The Journey of Man: A Genetic Odyssey

by Spencer Wells

  What I have just described is a process that could have occurred around 60–70,000 years ago in Africa, as a single fortuitous event changed the course of human evolution. As with many historical events, it depended on the right person being in the right place at the right time – a coincidental triptych that provided the spark for a revolution. But is this necessarily the way things happened?

  The short answer is that we simply don’t know. The anthropological term ‘Great Leap Forward’, coined by Jared Diamond, was borrowed from Mao Tse-tung’s 1950s plan for the industrialization of China to describe the development of radical shifts in technology at the onset of the Upper Palaeolithic, around 50–70,000 years ago. These ‘killer applications’, as we called them in the last chapter, marked a radical departure from the way of life that had gone before, and they deserve an explanation. What caused human behaviour to change so significantly?

  Richard Klein, one of the strongest supporters of the Great Leap Forward theory, cites three significant archaeological shifts that occurred around this time. First, the tools used by humans became far more diverse and made more efficient use of stone and other materials. Second, art makes its first appearance, with a presumed leap in conceptual thought. And finally, it is around this time that humans began to exploit food resources in a far more efficient way. All-in-all, the evidence points to a major change in human behaviour. And Klein points to our DNA as the reason.

  The sorts of changes we see at the onset of the Upper Palaeolithic could only have come about, he argues, if we began to communicate with each other more effectively. He infers from this that the onset of the Upper Palaeolithic marks the origin of modern language, with its rich syntax and multitude of ways to express oneself. This flowering of language skills is thought by most anthropologists to be a critical prerequisite for further social development. The development of complex social networks is almost certainly the spark that brought about the changes in Upper Palaeolithic behaviour. And this, Klein believes, happened because of a change in the way our brains are wired, set in motion by a genetic event.

  We can gain some insight into what these changes may have been by looking at modern children. Swiss psychologist Jean Piaget, working in the mid-twentieth century, developed a detailed scheme for normal child development. It involved a progression from object recognition to a gradually more complex understanding of the way in which objects relate to each other. Most of the earliest stages focus on organization of real-world objects (such as bottle, rattle, or Daddy’s face) into ever more complex systems through the adaptation of behaviours (when I see Daddy’s face, I usually get a bottle, or sometimes a rattle). It sounds complicated, but it does seem to explain the trial-and-error way in which children learn to interact with their world. It also provides a framework for the acquisition of language skills, the most uniquely human behaviour.

  Children begin to speak by ‘babbling’ – random sounds that roll off the tongue. This babbling phase gives way at around twelve months to actual words. Many psychologists and linguists think that children’s first words, such as ‘mama’ and ‘papa’, are the easiest to learn, genetically programmed into human vocal anatomy in some way. They are found almost universally in all languages, suggesting that there may be a grain of truth in this. The American linguist Merritt Ruhlen, however, argues that the universality of these words is the evolutionary remnant of a common origin for all human languages – a trace of the original language spoken tens of thousands of years ago – rather than a programmed anatomical by-product. It is likely that both attributes play a role, with the most basic sounds having been used in the first human language because they are the most basic sound combinations produced by our vocal machinery.

  The babbling and single words continue for another year, with a massive expansion of the child’s vocabulary. The first two-word sentences begin to emerge during this process, as the child combines different words to form a clause with a new meaning. My older daughter’s name is Margot, and during this phase she began to say things such as ‘Margot kiss’ and ‘Mummy hold’. Then, around age two, a massive leap in spoken language occurs. It is at this age that most children begin to put together three words into complex sentences – ‘Margot kiss Daddy’, rather than simply ‘Margot kiss’ or ‘Kiss Daddy’ – with the subject-verb-object (SVO) structure, or syntax, that characterizes English and most other human languages. The structure SOV (‘Margot Daddy kiss’) is used by a few languages (Japanese, Korean and Tibetan among others), while VSO and VOS structures are used by around 15 per cent of languages (Welsh is an example of the former and Malagasy of the latter). The rarest structure of all is OSV, perhaps best known from the film The Empire Strikes Back as the language of Yoda the Jedi master: ‘Sick have I become’ and so on, used by only a handful of languages spoken in the Brazilian Amazon. The important thing to glean from this syntactic diversity is that word order plays a crucial role in our understanding of a sentence. As the old saw goes, ‘dog bites man’ is mundane trivia, while ‘man bites dog’ is newsworthy.

  So, the explosion of linguistic complexity in a two-year-old is a result of the mastery of syntax, and from then on it is a never-ending barrage of ever more complex sentences. The great leap forward in understanding, however, involves crossing the syntax barrier-without a mastery of this, the rest will never happen. This is what we see with chimpanzees taught to use American Sign Language, such as Kanzi the bonobo. Kanzi was able to create and understand a wide variety of two-word sentences, like an eighteen-month-old human infant, but he never mastered the complex syntax of a two-year-old’s speech. The significant difference in human vs. ape communication seems to have been the creation of brain structures that allowed an understanding of syntax, and thus the communication of complex meaning.

  To see why this might be, let’s try another thought experiment. Imagine that you are cast away on a remote island into a tribe of people speaking a language unintelligible to you. Nothing in the language makes sense – there are no cognates with anything in your mother tongue. Your goal is to find out where you are and how to return home. How would you do it? Initially, it is likely that you would try to communicate using the skills that you developed as a young child – trial and error, focusing on nouns and verbs in isolation. Pointing to a tree, you raise your eyebrows questioningly, relying on the near universality of many human facial expressions (themselves perhaps an evolutionary remnant of a time before complex speech developed). Soon you learn enough words to develop basic sentences – ‘I drink’, or ‘Eat now’. The final leap will be to create complex sentences that convey much more information than single nouns and verbs alone. You congratulate yourself on the achievement of two-year-old speech when you can finally say ‘I go home now’. At this point, the locals seem to have a ‘eureka moment’, whereupon they take you to the other side of the island, to the local airstrip where you can catch a flight home.

  This imaginary shipwreck scenario serves to demonstrate the utility of syntax for human communication, and gives us a good idea of why it might have been such an enormous leap forward for our early human ancestors. What it fails to do, though, is to explain what may have caused it to happen. If the intellectual chasm between humans and apes is spanned by a syntactic bridge, we need to ask why it appeared in our ancestors but not those of chimps and gorillas. Here again we obtain some help from primate behavioural research. One of the things that prevents chimps from developing complex syntax, according to Sue Savage-Rumbaugh, is limited short-term memory. To understand the meaning of a complex sentence, you must remember the beginning when you reach the end in order to integrate them. Not difficult, perhaps, for ‘man bites dog’, but a little tougher for a complex past-tense construction in German, where the active verb in the sentence only shows up at the end! Limited short-term memory may be the root cause of chimpanzees’ minimal language skills.

  The reason why our ape cousins never evolved short-term memory comparable to ours may have to do with their lifesty
le. All of our simian relatives live in forests, and are at least partially arboreal. Our ancestors, on the other hand, appear to have given up a life in the trees several million years ago. Australopithecines had an upright stance, something that would only have been evolutionarily useful in a tree-free environment. The structure of the African ecosystem, with its vast savannahs in close proximity to forests, is in fact an ideal habitat for a primate making the transition from trees to the ground. And it was this leap beyond the trees that set in motion the evolutionary trajectory that would eventually lead to syntax and modern language.

  Most anthropologists now accept that early hominids walked upright before they developed higher mental capabilities. As with Raymond Dart’s Taung baby, the brains of the earliest human ancestors were comparable in size to those of apes, while they already showed the skeletal modifications that indicate bipedalism. Bipedalism would have conferred, in a treeless environment, the advantages of height (allowing improved vision), efficient overland movement and free hands for tool use – none of which would be terribly important if you moved primarily by climbing from branch to branch in a forest. As the saying goes, necessity is the mother of invention – and this is certainly true of evolution. But what drove us to the grasslands in the first place?

  Climatic changes have periodically wreaked havoc on Africa’s forests, with low rainfall reducing their area substantially several times over the past 10 million years. One particularly dry spell, between 5 and 6 million years ago, actually resulted in the disappearance of the Mediterranean, with significant knock-on effects on the African climate. During this prolonged drought some of the tree-dwelling apes may have moved to the edge of the forest to take advantage of resources offered by the grasslands. But while forest-dwelling apes are gatherers (chimpanzees occasionally kill and eat monkeys, but their diet consists primarily of fruit and insects), those who moved on to the savannah had to become hunters. This is because it is quite difficult for large primates to live on the savannah by gathering alone – plants and insects simply don’t provide enough nourishment. Animals, particularly mammals, provide a high-calorie diet rich in protein. And it was the necessity of hunting and killing the mammals of the grasslands, as well as escaping the attentions of the other carnivores living there, that probably drove the development of the human brain.

  If you imagine life as a chess game, then the causes and effects of brain evolution make a bit more sense. When times are good, and the environment is constant, chess can be pretty basic – perhaps even defaulting to fool’s mate. If you are hungry, you find a piece of fruit or use a blade of grass to fish termites out of a hole. Simple. Life in the forest is like this, day in and day out. The reason why so many species become extinct when forests are destroyed is that they are simply unable to cope with the new environment – they are too well adapted to their local habitat. Orang-utans are gloriously suited to life in the south-east Asian rainforest, but they do not manage very well in deforested slash-and-burn fields. When times become more difficult and the environment changes, you must start to anticipate your moves in advance – and chess becomes a more challenging proposition. This is what humans thrive at, precisely because of our birth as a species in the crucible of a marginal and changing environment. In a sense, we are biologically adapted to adapt. But while other animals have complex physical adaptations, we have only our minds, and our adaptations come in the form of behavioural changes.

  One of the results of having a highly adaptive mind is the development of a complex culture. Initially perhaps an extension of cooperative hunting technology, with its strong selection for intelligence and social interaction, human culture reached beyond the merely practical to encompass art, science, language and all of the other accoutrements of the ‘humane’ life. While we are not the first hominid to display extraordinary cultural adaptations, we are the only one to have taken them to such extremes. The Neanderthals, for instance, show evidence for group care of the sick. They also, at sites such as Teshik-Tash in present-day Uzbekistan, hint at deeper conceptualization of their place in the world, as suggested by the ritualized burial of a Neanderthal child surrounded by goat horns. But, more than any other species, it is complex culture that uniquely defines Homo sapiens, that makes us what we are. Without the early sparks of it, our hominid ancestors would never have ventured beyond the African forest margin into the savannah. And without having it in spades, we would never have survived what we encountered when we moved out of Africa into Eurasia, around 50,000 years ago.

  Bacterial soup

  When a single bacterium is placed into a nutrient-rich broth and allowed to divide to form two bacteria, then four, then eight and so on, an interesting thing happens. As we have seen, whenever DNA is copied – during reproduction – there are random mistakes known as mutations. These are the changes in soup recipe that occur naturally as a part of passing it on to the next generation. The same pattern is seen for dividing bacteria. Thus, in our rapidly propagating bacterial soup, we begin to see unique genetic lineages taking shape as a result of the small changes in their genomes. If we examine a sample of DNA sequences from the bacterial population after a few generations, we see barely any differences among them. But if we wait a few hundred generations (only a couple of days for bacteria) we see an enormous amount of variation. As with Zuckerkandl and Pauling’s insight into protein evolution, the longer the population has been growing, the more variation we see. Simply put, there are more genetic differences between two bacteria chosen randomly from the older population than from the younger.

  The experiment we have just performed with our bacterial soup illustrates what happens in any exponentially growing population, where we double the number of offspring each generation. Most obviously, the population increases in size rapidly – if we actually allowed the bacteria to divide without constraint for a few days, they would take over the planet. Far more important for our story, though, is the reason for this massive population explosion: every individual in the population leaves offspring. No one loses out in the evolutionary lottery – they all have bacterial babies, and their babies all have babies, and so on. This has an interesting knock-on effect on the genetic structure of the population.

  If we ask how many genetic differences, on average, distinguish the bacteria that comprise the growing population, we now know that the answer depends on how long the population has been growing. In fact, there is a distribution of differences among the individual bacteria, rather like the bell-shaped Gaussian curve that tormented us in our mathematics classes at school. The mean of this distribution – the average number of differences between individuals in the sample – depends on the length of time that the population has been growing. If we imagine the curve as a wave, moving from left to right as it accumulates more and more differences, then the further to the right it is (in other words, the further from zero), the more mutations the population has accumulated. And like the comparisons of haemoglobin sequences from horses and gorillas, the rate at which the wave moves from left to right is predictable, because the rate at which mutations occur is constant – our molecular clock tolling in A (as well as C, G and T). Because of this, we can calculate how long the population has been growing exponentially by measuring the mean of the distribution – the midpoint of the wave. Fine, you may be saying, this may make an interesting laboratory exercise for a university genetics course, but it isn’t terribly pertinent … unless, of course, we see the same pattern for other organisms.

  Figure 5 Mitochondrial DNA (mtDNA) mismatch distributions of two expanding populations. The longer the population has been growing, the greater is the average number of sequence differences.

  Henry Harpending, an anthropologist at Pennsylvania State University, and his colleagues did precisely this analysis for the distribution of genetic differences among human mitochondrial DNA sequences and found a striking pattern. First, the distribution of differences – called the mismatch distribution – indicated quite clearly that human po
pulations had indeed been growing rapidly, like bacteria. This was because the telltale wave was there in the data – a smooth, bell-shaped curve that indicated the human species had been expanding at a great rate. In populations of constant (or shrinking) size, the distribution begins to deteriorate, becoming ever more saw-toothed as time goes on owing to the uneven loss of genetic lineages – the result of genetic drift, or perhaps selection. So, there was a clear genetic signal that humans had expanded rapidly. The exciting result came when Harpending calculated the estimated start of the expansion: approximately 50,000 years ago, corresponding very well with our estimate of the time at which modern humans started to migrate out of Africa, and almost exactly with the onset of the Upper Palaeolithic.

  Harpending and his colleagues examined mtDNA data collected from twenty-five worldwide populations, and all but two of them showed evidence for exponential growth over the past 50,000 years. The two populations with saw-toothed distributions had (on the basis of other evidence) recently been subject to drastic reductions in population size, so the analysis was clearly capable of differentiating between the two scenarios. Furthermore, the populations seemed to have expanded nearly independently of each other. Africans started the ball rolling around 60,000 years ago, followed by Asians at 50,000, and finally Europeans at 30,000 years ago. It was a stunning result. The mtDNA data agreed perfectly with archaeological evidence for the progress of Upper Palaeolithic technology: first in Africa, followed by Asia, and finally Europe – even the dates were the same. It seemed that the Great Leap Forward had left its genetic trace in our DNA, tracing the progress of the ‘killer app’ around the world. It also hinted at a route – but the details of the journey would have to wait until Adam’s sons showed the way.