In human history, no animal has had a greater impact on human societies than the horse. But when and how humans domesticated horses is an enduring scientific mystery.
Half a million years ago or more, early human ancestors hunted horses with wooden spears, the very first weapons, and used their bones as early tools. During the late Paleolithic, some 30,000 years ago or more, ancient artists chose wild horses as their muse: horses are the most frequently depicted animal in Eurasian cave art.
After their initial domestication, horses became the basis of pastoralism in the grasslands of Inner Asia, and major technological advances such as the chariot, saddle, and stirrup helped make horses the primary means of travel, communication, agriculture, and warfare throughout much of the ancient world. With the help of sea voyages, these animals eventually reached the shores of every major landmass—even Antarctica, briefly.
As they spread, horses transformed ecology, social structures, and economies on a scale never seen before. Eventually, only industrial mechanization replaced their near-universal role in society.
Because horses have had a tremendous impact on shaping our collective human story, understanding when, why, and how horses were domesticated is important for understanding the world we live in today.
That turns out to be surprisingly challenging. In my new book, “Hoof Beats: How Horses Shaped Human History,” I marshal new archaeological evidence that rethinks what scientists like me thought we knew about this story.
A Hypothesis on the Domestication of Horses
Over the years, virtually every era and place on earth has been suggested as a possible origin of horse domestication, from Europe tens of thousands of years ago to countries such as Saudi Arabia, Anatolia, China and even America.
By far the most dominant model for horse domestication, however, is the Indo-European hypothesis, also known as the “Kurgan hypothesis.” It posits that sometime in the fourth millennium BCE or earlier, inhabitants of the Western Asian and Black Sea steppes known as the Yamnaya, who built large burial mounds called kurgans, mounted horses. The newfound mobility of these early horsemen, the story goes, helped catalyze massive migrations across the continent, spreading ancestral Indo-European languages and cultures across Eurasia.
But what is the actual evidence to support the Kurgan hypothesis for the first domestication of horses? Much of the key evidence comes from the bones and teeth of ancient animals, via a discipline known as zooarchaeology. Over the past 20 years, zooarchaeological data has seemed to converge on the idea that horses were first domesticated at Botai culture sites in Kazakhstan, where scientists have found large quantities of horse bones at sites dating to the fourth millennium BCE.
Other types of compelling circumstantial evidence began to mount. Archaeologists discovered evidence of what appeared to be holes in fence posts that could have been part of ancient enclosures. They also found ceramic fragments containing fatty horse remains that, based on isotope measurements, appeared to have been deposited in the summer months, a time when milk from domesticated horses could be collected.
The scientific evidence for early horse domestication, however, was a series of changes found on the teeth and jaws of some Botai horses. Like the teeth of many modern and ancient ridden horses, the teeth of the Botai horse appeared to have been worn down by a bridle, or bit.
Together, the data strongly pointed to the idea of horse domestication in northern Kazakhstan around 3500 BCE – not quite the Yamnaya homeland, but geographically close enough to keep the basic Kurgan hypothesis intact.
There were, however, some aspects of the Botai story that never quite gelled. From the outset, several studies showed that the mix of horse remains found at Botai was unlike that found in most later pastoral cultures: Botai are evenly divided between male and female horses, usually of healthy reproductive age. Regularly killing healthy, breeding-age animals in this way would devastate a breeding herd. But this demographic mix is common among animals that have been hunted. Some Botai horses even have projectile points in their ribs, indicating that they died as a result of hunting rather than controlled slaughter.
These unresolved loose ends hung over a basic consensus linking Botai culture to horse domestication.
New scientific tools raise more questions
In recent years, archaeological and scientific tools have rapidly improved, overturning important assumptions about the cultures of Botai, Yamnaya, and the early chapters of the human-horse story.
First, improved biomolecular tools show that whatever happened at Botai had little to do with the domestication of the horses that live there today. In 2018, nuclear genome sequencing revealed that Botai horses were not the ancestors of domesticated horses, but of the Przewalski horse, a wild relative and steppe dweller that has never been domesticated, at least not in recorded history.
When my colleagues and I subsequently reexamined skeletal features associated with riding in Botai, we found that similar problems were also evident in wild Ice Age horses from North America, which had certainly never been ridden. Although riding can cause recognizable changes to the teeth and jaws, we argued that the minor problems we see in Botai horses could reasonably be attributed to natural variation or life history.
This finding raises the question again: did horse transport even take place in Botai?
Leaving the Kurgan Hypothesis in the Past
In recent years, understanding the archaeological finds surrounding horse domestication has become increasingly complex.
For example, archaeologists noted in 2023 that human hip and leg skeletal problems found in Yamnaya and early Eastern European burials closely resembled those found in mounted horsemen, consistent with the Kurgan hypothesis. But problems like these could also be caused by other types of animal transport, including the cattle carts found at Yamnaya-era sites.
How should archaeologists interpret these conflicting signals?
A clearer picture may be closer than we think. A detailed genomic study of early Eurasian horses, published in June 2024 in the journal Nature, shows that Yamnaya horses were not ancestors of the first domesticated horses, known as the DOM2 lineage. And Yamnaya horses showed no genetic evidence of tight control over reproduction, such as changes associated with inbreeding.
However, the first DOM2 horses do not appear until just before 2000 BCE, long after the Yamnaya migrations and just before the first horse and chariot burials also appear in archaeological finds.
For now, all lines of evidence seem to converge on the idea that horse domestication probably did occur in the Black Sea steppes, but much later than the Kurgan hypothesis requires. Instead, human control of horses began just before the explosive spread of horses and chariots across Eurasia during the early second millennium BCE.
Of course, there’s more to be solved. In the latest study, the authors point out some funny patterns in the Botai data, specifically fluctuations in genetic estimates for generation time—essentially, how long it takes, on average, for a population of animals to produce offspring. Could these suggest that Botai people were still raising those wild Przewalski’s horses in captivity, but only for meat, with no role in transportation? Perhaps. Future research will tell us for sure.
In any case, one thing has become clear from these conflicting signals: the first chapters of the story about the relationship between man and horse are ready to be retold.
This article is republished from The Conversation, an independent nonprofit organization that brings you facts and analysis to help you understand our complex world.
It was written by: William Taylor, University of Colorado Boulder.
Read more:
William Taylor is not an employee of, an advisor to, an owner of stock in, or a recipient of funding from any company or organization that would benefit from this article, and has disclosed no relevant affiliations beyond his academic appointment.