Wednesday, July 27, 2016

Coppinger on dogs, their evolution, their behavior

Raymond Coppinger's thinking on the evolution of dogs, and on dog behavior

"If dogs disappeared from the face of the earth tomorrow, humans would survive the tragedy without much stress. But if humans disappeared tomorrow, dogs would likely become extinct shortly thereafter. "

A review of two books:
Dogs: A Startling New Understanding of Canine Origin, Behavior & Evolution; Coppinger and Coppinger[1]
How Dogs Work; Coppinger and Feinstein

The value of a theory lies in whether one can use it to accurately predict what will happen. If you have to modify the theory, after the fact, to fit the facts as they become history, the theory is not of much use.

Raymond Coppinger's books contain a  number of ideas that work as useful theories. However, while quite a lot of Coppinger's writing is filled out with entertaining and enjoyable anecdotes, I found his ideas were often buried in obtuse and opaque writing technique. It took me some effort to tease out "what is your point?" One could miss the point fairly easily, and you could feel that the writing was technical and dry, if you weren't "just getting it".

Coppinger's primary concept in the first book, and the most transformational, in my opinion, is the nominal subject of that book (Dogs: A New Understanding . . .). The concept is simple. Mankind, at some point in history, made one of our "one small step for a man, one giant leap for mankind" moves, when we began building semi-permanent communities. As an aside, did you know that cities are one of the greatest inventions of mankind, possibly the single greatest invention of all time? But the first thing that happened was villages. And villages created waste. That waste was a food source for other critters.

Some wolves evolved in ways that enabled them to use this food source, and they became dogs. Self-domesticated, if you will. Domestic cooperation as a survival technique. The old Aesop's fable about the dog and the wolf, where the wolf rejects servitude in favor of liberty - but that liberty leads to starvation - takes on a new meaning. The dog as evolved village wolf scenario is blindingly more simple, by far, than any scenario where man intentionally domesticates a wild wolf. As Coppinger says in the first chapter of "Dogs", " . . .It takes a herculean effort to get a wolf to develop in a way that makes it possible for the animal to be a tractable companion of humans. It's hard not to tame a dog".

Coppinger moves on to other concepts, more or less in support of his (supposedly) primary theory about dog evolution. He talks about how cross-breeding occurred, historically and naturally. My notes say something about the historically recorded types of dogs that we find in written history, as opposed to breeds: herding, guarding, etc.

He continues into the realm of behavior. And, even after reading two of his books, I am not entirely clear on why he thinks that what he describes is so significantly different from current thinking as he obviously wants us to believe. Here is one point I wrote down: 'behavior is not genetic, but it is inherited, and a dog can't learn anything that it doesn't inherit.'

Speaking of difficult, how self-contradictory and confusing is that?

But let's continue to some other ideas he presents.
behavior fits into pre-determined motor patterns
behavior is dependent on age and develops at certain ages
motor patterns (nature) can be suppressed, or encouraged, by age timing (nurture)
motor patterns are affected by breed
breed individuality is dependent on selection for certain motor patterns

What he has to say about the flexibility of physical attributes, and the flexibility of behavior (or lack of it), is, like the evolutionary bit, transformational, but to a lesser degree. What I think he says is that anatomy is not hard-coded. It is, instead, based on hard-coded points, around which the rest of the anatomy is adaptive. (E.G. Eyeball size is hard-coded. The eye socket development depends on the eyeball size. The palate is hard-coded. The rest of the jaw and the skull grows into place to fit that palate.) Behavior follows the same pattern. Some bits are hard-coded, others follow, or not, based on how everything else develops. Development is time-sensitive, and environment sensitive. I.e. Instinctual behavior can be (sometimes) extinguished if ignored during a certain period of a young dog's life. He argues that behavior is less flexible than is typically thought, but that it is still flexible, within certain parameters.

I believe the gist of what he wants to say on this topic is found in these two sentences:
" . . .a good way of beginning to understand the fundamental insight of ethology: that a behavior is a physical trait just like any other taxonomic character of a species (or a breed). "
" . . .behavior is always a synergistic result of both (nature and nurture) . . ."
And, these two sentences are a good example of why I am criticizing Coppinger's writing style. I had to get all the way to the end of his 2nd book to find those two simply stated points. They would have been far more useful to the reader had they been placed up front. And, while this topic is interesting by itself, I don't quite see how it relates to the thesis of his first book, or how it is particularly different in practice than more traditional concepts of instinctual behavior. The topic is relevant to the second book, where it is developed at greater length than in his first book.

The timing of onset behavior is a useful topic for breeders, trainers, and owners of young puppies. I am again, not quite sure here what is new, or different. However, I will say that Coppinger, in his rather difficult way, does set these points out more clearly than training volumes I've read over the past few years.

And, he pretty clearly sets out this point: there is a differentiation between classical reward conditioning, and instinct based behavior pattern training. Instinct based behavior training requires no reward because it is internally rewarding. Dunbar and Clothier have talked about this, mostly rather indirectly - but reading Coppinger I got it. Clearly. And, this point is an exceptionally useful one to understand when training dogs!

Another written bit that I found to be extremely useful for me, to understand my dogs behavior, is the instinctual predatory sequence. For Coppinger it is, I think, a minor tool, used to describe a larger concept, as a sidebar of his behavior arguments. However, I found this sequence to be another that is exceptionally useful.

Many breeders and working dog trainers will instantly recognize parts of the sequence, but I have found very little written directly about them elsewhere! (In spite of having been reading training books extensively over the past few years.) Take this sequence: ORIENT > EYE > STALK > CHASE > GRAB > KILL > DISSECT > CONSUME. Every predator has that sequence. Some of the details vary. GRAB can be GRAB-BITE , FOREFOOT-STAB, HEAD-SHAKE, FOREFOOT-SLAP, and others.

What makes a herding dog a herding dog is that they are bred for hyper-development of EYE > STALK > CHASE > GRAB. With little or no "KILL" and reduced "DISSECT". Livestock guarding dogs are bred to have nearly NONE of those. A livestock guarding dog should be incapable of even a low level "DISSECT " (and this is generally so). A pointer will have strong "EYE", but little "CHASE". Not all working dogs are created the same.

That sequence has been invaluable to me. Because I know that sequence, I can recognize "EYE" in a dog. And I can predict what is coming next. Experienced shepherds know these things, too, but I found little in writing that told me about these things. I had heard of "EYE" from reading about training herding dogs - but had no idea what they were talking about until I got the sequence. Then I knew exactly what they were talking about. What I found interesting is that I actually recognized this sequence on first reading what Coppinger wrote about it. I read about this sequence decades ago, but then it was a piece concerning the learning behavior of cats. It never occurred to me that the predatory sequence would be so much the same in dogs. But I knew a cat had to learn to stalk, and that was a separate learning from "catch", which was also separate from "kill", and ditto "consume".

Thus, as I was watching a training session last week, I could see that the dog exhibited no "EYE". The only "EYE" the dog exhibited was to watch its human, which is not really the same thing at all. While I could see this was a smart dog who will eventually "get" the tricks necessary (for him) to herd, that owner has a hard road ahead.

In his first book, Coppinger digresses from his main point early and often. I think the behavioral and genetics points are meant to be supportive, but I found them somewhat distracting. But perhaps he intended to discuss several points. His sections seem more like digressions than a contiguous whole. Example: at one point in the first book, he looks at the morality of adopting dogs - and when it might not be the moral answer some think it is. But, another idea, and I think it is another important idea, comes up when he discusses the importance of cross-breeding. It is clear from his writing that he thinks closed stud-books (lookin' at ya, AKC) are a mistake. That they inevitably lead to genetic, maladaptive issues. Like German Shepherd hips. Open stud books and cross-breeding leads to healthier, stronger dogs.

This same topic is also a prime example of why I found Coppinger's writing style annoying. I repeatedly found myself going back over a section trying to figure out what his point was. In the case of cross-breeding, it was buried in front of me, but still buried. He insists on using the word "transhumance". If you knew the meaning of that word before reading either this, or Coppinger, then more power to you. I know my vocabulary is highly exceptional, yet I had never heard the word. What he means is simply "seasonal herd (or livestock) migrations". The unfamiliar and little used word "transhumance" makes his meaning difficult to follow, because the word is so unfamiliar. If he had just used the far more readily understood phrase "seasonal livestock migrations", his meaning would come through far more quickly and clearly.

Coppinger has written a relevant book, with some very useful ideas. The evolution of dogs is a burgeoning area of study. The Coppingers were early to the field, and at this moment, may well be the most knowledgeable experts in existence. I notice that his ideas have created discussion, and blowback. However, similar to how Porter's ideas on competitive advantage transformed thinking for the fields of economics and business, Coppinger's ideas are transforming the nature of the discussion about dogs.

Separating the dog from the wolf means that looking at wolves as a yardstick to measure dog behavior (or eating habits) is ultimately not very productive, and certainly not worthy of the center stage position we today find the whole "dog as wolf" set of ideas. I've also noticed that some people want to cling to their Alpha-wolf alpha-dog concepts. I've never found the "you have to show the dog who is boss" idea to be particularly useful in training. Not that at times you don't have to use some discipline, but trying to teach a dog how to behave by being dominant doesn't make much progress. And taking the "wolf" out doesn't mean there isn't dominance. It just means that the patterns we should be looking for are less likely to be found in wolves. I also found some reviewers, who, having read the book, don't really understand some of Coppinger's points. Which could well be because Coppinger's presentation is difficult and opaque.

For further reading, pro and anti:

An interview with the Coppingers:

Another scientist's rather negative review. He seems to me to be more irritated by Coppinger's obtuse writing style and perhaps their manners vis-a-vis the scientific publishing world, than by the concepts Coppinger presents.

One in the anti-crowd, clinging to Alpha wolf (modifying the theory to fit the facts):

A recent review of Coppinger's latest book, which presents some of the best reasoning supporting the village dog as the original "super breed"

A recent study titled "Origin of mongrels", also supportive of the village dog as a "super-breed".

[1] The volume's title changed in later printings, dropping the use of the word "Startling" in the title. The copy I used is the 2001 Scribner printing. I have used a shortened version of the full title to refer to the book as a matter of convenience. Going from a one-word title to an eleven (or twelve) words, including sub-title, is a bit inconvenient.