IF YOUR DOG DROPS A TENNIS BALL in front of you and looks up at you with tail wagging, do you figure she wants to play? How naive! Who says dogs have desires and intentions? Her behavior is merely the product of reinforcement: she has been rewarded for it in the past.

Many scientists have grown up with the so-called law of effect, the idea that all behavior is conditioned by reward and punishment. This principle of learning was advocated by a dominant school of twentieth-century psychological thought known as American behaviorism. The school’s founders, John B. Watson and B. F. Skinner, were happy to explain all conceivable behavior within the narrow confines of what Skinner called “operant conditioning.” The mind, if such a thing even existed, remained a black box. In the early days, the behaviorists applied their doctrine in equal measure to people and other animals. Watson, for instance, to demonstrate the power of his methods, intentionally created a phobia for furry objects in a human baby. Initially “little Albert” was unafraid of a tame white rat. But after Watson paired each appearance of the rat with sharp noises right behind poor Albert’s head, fear of rats was the inevitable outcome. Even human speech was thought to be the product of simple reinforcement learning.

The behaviorists’ goal of unifying the science of behavior was a noble one—but alas, outside academia the masses resisted. They stubbornly refused to accept that their own behavior could be explained without considering thoughts, feelings, and intentions. Don’t we all have mental lives, don’t we look into the future, aren’t we rational beings? Eventually, the behaviorists caved in and exempted the bipedal ape from their theory of everything.

That was the beginning of the problem for other animals. Once cognitive complexity was admitted in people, the rest of the animal kingdom became the sole standard-bearer of behaviorism. Animals were expected to follow the law of effect to the letter, and anyone who thought differently was just being anthropomorphic. From a unified science, behaviorism had become a dichotomous one, with two separate languages: one for human behavior, another for animal behavior. Human rationality and superiority are not really the issue, however—one only needs to read the latest Darwin Awards to notice that our species can be less rational than advertised. The issue is the dividing line between us and the rest of nature.

Radical behaviorists adamantly insist on this line, and look across it with entirely different eyes than the ones they reserve for their fellow human beings. They speak about animals as “them” and compare “them” with “us,” as Clive D. L. Wynne does at the beginning of Do Animals Think? (“What are animals—really? What should we make of them?”). Other behaviorists, however, intentionally blur the line. They apply the same well-tested behaviorist methodology to reconnect human and animal behavior, daring to mention the words “animal” and “cognition” in the same breath. They write books such as Duane M. Rumbaugh and David A. Washburn’s Intelligence of Apes and Other Rational Beings.

Of the two, Wynne’s book is by far the more readable. Wynne has a pleasant writing style and a knack for engaging the reader. He begins with the story of a mad animal-rights activist who threatened the lives of people on the Isle of Wight, where Wynne grew up. The man was convinced that animals are sentient beings, a certainty Wynne says he wishes he could share.

This story sets the tone of doubt and reserve that permeates the book. Wynne includes numerous insightful accounts of remarkable animal behavior, but he invariably concludes on a note of caution: one should not infer too much from these accounts. He is not so radical a behaviorist that he excludes all forms of reasoning by animals, but he takes greater pleasure in explaining what animals cannot do—monkeys fail to understand relations between cause and effect, apes can sign but lack the syntax that defines human language—than in describing what they can do. Capacities unique to a particular species, such as echolocation in bats, get Wynne’s full admiration. But anything that seems to elevate other animals close to the lofty cognitive level of humankind he regards with utmost skepticism. He seems to take delight in animals, and possesses great knowledge about them, yet he prefers them at arm’s length. The constant message is that animals are not people.

That much is obvious. But it is equally true that people are animals. The dichotomy Wynne advocates is outdated, lending his book a pre-Darwinian flavor. Take the case of animal culture, currently one of the hottest areas in the study of animal behavior. The idea goes back to the pioneering work of Kinji Imanishi, who proposed in 1952 that if individuals learn from one another, their behavior may grow so different from behavior in other groups of the same species that they seem to have their own culture. Imanishi thus reduced the idea of culture to its most basic feature: the social rather than the genetic transmission of behavior.

Many examples of animal culture have been documented. The classic case emerged among wild macaques on Japan’s Koshima Island. During their fieldwork with the monkeys there, investigators provisioned them with sweet potatoes, which a juvenile female named Imo soon began washing; she would bring her potatoes to a small river and clean them off before eating them. Imo’s washing behavior spread first to her mother and then to her age peers, before affecting the rest of the group. Later Imo moved her operation to the shoreline, washing the potatoes in the ocean, and, again, the other monkeys followed.

Some psychologists have objected to this example, pointing out that it is uncertain whether the monkeys learned their skill by copying others or by discovering the behavior individually, without anyone’s help. Wynne supports the second view. But instead of basing his opinion on the actual data published by a team of Japanese primatologists, who have worked on the problem for fifty years, he relies on the word of a skeptical Westerner who has never set foot on the island. This scientist, a specialist in rat behavior, suggested that potato washing spread because performers were selectively rewarded by the people who handed out the potatoes.

A few years ago I went to Koshima Island to verify the idea of selective rewarding. I talked with some of the people who had actually witnessed Imo cleaning her first spud. They told me that initially the monkeys were fed far away from any water, so there was no question of rewarding any washing behavior. Imo herself came up with the idea of transporting the potatoes to the river for cleaning. They also pointed out that one cannot feed a group of monkeys any way one wishes. The dominant males have to be fed first, the females second, and the little ones last; changing the order sparks bloodshed. Thus, except for Imo’s mother, the monkeys that learned the behavior first, the juveniles, were the last to be rewarded. In fact, the only monkeys on the island that never learned potato washing were the adult males: precisely the best-rewarded group.

Wynne invariably favors interpretations that widen the assumed cognitive gap between human and animal. For example, he uncritically accepts the uniqueness claim du jour: that only human beings possess a theory of mind (ToM), or the cognitive ability to understand that others, too, have mental states such as thoughts and knowledge. Ironically—given Wynne’s dismissal of an ape ToM—the concept got its start with a 1970s study of chimpanzees. A female showed she had grasped the intentions of others by, for example, selecting a key from among several tools if she saw a person struggling to open a locked door.

Evidence for a theory of mind in apes has gone through its ups and downs ever since. Some experiments have failed spectacularly, leading the proponents of one school of thought to contend that apes simply lack the capacity. Negative results are inconclusive, though: as the saying goes, absence of evidence is not evidence of absence. Furthermore, the performance of apes is often assessed by comparing it with that of children. Because the experimenter is invariably human, however, only the apes face a species barrier. When an ingenious experiment conducted at Emory University’s Yerkes National Primate Research Center in Atlanta got around that problem, the evidence for an ape ToM was more positive: chimpanzees seemed to realize that if a member of their species had seen hidden food, this individual knew where the food was, as opposed to one who had not seen it. That finding threw the question of a ToM in nonhuman animals wide open again.

In an unexpected twist (because the debate has focused on humans versus apes), a capuchin monkey in a laboratory at Kyoto University in Japan recently passed a series of seeing-knowing tasks with flying colors. The least one can conclude is that it is premature to settle on ToM capabilities as the ultimate Rubicon.

In spite of Wynne’s dismissal of an ape ToM, his book offers many insightful descriptions of animal behavior. A wonderful chapter on the role of messenger pigeons during the First World War includes a picture of the stuffed body of Cher Ami, a genuine war hero. The pigeon kept flying after its leg had been shot off, delivering its message and thus rescuing an entire battalion.

Rumbaugh and Washburn are considerably more open-minded about the mental accomplishments of animals than Wynne is. Their book celebrates Rumbaugh’s lifetime of research on monkeys and apes. In fact, what fascinates me the most about Intelligence of Apes and Other Rational Beings is its historical overview of experimental work with primates, first with the Wisconsin General Testing Apparatus (WGTA) and later with joysticks and computers.

The WGTA was developed at the University of Wisconsin in the 1940s, and is still being used today. In this set-up, a primate subject in a cage faces an experimenter across a platform, on which differently shaped or colored stimuli are arrayed. Both experimenter and primate can reach the stimuli; the experimenter baits them with rewards, and the primate selects among them. I remember working with such an apparatus as a student, testing chimpanzees to see if they could discriminate shapes by touch alone. The task was so incredibly simple and repetitive that the apes invariably got tired of the whole thing five minutes into the testing. In fact, they got so bored that they performed worse than macaques tested on the same stimuli.

I mention this episode because test performance is often taken as a measure of intelligence, even though attention and motivation are equally important to the outcome. As a result, failure is open to interpretation. Rumbaugh and Washburn understand these points better than most scientists, and they are at pains to remind the reader how the questions one asks tend to constrain the answers one gets.

Indeed, some testing paradigms positively suppress the phenomena being tested. When Rumbaugh replaced the WGTA with an innovative testing setup in which the monkeys move a joystick to select stimuli on a computer screen, their performance improved dramatically. Rumbaugh’s work on the connection between method and outcome should be required reading for anyone who attaches significance to negative evidence.

One learning paradigm discussed by Rumbaugh and Washburn has special interest. Some animals learn how to learn—that is, once they have mastered a particular task, they can more quickly learn future tasks that have the same design but rely on different stimuli. Trial-and-error learning cannot explain improved performance in reaction to new stimuli, hence the level of learning must be higher. But generalization across tasks is precisely what the founders of behaviorism thought animals could not do.

Rumbaugh and Washburn discuss many forms of advanced problem-solving, which they classify as “emergents.” The term is slightly awkward, but the authors apply it to cases in which animals flexibly apply accumulated knowledge to new situations, resulting in an “emergent” solution. The classic example is the chimpanzee in a room with a few sticks and boxes in one corner and, for the first time in the chimp’s experience, a banana hanging from the ceiling. The solution emerges as the old bits of previous knowledge combine until, as if a lightbulb suddenly goes on in the chimpanzee’s head, he climbs on top of the boxes and reaches for the banana with a stick.

The two authors rightly speak of reasoning and rationality, and so adopt a terminology that is anathema to radical behaviorism. They discuss the behaviorist view at length but choose to deviate from it, stressing continuity between animal and human. For the reader, though, it is frustrating that they focus almost entirely on apes and other primates, without examining how the concept of emergents could apply equally well to other animals. Crows, dolphins, elephants, and parrots have been credited with creative problem-solving as well.

There will always be tension between those who view animals as only slightly more flexible than machines and those who see them as only slightly less rational than human beings. The views discussed in these two books are by no means as far apart as they could be; both, after all, come out of the same tradition of experimental psychology. Throw in a few naturalists and neuroscientists, and the debate gets even more complex. That said, however, the two books range widely enough across the spectrum of views to make a powerful case that there is still plenty to be discovered, and that human uniqueness is largely in the eye of the beholder.