I would wager that among the one thousand most frequently uttered phrases in the United States of America are the words, “Good dog!” These words are, of course, the common praise/reward a dog receives when it does something its owner approves of. They are typically coupled with a scratch on the head, a pat or an outright bribe of food. It is estimated that there are some 200 million pet dogs in the United States, demanding a good deal of human praise daily.
All domestic dogs trace their ancestry back to a handful of gray wolves in Asia, and the newest thoughts on how they became domesticated is that they undertook it themselves, beginning about 14,000 years ago. People were beginning to settle down, creating small villages and, therefore, garbage middens. The wolves least scared of humans would sneak into the garbage heaps to eat, and before too long they would be self-selecting for tameness and propinquity to humans.
In short, dogs came about with the urge to ingratiate themselves to us. The most ingratiating would get the best treatment. So dogs have an inborn tendency (which wolves definitely do not) to want to please humans. And who disagrees with the late Charles Schulz that happiness is a warm puppy? Your dog pays an amazing amount of attention to you, reading your body language, your moods, your words, your habits — just as would a lover — in order to please you, to comfort you if you are down. That’s why getting a dog to obey is a whole lot easier than getting a possum or bison to obey, or, for that matter, a house cat.
Besides simply obeying, dogs can learn to do all kinds of proactively good things for human society, such as learning to see for the blind, to sniff out narcotics at the border, to guard property, to alarm approaching burglars. The list goes on and on and includes rescues from avalanches and sniffing out corpses and, amazingly, even cancers. Veterinarians insist that patting a dog lowers one’s blood pressure.
Well and good, but the question remains: Are dogs intrinsically good in the sense of ethics? After all, if a dog learns by rote or by a cold and abstract conditioning to do good things, it can be argued that all or most such acts arise only from the dog’s perception that it will be rewarded. This perception (and hunger for praise and treats) is sufficient for many dogs to eschew acts that lead to punishment, even of the mildest sort, such as a deep-voiced “BAAAD dog!” (There may even be dogs who, when the folks are out, don’t get up on the sofa or swipe some food from a countertop, though I’ve never met one.)
Have they no shame?
Scientists have long assumed that dogs and other beasts are pretty much biological automatons that, with proper conditioning, can learn to achieve rewards by following orders. Dogs, this mechanistic school of thought suggests, do not know shame, embarrassment, remorse or any other complex emotion such as love, pride, fun or satisfaction in a job well done.
We have bribed them to be good, many still say. They are not intrinsically, ethically good. Some even argue that dogs are really evolutionary con artists, making of themselves beloved parasites on humanity — albeit parasites that can be trained to rise above sheer parasitism and, say, fetch the newspaper every morning for the Master. In fact, I knew a dog who — once trained to bring the newspaper to his Mistress — went out every day to collect all the newspapers along the Mistress’ street and bring them to her.
All of this put-down of what makes dogs lovable to people may be wrong. One of the reasons scientists have long thought dogs were automatons was that they couldn’t figure out any way to learn what was going on in a dog’s head. That is now changing. More sophisticated analyses of the minds of many animal species — from monkeys and apes to elephants, dolphins and dogs — are beginning to open doors on their mental differences from and, importantly, their mental similarities to us.
This could mess up a lot of philosophical and theological and moral theories about the animal kingdom, including the species that — perfectly obviously — resides at the pinnacle of the Great Chain of Being, more recently dubbed the evolutionary Tree of Life and even more recently turning out to be a rather Tangled and Shapeless Bush.
In other words, if without human intervention a lowly dog can be shown to act in an ethical manner in some instances, on its own exhibiting empathy and fairness, we have what you might call a whole new ballgame.
I should point out here that many kinds of animals, from apes to dolphins and whales to even parrots, show a remarkable ability to make use of something analogous to a few of human language’s many complex facets, but that is not what we are talking about here. Among other things, we are talking about untrammeled dog behavior, especially how dogs play.
People almost always smile when watching two dogs — usually young ones but by no means always — cavort, bumping each other, bouncing, biting each other’s necks, running, rolling over and generally having a wonderful roughhouse. Biologists certainly took note of this aspect of animal behavior but mostly didn’t think it was worth studying because they could think of no earthly evolutionary purpose it could serve. It seemed basically a waste of energy, possibly even dangerous. It wasn’t training for adult fights because adult dogs who play presumably don’t need continuing education in fighting. (Cats who, as kittens, never got to play with another kitten, still grow up perfectly capable of fighting and hunting.)
Another reason biologists steered clear of play is that it looks like fun, and attributing a sense of fun or any other human emotions to nonhumans was considered a scientific sin, unprovable one way or the other.
The first breakthrough came when a young ethologist at the University of Colorado, Marc Bekoff, began filming the play of dogs (and coyotes) in the 1970s and then, frame by frame by thousands of frames, analyzed their every move.
Bekoff was studying the structure or form of dog play, specifically play-fighting. It is an axiom in biology that form and function are inextricably linked: all legs, for example, are for locomotion, and no leg could pump blood through the aorta. There was some evidence that the axiom applied to animal behavior as well. Bekoff figured he might be able to get at the real function of play-fighting from analyzing what form it took. Certainly there had to be some “rules” to play-fighting, since it rarely erupts into real hostilities.
Bekoff found that there is a great deal more happening in play-fights than a random roughhouse. He described such play as a kind of dance. Typically the dance always begins in the same way, as structured as a barn dance with its do-si-dos or a ballroom tango or a waltz. It begins with one dog approaching another and performing what’s now called a play-bow. In this maneuver, the dog puts its front legs out, its chest near the ground and its head lower than its playmate’s, all the while keeping its rear end high. It may bark and wag its tail. From that posture, the dog can run or leap in virtually any direction as it launches the melee. Or the other dog may respond to the play-bow signal by launching the non-hostilities itself.
The play-bow also maintains the mood necessary to keep on every participant’s mind that even though the last bite on your leg was a bit rougher than I meant, we’re still playing, right? Watch dogs playing, and when things seem to be getting awfully rough, like a bite followed by head-shaking (the way a dog destroys a stuffed animal toy), the shaker will throw in a quick play-bow. The play-bow is evidently always understood.
Other moves in play-fights are also what might be called stereotyped or ritualized. The feints and sideways leaps, the soft-mouthed bites, mounting, hip slams and racing around each other — all these moves are sort of, but not exactly, what takes place in actual fighting. More than likely, it is all less a matter of learning how to fight as an adult than developing the overall motor and neurological activity needed to be a successful dog in general.
Evidence exists that dogs who as puppies do not get to play have smaller and maybe less complex brains. They also are likely to be ill-at-ease with other dogs, if not downright aggressive toward them. But there’s more to it than that.
Hey, no fair
What if someone cheats? Say Dog A play-bows and Dog B responds. As play goes on, Dog A cheats, biting aggressively and severely hurting Dog B. Such things happen from time to time, and the usual outcome is that Dog B henceforth avoids playing with Dog A. Dogs are not stupid.
What Bekoff found most remarkable about the play-fighting of dogs is that the biggest and strongest of the contestants do not always dominate or win. They often let the wimpier members prevail. They take a fall, at least momentarily. They throw the fight, at least for a few rounds.
In other words, the dog that is superior handicaps itself — but to what end? One effect of such sacrifice is to prolong the period of play. If Dog A always loses, it might grow discouraged and quit. Bekoff also proposes, however, that the play-bow and the self-sacrificing may imbue in dogs a sense of fairness, a rudimentary form of right and wrong, otherwise called morality.
Fairness? In a recent series of experiments at the University of Vienna, Friederike Range rewarded dogs with a food treat if they held up a paw. Then when a lone dog was asked to hold up its paw, did so, and didn’t get a treat, it would keep on trying as many as 30 times. But when two dogs together were tested, with one of them not receiving a reward, the dog who was unrewarded made a big scene and soon refused to play. “Dogs,” said Range, “show a strong aversion to inequity.”
Fairness, inequity, unequal treatment — these all imply that a dog has a sense of itself, a degree of self-awareness. The fairness that a play-fighting dog shows to its playmate suggests that it has a sense of itself and its playmate as different individuals. This is the first step toward the mysterious quality only humans possess (so far as we know) called consciousness.
Taking it one more early step further, do dogs empathize with each other? This is a tricky region of inquiry. It has been suggested that rats have empathy for others if they are cage mates. Rats who saw their cage mates in pain were more sensitive to pain than those who were tested alone. The researcher, Jeffrey Mogil of McGill University, spoke of an “emotional contagion” between the animals. Others have pointed out that this might be no more a specific form of empathy than a newborn baby crying when it hears others crying.
When it comes to more highly social animals like wolves, dolphins, elephants and dogs, there is plenty of anecdotal evidence of empathy among them, such as a Diana monkey teaching another one how to use the slot machine that rewards them with a morsel of food, or the older female elephant chasing a rough young male away from an injured young female. More “scientific” than such anecdotes is the recent finding by the research biologist L. David Mech that the size of wolf packs is determined not by the need to gang up on such large prey as moose but on the number of wolves with whom individual wolves can bond.
As Marc Bekoff wrote recently, “Do these examples show that animals display moral behavior? . . . Yes they do. Animals not only have a sense of justice, but also a sense of empathy, forgiveness, trust, reciprocity, and much more as well.”
It seems extremely unlikely that the sense of fairness Bekoff found in play-fighting could come about without some kind of personal concern for others — that is, a rudimentary empathy.
I have noticed, for instance, that whenever one of my dogs is hurt and yelps, the five others all immediately rush up and sniff its body and face. Are they concerned for the dog and trying to see how seriously it hurts? This seems more likely than a kind of doggish version of “there but for the grace of God go I” or “serves you right, dogface.”
The existence of empathy among dogs suggests in turn the existence in dogs of what is called a “theory of mind.” This refers to the theory (or lack of it) that each individual has about other creatures and their minds. Does that group member have a mind that is enough like mine that it can read my signals and act accordingly? It can be argued that the use of the play-bow suggests that even puppies have a sense that their littermates all think pretty much alike. Developmental psychologists tell us that in humans the theory of mind doesn’t come about until about age 4.
There is a simple test for it, suggested by psychologist Stanley Coren of the University of British Columbia. Throw a ball several times for a dog to retrieve and then turn your back. The dog, realizing that you won’t throw the ball if you can’t see it, will run around you and place the ball in front of you where you can see it. So, Coren concludes, “If we credit 2- and 4-year-old humans with consciousness and reasoning, then in absence of data to the contrary, it seems appropriate that we grant the same to dogs.”
Reasoning? Plenty of scholars say that is just not something dogs do. As for a theory of mind, some say dogs simply observe body language and respond (or not) accordingly. No doubt these matters will continue to be fussed about for some time to come, but it is a big deal that reputable scientists now take note of dogs’ sense of fair play and the likelihood that they are at least minimally empathetic.
So it seems only fair on our part to conclude that dogs are, intrinsically, good — at least in some realms of dog life. Goodness comes naturally to them.
Jake Page is the author of Do Dogs Smile?, a natural history published by Smithsonian Books, and also the owner of a six-dog mob.