My first post on this site was about dominance theory and how it doesn’t apply to our pet dogs. This post will be dedicated to rebutting the most common arguments I’ve encountered for sticking with it. Any old hand at dog training used to use these methods because that’s all there was. It can be hard to change, but behavior science has moved on to bigger and better things and we should too.
Dominance Isn’t A Dominance Hierarchy
In a dominance hierarchy there is linear social status. Your access to one resource corresponds with your access to another resource. The highest animal in the hierarchy has priority access to each and every resource. The lowest animal in the hierarchy gets the worst of everything. It’s an all or nothing system and the highest ranking member isn’t allowed to slack off. If dogs followed this model then permissiveness in one area would imply that the dog gets the same access to everything he wants.
A social species, such as dogs, can have dominant behavior, meaning that a dog will work to attain priority access to a specific resource. Because they do not form a dominance hierarchy that does not mean that they gain priority access to all resources. That also means that letting your dog on the best napping spot, such as your bed, will cause him to suddenly start trying to dominate you or other dogs in other contexts.
If you have multiple dogs you’ll be able to observe this for yourself. An established group of dogs will most often have a hierarchy of access for each resource. One dog wants a specific toy most and the other dogs in the group defer to him, but only for that item. Another dog likes to nap in a specific area and she’s allowed to boot other dogs out of her spot because they just don’t care as much as she does. Hierarchy in domestic dogs tends to be based on what’s the most important to who, and it’s very fluid. Today’s bed queen could be tomorrow’s toy hog and the social structure will adjust without too much fuss.
Another important factor is that there is zero evidence that dogs can’t tell we’re not other dogs, so even if they did form a dominance hierarchy, we wouldn’t be part of it. Dogs are somewhat unique in that they’re able to imprint on multiple species, and treat each species differently. They understand that we’re not sheep or cows or dogs and act accordingly, even if they’ve been raised equally around livestock, other dogs, and people.
Captive Wolves Form A Dominance Hierachy
L. David Mech popularized the dominance hierarchy model of wolf packs when his book The Wolf: The Ecology and Behavior of an Endangered Species was published in 1970. Most information about the social behavior of wolves at that time was obtained by studying captive wolves in incredibly unnatural packs. Not only are the composed of unrelated adults, but they’re in a relatively small area and in regular contact with humans.
Mech has since recanted his opinion, and new research studying wild wolves shows that they live in small family groups. The previous year’s litter will stay with their parents and help rear a litter of wolf pups before finding their own mate and forming their own pack, which generally consists of a mated pair and their offspring.
Because most people who own multiple dogs own unrelated individuals some owners assume that the captive wolf model should apply to their pet dogs. This is missing the point entirely. Dogs are not wolves. They don’t have the same requirements or stressors in a captive environment.
Captive wolves exist in stressful and unnatural living conditions for their species. Wolves are predators who are rarely seen by humans even when they live nearby. In order to survive, a single wolf, a pair, or a family group has to defend a large territory from other wolves to have enough access to prey and avoid starvation.
Dogs are domesticated scavengers and opportunistic hunters. Feral or wild dogs don’t kill most of what they eat, and they have a much higher population density than wild wolves, congregating around food sources, such as dumps and roadways. In places where there are still village dogs they’ve been observed defending a territory, but it tends to be small, such as a yard or a specific piece of a larger feeding area. It’s natural for them to interact daily with unrelated individuals and live in close proximity to people. The population density of the average pet dog household isn’t actually that out of line with village dogs.
Wolves are assholes to each other and form an unnatural social structure in captivity because they’re severely stressed by large numbers of unrelated wolves and proximity to people. My dogs want to lick you and cuddle with dogs they met yesterday, I don’t think they’re stressed by proximity to humans or unrelated dogs, so an unnatural hierarchy need not apply.
The Old Way Mostly Worked
No one is saying that the punishment based training that traditionally goes hand in hand with dominance theory based training doesn’t work. If it didn’t something else would have become popular a long time ago. What modern trainers are saying is that outdated techniques are very strongly correlated with fear and aggression issues, and that positive reinforcement based techniques are not. Obviously every dog trained in this way doesn’t become fearful or aggressive, but an unacceptable number will.
Since aggression is overwhelmingly fear based using punishment to inhibit aggressive behaviors will actually increase that fear. Because the underlying emotion of fear hasn’t been addressed the dog is at risk for an even more aggressive reaction if he’s pushed too far later. “Alpha rolling” your dog to teach him not to attack another dog is scary enough that he may eventually learn to inhibit his reactions around that dog, but because he’s still fearful of other dogs he may become more aggressive with other dogs or in other frightening circumstances.
The difference in body language between dogs who have been “corrected” until they “shut down” and just tolerate frightening things versus dogs who have been counter conditioned and desensitized correctly to learn a new emotional response is amazing. Relaxed happy dogs are not a bite risk. Terrified, overwhelmed dogs are.
My Dog Needs Consequences!
That’s cool, just change your preferred type of punishment and consequences will happen without all those nasty side effects. Modern dog training relies on negative punishment, in conjunction with positive reinforcement for correct behavior. Negative punishment removes a reward rather than adding something unpleasant. If I turn around and ignore him whenever my dog jumps on me for attention I’m using negative punishment. He wants attention, so I’m removing it when he’s behaving in a way that’s unacceptable. When he sits politely to ask for attention, I’ll give it to him, reinforcing the correct behavior. Eventually he’ll learn what does and does not work, and I haven’t done anything to hurt or scare him and possibly create a bigger behavioral issue for myself.
If you’re changing your reaction to an unwanted behavior that previously “worked” for your dog be ready for an “extinction burst”. If something used to work and now it doesn’t the first thing an animal will do is try to do it more. If you’re consistent and the behavior just stops being rewarding they’ll stop soon. If you reinforce on a variable schedule your dog will likely begin to do it even more consistently. This is a common reason people say negative punishment doesn’t work. They expect their dogs to realize right off that the rules have changed, but if your dog has a long history of rewarding experiences doing something it’s going to take a little time to train them out of it. Be realistic.
If you’re interested in learning more about wild dogs, wolves and their relatives and how that information applies to our pet dogs I suggest Dogs: A New Understanding Of Canine Origin, Behavior and Evolution and Dog Sense, both by biologists who have devoted much of their professional lives to understanding dogs. There are many things we can learn by studying the closest relatives of our pet dogs, but the bulk of our information should come from studying dogs themselves.