Dominance—Making Sense of the Nonsense

Roger Abrantes and Wolf

A relationship is a natural thing! (Photo by Monty Sloan)

Stable and profitable relationships
are not built in the long run by means of a series of dominant and submissive displays. Instead, these behaviors are necessary for resolving inevitable social conflict. Both humans and dogs (and wolves, of course) build relationships on the need for partnership in overcoming common problems related to surviving and, preferably, achieving an acceptable level of comfort. Relationships are not built on hierarchies, but they do exist and they do play an important role in certain circumstances—for humans as well as dogs (and wolves of course)—sometimes more, sometimes less and sometimes not at all.

Illustration showing the possible combinations of aggressive, fearful, dominant and submissive behavior in social canines (From "Dog Language" by Roger Abrantes, illustration by Alice Rasmussen). Copyrighted illustration.

Illustration showing the possible combinations of aggressive, fearful, dominant and submissive behavior in social canines (From “Dog Language” by Roger Abrantes, illustration by Alice Rasmussen). Copyrighted illustration.


In everyday language,dominance means “power and influence over others.” It means supremacy, superiority, ascendancy, preeminence, predominance, mastery, power, authority, rule, command and control. The word has so many meanings and connotations that we cannot simply pick a dictionary definition and employ it as a scientific term in the behavioral sciences. Terms need to be accurately defined in order to avoid misunderstandings, meaningless discussions and nonsensical claims. Unfortunately, the scientists who use this term (as well as those who repudiate it) have not managed to define it satisfactorily, thus contributing to the current confusion about the nature and function of dominant behavior.

I intend to remedy this by:

(1) demonstrating that dominance is an observable characteristic of behavior;

(2) establishing that it refers to one and the same class of behaviors independent of species;

(3) presenting a precise, pragmatic and verifiable definition of the term, which is compatible with evolutionary theory and our body of biological knowledge;

(4) arguing that, even though it is true that a good (in terms of being profitable and stable) relationship does not rely on continuous displays of dominance/submission from the same individuals toward the same other individuals, this does not imply that dogs cannot show dominant behavior.

Denying that dominant behavior exists in dogs has become a popular argument to defend the claim that we must not use ‘dominance’ to build a relationship with our dog.

Indeed, the discussion on dominance has run away with us; and there is only one thing more absurd and futile than attempting to prove that dominant behavior exists and that is attempting to prove that dominant behavior does not exist. In the following, I shall commit the first of these futile acts.

Wolf Pack

In a stable pack, wolves mostly display dominant and submissive behavior and seldom aggressive and fearful behavior.


It is absurd to argue that dominance (as an attribute, a property) does not exist when there are so many words for it, depending on context and nuance. If it didn’t exist, neither would these terms. The numerous synonyms and connotations suggest, not only that the term is hard to define, but also that a property of behavior has been observed whose features are sufficiently dissimilar to other properties to make it worth classifying and naming. Whether or not the name or names given are appropriate or well defined is another story, and has no influence whatsoever on the behavior in question. We can argue that this attribute (dominance) has been observed and that (1) it only refers to particular human relations, or that (2) it refers to particular relations among humans as well as some other animal species. The second option seems more appealing, considering that it is highly improbable that a particular condition would only exist in a single species. This would go against all we know about the relatedness and evolution of species.

However, there is nothing implausible about stating that the term does not apply to describe the behavior of a particular species. On the contrary, two species that diverged from a common ancestor billions of years ago have evolved and developed characteristics of their own and now differ, both from the common ancestor and from one another. By the same token, closely related species, which diverged from a single common ancestor a few thousand years ago, will show various characteristics, similar or equal to the common ancestor and to one another. Some species share many common attributes in terms of phenotype, genotype and behavior, others less, some none at all. It all depends on their common ancestry and their adaptation to the environment.

English: Saarloos Wolfdog male Polski: Samiec ...

Wolf-dog hybrid (Image via Wikipedia).

Humans and chimpanzees (Homo sapiens and Pan troglodytes) diverged from a common ancestor about 6 million years ago, so we can expect them to have fewer common attributes than wolves and dogs (Canis lupus lupus and Canis lupus familiaris) which  only diverged from a common ancestor  about 15 to20 thousand years ago according to recent studies (and by no means, more than 100 thousand years ago). The DNA of humans and chimpanzees differs to a greater degree than that of wolves and dogs (which is almost identical except for a few mutations). Humans cannot interbreed with chimpanzees; wolves and dogs can interbreed with each other and produce fertile offspring. Humans and chimpanzees are two completely distinct species. Wolves and dogs are two sub-species of the same species.

These facts considered, we could expect wolves and dogs to show a great number of similarities, which indeed they do, not only physically but also behaviorally—and any laymen will attest to that. Their similarities on certain levels are what makes it possible for them to mate, produce fertile offspring, and communicate. Nobody questions that wolves and dogs have a very large, common repertoire of communication behaviors; and rightly so, for multiple observations have confirmed that they do communicate perfectly well. Their facial expressions and bodily postures are remarkably similar (except for a few dog breeds) with only a few small differences, these being smaller than cultural differences between humans from some geographically separated settlements.

If wolves and dogs can communicate, it follows that the basic and crucial elements of their languages must be the same. This means that even though they evolved in apparently distinct environments, they retained the most anchored elements of their genotypic characteristics. This could be for the following reasons: (1) the common genotypes are vital to the organism, (2) the environments were not so crucially distinct after all, (3) evolution needs more time and more selective conditions (since it operates on phenotypes) for the genotypes to begin to differ radically.

Point (1) above means that there are more ways of not being alive than being alive, or, in other words, that evolution needs time to come up with different, viable life forms. Point (2) means that even though wolves and (pet) dogs now live in completely different environments, the phenomenon is still too recent. It is only in the last century that dogs became so overly domesticated. Until then, they were our companions. They were domestic animals that still maintained a high degree of freedom and depended (mostly) on the same successful selective factors as always. They were still not pets and breeding was not totally (or almost totally) controlled by human selection. Point (3) means that we might one day (in a million years or so) have two completely distinct species: wolves and dogs. By that time, they will not mate, will not produce fertile offspring and will perhaps show some completely different characteristics; and we may change the dog’s name to Canis civicus, or Canis homunculus. However, we are not there yet!

Recent trends claim that “dominant behavior” does not exist in dogs, which poses some serious problems. There are two ways to argue in favor of such thinking. One is to dismiss “dominant behavior” outright, which is absurd as, for the aforementioned reasons, the term does exist, we know roughly what it means and we use it in conversation. It must, therefore, refer to a class of behaviors that we have observed. Another argument is to claim that wolves and dogs are completely different and therefore, even though we can apply the term to explain wolf behavior, we cannot use it to describe dog behavior. If they were completely different, the argument would be valid, but they are not, as we have seen. On the contrary, they are very similar.

A third alternative is to propose a brand new theory to explain how two such closely related species, as the wolf and the dog (actually sub-species), can have developed in such a short period (thousands of years) with so many radically different characteristics in one aspect, but not in others. This would amount to a massive revision of our entire body of biological knowledge, with implications far beyond wolves and dogs—an alternative I find unrealistic.

A far more appealing approach, it seems to me, is to analyze the concepts we use and define them properly. This would allow us to use them meaningfully when dealing with different species without running into incompatibilities with the entire body of science.

An accurate definition of “dominant behavior” is important because the behavior it encompasses is crucial to the survival of a certain type of individuals, as we shall see.

To dismiss the existence of facts underlying a term, just because that term is ill-defined, not to mention politically incorrect (which means that it doesn’t suit our immediate goals) seems to me to be a poor approach. Dominant behavior exists, but it is badly defined (when defined at all). Most discussions involving dominant behavior are meaningless because none of the parties know what exactly the other is talking about. However, we don’t need to throw the baby out with the bath water! Therefore, I suggest precise definitions of dominant behavior and the terms we need to understand: what it is, what it is not, how it evolved and how it functions.

Dominant behavior is a quantitative and quantifiable behavior displayed by an individual with the function of gaining or maintaining temporary access to a particular resource on a particular occasion, versus a particular opponent, without either party incurring injury. If any of the parties incur injury, then the behavior is aggressive and not dominant. Its quantitative characteristics range from slightly self-confident to overtly assertive.

Dominant behavior is situational, individual and resource related. One individual displaying dominant behavior in one specific situation does not necessarily show it on another occasion toward another individual, or toward the same individual in another situation.

Resources are what an organism considers to be life necessities, e.g. food, mating partner, or a patch of territory. The perception of what an animal may consider a resource is species as well as individual related.

Aggressiveness (aggressive behavior) is behavior directed toward the elimination of competition while dominance, or social-aggressiveness, is behavior directed toward the elimination of competition from a mate.

Mates are two or more animals that live closely together and depend on one another for survival. Aliens are two or more animals that do not live closely together and do not depend on one another for survival.

Dominant behavior is particularly important for social animals that need to cohabit and cooperate to survive. Therefore, a particular social strategy evolved with the function of dealing with competition among mates, whilst conferring greatest benefit at the least cost.

Animals show dominant behavior with various signals: visual, auditory, olfactory and/or tactile.

While fearfulness (fearful behavior) is behavior directed toward the elimination of an incoming threat, submissive, or social-fearfulness, is behavior directed toward the elimination of a social-threat from a mate, i.e. losing temporary access to a resource without incurring injury.

threat is a stimulus that most often precedes a behavior that may harm, inflict pain or injury, or decrease an individual’s chance of survival. A social-threat is a threat (a threatening behavior) from another individual or group of individuals that may cause submissive behavior or flight resulting in the temporary loss of a resource, but not injury.

Animals show submissive behavior by means of various signals: visual, auditory, olfactory and/or tactile.

Persistent dominant or submissive behavior from the same individuals toward the same other individuals may or may not result in a temporary hierarchy of a certain configuration, depending on species, social organization and environmental circumstances. In stable groups confined to a defined territory, temporary hierarchies will develop more readily. In unstable groups under changing environmental conditions or, in undefined or non-established territories, hierarchies will not develop. Hierarchies, or rather the strategies involved, are Evolutionarily Stable Strategies (ESS), which are always slightly unstable, swinging forth and back around an optimal value, depending on the number of individuals in the group and the strategy each individual adopts at any given time. Hierarchies are not necessarily linear, although in small groups and over time, non-linear hierarchies seem to have a tendency to become more linear.

Some individuals have a stronger tendency to show dominant behavior and others to show submissive behavior. This may depend on their genetic makeupearly learningexperience, etc. There is no one single factor to determine this, rather a complex mixture. Let us call this a natural tendency; this is not to say it is not modifiable. It is a fact that some individuals can be more assertive than others, while others can be less so. Neither is ‘good’ or bad’ in a moral sense, simply more or less advantageous, depending on context. In one-to-one encounters, all things being equal, individuals are more likely to adopt the strategy they feel most comfortable with, hence maintaining their history of mostly displaying dominant behavior or mostly displaying submissive behavior.

When in a larger group, they will have the same tendency to play the roles they feel most comfortable with. However, this may change due to the accidental makeup of the group. Imagine a group with a large proportion of individuals that are prone to showing submissive rather than dominant behavior, and with only a few members showing the opposite tendency. In this scenario, an individual with a tendency to mostly show submissive behavior would have more chance of gaining access to resources by showing more dominant behavior and being successful. Success breeds success and, progressively, this individual with a tendency to display submissive behavior finds itself more frequently opting for a dominant strategy. If the scenario gives rise to an individual changing its preferred strategy, then others will also have the same opportunities. The number of individuals showing dominant behavior will increase, but only to a certain point, as the group cannot sustain too large a number of individuals adopting a dominant strategy. In order to avoid the risk of injury, it will eventually be more advantageous to adopt a submissive strategy, depending on benefits and costs.

Therefore, the number of dominant and submissive individuals in a group (which means individuals adopting one or other strategy as their preference) depends, not only on the natural tendency of the individuals, but also on the proportions of behavioral strategies within the group. Whether it pays off to play a dominant or a submissive role is ultimately a function of benefits and costs and the number of individuals that adopt one particular strategy.

Understanding the relationship between dominant and submissive behavior as an ESS (Evolutionarily Stable Strategy) opens up exciting perspectives and could help us to explain the behavior adopted by any given individual, at any given time. An individual will learn to display submissive behavior toward the more dominantly acting ones, and display dominant behavior toward the more submissively acting. This means that no individual always behaves dominantly or submissively as a principle, instead it all depends on the opponent and, of course, the value of the potential benefits and estimated costs.

As a corollary, hierarchies (when they exist) will always be slightly unstable depending on the strategies adopted by the individuals in the group; and they won’t be linear except in small groups or sub-groups.

In the opinion of this author, the mistake we have committed hitherto has been to regard dominance and submission as more or less static. We haven’t taken into account that these behavioral characteristics, like phenotypes and all other traits, are constantly under the scrutiny and pressure of natural selection. They are adaptive, highly variable, and highly quantitative and quantifiable.

As such, dominance and submission are dynamic features depending on different variables, a view which is compatible with the development of behavior at individual level, genetic functions, the influence of learning and, not least, evolutionary theory.

Dominance and submission are beautiful mechanisms from an evolutionary point of view. They enable (social) animals to live together and survive until they reproduce and pass their (dominant and submissive behavior) genes to the next generation. Without these mechanisms, we wouldn’t have social animals such as humans, chimpanzees, wolves and dogs to name just a few.

If an animal resolves all inter-group conflicts with aggressive and fearful behavior, it will be exhausted when subsequently compelled to go and find food, a mating partner or a safe place to rest or take care of its progeny (all of which decrease the chances of its own survival and that of its genes). Thus, the alien and mate strategy originated and evolved. It is impossible to fight everybody all of the time, so a mate is confronted using energy-saving procedures.

Submissive and dominant behavior also control population density, since they rely on individual recognition. The number of individuals an animal is capable of recognizing must have a limit. If this number exceeds a certain level, it makes recognition inefficient and impedes the alien/mate strategy; fearful/aggressive displays then replace submissive/dominant behavior.

The strategy of submission is sound. Instead of vainly engaging in a desperate fight, waiting may prove more rewarding. By employing pacifying and submissive behavior, subordinates are often able to shadow dominantly behaving animals and profit from opportunities to gain access to vital resources. By showing submissive behavior, they retain membership of the group, which also confers several advantages—particularly defense against rivals.

Hierarchies work because a subordinate will often move away, showing typical pacifying behavior, without any obvious signs of fear. Thus, the higher ranking animal may simply displace a lower ranking when feeding or at a desirable site. Hierarchies in nature are often very subtle, being difficult for an observer to decipher. The reason for this subtlety is the raison d’être of the dominance-submission strategy itself: the lower ranking animal (adopting the submissive strategy) generally avoids encounters and the higher ranking (adopting the dominance strategy) is not too keen on running into skirmishes either.

Fighting involves a certain amount of risk and can lead to serious injury, or even death. Evolution, therefore, shows a tendency towards favoring and developing mechanisms, which restrain the intensity of aggressive behavior. Most species have clear signals that show acceptance of defeat and end combat before injury occurs.

Sign-stimuli are the stimuli that produce an instinctive behavior sequence. To recognize sign-stimuli is life saving for the infant immediately after birth. Compromise is the most relevant lesson a social youngster may learn after the fundamental life saving sign-stimuli. It maintains the fitness of a group. Natural selection has proved this, favoring the individuals that develop behavior that enables them to stay together—when they need to stay together for better survival. Other animals, e.g. the solitary predators, do not need such social traits. These organisms found other ways of dealing with the maintenance of their metabolism and reproduction.

Learning to be social means learning to compromise. Social animals spend vast amounts of time together and conflicts are inevitable. It is crucial for them to develop mechanisms to deal with hostilities. Limiting aggressive and fearful behavior by means of inhibition and ritualization is only partially safe. The more social the animal is, the more efficient its mechanisms for avoiding injury need to be. Inhibited aggression is still aggression; it is playing with fire on a windy day. It works well for less social or less potentially aggressive animals, but highly social and potentially aggressive animals need other mechanisms.

In the long run, it would be too dangerous and too exhausting to constantly resort to aggression and fear to solve banal problems. Animals begin to show signs of pathological stress when under constant threat or when constantly needing to attack others. This suggests that social predators need mechanisms other than aggressiveness and fear to solve social animosities. It is my suggestion that social animals, through the ontogeny of aggressiveness and fear, developed two other equally important social behaviors. If the meaning of aggression is “go away, drop dead, never bother me again,” the meaning of social-aggression is “go away, but not too far, or for too long.” Equally, social-fear says “I won’t bother you if you don’t hurt me,” while existential-fear does not allow any compromise—“It’s either you or me.”

The significant difference between the two types of aggressive behavior seems to be the function. Aggressiveness deals with the alien and social-aggressiveness with the mate. Conversely, fear and social-fear deal with alien and mate. These are qualitative differences that justify the creation of new terms, hence dominance and submission.

What does this mean for our understanding of our dogs and our relationship with them?

This means that we all show dominant (self-confident, assertive, firm, forceful) behavior as well as submissive (insecure, accepting, consenting, yielding) behavior depending on many factors, e.g. state of mind, social position, resources, health status, opponent—humans as well as dogs (and wolves of course). There’s nothing wrong with it, except when we show dominant behavior where it would be more advantageous to show submissive behavior and vice versa. Sometimes we may act more dominantly or submissively and other times, less so. These are highly quantitative and quantifiable behaviors, with many variables. There is not one single, correct strategy. It all depends on flexibility and the strategy adopted by others.

Stable and profitable relationships are not built in the long run by means of a series of dominant and submissive displays. Instead, these behaviors are necessary for resolving inevitable social conflict. Both humans and dogs (and wolves, of course) build relationships on the need for partnership in overcoming common problems related to surviving and, preferably, achieving an acceptable level of comfort. Relationships are not built on hierarchies, but they do exist and they do play an important role in certain circumstances—for humans as well as dogs (and wolves of course)—sometimes more, sometimes less and sometimes not at all.

We build our particular (good) relationship with our dogs on partnership. We need them because they give us a sense of accomplishment that we don’t seem to get anywhere else. They need us because the world is overpopulated, resources are limited and an owner provides food, protection, healthcare, a safe place and companionship (they are social animals). It’s too hard to be a little dog all alone out there in the big world! Sometimes, in this relationship, one of the parties recurs to dominant or submissive behavior and there’s nothing wrong with that, as long as they do not both show the same behavior at the same time. If both recur to the same behavior, they have a problem: they either run into a conflict that they will usually solve without injury (the beauty of the mechanisms of dominance and submission), or one of them will have to get his act together and find the bearings for both.

A good relationship with our dogs does not involve any mysterious mechanisms. It’s basically the same as with all good relationships, whilst taking into account the particular characteristics of the species and individuals involved. We don’t need any new terms. We don’t need any new theories to explain it. We aren’t, after all, that special, nor are our dogs. We are all built from the same concept and with the same basic ingredients. All we need are good definitions and a less emotional, more rational approach. Use your heart to enjoy your dog (and life), and your reason to explain it (if you need to), not the other way around. If you don’t like my definitions, feel free to come up with others which are better (with more advantages and less disadvantages), but don’t waste your time (or anyone else’s) on meaningless discussions and knee-jerk reactions. Life is precious and every moment wasted is one less bite of a cake that you’ve devoured without even realizing it.

That’s how I see it and it looks beautiful to me—enjoy your cake!


Related articles


  • Abrantes, R. 1997. The Evolution of Canine Social Behavior. Wakan Tanka Publishers.
  • Coppinger, R. and Coppinger, L. 2001. Dogs: a Startling New Understanding of Canine Origin, Behavior and Evolution. Scribner.
  • Creel, S., and Creel, N. M. 1996. Rank and reproduction in cooperatively breeding African wild dogs: behavioral and endocrine correlates. Behav. Ecol. 8:298-306.
  • Darwin, C. 1872. The Expressions of the Emotions in Man and Animals. John Murray (the original edition).
  • Estes, R. D., and Goddard, J. 1967. Prey selection and hunting behavior of the African wild dog. J. Wildl. Manage. 31:52-70.
  • Eaton, B. 2011. Dominance in Dogs—Fact or Fiction? Dogwise Publishing.
  • Fentress, J. C., Ryon, J., McLeod, P. J., and Havkin, G. Z. 1987. A multi- dimensional approach to agonistic behavior in wolves. In Man and wolf: advances, issues, and problems in captive wolf research. Edited by H. Frank. Dr. W. Junk Publishers, Boston.
  • Fox, M. W. 1971. Socio-ecological implications of individual differences in wolf litters: a developmental and evolutionary perspective. Behaviour, 41:298-313.
  • Fox, M. 1972. Behaviour of Wolves, Dogs, and Related Canids. Harper and Row.
  • Lockwood, R. 1979. Dominance in wolves–useful construct or bad habit. In Symposium on the Behavior and Ecology of Wolves. Edited by E. Klinghammer.
  • Lopez, Barry H. (1978). Of Wolves and Men. J. M. Dent and Sons Limited.
  • Mech, L. D. 1970. The wolf: the ecology and behavior of an endangered species. Doubleday Publishing Co., New York.
  • Mech, L. David (1981). The Wolf: The Ecology and Behaviour of an Endangered Species. University of Minnesota Press.
  • Mech, L. D. 1988. The arctic wolf: living with the pack. Voyageur Press, Stillwater, Minn.
  • Mech, L. D., Adams, L. G., Meier, T. J., Burch, J. W., and Dale, B. W. 1998. The wolves of Denali. University of Minnesota Press, Minneapolis.
  • Mech, L. David. 2000. Alpha status, dominance, and division of labor in wolf packs. Northern Prairie Wildlife Research Center,
  • Mech. L. D. and Boitani, L. 2003. Wolves: Behavior, Ecology, and Conservation. University of Chicago Press.
  • Packard, J. M., Mech, L. D., and Ream, R. R. 1992. Weaning in an arctic wolf pack: behavioral mechanisms. Can. J. Zool. 70:1269-1275.
  • O’Heare, J. 2003. Dominance Theory and Dogs. DogPsych Publishing.
  • Rothman, R. J., and Mech, L. D. 1979. Scent-marking in lone wolves and newly formed pairs. Anim. Behav. 27:750-760.
  • Schenkel, R. 1947. Expression studies of wolves. Behaviour, 1:81-129. 
  • Scott, J. P. and Fuller, J. L. 1998. Genetics and the Social Behavior of the Dog. University of Chicago Press.
  • Van Hooff, J.A.R.A.M., and Wensing, J.A.B. 1987. Dominance and its behavioral measures in a captive wolf pack. In Man and wolf: advances, issues, and problems in captive wolf research. Edited by H. Frank. Dr. W. Junk Publishers, Boston.
  • Wilson, E. O. 1975. Sociobiology. Belknap Press of Harvard University Press, Cambridge, Mass.
  • Zimen, E. 1975. Social dynamics of the wolf pack. In The wild canids: their systematics, behavioral ecology and evolution. Edited by M. W. Fox. Van Nostrand Reinhold Co., New York. pp. 336-368.
  • Zimen, E. 1976. On the regulation of pack size in wolves. Z. Tierpsychol. 40:300-341.
  • Zimen, Erik (1981). The Wolf: His Place in the Natural World. Souvenir Press.
  • Zimen, E. 1982. A wolf pack sociogram. In Wolves of the world. Edited by F. H. Harrington, and P. C. Paquet. Noyes Publishers, Park Ridge, NJ.
Thanks to Simon Gadbois (merci), Tilde Detz (tak), Victor Ros (gracias), Sue McCabe (go raibh math agate), Parichart Abrantes (ขอบคุณครับ) and Anna Holloway (thank you) for their suggestions to improve this article. The remaining flaws are mine, not theirs.