When our sheep are lambing we add grain to their diet.  Once a day in the evening we go out to the barn on the ATV and feed them.  We fill buckets with corn and carry them to the feeders and pour it into the plastic troughs.  Within a few days, no matter what time of day it might be, the sound of the ATV sets them into a frenzy, racing to the barn, bleating, carrying on as if they are starving.  The sight of my husband or me carrying a bucket across the yard will send them racing back to the barn, for they can actually see us from several hundred yards away from the house.  The sound of corn pouring into a bucket or into a plastic trough will result in the same response, even if it is early in the day and they are not the intended recipients of the grain.  Within a week of beginning their added nutrition, I get the feeling that they perceive me as one huge corn cob.

A couple weeks before the lambs are to be weaned, we begin to reduce the corn in their diet until they are receiving no additional grain on the day we take their lambs from them.  The process of reducing their caloric intake helps them through weaning and the eventual complete lack of lactation.   By the time the lambs are weaned, the sound of the ATV, the image of my husband walking with a grain bucket, or the sound of pouring grain no longer elicits the frenzied bleating or racing to the barn.  The triggers no longer are followed by the reward and the behavior is extinguished.

Even though our sheep are crazy about eating corn, I have learned that I cannot count on using grain to get them to go where I want them to go.  Regardless of how hungry they may be, they will not follow me with a bucket of grain if they simply do not want to do so. 

We use herding dogs to manage our sheep, and even if the sheep are greedily eating their dinner, the dogs can make them move off the feeders if we instruct them to do so.   That is because the sheep have learned to respect the dogs.  When necessary, the dogs will “punish” a sheep that is not staying in line.  Herding dogs nip sheep that are challenging the dog’s authority and are not moving where the dog is directing them to go.  The worse the challenge from the sheep - the harder the grip will be.  A ram that maliciously charges a dog can expect a much harder grip than a ewe that is merely trying to double back to the grain or hay.  This ability to measure the correction is programmed into a well bred herding dog.  The sheep learn that they can avoid a nip by complying with the dog’s demands and they understand that the punishment fits the crime, so to speak.  So, they do not panic at the sight of the dog, they simply assume the posture of well seasoned students lining up to go indoors after recess. 

There is a contemporary method of dog training that is strongly based on using treats to reward specific behaviors.  This, in and of itself is not a bad thing.  But, like our sheep at weaning time, when the treat goes away, so may the behavior.  The method was developed in the captive aquatic mammal realm, where dolphin trainers were attempting to get their wild subjects to perform circus tricks.  Since a killer whale, walrus or a porpoise cannot be physically controlled, and because it has no natural affinity to subordinate to human keepers, they needed a way to communicate their pleasure when the animals performed natural behaviors.  It is now used routinely by wild animal trainers in performance areas like the motion picture industry or animal theme parks.

At some point, a dolphin trainer decided to try the method with her domestic dog.  Through good marketing, the technique has become quite popular, sometimes referred to as “click and treat” or “all positive” training.  The method is not based on leadership, loyalty or genetically programmed subordination.  It does not acknowledge the thousands of years of genetic selection that we have designed into our domestic dogs.  It works for pretty much any species with a nervous system.  It disregards the very unique status of dog, and puts him in the same category as a porpoise or a grizzly bear.

To perform the method, you just need a stimulus (the sound of the ATV for our sheep, or perhaps a small clicker device for a dog, a whistle for a porpoise) and you need to pair that with the treat (corn for our sheep, a piece of hotdog for your pup, a mackerel for the dolphin).  First, you teach the animal that the sound is associated with the treat.  The sound of the ATV engine preceding the sound of pouring grain worked for our sheep.  Dolphin trainers whistle and then toss a fish at the dolphin’s smiling face.  You click then offer the hotdog to your pup.  Once the animal has figured out that the sound is associated with the treat you can begin to use that sound to inform the animal that you are happy with a particular behavior when the animal presents it.  You wait for the animal to perform the desired response (a dolphin jumping out of the water, a bear standing on his hind legs, a chicken scratching her head, a lizard licking his eyeball) and then you click or whistle to inform the animal that behavior was “good”.   Upon hearing the sound, the animal’s brain says “yummy!”, he comes to get his goodie, and you give it to him!  This reinforces the behavior.  The animal may then choose to perform the behavior again to see if it can get the reward once more.  If you click or whistle again as the animal is exhibiting the behavior, you begin to reinforce the behavior.  Then, you can begin to pair a command (hand signal or voice command) to try to “control” when the animal performs the behavior.  It is a method that works on dolphins, bears, chickens, sheep and even lizards because it is based on a response that is found in animals with central nervous systems!

But, in the same way that the sheep consider me a big cob of corn and have no sense of respect for me or a need to do my bidding, dogs that are trained exclusively with this method begin to view their human as a big hotdog, or a gumball machine filled with Jerky Treats.  This method completely excludes the use of negative re-enforcers for behaviors that you find objectionable.  It excludes the concept of punishing bad behaviors and only relies on the learning that takes place when a dog chooses to present a behavior and when the dog wants to receive a goodie.  If the reward begins to loose meaning, the behavior may begin to extinguish.  A dog that is not food motivated, is not hungry, or just doesn’t care to play the game will not comply.   The method does not utilize the thousands of years of designed selection that have gone into creating modern dog.  There is no need for a special relationship with the animal.  I find that a real shame, since the whole point of that selection was to make the training and management of dogs easy, and to make living with a dog far superior to living with, say, a Brown Bear.

The clicker method works well for teaching dogs to do tricks or to do behaviors that are not based on compliance.  If you want to teach a dog to walk up a ladder, using the click and treat method is a good choice because the behaviors can be broken down into many steps, the dog can be rewarded for the little steps and then for adding steps together.  Tricks usually have no meaning to the dog, so the click and treat method can be used successfully, albeit it is not the only possible method.  And, depending on the dog, it may not be the best method, to accomplish that task. 

In my opinion, the clicker method does not work well to extinguish existing “bad” behaviors.  The method actually ignores half of the learning strategies that animals (including us humans) use to survive in the world.  Avoiding negative events based on receiving uncomfortable responses for the behavior is a valid method for training animals, including humans.  In fact, the ability to avoid discomfort is a basis for many behaviors.   When used appropriately and fairly, it puts the animal in complete control of his own destiny, resulting in a very confident dog.  When he knows the ramification of a behavior and he can avoid it by controlling his own behavior, the dog develops into a self-assured animal with a heightened loyalty for his leader (who he sees as a deliverer of information about the rules of the household). 

We have twelve dogs.  It’s a pack, one could say.  It is comprised of dogs as young as 6 months old and as old as thirteen.  There are intact males and neutered ones.  There are intact females and spayed ones.  We have dogs that are rescues that are obvious results of poor early management.  They tend to lack some basic dog social skills that the pups we have raised seem to exude.  We have dogs that we have bred or raised ourselves that appear to have good dog-social skills.

I often watch my dogs play in the yard.  They are constantly reminding each other of their rank in the pack.  For example, Breeze (a neutered, nine year old male) and Sage (a two year old, intact male) are very good at peace keeping.  Several times in an hour they might be seen running up to certain dogs and licking them under the chin to affirm, at least it seems, that they are still “cool” with the more dominant dog.  They also perform calming type behaviors (like yawning, turning their eyes away, bowing) when there are little uprising amongst the dogs.   And, yet, at the same time, when a subordinate dog or pup acts too silly, they are the first to take the youngster’s muzzle in their own jaws and clamp down at the perfectly appropriate level to demand the sort of respect that the pup failed to present. 

I have also seen a dog demand a “down-stay” from a subordinate, putting the dog in its place and then reinforcing that position for some period of time by the turning up of a lip, showing of teeth, staring down the subordinate, uttering a low growl or even physically punishing it with a nip.  I see this turf control when we have dogs in the house, as well.  Much of the dog-on-dog communication that I observe is based on punishment, or avoiding it.  It does not rely upon seeking rewards from other dogs.  Certainly, when two or more dogs are playing it appears that they are “getting” something good from the other dog, but the obvious communication between dogs appears to be based on negative reinforcement, not the “all positive” training that is so in vogue today.

Along with playing games of “keep-away”, “chase-Me!” or “I’m gonna chase you”,  dogs spend a lot of their time establishing rules, enforcing them, avoiding getting punished by another, making peace to avoid getting punished by showing subordination, and peace keeping behaviors that usually include measured “corrections” for the dog that steps over the line.  Dogs understand that language.  It makes sense to them.  It works for them.  They appear happiest when all the players know the rules and play by them.   So, I use that approach in my dog training.  It makes sense to me to do so. 

I want to be my dog’s ultimate leader, not a hotdog dispensing machine.  Using their own language that is heavily based on providing rules, setting boundaries, enforcing those lines, correcting the negative behaviors when necessary with fair and measured feedback; those approaches go a very long way in establishing a comfortable environment for the dog.  Most “behavior problems” fall into the “extinguish an existing behavior” column, not the “create a new behavior” bucket.  So, using a correction based method makes sense, is fast, fair, and highly understandable for the dog.  Once the dog recognizes me as his very faithful and competent leader, learning of the other behaviors or tricks (where we can use treats, toys or even lures and bribes) goes quickly and smoothly.

© 2005  Tammie Rogers - all rights reserved.   For permission to reprint email Tammie.