Could a Dog Truly Love a Robot? - Slate

The author of Dog Is Love: Why and How Your Dog Loves You responds to Andrew Silverman’s “Furgen.”

It doesn’t take any special technology to see that dogs love people. Hildegard von Bingen, in the 11^th century, noted that “a certain natural community of behavior binds [the dog] to humans. Therefore, he responds to man, understand him, loves him and likes to stay with him.” It could fairly be said that, like Othello, dogs love not wisely, but too well. Their loyalty to our capricious species has seen dogs led into wars, ill-fated Arctic expeditions, and many other tragic misadventures.

But are there limits on dogs’ capacity for love? What Andrew Silverman has been able to do in his wonderful tale of Caro (human), Tucker (dog) and Furgen (machine) is explore in fiction the possible limits of dog love. Could a dog love a box? Would a dog follow a piece of clever machinery off into the sunset the way dogs have followed humans to the furthest corners of the planet?

In Dog Is Love: Why and How Your Dog Loves You, I mustered all the scientific evidence I could find for the contention that what makes dogs special is their exceptional readiness to form strong emotional bonds with members of other species—ourselves in particular. And yet, although it is obvious that dogs form attachments very easily, the limits of that capacity have hardly been explored by behavioral scientists. There was a study back in the 1960s, when ethical standards were very different than what they are today, in which researchers raised litters of pups without any human contact except for 30 minutes a day for one week. Some pups first met people in their third week of life, some in their fifth, some in their seventh, and so on. The crucial point is that each group of pups had just one week of experience around humans, and even during that week they only saw people for half an hour per day. Once all the pups had reached 14 weeks of age, the researchers tested how comfortable these dogs were around people. Even though the pups had received such limited experience of humankind, most of them were pretty relaxed with people. Only dogs that received no human exposure at all until they were 14 weeks old were too shy to eat while people were watching. The researchers recorded that those dogs were “like little wild animals” and euthanized them.

No one today would entertain a study that would end up with dogs that could not accustom themselves to human proximity and had to be put to sleep, but that doesn’t explain why this fascinating question has not been the focus of more research. We know from the millennia-old practice of raising dogs with livestock that puppies will form strong attachments with whatever species of animal they are around early in life. In Australia there are even dogs that take care of penguins on an offshore island. Before the dogs came to live with them, these penguins were routinely decimated by foxes that ransacked the island at low tides.

To understand what domestication did to dogs and bolster the claim that there is something remarkable about dogs’ willingness to form attachments to other species, we need to compare dogs to wolves, the beast from whom all our dogs are descended. And yet, although the many people who have hand-reared wolves all agree that it is a lot more difficult to get a young wolf than a dog pup to accept you as a friend, there is a complete absence of systematic study of exactly how much exposure to a person a wolf pup needs to form bonds with people. Standard practice for hand-rearing wolves, established through decades of trial and error, is to keep the pups with surrogate human mothers 24/7 from around 10 days of age for several months. Even then, the bond that the wolves develop with humans is never as easy and promiscuous as what we take for granted with dogs.

A lot of what we know about how animals form strong attachments dates back to the work of Konrad Lorenz—the only Nazi ever to win a Nobel Prize. It was Lorenz who, in the 1930s, first systematized how young animals bond with the first living thing they see. He called this process Prägung, rendered into English as “imprinting”—emphasizing the automatic and apparently irreversible nature of this connection. It took an American to demonstrate that imprinting can occur toward inanimate objects. Howard S. Hoffman raised ducklings so that the first thing they saw after hatching was a foam cube mounted on model train tracks. Hoffman demonstrated that if the cube paraded back and forth on the tracks in front of the new-hatched ducklings, they would develop an attachment toward it. Just like the lost little bird in P.D. Eastman’s charming children’s book Are You My Mother? (published just as these studies were going on), they showed their emotional connection by crying when their foam cube “mother” disappeared.

While Hoffman’s research showed that ducklings could form attachments to foam cubes, he also demonstrated that mother ducks and other living forms were more effective imprinting stimuli. Only the absence of all other possibilities rendered the ducklings desperate enough to form an emotional bond to an inanimate cube on a model rail track. Consequently, if the dog in Silverman’s story, Tucker, had been a duckling, it is very unlikely that a smart box—no matter how charming—would have been able to alienate his affections. But we can’t be sure. As we love to end our papers, “more research is necessary.”

At the present time, the marketplace of dog training devices is rapidly filling up with ever-smarter pieces of machinery. Furbo, for example, will let you see what you dog is doing through a built-in camera, and it can deliver treats too—all through an app on your smartphone. GoBone is a plastic “smart bone” that entertains your dog through playful movement. Companion is a quite Furgen-like device that uses built-in cameras, a treat dispenser, and machine learning algorithms to actually train your dog to perform various behaviors like sit, down, and more. A real-life Furgen may not be so far off. As clever tech starts competing for our dogs’ affections by gently and generously leading them with treats and consistent praise, we may have to try harder to hold onto our dogs’ love. Given the choice between a generous and affectionate cube and a human who wants to draft him into war, a dog may indeed choose the inanimate but smart technology—as Tucker did. Some competition for dogs’ affections might nudge our species to treat the canine kind better. And that would be no bad thing.

Future Tense is a partnership of Slate, New America, and Arizona State University that examines emerging technologies, public policy, and society.

Adblock test (Why?)