Towards a New Model for Dog Domestication

German shepherd mating Carpathian wolf

Original crossbreeding between a German shepherd and Carpathian wolf to found the Czechoslovakian Vlcak.

If you have been reading this blog for a long time, I used to post historical and indigenous accounts of wolves, coyotes, and dingoes being used as working animals. I also would post accounts different breeders of domestic dogs crossing their stock with wolves to improve their strains.

I have long been critical of the Raymond Coppinger model of dog domestication, which posits that wolves scavenging from Neolithic dumps created the dog as an obligate scavenger that then became selectively bred for human uses. In this model, the tropical village dog is the ancestral form of all canines, a position that has emboldened the “Dogs are not wolves” theorists to suggest some tropical Asian Canis x is the actual ancestor of the domestic dog.

This model also posits that all dogs are just obligate scavengers, and unfortunately, this obligate scavenger designation means that what could be otherwise good books and research on dogs essentially denies their predatory behavior.

Last year, I kept hearing about a book that took on Coppinger’s model head-on. This book took Coppinger’s task for having distinct Eurocentric biases and that Coppinger essentially ignored vast amounts anthropological data on how different human societies relate to wild and semi-domestic canids.

So I finally ordered a copy of this book, which is called The First Domestication: How Wolves and Humans Coevolved by Raymond Pierotti and Brandy Fogg. I do recommend this book, but I readily admit that I don’t agree with quite a bit of it. I agree with it more than Coppinger, though, because they rather clearly show massive holes in Coppinger’s model.

Pierotti and Fogg have produced a model that relies heavily upon humans and wolves encountering and then benefiting from a hunting mutualism. Humans have a long history as scavengers, and even today, there are people who follow large predators, including lions, to rob them of their kills. Dholes are targeted by certain people as well, and it is very likely that humans entering Eurasia would have done the same with wolves.

The difference between the lions and the dholes and ancient wolves is that the lions and dholes resent having humans come near their kills.  The ancient wolves, however, came to work with people to bring down more prey. These wolves and humans came to be the dominant predators in Eurasia.

Pierotti and Fogg’s model posits the domestication process as beginning with ancient hunter-gather societies. It relies upon the wolf’s predatory nature as an important catalyst in allowing this partnership to thrive.

Further, the authors are rather clear in that our Eurocentric understanding of a clear delineation between wolves and dogs is a rather recent creation. Most cultures who have existed where there are wolves and dogs have a much more plastic understanding of the differences that separate the two or they have no separation of all.

The most compelling analogies in the work are the discussions about the relationships among hunters in Siberia, their laikas, and wild wolves and the relationships between indigenous Australians and dingoes.

In the Siberian laika culture, the dogs have extensively exchanged genes with wild wolves, enough that laikas and wolves do share mitochondrial DNA haplotypes. The laikas (or laiki, as they are known in Russia) do hunt the sable and other small game. They also protect the camps from bears, and in some areas, the laikas are used as not particularly specialized livestock guardian dogs. The authors see these dogs a very good analogy to describe how the earliest people and dogs would have lived. These dogs would have been cultured to humans, but they would still be getting an influx of wild genes as they lived in the wild.

In the dingo example, the authors discuss how these hunter-gather cultures would keep dingo pups and treat them almost exactly as we would our own domestic dogs. They also would use the dingoes to hunt kangaroos, but during mating season, they would allow their companions to leave the camps or stay.  They often would leave, but some would go off for a time in the bush and return. This suggests that early humans might not have forced their socialized wolves to stay in camp and that relationship could have been a lot more libertarian than we might have assumed.

These relationships are very different from the scavenging village dogs that Coppinger contends were like the original dogs. These animals are not obligate scavengers. They are hunters, and what’s more, it is their hunting prowess that makes the relationship work.

Further, the authors make a convincing argument that we can no longer use the scientific name Canis familiaris, because many cultures have relied upon wolf-like dogs and dog-like wolves for survival.  These animals are virtually impossible to distinguish from each other, and therefore, it would make sense that we would have to allow dogs to be part of Canis lupus.

The authors contend, though I think rather weakly, that dogs derive from multiple domestication events from different wolves. I remain fully agnostic to this question, but I will say that full-genome comparisons of wolves and three dogs that represent three distinct dog lineages suggest that dogs represent a clade. They are still very closely related to extant Canis lupus, especially Eurasian ones, and still must be regarded as  part of Canis lupus.  Therefore, one does not need multiple origins for domestic dogs from wolves to make the case that they are a subspecies of Canis lupus.

I am, however, quite glad to see that the authors reject this Canis familiaris classification, even if I think the reasoning is better explained through an analysis that shows how dogs fit within a clade called Canis lupus than one that relies upon multiple origins.

Also, one should be aware that every argument that one can make that says dogs are wolves can be applied to coyotes to suggest that they are wolves. Wolves and dogs do have a significant gene flow across Eurasia, but coyotes and wolves have a similar gene flow across North America. The most recent ancestor between wolves and coyotes lived 50,000 to 70,000 years ago, which is far more recent than the proposed divergence between Old World and North America red foxes and the divergence between Qinling and other giant pandas.

I really have no problem thinking of coyotes as being a form of Canis lupus in that a pug is a form of Canis lupus. All the acceptance of this classification does is allow for a positing that this species Canis lupus has thrived because it possesses both phenotypical and behavioral plasticity.

The authors, however, would have a problem with my classification. They make regular reference to red wolves, which have clearly been shown to be hybrids between coyotes and wolves, which themselves are probably better regarded as divergent forms of a phenotypically plastic species. They also contend that coyotes and people have never formed relationships like people have formed with wolves, because coyotes are too aggressive.

However, I have shown on this space that coyotes have been trained to do many of the things dogs have, including pointing behavior. They also have ignored the enigmatic Hare Indian dog, which may have been a domesticated coyote or coydog.

But that said, I think the authors have clearly shown in their text that dogs and wolves are part of the same species.

The authors also make some controversial arguments about dog paleontology and archaeology.  One argument they rely upon heavily is that wolves could have become behaviorally very much like dogs without developing all the morphological changes that are associated with most domestic dogs. Some merit certainly does exist with these arguments, but it also puts paleontology and archaeology in a position that makes it impossible to tell if a wolf-like canid found near human camps is a truly wild animal or creature on its way to domestication.

This argument does have some merit, but it still will have problems in those fields of study, because it becomes impossible to tell semi-domesticated wolves from wild ones in the fossil and subfossil record.

However, the authors do make a good case, which I have also made, that argues that the original wolf population had no reason to show fear or aggression towards people. The best analogous population of wolves to these original ones are those found on the Queen Elizabeth Islands of Northern Canada. These large arctic wolves have never experienced persecution, so they are quite curious and tolerant of the humans they encounter. Wolves like these could have easily been the basis for a mutualism that would eventually lead to domestication.

The authors also contend that the reason wolves in Europe are reviled is the result of the Western church’s propaganda that was working against traditional totemic animals of the pagans. Wolves were among those totems, and the church taught that wolves were of the devil.

However, I think this argument is a bit faulty, because Europeans are not the only people who hate wolves. Many pastoralist people in Asia are not big fans of wolves, and their hatred of wolves has nothing to do with the church. The traditional religions of the Navajo and Hopi also do not hold the wolf in very high regard, and these two cultures have been in the sheep business for centuries.

Further, we have very well-documented cases of wolves hunting and killing people in Europe. These wolf attacks were a major problem in France, where notorious man-eating wolves were often named, and they were not unknown in other parts of Europe as well.

The authors focus heavily on the benign relationship between wolves and people, including the wolf that hunted bison calves and deer to feed survivors of the Sand Creek Massacre, but they ignore the stories that do not posit the wolf in a good light.

The reason wolves in Eurasia have sometimes taking to hunting people is really quite simple:  Eurasia is a land where people focused much more on domesticating species to create animal agriculture. Agriculture has a tendency to reduce biodiversity in a region, and when people kill off all the deer in an area to make room for sheep, the wolves turn to hunting sheep. If you live in a society in which people do not have ready access to weapons, then the wolves start targeting people. Feudal societies in Europe would have been open target for wolves living in such ecosystems. By contrast, the indigenous people of North America, did not domesticate hoofed animals for agriculture. Instead, they managed the land, often with the use of fire, to create biodiversity of which they could hunt.

The authors do show that dogs and wolves are intricately linked animals. They show that dogs and wolves are the same species. They use many wonderful anecdotes of captive wolves and wolfdogs to make their case, and in making this case, they have made the case clear that dogs are the produce of hunter-gatherer societies and still are conspecific with the wolf.

I do, however, have some quibbles with some of the sources they use in the text. For example, when they discuss Queen Elizabeth Islands wolves, they focus on an account of Elizabeth Marshall Thomas on Baffin Island. She was on Baffin Island for one summer and observed one wolf pack. She is a fine observer of animals, but much of her analysis about dog and wolf behavior is still controversial. The authors also regularly make reference to Cesar Millan as a dog expert, when virtually no credentialed dog behavior expert thinks he is, and to the notorious dogsbite.org website, which is of even more contentious. These authors are making serious and well-reasoned arguments about dog and wolf behavior and relying upon these sources detracted from the work. I would have liked if they had referred to L. David Mech’s wolf observations on Ellesmere or to John Bradshaw as an expert on dog behavior.

I also had some issues with their contention that the Ainu people of Japan are Turkic or Altaic. No one knows exactly who these people are, but they are interesting in their relationship with wolves. Traditional Japanese society, distinct from the Ainu, was actually quite similar to the Siberian cultures that have produced laika dogs that still interbreed with wolves. However, I don’t think anyone still thinks that the Ainu are Turkic or Altaic.

Finally, the authors do make a good case against Coppinger’s model, but they go on to accept Coppinger’s fixed motor pattern dependence model to describe breed specialization. It is certainly true that Coppinger was Eurocentric in his understanding of dog domestication, but both Coppinger and the authors are Anglocentric in their understanding of dog hunting and herding behavior. The authors think this is Coppinger’s strongest argument. I think this is among his weakest.  This model states that pointing, herding, and retrieving are all just arrested development of a full predatory sequence. A dog that can point just stalks. It never learns to use its jaws to kill. A border collie stalks but also engages is a type of chasing behavior. It will also never learn to kill. A retriever will run out and grab, but it lacks the killing bite.

The biggest problem with this model is that everyone knows of border collies that have learned to hunt, kill, and eat sheep. I had a hard-driven golden retriever that would retrieve all day, but she would kill rabbits and even fawns.

The Anglo-American concept of specialized gun dogs affected Coppinger’s understanding of their behavior. He never really looked into continental HPRs. For example, Deutsch-Drathaars, the original German variant of the German wirehair, are bred to retrieve, point, track, and dispatch game. Such an animal makes no sense in this model, for it would suggests that an animal that would point would only ever be stuck in that stalking behavior. It would never be able to retrieve, and it certainly would never use its jaws to kill.

A better model says that dogs are born with a tendency to show behaviors, such as exaggerated stalking behavior that can be turned into pointing through training. There are countless stories of pointing dogs that suddenly lost their pointing behavior after running with hard-driving flushing dog. The dog may have been born with that exaggerated stalking behavior, but the behavior was lost when it entered into social interaction. Indeed, much of these specialized hunting behaviors are developed through training, so that what actually happens is the dog’s motor patterns are refined through training rather than being solely the result of being arrested in full.  This is why all the old retriever books from England tell the sportsman never to allow his dog to go ratting. As soon as that dog learns to use its jaws to kill, it is very likely that this dog will start using its jaws on the game it is sent to retrieve.

Despite my quibbles and reservations, Pierotti and Fogg have made a convincing case for the hunting mutualism between wolves and humans as the basis for the domestication of dogs. I was particularly impressed with their use of ethnography and non-Western histories to make their case. I do recommend this book for a good case that we do need a new model for dog domestication, and the questions they raise about taxonomy should be within our field of discussion.

Natural History

Posted in Pet Care And Pest Control Articles | Tagged , , | Leave a comment

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>