an important activity for your tal.

the next half-century. They are mostly languages which currently have fewer than a thousand native speakers, most of whom are already elderly. The time may come, it has even been suggested, when the world will be dominated by just two languages; on pres­ ent perform ance, these will almost certainly be English and Chi­ nese. The loss of all these languages will, of course, be a pity. As we lose them, we lose fragments of our past, for languages repre­ sent the history of peoples, the accumulation of their experiences, their migrations and the invasions they have suffered. But this observation overlooks one curious feature of human behaviour: our tendency to generate new dialects as fast as we lose others. English has spread around the globe to become the

- 1 56 -

- 1 57 -



The reconstruction of long-dead languages is, of course, a chancy business, and attempts such as those I've described have not been without their critics. Indeed, they have often been con­ demned as fanciful. Some linguists point out that the rate at which languages change naturally over time is such that after about 6000 years it would be virtually impossible to identify any common terms. Whether we will ever be able to reconstruct the original languages is perhaps less important than the fact that we can trace common a ncestries for many of them. Although it may be fun to converse in Nostratic ( as the Russian linguists regularly did ) , the more significant question is why languages should have diversified to produce so many thousands of mutually incompre­ hensible tongues from a single common origin. The Dynamics of Chaos

There are estimated to be about 5 000 languages currently spoken in the world today, depending on which you count as dialects and which as fully fledged languages. To these, you can perhaps add a handful of ' dead' languages that are still taught in schools (ancient Greek and Latin) or used in religious services (Sanskrit and Ge'ez) . Linguists reckon that well over half of all these languages will become extinct, in the sense of having no native speakers, within

Grooming, Gossip and the Evolution of Language

Babel's Legacy

lingua franca of trade, government and science, as well as the

dispersal, he suggests, include Khoisan (the language of the Southern African Bushman and their relatives), Basque ( arguably descendants of the original inhabitants of Europe - heaven help us if they demand their rightful tribal lands back ! ) , Caucasian ( the languages spoken in the region between the Caspian and Black Seas ) , the Indo-Pacific languages of New Guinea, and the Austric language group (the ancestral languages still spoken by hill tribes in parts of Vietnam, Cambodia and Thailand) . The second great event occurred about 1 0,000 years ago, follow­ ing the more or less simultaneous development of farming at vari­ ous sites in the New and Old Worlds. The discovery that certain plants could be farmed at will freed the peoples concerned from the inexorable need to follow the migrating game and the fruiting cycles of the woodland trees. Farming allowed groups to stay longer in the same place; moreover, the surpluses produced by farming allowed their populations to grow faster. As these peoples spread outwards from the regions of initial settlement in search of new land for their burgeoning families, they displaced (or probably less commonly absorbed) the hunter-gatherer communities they encountered. We have ourselves witnessed analogous cases of physical displacement in recent historical times: the migrations of the Saxon peoples into western Europe during the fifth and sixth centuries AD, when they gradually pushed the Celts into the peripheral hill regions in the north and west of the British Isles, and the nineteenth-century European migrations to North America and Australia. The most important of these movements were associated with grain farming in the Near East - which gave rise to the Afro-Asiatic language group that spread south-westwards into Arabia and North Africa, and to the Indo-European group that spread west into Europe - and with rice farming in the Far East, which produced the Sino­ Tibetan languages of southern China and, indirectly, the Austro­ nesian languages of the Pacific rim, including offshoots like the lan­ guages spoken as far apart as New Zealand and Madagascar. 2

national language of countries on every continent; yet, at the same time, it has diversified to produce local dialects that border on the mutually incomprehensible. Most linguists now recognize Pisin (the 'pidgin English' of New Guinea), Black English Vernacular (B EV, the form of English spoken by blacks in the major cities of the US), Caribbean creoles (the English of the various Caribbean islands) and Krio (the creole of Sierra Leone in West Africa) and even Scots (the English spoken in the Scottish lowlands) as distinct languages. A thousand years from now, some historical linguist may trace their origins back to an obscure island off the north-west coast of Eu r ope, and wonder how it came to be that the language of one unimportant member of the vast Indo-European language family came to oust all the others.

In fact, this process of dialectization is not unique to human languages. We now know that the 'languages' of other species also have dialects. Crows in eastern Europe pronounce their caws in a noticeably different way from those in western Europe, while the Japanese macaques in the northern parts of their range pro­ nounce their coo contact calls differently from those in d i e south. Of course, the scale of this is quite limited by comparison with the extent and rapidity of the dialect evolution in human languages, but the pattern is closely analogous. This flexibility is so striking and so universal that it cannot be a simple accident of evolution: it must have a purpose. A clue to this may lie in the history of lan­ guage evolution itself. The archaeologist Colin Renfrew has argued that the modern language groups evolved as a result of four major migration events in human history. The language groups of Australia and the Amerind languages of the New World owe their origins to the initial migration that saw modern humans spilling out of Africa around 1 00,000 years ago. The leading edge of this dispersal moved gradually eastwards until a bout 40 ,000 years ago it more or less simultaneously slipped across the Bering Straight into North America and across the Arafura Sea that separates Australia from the islands of the Sunda Shelf which j uts out below the Indo-Chinese peninsula. Other remnants of that same early


Despite being off the south-east coast of A frica, the i sland of Madagascar was i n

fact first settled by people from t h e Pacific rim a bout


years ago. Their l an ­

guage, Malagasy, i s closely related to s o m e of those spoken today i n Borneo, a n d they probably reached t h e then - u n i nhabited Madagascar as a result of e a r l y trading expeditions.

- 1 59 -

Grooming, Gossip and the Evolution of Language

Babe/'s Legacy

The third series of migrations occurred around 8000 years ago. The global warming that spelled the end of the last Ice Age opened up the Arctic regions to inward migration by a group of people who proceeded to disperse westwards into northern Scandinavia, where they are now represented by the reindeer­ herding Lapps, and eastwards across the Bering Strait into the Canadian Arctic, where they displaced the native Amerindians southwards into what is now the USA. Clearly, the latter migra­ tion must have taken place in at least two major waves: an earlier group represented by the Na-Dene language group of ( mainly) Canadian Indian peoples, and considerably later the Eskimo (or more correctly Innuit) peoples who now occupy the Arctic regions from Alaska to Greenland. Finally, within historical times a number of major migrations gave rise to what Renfrew terms the 'elite dominance' process. The development of complex societies, often associated with the use of the horse as a means of travel and warfare, led to rapid movements of Indo-European peoples eastwards into the Middle East and northern India, and of the Altaic-speaking peoples of Mongolia eastwards from central Asia into northern China, Siberia and the Japanese archipelago. This latter group were responsible for a second major expansion, this time westwards as far as eastern Europe, during the twelfth century AD as a result of the empire-building activities of the much-feared Mongol leader Genghis Khan. In most of these instances, the invaders enslaved or absorbed the native populations of the regions they took over rather than driving them out. As a result, the invaders imposed their language on their vanquished subjects ( much as the Spanish and Portuguese imposed their languages on the Indians of South and Central America) . What this seems t o suggest i s that, a t least until recently, i t has been the wholesale migration of peoples that has been largely responsible for the spread and diversification of languages. Elite dominance is a recent phenomenon, reflecting developments in transport and technology; the wholesale replacement of one group of people by another, bringing their own language and culture with them, seems more often to have the been norm. This sug-

gestion finds dramatic support from an unexpected direction. During the late 1 9 80s the a bility to sequence the base pairs on segments of DNA made it possible to carry out detailed compar­ isons of the genetic codes of different species. In a startling break­ through, Rachel Cann, the late John Wilson and their colleagues at Stanford University in California mapped out the sequence of base pairs on a segment of the mitochondrial DNA of 1 20 or so women who had given birth in local hospitals. The important thing a bout mitochondria 3 is that they are passed down only through the female line, so a pedigree constructed from their simi­ larities is a direct reflection of the mother-daughter descent lin­ eages involved. By comparing the sequences from different indi­ viduals they were able to reconstruct the genetic relationships between those individuals, and hence between the many different racial groups represented among the mothers in the sample. These a nalyses revealed that the range of variation in mitochon­ drial DNA is much greater in Africa than anywhere else in the world. Everyone from Europe, Asia, Australia and the Americas, plus some north African peoples, seems to be part of a single, closely related family group, which itself is a subset of the broad family of African peoples. By working backwards through the hierarchy of relationships, Cann and Wilson arrived at a common female ancestor for the women, who was inevitably named the 'African Eve'. Finally, by determining the number of mutations along each of the descent lines, they were then able to use the nat­ ural rate of mutation for mitochondrial genes, the so-called 'mole­ c ular clock', ( see page 3 3 ) to estimate that this common female ancestor probably lived between 1 5 0,000 and 200,000 years ago. Although there has been some dispute about how the reconstructions were done and about the methods used to deter­ mine when the ancestral Eve lived, the original hypothesis has largely been vindicated by subsequent analyses based on larger samples of women. More importantly perhaps, it is in good agree-

- 1 60 -

3. Mitochondria are the tiny power-houses within each cell that provide its energy.

It is now bel ieved that they were originally viruses that invaded the cell s of some very ancient single-cel led organisms and stayed on to develop a symbiotic relation­ ship with the cell's own nuclear DNA.

- 161 -

Babel's Legacy

Grooming, Gossip and the Evolution of Language ment with the fossil record : the only fossils that could be a ncestral to modern humans come from Africa during the period between 2 5 0,000 and 1 5 0, 000 years ago. Strictly speaking, these ana lyses do not identify a single com­ mon ancestor: they merely tell us that the mitochondrial DNA of all living humans - males get theirs from their mothers too derives from a very small number of females living at a particular time. Other calculations suggest that the ancestress(es) of all mod­ ern humans were mem bers of a population that totalled only a bout 5 000 individuals of both sexes and all ages. These would not have been the only members of their species ( our species) a live at the time, but they were the only ones whose lineage ( s ) survived to pass on their mitochondrial DNA to us. The real surprise came when the geneticist Luigi Cavalli-Sforza showed that if you plotted the tree of language groups on top of Cann's tree of the genetic relationships between human racial groups, the fit was surprisingly good. The dispersal and diver­ gence of the main language groups appeared to mirror the disper­ sal and divergence of the racial groups concerned. This suggests that when people migrated, they took both their genes and their languages with them, replacing wholesale the other human popu­ lations that they came across. Thereafter, as their genetic make-up diverged through the accumulation of mutations, so the ir lan­ guages diverged by a process of dialectical drift. And this brings us back to the question of why human lan­ guages develop dialects so readily. The changes in pronunciation over time documented by the brothers Grimm and other historical linguists are one of the mechanisms that have produced the extra­ ordinary diversity of languages we have today. It raises the funda­ mental question of why dialects evolve. !v1 y Brother and Me

It is widely recognized that dialects are intimately related to local subcultures. Speaking a particular dialect is a badge of group membership. It shows we belong. But why should we be so con ­ cerned to establish our group membership ? And why should - 1 62 -

dialects be so effective in this respect? The answer to the first question has to do with the Enquist-Leimar free-rider problem. But the answer to the second question - why dialects are an effec­ tive solution to that problem - depends on an important feature of evolutionary biology we have not so far encountered: the theory of kin selection. Mating and reproduction are obviously important from an evo­ luti ? nary point of view. They are, after all, the way you pass your vanous ge� es on to the next generation. But they are not the only way of domg so. In the mid- 1 9 6os the New Zealand entomologist Bill Hamilton ( then a young postgraduate student at Imperial College London, but now a professor at Oxford University) point­ ed out that an individual can ensure that the genes it carries are tra � smi tted to the next generation in either of two ways: by repro­ . ducmg itself, or by helping a relative who carries the same gene to reproduce more successfully. Provided the cost of helping a relative to reproduce (measured n term � of the lost opportunities to reproduce that the helper mcurs) IS less than the gain to the relative (when devalued by the degree of relationship between them), then it will pay an individ­ ual to behave in a helpful way. This principle is known to evolu­ tionary biologists as 'Hamilton 's Rule': it specifies the conditions under which helping behaviour can be expected to evolve within a population of animals. The mechanism is known as kin selection' it is not the only Darwinian mechanism for the spread of altruisti behaviour, but it does have important implications for the behav­ iour of organisms in general. The key consideration in Hamilton's Rule is whether the two individuals share a particular gene. The more closely related they are, the more l ikely it is that they will have inherited the same gene from a common a ncestor, and hence the more worthwhile it is for the one individual to help the other. The main point here is that, all things being equal, it is more beneficial to help a close rel­ ative than a distant one, because close relatives are more likely to share a given gene than distant relatives or, worse still, unrelated individuals. The importance of this for our story is that most higher organ-

- 1 63 -

Grooming, Gossip and the Evolution of Language

Babel's Legacy

isms, including humans, show a strong preference for relatives. In general, humans prefer to live near relatives rather than non-rela­ tives - although there obviously has to be some negotiation over which set of relatives a married couple lives near. It is relatively rare in pre-modern societies, for example, for newly married cou­ ples to live apart from both sets of relatives. Even in modern post­ industrial societies, where market forces mean that many people have to move away from their parental homes to find j obs, signifi­ cant numbers continue to live near their relatives if they can, and those who have moved maintain contact with relatives for longer than they do with unrelated friends. Moreover, people are decided­ ly more likely to help relatives than less closely related individuals. 'Blood is thicker than water' is a sentiment repeated in almost every human culture. As the Arab proverb so graphically reminds us, 'Me against my brother; me and my brother against my cousin; me and my brother and my cousin against [our common enemy] ' . There is a great deal of evidence t o show that humans take relatedness into account in their dealings with other humans, especially when the costs of behaving altruistically are high. This is not to say that humans are never altruistic towards other humans. They quite often are, but normally only when the costs of doing so are minimal. People like Sydney Carton, from Dickens' stirring story of selflessness A Tale of Two Cities, are comparatively rare in real life. Were altruistic behaviour towards all and sundry truly 'natural', we would not have to be exhorted to act altruistically as often as we are; such behaviour would sim­ ply be taken for granted. And we would all willingly pay our tax­ es in full and on time. In order to study Hamilton's Rule at work in humans, Henry Plotkin, Jean-Marie Richards, George Fieldman and I carried out a simple experiment that had originally been suggested some years earlier by the biologist David McFarland. Subjects were asked to perform an isometric skiing exercise that involves sitting against a wall with nothing to support them . The thighs are thus parallel to the floor, while the shins and back are at right angles to them. Initially, it's a pleasant enough position to take up. But the leg muscles are under intense stress, and after about a minute the

position becomes increasingly painful. Most people can only hold it for two minutes or so before collapsing on the floor. We offered subjects 7 5 pence for every 20 seconds they could maintain the position. The catch was that the money they earned was on most occasions given directly to someone else. The identi­ ty of the recipient was chosen at the start of the exercise, and so was known to the subject. Each subject performed the exercise six times, each time with a different recipient nominated for the rewards o f his or her efforts. The subjects themselves were always one of those recipients, and a maj or children's charity was always a nother. The other four recipients were individuals of specified degrees of relatedness - a parent or sibling (related by a half), an aunt or uncle (related by a q uarter) , a cousin (related by one eighth) , plus a best friend of the same sex ( relatedness zero ) . The results of the experiment were striking: subjects worked much harder for close relatives (parents, siblings ) and themselves than they did for less closely related individuals or the charity. This is of course a very simple situation. But it does capture the essence of real altruism - bearing a cost to benefit someone else and is probably comparable to many small acts of altruism in which we engage daily, such as lending small quantities of money or giving time to help others. In another study, Amanda Clark, Nicola Hurst and I looked at thirteenth-century Viking sagas. The histories of the Viking com­ munities in Iceland and Scotland tell of feuds and vendettas that rumbled on for decade after decade. We found that the Vikings were significantly less willing to murder close relatives than more distant relatives, relative to the n um bers of individuals available in each category in the population. Only if the stakes were very high, something like the inheritance of a title or a farm, were they pre­ pared to murder close relatives. The relatives of the victim were also more likely to insist on their right to a revenge murder for a close relative than for a distant relative ( where blood money was often accepted ) - unless the murderer was a particularly violent man, which obviously raised the risks of attempting a revenge killing. These tendencies are reflected even in modern homicide statistics, as Martin D aly and Margot Wilson have pointed out. - I 65 -

Grooming, Gossip and the Evolution of Language

Babel's Legacy

Canadian and US murder statistics show that people are twenty times more likely to murder an unrelated person they live with than a genetic relative. More starkly still, the Canadian data show that children are sixty times more likely to die before the age of two years at the hands of a step-parent than at the hands of a biological parent. This doesn't mean that every step-parent is an ogre; to put all this into perspective, we are only talking about a murder rate of around 600 per hundred thousand children born. But it does mean that the risks step-children run are considerably higher than those run by children living with both biological parents. Something holds us back from uncontrollably venting our frustra­ tions on other people when the source of those frustrations is genetically related to us. The infl uence of kinship emerges in other aspects of social life too. When the Vikings made alliances, for example, those formed with relatives were more likely to be stable than those between unrelated people, and they were more likely to be entered into vol­ untarily without demanding a price in return. Unrelated allies often demanded a favour or property as a precondition, whereas relatives would willingly lend a boat for an expedition or take part in a revenge killing out of a sense of obligation alone. This willingness to help relatives can even be seen in contemporary populations. Catherine Panter-Brick, of the University of D urham, found that among Nepalese hill-farmers the women would will­ ingly help relatives harvest their fields without obligation, but would expect ( and demand) a strict reciprocation of help when it came to helping unrelated members of the community. This emphasis on living with and helping kin is not, I think, sim­ ply a reflection of the desire to help relatives reproduce more effec­ tively, in the way that sterile worker bees selflessly help the queen to produce more and more sisters for them. Rather, it seems to be a consequence of the fact that relatives are more likely to co-operate with you in the formation of alliances, as well as in all the minor daily opportunities to help out, because they have a stake in your ability to reproduce successfully. When you need help, you are more likely to get it from a relative than an unrelated individual.

O ur kin seem to be of such overriding importance to us that we use the language of kinship to reinforce a sense of group identity even when the other people concerned are not actually related to us. When we do this, it is invariably because those people are valuable allies in some respect. We often refer to co-religionists as brothers and sisters, for example, particularly when we are mem­ bers of a beleaguered minority. The New Testament letters of the apostles are full of terms of endearment. 'To Timothy, my dearly beloved son', wrote Paul to his completely unrelated colleague; and he began his letter to the Colossians, 'Paul, an apostle of Jesus Christ by the will of God, and Timotheus our brother. ' The apos­ tle John opens his First Epistle with the words, 'My little children, these things I write unto you, that ye sin not.' Christianity in par­ ticular is infused with this sense of family membership. The most important prayer in the Christian canon opens with the words ' Our Father, which art in Heaven', while on the more earthly level the honorific title 'Father' is given to priests of the Roman Catholic and Orthodox churches. Interestingly, we resort to the same tactic when rousing our countrymen to the defence of the realm. Terms of kinship and appeals to the protection of our nearest and dearest are so com­ mon in this context we barely notice them: 'The Fatherland needs you ! ' ; 'Come to Mother Russia's defence ! ' ; 'They will rape your daughters and sisters ! ' In order to persuade others to take part willingly, we go to considerable lengths to persuade them that we have kinship ties, even though those ties are at best nebulous. One problem humans face with their large, diffuse groups is the ease with which free-riders can cheat the system by claiming to be a relative in order to beg a favour. One solution might be to have your DNA fingerprint stamped on your forehead, much as the character Rimmer in the television sci-fi comedy Red Dwarf has an H on his forehead to indicate his status as a hologram. Needless to say, this would be rather difficult to arrange; although it's worth noting that Hindu caste marks are j ust such a sign - and caste membership is, of course, inherited. One way to be sure that someone you encounter really is a relative is to require them to exhibit a token of group membership, some kind of badge. It is, of

- 1 66 -

- 1 67 -

Grooming, Gossip and the Evolution of Language

Babel's Legacy

course, only too easy to cheat with most badges. To be effective, badges either h ave to be costly to possess ( the equivalent of a very expensive motor car) or difficult to acquire ( because they require years of practice from a very early age ) . I n fact, most animals rely on familiarity. Statistically speaking, most individuals with whom you grow up are likely to be rela­ tives. While there will o bviously be mistakes from time to time, errors of recognition will only be serious enough to undermine Hamilton's Rule if they become very frequent. The evolutionary process can tolerate a surprisingly high degree of error because it is concerned not with a bsolute values but with relative benefits. D ialect is an obvious badge, because language is learned at a critical period early in life. Someone who speaks in the same way as you do, using similar words with the same accent, almost certainly grew up near you, and at least in the context of pre­ industrial societies, is likely to be a relative. It's not a 1 00 per cent guarantee, of course, but it's a lot better than simply guessing. But dialect has another a dvantage: it can change relatively quickly, at least on the scale of generations. This makes it possible to keep track of the movement patterns of individuals over time . A group that emigrates will evolve its own accent and style of speech over a generation or so, even though it uses the same words. Consider how different the Australian and British accents are now, despite the fact that most of the emigrants to Australia went there within the last hundred years. The obvious suggestion, then, is that dia lects were an adaptation to cope with the problem of free-riders . By constantly developing new speech styles, new ways of saying the s ame old things, a group ensures that it can easily identify its members. And it does so with a badge that is difficult to cheat because you have to learn it early in life . It is not easy to pick up a group's accent or style of speech without living in the group for a prolonged period of time. To test this idea, Dan Nettle developed a computer model that a llowed different strategies to compete against each other in a 'world' where they could only reproduce once they had acquired a certain level of resources ( in effect, eating enough food to do more than just keep body and soul together ) . Individuals were a ble to

co-operate in order to acquire the resources in question, and often had to do so in order to obtain enough to reproduce. Some indi­ viduals were co-operators, others were cheats who obtained help but then refused to pay it back. Some of the co-operators only helped individuals who had similar dialects to themselves. But there was a brand of free-riders who quickly learned to imitate the dialects they came into contact with. The model showed that so long as dialects remained constant through time, the imitator­ cheat strategy was a very successful one and thrived at the expense of the others . However, if dialects changed moderately quickly ( on the scale of generations ) , cheating strategies found it almost impossible to gain a foothold in a population of co-operators provided at least that individuals can remember whom they have p ayed against in the past. A dialect that continues to evolve pro­ VIdes a secure defence against marauding free-riders. It seems likely, then, that dialects arose as an attempt to control the depredations of those who would exploit people's natural co­ operativeness. We know instantly who is one of us and who is not the moment they open their mouths. How many times in history, for example, have the underdogs in a conflict been caught out by their ina bility to pronounce words the right way. As chapter 1 2 of the Book of Judges tells us:

- 1 68 -

And the Gi leadites took the passages of Jordan before the Ephraimites: and it was so, that when those Ephraimites which were escaped said, ' Let me go over'; that the men of Gilead said unto him, 'Art thou an Ephraimite ? ' If he said, 'Nay'; Then said they unto him, 'Say now Shibboleth' : and he said 'Sibboleth' for he could not frame to pronounce it right. Then they took him and slew him at the passages of Jordan : and there fell at that time of the Ephraim ites forty and two thousand. In another famous incident, in the courtyard outside the high priest Caiaphas's house in Jerusalem, the disciple Peter was soon identi­ fied as an outsider, a Galilean. In the words of the evangelist Mark: And a little after, they that stood by said again to Peter, ' Surely thou art one of them r a follower of Jesus, who had j ust been - 1 69 -

Grooming, Gossip and the Evolution of Language arrested] : for thou art a Galilean, and thy speech agreeth there­ to. But he began to curse and to swear, saying, 'I know not this man of whom ye speak.' Nor are these examples confined to the long-distant past. In the Netherlands at the end of the Second World War, not a few German soldiers trying to escape in the melee of fleeing civilians came badly unstuck when challenged by Dutch resistance fighters to pronounce complicated Dutch place-names. Language, then, is very much a social tool. Not only does it allow us to exchange information relevant to our ability to survive in a complex, constantly changing social world, but it also allows l!S to mark other individuals as friend or foe. And herein probably lie the origins of the historical language evolution we discussed in Chapter 7 . Languages diversify initially as local dialects, but even­ tually these become mutually incomprehensible because local groups need to maintain their identities in the face of competition from other groups. There is some evidence to suggest that, at least in West Africa, language diversity is higher (that is, there are more different languages per square mile, and individual languages have fewer native speakers) in the high-density populations near the Equator than among the low-density populations further to the north, where the proximity of neighbours is much less of a problem. If this result is confirmed (Dan Nettle is currently testing it), it suggests that the rate at which dialects evolve is not constant, but is directly related to population density. The higher the density of people, the faster their dialects change. The discovery of agricul­ ture a oare 1 0,000 years ago marks a watershed in human ecolo­ gy. It had a dramatic effect on population growth-rates, because it allowed people to live at very much higher densities than they had previously been able to do as mobile hunter-gatherers. That being so, it seems plausible to suggest that dialects might even have a rather recent origin. Prior to the agricultural revolution, people might well have spoken the same dialect over a very wide area, with dialect evolution being a slow process of change by gradual drift. Babel may not, in fact, have been so very long ago. - 1 70 -



The Little Rituals of Life

We often underestimate j ust how much of human language depends on an interpretation of the speaker's intentions. Without theory of mind (ToM) and the higher orders of intensionality ( see Chapter 5 ) , we would not be a ble to make more than the barest sense of what others say. Conversations would be factual and dull; they would have all the warmth and poetry of a conversation with Startrek 's Mr Spock. We would not have even the most rudi­ mentary literature; the best we could aspire to would be some rather dull narrative poetry. As it is, we use language daily to try to influence the lives of those around us, ultimately for our own benefit. And herein lies the great enigma of language: what we use for good we can as easily use for evil. With the benefits of Machiavellianism and deep ToM to aid us, we can use language to outwit and bamboozle, to lay propaganda trails to mislead, or to inveigle and caj ole. On the whole, I have shied away from exploring these slightly disreputable features of language, prefer­ ring instead to concentrate on the more general benefits that lan­ guage confers in terms of bonding groups. But the time has now come to explore these aspects of human behaviour in more detail. The Black Art of Propaganda

Free-riders are, as we have seen, an especially acute problem in the large dispersed groups like those typical of modern humans. Trying to prevent them gaining the upper hand becomes a critical problem if your survival in the rough and tumble of the real world depends on maintaining large cohesive groups. Enquist and - 171 -

Grooming, Gossip and the Evolution of Language

The Little R ituals of Life

Leimar have suggested that gossip may have evolved as a mecha­ nism for controlling the activities of free-riders. By exchanging information on their activities, humans are a ble to use language both to gain advanced warning of social cheats and to shame them into conforming to accepted social standards when they do misbehave. This is a powerful mechanism for deterring cheats, and Enquist and Leimar were able to show mathematically that free riders would be less successful in a community of gossiping co-operators. Perhaps language evolved not so much to keep track of your friends and acquainta nces as to keep track of free-riders and coerce them into conforming. There is, in fact, some experimental evidence to support this suggestion. Lida Cosmides, of the University of California, Santa Barbara, has suggested that the human mind contains a special module that is designed to detect people who cheat on social agreements. She used an old psychological test called the Wason Selection Task to demonstrate this. In the original Wason Task, subjects were presented with four cards marked with four symbols - say A, D, 3 and 6 . The subject is told that there is also a symbol on the back of each card; in addition, they are told that there is a general rule which states that a vowel card always has an even number on its obverse. Which card or cards should they turn over to check that the rule is true? The logically correct answer is that they should turn over the A card and the 3 card. The A card must have an even number on its reverse and the 3 card must not have a vowel. About three-quar­ ters of those tested , on this problem get it wrong ( roughly about the number you would expect if people chose the cards at ran­ dom ) . Most people choose either the A card or the A card plus the 6 card . But the rule they were given does not say that an even numbered card has to have a vowel on the other side, merely that vowels cards must have an even number on their obverse . An even-numbered card could have either a consonant or a vowel on its back without breaking the rule. Cosmides was a ble to show that if you give subj ects the same logical problem dressed up as a social contract, they generally get the right answer with no trouble. One of her social problems was

the under-age drinking rule. Instead of four cards, subjects are told there are four people sitting round a table; one is sixteen years old, one twenty, one is drinking Coca-Cola and one is drink­ ing beer. The social rule is that only those over eighteen can drink alcohol. Which individuals do you need to check to ensure that this rule is not being violated? The answer is trivially obvious: the sixteen-year-old ( because sixteen-year-olds are not allowed to drink alcohol) and the beer drinker ( because he or she must be over eighteen ) . Twenty-year-olds can drink whatever they like and anyone can drink co*ke. Nearly everyone gets this version of the problem right, despite the fact that they fail miserably on the abstract version of the same problem. Cosmides has argued that we have an in-built mental set designed to recognize social contract situations and to detect vio­ lators. Without this, human social groups would collapse into j ust the kind of black hole of self-interest that Enquist and Leimar identified. Since cooperativeness is essential to our survival (indeed, it might be regarded as the key human evolutionary strat­ egy) , we have to have mechanisms for policing observance of the rules we agree for the collective good (where the collective good really means: what's best for each individual in the long run ). This overriding concern with social cheats forces us to consider the fact that there are several ways in which language might work as a social device. I have tended, so far, to assume that language's social function is broadly the exchange of information on friends and acquaintances. But it may be that, as a device for ensuring the stability of large groups, language may in fact produce its bonding effects in a number of different ways. Allowing you to keep track of what your friends and allies are doing is one possibility. But another clearly lies in the exchange of information a bout free­ riders . A third is that language provides us with a device for influ­ encing what people think a bout us. The psychologist Nick Emler, for example, has argued that much of our daily use of language is in fact concerned with repu­ tation management. You can pass on information about yourself in order to influence your listeners' perceptions of you. You can tell them about your likes and dislikes, how you would behave ( or

- I 72 -

- I 73 -

Grooming, Gossip and the Evolution of Language

The Little Rituals of Life

how you think you ought to behave ) in different circ*mstances, what you believe in and how strongly you believe it, what you dis­ approve of, and so on. You can be deliberately rude or obse­ quiously nice; you can insult them or flatter them. It can a llow you to sort the sheep from the goats very quickly by driving away those whom you know you would never get on with or encourag­ ing those who might be of interest to stay and further their acquaintance with you. Or, of course, you can engage in black propaga nda, sowing the seeds of doubt about enemies in people's minds or praising a slightly dubious friend to the hilt so that he or

something that is only needed occasionally. To be able to admon­ ish some miscreant may be critical to the smooth running of a group, but it may not be necessary to do this more than once in a month of Sundays. That the machinery remains unused the rest of the time may be a cost worth bearing if the benefit is large enough. This would imply that all the wittering we do, all that social gossip, was j ust a matter of keeping the machinery of speech in trim, oiled and ready for the unpredictable moment when it suddenly becomes essential. It seems like a plausible suggestion, but the enormous cost of having all this machinery in place seems rather excessive. Evolution is not usually so wasteful of resources, never mind time. Remember, the brain uses a fifth of all the energy of the body, about ten times more than you would expect on the basis of its size alone. Moreover, there are much simpler solutions to the problem of free-riders that are much more cost effective . Why don't we simply resort to the tactic that seems to work so well for all the other monkeys and apes - giving miscreants a j olly good thum p ? It would seem to be a great deal less costly in terms of brain size. In short, however valuable the policing function may be, it's unlikely to have been the precipitating cause of the evolu­ tion of large brains or language. Indeed, it seems logically wrong: the free-riders problem is a consequence of the fact that we live in large groups, and living in large groups seems not to be possible without big brains and language. The suggestion that self-publicity may be an important consider­ ation remains a more serious possibility. In fact, it is clear from our analyses of people's conversations that we do sometimes exploit the opportunities for advertising that language offers. Two particu­ lar findings from our studies point strongly in this direction. One of the surprises of our study was that we found very few differences between males and females in the topics they talked about. Both sexes spent j ust as much time discussing personal relationships and experiences, and - contrary to popular myth both spent time discussing other people's relationships and behaviour. Nor were the men in our samples more likely to dis­ cuss politics or high art (or for that matter low art) than the

she gets the j ob . The fact that w e can identify a n umber of different advantages to language raises the question of whether any one of them was the key selection pressure for language evolution ( with the others simply being the metaphorica l icing on the evolutionary cake usefu l additional benefits, but not of sufficient magnitude of themselves to have driven language evolution on their own ) . An unequivocal answer would involve demonstrating that, if the oth­ er benefits were eliminated completely, language would survive only if, say, the policing function was left in place. With some­ thing as complex as language, it is often difficult to disentangle the various functions it can now serve. However, simply asking whether one function is more common than the others might at least provide us with a n indication of the likely answer. In an attempt to throw some light on the problem, Anna Marriott undertook a more detailed study for me of what people talk about. She discovered that criticism and negative gossip accounted for only about 5 per cent of conversation time, with a similar amount of time devoted to soliciting or giving advice on how to handle social situations. By far the most common topics of conversation were who-was-doing-what-with-whom and personal social experiences. About half of this was concerned with other people' s doings and about half with the activities of the speaker or immediate audience. This suggests that, whatever else may be going on, monitoring the activities of free-riders and social cheats may not be the primary use to which we put our linguistic abilities. It is, of course, possible that the primary function of language is - 1 74 -

- 175 -

The Little Rituals of Life

Grooming, Gossip and the Evolution of Language

tions for what we might think is going on when people converse with each other. The most plausible interpretation is that the women are engaged in networking, while the men are engaged in advertising. In terms of creating the right kind of environment for the suc­ cessfu rearing of offspring, networking is probably the single most Important activity that women engage in. The support net­ works that these provide make it possible for the reciprocal exchange of information on the processes of childbirth and child rearing, help with foraging and horticulture, psychological sup­ port during times of emotional crisis and a dozen other support . serVIces large and small. The male world, in contrast, is more directly competitive and much less co-operative. Directly or indirectly, much of its focus lies in mating or in the acquisition of the resources or status that will create future opportunities for mating. Advertising becomes a crucial factor in that process. In the intellectual world of the university, demonstrating your .mtellectual skills by showing off your knowledge of Kant or the Romantic poets, or by being a ble to explain yesterday's lecture on the second law of thermodynamics, may be quite acceptable hall­ marks of competence and status. They mark you out as a cut above the rest, the obvious choice in the mating stakes. In that kind of environment, intellectual prowess is as appropriate a crite­ rion of future status or earning power as being the best card play­ er in a bridge club or the best musician in a music club. Knowledge, as it is so often said, is power.

women were. There was, however, one striking difference: the proportion of time spent talking about work and academic mat­ ters or religion and ethics increased dramatically when males were in mixed sex groups. In each case, the proportion of total conver­ sation time devoted to these topics increased from around 0-5 per cent in all-male groups to 1 5-20 per cent in mixed sex groups, with males showing a much more dramatic change in this respect than females. Our interpretation of this result was that conversations often function as a kind of vocal lek. Leks are display areas where males gather to advertise their qualities as potential mates to the females. They occur widely among animals like antelope and birds, though usually only in species where the male does not ( or cannot) contribute to the business of rearing the young. The pea­ co*ck is a very typical lekking bird. The males defend small territo­ ries in an area that females frequent and display for all they are worth whenever any females come near them. The females wan­ der from one male to another, assessing the qualities on offer. Eventually, having settled on the best of what may well be a bad bunch, each female mates with the male of her choice and then heads off elsewhere to get on with the business of laying eggs and rearing chicks at her leisure. The suggestion that lekking may be an explanation for much of what passes as conversation in humans was given added weight by a second result from our conversation studies. We had found no differences between men and women in the amount of time devoted to talking about social topics: in both cases, around 6 5 per cent of speaking time was taken u p with talking about social experiences of one kind or another. However, there was one respect in which they did differ, and that was whose social experi­ ences they talked about most. At least among the younger group of subjects, the women tended to devote a bout two-thirds of their social topic time to other people's social experiences and activities (and about a third of the time to their own ) , whereas the men spent two-thirds talking about themselves ( and only a third talk­ ing a bout other people) . This difference between the two sexes has important implica-

It has been said that as much as two-thirds of the meaning in the sentences we utter is in fact conveyed in the nonverbal signals that accompany speech. Much is given away by the unspoken body language that we exhibit, sometimes deliberately, sometimes unin­ tentionally, and we seem to be very sensitive to these cues. I was particularly impressed by people's sensitivity to these aspects of the environment while carrying out a study of vigilance

- I 76 -

- 1 77 -


The Eyes Have It

Grooming, Gossip and the Evolution of Language behaviour. The study was concerned with the rates at which people checked the environment even while engaged in conversa­ tions. To do this, we had to choose a subject and record every glance out at their environment. Look-ups are often very s � bt e and to be sure of not missing any, we had to stare at the subJect s . eyes for periods of up to five minutes, counting each look-up as It . occurred. We soon discovered that people often noticed that they were being stared at, even from the far side of a large hall. People . constantly monitor what is going on around them, sometimes by rapid glances around the room, but often simply out of the corner of an eye. As a result, we had to a lter the way we collect� d our . data in order to avoid upsetting subjects or disruptmg theIr n a � ­ ural pattern of looking up, since being aware that someone IS watching you causes you to look up more frequently. . ' In fact, it is clear that we both use and monitor cues of thIS kmd constantly in everyday life. Eye contact, in particular, seems to be especially important as a sign of someone s honesty, as well a s their interest in us. In her song 'The ButterflIes Have Gone Away , , the country singer Helen D arling is concerned because er lov�r's ' lonely eyes don't follow me no more', recognizin� in t IS the first subtle sign of a dying love affair. One of the first sIgns of the . impending collapse of the relationship betwe n Pnnce Charles and Princess D iana picked up by the paparaZZI was the fact t at they neither touched nor looked each other in the eye dunng pub lic appearances. We set great store by eye contact . As the saying goes: never trust a man that cannot look y u in the eye . The use of eye contact is particularly important m the context ? f . initiating new relationships, especially for women. Contr? llmg SIt­ uations that could develop undesirable consequences IS one of many vitally important tasks. Studies of behaviour in singles b rs . reveal that in these contexts at least, women exert a surpnsmg amount of ontrol over the business of courtship and mating. With a few conspicuous exceptions ( and in the absence of a lot of a l o­ . hoI ) , males are surprisingly reluctant to pursue interactio�s WIth women unless they receive tacit encouragement with cues lIke eye contact. The two most important signals in this respect are strong steady eye-contact and the so-called 'coy' signal, in which eye con-

� �

- 178 -

The Little Rituals of Life tact is held for j ust a second, then followed by a rapid look away accompanied by a slight smile or blush (often with a second fleet­ ing look back out of the corner of the eye a moment later) . Reading the signals is, i n fact, a very ancient primate h abit. This was demonstrated rather nicely in a study of hamadryas baboons by the Swiss zoologists Hans Kummer and Christia n Bachmann. Unusually for baboons, male hamadryas hold small harems of females. Mating access to the females is j ealously guarded by the harem-h older and males do not normally bother to challenge a harem male's hegemony over his females. Indeed, when near a harem, potential rivals will go out of their way to avoid giving even the slightest suggestion that they might be interested in the females: they will sit peering intently out into the distance, fid­ dling with grass blades and generally avoiding all eye-contact with either the male or his females. Bachmann and Kummer found that, when the rival was very much more powerful than the harem male and knew from past experience that he could defeat the owner, he would sometimes attempt to take a female from the male. But he would only do so when the female was giving clear signals that she was not all that interested in her current male. The main cue seemed to be the alacrity with which the female followed the harem male when he moved and the frequency with which she glanced at him. Normally, hamadryas females will follow their males closely at heel. But her momentary hesitations, the dilitoriness that forces the male to stop and look back because she has not yet followed, are subtle cues that rival males read as showing that the female is less interested in her male than the conventional niceties of hamadryas society might normally require . The American prima­ tologist Barbara Smuts noticed a very similar phenomenon in the olive ba boons she studied in Kenya. Even though olive baboons are more promiscuous and informal in their relationships th an the hamadryas, the males seemed to rely on the same subtle cues to determine whether a female was really interested in her current male consort. But relieving a rival of his mate is not a simple matter. The mate in question needs to be persuaded that the change of male is worth- 1 79 -

The Little Rituals of Life

G rooming, Gossip and the E volution of Language while: when all else is equal, the devil you know is invariably better than the devil you don't. Prospective mates need to be impressed by a male's fitness to be their partner. Much of the business of mate selection, in humans as in other mammals, ends up with males advertising their wares and females choosing among those on offer. Finding a mate is, at root, a game of advertising. The American anthropologist Kristin Hawkes has argued that hunting is j ust such a form of 'showing off' in traditional hunter­ gatherer societies. She calculated the energy returns from hunting large game like antelope and concluded that it just isn't worth the energy and time expended on it. When the men do catch something, they promptly bring it back to camp and conspicuously share it with everyone else. In fact, they would do much better by setting out a dozen traps that they could visit every other day for five minutes. Yet the men insist on the importance of hunting and devote a great deal of time and effort to it, even though it makes no economic sense. Hawkes argues that our mistake is to think of hunting as an economic activity related to parental investment - the archetypal prehistoric man going out hunting to feed the wife and kids. In fact, she suggests, it is really part of the mating game . Hunting large mammals is a difficult and risky business, and great skill is needed to pull off a kill successfully. A hunting male - and in most hunter-gatherer societies, men commonly hunt on their own or in groups of two or three - inevitably exposes himself to risk of ambush by predators like lions, as well as such dangers as snakes and elephants. Among the Eskimo in the days before snowmobiles made hunting less dangerous, males suffered extraordinarily high mortality rates out on the winter ice : a male's life expecta ncy could be less than half that of a woman's in some of the more extreme ha bi tats of the Arctic north. In addition to these risks to his person, a hunting male has to exercise considera ble skill in tracking and stalking prey animals in order to have a chance o f a kill. As a demonstration of courage, stamina and skill, hunting provides infalli ble proof of how good your genes are. Hunting has all the hal lmarks of the challenges set to aspiring young knights in the chivalric tales of medieva l Europe. In these stories, the young knight is given a superhuman task to test his - 1 80 -

mettle - rescue a damsel in distress, wake the sleeping beauty, kill the dra�on that has been terrorizing the village, find the Holy . Grall, fIght the knight who has never been defeated, extract the sword from the stone. Such challenges are by no means confined to the folktales o f Europe. Young Maasai warriors volunteer to thro,:" a ,:"ay their spears and act a s bait for a cornered lion. By walkmg m towards it holding his shield before him, the warrior f?rce s the li? n to j ump him; his companions can then spear the . relatIve safety. By the time they h ave done this however hon m ' the l on 's hind claws - scra b ling for p urchase beneath the youn . man s shIeld - h ave done theIr best to disem bowel him. If he sur­ vives, he is feted as the hero of the village and becomes a much sought-after prize among the girls looking for husbands. In a rather spectacular example of the same phenomenon, the . young Captam Ewart Grogan walked the 4,50o-mile length of Africa from the Cape of Good Hope to Cairo in 1 899 to gain the hand of the woman he loved. Her family had dismissed him as a ne'er-do-well who would be unable to keep their daughter in the manner to which they thought she should be accustomed. Grogan banked on the fame (if not the fortune) that a dramatic adventure would bring him to persuade them to reconsider. They were duly impressed. In � ur own societies, young males take risks racing cars at exceSSIve speeds and play sports with a commitment and determi­ natio � that few women would consider worthwhile . Yet, despite showmg much less interest in performing these activities them­ selves, women continue to remain impressed by the performances of the males and compete with scarcely less vigour to a ssociate with the champions of the game . The late American basketball player Magic Johnson was no exception in being overwhelmed by the women who wanted to sleep with him. The fact that these activities are largely male mating displays may explain why, when women attempt them (walking Cape to Cairo or sailing single­ handed around the world, to take two recent examples ), it attracts a great deal less attention. What all these 'heroic tasks ' have in common is that they are difficult to cheat. They sort the men from the boys, those who are all mouth from those who gen uinely can do. As a demonstration


- 181 -

Grooming, Gossip and the Evolution of Language

The Little R ituals of Life

of fitness to be the father of your children and, perhaps more importantly, to provide for them in a capricious and uncertain world, it is difficult to imagine any more exacting test than hunt­ ing in the world of our ancestors. The cognitive scientist Geoff Miller has thrown another light on the question by suggesting that the evolution of the human brain was driven mainly by the demands for sexual advertising. The ability to entertain a prospective mate, to charm them with poetry and song, to make them laugh - these, he suggests, are the things that the human brain has been designed to do. And it is not j ust a matter of catching your catch, for he or she can at a ny time become more impressed by someone else. Like the Red Queen whom Alice encountered in Through the Looking Glass, you have to keep running j ust to stay still. Just as the hunter has to keep hunting to prove that he is still the best catch in the community, so the modern male has to keep his partner smiling. What makes this idea particularly appealing is a property of smiling and laughter that few people know a bout: they are both particularly good at stimulating the production of endogenous opiates. Both involve unusual muscle movements, while laughter, in particular, is surprisingly expensive in energetic terms. A bout of raucous laughter leaves us exhausted and gasping for breath. The staccato pumping of air through the windpipe requires a great deal of control and a lot of effort. Being morose and down­ at-mouth makes you unhappier. So the best recipe for happiness in life is to smile as much as possible - thanks to the surge of opiates flooding through your veins, it makes you feel warm and contented. By the same token, making a prospective mate laugh lulls them into a sense of narcotic security. Smiling and laughing have their own fascinating natural histo­ ry. American psychologist Bob Provine recorded the frequency with which speakers and listeners laughed during conversations. He found that women are more likely to smile and laugh than men, that they are more likely to do so when they are listeners than when they are speaking, and that they are more likely to laugh in response to male speakers than to female speakers. By comparison, men are much less likely to laugh at something a

The pervasive importance of mate choice and sexual selection in our lives is illustrated by another well documented but surprising

- r 82 -

. - r 83 -

woman says than they are at something a man says. These are interesting findings for several reasons. One implica­ tion is that women comics a re likely to be less successful than male comics because they will find it more difficult to raise a laugh from both male and female a udiences. Women comics may have to act i n a more exaggerated way than male comics do, breaking more strongly with conventional stereotypes for how men and women should behave. These sex differences in smiling and laughing responses h ave been interpreted as reflections of the way society is dominated by males: women smile and l a ugh at men more because smiling and laughing express subordination. These behaviours are presumed to be the human equivalent of animal appeasem*nt patterns like cowering with your tail between your legs. Certain forms of ' false' smile or laugh may indeed be forms of appeasem*nt in some circ*mstances. One study of doctors in a hospital, for example, showed that j unior doctors were much more likely to smile at their seniors than vice versa, and were much more likely to laugh at their superiors' j okes. But there are many different kinds of smiles and laughs and not all of them are appeasem*nt gestures. After all, we spend a great deal of time laughing and smiling at utterly helpless babies, as well as with our friends, without suffering from any sense of inferiority. A much more plausible explanation is that women smile at men to encourage them to take an i nterest. They are engaged in a constant game of assessment, comparing the current partner with other males that hove into view. Most of the time they are happy to settle for what they already have, but it's important to keep testing the waters (after all, no one is perfect and you never know when your current mate might desert you ) . Testing a male's a bility to make you laugh may be as good a way of covertly assessing his qualities as any. The Mati ng Game

Grooming, Gossip and the Evolution of Language

The Little Rituals of Life

observation: men and women differ rather strikingly in the way they learn accents. As they mature, boys tend to adopt their loc al regional working class accents, whereas girls tend to pick up a more neutral middle-class standard form of English ( known tech­ nically as Received Pronunciation, or RP for short ) . This curious fact has puzzled social scientists for years, since there is no obvi­ ous reason why the two sexes should differ in this way. The con­ ventional explanation is that there is more social pressure on daughters to 'speak nicely': girls a re more likely to be told they sound 'common' than boys are. It seems to be yet another exam­ ple of the double standard where the boys get away with murder while the girls pay the price. However, this is at best only half an explanation: it doesn't real­ ly tell us why these very different pressures should be exerted on the two sexes. So what i s going on here ? The answer is obvious once you appreciate that much of what we do (especially during the early adult years ) is related to the business of mating and mate choice. The difference between the sexes reflects crucial differ­ ences in their respective reproductive strategies. The main constraint on a female mammal 's a bility to reproduce is the resources she has available for rearing her offspring. Humans are no different, and women in almost all cultures show a striking preference for marrying males who are relatively wealthy (or high status, which comes to much the same thing as far as the opportunities of life are concerned ) . Jane Austen's nov­ els of social life during the opening years of the nineteenth century illustrate this rather nicely, showing the young middle-class women of the day holding out for the ideal catch. The sons of the local vicar are rarely the obj ects of their attentions, but the dash­ ing young army officers ( conventionally the finishing school for the nobility or a passport to wealth and success for the middle classes) and the sons of landed gentry were much sought after. Unfortunately, of course, there are never enough of them to go round, and the girls cannot afford to wait for ever for fear of being left 'on the shelf' . Eventually some of them have to settle for second best - luckily for the local vicar's sons. It is a surprise to most people to find that you still see much the

same pattern in modern western societies . We have carried out three studies of mate choice preferences in contemporary society ( one in the USA in colla boration with David Waynforth, and two in England ) . To do this, we analysed personal advertisem*nts in newspapers and magazines, because these provide a neat encapsu­ lation of what people would ideally like in a partner. As many as a quarter of the women mention cues of wealth and status - ' profes­ sional', 'home owner' , 'college-educated', 'independent means' as being desirable in a male partner, and 60-70 per cent of men mention these cues in writing a bout themselves, whereas women seldom mention them as descriptions of themselves and men sel­ dom ask for them in women. Given the fact that wealth and status are concentrated up the social ladder, it seems obvious that women will greatly improve their options in the marriage stakes by speaking in a way that will allow them to fit into the social world of the classes a bove them more easily. Hypergamy ( or mzrrying up the social scale) is wide­ spread in almost all human societies, and the behaviour patterns portrayed in Jane Austen 's novels extended far beyond the con­ fines of English county society. In rural Friesland in northwest Germany, detailed analyses of parish marriage registers from the last two centuries carried out by Eckart Vo land ( now at the University of Giessen ) and his colleague Claudia Engel show that women tended to marry up the social scale whenever they could; marriages into the social class a bove were m uch more common than marriages into the social class below. The wealthier petty farmers seem to have been highly sought after catches - and with good reason: the wealth they had to offer ( small though it was in absolute terms) was enough to ensure signif­ icantly higher survival rates for their children. Moreover, women who married up the social hierarchy typically did so at a younger age than those who married males from their own social class. Even though most girls would eventually be forced to marry into thei r own social class, it was always worth hanging on that little bit longer j ust in case a better catch came along. But they could not afford to wait for ever, lest they miss out on the marriage stakes altogether. Bear in mind that we are not talking a bout Jane Austen's

- r 84 -


-:- r 8 S -

Grooming, Gossip and the Evolution of Language

The Little Rituals of Life

upper classes here, but about an essentially peasant society. Marrying up the social (or economic ) scale is still common today. This doesn't mean that every working-class girl marries an upper-class boy, but it does mean that marrying into the class a bove is more common among women than it is among men, and it is more common among women than marrying down the social scale. The daughter of an earl marrying her local dustman attracts far more attention than does the earl's son and heir marrying the dustman's daughter. Given this, it pays girls to develop an all­ purpose accent that allows them to move more easily up the social scale when the opportunity arises - or at least it pays their parents to encourage them to do so. The boys, on the other hand, face a rather different problem. Middle- and upper-class boys are in demand because they hold out the best hopes of being able to provide an adequate resource base on which to raise children; as a result, they often have to try less hard in the mating game. Lower-class boys, on the other hand, h ave less to offer in this respect; and because they are much more dependent on their community networks, it becomes impor­ tant for them to ensure that they are seen to belong, that they have the right accents and dialects to mark them out as members of the group, where friends will provide them access to j obs or services that they would not otherwise be a ble to afford. Being poor with the wrong accent is the kiss of death, for it denies you access to the self-help network. This emphasis on wealth and status has a simple utilitarian pur­ pose. In society after society in the pre-industrial world, the single factor that most affects the mortality rates of children is the wealth of the husband. It doesn't seem to matter whether this comes in the form of land, cattle or money. A correlation between family resources and infant survival has been demonstrated among contemporary Kipsigis agro-pastoralists in Kenya by Monique Borgerhoff Mulder, in eighteenth- and nineteenth-centu­ ry German peasant farmers by Eckart Voland, and in South American Ache hunter-gatherers by Kim Hill and Hilly Kaplan. The American psychologist David Buss has carried out a series of studies of mate choice preferences in some thirty-seven cultures

drawn from all around the world: for the women, status and future earning potential were two of the most important criteria that they looked for in future husbands in virtually every culture. Having access to more wealth and resources enables you to feed your children better and, in modern economies, to create enough surplus to pay for more medical treatment and better education. However, there is evidence to suggest that women's demands may be changing: in our samples of US and UK Lonely Hearts advertisem*nts we found that around half the women were asking for family commitment in addition to or instead of wealth and status. This seems to be pointing to a shift in what women of reproductive age need for successful reproduction in modern economies. Where it was once resources, it is now increasingly the social input into the business of rearing children: help with child­ care, a contribution to the socializing of the children. The contrast with the consistency of the preference for wealth and status in numerous studies of traditional societies is so striking that it cannot be an accident. This change is of relatively recent origin ( and a quarter of the women in our samples were still emphasizin g wealth and status) . I t has come a bout, I think, because o f two key changes that have taken place in modern industrial economies this century. One is that dramaticall y improved hygiene and health services have almost obliterated childhood mortality, so making it more or less certain that every child you give birth to will survive to adulthood. The other is that the general level of wealth is much greater, so that the differences between the wealthiest man in the community and the average one are no longer great enough to make the dif­ ference between a life of poverty and one that is good enough to provide for your children's needs. This second factor is , of course , compounded by women's own greater economic opportunities: they are no longer so dependent on their husbands to provide all the income for the household. Men, it seems, have apparently not yet caught up with this shift in attitudes: the advertisem*nts clearly show that they are still busily hammering away at the old virtues of wealth and status. Of course, real wealth still carries a lot of weight - witness the appar-

- 1 86 -

- 1 87 -

Grooming, Gossip and the Evolution of Language

The Little Rituals of Life

ent ease with which millionaires of almost any age are able to attract beautiful young women. But the rest of us would probably do better to go for nappy-changing. Male behaviour will undoubt­ edly change, but it takes time; Eckart Voland and I have shown that, among rural German populations in the last two centuries, it took roughly a generation ( thirty years ) for rearing patterns to change in response to the precipitating economic factors that were driving change. A further social implication of women becoming less dependent on wealth and resources for successful rearing is that there will be less need for hypergamy, less pressure to try to marry up the social scale. And if this happens, we should see an increasing tendency for girls to acquire their regional or class-based accents and less preference for RP. Of course, these kinds of social change are predicated on the continued creation of sufficient wealth to allow all sectors of soci­ ety to benefit. If economic slumps prevent us from achieving full employment and a wider distribution of wealth, mate choice pat­ terns will return to the well-trod paths of yesteryear. Social change is driven by economic change. Sitting at the hub of all this frenetic activity is the evolutionary mechanism known as sexual selection. The idea of sexual selec­ tion was first discussed by Charles Darwin over 1 20 years ago. He pointed out that some features in the natural world appear to have no survival benefits of any kind; indeed, they are very often counter-productive if measured purely in terms of their impact on an animal's survival. The classic case he had in mind was the pea­ co*ck's tail: the male peafowl's long train makes its flight ungainly and awkward, and impedes its escape from pursuing predators. So why, asked Darwin, has the peaco*ck's tail evolved ? The answer, he suggested, is intense selection from peahens preferentially choosing males with the longest trains as mates. If the intensity of female choice is great enough, it will overpower the counter-selec­ tion being imposed by predators and other more mundane consid­ erations like the energetic costs of flying while towing the equiva­ lent of a barrage balloon. Sexual selection has turned out to be a much more potent force

in evolution than Darwin ever imagined. It may even be a more important force in the generation of new species than Darwin's original mechanism of environmentally driven natural selection. During the last thirty years a considerable amount of experimen­ tal and theoretical work has been done on this remarkable process, and we now know a great deal about it. The English biol­ ogist Marion Petrie, for example, has shown that peahens selec­ tively prefer peaco*cks with the largest number of eye-spots on their tails. These males gain more matings, fertilise more eggs and have more surviving chicks than males with fewer eye-spots. These findings from real life were later confirmed experimentally by removing eyespots from some males' trains or by adding eye­ spots to the trains of other males. Those males who lost eyespots gained fewer matings than they had previously achieved, and those who gained eyespots had more matings. In another study, the Swedish biologist Malte Andersson demonstrated the same effect by reducing and lengthening the long trailing tail feathers of male widow birds in Kenya. At least two mechanisms have been proposed to explain how this effect comes about. One is known as Zahavi's 'Handicap Principle' after the Israeli biologist Amotz Zahavi who first pro­ posed it. Zahavi argued that what the males are, in effect, doing is saying: 'Just look at me ! I'm s o good that I can afford to weigh myself down with all this baggage and still outfly predators ! tvlate with me if you want sons and daughters as good as me ! ' This is Kristin Hawkes's showing off in another guise. The other mechanism is known as Fisher's 'Sexy Sons Hypothesis' . The formidable English geneticist and statistician Ronald Fisher ( one of the architects of the modern neo-Darwinian theory of evolution ) suggested that female choice for quite arbi­ trary male characters might drive sexua l selection sufficiently intensely to produce the kinds of useless characters like peaco*cks' tails. The point is a very simple one: if females happen to take a shine to a particular trait, such as eyespots, then their daughters are likely to inherit the same predisposition. Since it would benefit females to produce sons that have the characters that females pre­ fer, it will pay them to mate preferentially with males with lots of

- 188 -

- 1 89 -


Grooming, Gossip and the Evolution of Language

The Little Rituals of Life

eyespots ( or whatever the character in question happens to be ) . This will lead to intense selection for males with many eyes pots, and the rapid evolution of this trait in the male population. In effect, Geoff Miller's poetic males hypothesis is a version of Fisher's Sexy Sons Hypothesis. Females who mate with males that carry these traits will produce sons that have these traits, who wi l then in their turn mate and produce lots of grandsons for theIr mothers. There is no intrinsic survival value to being a poet or a raconteur: it's simply a trait that females happen to h ave latched on to. Fisher's Sexy Sons Hypothesis is capable of producing very rapid evolution over relatively short periods of time. Whi � h, of course, is j ust what we see in the evolution of the superbrams of modern humans: for nearly a million and a half years, brain vol­ ume remained roughly constant at around 700-8 00 cc, then in the space of j ust half a million years it all but doubled in size. And this, Geoff Miller argues, was all a consequence of intense selec­ tion for skills to keep your mate entertained. There is another possibility, however, and it is suggested by the fact that smiling and laughing causes the brain to flood the body with endogenous opiates. Remember how, in Chapter 3, we saw that grooming is also very good at stimulating the production of endorphins ? Think of the implications for bonding if the way grooming works is to make you feel very relaxed and mildly euphoric in the company of friends . If the intensity of a . relation­ ship is related to the amount of grooming effort put in ( and hence to the quantities of opiates released ) , then our ancestors faced a serious problem when trying to push group sizes up beyond the levels observed in other primates. Leslie Aiello and I suggested that they initially did this by using vocalizations as a form of vocal grooming, so that they could keep 'grooming' with a friend even while b usy feeding at a distance from them - j ust as the gelada in fact do today. The problem here is that vocalizations are j ust, well , vocalizations. They don't h ave the same opiate-releasing properties as grooming. If opiates are a crucial part of the mechanism of bond­ ing, it is likely that vocal exchanges will only allow you to increase the size of the group beyond the maximum for primates

- 1 90 -

in general by a very limited amount. Sooner rather than later, you will hit an upper limit because the limited quantities of physical grooming cannot generate enough opiate production to maintain the reinforcement. However, suppose that as language develops, signals associated with langu age themselves begin to stimulate opiate production . Smiling and, particularly, laughing do j ust this, and this may well explain why smiling a nd laughing are such important components of conversation. They may well have begun as signals of submis­ sion during the earliest phases. There are facial signals that are structurally very similar to both smiles and laughter in chim­ panzees . But it seems that, at some point, they were captured and built into the business of bonding in social groups. We can now, quite literally, groom at a distance. Telling j okes allows us to stim­ ulate opiate production in our grooming partners even when we don't have the time to sit there doing it physically. We can get on with the other important activities of ecological survival - travel­ ling, hunting, gathering, preparing and eating food. Looking back over the ideas we have explored in this chapter, it seems to me that what we have is not necessarily alternative hypotheses for the evolution of language and large brains in mod­ ern humans, but rather valuable new components that were added into the system as it developed. Once large brains and a language capacity had evolved as a way of bonding large groups, it opened up windows of opportunity in new directions. Deception and advertising were now possible, where they had not been before. They must surely have reinforced the processes of selection acting on large brains and improved language skills; they may even have driven these beyond the levels which would have been possible on the basis of social bonding alone. But without the conventional forces of social bonding to underpin them, they would not have been sufficiently powerful to drive the evolution of brain size upwards as rapidly as it actually occurred.

- 191 -

The Scars of Evolution CHAPTER 1 0

The Scars of Evolution

maintain a large brain); others were cognitive (creating the brain modules needed to support ToM as well as the mechanical pro­ duction of speech). In this last cha pter, I want to explore some of the implications of these findings for the way we live. I have borrowed my title from that of a book by Elaine Morgan in which she describes the many bits of the human body that are hangovers from our evolu­ tionary past. Everything from our useless appendix to the weak backs we have through standing upright should remind us that evolution is not a process of inevitable perfection; in fact, evol u­ tion is a process of Heath Robinson make-do, a series of compro­ mises by which we attempt to do the best we can with a set of incompatible goa ls. We are imperfect creatures, stuck with our evolutionary heritage, far from the perfection of design that eigh­ teenth-century evolutionists interpreted as evidence for the handi­ work of God. The human mind is in no better shape than the human body. We are not quite Pleistocene minds trapped in space-age bodies, but there are some elements of our behaviour that reflect our evo­ lutionary past and it probably is true that, in some cases at least, our cultura l evolution has outstripped our ability to deal with the conseq uences . This chapter, then, rather than presenting facts that we know to be true, comes more by way of speculation on what might be the case. The story I have told clearly has implications for much of

Our j ourney has been a long and complicated one. We have cov­ ered five million or so years of evolutionary history; we have dipped into aspects of human biology as different as neurobiology and endocrinology on the one hand, and social psychology and anthropology on the other. Some of these will have struck a chord of familiarity. Others will have been new and surprising. So let me begin this final chapter by recapping the argument of the book. The central argument revolves around four key points: ( I ) among primates, social group size appears to be limited by the size of the species' neocortex; ( 2 ) the size of human social networks appears to be limited for similar reasons to a value of around 1 5 0; ( 3 ) the time devoted to social grooming by primates is directly related to group size because it plays a crucial role in bonding groups; and finally, ( 4 ) it is suggested that language evolved among humans to replace social grooming because the grooming time required by our large groups made impossible demands on our time. L anguage, I argue, evolved to fill the gap because it allows us to use the time we have available for social interaction more efficiently. Language fulfils this role in a number of different ways. It allows us to reach more individuals at the same time; it allows us to exchange information a bout our social world so that we can keep track of what's happening among the members of our social network (as wel l as keeping track of social cheats ); it allows us to engage in self-advertising in a way that monkeys and apes cannot; and, last but not least, it seems to allow us to produce the rein­ forcing effects of grooming ( opiate release) from a distance. For language to evolve, a number of key changes were required. Some of these were physiological ( freeing off the energy required to

Despite its extraordinary sophistication, human language is much more limited than we tend to give it credit for. Words fa i l us at crucial moments; we are unable to express the turmoil of inner thoughts that threaten to overwhelm us, so we resort to ancient modes of physical intimacy to express what we cannot or dare not say a loud. Of these limitations we are all painfully aware. But

- 192 -

- 193 -

what we do, but these implications have yet to be worked out in any detail. My aim is really to highlight some of the directions in which we might go. Small is Best

Grooming, Gossip and the Evolution of Language

The Scars of Evolution

there are other respects in which the machinery of speech imposes quite serious constraints on how we talk to each other that are perhaps less familiar to us, even though we cope with some of them daily. In Chapter 6, I pointed out that we are limited in the number of individuals whose attention we can hold in a conversation. Casual conversation groups are limited to around four people. This appears to be due to the fact that we cannot get more people into a circle small enough to hear what the speaker is saying. Two interesting features of our behaviour seem to arise from this. One is that conversation groups never contain more than one speaker at a time. When they do, no one can keep track of the conversations, and the group either breaks up into two separate conversations or one speaker tries to dominate the other - either by speaking more loudly or by actually demanding silence from the others . D uring our studies of conversational behaviour, it became apparent that there are striking sex differences in who does the talking in conversations that involve members of both the sexes . It has often been noted that, in mixed-sex groups, women tend to listen while men speak, and this has sometimes been interpreted as domineering behaviour on the part of men intent on forcing women into subservience. However, it is clear from our work that this explanation cannot be right (or at least not wholly right ) , for women are not more likely to listen on all occasions. In fact, the proportion of time that women spent speaking in male-female dyads ( or pairs) is exactly 5 0 per cent when the dyad is the entire group, but it declines as the size of the group within which the conversing dyad is embedded increases. In groups of eight to twelve, a woman engaged in a 'private' conversation with a man is likely to be talking for only 25 per cent of the time. There are two likely explanations for this. One is that, because women's voices are lighter than men's, they find it more difficult to be heard when the hubbub of conversation around them increases as the group they are in gets larger. When your efforts at conversation too often result in 'I'm sorry, I didn't catch that, could you say it again ? ', it's easier to sit back and listen. Since

men's deeper voices carry further, it is inevitably the men that get left to do the talking. There is, however, an a lternative possibility. Since much of con­ versational behaviour among young adults has all the hallmarks of a mating lek (a display area on which males advertise their qualities so that females can choose among them ) , it stands to rea­ son that when there are lots of men in the group, women should prefer to sit back and assess the bids on offer. You cannot assess the competition if you spend all the time talking - in fact, since speaking is a complex business, you probably won't have time to assess a nythi ng other than your own performance. In contrast, when a conversational dyad is on its own (rather than being embedded in a larger group) , it is usually there for a very good reason: things have advanced from pure sales pitch to an attempt to build a relationship. A second respect in which the mechanisms of speech impose con­ straints on our conversational behaviour concerns the way we han­ dle groups that are larger than normal. In order to prevent Babel breaking out in committees or lecture halls, we have to impuse very strict social rules on how people behave in these situations. In ser­ mons or lectures, the bulk of the people present have to agree to suspend their right to speak in favour of one particular individual. This agreement is very fragile: if irate members of the audience real­ ly want to, they can easily prevent a speaker from continuing. In extreme cases, it may become necessary to evict troublesome mem­ bers of the audience if business is to proceed at all. These negotiated arrangements override the natural patterns of human behaviour in order to permit some broader collective advantage to emerge a bove the disruptive consequences of imme­ diate self-interest. In effect, they are instances of what biologists call reciprocal altruism: this form of 'I'll scratch your back if you scratch mine' arrangement is another of the biological mecha­ nisms by which altruistic behaviour can evolve in a D arwinian world. But, like all such forms of cooperation, it is open to inva­ sion by cheats who may gain the benefits of cooperation without paying the price. By capitalizing on everyone else's willingness to adhere to the rules and remain silent, free-riders ( for that is j ust

- 1 94 -

- 195 -

Grooming, Gossip and the Evolution of Language

The Scars of Evolution

what they are ) can ensure that they get heard by forcing them­ selves on the assembled company, shouting louder than anyone else or intervening more vigorously. Formal arrangements in which the a udience suspends its right to speak are clearly necessary to allow certain kinds of important social functions to occur at all. Among the many such functions that would otherwise become impossible in traditional societies are religious instruction, rabble-rousing, proceedings in courts of law and the more formal kinds of political decision-making. Even things as simple as negotiating the arrangements for a marriage would become impossible if everyone present at these events insisted on speaking at the same time. We encounter this problem in a particularly acute form in com­ mittees, where everyone present expects to be a ble to make a con­ tribution. Committees require a n effective chairman to exercise control over the members, allowing each to speak in turn and pre­ venting unnecessary interruptions . As anyone who has sat in com­ mittee meetings knows only too well, the chairman's control of the meeting is fundamental. The moment the chairman's attention is distracted, the committee very quickly breaks up into a set of small conversations, often a bout a nything except the business in hand ( and as often as not, comprising gossip a bout mutual acquaintances) . The difficulty o f controlling conversational interactions has been largely responsible for an important informal rule in com­ mittee formation. It is a well-established principle that if you want business done and real decisions made quickly, then you should have a committee of no more than six members; if you want a committee to do some brainstorming in order to generate some new ideas, then you need more than six people in your committee. The two functions seem to be wholly incompatible . The bigger the committee, the longer it will take to come to any conclusions. Too many people will want to have a say - there will be too many irs, buts and howevers - and numerically powerful factions with opposing views may emerge. A small committee, on the other hand, l acks a sufficiently l arge range of opinions to gen­ erate new ideas: it is more likely to come to a decision because

everyone has had a chance to have their say and there is nothing more to add. And if two radically opposed views do emerge, there won't be enough people to create sizable factions in support of both; consequently, it is easier to isolate someone who proposes an a lternative view. Without the benefit of moral support, the odd man out is likely to accept the maj ority decision. This might, indirectly, give us an explanation for the peculiar­ ly paternalistic behaviour of the Victorians. Thanks to the suc­ cess of their primary health programmes and the rapid develop­ ments in medical science, the Victorian middle and upper classes achieved previously unheard-of levels of infant survival. Family sizes increased from the 2-4 surviving offspring characteristic of most traditional peasant societies to the 4-8 typical of the bet­ ter-off classes during the nineteenth century. With something like six children and two parents at the dinner table, plus the odd maiden a unt, the noise must have been deafening. It seems to me no surprise at all that the Victorians should have taken the view that children 'should be seen and not heard', and that they should speak only when spoken to. It was surely the only possi­ ble response in the face of what would otherwise have been bed­ lam. In contrast, with our 2.4 children and two adults, we can anticipate a more genteel discussion at the dinner table, and so treat children in a more liberal fashion. An obvious implication is that the a uthoritarian paternalism of the Victorians was a n inevitable extension of domestic attitudes into the wider world of adult life . One final example o f t h e way our mental machinery seems to limit what we do concerns our attempts to create virtual confer­ encing systems. The technology for setting up conferencing calls in which several people are hooked up together through the tele­ phone system has been available for some time. The telephone chat-line is one derivative of that technology. As a spin-off from that, much e ffort is now being put into building the technology for virtual conferencing systems, using video links to allow peo­ ple in different parts of the world to work together on the same document or to discuss policy issues a ffecting a multinational company. This is obviously a much cheaper ( and less exhaust- 197 -

Grooming, Gossip and the Evolution of Language

The Scars of Evolution

ing) solution to the management problems of multinationals than flying people half-way round the globe for a two-hour meeting. Unfortunately, it seems that the same kind of restrictions are likely to apply: it will be difficult to get more than a bout four people to interact successfully in these systems. The technology can handle an almost unlimited number of people, but the people cannot. Once there are more than four of them interacting, some­ one always gets left out a nd their contribution to the discussion becomes progressively marginalized . It's as though our mental machinery for handling a conversa­ tion group has been set at the maximum number of people that we normally need to integrate into a n interacting whole. Even though electronics can make it seem a s though a dozen people are sitting next to each other, we j ust don't have the cognitive machinery necessary to keep more than a bout three of them in mind at the same time. This has serious practical implications in areas like education. Most governments are predisposed to promote the development of large classes in order to minimize the costs of publicly funded education. But there is a cost. With large classes, the usual teach­ ing strategy is a form of lecture, because the hubbub of voices cre­ ated by any other form of teaching is too disruptive. In a universi­ ty context, much teaching is designed to try to provoke discussion, to get students to learn how to argue a case, to think through issues as they go along, weighing the evidence for and against a lternative hypotheses or courses of action - but that is only possible with very small groups ( six students and a teacher is usually considered the maximum viable group size for this kind of teaching ) . If the size of the group increases significantly beyond this, the discussion often becomes dominated by j ust a few of those present; the rest drift quietly into the metaphorical corners of their minds and gain little, or may set up competing conversa­ tions of their own. What you can teach as well as how you teach is affected. In large classes, the exercise can become one of simply shovelling prepackaged information down open mouths. There is very little

else you can do. The direct personal involvement in the to-and-fro of argumentation - the very thing that actively fires the mind - is lost, because the teacher's attention span is limited. The child's naturally questioning mind is forced to remain silent. And with that, the quality of education slips downwards from training minds that can think for themselves to training technicians who know the right response to give in a particular situation ( but don't really know why it's the right one) . Education comes to be a process of rote-learning rules of behaviour.

I wonder how many of us can honestly claim that we have not, at some point in our lives, become hooked on one or other of the TV serials. Bad as they often are, very few people escape them. Aside from our intrinsic fascination with other people's doings, there is an interesting issue here about why this particular form of enter­ tainment should be so popular. One of the more peculiar features of modern urban life is the extent to which we a re locked into the tiny worl d of our own homes. Separated from relatives and with limited opportunities to create circles of friends, the modern city dweller is forced increas­ ingly to draw on the ready-made imaginative family of the soap opera for a social life and a sense of community. It is conspicuous that the largest audience for these programmes is found among housebound women, trapped at home by young children . Those with active social lives, by contrast, rarely have an interest in these kinds of programmes. No one has looked at this in any detail yet, but it won't surprise me if, when they do, they find that soap opera characters begin to fill the actual role of network members for those whose rea l social networks are kept well below the natural limit of 1 5 0 by their social or economic situation. Even TV newscasters and personali­ ties can come to fill this role: they become part of our social net­ works, half-real friends whom we feel we know not j ust because we see them so often but also because they actually speak to us as individuals when they present the news. In fact, many successful

- 1 98 -

- 199 -

Coronation Street Blues

Grooming, Gossip and the Evolution of Language

The Scars of Evolution

newscasters deliberately try to speak in a way that makes it seem as though they are speaking to the individual listener across the dining room table. In traditional peasant communities the world over, everyone lives in everyone else's pocket. They have to, of course, because houses are crammed together and walls are like paper. But more than that, people want to: the community is a genuine communi­ ty, a co-operative whose members share the same problems of day-to-day survival. They are also bound by ties of kinship, at the very least through one sex and often through both. Modern industria l conurbations often lack that sense of com­ munity because they have been created anew out of nothing: housing is built, and people trickle in from many different places to fill up the plots. They have no social ties, no common history to bind them. Their networks of friendship and kinship may be stretched far out beyond the confines of the housing estate, a problem that is exacerbated by the high rates of mobility that force people to move long distances in search of work. One important consequence is that social networks become fragmented. In traditional societies, both peasant and hunter­ gatherer, communities are tightly integrated units. Everyone shares the same wider network of acquaintances, everyone knows everyone else. Two individuals may not have the same circle of immediate friends and relatives (the dozen or so people with whom they interact most often ), but their wider networks of I 5 0

mumtles, that benefit reverberates around the community in a series of overlapping waves as you pass on the benefit you received from me to your a unt, who in turn passes it on to her cousin, who passes it on to his friend - who eventually passes it back to me. My momentary generosity to you is repaid to me not once but repeatedly in the round-robin o f social life in small com­ munities. Despite the inevita ble petty frustrations of life in small communities, the benefits of social obl igation a nd reciprocation are magnified over and over again. I a m not suggesting that large towns are a bad thing or that people shouldn't move in search o f work. Humans have been on the move since time immemorial. Large towns have been an eco­ nomic and social magnet since at least the founding of the Middle Kingdom in Egypt and the heyday of the Mayan empire. Throughout the eighteenth century, London and other European capitals drew people in from far a field in search o f work. They grew in size and power on the industrious backs of these immi­ grants. But what they had to offer was not always fame and for­ tune. Most of these cities grew despite mortality rates that exceed­ ed their birth rates: they grew only because the number of immigrants continued to exceed the rate at which people died off. Poor hygiene and low wages were a principal cause of mortality in the city slums. But these effects were unquestiona bly exacerbat­ ed by another factor that has been overlooked by demographers, and this was the lack of kinship and other support structures in migrant communities. The a bsence of kinship networks has a sur­ prisingly bad effect on people's health . This was dramatically highlighted among both Captain Smith's Virginia colonists in I 6 26 and the famous Donner Party wagon tra in that set out to cross the American West in the I 8 4 6 . In both these cases, mortality was heaviest o n those who had no relatives in the group. Despite often being fit young men at the outset, many of those who travel led al one with the Donner Party were una ble to cope with the depredations of the j ourney. They died earlier and they died in significantly greater numbers. The same effect has been noted in a study of slum dwellers in north­ eastern England during the I 9 5 0S : those families with the smallest

friends, relatives and acquaintances overlap almost completely. In post-industrial societies, this is virtually never the case. You and I share the same su bset of acquaintances at work, but our spouses may not. You and your spouse may share another subset of acquaintances by virtue of belonging to the same church, but I do not. Rather than having a single large shared network, we have sets of sub-networks that only partially overlap. Each of us still has I 50 people in our individual networks, but we may share only I 5 -20 members in common. Our ties of common interest are weakened. By cooperating with you, I gain only from my immediate sel f-interest and by the benefits that you return to me in due course. In traditional com- 200 -

- 20 I -

Grooming, Gossip and the Evolution of Language kinship networks suffered the h ighest levels of both child mor­ bidity ( sickness) and mortality, as well as being generally more susceptible to conditions like depression. Networks of close social ties seem to be fundamental to our survival. Similar results have recently been reported from a study of a rural population in Dominica. This same loss of natural support networks seems to have been responsible for the extraordinary rise in the number of religious and pseudo-religious sects that have proved to be such magnets for the young in the past half century. From Charles Manson to David Koresh and the Rev. Chris Brain, from the Moonies to the Hare Krishnas, it seems that it is the sense of belonging, of com­ munity, of family, that is of overriding importance in drawing in young people. Indeed, some of the more aggressive of these groups deliberately target lonely youngsters for this very reason. In all these cases, beguiling l anguage offers hope for a commu­ nal life that is more welcoming and more secure. Language preys on the emotions, capitalizing on the fact that words can be used to stir deep emotional feelings, to generate opiate highs when used in the right way. History has many examples: religious fundamental­ ism sweeping whole nations, the rise of Fascism, witch-hunts, pogroms and crusades all bear eloquent testimony to this process. They are all consequences of the ease with which we are willing to surrender our individuality to the collective will (or perhaps a sin­ gle charismatic individual's will ) , spurred on by the hype of emo­ tional speech . The psychological mechanism that evolved to facili­ tate the bonding of communities has lost its way because those communities of common interest no longer exist. We are exposed to the risk of exploitation by strangers. In the small community, we h ave long-established bonds of trust, obligation and kinship to guarantee that one person's fiercely argued view will not harm others' interests . In the fragmented communities of the modern world, we no longer have that guarantee. Yet the mechanisms that engender trust in those who claim to be at one with us remain firmly in place. Free-riders have never had it so good. We can see this same effect spilling out in many different aspects of our social lives. The extent to which Lonely Hearts - 2 0 :z. -

The Scars of Evolution columns and dating agencies have boomed in the past two decades is indicative of the fact that people no longer h ave the kinds of social networks available that would normally provide them with access to prospective mates. The village match-maker has vanished with the village. As increasing numbers of people are thrown into a social vacuum through moving to a new city or town in pursuit of a j ob, more of us find ourselves in situations where we lack the social contacts needed to provide us with access to companions and partners. Where do you go to meet people without the risk of undesirable predators ? Personal-ads columns and dating agencies are increasingly becoming a part of our nor­ mal social lives. In somewhat similar vein, I am continually surprised at the extent to which friendships among adults in modern urban com­ munities seem to stem not from the adults' own social contacts but from contacts established by their children through schools and clubs. It may not be too much of an exaggeration to suggest that improved nursery school provision may be more important for the parents than for the children. This is not to say that there is anything intrinsically wrong with any of these phenomena, merely that they reflect the extent to which our psychological baggage may predispose us to certain kinds of social outlets. Nonetheless, the l ack of social contact, the lack of a sense of community, may be the most pressing social problem of the new millennium. Hard Sell around the Photocopier

It is often said that more business is cond ucted on the golf course than at the office desk . There is a very good reason why this should in fact be so. Business deals are personal interactions between individuals. The parties to a deal have to size each other up, assess the likelihood that the other party really means what they say, that they will stick to their word on a deal . You do not acquire that kind of information over the telephone or across a desk in a brief meeting. The golf itself is irrelevant: its purpose is simply to provide an opportunity for bonding. - 20 3 -

Grooming, Gossip and the Evolution of Language The diamond dealers of New York and Amsterdam are in many ways an archetypal example of how a b usiness community of this kind works. A man's word is his bond, because everyone in the community of dealers knows who he is, his past history, his hon­ esty and reliability. The world of the gem dealer is a small, closed world of familiar faces and personal introductions. There is no need for contracts and documents. It all works by trust. But it can only work by trust because it is a small community. It will all col­ lapse if too many people are allowed to j oin the club. Contrast this with the amorphous super-networks of the inter­ national money markets. Large n um bers of complete strangers are linked across the world through modern technology. How much of the current chaos o f the money and insurance markets is a con­ sequence of their size ? Rogue dealers are able to get away with what they do because they are operating in a large anonymous market where o bligation and trust do not exist, while at least some of their colleagues assume that they are still functioning in the small communities whose dealings are built on personal trust. In the modern dispersed electronic markets, dealers can never know all of those they come into contact with. Since trust between strangers is at best a fragile thing, behaviour will inevitably shift towards a new and less comfortable norm. Proponents of the information superl:ighway have always hoped that access to networks of potentially unlimited size all around the world will open up wonderful opportunities for the mass communication of ideas: the global network at the forefront of information technology. Well, it's true that the flow of infor­ mation will generally be greater: I can pick up things that some­ one I've never met (and probably never will meet) has deigned to make available on the Internet. But this won't necessarily open the doors to a worldwide network of associates and colleagues. For one thing, the impersonality of the electronic highway seems to make people less discrete in their interactions with others than when they communicate face to face. They are more likely to be a busive when angry and more likely to make suggestive remarks in passing. What happens is somewhat akin to the 'road rage ' with which we are becoming increasingly familiar. - 204 -

The Scars of Evolution Cocooned in their metal fortresses, people in cars escalate into anger much more quickly than they would had they been involved in an altercation as pedestrians on a sidewalk; cut off from direct face-to-face contact, where su btle cues are read rapidly and care­ fully, they lose the control that social interaction normally impos­ es in the interests of cooperation and bonding. Separated even fur­ ther by the apparent anonymity of the computer link, there is even less to constrain us. The inevitable result will be 'Net rage'. Safe in the knowledge that our opponent cannot get at us, we feel confi­ dent about escalating fights we wouldn't dare risk in a car, never mind a face-to-face encounter. Nor is it likely that electronic mail will significantly enlarge peo­ ple's social networks. It may be faster than snail mail (as computer buffs refer to the conventional letter post), but it can have little effect on the human mind's a bility to handle information a bout other people (as opposed to mere cyphers ) . The information super­ highway's only real benefit in the end will be the speed with which ideas are disseminated. Whenever person-to-person interaction is a necessary feature of the process (as in the striking of deals) , the old and trusted cognitive mindsets will come into play. Suspicion of the unknown and the fear of being duped by untrustworthy strangers will continue to dictate our decisions. As a result, negoti­ ations in large amorphous populations are more likely to be con ­ ducted under the straitjacket of the rule book rather than by intu ­ ition. And where it is really important, we will resort to the trusted age-old machinery of direct personal contact. The old-boy and old­ girl networks will never have seemed so important. Sociologists have long recognized that businesses of less than 200 individuals can operate through the free flow of information among the members. But once their size exceeds this figure, some kind of hierarchical structure or line management system is neces­ sary to prevent total chaos resulting from failures of communica­ tion. Imposing structures of this kind has its costs : information can only flow along certain channels because only certain individ­ uals contact each other regu larly; moreover, the lack of personal­ ized contacts means that individuals lack that sense of personal commitment that makes the world of small groups go round. - 205 -

Grooming, Gossip and the Evolution of Language

The Scars of Evolution

Favours will only be done when there is a clear quid pro quo, an immediate return to the giver, rather than being a matter of com­ munal obligation. Large organizations are j ust less flexible. One solution to this problem would, of course, be to structure large organizations into smaller units of a size that can act as a cohesive group. By allowing these groups to build reciprocal alliances with each other, larger organizations can be built up. However, merely having groups of, say, 1 5 0 will never of itself be a panacea to the problems of organization . Something else is needed: the people involved must be able to build direct personal relationships. To allow free flow of information, they have to be able to interact in a casual way. Maintaining too formal a struc­ ture of relationships inevitably inhibits the way a system works. The importance of this was drawn to my attention a couple of years ago by a TV producer. The production unit for which she worked produced all the educational output for a particular TV station. Whether by chance or by design, it so happened that there were almost exactly 1 5 0 people in the unit. The whole process worked very smoothly as an organization for many years until they were moved into new purpose-built accommodation . Then for no apparent reason, everything started to fall apart. The work seemed to be more difficult to do, not to say less satisfying. It was some time before they worked out what the problem was. It turned out that, when the architects were designing the new building, they decided that the coffee room where everyone ate their sandwiches at lunchtime was an unnecessary luxury and so dispensed with it. The logic seemed to be that if people were encouraged to eat their sandwiches at their desks, then they were more likely to get on with their work and less likely to idle time away. And with that, they inadvertently destroyed the intimate social networks that empowered the whole organization. What had apparently been happening was that, as people gathered informally over their sandwiches in the coffee room, useful snip­ pets of information were casually being exchanged. Someone had a problem they could not solve, and began to discuss it over lunch with a friend from another section. The friend knew j ust the per­ son to ask. Or someone overhearing the conversation would have

a suggestion, or would go away and happen to bump into some­ one who knew the answer a day or so later; a quick phone call and the problem was resolved. Or a casual comment sparked an idea for a new programme. It was these kinds of chance encounters over the coffee machine, idle chatter around the photocopier, that made the dif­ ference between a successful organization and a less successful one. By encouraging casual contacts, the old system had created a network of relationships around each member of the staff and these acted like a parallel-processing supercomputer: several dif­ ferent brains could be working on a problem at the same time independently of each other.

- 206 -


Language permeates human culture, underpinning our societies as much as our science and our art. Its roots go back into the distant past, and that ancient history is part of our mental baggage . We can do the most remarkable things with language. Yet underlying it all are minds that are not infinitely flexible, whose cognitive predispositions are designed to handle the kinds of small-scale societies that have characterized all but the last minutes of our evolutionary history. That need not be a signal for despair. It is simply something we have to work with, something we have to adjust our social practices to take account of rather than fight against. Nor does it mean that human behaviour is incapable of change. That is a common but naive reading of the evolutionary message. Like all primates and many mammals, humans are characterized by behavioural flexibili­ ty and the ability to adj ust within the constraints of the machinery'S design. The future of our species will be determined by our ability to recognize where those limitations lie and how we can circumvent them, if necessary by recreating the kinds of social environments in which we work best. If we can achieve that, the modern world may seem less alienating and become less destructive.

- 207 -



ourn al arti� l � s in w.hich the The fol lowi ng list inclu des the main j be f? und; It IS not mten ded key resul ts discu ssed in this book can stud Ies quoted. F?r the gen­ as a com plete refer ence list for all the artic les that prOV Ide furth er eral read er, a selec tion of book s and readi ng on speci fic areas is inclu ded. Chapter I: Talk ing Head s l Networks. Black well, Oxfo rd. Milro y, R. ( 1 9 8 7 ) . Language and Socia on. Lyon s, j. ( 1 9 7 0 ) . Chom sky. Fonta na, Lond ct. Allen La �e, Lond on. Pinke r, S. ( 1 99 4 ) . The Language Instin Introductton to Language and Tudg ill, P . ( 1 9 8 3 ) . Sociolinguistics: An Society. Peng uin, Harm ondsw orth. . Chapter 2: Into the Socia l Whir l . rl1la . of Cahfo . rslty tion. Ul1lve Bowl er, P. J. ( 1 9 8 6 ) . The Idea of Evolu Press, Los Angel es. . . rd Ul1lv� rslty Pres � , Oxfo rd. Byrne , R. ( 1 9 9 5 ) . The Thinking Ape. Oxfo . mg pnma te , s gUide to decep Byrne, R . , and Whit en , A. ( 1 9 8 7 ) . 'The think . tion.' New Scientist I I 6 ( No. 1 5 8 9 ) : 5 4-7· . 8 ) . Machiavelltan Intelltgence. B yrne, R ., an d Whit en , A. (eds. ) ( 1 9 8 Oxfo rd Unive rsity Press, Oxfo rd. . How Monk eys See the World. Chen ey, D. L., and Seyfa rth, R. M. ( 1 99 0 ) go. Chica Chica go Unive rsity Press, B. ( 1 9 9 5 ) · :The role of grunt s Chen ey, D. L., Seyfa rth, R. M., and Silk, j. intera ctIOns amon g adult in recon ciling oppon ents and facilit ating 7· 249-5 0: 5 femal e baboo ns.' Animal Behav iour rsi:y ,r ress, Oxford. Unive d Oxfor . Gene Dawk ins, R. ( 1 9 7 6 ) . The Selfish . hen and P. Smge r Cava P. In: .' mind the in 'Gaps . ) 3 Dawk ins, R. ( 1 9 9 h Estate , Londo n. (eds . ) , The Great Ape Project, pp. 80- 8 7 . Fourt Idea. Allen Lane, erous Dang Denn ett, D . ( 1 9 9 5 ) ' Darwin's Harmo ndswo rth .

- 208 -

Dunbar, R. I. M. ( 1 9 8 4 ) . Reproductive Decisions: An Economic Analysis of Gelada Baboon Social Strategies. Princeton University Press, Princeton. Dunbar, R. I. M. ( 1 9 8 8 ) . Primate Social Systems. Chapman & Hall, London. Fleagle, j . G. ( 1 9 8 8 ) . Primate Adaptation and Evolution. Academic Press, New York. Goodall, j. ( 1 9 8 6 ) . The Chimpanzees of Gombe: Patterns of Behaviour. Harvard University Press, Cambridge ( Mass. ) . Gribbin, j . , and Gribbin, M . ( 1 9 9 3 ) . Being Human: Putting People i n an Evolutionary Perspective. Dent, London. Harcourt, A., and de Waal, F. (eds . ) ( 1 9 9 3 ) . Coalitions and Alliances in Humans and Other Animals. Oxford University Press, Oxford. Hinde, R. A. ( ed. ) ( 1 9 8 3 ) . Primate Social Relationships. Blackwells Scientific Publications, Oxford. Jones, S., Martin, R. D., and Pilbeam D. (eds . ) ( 1 9 9 2 ) . The Cambridge Encyclopedia of Human Evolution. Cambridge University Press, Cambridge. Martin, R. D. ( 1 9 9 0 ) . Primate Origins and Evolution. Chapman & Hall, London. Pearson, R. ( 1 9 7 8 ) . Climate and Evolution. Academic Press, London. Richard, Alison F. ( 1 9 8 5 ) . Primates in Nature. W. H. Freeman, San Francisco. Smuts, B. B., Cheney, D. L., Seyfarth, R. M., Wrangham, R. W., and Struhsaker, T. T. ( eds . ) ( 1 9 8 7 ) . Primate Societies. Chicago University Press, Chicago. Smuts, B. B. ( 1 9 8 5 ). Sex and Friendship in Baboons. Aldine, New York. de Waal, F . ( 1 9 8 2) . Chimpanzee Politics. Allen & Unwin, London. de Waal, F., and van Roosmalen, j. ( 1 97 9 ) . 'Reconciliation and consola­ tion among chimpanzees. ' Behavioural Ecology and Sociobiology 5 : 5 5-66.



The Importance of Being Earnest

Abott, D. H . ( 1 9 84 ) . ' Behavioural and physiological suppression of fertil­ ity in subordinate marmoset monkeys.' American Journal of Primatology 6: 1 69-8 6. Abbott, D . H., Keverne, E. B., Moore, G. F., and Yodyinguad. U. ( 1 9 8 6 ) . 'Social suppression o f reproduction in subordinate talapoin monkeys, Miopithecus talapoin.' In: j. Else and P. C. Lee (eds . ) , Primate Ontogeny, pp. 3 29-4 1 . Cambridge University Press, Cambridge. Barton, R. ( 1 9 8 5 ) . 'Grooming site preferences in primates and their func­ tional implications. ' International Journal of Primatology 6: 5 1 9-3 1 . Bowman, L . A., Dilley, S . R., and Keverne, E . B . ( 1 97 8 ) . 'Suppression of oestrogen-induced LH surges by social subordination in talapoin

- 209 -

Grooming, Gossip and the Evolution of Language


monkeys.' Nature, London, 2 7 5 : 5 6- 8 . Cheney, D . L., and Seyfarth, S. M. ( 1 9 9 0 ) . How Monkeys See the World. Chicago University Press, Chicago. Dunbar, R. I. M. ( 1 9 8 5 ) . 'Stress is a good contracep tive.' New Scientist 1 0 5 ( 1 7 January ) : 1 6- 1 8 . Dunbar, R . I . M . ( 1 9 8 8 ) . Primate Social Systems. Chapman & Hall, London. Dunbar, R. I . M. ( 1 9 9 1 ) . 'Functiona l significanc e of social grooming in

Byrne, R. ( 1 9 9 5 ) . The Thinking Ape. Oxford University Press, Oxford. Byrne, R., and Whiten, A. (eds . ) ( 1 9 8 8 ) . Machiavellian Intelligence. Oxford University Press, Oxford. Coleman, J. S. ( 1 9 64 ) . An Introduction to Mathematical Sociology. Collier-Macmillan, London. Dunbar, R. I. M. ( 1 99 2 ) . 'Neocortex size as a constraint on group size in primates. ' Journal of Human Evolution 20: 4 6 9-9 3 . Dunbar, R . I . M . ( 1 9 9 3 ) . ' Coevolution o f neocortical size, group size and language in humans . ' Behavioural and Brain Sciences 1 6: 6 8 1 -7 3 5 . Dunbar, R . I . M . ( 1 9 9 2 ) . 'Why gossip i s good for you.' New Scientist

primates.' Folia primatologica 57: 1 2 1 -3 I . Enquist, M., and Leimar, O . ( 1 9 9 3 ) . 'The evolution of cooperatio n in mobile organisms .' Animal Behaviour 4 5 : 7 4 7-5 7 · Goosen, C. ( 1 9 8 1 ) . ' O n the function o f allogroom ing in O l d World mon­ keys.' In: A. B. Chiarelli and R. S. Corruccini ( eds . ) , Primate Behaviour and Sociobiology, pp. I I o-20. Springer, Berlin. Hockett, C . F. ( 1 9 60 ) . ' Logical considerat ions in the study of animal communic ation. ' In: W. E. Lanyon and W. N . Tavolga ( eds . ) , Animal Sounds and Communication, pp. 3 9 2-4 3 0 . American Institute of Biological Sciences, Washingto n. Howlett, T., Tomlin, S., Ngahfoong , L., Rees, L., Bullen, B., Skrinar, G., and MacArthu r, J . ( 1 9 8 4 ) . 'Release of B endorphin and met-enkep halin during exercise in normal women: response to training.' British

Medical Journal

2 8 8 : 1 9 5 0- 2 .

Keverne, E. B., Martensz, N., and Tuite, B. ( 1 9 8 9 ) . ' Beta-endorp hin con­ centrations in cerebrospin al fluid on monkeys are influenced by groom­ ing relationship s.' Psychoneuroendcrin-ology 1 4 : 1 5 5-6 I . Mason, H . ( 1 9 8 4 ) . 'Everything you wanted to know about sperm banks.' Observer ( 20 August), p . 3 5 · Silk, J . ( 1 9 8 2 ) . 'Altruism among female Macaca radiata: explanation s and analysis of patterns of grooming and coalition formation.' Behaviour 79: 1 6 2-8 8 .

Sparks, J . ( 1 9 6 7 ) . 'Allogroom ing i n primates: a review. ' In: D . Morris (ed . ) , Primate Ethology, pp. 1 4 8-7 5 . Weidenfeld & Nicholson, London. Smuts, B. B., Cheney, D. L., Seyfarth, R. M., Wrangham, R. W., and Struhsaker, T. T. (eds . ) ( 1 9 8 7 ) . Primate Societies. Chicago University Press, Chicago. Wasser, S. K., and Barash, D. P. ( 1 9 8 3 ) . 'Reproductive suppression among female mamma ls: impl ications for biomedicine and sexual selection theory. ' Quarterly Reviews of B iology 5 8 : 5 1 3 -3 8 .



Of Brains and Groups and Evolution

Buys, C . J., and Larsen, K . L.

Psychology Reports

( 1 979).

4 5 : 5 4 7- 5 3 ·

- 210 -

'Human sympathy groups.'

1 3 6 : 2 8-3 I .

Dunbar, R . I . M . , and Spoors, M . ( 1 9 9 5 ) . 'Social networks, support cliques amd kinship. ' Human Nature 6: 2 7 3 -9 0 . Friedman, J . , and Rowlands, M. j. ( eds. ) ( 1 9 7 7 ) . The Evolution of Social Systems. Duckworth, London. Hayes, K., and Hayes, C. ( 1 9 5 2 ) . 'Imitation in a home-raised chim­ panzee. ' Journal of Comparative Psychology 4 5 : 4 5 0-9 . Jerison, H. J . ( 1 9 7 3 ) . Evolution of the .Brain and Intelligence. Academic Press, New York. Johnson, G. A. ( 1 9 8 2 ) . 'Organizational structure and scalar stress.' In: Renfrew, c., Rowlands, M., and Abbott-Seagram, B. ( eds . ) , Theory and Explanation in Archaeology. Academic Press, London. Kellogg, W. N., and Kellogg, L. A. ( 1 9 3 3 ) . The Ape and the Child: A Study of Environmental Influence upon Early behaviour. Whittlesey House, New York. Killworth, P. D., Bernard, H. R., and McCarty, C. ( 1 9 8 4 ) . 'Measuring patterns of acquaintanceship.' Current Anthropology 2 5 : 3 9 1 - 7 . Kudo, H . , Bloom, S., and Dunbar, R. I. M . ( submitted ) . 'Neocortex size as a constraint on social network s ize in primates.' Behaviour. Lewin, R. ( 1 9 9 2 ) . 'The great brain race.' New Scientist ( 5 December ), pp. 2-8 .

Naroll, R.

( 1 9 5 6).


'A preliminary index of social development.' American 5 8 : 6 8 7-7 1 5 .

Premack, D., and Premack, A . J . ( 1 9 8 3 ) . The Mind of an Ape. Norton, New York . Savage-Rumbaugh, S. ( I 9 8 0 ) . Ape Language: From Conditioned Response to Symbol. Oxford University Press, New York.



The Ghost in the Machine

Astington, J . W. ( 1 9 9 4 ) . The Child's Discovery of the Mind. Fontana, London. Barkow, J . H., Cosmides, L., and Tooby, J . (eds . ) ( 1 9 9 3 ) . The Adapted Mind. Oxford University Press, Oxford. Baron-Cohen, S. ( 1 9 9 1 ) . 'The theory of mind impairment in autism.' In: - 211 -

Grooming, Gossip and the Evolution of Language Whiten, A. (ed . ) , Natural Theories of Mind, pp. 2 3 3 - 5 2 . Oxford University Press, Oxford. Byrne, R. ( 1 9 9 5 ) . The Thinking Ape. Oxford University Press, Oxford. Cheney, D. L., and Seyfarth, S. M. ( 1 9 9 0 ) . How Monkeys See the World. Chicago University Press, Chicago. Dennett, D. ( 1 9 8 3 ) ' 'Intentional systems in cognitive ethology: the " Panglossian paradigm" defended . ' Behavioural and Brain Sciences 6: 3 4 3-90.



Donald, M. ( 1 9 9 1 ) . Origins of the Modern Mind. Harvard Umverstty Press, Cambridge (Mass ) . Dunbar, R . I . M . ( 1 9 9 5 ) . The Trouble with Science. Faber & Faber, London. Gallup, G. G. ( 1 9 7 0 ) . 'Chimpanzees: self-recognition.' Science 1 67 : 4 1 7- 2 1 .

Gallup, G . G .

( 1985).

'Do minds exist in species other than our own ? '

Neuroscience and Biobehavioural Reviews 9 : 6 3 1 -4 1 . Happe, F . ( 1 9 9 4 ) . Autism: A n Introduction to Psychological Theory. University College London Press, London. Kinderman, P., Dunbar, R., and Bentall, R. ( submitted) . 'Theory-of-mind deficits, causal attributions and paranoia: an analogue study.' British

Journal of Psychology.



Leslie, A. M. ( 1 9 8 7 ) . 'Pretence and representation in infancy: the ongms of theory of mind.' Psychological Review 9 4 : 8 4- 1 0 6 . . Parker, S., Mitchell, R. W., and Boccia, M. L. ( eds. ) ( 1 9 9 4 ) , Self-aware­ ness in Animals and Humans. Cambridge University Press, Cambridge. Povinelli, D. J . ( 1 9 8 9 ) . 'Failure to find self-recognition in Asian elephants (Elephas maximus) in contrast to their use of mirror cues to discover hidden food.' Journal of Comparative Psychology 1 0 3 : 1 2 2-3 1 . Povinelli, D . j . , Nelson, K . E., and Boysen, S . T . ( 1 9 9 0 ) . 'Inferences a bout guessing and knowing by chimpanzees (Pan troglodytes).' Journal of

Comparative Psychology

1 04 : 20 3 - 1 0 .

Povinelli, D . J., Nelson, K . E. and Boysen, S. T. ( 1 9 9 2 ) . 'Comprehension of social role reversal by chimpanzees: evidence of empathy ? ' Animal


4 3 : 6 3 3-4 0 .

Premack, D . , and Woodruff, G. ( 1 9 7 8 ) . 'Does the chimpanzee have a theory of mind ? ' Behavioural and Brain Sciences 4: 5 1 5 - 2 6 . RACTER ( 1 9 8 5 ) . The Policeman's Beard is Half Constructed. Warner Books, New York. Weiskrantz, L. (ed . ) ( 1 9 8 5 ) ' A nimal Intelligence. Oxford University Press, Oxford. Wolpert, L. ( 1 9 9 4 ) . The Unnatural Nature of Science. Faber & Faber, London. de Waal, F. ( 1 9 8 2 ) . Chimpanzee Politics. Allen & Unwin, London. Whiten, A. (ed. ) ( 1 9 9 1 ) . Natural Theories of Mind. Blackwell, Oxford. - 212 -

Bibliography Whiten, A. , and Byrne, R.

'Tactical deception

( 19 8 8 ) .

Behaviou ral and Brain Sciences

I I:


primates . '

2 3 3 -4 4 .

Chapter 6: Up Through the Mists of Time

Aiello, L., and Dunbar, R. I . M . ( 1 9 9 3 ) . 'Neocort ex size, group size and the evolution of language . ' Current Anthrop ology 3 4 : 1 8 4-9 3 . Aiello, L., and Wheeler, P . ( 1 9 9 5 ) . 'The expensive tissue hypothes is.'

Current Anthropo logy

3 6: 1 9 9-2 1 1 .

Alexande r, R . D., Hoogland , j . L., Howard, R . D., Noonan, K . M., and Sherman , P. W. ( 1 9 7 9 ) . 'Sexual dimorph isms and breeding systems in pinnipeds ., ungulates , primates and humans.' In: N. Chagnon and W. Irons ( eds . ) , Evolutionary Biology and Human Social Behaviour, pp. 4 0 2-3 5 . Duxbury , North Scituate ( Mass. ) . Arensbur g, B . , Tillier, A. M . , Vanderm eersch, B . , Duday, H . , Schepartz , L. A., and Rak, Y. ( 1 9 8 9 ) . 'A Middle Palaeolith ic human hyoid bone.' Nature, London , 3 3 8 : 7 5 8-9 . Bischopin g, K. ( 1 9 9 3 ) . 'Gender differences in conversat ion topics, 1 9 22- 1 9 9 °.'

Sex Roles

28: 1-1 7.

Clarke, R. j., and Tobias, P . V . ( 1 9 9 5 ) . 'Sterkfon tein member 2 foot bones of the oldest South African hominid.' Science 2 6 9 : 5 2 1 -4 . Cohen, J . E . ( 1 9 7 1 ) . Casual Groups o f Monkeys and Men. Harvard University Press, Cambridg e ( Mass. ) . Dunbar, R. I. M. ( 1 9 8 8 ) . Primate Social Systems. Chapman & Hall, London. Dunbar, R. I. M. ( 1 9 9 3 ) . 'Coevolut ion of neocortica l size, group size and language in humans.' Behaviour al and Brain Sciences 1 6: 6 8 1-73 5 . Dunbar, R . I . M . , Duncan, N . D . c., and Nettle, D . ( 1 9 9 5 ) . 'Size and structure of freely forming conversat ional groups.' Human Nature 6: 67-7 8 .

Dunbar, R . I. M . , and Spoors, M. ( 1 9 9 5 ) . 'Social networks, support cliques and kinship. ' Human Nature 6: 27 3 -9 0 . janis, C. ( 1 9 7 6 ) . 'The evolutiona ry strategy of the Equidae and the ori­ gins of rumen and caecal digestion.' Evolution 3 0 : 7 5 7-74 . jones, S., Martin, R. D . , and Pilbeam D . ( eds. ) ( 1 9 9 2 ) . The Cambridge Encyclopedia of Human Evolution . Cambridge University Press, Cambridge. Legget, R. F., and Northwoo d, T. D. ( 1 9 6o ) . 'Noise surveys at co*cktail parties. ' Journal of the Acoustica l Society of America 3 2: 1 6- 1 8 . Lieberman n, D . ( 1 9 8 9 ) . 'The origins o f some aspects o f language and cognition.' In P. Mellars and C. Stringer (eds.), The Human Revolution , pp. 3 9 1-4 1 4 . Edinburgh University Press, Edinburgh. Mellars, P., and Stringer, C. ( eds . ) ( 1 9 8 9 ) . The Human Revolution . Edinburgh University Press, Edinburgh . - 213 -


Grooming, Gossip and the Evolution of Language Nettle, D.

( 1 99 4 ) .

'A behavioural correlate of phonologica l structure.'

Language and Speech 3 7 : 4 2 5-9 . Pinker, S. ( 1 9 9 4 ) . The Language Instinct. Allen Lane, London.

Sigg, H., and Stolba, A. ( 1 9 8 I ) . 'Home range and daily march m a hamadryas b aboon troop.' Folia Primatologica 3 6 : 40-7 5 . van Soest, P . j . ( 1 9 8 2 ) . The Nutritional Ecology of the Ruminant. Cornell University Press, Ithaca. Stringer, c., and Gamble, C. ( 1 9 9 3 ) . In Search of the Neanderthals. Thames and Hudson, London. Wheeler, P. E. ( 1 9 8 8 ) . 'Stand tall to stay cool.' New Scientist ( December) pp. 6 2-5 ·

Wheeler, P. E. ( 1 9 9 1 ) . 'The influence of bipedalism on the energy and water budgets of early hominids.' Journal of Human Evolution 2 I : I 07-3 6 .



First Words

Asch, S . E. ( 1 9 5 6 ) . 'Studies of independen ce and conformity. A minority of one against a unanimous maj ority.' Psychological Monographs 70, NO · 9 · Bever, G . , and Chiarello, R. j . ( 1 9 7 4 ) . ' Cerebral dominance in musicians and non-musicia ns.' Science 1 8 5 : 1 3 7-9 · . Bott, E. ( 1 9 7 1 ) . Family and Social Network. Tavistock Publications , London. Bradshaw, j., and Rogers, L.

( 1 9 9 3 ) . The Evolution of Lateral Asymmetries, Language, Tool Use and Intellect. Academic Press, New

York. Calvin, W. H.

Dunbar, R. I. M., and Spoors, M. ( I 9 9 5 ) . ' Social networks, support cliques amd kinship .' Human Nature 6 : 2 7 3 -9 0 . Foley, R. A. ( 1 9 8 9 ) . 'The evolution of hominid socia l behaviour.' In: V. Standen and R. Foley (eds . ) , Comparative Socioecology, pp. 4 7 3 -9 4 Blackwell Scientific Publications, Oxford. Foley, R. A., and Lee, P. C. ( 1 9 8 9 ) . ' F inite social space, evolutionary pathways and reconstructing hominid behaviour.' Science 24 3 : 9 0 I -6 . Hauser, M. ( 1 9 9 3 ) . 'Right hemisphere dominance for t h e production of facial expression in monkeys.' Science 2 6 I: 4 7 5-7. Hauser, M., and Fowler, C. ( 1 9 9 1 ) . 'Declination in fundamental frequen­ cy is not unique to human speech.' Journal of the Acoustical Society of


9 I : 3 6 3 -9 .

The Origin of Conciousness i n the Breakdown o f the Bicameral Mind. Houghton Mifflin, New York. Kinzey, W . (ed. ) ( 1 9 8 7 ) . The Evolution of Human Behaviour: Primate Models. State University of New York Press, Albany. Knight, C. ( 1 99 0 ) . Blood Relations: Menstruation and the Origins of Culture. Yale University Press, New Haven ( Conn . ) . jaynes, j .

( I 99 0 ) .

Manning, J . T. , Chamberlain, A . T . , a nd Heaton, R. ( 1 9 9 4 ) . ' Left-sided cradling: similarities and differences between apes and humans.'

Journal of Human Evolution

3 2: 22 1 - 5 5 ·

Schumacher, A . ( 1983)'

The Throwing Madonna: From Nervous Cells to

Hominid Brains. McGraw-Hill, New York. Casperd, j., and Dunbar, R. 1. M. ( 1 9 9 6 ) . 'Asymmetries in the visual pro­ cessing of emotional cues during agonistic interactions by gelada baboons.' Behavioural Processes ( in press) . Cheney, D . L . , and Seyfarth, R . M. ( 1 9 8 2 ) . ' How vervet monkeys per­ ceive their grunts: field playback experiments.' Animal Behaviour 3 °: 7 3 9-5 1 ·

Corballis, M . C . ( 1 9 9 1 ) . The Lopsided Ape. Oxford University Press, Oxford. Davies, N. B., and Halliday, T. R. ( 1 9 7 7 ) . ' Optimal mate selection in the toad Bufo bufo.' Nature, London, 2 6 9 : 5 6- 8 . Denman, J., and Manning, j. T. {submitted}. ' Lateral cradling preferences and left-eye-mediated perceptions of e motions.' Ethology and

Sociobiology. Dunbar, R. I. M. London.


Primate Social Systems. Chapman

& Hall,

( 1 982).

'On the significance o f stature i n human society.'

Journal of Human Evolution I I : 6 9 7-7 0 I . de Waal, F . ( 1 9 8 2 ) . Chimpanzee Politics. Allen

& Unwin, London. de Waal, F. ( 1 9 8 4 ) . 'Sex differences in the formation of coalitions among chimpanzees.' Ethology and Sociobiology 5: 2 3 9-5 5 . Wallach, M . A., Kogan, N., and Bern, D . j . ( 1 9 6 2 ) . ' Group influence on individua l risk-taking.' Journal of Abnormal and Social Psychology 6 5 : 7 5- 8 6 .

Wallach, M. A., Kogan, N., and Bern, D . j . ( 1 9 64 ) . 'Diffusion of respon­ sibility and level of risk-taking in groups.' Journal of Abnormal and

Social Psychology

6 8 : 2 6 3 -7 4 .

Chapter 8: Babel's Legacy Cavalli-Sforza, L. L., Piazza, A., Menozzi, P., and Mountain, j. L. ( 1 9 8 8 ) . 'Reconstruction o f human evolution: bringing together genetic, archae­ ological and linguistic data. ' Proceedings of the National Academy of

Sciences, USA

- 214 -

2 6 : 7 7-8 3 .

Richman, B . ( 1 9 7 6 ) . 'Some vocal distinctive features used by gelada mon­ keys.' Journal of the Acoustical Society of America 60: 7 1 8-24. Rodseth, L., Wrangham, R. W. , Harrigan, A., and Smuts, B. B. ( 1 9 9 1 ) . 'The human community as a primate society.' Current Anthropology

8 5 : 6002-6.

- 215 -

Grooming, Gossip and the Evolution of Language


Daly, M., and Wilson, M. ( 1 9 8 8 ) . Homicide. Aldine, New York. Dawkins, R. ( 1 9 7 6 ) . The Selfish Gene. Oxford University Press, Oxford. Dunbar, R. I. M., Clark, A., and Hurst, N. L. ( 1 99 5 ) . 'Conflict and cooperation a mong the Vikings: contingent behavioural decisions. '

Coser, R. L. ( 1 9 60 ) . 'Laughter among colleagues.' Psychiatry 2 3 : 8 1 -9 5 . Cosmides, L., and Tooby, J. ( 1 9 9 3 ) . 'Cognitive adaptations for social exchange. ' In: Barkow, J. H., Cosmides, L., and Tooby, J. (eds . ) , The Adapted Mind, pp. 1 6 2-2 2 8 . Oxford University Press, Oxford. Daly, M., and Wilson, M. ( 1 9 8 8 ) . 'Evolutionary psychology and family homicide.' Science 2 4 2 : 5 1 9-24 . Dunbar, R. I. M. ( 1 9 9 3 ) . 'The co-evolution of neocortical size, group size and the evolution of language in humans. ' Behavioural and Brain

Ethology and Sociobiology

1 6 : 2 3 3-4 6 .

Green, S. ( 1 9 7 5 ) . 'Dialects in Japanese monkeys: vocal learning a n d cul­ tural transmission of locale-specific vocal behaviour? ' Zeitschrift fur


3 8 : 3 04- 1 4 .

Hughes, A. L. ( 1 9 8 8 ) . Evolution and Human Kinship. Oxford University Press, Oxford. Johnson, G. R., Ratwick, S. H., and Swyer, T. J. ( 1 9 8 7 ) . 'The evocative significance of kin terms in patriotic speech.' In: Reynolds, V., Falger, V., and Vine, I. ( eds. ), The Sociobiology of Ethnocentrism, pp. 1 5 7-74 · Chapman & Hall, London. Knight, C. ( 1 9 9 0 ) . Blood Relations: Menstruation and the Origins of Culture. Yale University Press, New Haven. Mitani, J., Hasegawa, T., Gros-Louis, J., Marler, P., and Byrne, R. ( 1 9 9 2 ) . 'Dialects in wild chimpanzees ? ' American Journal of


2 3 : 2 3 3-4 3 ·

Nettle, D., and Dunbar, R. I . M . ( submitted) . ' Social markers and the evolution of reciprocal exchange.' Current Anthropology. Panter-Brick, C. ( 1 9 8 9 ) . 'Motherhood and subsistence work: the Tamang of rural Nepal.' Human Ecology 1 7 : 2 0 5 - 2 8 . Pinker, S. ( 1 9 9 4 ) . The Language Instinct. Allen Lane, London. Renfrew, C. ( 1 9 9 4 ) . 'World linguistic diversity. ' Scientific American 270: 1 04- 1 1 .

Ridley, M. ( 1 9 9 4 ) . The Red Queen. Viking, London. Shaw, R. P., and Wong, Y. ( 1 9 8 9 ) . Genetic Seeds of Warfare: Evolution, Natiuonalism and Patriotism. Unwin Hyman, Boston. Stoneking, M., and Cann, R. ( 1 9 8 9 ) . ' African origin of human mitochon­ drial DNA.' In: Mellars, P., and Stringer, C. (eds . ) . The Human Revolution, pp. 1 7-3 0 . Edinburgh University Press, Edinburgh.



The Little Rituals of Life

Bachmann, c., and Kummer, H. ( 1 9 8 0 ) . 'Male assessment of female choice in hamadryas baboons. ' Behavioural Ecolology and


6: 3 1 5-2 1 .

Betzig, L., Borgerhoff Mulder, M., and Turke, P . ( 1 9 8 8 ) . Human Reproductive Behaviour. Cambridge University Press, Cambridge. Bischoping, K. ( 1 9 9 3 ) . ' Gender differences in conversation topics, 1 9 2 2 1 990.'

Sex R oles

28: 1-1 8 .

Buss, D . ( 1 99 4 ) . The Evolution of Desire. Basic Books, New York. Coates, J . ( 1 9 9 3 ) . Women, Men and Language. Longman, New York.

- 216 -


1 6: 68 1 -7 3 5 .

Dunbar, R . I . M., Duncan, N . D . c., and Marriott, A . (submitted) . ' Human conversational behaviour: a functional approach. ' Ethology

and Sociobiology. Eakins, B. W., and Eakins, R. G. ( 1 9 7 8 ) . Sex Differences in Human Communication. Houghton Mifflin, Boston. Emler, N. ( 1 9 9 0 ) . 'A social psychology of reputations.' European Journal

of Social Psychology Emler, N .

( 1 99 2 ) .


1 7 1 -9 3 .

'The truth about gossip.' Social Psychology Newsletter

2 7 : 2 3 -3 7 ·

Goodman, R. F., and Ben-Ze'ev, A , ( eds. ) ( 1 99 4 ) . Good Gossip. University of Kansas Press. Grammer, K. ( 1 9 8 9 ) . 'Human courtship behaviour: biological basis and cognitive processing.' In: A. Rasa, C . Vogel and E. Voland (eds . ) , The Sociobiology of Sexual and Reproductive Behaviour, pp. 1 4 7-69 . Chapman & Hall, London. Hawkes, K . ( 1 9 9 1 ) . 'Showing off: tests of another hypothesis about men's foraging goals.' Ethology and Sociobiology I I : 29- 5 4 . Huxley, E. ( 1 9 8 7 ) . Out in the Midday Sun. Penguin, Harmondsworth. Kipers, P. S. ( 1 9 8 7 ) . 'Gender and topic.' Language and Society 1 6 : 5 4 3 -5 7 ·

Miller, G. ( 1 9 9 6 ) . 'Sexual selection in human evolution: review and prospects. ' In: C. Crawford and D. Krebs ( eds . ) , Evolution and Human Behaviour: Ideas, Issues and Applications. Lawrence Erlbaum, New York. Moore, M. M . ( 1 9 8 5 ) . 'Non-verbal courtship patterns in women: context and consequences.' Ethology and Sociobiology 6 : 2 3 7-4 7 . Petrie, M. ( 1 9 9 4 ) . 'Improved growth a n d survival o f offspring of pea­ co*cks with more elaborate trains.' Nature, London 3 7 1 : 5 9 8-9 . Petrie, M . , and Halliday, T. ( 1 9 9 4 ) . 'Experimental and natural changes in the peaco*ck's (Pave cristatus ) train can affect mating success.'

Behavioural Ecology and Sociobiology

3 5 : 2 1 3-17.

Provine, R . R . ( 1 9 9 3 ) . 'Laughter punctuates speech: linguistic, social and gender contexts of laughter. ' Ethology 9 5 : 29 1 - 8 . Ridley, M . ( 1 9 9 4 ) The Red Queen. Viking, London. Smuts, B. B. ( 1 9 8 5 ) . Sex and Friendship in Baboons. Aldine, New York

- 217 -

Grooming, Gossip and the Evolution of Language Voland, E., and Engel, C. ( 1 99 0 ) . 'Female choice in humans: a condition­ al mate choice strategy of the Krummhorn women ( Germany 1 7 20-1 8 7 4 ) . '


8 4 : 1 4 4- 5 4 ·


Waynforth, D., and Dunbar, R. I. M. ( 1 9 9 5 ) . 'Conditional mate choice strategies in humans: evidence from lonely-hearts advertisem*nts.'

Behaviour Zahavi, A.

1 3 2 : 7 5 5-7 9 ·

( 1975 ).

'Mate selection - a selection for a handicap.' Journal

of Theoretical Biology Chapter


5 3 : 205-14.

The Scars of Evolution

Bott, E. ( 1 9 7 1 ) . Family and Social Network . Tavistock Publications, London. Cohen, J. ( 1 9 7 1 ) . Casual Groups of Monkeys and Men. Harvard University Press, Cambridge ( Mass. ) . Coleman, J . S . ( 1 9 6 4 ) . Introduction to Mathematical Sociology. Collier­ Macmillan, London. Dunbar, R. I. M., Duncan, N . D. c., and Nettle, D. ( 1 99 5 ) . 'Size and structure of freely forming conversational groups. ' Human Nature 6 : 67-7 8 .

Flinn, M . , and England, B . G . ( 1 9 9 5 ) . 'Childhood stress and family envi­ ronment.' Current Anthropology 3 6 : 8 5 4-66. Grayson, D. K. ( 1 99 4 ) . 'Differential mortality and the Donner party dis­ aster.' Evolutionary Anthropology 2 : 1 5 1 -9 . Legget, R., F . , a n d Northwood, T . D. ( 1 9 6 0 ) . 'Noise surveys a t co*cktail parties.' Journal of the Acoustical Society of America 3 2 : 1 6- 1 8 . McCormick, N. B . , and McCormick, J. W . ( 1 9 9 2 ) . 'Computer friends and foes: content of undergraduates' electronic mail. ' Computers and

Human Behaviour

8: 3 7 9-40 5 .

Milardo, R . M . (ed. ) ( 1 9 8 8 ) . Families and Social Networks. Sage, Newbury Park ( Ca . ) . Morgan, E. ( 1 9 9 0 ) . The Scars of Evolution. Souvenir Press, London. Young, M., and Willmott, P. ( 1 9 5 7 ) . Family and Kinship in East London. Routledge and Kegan Paul, London.

Abbott, David, 4 2

I 2, 6 3 ; brain size/body weight, 5 8;

accents, 1 8 3 -4 , 1 8 6, 1 8 8

chest shape, 1 3 3 , 1 3 4; common

Ache hunter-gatherers, 1 8 6

ancestor of monkeys and apes, I I ;

'acoustic hiding', 9 2-3

comparing understanding o f anoth­

advertising, sexual, l S I , 1 77, 1 80-82,

er's intentions with that of mon­

1 8 5 , 1 8 7, 1 9 1 , 1 9 2, 1 9 3

keys, 9 8-9; contact calls, I I 5; use

Afar desert, 1 08

of gestures, 1 3 5 ; and grooming, 4,

'African Eve', 1 6 1

2 1 , 6 5 , 77; and groups, 7 1 , I I 2;

afterlife, I I I

and hairlessness, 1 07-8 ; heat stress

Age of the Dinosaurs, I I , 1 2

and walking upright, 1 07; and

agriculture: discovery of, 1 70; see also

intensionality, 9 1 , 1 04; in marginal


habitats, 1 4, 1 5 , 1 06, 1 07, 1 09 ;

Aiello, Leslie, I I I , 1 1 3 , 1 24, 1 90

monkeys out-compete, 1 3 , 1 4, 1 5 ,

'alliance fickleness', 2 5

1 06; neocortex size and group size,

alliances: o f apes, 1 9, 2 1 , l S I ; coali�

6 3 ; and observation of others'

tion size, 66, 67; and grooming, 20,

behaviour, 79; Old World, I I , 3 5 ,

2 1 , 22, 3 5 , 44, 4 5 , 46, 6 5 , 67, 6 8 ;

I 1 5 , 1 40; and predation, 1 09 ; pub­

and harassment, 4 3 -4; of monkeys,

lic behaviour, 7; social knowledge,

I 9 , 2 1 , I F ; with neighbouring

60; social world, 1-3 , 4 , 9 , 28; and

groups, I I 9 ; and primates, 1 9, 2 3 ,

theory of mind, 90-9 1 ; vocaliza­

4 3 -4, 66; and rclative� 1 66; and

tions, 1 3 5 , 1 40; and vowcl sounds, 14 I; see also chimpanzees; gibbons;

social inference, 22-3 altruism, 1 64 , 1 6 5 ; reciprocal, 1 9 5

gorillas; orang-utans

Amboscli National Park, Kenya, 2 1 , 101

Aristotle, 3 0 arithmetical skills, 5 5 , 9 0

American Indians, I 1 7

Arnhem Zoo, 24, 1 5 1

Andersson, Malte, 1 89

art objects, I I O

animals: behaviour, 3 4 ; caged, 3 7; and familiarity, 1 68 ; and language, F;

as machines, 82, 8 3 ; and the mind,

artificial keyboard language, 5 3 , 9 3 A S L (deaf-and-dumb language), 5 2 Asperger's Syndrome, 8 8 , 8 9 , 90

8 2; and mirror test for self-aware­

Austen, Jane, 1 84, 1 8 5

ness, 9 1 ; tactical deception by, 9 2-5

Australian Aboriginals, 7 1 , I I 7,

antelope, 3 1 , 1 08 , 1 26, 1 7 6, 1 80

1 5 3-4

anthropology, 1 9 2

a ustralopithecines, 1 1 3 , 1 27, 1 3 0

apes: ability t o calculate the likely

autism, 8 8-90, 9 5 , 1 00

effect of their actions, 2 5 ; alliances, 1 9 , 2 1 , 1 5 1 ; as anthropoid primates,

- 218 -

Babel, Tower of, 1 5 2-3 , 1 5 4, 1 70

. - 219 -

B ibliography

Grooming, Gossip and the Evolution of Language 3 4 , 1 6 3 ; infra-organismic, 2 8 ; mole­

babies: brain growth after birth, 1 28 ;

cular, 28, 69; organismic, 28

first words, 3 ; incapable of fending

birds, 57, 1 27, 1 4 2, 1 7 6

for themselves, 1 28 , 1 29 ; as 'not

birth, and excited contact calls, 4 9

fully human', 8 8 ; premature, 1 29

Bloom, Sam, 6 7

baboons: ability to eat unripe fruit, 1 4; and alliances, 22; and brain

body language, 8 1 , 1 77-8 3

size/group size, 94

bonding, 7 8 , 1 1 5- 1 6, 1 20, 1 23 , 1 3 0, 1 4 7, 1 4 8, 1 49 , 1 7 3 , 1 90, 1 9 1 , 1 9 2,

chacma: and exploitation, 23; and

203 ; female, 1 4 8-5 1

reconciliation, 26 as fruit-eaters, 60, 1 26-7

books, and gossip, 5-6

gelada ( bleeding heart ) : alliances,

brain: assymmetry and language, I I I ,

1 9-2 1 ; contact calling, 5 4 ,

1 3 6; brain volume doubles, 1 90,

1 40-4 1 , 1 5 0-5 1 ; grooming, 4 9 ,

1 9 1 ; consumption of energy, 3 , 1 2 3-5 , 1 7 5 , 1 9 2-3 ; mammalian,

6 8 , 1 1 4, 1 1 6, 1 90; a n d groups, I I 8, 1 5 0; open habits, 1 1 8; and

6 1 -2; and music, 1 3 9-40; and opi­

predation, I I 8 ; and reconcilia­

ates, 3 6, 3 7, 1 4 6; primitive, 6 1 ;

tion' 2 6-8; reproductive suppres­

right hemisphere specialized for

sion, 4 1-2; sensitivity to risks,

processing emotional information,

2 5 -6; social structure, 24; and

1 3 6, 1 3 8, 1 3 9 ; speech localized in

tightly bonded harems, 92, 1 79 ;

the left hemisphere,

I 3 6,

1 3 8-9

and visual/emotional cues, 1 3 7;

Brain, Rev. Chris, 202

vocal exchanges, 4 9 , 9 2-3 ; and

brain size: animals with bigger brains than humans, 5 6; at birth, 1 28 ; and

vowel sounds, 1 4 1 and grooming, 2 , 22; and groups,

coalition size, 66; dinosaur, 3; and

77, 1 1 8 ; gruntin� � 2� 4 6

energy, 3 2, 5 8 , 5 9 ; and free-riders,

hamadryas: exploitation, 23-4; as

1 7 5 ; hominids, 1 2 3 , 1 2 7; of hom*o

nomads, 1 1 9 ; social structure, 24;

erectus, 1 1 2; human, 3, 5 6, 5 8 , 1 0 9 ,

and tightly bonded harems, 92

1 1 1 , 1 24, 1 2 8, 1 3 6, 1 7 5 , 1 90, 1 9 1 ,

and intensionality, 9 5 ; and leop-

1 9 3 ; measured against body weight,

ards, 1 8 ; and a marginal habitat,

5 6-8, 5 7; and meat-eating, 1 27-8;

3 9; movement of, 1 09 ; as 'new­

Neanderthal, I I 7

comers', 1 5 ; as Old World mon­

need for bigger brain, 5 8 , 1 29 ; eco­ logical hypothesis, 59, 60;

keys, 1 3 ; olive, 1 79 ; and stress, 40 Bachmann, Christian, 1 79

Machiavellian Intelligence

barking, 4, 2 1 , 5 0

hypothesis, 60, 64, 66, 68, 94, 9 5

Barton, Rob, 3 6, 6 5

neocortex size and group size, 62,

bats, 1 2, 64- 5 ; vampire, 6 5

6 3 , 64-6, 6 8-9, 76, 94, 1 1 0,

bee-eaters, 4 5

1 1 1 - 1 2, I I 3 , 1 9 2; of porpoises and

bees: honey, 50-5 I ; worker, 1 6 6

dolphins, 3 ; prosimian increase in, 12

behaviourism, 8 3

British Broadcasting Corporation

belief/desire psychology, 8 7


Bentall, Richard, 8 4 Bever, Thomas, 1 3 9

Bronx Zoo, New York, 4 2

Biblical creation story, belief in literal


truth of, 29

I I O- I I

Bush, George, 1 44

binocular vision, 13 6n

business community, 203-4

biographies, 5-6

Buss, David, 1 8 6 Byrne, Dick, 2 3 , 60, 9 2, 9 3 -4, 9 5

biology: cell, 2 8-9; evolutionary, 3 3 ,

- 2 20 -

caged animals, 3 7

tactical deception, 9 3 , 94, 9 6; and

canine teeth, 1 3 0

theory of mind, 9 5 , 98, 1 00, 1 02

Cann, Rache� 1 6 1 , 1 6 2

Chinese language, 4 7-8, 1 5 7

Canterbury Tales ( Chaucer), 1 5 2

clans, 70-7 I

career women, 4 3

Clark, Amanda,

carnivores, 5 8 , 5 9 , 6 5-6, I I I , 1 27

class, a n d marriage, 1 8 5-6

Casperd, Julia, 1 3 7

climatic change, 1 2, 1 3 , 1 5 , 3 1 -2,

caste membership, 1 67

I 65

1 06, 1 60

cats, 1 7, 5 0, 5 8 ; sabre-toothed, 1 6

coalitions, see alliances

Cavalli-Sforza, Luigi, 1 6 2

cognitive predispositions, 207

cave paintings, 1 1 0

committees, 1 9 6-7

cell biology, 2 8-9n

communication: between animals 5 0 ''

Chamberlain, Bill, 80

and businesses exceeding opti

Charles, HRH The Prince of Wales,


size, 205 ; the electronic highway,

1 78

204; language as, 8 1 ; non-linguistic

Chaucer, Geoffrey, 1 5 2

transforms into language, 1 3 2

cheetahs, 1 7

community: business, 203-4; close­

Cheney, Dorothy, 2 1 , 26, 46-7, 4 8 ,

knit, 1 8 , 200-20 1 , 202; lack of a

6 8 , 1 0 1 , 1 1 5 , 1 40

sense of, 1 99-20 3 ; and language, 7

Chiarello, Robert, 1 4 0

competition: for food/water resources,

children: and belief/desire psychology,

I I 8 ; and human group size, 1 1 8 ;

87; and dolls, 86, 8 8 ; language

and reproduction, 4 3

development, 3, 7, 5 3 -4 ; and lying,

conferencing calls, I 9 7

8 5 ; and Piaget's theories of develop­

consciousness, 1 3 9

ment, 8 7-8; rearing, 1 8 7 ; and the

contact calling, 46, 4 9 , 5 4 , 1 1 5 ,

social world, 87; and a theory of

1 4 0-4 1 , 1 5 0-5 1 continental masses, 3 I

mind, 8 5-6, 9 8, 1 00

Chimp Talk ( BBC Horizon film), 1 00

conversation, 1 3 2, 1 7 1 ; audibility of,

Chimpanzee Politics ( de Waal ) , 24

1 2 2-3 ; dyads (pairs ) , 1 94 , 1 9 5 ; and

chimpanzees: and alliances, 22, 24- 5 ,

exchange of social information,

1 5 1 ; attempts to teach English to,

1 20, 1 23 ; and gossip, 4-5 ; groups

5 1 -4; bonobo (pygmy ), 1 0, 5 3 ; and

contain only one speaker, 1 2 1 , 1 9 4;

brain size/group size, 69; common,

interpretation of, 89; and lekking,

1 0; and communication through

1 76, 1 9 5 ; male/female, 1 7 5-7,

language, 1 0 5 ; and distinguishing

1 9 4-5 ; networking, 1 77; size of

between knowledge and ignorance

groups, 1 20-23 , 1 94, 1 9 5-9; timing

in other individuals, 9 8-9; and false

of, 1 4 1 ; and vigilance behaviour,

belief test, 99-1 00; females, 1 4 8 ; as fruit-eaters, 60, 1 2 6-7; in groups,

1 77-8 ; as vocal grooming, 7 8 , I I 5 cortex, 6 1

22, 77, I I 2, I 20, 1 2 1 ; and inten­

Cosmides, Lida, 1 72, 1 7 3

sionality, 97; lacking vocal appara­

courtship rituals, 3 8-9, 4 5-6

tus for language, 5 1 , 1 1 6- 1 7; and

cows, 1 26

marginal habitat, 3 9, 1 1 8 ; and mir­

'coy signal', 1 7 8-9

ror test, 9 1 ; movement of, 1 09 ; and

Cro-Magnon peoples, 1 1 3 , 1 1 6, I I 7,

predation, 1 6, 1 8 ; and pretend play, 9 5 ; and reconciliation, 26; shares a

118 crowd effect, 1 4 3

common ancestor with humans, 1 0,

crowding, 1 8-1 9

3 3 ; and sign language, 5 2; and

crows, 1 5 8

- 221 -



Grooming, Gossip and the Evolution of Language


dabchicks, 4 5

endogenous opiates, 3 6, 4 1 , 1 0 3 , 1 90

tation of, 1 26; finding, 1 6, 3 5 , 1 ° 7,

Grimm, Wilhelm, 1 5 4, 1 62

Daly, Martin, 1 6 5

endorphins, 3 6, 3 7, 4 1 , 1 90

I I 9 ; and groups, 3 9-40; stealing, 1 8

Grogan, Captain Ewart, 1 8 1

dance, 1 4 2, 1 4 3 , 1 4 6-7

energy: brain's consumption of, 3, 5 8 ,

Darling, Helen, 1 7 8 Darwin, Charles, 2 8 , 29, 8 2, 1 8 8-9

Fossey, D ian, I I

5 9 , 1 2 3-4; and fruits, 1 27; and

fossil foot ( from southern Africa ) , 1 0 9

groups' feeding, 40

free-riders, 44-5, 7 8 , 1 4 6, 1 63 , 1 67,

grooming, 3 9; and alliances, 20, 2 1 , 22, 3 5 , 44 , 4 5 , 46, 6 5 , 67, 6 8 ; and apes, 4, 2 1 , 65, 77; and bats, 6 5 ;

1 68 , 1 7 1 -� 1 74 , 1 7 5 , 1 9 5 , 202

Darwinian biology, 7

Engel, Claudia, 1 8 5

Darwinism, 3 3

English language, I 5 7-8

Frisch, Karl von, 50

dating agencies, 202-3

enkephalins, 3 6

fruit-eaters, 5 9 , 60-6 1 , 1 2 5 , 1 26-7

Datta, Saroj, 2 5

Enquist, Magnus, 44, 4 5 , 1 63 , 1 7 1 -2

fur, 1 07-8

Dawkins, Richard, 1 0, I I , 3 2n

Eskimo ( Innuit) peoples, 1 6o, 1 8o

deaf-and-dumb language, 5 2

Etemenanki, 1 5 3

G factor, 5 5

death a s a concept, I I I

ethologists, 5 9

galagos, 1 2, 22, 3 6, 5 8 , 9 4

decision-making, 3 4

eugenics movement, 3 3

Gallup, Gordon, 9 1-2

'deep croak', 1 44 , 1 4 5-6

EV� 9 , 1 0 , I I , I � 1 6 1

Galton, Francis, 3 3

deep-sea creatures, 3 1 -2

evolution, 1 6, 29-3 0, 1 7 5 ; and diver­

Gardner, Alan, 5 3

conciliatory, 2 6 ; conversations as, 78, I I 5; and groups, 68, 77, 1 1 0, 1 1 2- 1 3 , 1 1 4, 1 20, 1 2 1 , 1 5 1 , 1 9 2; and grunting, 2, 26, 49; and humans, 77, 1 3 6, 1 4 6, 1 4 8 ; and hygiene, 2 1 , 3 6, 44; language as vocal grooming, 78, 79, 1 9 2; and monkeys, 1 , 2, 4, 2 1 , 3 6, 3 8 , 6 5 , 6 8 , 77, I I I ; Neanderthal, I I 7; and neocortex size, 68; and opiates, 3 6,

deer, red, 1 4 4

sity of life, 29-3 0; human, 7; neo­

Gardner, Trixie, 5 2-3

Denman, Jim, 1 3 8

Darwinian theory of, 1 8 9 ; of pri­

geladas, see under baboons

Dennett, Dan, 8 4

mates, 1 5 ; as a series of

general knowledge, 5 6

and reconciliation, 26; and scream­

Descartes, Rene, 8 1 -4

compromises, 1 9 3 ; theory of evolu­

gene ( s } : a s the fundamental unit of

ing, 2 1 -2; vocal, 78, 1 1 5 , 1 90

tion by natural selection, 28-3 4

dialect, 70, 1 5 1 , 1 5 7, 1 5 8 , 1 62-3 ,

evolutionary biology, B , 3 4 , 1 6 3

1 6 8-70 D iana, HRH The Princess of Wales, 178

genetic codes, 1 6 1

exploitation, 23-5 , 1 4 6

genetic determinism, B , 3 4

diving, a n d courtship rituals o f ducks

facial expressions, 2, 20, 1 3 7, 1 3 8, 191

DNA, 28n, B , 1 6 1

false belief tests, 8 5-6, 8 8 , 89, 99- 1 00

DNA fingerprints, 1 6 7

family trees, 7 I

dog� 5 0 , 5 � 5 8 ; hunrin& 1 � 1 8 , 6 5

farming, development of, 1 5 9; see also

dolphins, 3-4, 1 3 7


Donner Party wagon train, 20 1

fatwa, 1 4 3

ducks, 3 8-9

feeding: and courtship rituals, 3 8 ,

Dukakis, Michael, 1 44

4 5-6; and predation, 1 6, 5 9 ; and time budget, 1 6, 77

eagles, 4 8 ; monkey-eating, 1 6 earth: distance to the sun, 3 I ; tilt of axis, 3 I

groups: apes, 7 1 , 1 1 2; and bonding, 7 8 , 1 1 5- 1 � 1 2� 1 2 3 , 1 4 8, 1 49 , 1 9 1 , 1 9 2; a n d centripetal/centrifu­ gal forces, 1 9 ; chimpanzees, 22, 77,

genetic sex, 1 3 6

I 1 2, 1 20, 12 I ; church congrega­

genetic traits, 3 3

dinosaurs, 3 , I I , 1 2, 3 1 , 5 7 and grebes, 3 8

evolution, 3 2; mitochondrial, 1 6 1 ; and reproduction, 1 6, 3 3 , 1 63

evolutionary change, 3 2 eye contact, 1 7 8 , 1 7 9

Dickens, Charles, 1 64

3 8 , 44, 77, 1 90-9 1 , 1 9 2; primates, 9 , 44, 67, 68 , 77, I I I , 1 1 4 , 1 20;

tions' 74; clans, 70-7 1 ; and conver­

genetics, 66

sation, 1 20-2 3 , 1 94 ; females' role,

gestation periods, 1 2 8

1 4 8-9 ; free-riders, 44- 5 , 1 7 5 ; and

gestural theory, 1 3 2, I B , 1 3 5 , 1 3 6,

grooming, 6 8 , 77, 1 1 0, I I 2-1 3 ,

1 4 0, 1 4 1

I I 4 , 1 20, 1 5 1 , 1 9 2; group size and

giant gelada, 1 27-8

appearance of language, I I I , I I 2;

gibbons, I I , 9 1 , 1 3 0, 1 3 4 , 1 4 2

group size and neocortex size, 62,

Gombe National Park, Tanzania, 1 6,

6 3 , 64-6, 6 8-9, 76, 94, 1 1 0, 1 1 2,


1 1 3 , 1 9 2; group size and social

gonadotrophic releasing hormone

complexity, 62-4, 1 7 5 ; groups of

( GNRH}, 4 1

1 5 0, 69-7 7, I I 3 , 1 2 1 , 146, 1 99 ,

fiction, 5 ; and third-order ToM, 1 0 2

Goodall, Jane, 1 6, 9 6

200, 206; hierarchical in structure,

Fieldman, George, 1 5 0, 1 64

gorillas, I I , 6 9 ; a n d mirror tests, 9 1 ;

72, 74, 20 5-6; human group size,

fimbulvinter, 1 5 4

education, and large classes, 1 9 8-9

Fisher, Ronald, 1 8 9

Einstein, Al bert, 5 5

Fisher's 'Sexy Sons Hypothesis', 1 8 9 ,

and predation, 1 6

1 1 3 , 1 1 5- 1 6, 1 1 8- 1 9, 1 20, 1 2 1-2,

gossip, 4-7; and evolution o f lan­

1 2 8 , 1 4 8, 1 9 2; hunter-gatherer,

guage, 7 9 , I I 5, 1 9 3 ; and free-riders,

I 1 9-20; military, 7 5-6; monkeys,

1 90

1 7 I -2; level in conversations, 1 2 3 ;

7 1 ; Neanderthals and, I I 7; and pre­

electronic mail, 20 5

fishes, 5 7

and the management o f reputation,

dation, 1 7- 1 8 , 1 9, 3 9, 4 3 ; primates

elephants, 5 6, 9 1 , 1 28 , 1 80

fission-fusion social system, 1 20

1 2 3 ; negative, 1 74

Eliot, T. S., 7

Fleagle, John, 7 2

elite dominance, 1 6o

Flood, Josephine, 1 5 3

grammar, 3 , 7, 50, 5 3 , 8 0, 1 3 2 Great Chain o f Being, 3 0

in, 1 8 , 1 9 , 3 9-40, 69,

Emler, Nicholas, 1 2 3 , 1 73

flying foxes, 1 2

grebes, 3 8-9

emotional cues, 1 3 6-7, 1 3 8

food: competition for, 40; and

Greek chorus, I S o

eland dances, 1 4 7

endocrinology, 1 9 2

decreasing gut size, 1 2 5-7; fermen-

- 222 -

I I O,

I I 2,

1 1 8 , 1 20, 1 2 1 , 1 9 0, 1 9 2; problems of, 3 9-40; sympathy group size, 76; tensions of living in, 1 8- 1 9 grunting, 4 , 5 0; conciliatory, 2 6 ; dif­

Grimm, Jakob, 1 5 4, 1 62

ferent kinds of grunt, 47; and differ-

- 223 -


Grooming, Gossip and the Evolution of Language dramatic change in appearance over

ent types of predator, 4 8-9, 1 40;

a short period, 3 I; flattened chest

and grooming, 2, 26, 49; monkeys

of, 1 3 4 ; giving way to bigger peo­

and apes, 2, 26, 4 6-7; and

ple, 1 44 ; and grooming, 77, 1 3 6,

Neanderthals, I 1 7

1 4 6, 1 4 8 ; and group size, I I 3 ,

guenons, 1 5 Gustavus Adolphus, King of Sweden,

1 1 5- 1 6 , 1 1 8- 1 9, 1 20, 1 2 1-2, 1 2 8 ,

75 gut: and size o f bra i n, 1 24-5 ; size of,

109; importance of height, 1 44-5;

1 4 6 , 1 4 8 , 1 9 2; hairlessness, 1 07-8, increase in stature, I I 8- 1 9 ; and

1 24, 1 2 5 , 1 27

intensionality, 1 02; learning process in, 1 29 ; and Neanderthals, I I 7;

hairlessness, 1 07-8, 1 0 9

nomadic, 1 1 7; and predation, I I 9 ;

Hamilton, Bill, 1 6 3

second phase o f evolution, 1 1 9 ;

Hamilton's Rule, 1 6 3 , 1 64 , 1 6 8

sharing a common ancestor with

Happe, Francesca, 8 9

chimpanzees, 10, 3 3 ; size of gut,

harassment, 40-44

1 24 , 1 2 5 ; social world, 2-3 , 4, 9,

Hare Krishnas, 202

28- as t h e 'third chimpanzee', 1 0;

l-Iauser, Mark, 1 3 8

vo al apparatus, 5 I; walking

Hawkes, Kristin, 1 80

upright, 1 0 8-9; see also hominids

Hayes family, 5 1 , 9 5

hunter-gatherers, 69-7 1 , 77, 1 1 6,

hearing system, 3 9

1 1 8, 1 1 9-20, 1 3 1 , 1 5 6, 1 5 9, 1 70,

herbivores, 1 3

1 8 0, 1 8 6, 200

Herodotus, I 5 3

hunting, 1 27-8, 1 3 4, 1 4 8 , 1 49, 1 5 0,

'heroic tasks', 1 8 0-8 2

1 8 0-8 2, 1 9 1

Hill, Kim, 1 8 6

Hurst, N icola, 1 6 5

Hitler, Adolf, 1 4 5

Hutterites, 7 2

Hockett, Charles, 5 0

hyenas, 1 6

hominids, I I I ; evolution o f large

hypergamy, 1 8 5 , 1 8 8

brains in, 1 2 3 , 1 27; and food

hypothalamus, 3 6, 4 I

sources, 1 09 , 1 27; fossil skulls,

1 1 1- 1 2; grooming, 1 1 2, 1 1 4, 1 1 4; and group size,


immune system, 40

1 2; neocortex size

and group size, 1 1 1 - 1 2, 1 1 3 ;

Indo -Europeans, 1 5 5

reduced sexual dimorphism, 1 3 0;

infertility, 40-43

reduction of gut size, 1 2 5 , 1 27; and walking bipedally, 1 0 8-9, 1 3 4; see

information superhighway, 204, 205 infra-organismic biology, 2 8 n insectivores, 1 2, 5 8

also humans

insectivorous diet, 1 2 5

hom*o genus, 1 5 , 1 1 2, 1 1 4, 1 27, 1 3 0 hom*o erectus, 1 1 2- 1 3 , 1 1 3 , 1 1 5 , 1 1 9

intelligence: defining, 5 5 ; measuring,

hom*o habilis, 1 1 3

5 5, 5 6 intensionality, 8 3 - 5 , 90, 9 5 , 9 6, 97,

hom*o sapiens, 69, I I I , I I 2, 1 1 3

98, 1 02, 1 04, 1 3 8, 1 7 1

honey bees, 5 0-5 1

international money markets, 204

'huma n interest' stories, 6

Internet, 204

humans: ancestry as apes, 7, 1 5 ; and

isometric skiing exercise, 1 64-5

apes' marginal habitat, 1 5 ; appar­ ently unique ability to use language,

2-3 , 49-50; behavioural flexibility,

j aw, the, 3 9

207; brain, see brain, brain size;

Jaynes, Julian, 1 3 9

common female ancestor, 1 6 1 -2;

Jerison, Harry, 5 6, 5 7

- 2 24 -

jihad, 1 4 3

size and appearance of language,

jogging, 3 7


Johanson, Don, 1 0 8

I I,


1 2; history of language evolu­

tion, 1 5 8-6 1 ; history of, 1 5 3 ,

Johnson, Magic, 1 8 1

1 5 4-6, 1 70, 207; and hom*o genus,

Jones, Sir William, 1 5 4

I I 4; and honey bees' dance, 5 0- 5 1 ;

Kapanora tribe, New Guinea, 1 1 6

to use, 2-4, 4 9- 5 0; as inadequate at

humans' apparently unique ability

Kaplan, Hilly, 1 8 6

the emotional level, 1 4 7-8, 1 9 3-4;

Kellogg family, 5 I

as learned, 5 0; location in the rrain,

Kenya, Mount, 1 0, I I

see under speech; Neanderthals and,

Khan, Genghis, 1 60

1 1 6- 1 7; networks, 1 77, 1 9 2; ? J

Kimura, Motoo, 3 3

non-iconic, 5 0; as permeating

kin selection, 1 6 3

human culture, 207; pictorial, 5 2-3 ;

Kinderman, Peter, 84

and propaganda, 1 7 1-7; reconstruc­

kingfishers, 45

tion of long-dead, 1 5 7-7; as refer­

kinship, see relatedness

ential, 5 0; and reputation manage­

Kipsigis agro-pasturalists, Kenya, 1 8 6

ment, 1 7 3-4; self-publicity, 1 7 3 ,

Knight, Chris, 1 4 7, 1 4 9 , 1 5 0

1 7 5 , 1 9 2; and sharing o f informa­

Koresh, David, 202

tion, 7, 78-9, 1 20, 1 2 3 , 1 7 3 , 1 9 �

Kudo, Hiroko, 6 7

1 9 3 ; sign, 5 2; and social cheats,

Kummer, Hans, 2 3 , 9 2, 1 79

1 7 2-3, 1 74, 1 9 2; as a social tool,

! Kung San people, I I 9-20

1 70; song theory of origins, 1 3 2, 1 4 2-4; symbolic, 1 1 6; as syntacti­

Laetoli footprints, Tanzania, 1 0 8-9

cal, 5 0; translating from hearing to

Lamarck, Jean, 3 0, 3 1 , 3 2, 3 3

speaking, 5 3 ; and tribal dialect, 70;

language: and abstract relationships,

vocal apparatus necessary for

5 3 ; and advertising, 1 5 1 , 1 77,

human language, 5 1 , 1 1 6- 1 7; as

1 80-8 2, 1 8 5 , 1 8 7, 1 9 1 , 1 9 2, 1 9 3 ;

vocal grooming, 78, 79, 1 9 2; vocal­

and appearance o f hom*o sapiens,

ization theory of origins, 1 3 2,


I 2; attempts to teach chimpanzees

1 40-42; vowel sounds, J 4 1 ; see also

to talk, 5 1-4; and autism, 8 8 , 8 9 ;

poetry; song

beguiling, 2 0 2 ; body, 8 1 , 1 7 7-8 3 ;

laughing, 1 8 2-3 , 1 9 1

and bonding, 7 8 , I I 5- 1 6 , 1 20, 1 2 3 ,

leaf-eaters, 5 9 , 1 2 5-6, 1 27

1 47, 1 4 8, 1 49 , 1 7 3 , 1 9 1 ; and brain

Leakey, Mary, 1 09

assymetry, I I I, 1 3 6; Chinese, 47-8,

Leakey, Richard, 1 1 0

1 5 7; 'click ', I I 9 ; a common, 1 5 2-4 ;

learning, 3 4

as communication, 8 1 , 1 04- 5 ; and

Leimar, Otto, 4 4 , 4 5 , 1 6 3 , 1 7 1-2

co:nmunity, 7; development in chil­

leks, 1 7 6, 1 9 5

dren, 3 , 7; dialect, 70, 1 5 1 , 1 5 7,

lemurs, 1 2, 2 2, 3 6, 5 8 , 6 1 , 94

1 5 8 , 1 6 2-3 , 1 6 8-70; diversification,

leopards, 1 6, 1 7, 1 8 , 4 8 , 49

1 3 2, 1 5 7; earliest, 8 , 1 3 2; English,

Leslie, A l a n , 88, 95

1 5 7-8; evolution of, 8, 79, I I 4-1 5 ,

leutenizing hormone (LH), 4 I

1 20, 1 3 2, 1 74, 1 9 2-3; exchanging

Lieberman, Philip, I I 6- 1 7

information about free-riders, 1 7 3 ;

life, diversity of, 29-3 0

extinct languages, 1 5 7; genuine ver­

'life force', 3 1

bal, 5 0; gestural theory of origins,

linguists, 49-50, 5 2

1 3 2, 1 3 3 , 1 3 5 , 1 3 6, 1 4 0, 1 4 1 ; and

lions, 1 6, 1 7, 1 8 , 1 9 , 6 5 , 1 80, 1 8 1

gossip, 6-7, 79, I I 5, 1 9 3 ; group

literature, 1 0 2 , 1 7 1

- 225 -


Grooming, Gossip and the Evolution of Language Miller, Geoff, 1 8 2, 1 90

logical reasoning, 5 5

mind: and animals, 8 2 ; and cheating

Lonely Hearts columns, 202-3

on social agreements, 1 7 2; and

lorises, 1 2

communication, 8 1 ; and cultural

Lowe, Catherine, 1 3 8

evolution, 1 9 3 ; as a number of sep­

Lucy ( Ethiopian hominid ), 1 08, 1 09

arate modules, 6 1 ; and theory of

lying: and autism, 8 8-9 ; children and,

mind, 90-9 1

88 Maasai warriors, 1 4 2, 1 4 5 , 1 8 1

mitochondrial DNA, 1 6 1 , 1 6 2

macaques: ability to eat unripe fruit,

molecular biology, 2 8 , 69

1 4; and alliances, 2 2; and groom­

'molecular clock', 1 6 1

ing, 26; and a marginal habitat, 3 9 ;

molecular genetics, 1 5

as 'newcomers', 1 5 ; as Old World

monkeys: baby, 1 2 8 ; able to calculate the likely effect of their actions, 2 5 ; alliances, 1 9 , 2 I , 1 5 I ; a s anthro­

McCann, Colleen, 4 2 McFarland, David, 1 64

poid primates, 1 2, 63; brain

Machiavellian Intelligence hypothesis,

size/body weight, 5 8, 59; chest shape, 1 3 3 -4; colobine, 3 9, 1 2 6

60, 64, 66, 68, 94, 9 5 mammals: behavioural flexibility, 207;

colobus: and alliances, 22; fermen-

brain sections, 6 1-2; brain size/

tation of food, 1 2 6; and neocor­

body weight, 5 7-8; as food, 1 27;

tex/group size, 94; as an Old World monkey, 1 3 , 22; and pre­

jaws of, 39; neocortex, 62; primates as one of the oldest lineages of,

dation, 1 6, 1 8

I I;

reptilian ancestors, 3 9; and seed dis­

common ancestor o f monkeys and

persal, 1 3 - 1 4 ; social lives, 66

apes, I I ; comparing understanding

Manning, John, 1 3 8 Manson, Charles, 202 Maoris, 1 4 2-3 , 1 4 5 marathon runners, 3 8

of another's intentions with that of apes, 9 8-9; contact calls, 46, I I 5 ; dominating the forest, 1 4 , 1 06; as fruit eaters, 5 9 ; use of gestures, 1 3 5 ;

marginal habitats, 1 4, 1 5 , 3 9

and grooming, 1 , 2, 4 , 2 1 , 3 6, 3 8 ,

marmosets: a s highly vocal, 4 6 ; and

6 5 , 68, 77,

suspended reproductive state, 4 2 ,

I I I;

and groups, 7 1

howler; and fermentation o f food, 1 2 6; and intensionality, 9 5 ; neo­


cortex/group size, 94

marriage, a n d class, 1 8 5-6 Marriott, Anna, 1 74

and intensionality, 9 1 ; langur, 1 3 ,

marsupials, 5 8

1 26; leaf, 5 8 ; monkeys out-compete

Marx, Karl, 1 0. 3

the apes, 1 3 , 1 4 , 1 5 , 1 06; neocortex

mass extinction, 3 I

size and group size, 6 3 ; New World,

mate selection, 1 79-8 2, 1 8 3-8

3 9 , 94, 9 8 , 1 2 6; and observation of

matrilineal relatives, 9, 1 0, I I , 1 6 1

others' behaviour, 79; Old World,

'mega-bands', 70

1 1 , 1 3 , 1 4 , 1 5 , 2 2, 3 5 , 46, 9 1 , 9 4 , 9 8 , I I 5 , 1 2 6, 1 40, 1 4 1 ; public

menstrual cycle, 4 I, 42

behaviour, 7 ; rhesus, 2 5 ; and seed

meat-eating, 1 27-8

mental states, 8 3 , 84

dispersal, 1 3- 1 4; size of gut, 1 2 5 ;

Menzel, Emil, 9 6

social knowledge, 60; social world,

mice, 1 2 8

1 -3 , 4, 9, 2 8 ; South American, 1 3 ,

mid-brain, 6 1

22; and theory o f mind, 90-9 I ,

migrations, major, 1 5 8-60, 1 6 2

1 00- 1 0 1

- 226 -

poor psychologists, 1 0 1 ; and pre­

'nuclear winter', 3 I numbers, 5 3

dation, 1 6, 4 8 ; sounds of, 47, 1 40, 1 4 1

O'Connell, Sanjida, 9 9 - 1 00

vocalizations, 1 3 5 , 1 40; and vowel

oestrogen, 4 I

sounds, 1 4 1 ; see also baboons;

Okavango swamps, Botswana, 2 6

guenons; macaques; marmosets;

Olorgasaillie site, Kenya, 1 27


mirror test, 9 1 -2

monkeys, 1 3 , 9 4 ; sounds, 1 4 1 , 1 5 8

vervet: and grooming, 2 1 -2, 6 8 ; as

On the Origin of Species ( Darwin), 2 8

monogamy, 1 3 0, 1 3 1 Moonies, 202

opiates, 3 6, 3 7, 3 8 , 4°, 44, 77, 1 0 3 , 1 4 6-7, 1 4 8 , 1 9 0, 1 9 2 , 202

Morgan, Elaine, 1 9 3

opium, 3 6

Mormons, 7 3


morphine, 3 6

and mirror problems, 9 I

mother-daughter relationships, 2, 9 ,

organismic biology, 2 8

1 8 , 2 5-6, 1 6 1


The Origin of Consciousness in the

'mouse-to-elephant' curve, 1 28

Breakdown of the Bicameral Mind

Mulder, Monique Borgerhoff, 1 8 6

(Jaynes), 1 3 9

murder of relatives, 1 6 5-6

oxygen, two forms of, 1 2

music, 1 3 9-40, 1 4 8 pain, 3 6-7, 40 Narikitome Boy, 1 1 0

pair-bonding, 1 3 0, 1 3 1

natural selection, 2 8 , 3 0, 3 2, 3 8 ,

Panter-Brick, Catherine, 1 66

5 8-9 , 1 8 9; and reproduction, 3 0,

paternalism, Victorian, 1 97

3 3 -4

patrilocality, 1 4 9

Nazism, 3 3

Pawlowski, Boguslaw, 9 4 peaco*cks, 1 76, 1 8 8, 1 8 9-90

Neanderthal peoples, 1 1 3 , I I 6- I 7 neo-Darwinian theory o f evolution,

pedigrees, 7 1 , 1 6 1

1 89

Petrie, Marion, 1 8 9

neocortex: described, 6 1 -2; as the

photographs of people, 1 3 7-8

'thinking' part of the brain, 6 I

Piaget, Jean, 8 7-8

neocortex size and group size, 62, 6 3 , 64-6, 6 8-9 , 76, 9 4 , 1 1 0, 1 1 1 - 1 2 ,

pictorial languages, 5 2-3 pituitary gland, 4 I

1 13, 192

plankton, 1 2n

Nepalese hill-farmers, 1 66

Pleistocene period, 1 5 1

nervous system: of lower-ranking ani­ mals, 40; and pain, 3 6, 3 7

Plooij, Frans, 9 6 Plotkin, Henry, 1 5 0, 1 64

Nettle, Dan, 1 2 2, 1 6 8, 1 70

poetry, 4 8 , 1 3 9, 1 40, 1 7 1 , 1 8 2; pro­

networks, 1 7 7, 1 9 2, 1 99, 200, 201-2, 2°3 , 2°4, 2° 5 , 206-7

duced by computer program, 80-8 I

The Policeman's Beard is

neurobiology, 1 9 2

Half-Constructed (RACTER com­

neutral selection, theory of, 3 3

puter program poems ), 80-8 1

newscasters, I 99-200

polygamy, 1 3 0, 1 3 1

newspapers, and 'human interest' sto-

porpoises, 3 , 9 I

ries, 6

Povinelli, Danny, 9 8 , 9 9

Nishida, Toshisada, 2 5 nomads,

I I7 ,

predation: attack distances of preda­


tors, 1 7; and feeding, 1 6, 5 9 ; and

non-fiction, 5

groups, 1 7- 1 8 , 1 9 , 3 9 , 4 3 ; and

Nostratic, 1 5 5-6, 1 5 7

grunting, 4 8-9, 1 40; and humans,

- 227 -


Grooming, Gossip and the Evolution of Language I I 9; primates and, 1 6, 1 8 , 1 9 , 3 9,

RACTER (computer program), 80

I I O; and reproduction, 1 6; and size

Ragnarok legend, 1 5 4

of prey, 1 6- 1 7 ; and stealth, 1 6, 1 7;

raiders, I I 8, I I 9

survival and, 1 6- 1 8 ; and visual

Received Pronunciation, 1 8 4, 1 8 8

cues, 1 3 7; see also under individual

reciprocal altruism, 1 9 5


reconciliation, 26-8

Red Dwarf (television programme),

pregnancy: length of, 1 2 8 ; and opi-

1 67

ates, 3 8 Pre mack, David, 5 3 , 9 8 , 1 0 5

relatedness, 1 6 3-8, 1 99, 200, 20 1 -2

pretend play, 8 8 , 9 5

Relativity, Theory of, 5 5

primates: and alliances, 1 9 , 2 3 , 4 3-4 ,

religion and science, 1 02-5

66; ancestral, 1 2; anthropoid,

Renfrew, Colin, 1 5 8, 1 60

1 2- 1 3 , 6 3 ; behavioural flexibility,

reproduction, 1 29; and genes, 1 6,


1 6 3 ; and natural selection, 3 0; and

brain: brain size/body weight, 5 7-8,

opiates, 4 1 ; and predation, 1 6 ; rela­

5 7; brain size/social complexity,

tives and, 1 6 6; resources for suc­

5 9-6o; neocortex ratio directly

cessful, 1 4 9-50, 1 84, 1 8 5 , 1 8 6,

related to total brain size, 6 3 ,

1 8 7; and stress, 40-43


reptiles, I I , 3 9 , 5 7 , 1 27

I I ; neocortex size a n d group

size, 62, 6 3 , 64, 6 5-6, 68-9 , I I O,

reputation management, 1 2 3 , 1 7 3 -4

1 1 2, 1 1 3

Richards, Jean-Marie, 1 5 0, 1 64

brain size at birth, 1 2 8 ; carnivores,

Richman, Bruce, 1 4°-4 1

66; and defence of food sources,

'risky shift', 1 4 3

I 1 8 ; distribution, 1 3 ; and effect of

ritual, I I 6, 1 4 2-3 , 1 47

stress on reproduction, 43; evolu­

'road rage', 204-5

tion of, 1 5 ; fruit-eating, 5 9, 60-6 I ,

rogue dealers, 204

1 2 5 , 1 27; grooming, 9 , 44, 67, 6 8 ,

Rumbaugh, Duane, 5 3

77, 1 1 0, I I I , 1 1 4, 1 20, 1 9 2;


rumination, 1 2 6 Rushdie, Salman, 1 4 3

groups, 1 8 , 1 9, 3 9-40, 69, I T O, 1 1 2, 1 1 8 , 1 20, 1 2 1 , 1 90, 1 9 2; growing physically bigger as a

'Sally and Ann' test, 8 6 , 9 9- 1 00

response to predation, 3 9 , I I O; leaf­

Sanskrit, 1 5 4

eating, 5 9 , 60, 1 2 5 -6; as one of the

Savage-Rumbaugh, Sue, 5 3 , 9 3 , 1 00

oldest lineages of mammals, I I ; and

scala natura ( 'scale of nature'), 3 0

predation, 1 6, 1 8 , 1 9 , 3 9 , I I O, l I 8 ;

Schumacher, A., 1 4 5

and social complexity, 62-3 ; social

science and religion, 1 0 2-5

world of, 9, 3 5 , 62, 66, 1 3 8 ; sociali­

scorpion flies, 4 5

ty of, 1 8 ; 'songs' of, 1 4 2; and tacti­

screaming, 4 , 6 7 ; a s exploitation, 2 3 ;

cal deception, 9 3-4 progesterone, 4 I prosimian primates, 1 2, 22, 3 6, 5 8 , 62, 94

and grooming, 2 1 -2; and Neanderthals, I 1 7 seed dispersal, 1 3-1 4

The Selfish Gene (Dawkins), 3 2

proto-World, 1 5 6-7

sexual dimorphism, 1 3 0

Provine, Bob, 1 8 2

sexual selection, 1 8 3 , 1 8 8-90; and

puberty, 4 1 , 4 2, 4 9

deep male voices, 1 4 6

pygmies, 1 49

Seyfarth, Robert, 2 1 , 26, 46-7, 4 8 ,

quantum physics, 29

Shakespeare, William, 7, 1 5 2

6 8 , 1 0 1 , I I 5 , 1 40

- 228 -

sharks, 3 I ,

I 37

'sympathy groups', 7 6 , 1 5 1

Shevoroshkin, Vitaly, 1 5 6 Sigg, Hans,


tachistoscope, 1 3 8

sign language, and chimpanzees, 5 2

tactical deception, 9 2-6

A Tale of Two Cities (Dickens), 1 64

Silk, Joan, 2 6 Simpson, O . J., 6

tamarins: as highly vocal, 4 6 ; and sus-

slum-dwellers, 20 1 - 2

pended reproductive state, 4 2

'small world' experiments, 7 3 , 74

tannins, 1 3 , 1 4

smiling, 1 8 2-3 , 1 9 1

telephone chat-lines, 1 9 7

Smith, Captain, 20 1

Thatcher, Margaret, Baroness, 1 4 5

Smuts, Barbara, 1 79

Theory of Mind (ToM), 8 3 , 8 5-6, 8 8 ,

snakes, 4 8 , 1 8 o

90-9 1 , 9 5 , 9 8 , 1 00, 1 0 1 , 1 02, 1 °4,

soap opera, 1 9 9

171, 193

social cheats, 1 7 2-3 , 1 74, 1 9 2, 1 9 5-6

Theory of Relativity, 5 5

social contact, lack of, 1 9 9-203

Through the Looking Glass (Carroll),

Social Darwinism, 3 3, 3 4


social inference, 2 2-3

throwing, 1 3 4-5 , 1 3 6, 1 3 9

social networks, 7 2-4, 1 5 0

Times, The, 6

social psychology, 1 9 2

toads, 1 4 4

social world: humans, 2-3 , 4, 9 , 8 7 ;

ToM, see Theory of Mind

monkeys and apes, 1-3 , 4, 9 ; pri­

tone-deafness, 90

mare� 9 , 3 5 , 6� 6� 1 3 8

tools, stone, I 1 0, I 1 6,

sociality, and primates, 1 8



trading, I 1 7

Societe de Linguistique, Paris, 1 5 5

tribes, 70

societies: hunter-gatherer, 69-70, 200;

trilobites, 1 3 7

post-industrial, 200 song, 80, 1 40, 1 4 2-4, 1 4 8 , 1 8 2

under-age drinking, 1 7 2-3

sound spectrograph, 4 7

ungulates, 5 8

Southern African Bushmen, 1 4 7 , 1 5 6,

Upper Palaeolithic Revolution, 1 1 0,

1 59


species: evolution of, 3 2; fate of all,

urban life, 1 9 9, 200, 201

3 0; Lamarck and, 3 0, 3 2 speech, 1 3 6; control o f breathing,

vampire bats, 65

1 3 3 ; fine motor control, 1 3 3 , 1 3 4 ,

Venus figures, l I O

1 3 5 , 1 3 6; location in the brain, 1 3 6,

verbal skills, 5 5

1 3 9 , 1 3 9-40, 1 9 3

Victorian paternalism, 1 9 7

Spencer, Herbert, 3 3

vigilance behaviour, 1 77-8

sperm banks, 4 3

Vikings, 1 6 5 , 1 6 6

sperm counts, 4 3

virtual conferencing systems, 1 9 7-8

Spoors, Matt, 1 5 0

Virunga Volcanoes,

Star Trek (television programme), 1 7 1

visual cues, 1 3 7



Stephan, Heinz, 6 5

visual problem-solving, 5 5

Stolba, Alex, 1 1 9

vocabulary, 3

stress: and harassment, 40; and living

vocal grooming, 7 8 , I l 5 , 1 90

in groups, 40; physical, 3 7; psycho­

Voland, Eckart, 1 8 5 , 1 8 6, 1 8 8

logical, 3 7, 40

vowel sounds, 1 4 1

sun, distance to the earth, 3 I survival, and predation, 1 6- 1 8

Waal, Frans de, 24, 26, 9 3 , 9 6, 1 5 1

- 229 -

Grooming, Gossip and the Evolution of Language walking upright: apes, 1 07; hominids, I oB-9, 1 3 4 Wason Selection Task, 1 7 2

wolves, 6 5 Woodruff, Guy, 9 B workaholics, 3 7-B

Waynforth, David, I B 5 weapons,



X-chromosome genes, 1 3 6, 1 4 9

whales, 5 6 Wheeler, Peter, 1 0 6-7, l o B , 1 24

Y -chromosome genes, 1 3 6, 1 4 9

Whiten, Andrew, 2 3 , 60, 9 2, 9 3 -4

Yerkes, Robert, s 3

wildebeest, 1 7- I B

'Yerkish' (computer keyboard lan­

Wilkinson, Gerry, 6 5 Wilson, John, 1 6 1

guage), 5 3 Young, Brigham, 7 3

Wilson, Margot, 1 6 5 Wisconsin Primate Center, 4 2

Zahavi, Amotz, 1 89

Wolpert, Lewis, 1 03

Zahavi's 'Handicap Principle', I B9

- 230 -

Grooming, Gossip, and the Evolution of Language - PDF Free Download (2024)
Top Articles
Latest Posts
Article information

Author: Madonna Wisozk

Last Updated:

Views: 6201

Rating: 4.8 / 5 (48 voted)

Reviews: 87% of readers found this page helpful

Author information

Name: Madonna Wisozk

Birthday: 2001-02-23

Address: 656 Gerhold Summit, Sidneyberg, FL 78179-2512

Phone: +6742282696652

Job: Customer Banking Liaison

Hobby: Flower arranging, Yo-yoing, Tai chi, Rowing, Macrame, Urban exploration, Knife making

Introduction: My name is Madonna Wisozk, I am a attractive, healthy, thoughtful, faithful, open, vivacious, zany person who loves writing and wants to share my knowledge and understanding with you.