Horses may have been domesticated much earlier than thought

Hinzugefügt: 2026-05-31

[00:00:00]Picture this. It's about 3,000 BC and you're standing on the vast and windswept Eurasian steps. No roads, no maps, nothing whatsoever. But you do have something that will change the entire trajectory of human civilization, a horse.

[00:00:15]When did humans actually start domesticating horses and when did we begin to ride them? The answer has enormous consequences and it is quite a complex and fascinating one.

[00:00:32]Before we get into it, let's establish why this question matters so much.

[00:00:36]Horses were of course a very efficient way to get around. Domesticating horses basically transformed everything we were doing. They enabled long-distance migration at speeds no human population had ever achieved before. They also changed warfare and they also allowed pastoralists to manage herds of cattle and sheep across enormous territories.

[00:00:57]Many researchers believe they were the vehicle quite literally by which the Proto-Indo-European language spread across Eurasia, eventually becoming the ancestor of English and Spanish and Hindi and Russian and hundreds of other languages spoken by half the world today. So, when riding began is a question that touches the very deep roots of human history. Now, in the past few years, a group of researchers using cutting-edge horse genetics have published a series of very influential papers arguing that effective horse domestication didn't really begin until about 2,200 BC. That's when a specific genetic lineage of horses called DOM2 spread explosively across Europe, the Near East, Central Asia and Anatolia.

[00:01:43]These DOM2 horses carried favorable mutations. One gene that likely eased back pain and improved endurance when when ridden and another linked to reduced fear and anxiety around humans.

[00:01:55]The genetics team argue that this genetic moment was the real beginning of domestication. And then another genetic team went even further arguing that widespread riding didn't really begin until about 1500 BC. Now, that is a very late date and not everyone is convinced.

[00:02:13]In May 2026, a new study is out that refutes the other theories or at least questions them. The central argument is this, domestication should not be considered one single genetic event, but rather a messy and regionally varied process that would have begun centuries, possibly millennia ago, so before 2200 BC.

[00:02:34]The paper organizes its evidence around three genetically distinct horse populations that existed around 3500 to 3000 BC.

[00:02:43]Think of these as three separate experiments in horse domestication happening all at the same time, but in different parts of Eurasia.

[00:02:53]The first is DOM 1. These would be horses from the Central Asian Steppes east of the Ural Mountains and they are the famous horses of Botai in Kazakhstan where hunter-gatherers had transformed into specialized horse keepers by around 3500 BC. [clears throat] At Botai, the evidence for domestication is quite remarkable. It seems like horses made up 99% of the food consumed and there are also signs of corralling, milking, and a form of butchery consistent with herd management rather than wild hunting. Most tellingly, some horse teeth show wear patterns on their second premolars consistent with being fitted with a rope bit, meaning that they were bridled and almost certainly ridden.

[00:03:35]DOM 1 horses survive today, by the way, as Przewalski's horses. You might have seen them in documentaries. They look kind of stocky and stiff-maned and this breed has for a very long time been considered the only remaining wild species, but turns out they're not really that wild, but rather the descendants of escaped once domesticated horses.

[00:03:55]There aren't really true wild horses left in the world. We rather have populations of feral animals including the famous Przewalski's horse. The second is DOM2, the Pontic-Caspian Steppe horses west of the Urals associated with the Yamnaya culture, which thrived from about 3200 to 2600 BC. This is the genetic lineage that would eventually give rise to every modern domesticated horse. And here's where the debate gets really sharp because the genetics researchers essentially argued that Yamnaya horses were ancestral to modern horses, but weren't yet truly domesticated. And the other study insists that the Yamnaya people ate horse meat, drank horse milk, um there's proteomic evidence from dental calculus, and they included horse bones in funerary rituals as well. So, yeah, there is a quite a debate there, and some of the skeletons also show physical signs of habitual horse riding.

[00:04:54]The Yamnaya expansion beginning around 3100 BC is one of the largest migration events in the past 5000 years of European prehistory. Their DNA would eventually contribute to about 75% of the ancestry of the Corded Ware people across Central and Northern Europe. And this more recent team argues that the kind of rapid long-distance movement across about 5000 kilometers would have been virtually impossible without the help of horses. The third and most novel contribution of this new paper is DOM3, a label that the authors coined for a group of indigenous European and Anatolian horses. These horses had largely been ignored in the domestication debate and treated as unimportant, but this team argues that around 3500 BC horse bones in Central and Southeastern Europe begin to show signs of management. Body sizes increase, size variability widens, which classic signs of human interference. And larger, possibly imported step horses also begin appearing. At a site called Salzmünde in Germany, there is a horse skull dated to about 3,300 BC that carries the gene for tobiano spotting, which is a coat pattern only found in domesticated horses.

[00:06:11]The genetic evidence suggests that this DOM3 lineage was eventually replaced by DOM2 after 2,200 BC, which is perhaps why it has been overlooked.

[00:06:21]So, we have three separate regions, three separate horse populations, and multiple independent lines of evidence for management and riding, all before the DOM2 genetic revolution that the genetics team placed at about 2,200 BC.

[00:06:37]One of the most fascinating sections of the paper concerns human skeletons, specifically the physical changes that habitual horse riding leaves on the human body.

[00:06:47]Bareback horse riding is biomechanically very unusual, and even riding with a saddle is physically very challenging.

[00:06:54]It requires hours of squeezing your thighs and knees together with legs spread wide while bouncing vertically on a firm surface and occasionally falling backward from a height. So, it doesn't really sound like the easiest thing in the world to do.

[00:07:08]Over the years, this leaves distinctive marks on the skeleton, like changes to the hip socket, stress reactions at thigh muscle attachment points, alterations to the shape of the femur, vertebral degeneration from repetitive impact, and yeah, also trauma due to falling. One of the researchers developed a diagnostic framework that identifies what he calls horse rider syndrome, a combination of four or more skeletal markers that taken together become very difficult to explain through any non-riding activity. And he applied this framework to 217 mostly Yamnaya individuals and the five individuals from Romania, Bulgaria, and Hungary dated between 3,000 and 2,500, they tend to show four or more rider traits.

[00:07:56]Two pre-Yamnaya individuals from Hungary and Romania also showed the syndrome um including one dated back to 4,400 which would push early horse riding to the fifth millennium. So, quite a large difference in comparison to the other studies.

[00:08:13]Overall, it was estimated that roughly 20% of the analyzed Yamnaya population might have been habitual riders and all of the diagnosed riders were male. The genetics team tried to dismiss the evidence. They offered alternative explanations for each individual.

[00:08:29]However, the other team pointed out that the power of the method lies precisely in combining multiple markers. A list of possible alternative causes for each individual trait doesn't address a diagnosis based on all of them together at the same time. And they also noted that the method has since been successfully applied to a Bronze Age population in Poland where both men and women showed rider syndrome and this um suggests that it could work beyond the Yamnaya context. One of the more creative counterarguments from the genetics side was that perhaps the Yamnaya people got their rider syndrome injuries not from horses but from riding in ox-drawn wagons. Chariots hadn't been invented yet but wagons had.

[00:09:11]And then the other team said that chariots don't involve a more active riding posture um that didn't exist until after the Yamnaya period so they're not really relevant. Secondly, early Bronze Age wagons because the authors do review some actual archaeological reconstructions from Yamnaya and related graves had no seats, no high boxes, and no elevated platforms. Wagon drivers in that era either walked beside their oxen or sat on the wagon bed itself. So, there is no biomechanical plausibility to the idea that sitting on a flat wagon floor would cause hip socket over-lization and thigh adductor enthesopathy in this specific pattern that is seen in the Yamnaya individuals. But, here's the question worth asking. If people were riding horses as early as 3,500 or even 4,000 BC, why does the archaeological evidence for riding, like horse equipment, uh depictions of riders, and so on, explode much later in about 1,200 - 1,000 BC?

[00:10:14]Well, the answer might be in the fact that what changed around the 1,000 BC wasn't riding itself, but mounted archery. That's when the composite recurve bow was developed. It was short enough to be used effectively from horseback and powerful enough to be military decisive. Simultaneously, around 1,000 BC, mass-produced socketed arrowheads appear across the steppes, standardizing arrow weights and increasing ammunition capacity. So, taken together, these innovations transformed horses from a tool of transportation and herding into a platform for deadly warfare. The Scythians and their successors then brought this combination to the attention of the ancient world's major civilizations from Greece to China, and the historical record explodes with horses. So, early riding was invisible because it didn't exist at all. It might have been invisible because it was primarily social and economic and not yet military. So, what this paper ultimately argues is that we should resist the temptation to define domestication as this one single moment, a genetic flip that happened once in one place at one time.

[00:11:28]Horse domestication was, in their words, a protracted, regionally varied process.

[00:11:33]People on the steppes were already incorporating horses into diet, ritual, and political symbolism centuries before any of the DNA signatures the genetic team focuses on appear. For example, artifacts like stone horse head maces dating back to about 4,000 BC. And horses from multiple distinct genetic backgrounds were being ridden and managed simultaneously with different communities sort of experimenting with different approaches in parallel. The genetic revolution of 2200 BC was real and important. Those more endurance capable DOM2 horses did spread rapidly and competitively in this time period, but on the other hand, that spread built on a longer foundation. It didn't come out of nowhere. These arguments are not without their problems. The most glaring data gap is that no horse from an actual Yamnaya or Corded Ware grave has yielded usable ancient DNA.

[00:12:32]The genetic argument for Yamnaya horse riding is therefore indirect. We know Yamnaya horses were ancestral to modern domestic horses, but we can't directly sequence the horses these specific people were buried with.

[00:12:45]The skeletal markers can also have alternative explanations, and the method hasn't been validated against large rider populations from the same era and region. The sample sizes are also quite small. Five Yamnaya individuals with four or more traits across 39 sites and 150 individuals is quite a thin evidentiary base. The other horsemen that are used as a comparison group lived about 4,000 years later and in a very different context with stirrups and saddles. Moreover, the horse milk peptides found in Yamnaya dental calculus, they only come from two individuals out of the 16 examined at one specific cemetery in the lower Don region. So, this might reflect a regional practice rather than some widespread Yamnaya custom.

[00:13:35]Okay, so where does that leave us, you might wonder? Well, the debate isn't fully resolved. It rarely is when dealing with such big topics, but the review paper does make a compelling case that the story of horse domestication might be older and messier than we imagine and then a single genetic timestamp can capture. At the same time, saying that horses were gradually managed and occasionally ridden across a wide range of populations doesn't straightforwardly contradict the genetics team's findings that effective and transformative horse-based mobility only became possible after the DOM2 mutations. Essentially, I think both things can be true simultaneously in the sense that early limited riding would have existed for centuries and it only became historically consequential after 2200 BC.

[00:14:29]And maybe that's the most important takeaway from this whole story. When we look for the origins of transformative technologies, whether horseback riding or writing or agriculture and so on, we almost never find one single inventor, one single moment, one single place. But we tend to find people across generations and across geographies and they're gradually trying to figure something out together.

[00:14:58]The horse changed the world and it turns out that change was a very long time coming. Well, I hope this was interesting and informative for you. If you enjoyed it, there is much more where that came from. I have a similar video about dog domestication or if you like stuff like the Viking Age, you're in luck because that is the main topic of this channel.

[00:15:16]Please consider paid membership or buy me a coffee if you want to support the project. This was Iryna. Take care, and until next time.