I wrote this as part of Animal Inkhaven. It's fairly speculative.
In 1959, a Soviet geneticist called Dmitry Belyaev started breeding silver foxes on a farm in Siberia. What looked like a fur farm was in fact a study of how wolves had become dogs.
Each generation, his team scored every fox on how tame it was, and bred the tamest. Within a few generations, some of the foxes were wagging their tails at the researchers. But over the following decades, something strange happened: they grew floppy ears, curly tails, white patches in their coats, and shorter, rounder snouts. Some even barked. They were bred for specific behaviours, but genetically, some seemingly unrelated physical traits came along for the ride.
One thing many people don’t appreciate about evolution is that mental (“cognitive”) and physical traits are not really different from an evolutionary perspective. To evolution, both are built by genes interacting with the environment. Both vary between individuals, and genes almost always play a role in those differences. And both respond to breeding (“selection”) in the same way.
Just as evolution shaped humans’ and animals’ bodies, it’s shaped our minds. Almost all mental illnesses have a genetic component; for example schizophrenia is about 79% heritable, a similar sized genetic component to height.
In fact, cognitive and physical traits are often “genetically correlated”: the same underlying genes affect both. For example, certain genes for disordered blood sugar, blood fats and body measurements also play a role in anorexia.
Now, the Silver fox case has been disputed, but the concept is well accepted: when you breed an organism for behavioural or cognitive traits, you often change their physical traits too. Vice versa: selecting an organism for certain physical traits can change their psychology and behaviour. And that’s what worries me.
Purebred dogs have many health problems. But they also have problematic breed behaviours: Labradors that always feel hungry, Chihuahuas that are psychotically aggressive, Italian greyhounds that are extremely needy.
Take Labradors. About a quarter of Labradors carry a gene deletion that may make them hungry all the time. The gene, POMC, makes several hormones: ones for appetite, one for coat colour, a stress hormone, and beta-endorphin, the body’s own painkiller. The Labrador version knocks out one of the appetite hormones and the painkiller. Imagine being a labrador, and feeling ravenously hungry all the time, never satisfied. Some labs are so hungry they’ll eat golf balls, hair bands, even stones.
Did we accidentally do this? One study found that the prevalence of this mutation in labrador service dogs was 76%. Why? Well, service dogs are often trained with treats, so maybe breeding the best service dogs also selected the dogs most responsive to treats, which accidentally selected more dogs with broken hunger genes.
In our breeding of dogs for specific physical and sometimes behavioural traits, have we accidentally bred in negative mental traits? I worry that some breeds may carry hidden suffering we can’t see. Potentially a lot of suffering.
Sure, we treat our dogs very nicely, and they live easier lives than in the wild. But that doesn’t guarantee we’re morally safe. Placing someone whose mind is broken in a 5 star hotel does little to make them feel better.
Artificial breeding can create “out of distribution” traits, exaggerated beyond what we’d see in nature. We can breed traits into animals that nature never would. Skull shape varies about as much across dog breeds as across all other carnivores, and the most extreme breeds differ more than any two carnivore species. A collie and a Pekingese differ more than a cat and a walrus. If bodies can go that far, minds plausibly can too.
Seriously broken minds that would never occur in nature are possible with artificial breeding. Evolution would probably never design an organism whose natural state is 100% mental anguish and suffering all the time. Suffering is something evolution deploys “tactically”, to motivate an organism to avoid death or other things bad for its “fitness” (i.e. having lots of healthy kids). When we artificially breed an animal, we have no such constraints. Evolution has a brake on suffering, and breeding takes the brake off. Flinders Sensitive Line rats were bred for sensitivity to a chemical, and are now so prone to depression they are used to study it in the lab.
Have we accidentally done that to purebred dogs?
There’s a current conversation in the animal movement about engineering animals for traits, including the controversial idea of modifying farm animals so they don’t suffer.
Firstly, it’s important to note that genetically engineering animals is not the same as selective breeding. Genetic engineering is very precise: you can change one or a few genes keeping others the same. So it would appear to eliminate the problem I’m talking about.
On the other hand, there’s still risk. Changing one gene can still have unintended effects if that gene controls lots of different traits. And most do. People born with two broken copies of the SCN9A gene apparently feel no pain, but also have no sense of smell. Mice missing two enzymes that work in a brain area behind the unpleasantness of pain lose the pain hypersensitivity that follows injury, but also have impaired memory. And genetic engineering can engineer for traits the species has basically never had, so we could get some extremely weird mental traits.
I honestly don’t know. But if it is, it’s pretty worrying. There are a ton of purebred dogs out there.
This was really easy to read with an interesting core idea, thanks for writing this up!