Hmm, well, I can't comment on the "electrician" example except to say that there's a big difference...lol. You're working with a LIVE being, and in working with a live being, you should have a great, deep understanding of that being's behaviour, instincts, signals, and language.
The fact is, people who have great experience in these areas do not get bitten. At least not more than once or twice, and ANY trainer who has gotten bitten, for any reason, will admit openly it is solely because
they moved too fast. They were not paying close enough attention to the dog. That's all there is to it. Because they know what they are doing, they know not to approach a dog if it's not safe. They know how to use their own body language to indicate a safe situation. It's not a "slip of the wrist and I touched an electric wire".
And I have to admit, I wholeheartedly agree that you
can train a dog solely on P+ or R- (although the two often go hand in hand). There's no question those methods can be effective, they are. The question for me lies in - do I really want to interact with my dogs that way? Knowing what I know about canid behaviour and canid learning, there are just waaay too many loopholes to worry about doing it "right".
Darn, I jerked the choke chain 4 seconds too late, now the dog hates children. Darn, I squirted the dog, and now he's terrified of hoses, running water, and water guns. Darn, I set mouse traps under newspaper on the couch, now he's terrified of the living room, or he's terrified of loud noises. The point is, corrections and P+/R- in training leave open way too much room for unwanted consequences. And the fact lies in that once we do it, it's too late to say "Oops".
It was once said that in order for punishment to truly be truly effective, it had to be
a) immediate - easy enough
b) harsh enough so that you only had to use it once, twice at the most. Any more than that is ineffective punishment (the case of the dog who is immune to the hardest of jerks from a choke collar, because the owner started out very gently, and got increasingly harder, till the dog stopped paying attention to it altogether, as an example). And in order for a person to only have to do it once or twice, it would have to be so extreme, that it would be considered cruel and inhumane.
However, you
can also train a dog on solely R+ and P- (and extinction). And you know the worst thing that can happen? You can accidentally give your dog too many treats. You can accidentally reward something you didn't want to. There is NO risk in the relationship with your dog, or the dog's well being.
I don't want to use a way that might or might not work, and that runs a risk of hurting my relationship with my dogs. Because I know that dogs do not ever need "corrections" in order to be a happy, healthy animal, I choose not to use them in my life, and especially not in my training program.
My dogs do not have the word "no" in any part of their human-word base. If you said it to them it'd be no different than saying "banana" or "mailbox". To me it just doesn't make sense. "No" does not teach the dog what it should do. For most people, "no" takes on some or all of the following:
a) get off me
b) get off the cat
c) don't jump on the bed
d) don't bark
e) don't lick me
f) (insert reason here).
"No" takes on so many forms I don't know how the dog can possibly learn what it means, because people use it as the all-knowing word, expecting their dog to know what "no" means in that situation. Would you ask your dog to "sit" in one situation, and then expect it to lay down when you say "sit" in another? It does not in any way teach the dog what it SHOULD do instead, and that is the basis of effective learning. Instead, one would teach Off, Quiet, Leave It, Get Your Toy, Place, etc, things the dog can actually DO, and things the dog can and will understand as different cues. I highly prefer to teach the dog what I want it to do INSTEAD of what it's doing now, not what I don't want it to do.
Kim MacMillan