Dogs and more dogs - More on why dog breeding is not a model for evolution

Recently, I featured comments on Richard Dawkins's (possible non-) review of Michael Behe's Edge of Evolution for the New York Times - in which Dawkins tried to evade bad news for Darwin's theory by appealing to human dog breeding.

In that post, I featured comments from Dr. David DeWitt, who pointed out that dog breeders select animals for juvenile canine traits that people happen to admire.

(Most people prefer to live with a happy puppy rather than the Big Bad Wolf.)

But that has nothing to do with survival in nature and does not confer or lead to new traits in canines. So, contrary to Dawkins's claims, it cannot be used as direct evidence for Darwin's theory.

(No one doubts that all sorts of changes may be possible - in theory. But which changes play a role in the development of life forms in nature itself? The weird choices of dog breeders hardly count there.)

After I posted that item, some correspondence followed from friends in science. I am posting some of it here, with permission:

One friend, who works in a lab, writes,

... many biologists use the mouse as a model organism of choice, and we use particular inbred strains.

Like inbred dogs, they are dumber and more prone to particular types of tumors/defects, so while they are more homogeneous in behavior from mouse to mouse, they are also just less fit.

I bring them up because like the observed experiments on E. Coli [ in Edge of Evolution], we are treating these organisms as relatively static. When I order a C57Bl/6 mouse from Japan, I don't worry that reproductive isolation has produced one that is different from the one here in Maryland.

And when we introduce mutations in mice, we don't assume that they'll evolve and improve upon them. If those genetic alterations were to disappear in a generation (which is only ~3 months in mice), people would make much note of it. But mouse colonies in research institutes throughout the world are telling us that this does not occur.

Does that mean that lab mice can never change? No! He adds,

it has been reported that sometime prior to 1984, a mutation did develop in the C57Bl/6 strain:
"A naturally occurring deletion in nicotinamide nucleotide transhydrogenase (Nnt) exons 7-11 occurred in C57BL/6J sometime prior to 1984. This deletion results in the absence of the NNT protein, and is associated with impaired glucose homeostasis control and reduced insulin secretion. This mutation is not found in C57BL/6JEi, C57BL/6N, C57BL/6ByJ, C57BL/10J, C57L/J, or C58/J (Toye AA, et al, Diabetologia, 2005)."

So the mutation was a health problem for the mouse.

Another friend writes to say that the dog genome has not been affected in any signfiiant way by Darwinian changes:

Following talk about dogs by Dawkins, there has appeared an excellent article about the dog genome in American Scientist (September-October 2007): Genetics and the shape of dogs by Elaine A Ostrander (National Human Genome Institute of the NIH). This details much of the research on the dog genome by her and her team and others in the field.

It is well written with excellent illustrations. It appears that most of the variations between breeds of dog are incipient in the dog genome with little due to recent mutations. I think this was already known but this article deals with it in an authoritative way. I hope Dawkins reads it!

You hope Dawkins reads it? But wasn't the whole point of his review to discourage people from reading Mike Behe's Edge of Evolution? Just today I was talking to a science teacher who had started to read it - and was not slow to recognize the devastating attack on ultra-Darwinism that it represents.

Also, another friend writes: Following talk about dogs by Dawkins, there has appeared an excellent article about the dog genome in American Scientist (September-October 2007): Genetics and the shape of dogs by Elaine A Ostrander (National Human Genome Institute of the NIH). This details much of the research on the dog genome by her and her team and others in the field.

Dog breeding - proof that Darwin was right? Hardly, says prof

In his review of ID biochemist Mike Behe's Edge of Evolution, which caused many to wonder whether he had actually read the book he was reviewing, Richard Dawkins indulged in a long and seemingly irrelevant riff on dog breeding. He hoped to convince his readers that complex and fantastical intracellular machines come about by chance (and mind comes from mud) on account of the vast variety that humans can produce by selective breeding of dogs.

Correspondents have pointed out that Dawkins is counting on his readers' ignorance of a fundamental fact about dog breeding- that is depends on existing traits and does not introduce new ones. One writes, for example,

The problem is that the variety of dogs obtained through breeding programs is an example of the variation possible within the dog genome, but (and this is a very big 'but') there are natural limits to variation.

Darwinism predicts that there are no taxonomic limits to variation. However, every breeding experiment of the last 100 years that attempts to see how far variation can go (E. coli, drosophila, etc.) always encounters limits beyond which further change is not possible. Thus, the fundamental prediction of Darwinian theory has been consistently falsified in a century's worth of experimental testing. Dog breeding, itself, encounters these limits.
The bottom line is that dog breeding, and the observed limits to variation within dogs, falsifies the most important prediction of Darwinian theory.

What is he talking about?
Another correspondent, David A. DeWitt, author of Unraveling the Origins Controversy, enlightened me further,

Many of the traits for different dog breeds are examples of neoteny.

Neoteny refers to the maintaining of juvenile characteristics into adulthood. Mutations can prevent proper development and maturation. Even though particular traits might seem like they are novel, in such cases it is really a loss of information since the animal has stunted development in one trait.

This is why some breeds of dogs are so cute and look like puppies even though they are full grown (Jack Russel, Shitzu etc).

Well, that makes sense. The disgusting little freaky-poos that infest my neighbourhood are really just immature? Makes sense, all right.*


I wrote back to ask,

David, is there not also some distortion involved, maintained by selective breeding? I am thinking in particular of the Basset hound, the bulldog , and the dachsund. Do these distortions not shorten life in many cases?

Also, the single most important trait in domestic dogs is that the animal not be aggressive around humans. (That would be the fastest way for a dog to get himself a one-way trip to the vet's office.) But that means selecting for a trait that would NOT aid survival in nature.

The breeds that are commonly trained to BE aggressive toward humans (intentionally) are wolfhounds like German Shepherds. But they have the most characteristics in common with wild animals like wolves.

In other words, domestic breeding not only does not employ natural selection, but it selects for traits that would not be chosen in any process that favoured survivability. Is that correct?

He replied, with a long, careful answer:

Pure breed dogs often do have shorter lives than "mutts". Presumably, this is because of severe inbreeding. The result is that mutations for particular diseases/defects become concentrated.

So when people have selected for those traits that comprise the poodle breed, they have also inadvertently selected several serious genetic defects. Certain breeds are prone to the same diseases and early causes of death.

Regarding the lack of aggression in dogs...this is also considered an example of a neotenous trait (juvenile traits that persist into adulthood).

When wolves are very very young, they are not so aggressive. Many of the behaviors of our dog breeds are also neotenous. There is plenty of information about this on the internet. The less a dog is physically like a wolf, the less aggressive the dog.

The most important thing to understand about dog breeding is that there is not new genetic information (from mutation) that is being supplied. Through breeding, humans are either shuffling genes that pre-exist in the population (like different poker hands from the same deck) or preserving mutations that amount to developmental defects.

While developmental defects can look like new traits (short stubby legs or a short snout for example or a Chihuahua that looks like an embryonic dog) they are not new at all since it is simply preservation of a previous stage.

Another example of a neotenous trait would be a mutation that leads to webbing between fingers in a human. During development, the cells between the fingers are supposed to go through a process of programmed cell death (apoptosis). If the cells do not die (because of a mutation), then the remaining tissue would be webbed fingers.

Since all human babies go through such a stage, it would not be a new trait even though it looks like it. It is preservation of a previous developmental stage because of a mutation in the normal developmental pathway. This highlights another aspect to the limits of Darwinian evolution.

Often, dogs are considered an exception because they are so "plastic". In reality, it is just that we have been able to preserve a wider array of developmental defects. Dawkins pulled a real bait and switch trick when he criticized Behe's Edge of Evolution using dog breeding. Dog breeds highlight the limits of evolutionary change, but Dawkins used the diversity of dogs (from developmental defects) to rebut this fact.

However, since most people do not understand the preservation of juvenile characteristics, they can be fooled into thinking that evolution really can produce new traits.

Hmmm. We hear plenty about Darwin's natural selection, but almost nothing about neoteny. And, to the extent that Dawkins was counting on our ignorance of neoteny, why SHOULD he bother to read Edge of Evolution before discouraging others from reading it?

(*Thanks, Dr. DeWitt! I've been looking for years for a way to insult the local infestation of little canine swine without being cruel. Like, neighbours, please, if you're going to have a dog, have a dog. Otherwise, be a cat person like me.)