Our Beastly Selves: Exorcising the Ghosts of Anthropomorphism

In his classic 1872 book, The Expression of Emotion in Man and Animals, Darwin wrestled with the challenges associated with interpreting animal behavior. Can we use psychological attributes to refer to animals in the same way we use them to describe other humans? When Darwin argued that emotions exist in both human and nonhuman animals, his work was branded as anthropomorphic, inappropriately projecting human characteristics onto animals. Ever since, many scientists have been reluctant to attribute mental states to animals, a trend that was exacerbated by the rise of behaviorism with its emphasis on external, directly observable behavior. Consequently, the specter of anthropomorphism continues to haunt attempts to study emotions, personality, and other social phenomena in non-human animals.

Of course, it is proper to exercise caution in interpreting animal behavior. However, a rigid refusal to acknowledge the possibility of emotions and personality in animals may itself be scientifically myopic and ultimately incur its own costs (Hebb, 1946). Most scientists unquestionably accept that the anatomy and physiology of humans has similarities with other animals. If there is cross-species continuity for morphological traits, shouldn’t there also be at least some level of continuity for psychological traits? The question that remains is the degree to which the continuities extend to psychological features; how many of the apparent similarities between humans and non-human animals reflect real similarities, and how many are merely anthropomorphic projections?

Animal studies have always played a central role in psychology. Starting with Pavlov’s classic work on conditioning in dogs, and including such landmark studies as Thorndike’s puzzling of cats and Skinner’s operant conditioning of rats, animal studies have long contributed to learning and other areas of psychology. Can animal studies also be harnessed to examine basic questions in social and personality psychology? The potential benefits of such work would seem to be enormous, contributing to such topics as the effect of animal companions on human behavior and its relation to mental health, as well as providing insight into basic social psychological processes. However, before such benefits can be realized, it is essential to determine the degree to which anthropomorphism obscures human judgments of animals.

With a little imagination, research methods from social psychology and cultural psychology can be mobilized to study anthropomorphism empirically. I have teamed up with Sam Gosling, University of Texas at Austin, and Oliver John, University of California, Berkeley, to address this central issue in animal research. To determine whether the charge of anthropomorphism in animal perception is warranted, we have adopted a comparative approach, comparing side-by-side perceptions of humans and dogs. In one comparative study of interpersonal perception using both dogs and humans as targets of perception, we have examined whether perceptions of animals are susceptible to inappropriate projections. Specifically, we have compared human-to-animal projections (i.e., anthropomorphism) with human-to-human projections (i.e., assumed similarity). Our findings so far suggest that animal caretakers do not project their self-views or their views of other people onto animals any more than they do to other humans. Indeed, projection seems to be a greater problem in judgments of humans than in judgments of animals.

Our research suggests that the idea of emotions and personality in animals cannot easily be dismissed as mere anthropomorphism. Instead, we argue that animal models may have much to contribute to social and personality psychology, just as they have done in other areas of psychology. Yet, very few animal studies appear in contemporary journals or books in social psychology. This has not always been the case. In the 1935 Handbook of Social Psychology edited by Carl Murchison, more than a quarter of the 23 chapters discussed social psychological issues within the context of nonhuman populations. Thus, we are arguing here not so much for a new idea but for the return to a set of ideas that once pervaded the field. We show how modern conceptual and analytical tools can be used to exorcise the ghosts of anthropomorphism, clearing the way for animal studies to make empirical contributions to basic research questions in social and personality psychology.