Animal Behavior’s Biggest Taboo Is Softening

At the start of Elizabeth Hobson’s career as an ecologist, she knew to stick to one rule: Never anthropomorphize the animals you study.

For plenty of people, assigning human characteristics to another living creature feels natural enough that we do it as a matter of course. But to many scientists who study animal behavior, anthropomorphism is a cardinal sin, and suggesting that a researcher has tiptoed in that direction is tantamount to saying they’ve resorted to uninformed speculation. Hobson, who studies birds at the University of Cincinnati, told me that when she was trying to get a foothold in her field, the mere accusation of anthropomorphism might have been enough to ruin her credibility.

But in recent years, a slow revolution has been unfolding among a contingent of animal-behavior researchers who argue that our impulses about other species, rooted in our own experiences of the world, are scientifically useful. Other animals do share our physiologies, habitats, and genes (to varying degrees); if anthropomorphism draws on those commonalities, it offers legitimate, testable ideas about other creatures’ experiences. For many animals, there’s even “a good case to be made that it’s the right approach to assume, until we know otherwise, that there’s similarity,” Amy Parish, a primatologist at the University of Southern California, told me. Besides, the idea that anthropomorphism, so ingrained in human nature, can be fully stamped out is a myth, Ambika Kamath, a behavioral ecologist who’s writing a book about animal behavior, said. If anthropomorphism can’t be eradicated, perhaps it can be tamed by scientists who learn to wield it wisely.

[Read: Do animals have fun?]

Just 150 years ago, many naturalists took for granted that animals could and should be much like us. Darwin described disappointment in dogs and cunning in cobras, and argued that there existed “no fundamental difference between man and the higher mammals in their mental faculties.” His protégé George Romanes wrote of rooks putting a jackdaw on trial, a pet snake that died from shock upon glimpsing its ailing master, a monkey guilt-tripping the hunter who shot it by smearing its hand with blood.

By the late 1800s, other scientists had begun to loudly protest these accounts, and called for a new era of behavioral research, ruled by empirical observations and only the most irrefutable evidence. Anthropomorphism became regarded as lazy; today, researchers such as Clive Wynne, a behavioral scientist at Arizona State University, contend that it amounts to “short-circuiting the real work of doing science.”

But that position had its shortcomings, too, Gordon Burghardt, an ethologist at the University of Tennessee at Knoxville, told me. Researchers focused only on external observations have dismissed (and still do dismiss) the possibility that animals might have tough-to-quantify emotions or complex internal lives. Rather than friendships, animals have affiliative relationships. They could experience only fear-like and anxiety-like responses; what looked like alarm was a creature perceiving a flight-eliciting stimulus. Laughter, too, was euphemized as “positive affective vocal responses to tickling,” or simply put in quotes. Even the idea of pain in other animals became controversial, with some scientists chalking up the jerks and twitches of creatures experiencing physical harm to reflexes. To this day, Margaret Gruen, a veterinarian and animal behaviorist at North Carolina State University, encounters colleagues who refuse to use the term suffering for other species.

But even with the moratorium on anthropomorphism, scientists were still letting human hang-ups shape their work. Researchers took decades to come around to the notion that bonobo societies were ruled by female dominance, Parish, who studies the apes, told me. And some have dismissed same-sex relationships in other creatures as paradoxical or maladaptive, Kamath said—despite evidence supporting the notion that those behaviors do come with benefits, such as improving social relationships among bonobos and dolphins.

[Read: Surprise! Snakes have clitorises]

Some scientists have now come to think that stamping out anthropomorphism was never feasible. The inclination for humans to see themselves in their surroundings is too automatic, Esmeralda Urquiza-Haas, a cognitive scientist in Austria who has studied the basis for anthropomorphism, told me. People see faces in architectural features; they give cars and  boats pronouns, and assign personalities and motivations to shapes moving across a screen. Anthropomorphism may just be a natural part of being a social creature, anticipating and inferring the motivations of others we interact with, including those of different species.

And the more that scientists have studied animal behavior, the more they have had to admit that other creatures are “more like us than we used to give them credit for,” Joshua Plotnik, a psychologist at Hunter College, told me. Octopuses can use tools; wasps can distinguish faces; orcas cooperate to hunt seals. Orangutans can tease; ravens exhibit self-restraint; dolphins even have a way to call each other by name. Humans, too, are animals, Burghardt said. So why wouldn’t it be the case that many of our traits—down to our motivations and needs—are shared across other life forms? To deny other animals that possibility would be its own fundamental error.

[Read: Great apes know just how much to annoy one another]

“The pressure to avoid anthropomorphism at all costs has lessened,” Plotnik told me. His current studies on elephants, which delve into concepts such as cognition and intelligence, would probably have gotten him laughed out of most psychology departments several decades ago. Now, though, many academics are comfortable describing his study animals as clever, cooperative, and capable of thinking and feeling. This more permissive environment does put that much more pressure on researchers to weigh exactly how and where they’re applying anthropomorphism—and to do so responsibly. But it’s also an important opportunity “to use our anthropomorphic lens carefully,” Kwasi Wrensford, a behavioral biologist at the University of British Columbia, told me.

Anthropomorphism can sometimes be spot-on. The key, Plotnik said, is actually gathering the evidence to back up your hunch. That’s become one of the basic tenets of what Burghardt calls critical anthropomorphism—using anthropomorphic tendencies as fodder for generating hypotheses that can then be tested. Plotnik, for instance, has shown that elephants can console each other, by documenting how they proactively caress other individuals showing signs of distress. Other scientists have found that bonobos are capable of foresight, by showing that the apes will stash tools that aren’t useful to them in the present but will become handy in the future. Still others have found that crows can remember individual faces—by donning rubber masks, temporarily trapping individual crows, and recording the birds later scolding people who are wearing the same getup. No single study will ever be airtight, and “plenty of people will never accept it regardless of how much objective evidence you give,” Plotnik told me. But the foundations for these findings may be stronger than they’ve ever been.

And when hypotheses do turn out to be wrong, as hypotheses sometimes are, these same careful experiments can leave scientists with new ideas, rather than back at square one. Alexandra Horowitz, a canine-cognition researcher at Barnard College, told me that she was in part inspired to run an experiment a few years ago by a sentiment many pet owners share: that dogs get a guilty look when they realize they’ve done something bad. But her research showed that the remorseful gaze was actually sparked by their owners’ chastising—“better understood as a really good reading of us,” Horowitz told me, than an understanding of right and wrong.

Plus, allowing for a degree of anthropomorphism can free scientists to describe their findings in less stilted ways. At the University of Cincinnati, researchers in Hobson’s lab debated how to describe the concerted aggression they observed when a high-ranking monk parakeet vanished from a social group, then attempted to reintegrate. They worried at first that the word bullying would project middle-school-esque dynamics onto the birds—the popular kids snubbing a former member of their circle because “we hate you now,” Hobson said. But the term was also excellent shorthand to describe what the birds were doing. “We’re just careful to define exactly what we mean,” she said: “an increase in aggression towards a specific individual from all the other birds in the group.”

Plenty of researchers, Wynne included, remain skeptical that anthropomorphism can accomplish net good. Even if there’s evidence to back the notion that an animal experiences, say, shyness, defaulting to that answer might stop scientists from finding additional, less intuitive explanations. Anthropomorphism can also narrow the lens through which researchers view other species, many of which are capable of some very nonhuman feats: Bats echolocate; birds use quantum effects to navigate; bees can sense electric fields; mosquitoes can see in infrared. Project too much of what we do, and scientists will miss the ways in which other animals experience the world. “I find it very disappointing to keep looking for ourselves wherever we go,” Wynne told me.

[Read: How animals perceive the world]

Many scientists are now trying to guard against these types of errors—following intuitions about animals’ cognitive complexity, but searching for answers through means that aren’t just primarily suited to us. Gruen’s work in cats, for instance, has found that feline pain manifests not as moaning and groaning, but as subtle changes in daily routine, including whether the animal has gotten worse at leaping onto high surfaces, or is hesitating to climb stairs. At the Max Planck Institute of Animal Behaviour, Alex Jordan’s lab is trying to confirm interpretations of certain cichlid behaviors by directly including the perspectives of the fish. To test whether a certain behavior is a threat display, for instance, the researchers use artificial intelligence to generate moving avatars of the animals, Jordan told me, then play back that action to cichlids in the wild and in the lab to gauge if their response matches up.

In the same way that scientists could never be certain that they were completely stripping anthropomorphism from their studies, there is no guarantee that they’re self-aware enough to catch themselves overusing it. We struggle enough to see the perspectives of other people; to do so with another creature, with its own sensory repertoire and its own evolutionary path, requires even greater leaps. Still, accepting the inevitability of anthropomorphism may be more responsible than insisting that it can be purged, Kamath told me. Researchers who do the latter may risk something worse: a false sense of their work’s objectivity.

Detachment, after all, shouldn’t always be the goal. Rejecting anthropomorphism too vehemently “can justify doing ethically questionable things,” Wrensford said: treating animals without mercy, or as expendable obstacles to our goals. The value of other species shouldn’t be  dictated only by how much they resemble us. But by ignoring all instincts to think of them like ourselves, we lose our best shot at empathy.