Are We Smart Enough to Know How Smart Animals Are

by Frans de Waal

  For the behaviorist, though, such findings are thoroughly puzzling. Two similar birds differing so starkly in what they learn makes no sense, because learning is supposedly universal. Behaviorism ignores ecology and has little room for learning that is adapted to the specific needs of each organism. It has even less room for an absence of learning, as in the kittiwake. Evidence for inborn learning specializations has been mounting, however.2 There are many different types, from the way ducklings imprint on the first moving object they see—whether it is their mother or a bearded zoologist—to the song learning of birds and whales and the way primates copy one another’s tool use. The more variation we discover, the shakier gets the claim that all learning is essentially the same.3

  Yet during my student days, behaviorism still ruled supreme, at least in psychology. Luckily for me, the professor’s pipe-smoking associate, Paul Timmermans, regularly took me aside to induce some much-needed reflection on the indoctrination I was being subjected to. We worked with two young chimpanzees who offered me my first contact with primates apart from my own species. It was love at first sight. I had never met animals that so clearly possessed a mind of their own. Between puffs of smoke, Paul would ask rhetorically, with a twinkle in his eyes, “Do you really think chimps lack emotions?” He would do so just after the apes had thrown a shrieking temper tantrum for not getting their way, or laughed their hoarse chuckles during roughhousing. Paul would also mischievously ask my opinion about other taboo topics, without necessarily saying that the professor was wrong. One night the chimps escaped and ran through the building, only to return to their cage, carefully closing its door behind them before going to sleep. In the morning, we found them curled up in their straw nests and would not have suspected a thing had it not been for the smelly droppings discovered in the hallway by a secretary. “Is it possible that apes think ahead?” Paul asked when I wondered why the apes had closed their own door. How to deal with such crafty, volatile characters without assuming intentions and emotions?

  To drive this point home more bluntly, imagine that you wish to enter a testing room with chimpanzees, as I did every day. I would suggest that rather than rely on some behavioral coding scheme that denies intentionality, you pay close attention to their moods and emotions, reading them the way you would any person’s, and beware of their tricks. Otherwise, you might end up like one of my fellow students. Despite the advice we gave him of how to dress for the occasion, he came to his first encounter in a suit and tie. He was sure he could handle such relatively small animals, while mentioning how good he was with dogs. The two chimps were mere juveniles, only four and five years old at the time. But of course, they were already stronger than any grown man, and ten times more cunning than a dog. I still remember the student staggering out of the testing room, having trouble shedding both apes clinging to his legs. His jacket was in tatters, with both sleeves torn off. He was fortunate that the apes never discovered the choking function of his tie.

  One thing I learned in this lab was that superior intelligence doesn’t imply better test outcomes. We presented both rhesus monkeys and chimpanzees with a simple task, known as haptic (touch) discrimination. They were to stick their hand through a hole to feel the difference between two shapes and pick the correct one. Our goal was to do hundreds of trials per session, but whereas this worked well with the monkeys, the chimps had other ideas. They would do fine on the first dozen trials, showing that the discrimination posed no problem, but then their attention would wander. They’d thrust their hands farther so as to reach me, pulling at my clothes, making laughing faces, banging on the window that separated us, and trying to engage me in play. Jumping up and down, they’d even gesture to the door, as if I didn’t know how to get to their side. Sometimes, unprofessionally, I would give in and have fun with them. Needless to say, the apes’ performance on the task was well below that of the monkeys, not due to an intellectual deficit but because they were bored out of their minds.

  The task was just not up to their intellectual level.

  The Hunger Games

  Are we open-minded enough to assume that other species have a mental life? Are we creative enough to investigate it? Can we tease apart the roles of attention, motivation, and cognition? Those three are involved in everything animals do; hence poor performance can be explained by any one of them. With the above two playful apes, I opted for tedium to explain their underperformance, but how to be sure? It takes human ingenuity indeed to know how smart an animal is.

  It also takes respect. If we test animals under duress, what can we expect? Would anyone test the memory of human children by throwing them into a swimming pool to see if they remember where to get out? Yet the Morris Water Maze is a standard memory test used every day in hundreds of laboratories that make rats frantically swim in a water tank with high walls until they come upon a submerged platform that saves them. In subsequent trials, the rats need to remember the platform’s location. There is also the Columbia Obstruction Method, in which animals have to cross an electrified grid after varying periods of deprivation, so researchers can see if their drive to reach food or a mate (or for mother rats, their pups) exceeds the fear of a painful shock. Stress is, in fact, a major testing tool. Many labs keep their animals at 85 percent of typical body weight to ensure food motivation. We have woefully little data on how hunger affects their cognition, although I do remember a paper entitled “Too Hungry to Learn?” about food-deprived chickens that were not particularly good at noticing the finer distinctions of a maze task.4

  The assumption that an empty stomach improves learning is curious. Think about your own life: absorbing the layout of a city, getting to know new friends, learning to play the piano or do your job. Does food play much of a role? No one has ever proposed permanent food deprivation for university students. Why would it be any different for animals? Harry Harlow, a well-known American primatologist, was an early critic of the hunger reduction model. He argued that intelligent animals learn mostly through curiosity and free exploration, both of which are likely killed by a narrow fixation on food. He poked fun at the Skinner box, seeing it as a splendid instrument to demonstrate the effectiveness of food rewards but not to study complex behavior. Harlow added this sarcastic gem: “I am not for one moment disparaging the value of the rat as a subject for psychological investigation; there is very little wrong with the rat that cannot be overcome by the education of the experimenters.”5

  I was amazed to learn that the nearly century-old Yerkes Primate Center went through an early period in which it tested food deprivation on chimpanzees. In the early years, the center was still located in Orange Park, Florida, before it moved to Atlanta, where it became a major institute for biomedical and behavioral neuroscience research. While still in Florida, in 1955, the center set up an operant conditioning program modeled on procedures with rats, including a drastic reduction in body weight and the replacement of chimp names with numbers. Treating apes as rats proved no success, however. Due to the gigantic tensions this program engendered, it lasted only two years. The director and most of the staff deplored the fasting imposed on their apes and constantly argued with the hard-nosed behaviorists who claimed that this was the only way to give the apes “purpose in life,” as they blithely called it. Expressing no interest in cognition—the existence of which they didn’t even acknowledge—they investigated reinforcement schedules and the punitive effect of time-outs. Rumor had it that staff sabotaged their project by secretly feeding the apes at night. Feeling unwelcome and unappreciated, the behaviorists left because, as Skinner later put it, “tender-hearted colleagues frustrated [their] efforts to reduce chimpanzees to a satisfactory state of deprivation.”6 Nowadays, we would recognize the friction as about not just methodology but also ethics. That creating morose, grumpy apes through starvation was unnecessary was clear from one of the behaviorists’ own attempts with an alternative incentive. Chimpanzee number 141, as he called him, successfully learned a task after each correct choic
e was rewarded with an opportunity to groom the experimenter’s arm.7

  The difference between behaviorism and ethology has always been one of human-controlled versus natural behavior. Behaviorists sought to dictate behavior by placing animals in barren environments in which they could do little else than what the experimenter wanted. If they didn’t, their behavior was classified as “misbehavior.” Raccoons, for example, are almost impossible to train to drop coins into a box, because they prefer to hold on to them and frantically rub them together—a perfectly normal foraging behavior for this species.8 Skinner had no eye for such natural proclivities, however, and preferred a language of control and domination. He spoke of behavioral engineering and manipulation, and not just in relation to animals. Later in life he sought to turn humans into happy, productive, and “maximally effective” citizens.9 While there is no doubt that operant conditioning is a solid and valuable idea and a powerful modifier of behavior, behaviorism’s big mistake was to declare it the only game in town.

  Ethologists, on the other hand, are more interested in spontaneous behavior. The first ones were eighteenth-century Frenchmen, who already used the label ethology, derived from the Greek ethos, “character,” to refer to the study of species-typical characteristics. In 1902 the great American naturalist William Morton Wheeler made the English term popular as the study of “habits and instincts.”10 Ethologists did conduct experiments and were not averse to working with captive animals, but still a world of difference lay between Lorenz calling his jackdaws down from the sky or being followed by a gaggle of waddling goslings and Skinner standing before rows of cages with singly housed pigeons, firmly closing his hand around the wings of one of his birds.

  Ethology developed its own specialized language about instincts, fixed action patterns (a species’ stereotypical behavior, such as the dog’s tail wagging), innate releasers (stimuli that elicit specific behavior, such as the red dot on a gull’s bill that triggers pecking by hungry chicks), displacement activities (seemingly irrelevant actions resulting from conflicting tendencies, such as scratching oneself before a decision), and so on. Without going into the details of its classical framework, ethology’s focus was on behavior that develops naturally in all members of a given species. A central question was what purpose a behavior might serve. Initially, the grand architect of ethology was Lorenz, but after he and Tinbergen met in 1936, the latter became the one to fine-tune the ideas and develop critical tests. Tinbergen was the more analytical and empirical of the two, with an excellent eye for the questions behind observable behavior; he conducted field experiments on digger wasps, sticklebacks, and gulls to pinpoint behavioral functions.11

  The two men formed a complementary relationship and friendship, which was tested by World War II in which they were on opposite sides. Lorenz served as medical officer in the German army and opportunistically sympathized with Nazi doctrine; Tinbergen was imprisoned for two years by the German occupiers of the Netherlands for joining a protest against the way his Jewish colleagues at the university were treated. Remarkably, both scientists patched things up after the war for the sake of their shared love of animal behavior. Lorenz was the charismatic, flamboyant thinker—he didn’t conduct a single statistical analysis in his life—while Tinbergen did the nitty-gritty of actual data collection. I have seen both men speak and can attest to the difference. Tinbergen came across as academic, dry, and thoughtful, whereas Lorenz enthralled his audiences with his enthusiasm and intimate animal knowledge. Desmond Morris, a Tinbergen student famous for writing The Naked Ape and other popular books, got his socks knocked off by Lorenz, saying that the Austrian understood animals better than anyone he’d ever met. He described Lorenz’s 1951 lecture at Bristol University as follows:

  To describe his performance as a tour de force is an understatement. Looking like a cross between God and Stalin, his presence was overpowering. “Contrary to your Shakespeare,” he boomed, “there is madness in my method.” And indeed there was. Almost all his discoveries were made by accident and his life consisted largely of a series of disasters with the menageries of animals with which he surrounded himself. His understanding of animal communication and display patterns was revelatory. When he spoke about fish, his hands became fins, when he talked about wolves his eyes were those of a predator, and when he told tales about his geese his arms became wings tucked into his sides. He wasn’t anthropomorphic, he was the opposite—theriomorphic—he became the animal he was describing.12

  A journalist once recounted how she had been sent into Lorenz’s office by a receptionist with the words that he was expecting her. His office turned out to be empty. When she asked around, people assured her that he had never left. After a while, she discovered the Nobelist partially submerged in an enormous aquarium built into the office wall. This is how we like our ethologists: as close to their animals as possible. It reminds me of my own encounter with Gerard Baerends, the silverback of Dutch ethology and the very first student of Tinbergen. After my stint in the behaviorist lab, I sought to enter Baerends’s ethology course at the University of Groningen to work with the jackdaw colony that flew around the institution’s nest boxes. Everyone warned me that Baerends was very strict and did not let just anybody in. When I walked into his office, my eyes were immediately drawn to a large well-kept tank with convict cichlids. Being an avid aquarist myself, I hardly took the time to introduce myself before we launched into a discussion of how these fish raise and guard their fry, which they do extraordinarily well. Baerends must have taken my passion as a good sign, because I was admitted without a problem.

  The great novelty of ethology was to bring the perspective of morphology and anatomy to bear on behavior. This was a natural step, because whereas behaviorists were mostly psychologists, ethologists were mostly zoologists. They discovered that behavior is not nearly as fluid and hard to define as it might seem. It has a structure, which can be quite stereotypical, such as the way young birds flutter their wings while begging for food with gaping mouths, or how some fish keep fertilized eggs in their mouth until they hatch. Species-typical behavior is as recognizable and measurable as any physical trait. Given their invariant structure and meaning, human facial expressions are another good example. The reason we now have software that reliably recognizes human expressions is that all members of our species contract the same facial muscles under similar emotional circumstances.

  Konrad Lorenz and other ethologists wanted to know how animals behave of their own accord and how it suits their ecology. In order to understand the parent-offspring bond in waterfowl, Lorenz let goslings imprint on himself. They followed the pipe-smoking zoologist around wherever he went.

  Insofar as behavior patterns are innate, Lorenz argued, they must be subject to the same rules of natural selection as physical traits and be traceable from species to species across the phylogenetic tree. This is as true for the mouth brooding of certain fish as it is for primate facial expressions. Given that the facial musculature of humans and chimpanzees is nearly identical, the laughing, grinning, and pouting of both species likely goes back to a common ancestor.13 Recognition of this parallel between anatomy and behavior was a great leap forward, which is nowadays taken for granted. We all now believe in behavioral evolution, which makes us Lorenzians. Tinbergen’s role was, as he put it himself, to act as the “conscience” of the new discipline by pushing for more precise formulations of its theories and developing ways to test them. He was overly modest saying so, though, because in the end it was he who best spelled out the ethological agenda and turned the field into a respectable science.

  Keeping It Simple

  Despite the differences between ethology and behaviorism, the two schools had one thing in common. Both were reactions against the overinterpretation of animal intelligence. They were skeptical of “folk” explanations and dismissed anecdotal reports. Behaviorism was the more vehement in its rejection, saying that behavior is all we have to go by and that internal processes can be safely ign
ored. There is even a joke about its complete reliance on external cues, in which one behaviorist asks another after lovemaking: “That was great for you. How was it for me?”

  In the nineteenth century, it was perfectly acceptable to talk about the mental and emotional lives of animals. Charles Darwin himself had written a whole tome about the parallels between human and animal emotional expressions. But while Darwin was a careful scientist who double-checked his sources and conducted observations of his own, others went overboard, almost as in a contest of who could come up with the wildest claim. When Darwin chose the Canadian-born George Romanes as his protégé and successor, the stage was set for an avalanche of misinformation. About half the animal stories collected by Romanes sound plausible enough, but others are embellished or plainly unlikely. They range from a story about rats forming a supply line to their hole in the wall, carefully handing down stolen eggs with their forepaws, to one about a monkey hit by a hunter’s bullet who smeared his hand with his own blood and held it out to the hunter to make him feel guilty.14

  Romanes knew the mental operations required for such behavior, he said, by extrapolating from his own. The weakness of his introspective approach was, of course, its reliance on one-time events and on trust in one’s own private experiences. I have nothing against anecdotes, especially if they have been caught on camera or come from reputable observers who know their animals; but I do view them as a starting point of research, never an end point. For those who disparage anecdotes altogether, it is good to keep in mind that almost all interesting work on animal behavior has begun with a description of a striking or puzzling event. Anecdotes hint at what is possible and challenge our thinking.