Cue, Not Command: Rethinking the Language of Dog Training
Why cooperation, communication, and choice matter more than obedience.
There is a small language change I have come to learn and care a lot about in dog training: I don’t give dogs commands. I give them cues. At first glance, that might sound like semantics. Sit is still sit, right?
But the distinction represents something much larger: what we believe training is for, what we believe our relationship with dogs should look like, and what we think is happening when a dog doesn’t do what we’ve asked.
Where the “alpha” idea came from
For decades, our understanding of dogs was heavily influenced by an older understanding of wolves.
In 1947, animal behaviorist Rudolf Schenkel published observations of captive wolves living together in Switzerland. These were not natural family packs roaming freely in the wild. They were wolves living together under artificial conditions, where competition and social tension produced visible dominance relationships. The work was valuable observation for its time — but the environment mattered enormously.
The idea of the dominant “alpha wolf” was later popularized widely, including through wolf biologist L. David Mech’s influential 1970 book. And then we upgraded our observations. After years spent observing wolves living naturally, Mech himself became one of the scientists arguing that the familiar alpha model was misleading when applied to wild wolf packs. Most natural packs aren’t collections of unrelated adults fighting their way up a hierarchy. They’re families — typically parents and their offspring. In thirteen summers observing one free-living pack, Mech reported seeing no dominance contests among the breeding adults and their offspring of the kind the popular “alpha” narrative suggests.
That distinction matters, because ideas about wolf dominance didn’t stay with wolves, they followed us home to our dogs.
From leadership to control
For a long time, dog training frequently centered around establishing control: the human gives a command; the dog complies. And in coercive versions of that model, there is an implied second half to the sentence:
Do this — or else.
The “or else” doesn’t have to mean extreme physical punishment. It might be a leash correction, pressure from a collar, an unpleasant sound, physical manipulation, intimidation, or another consequence the dog learns to avoid. The behavior can absolutely occur. That doesn’t necessarily mean the dog understands the situation the way we think they do. It may simply mean they have learned which behavior makes the unpleasant thing stop — or prevents it from happening in the first place.
Contemporary research comparing training styles has found greater stress-related behavior and, in some studies, higher post-training cortisol among dogs trained with a greater proportion of aversive methods. The American Veterinary Society of Animal Behavior currently recommends reward-based methods for dog training and behavior modification.
But there is another way to have the conversation.
A cue gives information
A cue isn’t a demand for submission. It’s information. When I say “sit,” I’m communicating something the dog has learned:
“This is the behavior I’m asking for right now.”
And through reinforcement history, the dog has also learned:
“When I hear this cue and offer this behavior, good things tend to happen.”
That creates a fundamentally different interaction.
The dog still makes a choice. Yes — even when we have trained the behavior beautifully. And I think that choice matters. Because when a dog chooses not to respond to a cue, instead of immediately asking, “Why isn’t this dog obeying me?” we can ask better questions.
Does the dog actually understand the cue in this environment?
Is the distraction greater than anything we’ve practiced around?
Is the dog frightened?
Overstimulated?
Tired?
In pain?
Is what I’m offering worth more to this dog right now than whatever has their attention?
Have I practiced this behavior enough for them to understand what I’m asking?
Suddenly, “disobedience” becomes information. And information gives us somewhere productive to go.
Choice doesn’t mean chaos
Giving a dog agency does not mean letting them do whatever they want. Dogs still need boundaries. They still need consistency. Sometimes humans have to make decisions for them — especially when safety is involved. The difference is that I don’t need my dog to believe I am stronger, scarier, or higher-ranking in order for me to provide structure.
I need to be clear, consistent and trustworthy. If a dog doesn’t respond to a cue during training, perhaps they don’t receive the reward associated with that behavior. We reset. We make the task easier. We reduce the distraction. We reinforce more effectively. We figure out what part of the communication broke down.
The goal isn’t:
How do I make this dog comply? It’s: How do I make it possible for this dog to succeed?
That may sound like a subtle change. In practice, it changes almost everything.
Training is a conversation
Dogs spend an extraordinary amount of their lives learning us. Our routines. Our facial expressions. Our tone of voice. The sound of the treat cabinet opening.
Which shoes mean we’re leaving the house. Which tiny movement toward the leash means a walk might be coming. They are constantly gathering information from us. Training gives us the opportunity to return the favor. To observe. To communicate clearly. To notice when something isn’t working. To build behaviors without making the relationship itself something the dog has to navigate carefully.
That’s why I prefer cue over command.
Not because one vocabulary word magically makes someone a kinder trainer. But because the word reminds me what I want the interaction to be: Not an order handed down a hierarchy. A piece of information passed between two species that have spent thousands of years learning how to live beside one another.
I ask. You understand. You choose. I listen to the answer. And then we keep learning each other.
Sources & further reading
L. David Mech, Alpha Status, Dominance, and Division of Labor in Wolf Packs — a discussion of why observations of unrelated captive wolves should not be assumed to represent the social structure of natural wolf packs.
Rudolf Schenkel’s 1947 captive-wolf work and its role in the development of the traditional “alpha wolf” concept are summarized by wolf researcher L. David Mech.
Vieira de Castro et al. (2020), Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare.
American Veterinary Society of Animal Behavior, Humane Dog Training Position Statement.