I try to stay positive. I really do. This is a rant.
I use Google Reader to monitor news stories and blog posts for information about dogs and dog training. There’s a lot of information out there and I try to filter it to stuff that I can either share on twitter, share in the weekly updates, or riff on in a blog post.
Every so often I see an article that actually makes me angry. Most of the time I just move on, but this time I saw two pieces so close together that I can’t help myself.
First, we have “Not All Trainers Are Competent”. (Isn’t that a lovely title? Doesn’t it just make you think of someone who has something to prove, I mean, someone who wants to help?)
The author hangs his entire point on what clients told him other trainers said. Do you remember playing telephone as a child? Remember how hilarious and ridiculous things quickly became as they were passed on from person to person? Similar things can happen with training instructions. I learned that very quickly when I started seeing clients, and as a result I’m very circumspect about criticizing other professionals based on second hand information.
The author goes on to refer to “Reward-based training — treats for tricks” – this is one of my favorites! There is a school of thought that says that as soon as treats are involved in training the dog is doing “tricks.” It’s based on….well frankly I don’t know what it’s based on. Certainly not science.
Behavior is behavior and anything that strengthens a behavior is a reinforcer, whether it’s food or the vaunted “affection and love” that is usually tied to the Disney-infused belief that a dog should simply obey because he wants to.
Food is a convenient “least common denominator” and using does not weaken (or strengthen) training on its own. It can be used effectively and it can be misused. Anyone who says otherwise should pick up a book on behavior science, or maybe just a book on dog training that was written some time after the Watergate scandal.
As said, normally I let stuff like this just fly on by, but within a few hours I came across this little treasure. <- Dead link. See below.
Again with a warm and fuzzy title: "Clicker training rarely works." (Update: Version with this title was pulled from site?) I’m not even a clicker trainer and I’m annoyed. (Note: I more recently came across the same article with the title “Clicker training teaches dogs tricks, not manners.” It’s been recycled. Like fertilizer!)
(Another new version, has shown up online.)
(Saved a copy in case it disappears again.)
It’s a question and answer column, (if actually answering the question doesn’t count.) Five months ago the questioners took their dog to a basic obedience class. They left satisfied. Now the dog is acting up, including knocking over the kids, growling, and showing his teeth.
The answer? The trainer you saw probably sucked because (s)he used a clicker.
Pulitzer, here we come!
More often than not when people pay for clicker training they are, without realizing it, paying someone to teach their dogs to do tricks. Mind you those tricks are called sit, down, stay, etc. I call them tricks when the end result is a dog won’t do it for any longer than it takes to swallow a treat or their owner to turn their back.
Funny, I call training unreliable behaviors “crappy training,” and I’ve seen it done with all kinds of tools, from clickers, to treats, to special collars named after self-proclaimed expert trainers.
The author then goes on and paints a picture of a clicker trainer that refuses to ever punish a dog, even going so far as to compare how these imaginary trainers raise their imaginary kids. Is it possible to train a dog without using any punishment at all? No. Punishment is a fact of life, and attempting to deny that you are using or that it is not needed at times is foolhardy.
Are all clicker trainers like this imaginary trainer? Are all elephants grey? Is everything grey an elephant?
The entire “answer” is dedicated to building a straw man and then setting him on fire. How this is going to help with the dog knocking over and snapping at the kids is beyond me.
If you are talking to a trainer and the main gist of his reason for being is how much better (s)he is than someone else, or how terrible or ineffective another trainer or his methods are, walk away. And yes, that goes for people who constantly berate Cesar Millan too.
Find someone that can tell you, and show you, how they can help you.
12 Comments
“Is it possible to train a dog without using any punishment at all? No. Punishment is a fact of life, and attempting to deny that you are using or that it is not needed at times is foolhardy.”
I think there are a lot of very well educated, highly experienced, well known trainers who would seriously dispute that claim. Are you using “punishment” in the training sense or the lay sense?
Rant on, Eric. Seriously. I know we don’t always see eye-to-eye, but on this particular subject, I think we’re in agreement.
BTW, your ‘crappy training’ comment reminded me of a terrific training acronym — Continuously Reinforcing Approximate Performance = CRAP. We use it to describe the situation where a well-meaning trainer continues to reinforce ‘almost’ right with the same frequency and intensity, instead of properly holding back and working to raise the performance criteria until the desired behavior is solid. And what does that trainer get by continuing to reinforce ‘almost’? You guessed it…CRAP.
Keep raising the bar on behavior knowledge, Eric, and work to elevate lay perceptions of behavioral language from perception to true understanding. Don’t fall into the trap of dumbing down behavioral language — that’s what’s gotten dog training into the mess we’re in right now where ‘positive’ is perceived as fun-good and ‘negative’ as destructive-bad. As trainers, we owe it to the people we work with to keep raising the bar of common behavior knowledge — and we can’t, with our own language, continue to reinforce approximate performance. That just leads to, as you put it, crappy training — not of just dogs, but of the owners.
Excellent post Eric! I agree with you wholeheartedly!
@barrie: Training sense, mostly. I don’t know if I can concisely describe where I stand on punishment in a brief blog post or comment.
Speaking broadly, I stay away from broad statements like “punishment is never necessary” or “punishment is always necessary.”
It’s certainly not a discussion to open in a Q&A in a newspaper.
I’ll side with Dr. Dunbar on this one: http://www.dogstardaily.com/blogs/non-aversive-punishment
Good points.
People spend a lot of energy bashing others. I find it’s a lot more productive to ignore rather than bash, and use that energy instead to do something productive. Trainers just end up looking childish if they spend too much of their time criticizing others methods. Some horse trainers are extremely guilty of this.
Besides, criticizing someone makes the person defensive and almost never leads the person wanting to change their behavior. If I tell a correction-based or force-based horse person that she’s “wrong,” I can probably get a short lecture in, but I bet she’ll never again ask me for advice or opinions on training.
I think clicker trainers and other positive trainers can best promote their ideas by examples. A really well trained dog or horse speaks much louder to quell criticism than words ever could.
Mary H.
http://stalecheerios.com/blog
Thank you for your comments. I spent a lot of time (way too much, I am sure) thinking about this before I posted it because even this is, to a certain degree, exactly what I am complaining about.
Setting the example is by far the best approach.
I think those trainers who bash positive training are suffering from fear based aggression.
Dr. Dunbar says:
This technique is extremely effective, works surprisingly quickly, and prevents the need for physical restraint or aversive punishment.
when describing repeatedly asking for a sit as non-aversive punishment which really is an ENTIRELY new definition of punishment to me and I believe a lot of other OC people.
So yes, if you are calling repeating a cue in a “sweet tone of voice” punishment then no, you probably can’t train a dog without the use of punishment unless you are a wicked great clicker trainer who never, ever adds a cue until you have a completely solid behavior such as Dr. Yin or Patrica McConnell.
I do believe it is very possible to not train a dog using what Dr. Dunbar calls “aversive punishment” which would be what OC trainers think punishment means.
For the record, I am not “bashing” you. I just want to understand what you meant by that statement because, if you stop and think about the fact that to the vast majority of trainers punishment is a loaded term which generally means to add an aversive unpleasant enough to the dog to make the behavior less likely to occur, it is actually a fairly bold statement.
Maybe you started off with Dr. Dunbar’s definition of punishment (it is one I only just read this year) but the vast majority of us are either straight up clicker trainers or like myself, cross over trainers who didn’t learn about OC until the late 80s and all of these terms had previous definitions for us.
I didn’t include strict definitions of what I meant by punishment for a couple of reasons, maybe I should have, but other than updates about broken links I made an agreement with myself that I would not go back and correct blog posts. They stand as they are and I’ll take the “hits.”
I do tend to think like an engineer (because I am one) and it sometimes gets me into trouble. I catch myself, usually a few paragraphs into a debate, arguing over terms because I am using a different meaning. I try to avoid that when I can.
While most of my feedback comes from other trainers and very savvy dog owners, I still try to make my most of my stuff accessible to “average pet owners” (APOs).
I think that most APOs are only barely interested in the 4 quadrants and debates about aversive vs. non-aversive punishment. I believe that most of them have a “line” somewhere that they won’t cross. Where that line is depends on a variety of factors, and how firmly anchored that line depends on a bunch of others. It’s difficult to negotiate that line on a web page, in posts that I try to keep short, and that are aimed at complete strangers.
I actually spent a lot of time debating with myself whether to address the original article’s brief dive into the quadrants because it was so incomplete. I ended up opting for sloppy language and kind of expected someone to call me on it.
Some thoughts on the word punishment:
A “time-out” is mostly negative punishment. But you could make a case that interrupting and moving the dog to another place, and/or leaving the room to actually execute the “time out” is positive punishment (At least I could.) I also think that most APOs would just call it punishment and not worry about the “type.”
Repeating a cue, as you said, can be positive punishment. Here again, most APOs would probably categorize it as functionally the equivalent as saying “No.”
“Body-blocking” at a door or cones to teach “wait” is probably positive punishment. (If it works, it stops rushing the door, right?) Again to an APO – whatever.
Putting a dog back into a down stay can be considered punishment, even if you do it with laying a finger on the dog or her leash.
At a certain point it’s word games, and like I said before, I think many APOs just turn off. They might be “technically” wrong, but it’s our job to communicate with them in terms they can understand first, improve how they communicate with their dogs next, and if we can educate them along the way, that’s even better.
I also think that the word games can end up in strange places like dropping the word trainer because it’s “associated with coercion” or not turning away from a dog that jumps up because it’s technically aversive.
P.S I don’t think you’re bashing me. I enjoy the feedback.
Ah, so really, you just meant it in a lay sense since you are writing this for the APO who doesn’t know or really care about OC which is a totally valid thing to do and which is why I asked for clarification
I would still argue that in the training sense, unless you are really committed to Dr. Dunbar’s definition of non-aversive punishment (i.e., that every time we repeat a cue we are in fact punishing our dogs – eek!) that I disagree with your statement that punishment-free training is not possible but a number of years ago I would have completely agreed with you.
Nice post! Trainers can be so competitive. For a *very* brief time I took competitive obedience classes with a woman who stood there and insulted every other trainer that she knew in my state. One of my favorite trainers had had a reactive dog, and this trainer kept saying how my favorite trainer (Leslie) had a “vicious” dog that Leslie had screwed up and failed to control, which I of course knew was incorrect completely. Leslie had had a difficult dog and she had turned that dog into a superstar. But, ya know, rumors are rumors and it’s an ugly marketing device. This person was trying to collect Leslie’s students, and since Leslie was extremely busy teaching agility, and dropping some of her obedience classes, this lady had a few of Leslie’s students temporarily in her classes, and she was busy trying to convert them.
It’s totally annoying! I felt that way when S. Clothier wrote that kinda mean (in my opinion) post about Jean Donaldson. Everyone is entitled to their opinions, but can’t we just say it’s OUR opinion, instead of trying to cut the liver out of fellow trainers?
I have written about Cesar Millan, but I always do it as a reviewer of the material, and I comments on the material presented, not the man, who lets face it, seems as likable as anyone. I can write about the method, and I can develop arguments about it, but I don’t need to turn anyone into an ogre.
Well apparently this behavior is being reinforced because it does seem to continue. Cheers,
Jenny Yasi
@Jenny
Thanks for your comment.
Most of the time (at least I’d like to think so) people that “market” this way in person and “traditional” channels tend to self-destruct, or at least are self-limiting since only so many people are enthralled by it. I’ve encountered this in a few different fields. It’s a very popular tactic in martial arts.
I think online is a little more insidious, especially when someone can slime an entire category, like clicker trainers, and create a “meme” like “clicker train rarely works” etc. Stuff like that can be durable.
While I though Clothier’s post was well – I’ll settle for unfortunate here in public – at least it is a focused attack on a specific thing that a specific trainer did that you can view for yourself, and she even provided the link to do so. (I think many people did and the first thing that came to their mind was “hyperbole,” but there I go….)
It’s similar with Cesar, he has a very public body of work that people can criticize. The last time I weighed in, it was on a specific show where he basically had to hang a dog to extricate himself from a situation he created.
That said, if my website, email postings, and tweets were all about how much I dislike Cesar’s show, I’d be really popular with a few hundred trainers and have very few clients.
One Trackback
[...] the rest here: It’s what you know, not what others don’t know Bookmark It Hide Sites [...]