what's new
what's new

the "D" word

("dominance"is not a dirty word)

The term "dominance" has become something of a "dirty word" or at least a controversial concept amoung dog trainers and dog writers. That's really unfortunate, because social status and hierarchy are a fact of canine nature (and also human nature).


the "D" word

understanding and misunderstanding the concept of "dominance"

by Pam Green, © 2019

(note : because this article is about the concept of "dominance" it seems to focus on the RESPECT that is paid to the higher ranked individual. Please keep aware that to be a good LEADER the high ranked individuals , whether dog or human, must be TRUSTED as well as RESPECTED. )


The term "dominance" has become something of a "dirty word" or at least a controversial concept amoung dog trainers and dog writers. That's really unfortunate, because social status and hierarchy are a fact of canine nature (and also human nature).

Basically all species which live in groups need to have a hierarchy of rank. They need a dominance order. This may be very rigid or it may be more flexible. The more intelligent the species, the more likely that there is some flexibility in social order. Species which are socially cooperative need this even more. The socially cooperative hunters or socially cooperative breeders need it even more. Humans and dogs are very much socially cooperative in hunting (and for humans inother food production) and are cooperative in rearing their young. Members of both of these two species also cooperate with members of the other one in these regards. Both species are highly intelligent. (It's hard to say whether the increased complexity of social interactions was the driving force for evolution ofgreater intelligence or whether greater intelligence allowed evolution of greater social complexity. I think that both processes influenced each other in an upwardly spiraling fashion.)

Thus the main point is still valid : dogs are genetically hard-wired to rely on a clear hierarchy of rank and privilege to maintain social harmony within the living group. The presence of an established hierarchy with humans at the top makes dogs more contented and peaceful, and prevents or reduces the potential for serious behavior problems, especially problems involving biting humans.

what is dominance and what it is not

"Dominance" has become a dirty word because some people misunderstand the concept and think that "dominance" means "domination" and is to be obtained by domineering behavior, by bullying and by physical force. That is simply NOT true. Those who try to frighten dogs into submission generally are "sowing the storm" and will "reap the whirlwind" of a dog who becomes a fear-inspired biter. Attempts of humans to dominate by use of the "Alpha roll-over" and similar techniqes are particularly dangerous (as will be explained below.)

the ethologist's definition

(taken from Patricia McConnell's "The Other End of the Leash")

Dominance means that the higher ranked individual has the "right to preferred access to scarce and valued resourses". That's the definition most ethologists would use. "The right to preferred access" means that this individual can choose to claim that resourse. The individual might or might not choose to do so. "Scarce" means that there is not enough of that particular resource for both the dominant individual to have it and for the other or others to also have it. Think of the difference between a plate full of chocolate versus only one piece of chocolate. "Valued" means that the dominant individual and the other individual bothreally want to have this resource right now. Think of the difference between wanting food when you are already over-stuffed full versus when you are hungry. Or the difference between a food you really love and thus want versus one you don't care about or actually dislike.

why (adult) humans can easily rank at the top

For many of the resources that dogs value and cannot obtain on their own , we (adult) humans not merely have "preferred access" but actually have ownership and control. We can give or withhold these valued resources to our dogs. We bring home the dog food (and treats), store it out of dog-reach, give it to the dog when we choose to do so, either as treats, training rewards, or as a meal. We give or withhold our caresses. We have the power to turn on the water tap and pour it into a bowl and provide it to the dog.(Note : normally we ensure that there is always water available, as that is a health requirement; but still our dogs do see us fill the bowl, and most dogs will try to draw our attention to a bowl that has gone dry in order to get us to refill it.) We hold the car keys and can give or withhold the delights of a car ride. We open the door or gate to the big wide world and grant or withhold the delights of a walk for the dog (usually first requiring the dog to come and get leashed.) We provide a lot of toys and we are wonderful playmates to the dog, which may include some games that another dog couldn't provide, such as Agility coursing. For herding heritaged dogs, we take them out to herd livestock and we make ourselves the leader of the dog-human livestock control team..

That is the basis of programs like "Nothing in Life is Free" (really "not much in life is free") and "Learn to Earn", programs in which the dog is asked (required) to respond to a cue in order to get the human to deliver some desired resource. These techniques drive home to the dog the fact that the human is truly high status, dominant, is both "She from whom all blessings flow" and "She who must be obeyed". The human who is a benevolent leader will also make clear that she is "She who can be trusted and relied upon".

(note : I emphasize "adult" human because younger children are usually viewed very indulgently by well socialized dogs who have had lots of experience with nice dog-savvy kids. (I've seen dogs "obeying" a child with an expression on the dog's face that says of the child "isn't he cute".) The child is not viewed as a higher ranked individual. Many behaviorists say that the dog views the young child as a "puppy" or as a "littermate". The child is trusted and viewed with affection, but not with a lot of the respect that would be given to a higher ranked individual. Now an older child, especially one who engages in training the dog, very much can become a higher ranked leader. Some children become better trainers and leaders than many adults.)

observing dominance interactions

Behaviorist Patricial McConnell urges dog owners to "be Jane Goodall in your living room" (and in your backyard). Be truely observant as you watch dogs interacting and watch yourself interacting with your dogs. If you do so, you will become aware of many instances, some obvious and some extremely subtle, in which dominance is in action.

Everyone who lives with two or more dogs has seen instances where one dog claimed something that another dog also would have wanted. This can be very obvious in the case of their person returning home and all the dogs want to greet her, but it's usually one particular dog who gets greeted first. Or one dog who is first in line to get petted while the person is on the telephone (or sitting on "the oval office")

Likewise everyone has seen instances where only one dog wanted something and the other did not. Perhaps it's a usually subordinate dog who is the one who likes to chase balls and the usually dominant one couldn't care less when a ball is thrown.

Everyone has seen instances where there was plenty for everyone, so rank didn't matter. For instance my dogs returning from a walk don't really care who drinks first from a water bowl because there are several bowls.

Of course there are plenty of situations where the person prevents any issue of rank from being active. I habitutally phsically separate dogs (crates, stretch gates, doors) before dishing out their meals. (That both avoids any chance for one to bully or steal food, and it ensures that I will know if a dog doesn't finish the meal thus displaying symptom of illness.)

Likewise the wise human leader will sometimes or often interrupt a situation in which rank is at issue. The human will assert her own rank as the "uber-alpha", ranking higher than any of the dogs. Often all that is needed is a sharp gutteral word (eg "aach !" or "eh eh !") plus a sharp eye contact. Sometimes dogs need to be given a "time out" by being sent to their crates or other station.

When dogs who don't already know one another well, the rank testing behavior is fairly obvious. Most humans can easily learn to see and understand these behaviors. The paw over the other's shoulder is one of the classics. Fights (ie genuine ones, not just "play fights") are relatively rare and invitations to play, via the classic "play bow" , the more common outcome.

When two or more dogs live together for an extended period, thus know each other well, the instances of one dog asserting dominant status and the other yielding are usually pretty subtle. Only a subtle reminder is needed, and often even that is unnescessary. One dog will claim a particular place to lie and rest and others simply seek other places. One dog is standing or lying somewhere and others make a bit of a detour around that one or hesitate to cross that one's path. One dog will give eye contact and the other averts his gaze, looks away. Humans who are not observant or are distracted will usually miss these events.

Note : when two dogs are playing together, there can be frequent "role switching" as to which dog is playing the dominant role and which the subordinate role. Thus in play-fighting games, sometimes one is on top and sometimes the other. Indeed a confidently dominant dog will very readily "self-handicap" in order to allow a lower ranked playmate to enjoy the game. A physically stronger or simply larger dog will self-handicap to play with a weaker or smaller one. This is a very endearing sight to behold. Likewise in chasing games (one of the absolutely favorite dog games, and one where humans are not good playmates because we are so slow), the roles of chaser and chased get switched back and forth.

Note also that often when dogs play wrestling games or tug of war, there is some growling by one or both. That is just "play growling" and it's absolutely nothing to do with aggression or dominance. Some dogs also play growl when playing tug with a human.

"dominance" as a diagnosis ?

Sometimes an owner who considers the dog's behavior to be a problem will be told that the underlying problem is that the dog is "dominant". This is especially likely if the problem involves the dog behaving in a manner that seems to the owner to be "aggressive". This diagnosis often results in the owner being advised (or deciding on his own) to act forcefully towards the dog, including tactics like the "alpha roll-over" . Wiser advice would be to begin some form of "Nothing in Life is Free".

A lot of behavior that is labled "aggressive" is FEAR-based. . This is self-defensive aggression. If that's the case, then acting in a physically forceful manner towards the dog will simply make him more and more afraid of the person doing it and thus will make the dog more and more self-defensive. The dog will be more and more sure that he must save himself by either Plan A = "avoidance" = fleeing or by Plan B = "bite" = fighting. This can escalate into a very dangerous situation.

(By the way , in my opinion the case of "food aggression" is really a form of fear-based self-defense. The dog is afraid that the person is about to take his food away, food his genes scream he needs to survive , so of course he defends his posession of that food. Since fleeing would mean abandoning the food, the usual tactic is to threaten to bite and to bite if it seems necessary.)

The highly confident dog who is acting in an assertive offensive manner IS indeed quite likely acting on the belief that he outranks the human involved. In this case the human is well advised to seriously practice "Nothing in Life is Free".

why "alpha roll-over" is a really really BAD idea

If you try to "alpha roll" a dog whose is fearful, you will make him that much more afraid of you. If you try to "alpha roll" a dog who is confident that he outranks you, you will probably get bitten.

The biggest reason NOT to try an "alpha roll" is simply that your timing of the release is very unlikley to be correct. Dog doing this to another dog will usually get the timing of release perfect. The release is the reward for the other dog yielding. If you release too soon, while the dog is still in physical or mental opposition, you are rewarding opposition and thus sending the message that resistance is useful, a winning tactic. If you release too late, after the dog has mentally yielded, you fail to reward yielding and send the message that yielding is futile.

Some behaviorists say that dogs don't actually do "rolls" on each other. That what looks like a "roll" is really the subordinate dog preventing any assertion (or responding to subtle assertion) by the dominant dog by flinging himself to the ground. I'm fairly sure that I have seen dogs do "rolls" on another dog : during my Bones yearling year, Chelsea and her brother Smokey pinned him down frequently. It's certainly true that plenty of dogs will voluntarily go belly up to a human, and that may be submission or it may be a request or a demand for a belly rub. (My FoxTheWicked does this a lot.)

are some dogs naturally "dominant" ?

This can be a "diagnosis" that implies that the dog cannot be changed. That can lead the human to give up on trying to solve problems. If the dog gets dumped at the Pound with the statement "he's too dominant" or , worse, "he's aggressive", that can be a fatal condemnation.

In my opinion , dogs who are naturally very self-confident or have learned to be very self-confident and who are naturally assertive or have learned that assertion is usually a good tactic are very likely to become dominant in a group where there is no other dog who is even more this way. A highly intelligent and "thoughtful" (the opposite of "impulsive") dog who is confident and assertive is almost certain to become the leader dog and to be a benevolent leader. (My adored Chelsea and Bones were this type.) A dog who is especially skilled at using his "native language" of social gestures probably also has an advantage. (My precious Chris was this type).

On the other paw, most dogs are NOT all that highly self confident and self assertive, thus will cheerfully let another dog who is more self-confident and /or more self-assertive be dominant over themselves. Some dogs (my Velvet is one of them) are very timid and would never think of asserting themselves, thus they become the lowest ranked member of the group.

Of course most dogs fall somewhere in the middle. But if there is no dog whose character naturally makes that dog the highest ranked, and if no human takes on the leader role, then the intermediate dog may have to take on that role but will be insecure in that role. Uneasy lies the head that unwillingly wears the crown. Insecurity can lead to problems for all concerned. In such situations, the human can easily use "Nothing in Life Is Free" type of protocol to become the leader.

Or as Shakespeare might have said, "some dogs are born dominant, some become dominant , and some have the dominant role thrust upon them".

By the way, I really love highly confident , assertive, highly intelligent, and thoughtful dogs. They are wonderful companions (for the human who also has these characters) and they can be fabulous working dogs or working competition dogs when the job is one for which they have inherited the relevant talents (eg herding instinct). My adored Chelsea and Bones were absolute paragons of this type.

are wolves a relevant model ?

Some behavior writers point out that a lot of the really dramatic displays of status behavior in wolves was observed in unnatural packs. Packs of unrelated wolves put together, often in adulthood , usually in an enclosed area (which may be large or small). Thus a hierarchy can only be established with a lot of displays, often dramatic, and with physical testing (ie fights of limited intensity). A natural wolf pack is really an extended family, with the pack parents being the alpha couple, and the rest being that couple's siblings or offspring or grand-kids. Thus everyone grows up into their social place and accepts that position.

But most multi-dog households get formed because the humans bring unrelated dogs together, either all being already adults or some being added as puppies. Such a group is much like the unnatural wolf pack. In spite of this, many multi-dog households get along well. (My own dogs have usually been very willing to accept my foster dogs into the group without major difficulties.) The nearest thing to the extended family wolf pack would be found in dog-breeders' homes, where several generations may accumulate. However while in the wolf pack only the Alpha Couple (highest ranked male and highest ranked female) get to produce a litter, in a breeder's home there may be more than one of each sex that gets to breed, and this can be a cause of dissention.

The other significant aspect is that wolves have to take life more seriously. They have to work at survival. Our companion dogs don't have to take life seriously and don't genrally have to work at surviving. Additionally over many generations many breeds of dogs have been genetically selected for "behavioral neotency", ie to retain juvenile behavior. (Some behaviorists refer to this as being "Peter Pan wolves", ie never really becoming adults.) While hierarchy begins to form between littermate puppies, status really doesn't become a serious affair until they reach the age of "social maturity" as young adults. So it's just not as serious for dogs as it is for wolves.

A lot of dog behavior is quite obviously similar to a lot of wolf behavior. No surprise, considering common ancestry. But there are some differences. Wolves are worth studying in their own right and also to see what dog behavior has evolved from.

Alexnder Pope might have said "Observe the wolf but don't get lost in fog : the very best model of the dog is dog."

then give me another word for it

The term "pack leader" has become almost as controversial and multi-meaninged and subject to mis-use as the term "dominance". Really the idea of leadership ought to be straight-foreward, but in the context of humans leading humans it is anything but reasonable.

Some prefer to think of themselves as their dog's "parent". That image works well for anyone whose parents were wise and benevolent. Not so much for those whose parents were abusive or merely tyranical. The same goes for the terms "teacher" and "coach". So many good ones but also some dreadful ones.

"Guardian" has become controversial. I was using the term long before it became politicised, because to me it gave the meaning of benevolence and wise leadership. The term "protector" might work well. It certainly works for me because it means that I have the benevolent responsibility and concomittant authority to take care of the welfare and happyness of my dogs.

The essential ingredience of human-dog relationship is that the dog both TRUST and RESPECT you. And you must trust the dog and respect his nature and needs. Also you should ENJOY one another's company and have at least a few activities you enjoy together. (That also sounds like a good description of human to human friendship, as well as essential to human marriage, with marriage having "benefits added" and financial partnership and potentially (but not inevitably) reproductive partnership.)



a footnote about "equality" and why humans are conflicted about dominance and hierarchy.

Something that complicates our human attitudes towards the need for hierarchy among dogs and between dogs and humans is that we humans , at least in the USA, proclaim a belief in the equality of all humans. Over two centuries ago, our forefathers proclaimed that "all men are created equal". Of course "men" didn't include female men, only male men, and it didn't include all males, just free white males. Our current Constitution still doesn't include women as equal, though it does include former male slaves. Two years ago today, on 1/21/2017, we inaugurated a POTUS elected by less than a majority of voters who gained election by fomenting racial and sexual bigotry and bigotry against immigrants.

Meanwhile many of our human institutions and social clubs are overt with hierarchy. The military of course as practical necessity must have ranks and chain of command. But Boy Scouts also have ranks, much respected. Social lodges such as Masons, Moose, etc have their degrees. Our universities certainly have ranks, from lowly student to Full Professor and Department head, and the infighting among academics is legendary.

In many practical situations we have a clear agreement as to who is the leader of a particular group doing a particular activity. That leadership can be based on the leader's superior knowledge or competence at the activity. If so, the role of leader may shift as the needs of the activity shift. My favorite example is a group going for a hike or camping trip. The one who knows the territory and the abilities of participants is the general leader. But if someone gets injured or ill, the leadership shifts without question to whomever has medical knowledge and skill.

(I've wondered whether to some extent dogs also consider knowledge and skill as ingredients of leadership. I'd bet wolves setting out on a hunt know very well the abilities of each pack member. Is the leader the one who best can deploy hunt members and coordinate their work ?)

Our human views about hierarchy vs equality are conflcted or confused at best. No wonder some of us resist the idea of canine hierarchy as well as the notion that humans naturally tend to form hierarchies. Dogs, however, have no doubts about the need for well defined social rank.

return to top of page

Related topics :

site author Pam Green copyright 2003
Pam's e-mail address has changed tips on site use
created 12/01/2019 revised 1/28/2020, 4/8/2021
return to Welcome page go to the Farewell page with links to other sites