So I have been alerted by a friend that perhaps I was a little vague in my previous post about whether or not I will use a clicker to teach Case to herd. I have since added a note about this to that post but thought a new post was in order for further clarification.
The short answer is, I will consider all the tools and skills I have learned since I first picked up a clicker and since I first watched a Border Collie herd sheep. I am ruling out nothing nor am I automatically including something.
I guess this could be read to mean I would use a shock collar if I thought it would work. That’s not what I meant, at all (to be even more clear, I will never, ever use a shock collar on a dog, period).
Breaking it down, I will consider all the tools and skills I have learned since I first picked up a clicker means that I have learned a lot in the last 16 years about how animals learn. Most simply, I have learned about the A -> B -> C of behavior. That each behavior has an Antecedent which precedes the Behavior and is then followed by a Consequence. That my best bet is to manipulate the Antecedents and set the animal up for success by training, training, training. In this case it will mean getting an unbelievably reliable “down” at a distance and an unbelievably reliable recall. Those two things will be necessary to protect the sheep from being unnecessarily harassed or injured.
I have learned we have to find out what is most reinforcing for the animal in each circumstance. My dog might willingly come to me across the room for just a scratch on his neck because we have a relationship and he loves having his neck scratched. If someone else he likes is in the room, I might need to pay with a piece of food. If the UPS man is in the driveway, I might need to reinforce with some really good food! He won’t know when he comes what I have to offer, but he’ll remember whether it was worth it last time.
One thing I have learned since I first watched a good Border Collie work sheep is that nothing, ab-so-lute-ly nothing, is more reinforcing than working sheep. They will drop from exhaustion, from heat stress, from illness before they volunteer to leave the sheep. We had one who suffered from heat stress and we had to learn to recognize the early signs, stop him, run to him and carry him as fast as we could to the nearest stream or water tub to drop him in and get him cooled off. This was not because we had pushed him to work that hard but because there was a job to do and he just kept at it until we asked him to stop.
So, what am I going to offer an enthusiastic adolescent pup if I call him off the sheep? There is only one thing and that was taught to me by a Scottish shepherd years ago, “give him his sheep back”. This is similar to a ball obsessed dog who will learn to drop the ball if you throw another one. The difference is that the ball doesn’t care. The sheep do. Thus the training, training, training and management, management, management that will be required.
Things I have learned since I first watched Border Collies herd sheep are how they know better than I do what needs to be done when. How many times have I heard a good shepherding coach say “trust your dog”. That’s because the dog has the better eye. They see the thought go through a ewe’s head that she might consider breaking from the flock. The dog sees that the cattle want to drift uphill so he needs to be on that side, rather than directly behind them. When I wrote in my previous post about the precision of the clicker, I meant I do not want my less skilled eye to precisely direct a dog to a spot when if I’d let him find his own balance, he’d be a far, far better working companion.
So while I may not use a clicker while he’s working, I will use all the good, proven science behind clicker training.
When I said, I am ruling out nothing, I was just trying to be realistic; I wasn’t suggesting I’d plan to punish. I don’t believe in grabbing children by the arm to discipline them. But if I saw a child headed for traffic, you bet I’d grab him by the arm to prevent him from being hurt. Likewise, I intend to do my best to set Case up for success, but that doesn’t mean I’m going to let him get run over by stock or hurt stock rather than tackling him as he goes by just because I don’t want to “punish” him.
If he was after a ball, I’d let him go by, let him grab the ball, recognize it was my error, train more and make a better plan for next time. If he is after sheep, I may have to resort to a tackle to protect puppy and stock, but then I’ll go back and train, train, train some more in a safer environment to build the strength of the behavior I want so another tackle isn’t required. Case showed when he was 8 weeks old that he could climb out of his three foot ex-pen. I have seen Border Collies scale fences, squirt under walls, squeeze through doors and any number of other gymnastics to get to stock. I’ll do my best to prevent that from happening. But there are hundreds of years of genetics in there telling him to get to those sheep.
Since getting this puppy I have heard from people who’ve used a clicker and say “you can’t use a clicker for herding”. But there are so many incredible things that people have trained so many different animals to do that I can’t believe there isn’t a way to utilize this technology.
I have also heard from committed clicker trainers who say of course they have used it and of course it works. But they are doing it as a hobby, not living with sheep nor teaching real-life herding. When it’s a hobby, you can give it up if it doesn’t work out. Case is going to be a critical part of our farm.
I have spoken with a world renowned dog trainer with shepherding experience who uses positive reinforcement and many markers. That’s a brain I’d like to pick some more.
I’ll finish with another quote from my previous post:
So how am I going to do it? I don’t know yet. I will need to feel my way step by step to see what works best.