Tamed

by Alice Roberts

  Towards the end of the fourth millennium BCE, the herders of the steppe were becoming more mobile again. A climatic improvement in the early centuries of that millennium was followed by a downturn. Large herds now needed to roam more widely in order to take in sufficient pasture. This seems to have stimulated the emergence of a new way of life, and a new culture. The herders couldn’t afford to be semi-settled any more, as they had been at Botai; they needed to move with their herds and flocks. The solution: wagons. These wheeled vehicles first appeared on the steppe around 5,000 years ago. This sounds very precise. How on earth could archaeologists make such a claim, about vehicles which leave so little trace on the ground? Wheel ruts don’t tend to last for millennia (and where they are found, they’re impossible to tell apart from sled-ruts).

  The answer lies inside the graves of these steppe people. They were still making kurgans, and under those mounds of earth, they buried their elite – mostly men – with their wagons. These extraordinary burials, with bodies and dismantled wagons placed in pits, appear across the Pontic-Caspian Steppe, dating to between 3000 and 2200 BCE. The burial rite gives this new culture its name, though it’s a name those people never knew – ‘Yamnaya’, after the Russian for ‘pit-grave’.

  The wheel itself may not have been invented on the steppe. It’s thought that the idea of wheeled vehicles spread there, either from the west, from Europe, or from the south, from Mesopotamia. The earliest known image of a wheeled vehicle comes from a site in Poland, and dates to around 3500 BCE, while a clay model of a wagon from Turkey dates to about 3400 BCE. With covered, ox-drawn wagons as their mobile homes, the herders could migrate around the landscape, following huge flocks and herds. And they were still riding horses, of course. Archaeologists suggest that the cycle of the year would have taken the herders out on to the open steppe in spring and summer, while in winter they’d set up camp in river valleys. Crucially, those valleys would have contained trees – providing wood for fuel, and for mending the wagons. Although this horse-riding, wagon-camping, kurgan-building culture stretched right across the Pontic-Caspian Steppe, there were regional differences in livestock and the types of plant being eaten. In the east, beyond the River Don, people were mainly tending sheep and goats, with just a few cattle and horses – which provided essential mobility. Along with mutton and goat, they were eating foraged tubers and the seeds of goosefoot – a plant very closely related to quinoa. In the western steppe, people were more settled – and they were herding cattle and pigs, and growing some cereals.

  But – like the earlier horse-riding nomads in the fifth millennium BCE – the Yamnaya didn’t stay on the steppe. By around 3000 BCE, they had begun a westward expansion, into the Lower Danube Valley, and on to the Great Hungarian Plain. The herders of the steppe were expanding eastwards as well, making contact with the early farmers of China. Crops and animals that had been domesticated in the west spread east. The idea of copper metallurgy may also have travelled to China from the west. After the Yamnaya, there seem to have been successive waves of steppe people expanding both east and west. Over five thousand years, this scenario played out again and again, with the last wave recorded in history – as the thirteenth-century Mongol invasion.

  The prehistoric expansions of steppe nomads appear to have had very different impacts on existing societies in the east and west. In China, the nomads seem to have merged with settled societies, but in the west they encroached on lands occupied by other nomadic pastoralists – and they caused a knock-on effect, pushing those nomads even further west.

  The Yamnaya expansion into Europe had a profound cultural impact, one that is still echoing today. Geneticists and comparative anatomists use patterns of similarity and difference between modern organisms, and ancient ones too, when they’re available, to construct phylogenies – family trees which represent evolutionary history. Linguists can do much the same thing with languages, using comparative grammar and vocabulary. Many ancient and modern languages, from English to Urdu, Sanskrit to ancient Greek, all group together, in the Indo-European language family. And linguists have traced the evolution of sounds back and back, until we have the closest thing to an original Indo-European language – in the form of about fifteen hundred distinct sounds. It’s very hard to test whether they really have found the traces of an ancient language, but archaeological discoveries have since revealed previously unknown words in Hittite and Mycenaean Greek – which were correctly predicted by the methods of the historical linguists, giving us some grounds to trust their reconstructions.

  The fragments of the Proto-Indo-European language contain words for otter, wolf and red deer as well as for bee and honey, cattle, sheep, pig, dog and horse. In other words, it’s a language root that clearly emerged after the beginning of the Neolithic – its speakers had words for domesticated animals. Still, it’s not clear that the word for ‘horse’ refers to a domesticated horse. But there are other clues. The reconstructed Proto-Indo-European also contains words for wheel, axle and wagon. It seems that the Yamnaya people, the horse-riding, wagon-driving nomads of the steppe, were speaking a language which would form the basis for all the Indo-European languages that we continue to speak, across Europe and western and south Asia, today. How wonderful to think we’re still using words that contain a faint echo of that ancient culture on the steppes.

  Close cousins and alternate histories

  Unravelling the origins of domestic horses has proved – this is now a familiar theme – very difficult indeed. Just as with wolves and early dogs, aurochsen and early cattle, it’s difficult to discern any differences between the bones of wild and domestic horses. Those metapodials from the Botai sites were only subtly different to wild horse bones. In fact, there are very few differences in the skeletons of any species belonging to the genus Equus much more generally. If you compared the skeleton of a zebra with the skeleton of an ass, you’d be hardpushed to tell them apart. Once again, this is where genetics has ridden to the rescue. And before looking at the origin of domestic horses, we need to make sure that we really understand the differences between the various equine species that exist today. Some have recently turned out to be even less ‘different’ than we’d thought.

  It seems that we’ve been taxonomically over-zealous in dividing up Equus in the past. Genetic analyses have prompted suggestions that some apparently discrete populations, traditionally seen as separate species, are actually much more closely related. For instance, the plains zebra and the extinct quagga are traditionally labelled as separate species, based largely on how they look. Modern genetics says: they are one and the same species. Similarly, the extinct ‘stilt-legged horses’ of America are, genetically, caballines – close cousins of modern domestic horses. But, as the family tree of horses collapses in on itself, with fewer branches and closer genetic relationships than previously suspected, one key part of the family tree is uncontested – and that’s the very close relationship between domestic horses, Equus caballus – together with their wild ancestors and relatives, Equus Ferus – and the surviving, truly wild horses of the central Asian steppe: Equus przewalskii – Przewalski’s horse. These small but robust horses have a sandy or reddish coat, pale muzzles and bellies, a bristly brown mane, and a stripe along the back.

  Genetic analyses have made it possible to reconstruct the family history of the caballines, or true horses, and pin some dates on it. The wild ancestors of domesticated horses emerged as a separate lineage some 45,000 years ago, when they diverged away from the ancestors of Przewalski’s horse – long before domestication. Despite that divergence, however, a small amount of interbreeding continued – the reciprocal gene flow shows up quite clearly in genomes today. Most of that interbreeding happened long ago, before the last glacial maximum, twenty thousand years ago. After the Ice Age, there was still some genetic input from Przewalski’s horses into the ancestors of domestic horses, and it continued even after domestication. Later still, right at the start of the twentieth century, there’s ev
idence for gene flow in the other direction – from modern horses into Przewalski’s horses. This last injection of domestic horse genes into Przewalski’s horse was precisely at the time when Przewalski’s horses first started to be kept and bred in captivity.

  The ability of these two horse populations to interbreed really is extraordinary. They’re distinct enough – morphologically and genetically – to be considered separate species. And they have different numbers of chromosomes – often considered to be a complete barrier to interbreeding. While domestic horses have sixty-four chromosomes (thirty-two pairs), Przewalski’s horse has sixty-six (thirty-three pairs). When a mammalian egg or sperm is made, it ends up with half the usual genetic complement found in other cells of the body. At fertilisation, the genetic material from the egg is combined with that from the sperm to create a full set once again. Each chromosome from the egg must pair up with its opposite number, from the sperm, before the fertilised egg can start to divide and make an embryo. If a domestic horse and a Przewalski’s horse mate – the resulting fertilised egg would have one set of thirty-two chromosomes and one set of thirty-three. But somehow (and even geneticists are astounded by this), the chromosomes manage to pair up – if they didn’t, there’d be no viable offspring. And the traces of interbreeding in the genomes of modern domestic horses and Przewalski’s horses shows that, not only were such offspring viable, they were fertile – they could reproduce.

  Of course, hybrids between equine species are well known. A hinny is a cross between a male horse and a female donkey. A mule is a cross the other way – between a male donkey and a female horse. And although hinnies and mules are usually sterile, they can occasionally reproduce successfully. That’s again quite remarkable considering that donkeys have thirty-one pairs of chromosomes and horses have thirty-two. But the genomes of different equine species contain evidence of an even more astounding feat – interbreeding and gene flow between the Somali wild ass, with thirty-one chromosome pairs, and Grévy’s zebra, with twenty-three. Such findings challenge our stereotyped views about how biology should work. The species boundary is turning out to be much more permeable than we’d anticipated, before genomics came along. Even differences in chromosome number don’t seem to be quite the barrier to successful reproduction that we’d previously imagined.

  As well as tackling questions of interbreeding, genetics provides an insight into fluctuations in the size of ancient populations over time. Both the ancestors of domestic horses, also known as Equus ferus, and Przewalski’s horse suffered a population crash in the late Pleistocene and early Holocene, around 10,000 to 20,000 years ago. Populations continued to dwindle – up until the point of domestication, some 5,000 years ago. Then, for Equus caballus, the domesticated horse, the future started to look decidedly rosy. But while this population of horses grew and grew, and spread around the world, their wild cousins would become endangered.

  The close wild relative of domestic horses, Equus ferus, also known as the Tarpan – with a characteristic sandy-grey-coloured coat, pale belly, black legs and a short mane – finally went extinct in 1909. The Przewalski’s horse was also spiralling towards extinction. These rare, shy horses were spotted by a Russian explorer and geographer, Nikolai Mikhailovich Przewalski, who was making his way across the central Asian steppe in 1879. By this time, the range of these wild horses had contracted, and there were only small herds roaming the steppe in Mongolia and Inner Mongolia. When Przewalski was preparing to leave Mongolia, he was presented with the hide and skull of one of the horses, which had been shot, and which he duly took back to St Petersburg. These remains were studied by the zoologist I. S. Poliakov, who published his description of this unusual horse in 1881. Poliakov determined that the remains of the beast from Mongolia were sufficiently different from domesticated stock to be considered new to science, and he gave the wild horses of Mongolia a new species name, honouring its discoverer. The horses immediately became collectable, and expeditions set out to Mongolia to capture specimens for zoos – further depleting the wild population. The last Przewalski’s horse to be caught was a mare, named Orlitza, captured as a foal. The horses were getting rarer and rarer in the wild. Being recognised as a new species was in some ways their downfall. The expeditions that fed zoological collections inevitably killed some animals, and dispersed others.

  In 1969, the last sighting of a wild Przewalski’s horse was reported, from the Dzungarian Gobi in south-western Mongolia. Extinct in the wild, a few Przewalski’s horses survived in zoos, long enough to breed. In the 1980s and 1990s, attempts were made to reintroduce these horses to the wild, using horses bred from a stock of just fourteen individuals – including Orlitza. And the attempts were successful. Between the 1960s and 1996, Przewalski’s horse was considered ‘extinct in the wild’, but by 2008 it was back in the game – though in such small numbers it was designated ‘critically endangered’. The numbers crept up – by 2011 it was considered to be merely ‘endangered’ – which meant that a population of more than fifty mature animals was living wild.

  It’s now estimated that there are a few hundred Przewalski’s horses living outside of captivity. These small numbers mean that the population remains very vulnerable to adversity – to disease and severe winters – but they have some help. In the Kalamaili Nature Reserve in Xinjiang, China, where Przewalski’s horses were released into the wild in 2001, the horses are rounded up into a corral each winter to give them extra food and to protect them from competition with domestic horses. By 2014, there were 124 individuals in just this one group of rewilded Przewalski’s horses, which has been described as the most successful of the reintroduction efforts in China.

  The population in captivity is looking healthy too – with around 1,800 horses in zoos worldwide, and growing. The reintroductions to the wild have mainly taken place in China and Mongolia, around the area where the horses were last seen living wild before going extinct. But some Przewalski’s horses have also been released into nature reserves and national parks – in Uzbekistan, Ukraine, Hungary and France.

  The story of these wild horses provides us with an alternate history for their domesticated cousins, Equus caballus. What would have happened had these horses remained wild? There’s no doubt that the course of human history would have been altered, but the fate of horses would have been significantly affected as well. Wild horses formed an important source of meat for our Palaeolithic hunter-gatherer forebears, but they are surely very likely to have been hunted to extinction, or close to it, had they not proved useful to our ancestors in other ways – carrying riders across the vast steppes and knights into battle; pulling chariots, wagons and cannons; becoming emblems of status and prestige in human society. The reintroductions of Przewalski’s horses into the wild seem to be a success story – a triumph for rewilding – but the global population of these horses, out in the wild and in captivity, numbers just a few thousand at most. In comparison, there are around 60 million domestic horses on the planet. There are concerns about diminishing genetic diversity amongst them, and about loss of breeds, but Equus caballus is far from being an endangered species.

  Leopard spots and horses’ faces

  Perhaps it seems that we’ve already nailed the origin of this particular domesticate. Yet while the earliest archaeological evidence for domesticated horses undoubtedly comes from the Pontic-Caspian Steppe, this doesn’t mean that all our horses today must have come from a single origin. There could have been later, independent centres of domestication. After all, horses ranged widely across Eurasia – there were plenty of other places where horses and humans had been in contact for millennia. In a more diffuse, multiregional model, separate herds would later have merged into a single, diverse population of domesticated horses, which continued to reflect regional differences and local origins. Just as with dogs, the apparent diversity of modern horses could be taken to suggest that multiple origins were most likely. In the past, similarities between some domesticated breeds and local, wild ponies we
re taken to support such a model. Studies of morphological characteristics – the shape and size of bones – suggest a strong similarity between the Exmoor pony, the Pottok pony of the Basque country, and the extinct Tarpan. Some have argued that the beautiful, semi-wild horses of the Camargue represent the direct descendants of the ancient, truly wild Solutré horses which featured in Ice Age cave paintings. But the genes say something different, and more interesting.

  In 2001, a piece of research was published, based on analysis of a particular stretch of mitochondrial DNA in samples taken from thirty-seven individual horses – and this bit of DNA was very variable indeed. But did this diversity represent lineages which had diverged from each other before or after domestication? If before, this would suggest multiple origins for modern horses. If after, that would support a single origin. In order to answer this question, the geneticists looked at the mitochondrial DNA of a donkey – which was 16 per cent different to that of horses. They assumed that donkeys and horses had split apart from each other some time between 2 million years ago (as the fossil record suggested) and 4 million years ago (according to the genetic estimates at the time). This gave them a form of calibration – over 1 million years, you’d expect to see genetic sequences diverging by 4 per cent (given a split 4 million years ago) to 8 per cent (for a 2 million-year-old split). They could then apply this rate to the differences they’d found within modern horse mitochondrial DNA, which amounted to about 2.6 per cent. The calibration suggested that the lineages found in those modern horses must have been diverging for somewhere between 320,000 to 630,000 years. Even the lower estimate places the origin of this genetic diversity well before the date of domestication around 6,000 years ago. The geneticists went on to suggest that wild horses had been captured over a huge area, and used for both meat and transport. Then later on, as wild populations began to disappear, domesticated herds became more important and were interbred, to form the genetic basis of modern horses. The researchers contrasted the domestication history of horses with that of dogs, cattle, sheep and goats. Firstly, those other species had been domesticated much earlier (still true), and they all came from a restricted origin, then spread out. The domestication of horses, on the other hand, seemed to have happened again and again, and in many places it represented the dissemination of an idea – a technology – rather than a spread of the animals themselves.