Merle's Door

by Ted Kerasote

  Unlike Gray Cat, who ate the choicest parts of his mice and left me the tails, Merle was leaving me what he valued most. And he was sticking around to watch my reaction. I never saw Gray Cat do that.

  I therefore always made a great fuss over the bones Merle left me, for I was touched by this thoughtfulness. Here was my dog giving me more than affection, loyalty, and devotion by the look in his eyes, his body pressed to mine, and his tracking me down when he found that I was gone. It appeared that he had moved into the realm where we operate when we offer symbols of our affection.

  No matter how many times I tried to explain his behavior by employing Morgan's Canon, I couldn't. If he were simply seeing me as his pup and was replaying the hardwired behavior of an adult wolf who brings food back to the den for its young, why not bring me one of the bones I gave him, or one of the elk or deer bones he found at other people's homes after they butchered animals, or, easiest of all, why not regurgitate some of his kibble? Why beef bones? It seemed that the complex explanation was really the simplest: He was giving me something I didn't have; he was giving me a gift.

  I responded in kind. I licked and gnawed at his bones a bit before trying to hand them back to him. His reaction was always the same—he wouldn't take them. In fact, he would draw back his head in surprise, his look saying, "No, no, that's for you." This from a dog who adored bones and would chew them for hours, manipulating them in his paws and sucking their ends with his eyes closed in rapture.

  I'd carry the bone into the bathroom and, after washing up, would take it downstairs and place it directly in front of his dog door. He'd watch me with a bit of chagrin—"You don't know what you're missing"—and then he'd pick it up, and slap-slap, head outside, where he'd lie on the porch and chew it into something that resembled a seashell that has been tossed by the waves until it is smooth and round and polished.

  Undoubtedly, Morgan and most of the researchers who followed his lead into the twentieth century would have found these two anecdotes—claiming to show reason and emotion in dogs—thin gruel. These men increasingly relied on controlled experiments done in the laboratory and, pulling their chins, would have said, "Interesting examples of reasoning and emotion in animals, Mr. Kerasote. Now show us some proof."

  In 1932, that's exactly what Edward Tolman did. A psychologist at the University of California, he used an experiment designed by one of his graduate students, H. C. Blodgett, to argue that animals retained memories of their territory and could make informed decisions based on this information. Hungry rats were allowed to run through a maze for six days, but without any food reward. On the seventh day they found food at the end of the maze and, within two trials, were completing the maze as quickly, and with as few errors, as the control rats who had gotten food every day.

  Tolman concluded that the first group of rats had learned the twists and turns of the maze while wandering around freely. He called this process "latent learning" and coined the term "cognitive map" for the mental geography that rats, dogs, and people assimilate in their travels and then retain so as to help them get economically from place to place. It was this very sort of mental map, Tolman would say, that Merle used when he cut me off at the pass.

  I also believe that because Merle wandered extensively and had the opportunity to build ever larger cognitive maps—in all senses of the term—he was able to make a leap of the canine mind and bring me beef bones as tokens of his affection. A bold statement, it's true, but I'm willing to give him the benefit of the doubt, as I imagine Tolman's contemporary, Robert Yerkes, would, if he were alive today.

  A Harvard psychologist who co-founded the Journal of Animal Behavior in the early 1900s, Yerkes stood out from his colleagues in that he understood an animal's behavior to be a window into its subjective experience: If you bring someone beef bones over a period of eight years to the exclusion of other bones that are readily available, it's more than chance; it's likely that you're saying something by this action. In a time when most psychologists believed that animals led impoverished mental lives, Yerkes sided with Darwin when he wrote, "There is no question, in the mind of the person who really knows animals, that the higher vertebrates possess a great variety of sense qualities and feelings....Of emotions, sentiments, associations, memory images, ideas, and even certain forms of judgement there are noteworthy evidences, and the more liberal among psychologists are at present inclined to believe that at least some animals, among them the dog and horse, the raccoon and cat, experience conscious complexes which are much like ours."

  Such liberal views, however, remained in the minority during the early part of the twentieth century, and not without reason. In the late 1800s, the Russian physiologist Ivan Pavlov had graphically demonstrated that if you showed a dog food and rang a bell while it salivated, it would soon salivate when the bell was sounded, even in the absence of food. Following Pavlov's lead, many animal researchers designed stimulus-response experiments, forgetting that even though a conditioned response could be readily produced in their subjects, these animals were in fact complex beings who might act differently in a natural environment where stimuli are never experienced in a vacuum, but against a background of other interacting stimuli. The result of relying mainly on laboratory observation was that animals came to be seen as simplistic creatures, influenced predominantly by positive and negative reinforcements like food and electrical shocks.

  It was the American psychologist John Watson who subsequently applied this view of animals to the social arena. If controllers could manipulate enough stimuli, his reasoning went, they could control the entirety of their subjects' lives. This notion would lead to profoundly tragic consequences for millions of people in the Soviet Union and China, but Watson was unable to foresee its political implications when he wrote, "Give me a dozen healthy infants, well-formed, and my own specified world to bring them up in and I'll guarantee to take any one at random and train him to become any type of specialist I might select—doctor, lawyer, artist, merchant chief and, yes, even beggar-man and thief, regardless of his talents, penchants, tendencies, abilities, vocations, and race of his ancestors."

  In the 1930s, the Harvard psychologist B. F. Skinner then took equal portions of Pavlov and Watson and combined them with his own laboratory experiments—trials of positive reinforcement with which he taught pigeons to play toy pianos and discriminate between playing cards—to produce the science of behaviorism. As Skinner enthusiastically reported, his techniques made "it possible to shape an animal's behavior as a sculptor shapes a lump of clay."

  Skinner was both a meticulous scientist and a great popularizer of his lab work. Not stopping with rats and pigeons, he gave the public an easy way to train their dogs when he invented clicker training, using an inexpensive device that makes a cricket-like sound and substitutes for the human voice saying "good dog" or "yes." In this way, the clicker gives the dog a precise, unvarying cue that it has executed the correct behavior and that a reward or reinforcement will follow. And the reinforcement worked both ways. People enjoyed clicker training because it was easy and got results.

  Skinner became controversial, however, when he made the case that animals perform their conditioned responses in a "purely mechanical" fashion, without really knowing what they're doing. He then extended this notion to human beings. The world of introspection, he claimed, was "vastly overrated" and wasn't an essential part of the fields of behavior or medicine. When Beyond Freedom and Dignity was published, its very title made the hairs of libertarians stand on end. Their suspicions of Skinner's totalitarian tendencies only increased when he approvingly quoted one of Jean-Jacques Rousseau's most chilling passages from Émile:

  Let [the child] believe that he is always in control, though it is always you [the teacher] who really controls. There is no subjugation so perfect as that which keeps the appearance of freedom, for in that way one captures volition itself. The poor baby, knowing nothing, able to do nothing, having learned nothing, is he not at your mercy? Can you not arr
ange everything in the world which surrounds him? Can you not influence him as you wish? His work, his play, his pleasures, his pains, are not all these in your hands and without his knowing? Doubtless he ought to do only what he wants; but he ought to want to do only what you want him to do; he ought not to take a step which you have not foreseen; he ought not to open his mouth without your knowing what he will say.

  Substitute the word "dog" for "baby" and it's not hard to see why dog trainers eagerly adopted Skinner's methods. After all, who doesn't want an obedient dog? And unlike babies, do we really need to worry about a domestic animal being subjugated? The word "domestic" implies that condition.

  Today, the language of Rousseau and Skinner can be found in a variety of dog-training manuals. As Kevin Behan writes in his book Natural Dog Training, "To Master a dog, we must be decisive and control everything that the dog learns so the dog will have no opportunity but to learn what we want him to." Not only should everything be managed, Behan continues, but the dog should also have "absolutely no unsupervised freedom."

  Jean Donaldson, in The Culture Clash: A Revolutionary New Way of Understanding the Relationship Between Humans and Domestic Dogs, is just as clear about the role humans should play in this ancient relationship: "You have control of your dog's access to everything he wants in life: food, the outside world, attention, other dogs, smells on the ground, play opportunities." She advises micro-managing every detail of a dog's life down to training it to eliminate on command. Why do we need to oversee dogs in this fashion? Because, she answers, they have "little, smoothish lemon brains," not at all like the "convoluted, melon brains" that allow humans to think.

  George Orwell warned of the perils of this sort of control in his novel 1984 (published in 1949), but few dog trainers have worried about becoming Big Brother to their dogs, especially when Skinner's methods—clicking for the correctly executed behavior followed by a treat—have produced hundreds of thousands of spit-and-polish dogs: field dogs, service dogs, and the family pet. Yet the nagging question remains: Does controlling a dog's life through micro-managing its behavior short-circuit its ability to think on its own?

  ***

  During the 1930s, Professor E. G. Sarris of the University of Hamburg investigated this issue by doing a series of innovative experiments with a dog named Argos. A male of Pinscher, Spitz, and Dachshund descent, Argos was twenty-three months old when Sarris began his experiments, suspending a piece of meat by a string high enough above the dog so that he couldn't reach it, even with a great leap. Close by were wagons with little and big boxes on them. Once Sarris showed Argos that the wagons could be towed by loops, the dog quickly realized that the wagon with the bigger box needed to be towed under the meat in order for him to stand on it and grasp the meat. Sarris made the problems harder and harder, putting obstacles in front of the wagon, which Argos needed to remove one by one so that he could roll the wagon under the meat. In addition, Sarris hid the wagon a dozen feet from the meat. Argos would run around in widening circles, searching for it, and, when he found it, pull it into position so he could stand on it and retrieve his prize.

  According to the animal behaviorist Marian Stamp Dawkins, whose book Through Our Eyes Only? is one of the more compelling explanations of animal consciousness, Argos's actions demonstrated the two attributes of thinking. "The first," she writes, "is that the thinker should have some sort of internal representation of the world in his, her, or its head. This means that it does not just respond to the stimuli immediately surrounding it but carries a memory of things that were there in the past but are now gone or are out of sight. The second is that something is done to that representation to enable the true thinker to work out what would happen under new circumstances—for example, if everything were turned upside down or one element were changed."

  Sarris did further experiments, trying to ascertain the extent of a dog's internal representation of the world. After chaining Argos and some of his kennel mates in place, he showed them a piece of meat attached to a string. Then he tossed the meat out of reach. The loose end of the string was left within their grasp, however. One dog, Phryne, tried to reach the meat by lunging, soon gave up, and yawned with "an expression of helplessness." Another, Ares, became frustrated, barked wildly, and bit at his chain. Both Niki and Argos initially lunged at the meat a few times before becoming quiet and studying the string. Within forty-eight seconds, each used the string to tow the meat toward them. When the tests were repeated, Argos and Niki reeled in the meat within four seconds of seeing it.

  Sarris then showed them the meat at the end of the string before tossing the meat over a wall. Both Niki and Argos pulled it in. Sarris proceeded to lengthen the string to five, ten, fifteen, twenty, and forty meters so as to see how far—both in distance and time—the dog's internal representation of "meat at the end of the string" extended. Niki was able to retrieve the meat when it was fifteen meters out, but Argos truly shone. Even when the meat was forty meters away, he pulled it in.

  Sarris's conclusion from these experiments was the opposite of the one reached by his contemporaries, Watson and Skinner. There are real differences in "temperament and intelligence," he wrote, that predispose individuals to certain activities. In other words, despite positive reinforcement, you can't train any child to become an artist or any pigeon to play the piano—at least not well—as the American behaviorists claimed. Dog training expert Steven R. Lindsay emphasizes this concept when he advises dog owners to ask themselves whether the dog under their care, "given a choice, would likely select the career being chosen for it."

  Sarris's experiments also showed him that "a widening of the dog's 'Umwelt' was possible." The German word Umwelt means "environment," but Sarris used it in the sense of "the surrounding world of the dog, which it is capable of understanding." Key to such an expanded understanding of the world was giving dogs the time and space to work through problems on their own. What they needed was less control, not more.

  ***

  One autumn day I saw firsthand where allowing a dog to make its own decisions on a regular basis might lead. Merle and I were traveling in the high mountains east of Yellowstone National Park, trying out two llamas as pack animals instead of horses. One of the llamas, its coat brown and white, was doing a workmanlike job of carrying our camp and food onto the Continental Divide. The other llama constantly wanted to sit down. And there are few creatures—maybe a camel—who can be more stubborn than a llama when it puts its mind to something. This llama, colored black and white, could not be cajoled; he could not be pulled; he could not be prodded.

  There we were, halted several miles from camp, on a high barren ridge in one of the more remote spots in the Lower Forty-eight, the sky becoming roiled with storm clouds, and the second llama refused to move. More. He lowered himself to the ground and folded his feet calmly beneath him as if he might wait a decade before the whim to move crossed his mind again.

  He was dallied to the first llama's packsaddle, so I figured if I got the first llama moving, the second might follow. I clucked the first llama into a walk; the second sat firm. The result was that the first llama's packsaddle was nearly torn off. Of course, this was nothing new—to a lesser degree, it had been happening all day. But whereas the first llama had previously become annoyed, stomping his feet, he now became infuriated. He turned and spit into the second llama's face. Placidly, the second llama gazed into the distance as a mix of phlegm and masticated vegetation dripped down his cheeks.

  In the meantime, Merle had been watching the unfolding of this little drama from a sitting position a dozen feet off to the side, his head twisting intently from the first llama to the second and then to me, pulling, pushing, and exhorting the stubborn beast in no uncertain terms to get to his feet. The next thing I knew, Merle was rushing at the second llama, woofing at him while lunging menacingly at his forequarters, as he had done to the bison and the moose. All to no effect. The llama stared beyond Merle into some ethereal distance where men and d
ogs and good pack llamas were not troubling him.

  Merle tried the woof-and-lunge maneuver once more—nothing. Turning, he seemed to give up and walk away, putting on an air of nonchalance as he ambled behind the llama. Then, without warning, he darted at the llama's rear and sank his teeth into his butt. No nip—this was an honest-to-goodness bite. The llama leapt to his feet and lunged forward, running into the first llama, who began to trot up the trail.

  In this way—Merle following the second llama and worrying his heels each time he slowed—we got to camp. Arriving at tree line in the next valley, I picketed the llamas and looked at Merle.

  "Thanks," I said gratefully. "I didn't think you had any Heeler in you."

  He panted happily, wagging his tail in the half-mast position that I had come to learn meant "I think it's time to eat." And it made me wonder if, back there on the trail, Merle had worked out that getting to camp meant eating and had prudentially decided to take matters into his own hands or, more accurately, his own teeth. Perhaps he had read my frustration with the llama and interpreted it as a call for assistance.

  Whatever his motives, his actions impressed me. He had seen a problem that neither I nor the other llama could solve, and—just like Argos using the right-sized box—had thought outside of the box, coming up with a solution that worked. His ability to solve our dilemma in an elegant way—and without any reinforcement from me—was unlike any behavior I had witnessed in a dog, a creature who normally is just along for the ride and, like a child, isn't expected to use its brain.

  So who is right? Are dogs mere trial-and-error learners, stimulus-response machines, who respond to clickers and dog biscuits? Are their brains accurately portrayed as "lemons," with all the negative connotations the word implies? Or might their brains be very much like ours, quite capable of reasoning and emotion, yet with a crucial difference?