Skip to main content

Verified by Psychology Today


Dominance and Pseudoscience: Making Sense of Nonsense

Dominance hierarchies in animals are real, and rigorous science tells us so.

Ongoing discussions about whether or not nonhuman animals (animals) form and maintain dominance hierarchies are very useful for learning more about this widespread and well-documented phenomenon. Numerous researchers who have spent upwards of thousands of hours watching a wide variety of animals in the field and in captivity have observed various sorts of dominance hierarchies and there is simply no doubt that they are "real." Detailed discussions and references can be found in two of my own posts (here and here) and in a detailed discussion by renowned primatologist and Psychology Today blogger Dario Maestripieri's post, "Social Dominance Explained Part I." Excellent introductory textbooks in animal behavior also attest to this fact.

On January 1, 2013, dog trainer and author Lee Charles Kelley published a post called "Do Dogs and Wolves Form Dominance Hierarchies? No!" in response to one of my earlier posts called "Dogs, Dominance, and Cesar Millan Redux: Dominance Is Real." Earlier Kelley wrote a post called "A Mea Culpa to Mech, an Apology to Bekoff" in response to another of my posts on social dominance in animals, titled "Social Dominance Is Not a Myth."

Kelley's latest essay has generated some interesting discussion. My purpose here is simply to make a few points that I hope will make sense and also show clearly that his claims and some of the material posted in comments on his post fly in the face of solid scientific data.

Kelley first writes, "Dominance = A Vertical Chain of Command" and "True dominance hierarchies are always top-down systems with a clear and invariable vertical chain of command." Nothing can be further from the truth, as I point out in a brief comment I posted.

One of the most important aspects of dominance hierarchies is that they are not always vertical affairs or linear but this does not mean they do not exist (for discussion of Landau's index, a measure of the degree of linearity of dominance hierarchies, please see here). Indeed, even within a group of animals, some dominance relationships can be linear and others non-linear. (Please see the references to the research of Stony Brook University's Ivan Chase below.) Ethologists and others who have studied this would never make the assumption that dominance hierarchies are always vertical or top-down affairs. Social relationships are formed in a complex array of social interactions that involve animals of different ranks both initiating and terminating these encounters.

Kelley also notes, "Dominant and Submissive Labels Are Meaningless." Of course, taken out of context they might be meaningless, but for those who have spent countless hours studying in detail numerous and diverse species, the labels are extremely useful for understanding the social dynamics of packs, herds, flocks, and other sorts of social groups of animals.

In all honesty, I had trouble following the logic and relevance of much of the rest of Kelley's post, but once again strong science shows clearly that dominance hierarchies are real.

Instinctive behavior can be modified due to individual experience

Let me make one comment about one of the comments posted in response to Kelley's post. Kevin Behan wrote, "The current science says dominance is variable according to a number of factors, and that it is also instinctive and unconscious. But then one will not be able to enunciate this factor of variability without projecting human thoughts and rationales into the minds of the animals, which immediately contradicts the notion of instinct. Which is it, instinct or psychology? If instinct, how then is it variable?" (my emphasis)

In any basic ethology or Animal Behavior 101 course, students learn that dominance is not "instinctive and unconscious." And, students also learn that just because a specific pattern of behavior is instinctive does not mean it cannot be modified due to individual experience. These sorts of individual tweaks lead to variability.

Concerning the variability of instinctive behavior patterns, not only did Nobel-prize winning ethologist Konrad Lorenz, a big fan of stressing the importance of instinctive behavior, point this out in his classic book called Evolution and Modification of Behavior, but renowned ethologist Jack Hailman also showed how instinctive behavior can be modified during early development as part of an individual's learning process as in the instinctive pecking response of young laughing gulls.

In Hailman's classic essay called "How an instinct is learned," he wrote, "The feeding behavior of laughing and herring seagull chicks was studied in both a natural and laboratory environment. Results show that the newborn chick reveals a poorly coordinated peck, motivated by hunger and elicited by the stimulus properties of shape and movement from a parent or sibling. The chick's aim and depth perception improve steadily through practice in pecking. Also, the chick's begging and feeding pecks become differentiated as it learns to rotate its head when begging from the parent. It is concluded that 'behavioral development is a mosaic created by continuing interaction of the developing organism and its environment.'" (my emphasis)

So, even if dominance were instinctive, which it is not, there's nothing to worry about concerning its variable expression by a single individual or by different individuals. Mr. Behan also claims, "The idea of variable dominance, means there is no dominance." This simply is not so. Dominance relationships may, and do vary over time, but this does not mean that dominance hierarchies do not exist nor that dominance does not exist.

I'm a big fan of informed discussions and debates and I hope my brief comments and also Dario Maestripieri's excellent post, some of the references above and below, and some of the comments from people who have performed detailed studies of dominance that show that dominance hierarchies are real are taken seriously. Claims that dominance hierarchies are not real are based on misinformed readings of the results of solid science conducted by many different talented researchers who really know the animals they study, and in some cases might be called pseudoscience.


Please click here and also see:

Bekoff, Marc. 1977. Quantitative studies of three areas of classical ethology: Social dominance, behavioral taxonomy, and behavioral variability. In B.A. Hazlett (editor), Quantitative methods in the study of animal behavior. Academic Press, New York. pp. 1-46.

Bekoff, Marc, and Dugatkin, Lee. 2000. Winner and loser effects and the development of dominance relationships in young coyotes: An integration of data and theory. Evolutionary Ecology Research 2, 871-883.

Bekoff, Marc, Marlene Tyrrell, Valerie E. Lipetz, and Robert A. Jamieson. 2001. Fighting patterns in young coyotes: Initiation, escalation, and assessment. Aggressive Behavior 7, 225-244

More from Marc Bekoff Ph.D.
More from Psychology Today