This is the blog of a Norwegian (Marianne, aka 'M') who moved to Australia in 2007 to be with Roger (aka R) and feels lucky to have escaped a comparatively communistic country with ridiculous weather!

I think the dog training world is confused about what negative punishment (P-) and negative reinforcement (R-) are. I am currently doing a course to be a dog trainer, and I have also done a lot of reading about dog training, both in books and online. And I think dog trainers have it wrong.

By definition, punishment weakens behaviour and makes it less likely to happen in the future. Reinforcement strengthens behaviour and makes it more likely to happen in the future. 

R- is something aversive that goes away when the dog does what you want, thus making the behaviour more likely to happen again. An example is when you push on the dog's butt to get him to sit. When he sits, the pressure goes away. 

P- is removal of privileges when the dog does something you don't like, thus making the bad behaviour less likely to happen again. This can be a time-out, or the end of play time. 

Currently, the dog training world also thinks of "withholding an expected reward" as negative punishment.

For example, if the dog is asked to sit, and does it slower or more crooked than normal, you withhold the reward because it wasn't good enough. Or if you're shaping a new behaviour and you've rewarded the same behaviour a few times, you then stop rewarding that behaviour in the hopes that the dog will offer more. 

Everyone who's into dog training will tell you these things are examples of P-. I disagree.

The purpose of withholding the reward for the slow and crooked sit, is to make the sit happen faster and straighter in the future. The dog learns that he only gets rewarded when he sits fast and straight. When you shape a behaviour and withhold the reward, the intention is for the dog to offer a stronger or more intense behaviour. So by definition, you must be reinforcing the behaviour. If you punish the behaviour, it will get weaker, not stronger.

I think withholding an expected reward should be classified as R-. Withholding an expected reward is the mild aversive that goes away when the dog offers a stronger behaviour.

If you think of it as P-, you must say that you are punishing the mediocre behaviour in order to get a stronger behaviour. That barely even makes sense.

I have been undecided on whether or not banning them is the right thing to do, and have been thinking very hard about it  for the last few days. My immediate thought was 'yep, ban them', but then I thought that they are actually a useful tool for gaining control of very large, reactive/aggressive dogs while training them, as long as they are used correctly and as a last resort. So I then thought maybe I don't support a ban after all.

But after thinking about it for several days, I have come to the conclusion that I do support a ban.

I have been watching the discussing of "ban vs no ban" in the Aussie dog forum for days. The arguments against the ban are usually that they are a training tool, and when used correctly they can mean the difference between being able to re-train a dog or putting the dog to sleep. They also say that when used correctly, there is no extreme pain and the prongs won't pierce the skin. The arguments in favour of the ban are that the prong collar is cruel and that a lot of people use them incorrectly, and that a lot of people even sharpen the prongs to cause more pain, and that there are other, gentler ways of controlling reactive dogs.

Sweden and Norway have both come up as countries that have already banned prong collars. What hasn't been mentioned is that in Sweden it is also illegal to crate a dog at home while you are out of the house. This obviously isn't because crates are evil, but because a lot of people were overusing the crates and the dogs were suffering because of it. When you look at banning something, you have to weigh the usefulness up against the potential for damage. And the Swedish government deemed the damage potential of crates higher than the usefulness, and so they were banned. No one has argued that crates are "not useful". 

The same thing applies to prong collars. Yes, they CAN be a useful tool, when used correctly. However, most people will not use them correctly, or use them too much, or use them when they could be using something gentler. People against the ban say that rather than banning them, people should simply be educated on the correct use of them. The problem is, I have seen very experienced trainers use prong collars as a first resort against bad manners. Bad manners! Not huge aggression problems where everything else has been tried and failed, but to fix a fidgety sit, which could easily have been fixed using nothing but treats. But the trainer immediately reached for the prong collar. 

So where is this good education about prong collars going to come from? This experienced trainer certainly should  not be trusted to educate people.

How many people will be using the prong collar as a quick and permanent fix, rather than as a temporary training tool? 90%? While maybe 10% will use them correctly and temporary, only long enough to re-train a particularly difficult dog. Should we sacrifice the mental and physical health of those 90% of dogs where the prong collar is used incorrectly, for the sake of giving the 10% this training tool? 

Also, how many dogs are really so large and aggressive that NO other tool can be used to control them? I am guessing not many.

And so, when you weigh the usefulness of the prong collar up against the potential for damage, a ban makes perfect sense.