Your pet dog’s ancestor was a fierce, wild animal. How was it tamed?

In 1881, zoologist John Murdoch took part in an expedition to Point Barrow, Alaska, at the northernmost tip of the US. The goal was to conduct a two-year, unbroken observation of meteorological and magnetic phenomena and to document the nature and wildlife of the Arctic along the way. Travelling north through Alaska, Murdoch and his crew witnessed a peculiar act: an Iñupiat family captured two wolf pups and took them back to camp. The family carefully fed and nurtured the pups, raising them to adulthood before killing them for their fur. Murdoch was observing an ancient tradition. It turns out this Iñupiat practice might also hold the key to understanding the origin of modern dogs.  

How ancestral grey wolves were transformed into humanity’s best friend has long been debated. For the past several decades, the prevailing hypothesis has been that wolves domesticated themselves. They initiated the process by first wandering around the periphery of human settlements and feeding on rubbish tips. Over time, they became habituated to the presence of people and formed mutually beneficial relationships with them. Only then did curious humans select and breed individuals with traits like docility and gregariousness, eventually giving rise to the pet canines we know and love today.   

Through this unguided and unintentional process, wolves scavenged their way to domestication. Or so the theory goes. Recent evidence, however, has led many scientists to abandon the idea of self-domestication. If the revisionist thinking is correct, then humans, not wolves, were the driving force – and the domestication of dogs is evidence that humanity has a deep and complex relationship with wild animals that was born long ago.  

Popular perceptions of dog domestication have been shaped in large part by the late Raymond and Lorna Coppinger, who wrote a series of highly readable books on the subject at the beginning of the 21st century. The husband-and-wife team of biologists based much of their argument on “pariah” dogs that prowl human garbage dumps, feeding off leftovers and sometimes receiving direct aid from local people. “These animals are ownerless and survive largely on scraps of food waste from human settlements,” says archaeologist Loukas Koungoulos at the University of Western Australia. “They make up a majority of the species Canis familiaris worldwide – in some estimates up to 70 per cent of all dogs presently alive.”   

Pariah dogs are the perfect analogue for wolves at the beginning of their domestication, the Coppingers argued. The idea they championed was that self-domestication occurred when humans transitioned from hunting and gathering to farming. This was when our ancestors became sedentary, living in larger groups and producing enough waste to attract wolves, in much the same way that pariah dogs are attracted to settlements today. This shift began around 12,000 years ago in the Middle East.   

“Finds of considerably earlier dogs would naturally disprove the idea,” says Koungoulos. And such finds have now emerged.   

Evolution of the first dogs

Palaeontologists have discovered around two dozen fossil specimens of dogs ranging in age from 35,500 to 13,000 years old across Eurasia, in countries including Spain, France, Belgium, Italy, Switzerland, Germany, Ukraine and Russia. These Palaeolithic dogs have a variety of physical characteristics that distinguish them from wolves. They weigh 31.2 kilograms on average, while Pleistocene wolves weighed around 41.8 kilograms. They also have shorter snouts, a slightly wider palate and shrunken canine teeth. These differences in morphology represent a changing body form that many scientists argue illustrates the budding signs of domestication. In addition, genetic analysis of ancient canid DNA points to south-west Asia and East Asia as the original centres of domestication. Although scientists are still calibrating the exact dates, it now seems clear that dogs emerged over 36,000 years ago in several locations independently. In other words, domestication long predates the transition from hunting and gathering to farming.  

Some supporters of the self-domestication idea have tried to salvage it by pointing out that palaeolithic hunter-gatherers killed large mammoths and herd animals like bison and deer, so they could have generated enough leftovers to attract and feed meat-hungry carnivores, including wolves. But this argument also has its problems. For one, we know Stone Age people were experts at using all kinds of animal resources and seldom left surplus waste, especially not close to where they were living. Besides, the practices of modern hunter-gatherers suggest that if our ancestors kept excess meat, they would have stored it away from scavengers on platforms or up in trees.   

An even bigger obstacle is research showing that wolves are often seen as dangerous pests. If they get too close and comfortable with humans, they will occasionally prey on young children or other vulnerable members of society. There are documented instances of people killing wolves when they feel threatened by their presence.  

It was evidence like this that turned Koungoulos firmly away from the self-domestication model. “I became pretty convinced that there are deep, consistent, almost structural barriers to self-domestication posed by the innate behaviour of wolves and the typical attitudes of traditional societies towards canids, which are, for the most part, rightly considered dangerous animals,” he says. “[Self-domestication] might make sense for some other domesticated species, but not for large carnivores like this.”  

If not self-domestication, then what?  

The appeal of puppies

One clue pointing to a different origin story comes from a growing understanding of wolf behaviour. Wolves are born blind and don’t develop eyesight until they are about 2 weeks old. During this critical period, they are highly adaptable and able to habituate to humans, which means they can form an attachment to a human caretaker, allowing pups to be safely nurtured and making it less likely they will attack anyone in the future. “If people are willing to put in the work, they can handle practically any type of canid as a companion,” says evolutionary biologist Raymond Pierotti at the University of Kansas, who has raised wolf pups himself. The key is to begin when they are still very young.  

Other clues can be found in the archaeological record. “Palaeolithic dogs are generally, but not always, found in Palaeolithic sites,” says archaeozoologist Mietje Germonpré at the Royal Belgian Institute of Natural Sciences. But they don’t just occur in close association with ancient humans and their settlements; there are also signs that these people had deep connections to dogs and other canids. At a site called ‘Uyun al-Hammam in Jordan, for instance, a fox was buried alongside two humans around 16,000 years ago. The excavators of this grave hypothesised the canine wasn’t a grave good, but a companion, buried together with its owners like part of the family. Numerous sites across Europe, Asia and North America suggest similar relationships.  

Germonpré’s work on Palaeolithic dogs has made her an early and prominent advocate of an alternative model of domestication that is now taking hold among a growing body of academics. It sees wolf pups slowly becoming domesticated after first being adopted by humans as pets. Stone Age people would have taken wolf pups from the wild and nurtured them to adulthood. Then, later, by selectively breeding those with the most desirable traits, domestication would have gradually been achieved. Germonpré calls this the human-initiative model of dog domestication.  

In fact, it isn’t a new idea. Probably the earliest and simplest version of it comes from Francis Galton, Charles Darwin’s cousin. Galton, a Victorian polymath most notorious for founding the field of eugenics, was an extensive traveller and documentarian. This, along with his many connections to the elite of his day, made him aware of the practices of Indigenous populations – including ones who took young wild animals as pets. Galton wrote about Indigenous peoples in North America who captured bear cubs and wolf pups, South Americans who caught and raised birds, and African populations who adopted young buffalo and antelope. This practice of taking young pets, thought Galton, could be the origin of domestication.  

Modern ethnographic reports paint a similar picture. There are accounts from Russia of groups, including the Khanty and Mansi, keeping fox pups as pets and later killing them for their fur. In North America, the Inuit regularly adopted bear cubs into their families, allowing them to play with children and even sleep in their igloos. The Siberian Ket families also adopted young bear cubs, particularly if they had no children. The Ainu of northern Japan and eastern Russia did something similar, adopting and raising young bear cubs to later sacrifice in ritual ceremonies. 

Anthropologists have recorded many instances of hunters taking young carnivores back to camp and these animals being breastfed by women, a practice also mentioned by Galton. We now also know the tendency for cross-species adoption isn’t even confined to humans. Dolphins have been observed to adopt individuals from other species, as have monkeys. This hints that the desire to affiliate with and care for the young of another species has deep evolutionary origins.  

Sacred animals of the Stone Age

However, the modern version of Galton’s hypothesis goes beyond humanity’s fondness for puppies. Germonpré became interested in the earliest manifestations of canid domestication when studying the relationship that Stone Age hunter-gatherers had with cave bears (Ursus spelaeus). Cave bear bones, including skulls, were often burned, painted with ochre and deposited beneath rock slabs, indicating that these fearsome beasts were imbued with symbolic and ritual meaning. This might have given people another reason to want to adopt and raise bear cubs from an early age. “My interest in the pet model as a hypothesis grew from there,” she says, “with the idea that other carnivores, such as wolves, must have had symbolic value for Palaeolithic people, together with other worths and utilities.”  

Archaeological evidence supports this idea. Many excavated prehistoric sites reveal that Stone Age people had a wide variety of uses for wolves. Their teeth were turned into ornaments, perforated skulls hint at ancient rituals and cut marks suggest people ate wolves and fashioned their long bones into tools. But perhaps the most important resource a wolf provided was its pelt.   

Like other species that live at high latitudes, wolves are adapted to thrive in cold conditions, and this includes sporting super-insulated fur with a mixture of long and short hairs. Historical reports recount hunter-gatherers living in northern latitudes across North America, Europe and Asia using wolf fur to line hoods, collars and other clothing. The Inuit historically captured wolf cubs, raised them and killed them for fur, which was also the fate of the Alaskan wolves adopted by the Iñupiat in Murdoch’s record. Likewise, wolf fur would have been a precious resource for people living during the last glacial maximum between around 26,000 and 19,000 years ago. For tens of thousands of years around this time, people had to endure some of the coldest and harshest climates of the past few million years. It is also the period to which many of the fossils of Palaeolithic dogs date.

Exactly how the process of domestication might have unfolded is unclear. “The specific Asian wolf that was ancestral to today’s dogs is gone forever, so the domestication of the dog can never be recreated under experimental conditions,” says Koungoulos. Nevertheless, he and others think there may have been parallels with a much more modern interaction between humans and wild canids. “One of the best analogues we have is the dingo and its relationship to Australian Aboriginal people – some of the only traditional hunter-gatherer peoples to have, until recently, maintained domestic relations with a wild canid,” he says.   

Living with dingoes

The dingo provides a clear illustration of what happens when small bands of mobile people adopt young wild canines but don’t selectively breed them. Until recently, Aboriginal Australians regularly captured dingo pups and cared for them, but then let them go once they reached adulthood. Dingoes haven’t become domesticated despite thousands of years of this sort of association with people. “But this modern example hints at how a long-term tradition of keeping wild-born canid pups as pets could alter the behaviour of the free-living population, or at least parts of it,” says archaeologist Adam Brumm at Griffith University in Australia.  

In a paper co-written with Koungoulos and Germonpré, Brumm speculates that when returned to the wild to breed, dingoes may have taken up residence near Aboriginal camps, forming a sort of human-associated subpopulation apart from other dingoes. Their pups then also tend to be the ones taken from nearby dens and adopted as pets. “Perhaps something similar happened tens of thousands of years ago with the grey wolf, giving rise to the first canids we would recognise as dogs,” says Brumm.  

We may never know for sure, but there are still looming questions that canid researchers hope to answer, including exactly where and when domestication originated. Germonpré wants to address these with further studies of ancient canid DNA. Whatever future studies reveal, however, the old story no longer seems plausible. “The self-domestication model still has a number of supporters and popular books out there,” says Koungoulos, “but my feeling is that, in the face of contrary evidence, they take increasingly fringe positions.”

Pierotti’s assessment is more direct: “Do not let yourself be beguiled by the Coppingers and their way of thought.”