How You Train Your Dog Is a Moral Statement — New Research Backs It Up

Doges Editorial · 2026-05-10 · 5 min read

When you ask your dog to sit, you're making a choice. When your dog does something you don't want and you decide how to respond — with a treat, a redirection, a sharp "no," or something more physical — you're making another choice. New research from the University of Copenhagen and the University of Edinburgh spent time asking 500 American dog owners about those choices, and found something that goes deeper than training philosophy: the methods people choose to use with their dogs map consistently and specifically onto their underlying ethical views about what kind of beings animals are. How you train your dog, the study finds, is a moral statement.

The Study: 500 Owners and the Ethics Hiding in the Training Session

Published in the journal Anthrozoös and covered by EurekAlert in May 2026, the study surveyed 500 US dog owners about both their training practices and their ethical orientations toward animals. The research team — led by Professor Peter Sandøe of the University of Copenhagen, with colleagues at the University of Edinburgh — designed the study to look for correlations between what people believe about animals and how they treat them in the specific context of training. They found those correlations, and they found them to be systematic enough to warrant a significant reframing of how the dog training community thinks about method choice.

Three Ways of Thinking About Animals

The study organized participants' ethical views into three broad orientations. An anthropocentric view holds that animals exist primarily in relation to human needs and interests — they matter insofar as they matter to people, and the obligations we have toward them flow from their utility or emotional value to us. An animal welfare orientation holds that animals have their own interests — particularly the interest in not suffering — that place real constraints on what humans can ethically do to them, regardless of whether the human benefits. An animal rights orientation holds that animals have a moral status analogous to persons and should not be treated as instruments of human purposes, even when that treatment would not cause direct suffering.

Most dog owners, the research found, fall somewhere along the spectrum between anthropocentric and animal welfare orientations. Fully developed animal rights perspectives are uncommon in the general population. But the differences between the anthropocentric and welfare-oriented stances produce meaningfully distinct training behaviors — and the data on those behaviors is where the research becomes genuinely interesting for anyone who has thought carefully about their own choices with their dog.

What the Survey Numbers Reveal About Real-World Training

The numbers are instructive. Ninety-seven percent of respondents reported using praise as a training tool. Eighty-six percent use treats or toys. Those near-universal figures mean that almost everyone, regardless of their ethical orientation, incorporates reward-based methods to some degree. The divergence appears when the study looks at the other end of the spectrum: 46 percent of respondents use some form of punishment. Twenty-five percent use physically aversive methods. And fewer than 18 percent rely exclusively on reward-based approaches.

That last figure is worth sitting with. Despite decades of behavioral science research establishing that positive reinforcement produces faster learning, stronger retention, and fewer behavioral side effects than punishment-based methods, fewer than one in five dog owners trains exclusively with it. The other four-fifths incorporate some form of punishment or aversive correction — often not because they have reviewed the evidence and consciously chosen otherwise, but because those methods were modeled by the people who first taught them about dogs, or because the immediate visible feedback of a correction can feel more decisive than the patient shaping that reward-based training requires.

Training as a Moral Statement

Professor Sandøe's framing of the findings is direct and worth quoting at length. "Training is not a neutral activity," he said. "It is an activity in which the owner's view of the animal becomes apparent." In his analysis, the training session is not a purely technical exchange between human and dog. It is a moment in which the owner's underlying assumptions about the dog's nature and moral worth become visible in concrete, observable choices. Do you believe your dog deserves to understand what you want? Do you believe their confusion is a communication problem worth solving? The answers to those questions show up in the training data.

Training is not a neutral activity. It is an activity in which the owner's view of the animal becomes apparent.

The implication runs in both directions, which is what makes the research genuinely actionable rather than merely descriptive. If training methods reveal ethical orientation, then changing how you think about your dog — extending more moral weight to their experience, developing a richer model of their cognitive and emotional complexity — should produce corresponding changes in how you train. You don't have to start with the training technique. You can start with the question of what kind of being you think is on the other end of the leash.

It is not only about learning theory — it is also an ethical discussion.

Why Punishment Persists When the Science Points Elsewhere

The persistence of aversive training methods in the face of a well-established evidence base for positive reinforcement is a puzzle worth examining. Part of the answer lies in immediacy: a physical correction or a sharp sound produces immediate cessation of the unwanted behavior, which registers as success in real time. The longer-term consequences — elevated anxiety, reduced trust, increased reactivity in contexts that trigger memories of correction — arrive later and are harder to link causally to the training session that produced them. Positive reinforcement operates on a longer timeline. The relationship between careful behavioral shaping and the eventual result is less viscerally obvious, especially in the early sessions.

Part of the answer also lies in cultural transmission. Training methods are learned, often from parents, breeders, early training classes, or the dominant method of whatever decade someone first acquired a dog. If your framework for understanding dog behavior was built around the assumption that corrections are necessary tools, that framework feels like common sense rather than a philosophical choice. The University of Copenhagen study's specific contribution is to make that choice visible — to surface the ethical orientation embedded in the technical practice and to ask owners to look at it directly, rather than treating it as a given.

What This Means for Your Next Walk, Your Next Command

The practical implication of this research isn't guilt — it's an invitation to self-examination. If fewer than 18 percent of owners currently train exclusively with reward-based methods, the gap between where the evidence points and where practice currently sits is substantial. Closing that gap is not primarily a technical challenge. There are excellent resources, courses, and communities built around positive reinforcement training. It is, as Sandøe suggests, a values clarification. Do you believe your dog has an inner life worth consulting? Do you believe their experience during training — their confusion, their frustration, their pleasure in getting something right — matters morally? If yes, the method follows from there.

The walk you track every day, the commands you give and the way you give them, the moment your dog does something unexpected and you choose how to respond — each of these is an instance of the relationship made concrete. The training session is not a separate activity from the walk, from the bond, from the years of accumulated trust and communication that make a dog and a person genuinely legible to each other. You are always practicing a set of beliefs about another creature's mind and moral worth. Getting those beliefs examined, or at least examined more deliberately, matters for the dog. As it turns out, it also shows up in the data.