Verified by 
There is considerable disagreement among ethologists, dog behaviorists, and dog trainers about the meaning and function of dominance, both in the world of dogs themselves and also in the world of closely related canid species such as wolves. (See, for example, Marc Bekoff's recent post on this.) Some of the debate centers on the extent to which wolf behavior and dog behavior can be compared. Some of the debate centers around whether or not dogs form dominant relationships or engage in dominance behaviors.
Two different—but intimately related—questions, also related to dominance, spin off from this debate:
1) How do dogs understand their relationships with humans? Does “dominance” play any role in this relationship, from the dogs’ point of view?
2) And how do we, in turn, understand our relationship with dogs? Should “dominance” figure in how we interact with our canine companions? Do dogs understand interactions with us through the lens of “dominance” or are we the ones with the dominance-shaded glasses? Do human “dominance” behaviors harm or help dogs?
These are exceedingly important questions and the welfare of millions of dogs living with humans can be impacted by our answers.
As for the first question, how our dogs understand and relate to us, I would like only to make a few general remarks. First of all, how dogs relate to each other certainly bears on our understanding of how they might relate to us, but dog-dog relationships are fundamentally different from dog-human relationships. As dog owners, we need to keep these two kinds of relationship separate in our minds. It seems that too often in the vast database of dog training advice available to dog owners these two types of relationship get confused. (“We are their pack and need to behave like the alpha.”)
Second, the behavior of dogs is related to the behavior of other canids, but it is also entirely unique and it is tricky to try to compare among species. This sort of exercise should be left to the scientists. The nuance of comparative ethology may be lost on many dog owners, who have been told by a trainer or who have read on the Web that “dogs are just small wolves.”
It will always be a mystery to us precisely how our dogs think of us and what their relationships with us mean to them. Science can go a long way in helping us understand what dogs are like and how to interpret their behaviors, and this knowledge is essential for forming healthy relationships that respect the needs of dogs and humans alike and that result in harmonious interactions. But science isn’t the end of the story; layers of value judgment get inserted into science and we need to be clear about where science and values intersect. Indeed, trying to take the values out of science is like trying to take the cream out of coffee. You can’t do it.
Now, on to the second and perhaps even more important question—and one that scientists and behaviorists and trainers and dog owners all need to consider. How do we think of ourselves in relation to our dogs? And why is this, at base, an ethical question? In other words, how should we think of ourselves in relation to our dogs—and how is this different from how we do think of ourselves?
Consider some possible models of the human-dog relationship, and think about the implications of each for how we understand power relations, proper interactions, the moral value of dogs.
HUMAN - DOG
Master - Slave
Ruler - Subject
Warden - Prisoner
Programmer - Computer
Engineer - Robot
Partner - Partner
Parent - Child
Teacher - Student
The way we understand our relationships with dogs shapes the kind of interactions we have and the kinds of treatment that we consider acceptable. There is certainly science involved: We want to build relationships based on what dogs are actually like, not what we imagine them to be like, and especially not on misconceptions about who they are and what they need, because these misconceptions can lead to harm. But our conceptualization of how we ought to relate to dogs takes us strongly into the realm of values and subjective judgments because the model we work from will shape the kinds of interactions we believe are ethically appropriate. For example, within a master-slave model, exploitation and domination are built into the relationship. (It doesn’t even seem right to call this dyad a “relationship,” does it, since it is so one-directional?)
These models are obviously open to a great deal of variation, and it isn’t necessarily true that one is better than another. And within each model, there are dangers. Ruler-Subject implies a benevolent relationship, with the ruler using his power judiciously and for the good of his subjects. Sometimes, though, rulers are autocrats, and demand absolute obedience and submission and care very little about what their subjects need. Within the parent-child or paternalistic model, there is a danger of emotional neglect and abuse. The parent understands what the child most needs (e.g., affection) and uses these needs against the child. A truly benevolent and wise “ruler” may take better care of a dog’s needs than a neglectful and inconsiderate “parent.”
Coming back briefly to the issue of dominance, we can ask whether and how it reflects a particular model of human-dog relationships and what this might mean for dogs. Dominance fits most easily within the master-slave model, and is a natural part of it. The use of power in this relationship is essentially self-serving: you use your dominance to get what you want, often through physical and emotional violence. What you seek in your dog is absolute submission. The danger of viewing dogs as our slaves shouldn’t need explanation.
Now, it is possible—indeed likely—that we have a variety of relationships with our dogs and that one model does not and cannot adequately describe all of them. And each dog owner will be different. For example, in the realm of training (or what some prefer to call “teaching”) I tend to think of my dogs as partners or collaborators. When I teach them, I am seeking to find the most effective means of communication and my dogs and I are engaged in collaborative learning. The teacher-student model doesn’t work as well for me, in this role, because at least half of the work that needs to be done is training myself to communicate clearly, to have impeccable timing, to understand how it is that dogs learn—including, especially, the individual learning needs and quirks of my two dogs. When it comes to feeding, providing daily care like exercise and grooming, and seeing to the health of my dogs, I work more strongly from the parent-child model. I am “paternalistic” and may, at times, violate the autonomy of my dogs for their own benefit (they don’t have a choice about whether or not to see the veterinarian). I see myself as responsible for their care and safety, in the same way that I am responsible for the well-being of my daughter. Because our dogs are completely reliant on us for daily care, I think the parent-child model is appropriate in this realm. The negative side of paternalism is a kind of patronizing attitude and an inclination to violate autonomy more often than necessary.
As the conversation continues about how to understand and best relate to our dogs—and whether dominance has any role in healthy dog-human relationships—we should keep the moral elements of the discussion in the forefront.
