Last year a raft of stories floated across the internet proclaiming that the questions of dog domestication would soon be answered. Those are the same questions every story worth its salt must ultimately answer—who, what, when, where, why, and how. Indeed, some writers on major publications went so far as to proclaim the who and how of dog origins were already settled. But something happened on the way to the unveiling, and we are still awaiting the promised answers to the mystery of the dog.
Why is that?
Well, It seems the more archaeologists and evolutionary biologists learn about the origins of the dog, the more questions they have despite a major, international collaboration that has been under way for several years now. (Large groups are notoriously hard to manage to consensus, especially when the egos are large, the evidence is sparse, and what there is appears not to match received wisdom.) Still, the project has uncovered more new material than the prevailing narrative can handle, and it is crumbling; but fully replacing it is difficult, in no small measure because people’s reputations rest upon the revealed truth. (One is reminded of the ways in which scholars who had spent their careers pondering the meanings of Michelangelo’s dark hues in the Sistine Chapel spent years decrying its “ruination” when cleaning revealed its bright, original palette.)
More specifically, the received wisdom in many English speaking countries is presented as a stark dualism—either Paleolithic people captured and tamed enough wolves to bring a population under human control; or wolves started following Paleolithic hunters and gatherers and through natural selection their descendants turned into a population of obsequious dump divers who endeared themselves to people who otherwise feared and loathed them.
Readers of this blog know that I am no fan of the self-domestication by self-taming theory, which I believe rests on two major conceptual errors.
The first holds incorrectly that between humans and wolves is an undying enmity. That means one or both had to change their nature for them to get together. According to the self-domestication theory the primary shape shifters were the wolves who turned themselves into perpetually juvenilized attention seekers, and that brings me to the second major conceptual error underpinning the standard model: The big brained creature in the dynamic is turned into a passive eejit who is taken in by a wagging tail and loose tongue.
Without repeating what I have said in my books and several blogs, including the two most recent ones, "Rats!" and "Becoming Dogs," the standard model also relies on an argument by analogy—that is, Russian geneticist Dmitri Belyaev’s domestication of silver foxes through intensive selective breeding solely for a characteristic he called “tameness.” That experiment is said by its fans to replicate the domestication of dogs and other animals.
The standard model’s specific problems date from its introduction when the wolves involved were said to have fed on the garbage dumps of Mesolithic villages. The problem there, of course, was that dogs arose in the Upper Paleolithic, if not before, when humans lived in bands of hunters and gatherers, well before the move to semi-permanent villages that marked the Mesolithic. To deal with that problem, defenders of dump divers proposed that self-taming wolves were fearless scavengers, who had taken to following the superior hunters and snatching their surplus from under their noses. The errors in that revision are almost too numerous to count. It very nearly is exactly backward.
By most accounts Paleolithic hunters and gatherers were not wasteful. They consumed what was edible and repurposed much of the rest, leaving little for the multiple scavengers who patrolled the edges of their camps—animals who, unlike dogs, did not join their society. Certainly, they didn’t leave enough to feed a growing pack of dogs, even if only a favored few were permitted to eat.
Ethnographic evidence suggests that the bands arriving in Europe for the first time were not initially skilled hunters of Pleistocene megafauna. It is far more likely that they learned to hunt from Neanderthals or Denisovans or some as yet unidentified early human ancestor they encountered. They could also have honed their skills by following wolves or following ravens, who still follow wolves on the hunt. The renowned Austrian ethologist Wolfgang Schleidt, has argued that anatomically modern humans became apex predators at the end of the last glacial maximum when they developed ways—the atlatl and the bow and arrow—to kill from a distance. Humans were profligate hunters, who tended to take animals in their prime, while wolves tended to cull the old (past reproductive age), young, and injured, ill, or infirm. In short, it appears that wolves were not only better at tracking and finding game but also better at animal husbandry. Good as wolves were at finding and cornering prey, they are not great killers. Humans closed the deal better.
In my view, the dog isn’t a lesser wolf, juvenilized, made safe for the home; rather, it is a different kind of wolf, one born on the trail and shaped by natural and artificial selection that have emphasized certain aspects of dogdom. Sociability more than any other characteristic, including tameness—the two are not the same—has long seemed to me the key to the transformation of dogs to wolves. An elongated socialization period and delayed onset of fear of the new allowed dog puppies to form lasting bonds with another species.
I have proposed in How the Dog Became the Dog that those bonds could have been forged between wolves of all ages who were inclined out of curiosity or their innate sociability to hang around people—and we know adult wolves can form strong social bonds with a variety of people and can be more sociable and gregarious than many dogs—and people, call them adepts who shared those characteristics. Over time, among a group or groups of people enough of these animals could have come together to form a reproducing pack of highly sociable doglike wolves. They would have been the camp hangers on favored with food—scraps from the “table.” The speed of the transformation is unknown, and in fact could well have varied from one population of dogwolves to another, depending on the degree of inbreeding or outcrossing to other wolf populations as they moved with their human companions throughout the land.
Living as part of nature, hunting and gathering people routinely adopted many kinds of animals, including wolves and, in Australia, dingoes, who are believed by some of the Aboriginal people to make them human. If the legends be true, the nurturing of human children by wolves also occurred. Even if the pups moved back among their kind upon reaching maturity—and some might well have stayed around—at the very least, these stories and reports point to the powerful and enduring bonds that have formed not just between individuals but between species as well. Those bonds do not spell natural enmity. That comes later, as a human invention.
It is from that perspective that I read a few weeks ago an article on a study Bridgett vonHoldt, an evolutionary geneticist at Princeton, and her colleagues at Princeton and Oregon State had just published in the journal Science Advances about finding genes responsible for hypersociability in dogs. I was particularly interested in the many news reports that quoted one or more of the collaborators speculating that domestication may have been a case of the friendliest dogs and friendliest people getting together. [See for example this CBC report.]
Then I looked at the paper and became troubled before looking again and deciding that if it breaks the stranglehold the standard model on our narrative of the transformation of some wolves to dogs, it will have accomplished a great deal.
That won’t be easy. The standard view provided the theoretical basis for this investigation. vonHoldt writes:
Hypersociability, one facet of the domestication syndrome, is a multifaceted phenotype that includes extended proximity seeking and gaze, heightened oxytocin levels, and inhibition of independent problem-solving behavior in the presence of humans. This behavior is likely driven by behavioral neoteny, which is the extension of juvenile behaviors into adulthood and increasing the ability for dogs to form primary attachments to social companions.
The view of dogs as perpetually juvenilized wolves has been repeatedly debunked. Some breeds have been shaped by artificial selection to resemble puppies, but that is human dictated. Wolves play into adulthood as a way of educating their young. In the paper, there is as well a troubling lack of rigor in the defining terms, such that “sociability” is sometimes used as a synonym for “hypersociability” and both are used interchangeably with “tameness.”
Essentially, vonHoldt and her colleagues associated structural variations—inserted or deleted genetic material, for example—in three genes believed involved in hypersociability in dogs with those believed involved in Williams Beuren syndrome in humans. More commonly known as Williams syndrome, it is a developmental disorder caused by the deletion of a region on Chromosome 7 that includes 26 to 28 genes and is characterized, according to the National Institutes of Health, by mild to moderate retardation, cardiovascular problems, distinctive facial appearance, and excessive attentiveness to others. Except in a few cases it is not inherited, raising the question of whether the genes they have identified in dogs might not be. vonHoldt and her colleagues refer to “excessive attentiveness to others,” as hypersociability, which they define as dogs paying relatively more attention to people than wolves hand reared from birth do. In total, 18 dogs representing breeds adjudged attention seekers and those considered to be aloof, and 9 wolves, were involved in the tests for hypersociability.
Examining chromosome 6, which is the site of the genomic deletions associated with Williams syndrome in canids, the researchers found that mutations in three genes associated strongly with hypersociability.The more variations there were in those genes, the more hypersociable, the animal was. More significantly, they had some ‘doglike wolves” in the genomes they sampled, although it is not clear whether they behaved more like dogs or wolves—an important point.
I won’t comment on the science itself, except to cite my genetics consultant, who has a Ph.D. in the field from Harvard and whose name I withhold to shield him from my errors in interpreting a difficult field. He points out the difficulty in defining hypersociability as a phenotype and the multiple problems, of which vonHoldt seems aware, involved in presenting a weak association of genes, or genetic variations, with a particular and complex phenotype as if they caused it. But association is not causation, he observes, and there is no guarantee they have the right genes or even the only genes involved, as would be the case if this were classic genetic linkage. “Our paper presents associations and clearly have not identified a functional consequence,” vonHoldt said in an email on July 23.
It would be interesting if researchers would begin to focus on the role of sociability in the birth of dogs, especially in light of evidence that adult wolves often can forge more solid bonds with people than hand-reared young wolves. To do that, they will have to define their terms and stop seeing wolves as enemies, threats to our existence. That any wolves have remained capable of forming a tight bond with a human after centuries of brutal persecution would be a testament to their strength of character and the centrality of sociability to their being, as well as to that of dogs.
Be sure to read the following responses to this post by our bloggers: