From Wolf to Woof

02nd April 2020

Dogs might be man’s best friend, but 15,000 years ago they may not have seen us as friends but as potential dinner. New DNA science reveals that inside even our sweetest pooch lurks an inner wolf. So how did our beloved pets make that genetic leap?

Words by Marek Kohn

  • dog

  • dogs

  • goodwoof

  • magazine

  • estate


Many kinds of animals have found ways to live with humans, whether by humans’ rules or against them, but no other animal has got the measure of us the way dogs have. Dogs have lived with us longer than any other domesticated creatures. They live with us on all the continents of the planet except Antarctica (from which they are banned, for fear of passing distemper to seals), and in total they may number up to a billion. There may be nearly ten million of them in the UK alone, according to PDSA (People’s Dispensary for Sick Animals) figures – which show an increase of more than 20 per cent in the five years to 2019. We house them, feed them and even clothe them. Yet when we contemplate a French bulldog trotting through a city park in a purple onesie, we may well ask: How did this triumph of evolution come about?

Scientists asking that question also have to ask when dogs began to emerge, and where. Genetic studies have provided them with a wealth of data – but the data raise more questions than they answer. James Serpell, who studies the relationships between humans and their companion animals, ruefully noted in a book surveying the state of canine science, The Domestic Dog, that researchers still don’t agree on where, when, or how many times dogs were domesticated. It might have been in Europe; it might have been somewhere in Asia; it might have been once, or it might have been many times. “Dogs seem to be particularly resistant to us trying to uncover their deeper history,” observes Greger Larson, who researches ancient DNA at the University of Oxford.


Attempts to trace their lineages by surveying modern dogs’ DNA are frustrated by their prolific history of interbreeding, whether with mates selected for them by humans – to reproduce or accentuate qualities that endear or make them more useful to us – or with mates they encountered on their travels with humans, or even with wolves on the fringes of human settlements. DNA from long-dead animals can help scientists see past this confusion and detect prehistoric traces of what might be the first dogs. “We’re getting closer,” says Larson. “We do have some hints about some things.”

There’s no doubt that dogs’ ancestors were wolves, rather than jackals or other members of the same family, the Canidae. Indeed, the relationship is so close that many scientists now classify dogs as a subspecies, Canis lupus familiaris, of the grey wolf, Canis lupus. But what those ancestral wolves were like remains shrouded in mystery. “One thing that’s absolutely clear,” Greger Larson says, “is that the population of wolves from which dogs were derived appears to be extinct.” Genome studies suggest that wolves underwent a drastic population collapse, losing much of their genetic diversity in the process. The wolves that gave rise to dogs may have had genetic qualities that are absent from modern wolves, and that helped them become domesticated.

Many researchers are comfortable with a date of around 15,000 years ago for the start of the domestication process.

Many researchers are comfortable with a date of around 15,000 years ago for the start of the domestication process. One dissident is Mietje Germonpré, a Belgian palaeontologist, who has argued that remains more than 30,000 years old found at several sites across Europe are those of dogs. But their resemblance to dogs doesn’t necessarily imply that they were domesticated; it might just illustrate the lost diversity of ancient wolves.

Whenever domestication happened, though, it couldn’t have been easy or straightforward. The two species had very good reasons to keep well away from each other. Humans and wolves were competitors for food, and humans were at risk of becoming food for wolves.  A visiting observer from Mars would hardly have marked humans and wolves down as potential associates, allies, or future best friends forever. And yet wolves turned into the first domesticated animals, several thousand years at least before humans took control of goats, sheep, pigs and cattle.

Broadly speaking, there are two ways to imagine how it might have happened. One is as a project directed by humans, who initiated domestication when they realised that they might benefit from keeping canines. The other is as a gradual coming together in which both parties developed the relationship, because both gained benefits from it. In this perspective, wolves might have been the initiators, approaching human groups and behaving in ways that allowed the association to develop. They could be said to have domesticated themselves.

Either view presents glaring difficulties. In one scenario, wolves started to scavenge from the refuse dumps that built up near human settlements. They gradually lost their fear of humans and drew closer to them, ending up as their companions. But James Serpell doesn’t buy it. “We know what happens when wolves lose their fear of humans – they kill people!” he says. “I just don’t see our Stone Age ancestors being stupid enough to encourage any kind of potential predator to be scavenging around their villages. There are children running around. It would make no sense.”


He considers that “Francis Galton got it right back in the 19th century when he said it was a product of pet-keeping”. Galton, a discipline-spanning Victorian scientist and statistician whose controversial legacy includes the concept of eugenics, suggested that domestication had arisen from the practice of capturing young wild animals and raising them as pets, which was widespread among peoples he and his contemporaries called “savages”. Serpell argues that a wolf cub taken at an early enough age would come to regard the people around it as its own social group, and so would pose no threat to them as it grew bigger. Indeed, it would help protect them from wild wolves.

The only way to find that out would have been to try it, though, and people would surely have been loath to take a chance on a member of a predator species – particularly one that, being of medium size, is likely to target children if it tries to prey on humans. Two species with such potential for conflict would need a lot of time to get used to each other. It might have been easier to overcome the suspicions if the genetic diversity of ancestral wolves had included more variations than present-day wolves carry, of the kind that made some of them ready to be friendly towards humans.

Such variations would certainly have been favoured as the canines adjusted to their new ways of life, and as people began to choose their mates for them, selecting ones that had amenable dispositions. Those choices might have had effects on the dogs’ physical forms as well as their behaviour. Domestication has been linked to a range of changes in various species, from reductions in brain size to shortened jaws, floppy ears and curled tails.

Scientific thinking about such observations has been powerfully influenced by the work of Dmitry Belyaev, a Soviet geneticist who conducted an intensive domestication programme on captive foxes. Selecting the tamest individuals for breeding, he produced a population of foxes that was not only tamer, but also included individuals with features such as floppy ears, curly tails and dog-like faces. It looked as though selection for tameness was acting on genes that shaped the development of physical traits as well as temperament.

The story may have been too good to be the whole truth. Greger Larson contributed to a critique, published last year, which suggests that claims about the so-called “domestication syndrome” have been overstated. For one thing, the foxes were from Canadian stock, acquired in the 1920s for Soviet fur farms. They had probably been selected for tameness, if only unconsciously, for decades before Belyaev began his study. Nevertheless, Larson still thinks there’s something in the theory. “Belyaev’s experiments nicely demonstrated that by selecting for a behavioural trait you can get a whole suite of traits that kind of come along for free,” he affirms.

At least one of these had shown up by the time some ancient hand engraved images of dogs on a cliff in what is now north-western Saudi Arabia. Shown in hunting scenes with humans, they look like the modern breed known as Canaan dogs, with distinctive curled tails. The artworks could be 8,000 or 9,000 years old. Decorations on pottery from south-western Iran, made around 6,000 years ago, depict dogs that resemble modern salukis.

Domestication has been linked to a range of changes in various species, from reductions in brain size to shortened jaws, floppy ears and curled tails.

Extinct Breeds: Five breeds that are gone, but not forgotten



The talbot was a common hunting hound in England during the Middle Ages. Some believe it was brought over from Normandy by William the Conqueror, although there is no definitive evidence to support this. A small to medium-sized dog with a white coat, short legs, long drooping ears and a long curled tail, the talbot featured in much of the art of the period (as well as on the signs of public houses) and, along with the greyhound, was uniquely used in heraldry.

An ancestor of beagles and bloodhounds, the breed disappeared around the late-18th century. Its legacy lives on in the names of many English inns and pubs: the “Talbot Arms”.

Old English bulldog

Believed to be descended from ancient war dogs such as the old mastiff or the alaunt, the old English bulldog was used for bull baiting and dog fighting in London in the early-19th century. The passage of the Cruelty to Animals Act in 1835 led to a decline in the sport, which eventually led to its extinction. Despite dog fighting becoming illegal, the activity continued for many years and in order to create a superior fighting dog, breeders created a cross between the old English bulldog and old English terrier. This new breed, called the “bull and terrier”, was a precursor to the Staffordshire bull terrier, English bull terrier and American pit bull terrier and speeded the end of the old English bulldog.

Turnspit dog

With a long body and short legs, the turnspit dog was bred to run on a wheel, called a turnspit or dog wheel, that allowed meat to cook evenly over a fire. Mentioned in Of English Dogs in 1576 under the name “Turnespete”, it was also known as the kitchen dog, cooking dog or underdog. 

As technology progressed, the  job became obsolete and the breed became rare by the  1850s and extinct by 1900.

Cumberland sheepdog

The Cumberland sheepdog was similar to the Welsh sheepdog and old working collie types and is thought to be an ancestor of the Australian shepherd. Black with a white blaze and a heavy, dense coat, the Cumberland was believed to be the favourite breed of the 6th Earl of Lonsdale.

It existed in his family for more than 100 years, but by the start of the 20th century, it had been absorbed by the border collie.

Alpine spaniel

Famed for their thick coats and large size, Alpine spaniels lived in the bitterly cold climate of the Swiss Alps, where they were used in mountain rescues by the Augustinian Canons, who ran hospices in the region around the Great St Bernard Pass. Sadly, disease wiped out this breed in the mid-19th century. However, modern-day St Bernards are their genetic descendants as a result of cross-breeding with Newfoundlands, and proudly bear the name of the place of their ancestors.


Other types of dog emerged over the course of millennia. A statue now on display in the British Museum affirms that muscle-bound, mastiff-like dogs, known as Molossians were a formidable presence in ancient Greece and Rome. Shakespeare had Macbeth observe that the variety of dogs included “hounds and greyhounds, mongrels, spaniels, curs, shoughs, water-rugs, and demi-wolves”. (Shoughs were a kind of lapdog; water-rugs a type of spaniel.) But today’s vast diversity of breeds – the Belgian-based Fédération Cynologique Internationale, or World Canine Organisation, recognises around 350 – is largely the product of the past couple of hundred years, in which dog breeding became an organised pursuit. And as well as the breeds recognised by the kennel clubs, crossbreed varieties such as cockapoos and labradoodles enliven modern canine diversity still further.

For geneticists, one of the most significant aspects of breed creation is that it involves small populations, which may be more vulnerable than large ones to harmful genes. As the farm fox story shows, breeders who select for one sort of characteristic may inadvertently be selecting for a number of other traits that tag along with it. A dog with a trait that breeders favour may end up with large numbers of offspring – many of whom may inherit any less desirable traits that it also happens to possess.

That, however, means that studying dog genetics may be of far more than just historical interest. An international collaboration called the Dog10K Consortium aims to analyse 10,000 dog genomes, ranging from pedigree specimens to the free-breeding village dogs that make up much of the world’s canine population. Wolves, coyotes and jackals will also be included for comparison. The project aims not only to explore intriguing questions about the origins of dogs and their breeds, but also to shed light on health in both dogs and humans. We have a lot of illnesses in common: heart disease, cancer, diabetes, and also behavioural problems such as obsessive-compulsive disorders.

Researchers involved in the collaboration (who include Greger Larson) argue that insights into the genetic factors underlying disease in dogs may translate into insights about human conditions. They envisage a new role for dogs, as models for humans in matters of health. That would be a new departure, after hunting, herding, guarding, companionship and all the other human needs that dogs have met. But once again, dogs would demonstrate their uncanny knack of finding a place in the most important parts of our lives.

Top Dogs: Doggy popularity contest


There are few clear stats for a canine hit parade, as the Kennel Club’s records cover registered dogs only, and recognised breeds, so wildly popular new crossbreeds such as the labradoodle don’t feature in its annual league tables. What’s clear, though, is that our dogs of choice are always changing.


For many years the UK’s most popular dog, last year the Labrador retriever retained its number one position in the Kennel Club’s league table – with more than 35,000 registered. However, there are regional variations in the doggy popularity contest, and the French bulldog is still the most popular breed in London.


“It must be a nuisance to go through life with a Father Christmas moustache,” wrote Vita Sackville-West in her book about dog breeds, Faces: Profiles of Dogs, “but no doubt the schnauzer gets used to it.”  In 1961 when Faces (recently re-released by Daunt Books) was written, the schnauzer “was not so often seen in this country, though a smaller edition… has recently begun to find favour.” Today the UK is home to almost half a million miniature schnauzers.

English pointer

Our canine cover star, the English pointer, is sadly in sharp decline – although the German shorthaired pointer’s numbers are up. A few other old favourites, the boxer and the Staffordshire bull terrier, are also down.


Today’s highly popular crossbreeds were too new in 1961 to appear on Sackville-West’s radar and their numbers are difficult to accurately assess today, but a recent Pets4Homes survey placed them in the following positions: cockapoo – number 4; cavapoo – number 18; labradoodle – number 22.

  • dog

  • dogs

  • goodwoof

  • magazine

  • estate

  • Goodwoof guests 2022. Ph. by Chris Ison.

    Latest News

    Gorgeous gifts for both hound and human

  • sheep-dog-goodwoof.jpg

    Latest News

    Award-Winning Sheepdog Teams Demonstrate Ancient Skills at Goodwoof

  • dogs-trust-main-image.jpg

    Latest News

    Goodwoof partners with Dogs Trust