Are We Smart Enough to Know How Smart Animals Are

by Frans de Waal

  Our experiment was inspired by a floating peanut task conducted on a large number of orangutans and chimpanzees, a subset of which cracked the puzzle at first sight.41 This is especially remarkable, since—unlike the crows—the apes had no pretraining; nor did they find any tools nearby. Rather, they must have conjured the effectiveness of water in their heads before going out of their way to collect it. Water doesn’t even look like a tool. How hard this task is became clear from tests on children, many of which never found the solution. Only 58 percent of eight-year-olds came up with it, and only 8 percent of four-year-olds. Most children frantically try to reach the prize with their fingers, then give up.42

  These studies have set up a friendly rivalry between primate chauvinists and corvid aficionados. I sometimes teasingly accuse the latter of “ape envy,” because in every publication they draw a contrast with the primates, saying the corvids are either doing better or at least equally well. Calling their birds “feathered apes,” they make outrageous claims such as “The only credible evidence of technological evolution in nonhumans to date comes from New Caledonian crows.”43 Primatologists, on the other hand, wonder how generalizable corvid tool skills are, and if “feathered monkeys” isn’t a better moniker for the birds. Are crows one-trick ponies, like the clam-smashing otters or the Egyptian vultures that throw rocks at ostrich eggs? Or do they have the intelligence to take on a broad array of problems?44 This issue is far from settled, because even though ape intelligence has been studied for over a century, corvid tool studies have come up only in the last decade.

  An intriguing new entry is the use of metatools by New Caledonian crows.

  A crow is presented with a piece of meat that it can retrieve only by using a long stick, but this stick is behind bars wide enough for the crow’s beak but not its head. The crow is unable to reach the tool. In a nearby box, however, lies a short stick suitable for retrieving the long one. To solve this problem, the right order is to pick up the short stick, use it to fetch the long one, and then apply the latter stick for the meat. The crow needs to understand that tools can be used on nonfood objects and to take steps in the right order. Alex Taylor and coworkers used wild New Caledonian crows on Maré Island, placed temporarily in an aviary. They tested seven crows, all of which managed metatool use; three followed the right sequence on the first attempt.45 Presently, Taylor is trying out tasks with even more steps, and the crows are keeping up with the challenge. This is most impressive, and considerably better than monkeys, which have trouble with stepwise tasks.

  Given the evolutionary gulf between primates and corvids, and the many ancestral species of mammals and birds in between that don’t use tools, we are dealing with a typical example of convergent evolution. Independently, both taxonomic groups must have faced a need for complex manipulations of items in their environment, or other challenges that stimulated brain growth, which led them to evolve strikingly similar cognitive skills.46 The arrival of corvids on the scene illustrates how discoveries of mental life ripple across the animal kingdom, a process best summarized by what I’ll call my cognitive ripple rule: Every cognitive capacity that we discover is going to be older and more widespread than initially thought. This is rapidly becoming one of the best-supported tenets of evolutionary cognition.

  As a case in point, we now have evidence of tool use outside mammals and birds. Primates and corvids may well show the most sophisticated use of technology, but what to think of partially submerged crocodiles and alligators balancing large sticks on their snouts? Crocodilians do so especially in pools and swamps near rookeries during the nesting season, when herons and other wading birds are in desperate need of sticks and twigs. You can imagine the scene: a heron lands on a log in the water from which it wants to pick up an attractive branch, but suddenly the log comes to life and grabs the bird. Perhaps crocs initially learn that birds land on them when branches float nearby and then extend this association by making sure to be near branches when herons are nesting. From there, it may be a small step to cover oneself with objects that attract birds. The problem with this idea, however, is that there are actually very few free-floating branches and twigs around. There is too much demand for them. Is it possible that the crocs—which the scientists lament are historically taken to be “lethargic, stupid, and boring”—bring their stick-lures with them from far away? This would be another spectacular cognitive ripple, one that extends deliberate tool use to the reptiles.47

  The final example, which may again stretch the definition of a tool, concerns the veined octopus in the seas around Indonesia. Here we are dealing with an invertebrate: a mollusk! It has been seen collecting coconut shells. Since octopuses are a favorite food of many predators, camouflage is one of their main goals in life. Initially, the coconut shells yield no benefit, however, because they have to be transported, which only draws unwanted attention. Stretching its arms into rigid limbs, the octopus tiptoes over the sea floor while holding its prize in some of its other arms. Awkwardly walking to a safe lair, it can then use the shells to hide underneath.48 A mollusk collecting tools for future protection, however simple, goes to show how far we have come since the days when technology was thought to be the defining characteristic of our species.

  4 TALK TO ME

  “Speak and I shall baptize thee!”

  —French Bishop to a chimpanzee, early 1700s1

  We associate research in the natural habitat with sacrifice and bravery, since fieldworkers must tackle the unpleasant and dangerous creatures of the tropical rainforest, from bloodsucking leeches to predators and snakes. By contrast, students of captive animals are thought to have it easy. But we sometimes forget how much courage it takes to defend one’s ideas in the face of staunch opposition. Most of the time this occurs just among academics, which is disagreeable rather than hazardous, but Nadia Kohts faced lethal risks. Her full name was Nadezhda Nikolaevna Ladygina-Kohts, and she lived and worked early last century in the shadow of the Kremlin. Under the sinister influence of the would-be geneticist Trofim Lysenko, Joseph Stalin had many a brilliant Russian biologist either shot or sent to the Gulag for thinking the wrong thoughts. Lysenko believed that plants and animals pass on traits gained during their lifetime. The names of those who disagreed with him became unmentionable, and entire research institutes were closed down.

  It was in this oppressive climate that Kohts, with her husband Alexander Fiodorovich Kohts—founding director of Moscow’s State Darwin Museum—set out to study ape facial expressions, inspired by The Expression of the Emotions in Man and Animals by that bourgeois Englishman, Charles Darwin. Lysenko was distinctly ambivalent about Darwin’s theory, some of which he labeled “reactionary.” Staying out of trouble became a major preoccupation of the Kohtses, who hid documents and data among their taxidermy collection in the museum basement. They wisely put a large statue of the French biologist Jean-Baptiste Lamarck—famous proponent of the inheritance of acquired characteristics—at the museum entrance.

  Kohts published in French, German, and most of all her native Russian. She wrote seven books, of which only one was translated into English, long after its appearance in 1935. The English version of Infant Chimpanzee and Human Child, edited by me, appeared in 2002. The book compares the emotional life and intelligence of a young chimpanzee, Joni, with that of Kohts’s little son, Roody. Kohts studied Joni’s reactions to pictures of chimpanzees and other animals, and to his own mirror image. Even though Joni was probably too young to recognize himself, Kohts describes how he would entertain himself in front of his reflection by pulling weird faces and sticking out his tongue.2

  Kohts is little known compared to Wolfgang Köhler, who conducted his groundbreaking ape research from 1912 through 1920. I wonder what she knew about it while working in Moscow from 1913 until Joni’s premature death in 1916. While Köhler is widely recognized as a pioneer of evolutionary cognition, pictures of Kohts’s work leave little doubt that she was on exactly the same track. One of the museum’s glass cases f
eatures Joni’s mounted body surrounded by ladders and tools, including sticks that fit into each other. Was Kohts overlooked by science due to her gender? Or was it her language?

  I learned about her from the writings of Robert Yerkes, who came to Moscow to discuss her projects through an interpreter. In his books, Yerkes described Kohts’s work with the greatest admiration. There is a good chance, for example, that Kohts invented the matching-to-sample (MTS) paradigm, a staple of modern cognitive neuroscience. MTS is nowadays being applied to both humans and animals in countless laboratories. Kohts would hold up an object for Joni, then hide it among other objects in a sack and let him feel around to find the first one. The test involved two modalities—vision and touch—demanding that Joni make a choice based on his memory of the previously seen model.

  Nadia Ladygina-Kohts was a pioneer in animal cognition, who studied not only primates but also parrots, such as this macaw. Working in Moscow at around the same time that Köhler conducted his research, she remains far less known.

  My own fascination with this unsung hero’s work took me to Moscow, too. I received a behind-the-scenes tour of the museum, where I leafed through private picture albums. Kohts was (and is) much beloved in her country, where she is widely recognized as the great scientist that she was. My biggest surprise was to learn that she owned at least three large parrots. Pictures show her accepting an object handed to her by a cockatoo, and her holding out a tray with three cups toward a macaw. The parrots would sit opposite her, on a table, while Kohts held a small food reward in one hand and a pencil in the other, scoring their choices as she tested their ability to discriminate among objects. I checked with our contemporary expert on Psittaciformes, the American psychologist Irene Pepperberg, but she had never heard of Kohts’s parrot studies. I doubt that anyone in the West ever suspected that bird cognition, too, was studied in Russia well before it became more widely known.

  Alex the Parrot

  I first met Alex, the African gray that Irene raised and studied for three decades, on visits to her department from a nearby university. Irene had bought the bird in a pet store, in 1977, and was setting up an ambitious project that would open the public eye to the avian mind. It ended up paving the way for all subsequent studies of bird intelligence, because until then the general opinion had been that bird brains simply don’t support advanced cognition. Due to their lack of much of anything that looks like a mammalian cortex, birds were viewed as well endowed with instincts yet poor at learning, let alone thinking. Despite the fact that their brains can be quite sizable—the African gray’s is the size of a shelled walnut, with a large area that functions like a cerebral cortex—and that their natural behavior offers ample reason to question the low opinion of them, the different brain organization of birds has been held against them.

  Having myself kept and studied jackdaws—members of that other large-brained bird family, the corvids—I have never had any doubt about their behavioral flexibility. On walks through the park, my birds would tease dogs by flying right in front of their heads, just out of reach of their snapping mouths, to the surprise and chagrin of the dog owners. Indoors, they would play object hiding with me: I would hide a small item, such as a cork, under a pillow or behind a flower pot, while they would try to find it, or vice versa. This game relied on the well-known food-caching talents of crows and jays but also suggested object permanence: the understanding that an object continues to exist even after it has disappeared from view. The extreme playfulness of my jackdaws hinted, as it does with animals in general, at high intelligence and the thrill of a challenge. Visiting Irene, I was quite prepared to be impressed by a bird, therefore, and Alex did not disappoint. Cockily sitting on his perch, he had begun to learn labels for items such as keys, triangles, and squares, saying “key,” “three-corner,” or “four-corner” whenever these objects were pointed out.

  At first sight, this came across as language learning, but I am not sure this is the right interpretation. Irene didn’t claim that Alex’s talking amounted to speaking in the linguistic sense. But of course, the labeling of objects is very much part of language, and we should not forget that once upon a time linguists defined language simply as symbolic communication. Only when apes proved capable of such communication did they feel the need to raise the bar and add refinements such as that language requires syntax and recursivity. Language acquisition by animals became a huge topic that drew enormous public interest. It was as if all questions about animal intelligence boiled down to a sort of Turing test: can we, humans, hold a sensible conversation with them? Language is such a marker of humanity that an eighteenth-century French bishop was ready to baptize an ape provided he could speak. It surely was all that science seemed to care about in the 1960s and 1970s, resulting in attempts to talk with dolphins and teach language to a multitude of primates. Some of this attention turned sour, however, when the American psychologist Herbert Terrace, in 1979, published a highly skeptical article about the sign-language capacities of Nim Chimpsky, a chimpanzee named after American linguist Noam Chomsky.3

  Terrace found Nim a boring conversationalist. The vast majority of his utterances were requests for desirable outcomes, such as food, rather than expressions of thoughts, opinions, or ideas. Terrace’s surprise at this was by itself rather surprising, however, given his reliance on operant conditioning. Since this is not how we teach children language, one wonders why it was used for an ape. Having been rewarded thousands of times for hand signals, why wouldn’t Nim use these signals to obtain rewards? He simply did what he was taught. As a result of this project, however, the voices pro and contra animal language were getting louder by the day. To find a bird voice among this cacophony threw many people off, because while apes obviously don’t talk, Alex carefully pronounced every word. Superficially, his behavior resembled language more than that of any other animal, even if there was little agreement about what it actually meant.

  Irene’s choice of species was intriguing since Doctor Dolittle, the central character of a series of children’s books, owned an African gray, named Polynesia, who taught the good doctor the language of animals. Irene had always been attracted to these stories and as a child already presented her pet budgie with a drawer full of buttons to see how the bird would arrange them.4 Her work with Alex grew straight out of her early captivation with birds and with their taste in colors and shapes. But before discussing her research further, let me briefly dwell on the desire to talk with animals—a desire often expressed by scientists working on animal cognition—as it relates to the deeper connection often assumed between cognition and language.

  Oddly enough, this particular desire must have passed me by, because I have never felt it. I am not waiting to hear what my animals have to say about themselves, taking the rather Wittgensteinian position that their message might not be all that enlightening. Even with respect to my fellow humans, I am dubious that language tells us what is going on in their heads. I am surrounded by colleagues who study members of our species by presenting them with questionnaires. They trust the answers they receive and have ways, they assure me, of checking their veracity. But who says that what people tell us about themselves reveals actual emotions and motivations?

  This may be true for simple attitudes free from moralizations (“What is your favorite music?”), but it seems almost pointless to ask people about their love life, eating habits, or treatment of others (“Are you pleasant to work with?”). It is far too easy to invent post hoc reasons for one’s behavior, to be silent about one’s sexual habits, to downplay excessive eating or drinking, or to present oneself as more admirable than one really is. No one is going to admit to murderous thoughts, stinginess, or being a jerk. People lie all the time, so why would they stop in front of a psychologist who writes down everything they say? In one study, female college students reported more sex partners when they were hooked up to a fake lie-detector machine than without it, thus demonstrating that they had been lying before.5 I am in fact relie
ved to work with subjects that don’t talk. I don’t need to worry about the truth of their utterances. Instead of asking them how often they engage in sex, I just count the occasions. I am perfectly happy being an animal watcher.

  Now that I think of it, my distrust of language goes even deeper, because I am also unconvinced of its role in the thinking process. I am not sure that I think in words, and I never seem to hear any inner voices. This caused a bit of an embarrassment once at a meeting about the evolution of conscience, when fellow scholars kept referring to an inner voice that tells us what is right and wrong. I am sorry, I said, but I never hear such voices. Am I a man without a conscience, or do I—as the American animal expert Temple Grandin once famously said about herself—think in pictures? Moreover, which language are we talking about? Speaking two languages at home and a third one at work, my thinking must be awfully muddled. Yet I have never noticed any effect, despite the widespread assumption that language is at the root of human thought. In his 1973 presidential address to the American Philosophical Association, tellingly entitled “Thoughtless Brutes,” the American philosopher Norman Malcolm stated that “the relationship between language and thought must be so close that it is really senseless to conjecture that people may not have thoughts, and also senseless to conjecture that animals may have thoughts.”6

  Since we routinely express ideas and feelings in language, we may be forgiven for assigning a role to it, but isn’t it remarkable how often we struggle to find our words? It’s not that we don’t know what we thought or felt, but we just can’t put our verbal finger on it. This would of course be wholly unnecessary if thoughts and feelings were linguistic products to begin with. In that case, we’d expect a waterfall of words! It is now widely accepted that, even though language assists human thinking by providing categories and concepts, it is not the stuff of thought. We don’t actually need language in order to think. The Swiss pioneer of cognitive development, Jean Piaget, most certainly was not ready to deny thought to preverbal children, which is why he declared cognition to be independent of language. With animals, the situation is similar. As the chief architect of the modern concept of mind, the American philosopher Jerry Fodor, put it: “The obvious (and I should have thought sufficient) refutation of the claim that natural languages are the medium of thought is that there are non-verbal organisms that think.”7