American Scientist

My title is borrowed from a book by Eugène Marais (1871–1936), published posthumously in 1969. Marais was one of the first to make a close study of the behavior of nonhuman primates in the wild. His analyses were far ahead of their time, because he attempted to understand the actions of these creatures in terms of the evolutionary continuity with Homo sapiens that Charles Darwin had earlier proposed. Although genetic analyses were still a long way off, it was obvious to Darwin and his converts in the 19th century that the great apes—chimpanzees, gorillas, orangutans and bonobos, or pygmy chimps —must be our distant cousins. "He who understands baboon," Darwin noted, "would do more toward metaphysics than Locke."

Although today we recognize baboons as monkeys, not apes, and therefore less closely related to us, Darwin's point still stands. And Marais was on the right track when he observed free-living and captive baboons in his native South Africa and puzzled over the similarities—and differences—he could see between their behavior and that of human children and adults.

Most of the questions with which Marais grappled remain unanswered. But in recent years attempts to understand primate minds have spawned a controversial new ethic. It is based on the notion that the differences in psychology between people and other great apes are too small to justify different treatment. That is, nonhuman apes should get the same ethical consideration that their human brethren receive. This view is most manifest in the Great Ape Project, which describes itself as working to raise the legal and moral status of these animals. The ultimate aim of this group is to have the United Nations adopt a declaration on the rights of nonhuman apes, one that would make all medical research on them impossible.

Supporters of the Great Ape Project are impressed by the genetic similarity between people and nonhuman apes and by the relatively short period of time since we and our closest relatives (chimpanzees) diverged from a common ancestor. Most advocates of this project point to what they consider the key psychological similarity between nonhuman apes and ourselves: Nonhuman apes, they argue, are self-aware. As a consequence of this self-awareness, gorillas, chimpanzees, bonobos and orangutans must suffer in captivity in ways not so different from what we would experience under similar circumstances.

This is not an abstract scholarly debate. There are about 1,600 chimpanzees held for biomedical research in the U.S., and these animals are essential to the study of several maladies, ones for which few if any other approaches are available. Probably the single most important example is liver disease. It was research on chimpanzees that provided the vaccine against hepatitis B. Carriers of hepatitis B are about 200 times more likely to develop liver cancer than the general population, so the hepatitis B vaccine can be considered the first cancer vaccine. More than a million people in the United States have been infected with hepatitis B, and nearly half the global population is at high risk of contracting this virus. Chimpanzees have also been crucial for the study of hepatitis C, a chronic disease for some four million Americans.

And hepatitis just heads the list. AIDS is another prominent example, because chimpanzees are the only nonhuman species that can be infected with HIV-1, the common form of the virus found throughout the world. The reason became clearer last year, when investigators found proof that HIV-1 spread to humans from chimps early in the 20th century. Now more than 36 million people around the world are infected with this virus, and some 22 million have already died from AIDS. Chimpanzees continue to be immensely valuable in the search for a vaccine. Chimps are also helping scientists to battle other dire health problems, including spongiform encephalopathy ("mad-cow disease"), malaria, cystic fibrosis and emphysema.

So the stakes are very high indeed. On the one hand, precisely because they are closely related to us, nonhuman great apes are the only suitable species for research on several diseases that cause immense human suffering. On the other hand, these creatures may be so similar to ourselves that making them the subjects of biomedical experimentation may be difficult to justify ethically. This could be a Faustian dilemma—are we selling chimp souls to better our human lives?

I don't pretend to have the answer to such a weighty moral question. But as a psychologist I do feel qualified to examine one of the reasons being cited for providing these creatures with rights: the idea that nonhuman great apes are self-aware. In my view, the evidence for this assertion is considerably exaggerated. Several lines of reasoning have been put forward to buttress this claim. Let me examine each of these arguments in turn.

First, it is said, the minds of nonhuman great apes are much like ours because some of these animals have been taught to use language—sign language. Research on the abilities of chimpanzees and bonobos to communicate using signs has been going on for more than 30 years. The early excitement that accompanied reports of signing chimpanzees has died down, perhaps because the details are not that compelling. In all these studies, the animal's vocabulary developed painfully slowly, and it never exceeded a couple of hundred signs (about two weeks' work for a healthy two-year-old child). Often the chimp involved could only repeat gestures, and in any case, chimpanzee "sentences" rarely extend beyond one or two signs—making any discussion of grammar or syntax seem rather forced.

The occasional anecdote of a chimpanzee forming a novel combination of signs to represent a word he had not been taught (such as signing water-bird for swan) prompted some to believe that these animals indeed have an innate capacity for language. But in reality such cases are extremely rare and quite hard to interpret. How does one know, for example, that this chimp did not intend two distinct utterances, water and bird, to indicate the two things then in view? In my estimation, such curiosities contribute little to the debate.

Somewhat more relevant are recent reports of a bonobo named Kanzi, whose linguistic abilities (as expressed by pressing buttons on a special keyboard) are alleged to far exceed those seen during the earlier sign-language studies with chimpanzees. A critical test for Kanzi's comprehension of sentence structure involved asking him to respond to an instruction such as "would you please carry the straw?" Sure enough, Kanzi picks up the straw. But there's a weakness here. Although it could be that grammar conveyed the correct meaning of the test sentence, it could equally be just the circumstances that made the requested action obvious (given that Kanzi knows what the words carry and straw refer to). After all, a chimp may carry a straw; a straw cannot carry a chimp. Investigators have also claimed to discern grammatical structure in Kanzi's own keyboard strokes. But because the average length of his utterances is around 1.5 pushes, I believe that a meaningful analysis of grammar or syntax is impossible.

The second tier of evidence for self-awareness in nonhuman apes comes from so-called mirror tests. These experiments can be conducted in several different ways. For example, while the subject is anesthetized or sleeping, the experimenter places a mark on the forehead or ear with an odorless, tasteless dye. Upon waking, the animal is shown a mirror. Will the creature recognize that the splash of dye is on his own face? The answer is taken to be yes if the animal touches the marked area of skin more often with a mirror in front of him than without.

Although there were methodological problems with some earlier studies, it is now broadly accepted that chimpanzees, bonobos and orangutans can recognize themselves in mirrors. Claims that dolphins and gorillas can pass this test are disputed. And all other species examined (including fish, dogs, cats, elephants and parrots) react to themselves in a mirror either not at all or as if the reflection is another animal.

The problem with the mirror test of self-recognition lies not in the results—clearly some nonhuman apes can recognize themselves in mirrors—but in the interpretation. Why should such self-recognition be equated with self-awareness? Some people cannot recognize themselves in mirrors (blind people are the most obvious group), but nobody suggests they lack any aspect of self-awareness. Conversely, in autistics self-awareness is clearly impaired. And yet autistic children develop the ability to recognize themselves in mirrors at the same rate as normal children. So these tests of self-recognition in mirrors are interesting and doubtless say something about how animals view their bodies, but they tell us nothing about the deeper question of self-awareness.

The final set of evidence for self-awareness in nonhuman apes comes from something called "cognitive perspective taking." Daniel Povinelli and his colleagues at the University of Southwestern Louisiana developed the best-known experiment of this sort. They based it on certain tests that are used to judge mental development in autistic children. In Povinelli's "guesser-knower" experiment, a chimpanzee watches a trainer put food into one of four cups. The chimpanzee knows that the cups are there (he was shown them before the experiment started) but cannot see which of them receives the food. Another trainer who has viewed the manipulations (the knower) then points to the baited cup. A third trainer who has not seen the food go in (the guesser) points to a different cup. A chimp—or child—who has an awareness that others have minds, can readily appreciate that one trainer knows where the food is hidden, and the other one doesn't.

Chimpanzees in these experiments were ultimately successful in picking the cup containing the food (selecting the cup pointed to by the knower), but they required many hundreds of training experiences before they could make the right choice reasonably consistently. This pattern suggests only that the apes involved gradually learned to associate one stimulus (the knower) with the reward, not that the animal is treating the trainers as people with minds.

In more recent experiments, Povinelli and his coworkers offered a chimpanzee a choice between begging from a person who could see various pieces of food and begging from a person who could not (because she was blindfolded, for example). To the experimenters' surprise, chimps were initially just as likely to approach the trainer who could not possibly see the food. With enough experience, the chimps gradually learned to ask only the person with unrestricted vision for a tasty morsel. But they showed no spontaneous understanding that being unable to see disqualified a person from providing snacks. So here again, I find little to indicate that apes have any awareness of the minds of others, much less that they have an awareness of their own thoughts.

In truth, scientists do not yet know much about the soul of the ape. But we've learned a few things since Marais made his pioneering observations.It is clear that these animals do not show self-awareness in anything like the ways a human does, even when given the tools to do so. But if their minds are not like ours, what are they like? I don't know. I don't even know how we would draw ethical conclusions from that knowledge if we had it. The only thing I'm sure of is that we should view apes as worthy of wonder for being what they are—not merely as reflections of ourselves.