Natural History

One of the problems in figuring out when and where dogs were domesticated is that people are forever searching for a single domestication event.  We have been able to figure out the time and place of origin for virtually every other domestic species, but when it comes to dogs, well, the dates and places are everywhere.

The initial mtDNA study that tried to calculate when dog and wolf lineages splits put the origin at 135,000 years ago, also known as a very, very long time ago.  When the dog genome was sequenced, it was estimated that the split occurred 27,000 years before present. The oldest dog remains that are no longer contested are from Russia and are 14,000 year old. 14,000-year-old dog remains have also been found in Germany, Switzerland, and the Czech Republic.

There is also the 31,000-year-old Goyet Cave “dog,” which is called a dog only because its skull morphology fits that of a dog much more than any wolf. It was found in Goyet Cave in Belgium, and unlike the other ancient dogs, its mtDNA was analyzed and compared to modern wolves and dogs. It did not share its matriline with any extant wolves or dogs, but because of its shortened muzzle, it still could be a dog. One should be a bit cautious before declaring this animal a “dog,” for it was not associated with any modern human remains or artifacts.

However, a recent article published in PLoS ONE found an even older possible dog at the Razboinichya Cave in the Altai Mountains of Siberia.  This would place the first possible morphologically distinct domestic dogs in Central Asia. This particular article, unlike the one on the Goyet Cave “dog,” clearly suggest that this animal is probably not ancestral to any domestic dogs, and the main reason why there is a gap between the remans of the Goyet Cave  and Razboinichya Cave dogs and those 14,000-year-old European dogs is that the last Glacial Maximum disrupted the domestication process.

That suggests that dog domestication had a lot of stops and starts. There was no single domestication event. Rather, dogs were domesticated several times, and these dogs then reverted back into the wild wolf population. This process probably occurred for tens of thousands of years.

Now, that’s what the archaeological literature says about dog domestication.

The various assays of DNA have all put dog origins as much older than the 14,000-year-old mark. The 14,000-year-old dogs from Russia have also been dated to as early as 17,000 years before present, so even these dogs could have been older than when the supposed modern dog lineages evolved. I should also point out that the 14,000-year-old dogs have never had their mtDNA or any nuclear DNA examined to see if they are related to modern dogs or wolves. It is very hard to find DNA any sort from animal remains of this antiquity, but as I mentioned earlier, it was possible to sequence the mitochondrial DNA sequence from the Goyet Cave canid, which was twice the age of the other European dog remains.

Now, I’m sure you’re all going to jump at the suggestion that dogs were domesticated in Central Asia or Western Europe. None of the genetic studies suggest that modern were developed in either of those places. The original studies that compared mtDNA diversity within domestic dog populations put the origins of the dog in East Asia, which was then refined to China south of the Yangtze. Of course, mitochondrial DNA studies examine only maternal lineage, and these particular studies were largely falsified with an important genome-wide analysis that was peformed at UCLA. This study used “SNP chip” technology to compare dog and wolf DNA sequences at 48,000 different places in their genomes. It found that modern dogs were more closely related to Middle Eastern wolves, but there were some European breeds that showed some European wolf ancestry and some East Asian breeds with affinity with Chinese wolves.

This finding suggests multiple origins for domestic dogs, which is exactly what the researchers concluded with Razboinichya Cave.

So it seems that this consensus is being confirmed between the more sophisticated DNA analyses and archaeological studies.

However, it doesn’t explain why dogs have such a strong relationship with Middle Eastern wolves, but keep in mind that the even if dogs are largely derived from Middle Eastern wolves, it doesn’t automatically follow that they became morphologically distinct there. Robert Wayne at UCLA points out that the first domestic dogs would not have been morphologically distinct from wolves at all, so it is possible that all of these wolf remains that are associated with humans that don’t have dog features are actually those of the earliest dogs.

So it is possible that tame Middle Eastern wolves could have followed humans into Central Asia and Europe for centuries. Occasionally they crossbred with native wolves and with tame wolves from other parts of Eurasia. If the Goyet Cave canid is an early dog, then ancient Europeans would have had populations of tame wolves, and it is likely that East Asians also had their own populations, too. For whatever reason, the ancient tame wolves that became modern domestic dogs are mostly derived from the Middle Eastern wolves.

The Altai dog study has some problems. Like the Goyet Cave canid, it was not associated with human remains, but that alone should not disqualify it for being a dog. However, within the literature, there is always an assumption that a shortened muzzle is a sign of domestication.  Domestic dogs, in general, have shorter muzzles than wolves, so it is believed to be a trait that defines a domestic dog or to be an effect of domestication within canids. I am a little skeptical of this assumption for the simple reason that wolf skull morphology varies greatly even within the same population. It is possible that there were fully wild wolves with short muzzles, just as it is possible that there were tame wolves that were essentially dogs that had normal wolf-like muzzles.  The latter would particularly be true if the wolves were eating raw meat from large prey species.  When modern wolves are kept in captivity and fed commercial wild dog diets, dog food, chicken, and butchered meat, their skull morphology becomes very similar to that of dogs. If these ancient tame wolves were eating from large carcasses, they would still have the environmental pressures on the muscles and bones in their heads to retain much of the wild wolf’s skull morphology– and they would not appear as morphologically distinct in the archaeological literature.

However, it is still possible that the shortened muzzle in both the Goyet Cave and Razboinichya Cave canids are the result of some level of domestication, but I don’t think the authors in the Altai dog study have made the convincing case that the dog domestication was interrupted by the Last Glacial Maximum. The fact that the Goyet Cave canid did not share mtDNA lineages with modern dogs does not mean that dog domestication was somehow interrupted. If the Goyet canid is a tame wolf that was becoming a dog, it could have been part of a population that was replaced by another tame wolf population. The revelation from the dog genome that modern domestic dogs split from wolves 27,000 years ago strongly suggests that dog domestication was not interrupted by the last Ice Age at all.

But archaeologists just deal with physical remains, and the background assumption that the first dog will have a shortened muzzle might mean that they could miss the first dogs. They might call the remains of a canid that is clearly associated with humans “a wolf” simply because it has a long muzzle. This is a big methodological error that I don’t think is ever fully considered in the archaeological literature, which is one reason why I am not impressed with the comparison between the Razboinichya Cave dog and thousand-year-old Greenlandic dogs. Skull morphology in dogs and wolves can evolve very rapidly, as anyone who has followed the history of modern dog breeds clearly understands. For example, if one looks at the early bull terriers, they had heads that looked nothing like the modern examples of this breed.

The truth is we have a very poor understanding of what ancient wolves were like.  One thing we do know from the Goyet Cave “dog” study is that wolves were originally much more genetically diverse than they are today, and we also know that the ecosystems of Eurasia were much more diverse and dynamic than they are today. It is possible that wolves adapted to fit all sorts of different niches within those ecosystems, and they evolved different adaptations to fit these niches.

But it is still possible that both the Goyet Cave and the Razboinichya Cave canids are actually early dogs. It is very much possible.

The genetic literature suggests that the clear split between dogs and wolves happened tens of thousands of years ago, and it is possible that the shortened muzzles really were signals of some level of domestication.

I don’t think the Last Glacial Maximum interrupted dog domestication at all. Instead, it is possible that the end of the Last Glacial Maximum was responsible for creating the features we associate with domestic dogs within these tame wolf populations. However, there will be more about that later.

We like to think of dogs as household pets, but they were with us long before we had houses.

***

It should be noted that not a single one of the oldest dog remains is 12,000 years old. At one time, it was said that the dog was domesticated during the Natufian, for in the Levant, there were massive dog burials.  There was also a burial of a human with a puppy that was dated to 12,000 years before present.  This is the Bin Mahalla site, and the puppy buried there was once considered to be the oldest dog. Some archaeologists still call it the oldest dog, including the researchers who recently discovered a buried fox in the Pre-Natufian in Jordan— even going as far to say that man domesticated the fox before the dog.  Of course, there is no connection between that ancient Jordanian fox and the farmed foxes in the Belyaev experiment, so it’s kind of a silly claim.

The Natufian culture was a proto-agricultural society.  They were starting to do some small scale farming, and they were staying villages.

These dogs and the economic and cultural conditions of the time were the main inspiration behind the Coppinger theory of dog domestication. Raymond Coppinger holds that the dog originates from this time, and it derives from wolves that evolved to scavenge in these early farming villages.

These wolves evolved shorter flight distances, which means they would allow people to approach them more closely before they will run. If a wolf can allow people to approach them, then they can be more effective scavengers. Over time, these wolves evolved smaller brains, weaker jaws, and smaller size to accommodate the less nutritious but more secure diet in the villages.

This theory is a theory of self-domestication, and it fairly easy to explain. That’s probably why it is so popularly understood, even if some of its background assumptions are in error. After all, dogs are older than the 12,000 year date that Coppinger holds posits their origins, and if they evolved 12,000 years ago, it actually would have been difficult for dogs to have appeared in the Americas. All Native Americans derive from East Siberian people who came across the Bering Land Bridge 12,000 years ago. If dogs were domesticated in the Levant at the same time, exactly how would they have made it to the Americas? East Siberia is not adjacent to the Levant in the least. Further, Coppinger assumes that ancient wolves were just like modern wolves in their critical periods for socialization, which may not be the case. He also assumes that it is impossible to tame adult wolves, which is also not the case.

This theory, which is so widely reported in the literature and is now such a popular meme within the Western dog culture, is actually based upon several faulty assumptions.  Coppinger uses this theory to proclaim a species status for the domestic dog, even though it is not a current accepted practice in modern taxonomy to give domestic animals full species status that is separate from their wild ancestors. Coppinger, to his credit, rejects dominance theory, but to reject dominance theory, he has created another myth– the one that says the dog isn’t a wolf. It’s just a village scavenger.

The truth is these older dog remains clearly put the origin of dogs within ancient hunter-gatherer societies, and dogs or wolves on their way to becoming dogs have been living with people for at least 20,000 years before Coppinger’s proposed domestication event.