25/10/2026
🐾 E-collars, ethics, and the future of gundog training 🐾
After hearing that a previously R+ trainer - who I've interviewed twice on my podcast - has now become an e-collar advocate, I’ve been thinking recently about e-collars, their use in gundog training, and the arguments that are made in their defence.
I’ve watched with interest as the debate around e-collars has continued to evolve. My focus has always been to highlight what is possible without the use of force. If something can be trained to the same standard either with or without an aversive, why would anyone (who's not a sadist) want to use the aversive?
What concerns me is not simply that individual trainers make different choices, but the arguments we use to justify those choices and the implications they have for the wider dog training community.
For me, this ultimately comes down to ethics. But I think there are some important points about learning theory that we need to consider first.
⚡ 1. If an e-collar doesn’t cause fear or pain, how does it stop a dog doing something they really want to do?
One of the arguments commonly made in favour of e-collars is that the stimulation doesn’t hurt. People demonstrate this by placing a collar on their own hand, selecting the level they intend to use on their dog, and reporting that they can barely feel it.
But here’s the problem.
If that same stimulation is sufficient to stop a dog from doing something they find extremely reinforcing, such as chasing a deer or a rabbit, then we need to ask what is motivating the dog to stop....? A little tickle isn't going to cut it.
The dog is engaged in something they find highly reinforcing. So the stimulation must be sufficiently aversive to interrupt that behaviour and to prevent the dog from continuing to access that reinforcer.
That doesn’t automatically tell us exactly what the dog is experiencing, and we shouldn’t pretend that we can measure a dog’s subjective experience simply by comparing it with our own. But it does raise a fundamental question.
If the stimulation is sufficiently aversive to outweigh something the dog desperately wants to do, can we really dismiss its significance by saying that it doesn’t hurt *us*?
An aversive doesn’t have to produce a visible injury, a yelp or an obvious reaction to affect behaviour. Nor does the absence of an obvious reaction tell us that the dog isn’t experiencing discomfort, fear or stress.
We need to look at the FUNCTION of the stimulus, not simply how it feels to the person holding the remote.
🧠 2. What about using a very low level of stimulation, or vibration alone?
E-collar users will often say 'I only use this very low level'. Or 'I just use it on vibrate'.
This is where learning history becomes particularly important.
When it is effective, a low-level stimulus works because the dog has previously experienced higher levels of stimulation. The lower level has acquired meaning through its association with what may follow if the dog doesn’t respond.
The same principle can apply to vibration.
A vibration may function as a warning signal: respond now, or something more aversive may follow.
In learning theory, this is an example of how a stimulus can acquire its function through conditioning. Its effect cannot be understood simply by looking at the intensity of the stimulus in isolation. In a single rep. Out of context.
So, yes, someone may genuinely be using a very low level of stimulation, or vibration alone, at the moment the dog responds. They can show you that.
But if that signal works because of the dog’s learning history with a more aversive consequence, we have to ask whether we have actually removed the aversive element from the training, or merely moved it into the dog’s history.
Are we really moving away from fear, just because the current situation appears to be a low level - or are we relying on the dog’s fear of what might happen next, if they don't respond?
And if the dog has never experienced anything more aversive, what exactly explains the effectiveness of that warning signal?
These are important distinctions. We shouldn’t assume that a low setting automatically makes the procedure ethically unproblematic.
🎯 3. Can a dog really be responding to positive reinforcement and punishment at the same time?
We often hear that e-collars can be combined with positive reinforcement. A dog is recalled, receives a nick if they don't respond, and gets a tasty treat when they return. The argument is that the dog learns to come back happily for the reward.
But we need to distinguish between the behaviour we observe and the processes maintaining it.
A dog can certainly learn about both reinforcers and aversive consequences within the same training programme. R+ and P+ and R- can all influence behaviour, and their effects can coexist across different situations.
However, when we look at a particular response in a particular moment, we need to consider which consequences are controlling that behaviour.
Is the dog returning because the recall cue predicts an opportunity to earn something wonderful? Or is the dog returning because continuing to chase risks an aversive consequence?
The dog may have learned both associations. We cannot simply assume that the presence of a food reinforcer means that the dog is primarily motivated by that reinforcer, or that the use of an e-collar means that fear is necessarily the sole motivator.
The important question is not whether the dog takes the treat. It’s what is actually maintaining the behaviour.
A dog can take food while also experiencing stress. A dog can perform a behaviour reliably while being motivated partly by avoiding an aversive. And a dog can learn to respond to a recall cue through positive reinforcement without the need for an e-collar.
We need to be honest about those distinctions rather than assuming that adding a reinforcer automatically resolves the ethical questions raised by the use of aversives.
🐕 4. Yes, e-collars can work well. But that isn’t the end of the argument.
I don’t think we need to pretend that e-collars don’t work.
They can suppress unwanted behaviour. They can interrupt chasing. They can influence recall. They can be used at a distance, precisely where we may have the least immediate control over what our dogs are doing.
Those are real training advantages, and pretending otherwise doesn’t help our argument.
But effectiveness and ethical justification are two different questions.
A method can be effective at producing a particular outcome without being the method we choose.
We could probably stop a great many unwanted behaviours by making the consequences sufficiently unpleasant. We could raise children the same way. That doesn’t mean we should.
The fact that a method works tells us something about its behavioural effects. It doesn’t, by itself, tell us whether those effects justify the means used to achieve them.
We also need to consider the welfare costs, the potential for unintended associations, the individual dog’s response, and whether less aversive alternatives are available.
This is particularly important when we’re talking about gundogs, where there are so many situations in which we want to interrupt behaviour that is naturally and powerfully reinforcing to the dog.
The ethical question is not simply, ‘Does it work?’ It’s, ‘What are we asking the dog to experience in order to make it work, and is that justified?’
⚖️ 5. What about the exceptional cases?
Whenever the use of e-collars is challenged, the discussion can quickly move towards the most extreme examples.
The dog that chases livestock. The dog that has killed sheep. The dog whose recall appears impossible. The dog for whom conventional training has failed and would otherwise have to spend their entire life on a leash.
These are serious situations. Livestock worrying, for example, can have devastating consequences, and we have a responsibility to keep dogs, other animals and people safe.
But we need to be careful about the conclusions we draw from exceptional cases.
Even if an e-collar has stopped a particular behaviour in a particular dog, that doesn’t establish that e-collars are necessary or appropriate for the vast majority of dogs. Nor does it establish that the same outcome could not have been achieved through other interventions, management, environmental changes or training methods.
There is also a danger in allowing the most difficult cases to define our approach to *every* dog. Or even most dogs.
The question of what to do in an exceptional, high-risk situation deserves careful, individual consideration. It shouldn’t become a general justification for introducing aversive methods much earlier in training, or for treating them as the default solution whenever progress is difficult.
We need to distinguish between what might be argued to be necessary in an exceptional case and what we should routinely recommend to the wider dog-owning public.
Those are not the same thing.
📚 6. What we can achieve through force-free training depends enormously on **the trainer’s skill**.
This is one of the most important points for me.
What is possible through positive reinforcement and force-free training is influenced, to a considerable extent, by the skill, experience and knowledge of the trainer.
A trainer who has spent years learning about reinforcement, motivation, stimulus control, generalisation, reinforcement schedules, arousal, predatory behaviour and the mechanics of teaching reliable responses has a very different set of options available to them from someone who has not.
That doesn’t mean every behaviour can be resolved easily, or that force-free training guarantees a particular outcome in every case. Individual dogs differ, environments differ, and some situations require extensive management as well as training.
But it does mean that we cannot judge the limits of force-free training solely by looking at what one particular trainer has been able to achieve. If anything, that says more about that trainer's limitations than it does about R+.
And this brings me to a question that I think deserves much more attention.
If what we can achieve through force-free training depends partly on our own level of skill, how can we justify turning to an aversive method simply because we haven’t yet developed those skills? (Why should dogs suffer simply because we are a bit rubbish at this training stuff?)
I’ve seen people put e-collars on very young puppies to stop them chewing furniture.
I’ve seen people turn to e-collars because they haven’t been able to train a basic recall around distractions.
I’ve seen people who repeat a cue over and over without having established the behaviour reliably, then look for a tool that will make the dog comply.
In gundog training, the temptation can be particularly strong when we want to stop a dog chasing, running in, breaking steadiness or ignoring a recall.
These behaviours require planning, an understanding of reinforcement, carefully structured setups, and the ability to progress at the dog’s pace.
An e-collar can look like a shortcut.
But if one trainer turns to an e-collar because they cannot achieve a particular behaviour, while another trainer can achieve it through positive reinforcement, what exactly are we saying when we claim that the first trainer’s use of an aversive is "justified"?
Are we saying that it’s acceptable to expose a dog to fear, pain or discomfort because the person training them hasn’t yet acquired the skills to do otherwise?
That makes very little ethical sense to me.
Of course, not every failure is attributable to poor training. Dogs can have complex behavioural needs, and even highly skilled trainers encounter difficult cases. But that makes individual assessment and appropriate support more important, not less.
The limitations of a trainer’s current skill set should not automatically become the dog’s welfare burden.
❤️ 7. Force-free training doesn’t have to be all or nothing.
I want to be clear about something else.
I’m not suggesting that everyone has to change every aspect of their training overnight, or that we should dismiss people who are at different stages of learning.
My work is about teaching force-free gundog training. That’s what I teach, and I have a clear understanding of what I mean by force-free training. It isn’t the absence of all physical contact or all management. It is an approach that seeks to teach and guide behaviour without relying on the application of aversive consequences to suppress it.
Every time we help someone replace an aversive procedure with a positive reinforcement-based alternative, we have an opportunity to improve a dog’s experience of training.
Take force fetch, for example.
If someone replaces force-fetch with the clicker retrieve, that is a substantial change.
That’s a win for the dog.
It doesn’t become meaningless simply because that person still uses other methods I wouldn’t choose.
We don’t have to insist on an all-or-nothing conversion before recognising progress. We don't have to shame people who are still using some aversives. The more people discover that R+ can achieve results in gundog training, the more likely they are to explore what else they can do with it.
And the more trainers who develop these skills, the more options we can offer people who are struggling.
Every aversive procedure we can genuinely replace with an effective, less aversive alternative is a step in the right direction. 🐾
🔮 8. We also need to prepare for the future.
E-collars are already prohibited or restricted in a number of countries and jurisdictions. It is extremely likely that further restrictions or a ban will (at some point) come into place, in the UK.
If that happens, people who have come to rely on e-collars may find themselves having to rethink how they train their dogs.
And that isn’t a particularly good time to start learning how to do everything differently.
Changes in legislation require preparation. They require trainers to develop alternative skills, owners to understand different approaches, and the wider community to become more familiar with effective methods that don’t rely on e-collars.
Why wait?
The best time to develop the skills to train without e-collars is before we find ourselves having to do so.
So, where does that leave us?
I don’t think we need to deny that e-collars can prevent behaviour. I don’t think we need to pretend that every dog is easy to train, or that every behavioural problem has a simple solution.
What I do think is that we need to be willing to look beyond the very basics; beyond whether a method produces the behaviour we want.
We need to ask what motivates the dog, what the dog may be experiencing, what learning history is involved, what alternatives are available, and whether the outcome justifies the means.
We need to recognise that our own training skills influence what we believe is possible.
And we need to be prepared to keep learning, especially when the alternative is to place the burden of our limitations on the dog.
For me, force-free training isn’t about claiming that every dog can be trained effortlessly, or that every problem has an easy answer.
It’s about the direction in which we choose to move.
Towards better understanding of learning theory. Towards greater training skill. Towards methods that minimise fear, pain and discomfort. Towards taking responsibility for our own development rather than reaching for a tool that makes compliance easier to obtain.
The fact that a method works is only the beginning of the discussion, not the end of it.
Our dogs depend on us to make decisions on their behalf. They can’t choose the training methods we use, or tell us in words what those methods cost them. That makes our responsibility to think critically about those methods all the more important. ❤️🐕🦺
