Boulder, Colorado -- When Marc Bekoff talks about his dog, Jethro, he frequently uses human qualities to describe its behavior.
Jethro is playing, he might say. Or Jethro appears happy. Or Jethro seems threatened or feels pain.
To most people, such phrases are perfectly reasonable ways to describe the actions of a dog. But to scientists, ascribing human emotions or qualities to an animal makes Mr. Bekoff, a biologist at the University of Colorado here, guilty of one of the worst sins in science -- anthropomorphism.
Long considered taboo among researchers studying animals, anthropomorphism violates a central tenet of science: that researchers should strive to be totally objective and dispassionate observers of nature. As a result, scientists who suggest that animals possess intentions, emotions, or other qualities assumed to be uniquely human are often viewed by colleagues as careless, gullible, or even irresponsible.
So despised is the practice that animal researchers can discredit others in their field by simply labeling their work anthropomorphic.
“It’s a degrading term,” says Helen E. Fisher, an anthropologist at Rutgers University who studies similarities in the behavior of animals and humans. “They’re sneering when they use it.”
While no one believes researchers who talk about such blatantly erroneous concepts as “marriages” among monogamous animals should escape scorn, a growing number of ethologists think it is wrong simply to dismiss anthropomorphic thinking outright. They have begun arguing in journals and other scholarly publications that anthropomorphism, when used selectively, can be a valuable scientific tool.
“There are certain patterns of behavior that simply can’t be made intelligible to humans without using human terms,” says Mr. Bekoff, a professor of environmental, population, and organismic biology. “I can describe to you until the cows come home the anatomy and physiology of play. Or I can describe to you anatomically and physiologically what my dog, Jethro, does if you put a match on his paw. But in the absence of a context of what’s going on, you’d have no idea.
“If I called you on the phone and said, ‘Jethro’s in pain because he stepped on a cactus,’ you could imagine what he’s doing. Basically, you purchase a lot with anthropomorphism, and it has to be used carefully. But it’s important because it motivates new studies by taking into account the richer cognitive life of animals.”
Gordon M. Burghardt, a psychology professor at the University of Tennessee, discovered how valuable anthropomorphism can be several years ago while conducting a study on deception in hognose snakes. When confronted with predators, such snakes stop breathing, bleed from their mouths, and flip on their backs with their tongues hanging out.
Some scientists believed the reaction was due to fright, but Mr. Burghardt showed that the snakes were consciously attempting to deceive their predators. In one experiment to demonstrate that this was so, he sought to determine whether the time hognose snakes played dead would be longer when a predator did not leave. He started by comparing the time it took the snakes to recover with and without a stuffed owl and, later, with a human. As expected, the snakes took longer to recover when both potential predators remained, instead of being taken away.
But then Mr. Burghardt did something unusual: Imagining himself as the snake, he asked how he would assess the potential danger of a predator. That prompted him to conduct an additional experiment in which he discovered that the hognose snake would take longer to recover if a human “predator” was looking directly at the snake, rather than away from it.
“It was selectively attentive to the eyes,” Mr. Burghardt says. “This is something I don’t think I would have ever done if I didn’t imagine myself in that situation and ask, What would make sense?”
Whether they admit it or not, Mr. Burghardt says, many other animal researchers, including critics of anthropomorphism, consciously use anthropomorphic thinking to construct useful, testable hypotheses in their research. In several publications, he called on his colleagues to openly acknowledge this practice, which he dubbed “critical anthropomorphism” to distinguish it from the uncritical application to animals of human qualities that violate what is known about an animal’s biology.
For example, he says, while it might be unreasonable for scientists to assume that a dog standing in the snow is cold, because canines can withstand much colder temperatures than humans, it might be reasonable for them to assume that dogs, like humans, enjoy the flavor of some types of food over others.
Frans B.M. de Waal, a professor of psychology at Emory University who studies chimpanzees, believes ethologists like himself would be seriously hampered if they didn’t use critical anthropomorphism in their work.
“We propose something,” he says. “We say it looks like this, and we could be using a very anthropomorphic term, but we would feel an obligation to back that up and see if we can find the evidence for it. Sometimes we do and feel it’s supported. Sometimes we don’t and then drop the term.”
The key is selecting the proper term. Ethologists say their choice of words or phrases to describe an animal’s behavior -- “play,” “appeasement,” “submission,” or “cooperation” -- can make all the difference between success and failure for a research paper.
Robert W. Mitchell, an assistant professor of psychology at Eastern Kentucky University who is writing a book about the history of anthropomorphism, says that in the 1940’s a battle ensued among biologists after a researcher claimed to have discovered that mosquitoes emitted “distress calls.” But when the scientist changed his description to “warning calls,” the controversy ended.
“They were thinking that when the person said ‘warning calls,’ he was talking about something functional,” Mr. Mitchell says. “But when he said ‘distress calls,’ that sounded psychological.
“People get obsessed with these words, and they think they know what the other person means,” he adds. “Even when the other person writes back and says ‘I just meant it functionally,’ they still end up saying this person doesn’t know what he’s talking about.”
Attempts to clarify a description by making it appear more objective or functional doesn’t always work. In his defense of anthropomorphism in the journal Behaviour, Mr. de Waal argued that substituting anthropomorphic terms like “play” -- with its psychological connotations of fun and lack of seriousness -- with a seemingly more accurate description such as “rearranged elements drawn from other types of behavior” doesn’t enlighten anyone.
“Such a label may sound more objective,” he wrote, “yet the problem then would be to decide which rearrangements of behavior to include, and which to exclude.”
Mr. Mitchell says feuds among ethologists over what terms are or aren’t anthropomorphic often boil down to their beliefs about the cognitive abilities of certain animals.
Take bears and chimps. Both hide food and other items under themselves. But while primatologists might label the chimps’ behavior deception, they probably wouldn’t do the same for the bear.
“When chimps do something, it seems really smart,” says Mr. Mitchell. “But when bears do something, it doesn’t seem very smart.”
“If you say an animal deceives,” he adds, “then it suggests that the animal is aware of what it’s doing and aware of the consequence of its own actions and that it’s thinking, and most people aren’t willing to concede that.”
Part of the reason is philosophical. In the 19th century, Darwin and many other scientists freely attributed emotions to animals. But with the rise at the turn of the century of the hard sciences and positivism, Western scientists became increasingly reductionist. “They began looking at animals as stimulus-response machines,” says Mr. Bekoff of Colorado.
In a study in the early 1980’s of attitudes toward anthropomorphism, Pamela J. Asquith, an associate professor of anthropology at the University of Calgary, found that Japanese scientists, unlike those from Western countries, had no problem accepting emotions and personalities in animals. “They weren’t hampered by the Judeo-Christian heritage of thinking of humans as separate and superior,” she says.
In a recent followup study, however, she found that Western scientists appear to have grown less rigid about denying anthropomorphism because of increasing recognition of the complexity of animals.
Mr. Bekoff is confident that this trend will benefit science.
“Over the next 15 or 20 years,” he says, “you’re going to see a lot of acceptance of anthropomorphism and its utility for motivating good, hard science.”
“More and more people are watching animals, and they’re seeing things that simply can’t be explained in simple stimulus-response types of explanations.”
“The other thing,” Mr. Bekoff adds, “is that people are really tuned in to evolutionary continuity and they’re realizing that there is also a continuity in brains. If brains are somehow associated with mind and cognition, then it would be absolutely ludicrous to think that humans de novo acquired certain cognitive abilities, whereas maybe a few monkeys did, but no dogs or cats.”
The question is where to draw the line. Why should scientists be able to assume monkeys or dogs have human qualities, but not rats or ants? Many scientists concede that common sense and the biology of animals are their only guides.
“Do I think ants experience joy?” asks Mr. Bekoff. “I doubt it. I don’t know that they don’t, but I doubt it. Do I think dogs and cats experience joy and pain and suffering? Yes, I do. Just by watching them, you can tell that there are things that they enjoy and things that they dislike.”
Linnda Caporael, a psychology professor at Rensselaer Polytechnic Institute, however, warns that ethologists should not be deceived into thinking that they can understand an animal’s intentions simply because its behavior may mimic or parallel that of humans. Scientists, she says, need to understand more about how their own minds filter this evidence before they can say with any conviction what an animal may be doing.
“Science is in the business of overturning common sense,” she says. “When a dog yawns, it’s not sleepy. It’s under stress.”
Some critics of anthropomorphism also see problems in the tendency of researchers to impose their own values on the animals they study. For example, the assumption that animals, like ourselves, prefer more spacious quarters has resulted in a revision of animal-care rules that requires bigger cages for laboratory rats, which actually prefer the confines of a smaller cage. In primatology, researchers are engaged in a fierce debate over whether it is appropriate to measure intelligence in primates by human standards.
“For some people, an intelligent animal is one that obeys or understands what we say,” says Andrew N. Rowan, director of the Center for Animals in Public Policy at the Tufts University veterinary school, rather than learning what is necessary to stay alive in the wild.
“Insisting that an animal uses information the way we use it is not only poor science, but reflects a kind of arrogance,” says Alan M. Beck, director of Purdue University’s Center for Applied Ethology and Human-Animal Interaction. “We don’t relate to odor, so we minimize it in studies. We insist that monkeys learn our language to be intelligent. Sure, other animals aren’t writing books, but we don’t fly, we don’t echo-locate.”
Despite such problems, many scientists believe animals will be the main beneficiaries of the increased acceptance of anthropomorphism. As more scientists acknowledge that animals have emotions and other human qualities, they believe, concerns over the welfare of laboratory animals will increase.
Mr. Beck of Purdue notes that one reason scientists have traditionally denied feelings to the animals on which they conducted experiments was the necessity to separate themselves emotionally from their research subjects. “For years, we didn’t say animals feel pain, because it made it easier to do the research we wanted to do,” he says. “We need barriers.”
Mr. Bekoff, a vegetarian who has mixed feelings about animal experimentation, agrees. “If you start attributing human characteristics to non-human animals,” he says, “if you’re in touch at all with your emotions, it’s going to influence what you would allow.”
