A Short History of Dog Training

People tend to think the way things are done now is how they were done forever. Not entirely so, especially in the arena of dog training.

A few key dates in the history of canine communication show that while operant conditioning is pretty old stuff, the mix of methods has changed and become better understood over time:

_ _ _ _ _ _ __

1700s: Truffle hunters learn to give their dogs bread when they locate truffles, which turns out to be cheaper than using pigs which cannot be stopped from eating all the truffles they locate.

1885: S.T. Hammond, a writer for Forest and Stream magazine advocates in his columns and in a book entitled Practical Training, that dogs should be praised and rewarded with meat when they do something right.

1880s: Montague Stevens trains his New Mexico bear dogs by rewarding them with pieces of bread instead of beating and kicking them as others of that era were generally doing. Stevens is a famous bear hunter and friend of Teddy Roosevelt and the sculptor Frederic Remington.

1886: Edward Thorndike develops a theory of learning based on stimulus and response. Thorndike shows that "practice makes perfect" and that if reinforced with positive rewards, animals can learn quickly.

1899: The first canine school for police dogs is started in Ghent, Belgium using Belgian Shepherds, which had recently been established as a breed.

1903: Ivan Pavlov publishes his experiments with dogs and digestion, noting that animals can be trained to have a physical response to stimuli. Pavlov called this learning process "conditioning," and in 1904 he was awarded the Nobel Prize for his research.

1903: The Germans begin schutzhund work, a competition devoted to obedience, protection, tracking and attack work.

1907: Police begin patrolling New York City and South Orange, New Jersey with Belgian Shepherds and newly reconstructed Irish wolfhounds.

1915: Baltimore police begin using Airedales from England to patrol the streets. The police suspend use of Airedales in 1917 as the dogs had helped make no arrests. The police failed to notice that no robberies had occurred where the dogs were on patrol.

1915: Edwin Richardson trains dogs for the military during WWI using some positive reinforcement, and the dogs prove to be quick studies. Many dogs are used for communication and for guard duty.

1917: The Germans begin to formally use dogs to guide soliders blinded in mustard gas attacks. The French soon follow suit.

1918: U.S. Army Corporal Lee Duncans find an abandoned war dog station in Lorraine, France which has five young puppies in a kennel. Duncan takes one of the pups and names it "Rin Tin Tin" after the finger dolls that French children were giving to the soldiers at the time. The dog travels to California, proves easily trainable, and is soon employed making movies that are so successful it saves Warner Brothers studio from bankruptcy. The dog dies in 1932 in neighbor Jean Harlowe's arms, and is buried in Paris, but its descendents work in the movies throughout the 1950s, inspiring many people to try to train their own dogs to do simple tricks.

1925: One of the very first German-trained guide dogs for the blind is given to Helen Keller.

1926: Propelled in large part by the popularity of Rin Tin Tin, the German Shepherd population in the U.S. explodes, and by 1926 it accounts for 36 percent of all the dogs in the AKC -- 21,659 animals. Due to rapid inbreeding and poor selection, however, the American German shepherd quickly degenerates and is soon deemed inferior.

1929: Dorothy Harrison Eustis establishes the Seeing Eye Foundation to train guide dogs for the blind. Eustis goes to Switzerland to get a better stock of German Shepherds than she can find in America. This same year the AKC tries to ban the importation of foreign purebred dogs in order to protect domestic dog breeders, but the plan fails.

1930: About 400 dogs are employed as actors in Hollywood, the majority of them mongrel terriers which prove to be small enough for indoor scenes, rugged enough for outdoor scenes, and exceedingly smart.

1938: B.F. Skinner begins research into operant conditioning as a scientifically-based learning model for animals and humans. His special focus is on teaching pigeons.

1939: The AKC begins obedience competitions designed by Helen Whitehouse Walker who wants to prove that her standard poodles can do something other than eat food.

1942: The U.S. military says it needs 125,000 dogs for the war, and asks people to donate their large breeds. The military manages to train only 19,000 dogs between 1942 and 1945. The Germans reportedly had 200,000 dogs in service.

1943: In 1943, Marion Breland and her husband Keller Breland form a company called Animal Behavior Enterprises (ABE) to teach animals for shows. The Brelands had been students of B.F. Skinner (see 1938) and began teaching animals to peform tricks for shows and for commercial clients such as dog-food maker General Mills. They pioneer the use of a "clicker" to teach animals at a distance and to improve timing for affirmations and delayed rewards. The Brelands were the first people in the world to train dolphins and birds using operant conditioning.

1943: The movie "Lassie Comes Home" is filmed, featuring a purebred male collie playing the female staring role. Ironically, the U.S. military considered purebred (i.e., AKC ) collies so stupid that they were specifically excluded from military service in World War II, while herding farm collies were actively recruited.

1947: The Brelands (see 1943) begin using chickens as learning subjects with which to train other trainers, as they are cheap, readily available, and "you can't choke a chicken."

1953: Austrian animal behaviorist Konrad Lorenz writes "Man Bites Dog" and "King Solomon's Ring," books which popularize animal behaviorism.

1954: Baltimore reestablishes its police dog program, and today it remains the oldest police K-9 program in the country.

1960s: During the early part of the 1960s, Marian and Keller Breland (see 1943) were hired by the U.S. Navy to teach other animal trainers how to train dolphins. The Navy was interested in using dolphins to patrol harbors, retrieve lost gear, and guide bombs (i.e. "suicide bomber" dolphins). During their work with the Navy, the Breland's meet Bob Bailey, the Navy's first director of animal training, and they began a partnership with him. Keller Breland dies in 1965, and in 1976 Marian and Bob Bailey are married.

1962: William Koehler publishes "The Koehler Method of Dog Training" which becomes a staple of AKC obedience competitors. Though often criticized today, Koehler's methods are the core of a lot of effective dog training systems still in use.

1970s: The U.S. Customs Service begins to use dogs to detect drugs, and they are subsequently employed to sniff out explosives and fire-starting chemicals.

1978: Barbara Woodhouse publishes "No Bad Dogs" one of the first popular books on basic dog training. It relies heavily on proper use of a choke chain, and says most "bad dogs" have inexperienced owners who are not training their dogs properly by being consistent, firm and clear.

1984: The U.S. Dept. of Agriculture begins to use beagles to patrol airports for contraband food and other perishable items.

1985: Dolphin trainer Karen Pryor publishes Don't Shoot the Dog: the New Art of Teaching and Training which focuses on timing, positive reinforcements and shaping behavior, and draws heavily on the work of Marian Breland Bailey and Robert Bailey (see 1943 and 1960s). Her book promotes "clicker training" of dogs to improve timing and to allow trainers to communicate and "reward" their dogs from a distance.

1995: The U.S. Dept. of Agriculture begins using Jack Russell Terriers to locate and kill invasive brown tree snakes on the island of Guam.

2000 and beyond: Various cable television shows feature various dog training and rehabilitation methods. The notion that there are "new" and "old" dog training methods obscures the fact that ALL dog training methods involve some form of operant conditioning which is, in fact, pretty old stuff (as old as dogs). None of the dog training shows actually explain the core principles of operant conditioning or their relative worth in different training situations.
.

Mountain's New Collar

 

Mountain's new e-collar came in, and it did not taken her very long to figure out that she can still hunt around, but she cannot barrel for Jesus down the hills to see if *maybe* a deer might be down there in the breaks. 

It was interesting watching her figure it out -- and how quickly she put it together. A little simple consistency, some use of the warning beeper, food treats whenever she did it right, and a few surprisingly low-volt "nicks" from the e-collar, and we seemed to be on a program for change as far as wide hunting was concerned. 

For the record, I started the collar off by putting it on myself, and at "1" I could barely feel it. At "2" I was aware of getting nicked, but it was not very bad and it was no big deal if my son wanted to juice me on the hand. It turned out that "2" was all it took for Mountain to get the idea, though I did turn it up to "4" once when I spotted a feral cat crouching in the grass waiting for Mountain to pass by. Mountain did not see the cat as she was very focused on doing the right thing with me, and so I was able to simply turn around and walk another away and all was well for cat and dog alike. All good! 

The Innotek e-collar I bought appears very well made, comes with a charger, a self-test, and -- most importantly of all -- a CD-Rom video on how to use the collar.  

E-collars are simply a tool for operant conditioning, same as food, a leash, a collar, and the human voice. The only benefit of an e-collar is that it works long distance, which is very useful if you have a strong prey-drive in a very smart hunting dog as I do with Mountain. Mountain is a hard nut, and this looks like a wrench that will work. I don't think I will have to use it on her too long -- dogs generally learn quickly once you get the right communication stream working.

How did I know I had the right collar? Simple -- I carefully researched them, checked all the features, price-shopped ... and then got the one that happened to have a picture of a Jack Russell on the box. Sometimes the purchaser and the marketing department agree!

Training Children Like Dogs

Click here to play part of the SouthPark Dog Whisperer episode


One of the core messages of "The Dog Whisperer" is that dogs are not children. The converse can also be said, but, in fact, most over-indulgent suburban parents could learn quite a lot about raising children by studying basic operant conditioning -- the kind of thing you learn in dog training.

In the long version of the Southpark program Cartman sends two TV reality show "Super Nannies" over the edge. Why had the Super Nannies failed? Well, in Southpark it's because the nannies tried to reason it all out and because they really had no serious consequences to offer -- a "time out chair" stops working at about age six.

With nowhere else to turn, Cartman's mom calls Cesar Millan, "The Dog Whisperer."

As in the National Geographic TV show, Millan is all about "dom-ee-nance." Millan says the problem is that the child (Cartman) is leading the adult, and it needs to be the other way around. Or, as Millan puts it, "This walk is about me, the child is merely lucky enough to come along."

Perfect!

Watch as Mrs. Cartman get her life back, and as Cartman learns that bad behavior has bad consequences.

In the real world of dogs, kids and adults, the rules are really not that different: "Do good, get good. Do bad get bad."

Really, isn't that what both job performance bonuses and the criminal justice system are all about?

If you watch the long version of this South Park episode (no longer on You Tube, but look around in Goolge Video, as the world will be free) you will notice that Cartman turns his life around, but at the end that Cartman's mother reverts back to enabling behavior as soon as Cesar Millan is out of the picture.

Even in the Southpark version of the "Dog Whisperer," the realities of the human condition intrude. That's good cartooning!

.

A Balanced Trainer for Unbalanced Dogs

Asking me my thoughts about dog training is a bit like asking a rodeo clown to bring in the herd: I might know my way around a cow, but I might not be quite what you need for this job.

And so, it was with some confusion that I got a call from my mother asking me what I thought of the article on Cesar Millan (the so-called "Dog Whisperer") in this morning's New York Times.

I had not see it.

A bit later, I got an email from Reid Farmer asking, "Who's gonna jump on this one?"

Well hell, I guess I will. If deep knowledge and expertise are required to type, there would not be too many typing classes.

So here goes. What does this New York Times piece criticizing Cesar Millan and "The Dog Whisperer" TV show actually say?

Here's the meat of it:

"Essentially, National Geographic and Cesar Millan have cleverly repackaged and promoted a simplistic view of the dog's social structure and constructed around it a one-size-fits-all, cookie-cutter approach to dog training. In Mr. Millan's world, dog behavioral problems result from a failure of the human to be the pack leader, to dominate the dog (a wolf by any other name) completely.

"While Mr. Millan rejects hitting and yelling at dogs during training, his confrontational methods include physical and psychological intimidation, like finger jabs, choke collars, extended sessions on a treadmill and what is called flooding, or overwhelming the animal with the thing it fears. Compared with some training devices still in use whips and cattle prods, for example these are mild, but combined with a lack of positive reinforcement or rewards, they place Mr. Millan firmly in a long tradition of punitive dog trainers."

Aha! Now I get it. Millan is a "bad and backward" punitive dog trainer! The author of The New York Times piece slyly manages to work in whips and cattle prods to suggest Millan is one of those types of trainers.

But wait a minute! What Mr. Millan actually does is simply use an ordinary choke chain as it should be used, a firm voice and a firm hand to get a dogs attention. There are no beatings of dogs here. This is the National Geographic Channel, for God's sake!

The over-educated yuppies who watch this channel do not have a problem being too coercive. Their problem is that they think of their dogs as human children and they are willing to have themselves and their children bitten rather than try to communicate with the dog as a dog.

Whatever a dog is, it is NOT a human child.

But, of course, this is news to a lot of folks that watch the National Geographic channel, Animal Planet, and public television. These are nice people who love their pets. Most of them have sat on the couch so long watching nature shows they think real-life lion kills in the Serengeti are narrated, sotto voce, by the same voice that announces par scores on the Golfing Channel.

But no matter how nice or over-educated you are, a dog is not a child. This is a particularly uncomfortable and threatening truth for women who have managed to displace their maternal instincts to the family pet. This is a point Millan does not address directly (please, not too much honesty Millan!) but he does suggest women often have a harder time asserting dominance over their "fur babies." He is right. Why does saying this obvious thing send Mark Derr clucking away like a hen? Anyone who works with dogs has seen the maternal displacement problem in action, and anyone who has seen the "Dog Whisperer" TV show has seen the problem repeated out again and again. The Daisy Fuentes episode was particularly memorable for some reason ....

The simple but harsh truth is that the psycho-demographic watching the National Geographic channel tend to be people with two types of common "dog problems": They think their dog is their child, and their dog is over-fed and fat.

The dog is, quite simply, being "loved to death."

Cesar has antidotes for both problems. He is the right wrench for this nut.

If Mark Derr, the author of the New York Times editorial, wants to put together a TV show about dog training his way, he is free to do it. But it will not be a show that makes it on air unless it is entertaining. And dramatic. And has a charistmatic human figure as the lead.

Can Mr. Derr supply that? I doubt it. Cesar Millan can.

What National Geographic has done, as far as I can tell, is nothing more than film a charismatic and smart dog trainer at work and then cut up the tape so that it makes for good television. What's wrong with that? Was the boring stuff left on the cutting room floor? Of course. Who wants to see two hours of click-and-treat? Not me!

Derr seems to take issue with Cesar Millan's admonition that dogs do best with "exercise, discipline, and affection," and he seems to disagree with Mr. Millan's definition of discipline as meaning "rules, boundaries, and limitations."

Derr is being stupid. No serious dog man would argue with Millan on these points. The only thing you should say after Milan's statements on these issues is "Of course." Or perhaps, if you are being a bit chatty, "It is also true for children."

Next point.

Derr goes on to criticize Millan because Millan talks of dogs as being primarily pack animals with an innate desire to have a leader. Derr says wolves are not what Millan says they are because he read a study once that says differently.

Hmmm. . . .

I would agree that a dog is not a wolf (I say exactly that in my own book on working terriers -- see Chapter One), but I would NOT say a dog is not a pack animal.

A dog IS a pack animal.

On this matter, I speak from experience -- experience that Mr. Derr may not have. You see, I have been attacked by a pack of feral dogs. I assure you it is not something you forget.

At the time I was attacked I had been walking solo in the woods for eight weeks. I smelled rank, and with a 60-pound pack on my back, and a rain fly over my pack, I am not sure the mixed pack of large farms dogs recognized me as a human.

I smelled wild, and these were deer-killing dogs that had decided to put me on the menu. I was lucky to "sort it out" the way I did. My pants were ripped through at the crotch by an attacking dog who managed (by miracle) to only find wool before he was batted off into a ditch. Yes, one dog died. There are no regrets here.

So when Derr says dogs are not pack animals he can take a hike (pun intended). I know better, and what I know I did not read in a book.

Derr correctly notes that some aggressive dogs have underlying genetic problems predisposing them to being aggressive.

Yes, that is true. But guess what? So too do humans. We lock up a lot of humans, but we assimilate, educate and train most of them to "do the right thing". A genetic load for a little aggression is not necessarily destiny. It should be a lot easier to deal with dogs than humans. Dogs do not have beer, whiskey and PCP to warp their minds.

Huge numbers of dogs are written off, not because the dog is a genetic failure, but because the owner is ignorant (and sometimes stupid) and has done a poor job of socializing, training and leading a perfectly fine animal as it needs to be lead.

Is there a place for Prozac in your dog's food? There is according to Mr. Derr. I would disagree. Most dogs need training. If a dog is so unremittingly aggressive that it needs to be medicated, it needs to be put down. There are too many fine dogs being put to sleep every day for want of a good home to waste food and space on keeping a genetic wreck alive and medicated with chemical clutches. Dogs are not humans! We are not talking about YOU. Get over it. The only REAL question is whether the dog is actually the problem. Cesar Millan says it is more likely the owner -- and he is probably right.

Mr. Derr's shock and dismay at mild coercion during dog training suggests he is a dog training fadist. We see these people in the world of dogs just as we see them in the world of food and fashion.

Some of the most extreme fadists are the "clicker training" folks. For a cult-like subset of this group, Karen Pryor's book, Don't Shoot the Dog is a virtual Bible, and clicker training their sole religion.

I am not against clicker training (quite the opposite!) and I am all for positive reinforcement. I will bet Cesar Millan is not against positive reinforcement either.

That said, pure positive reinforcement is NOT how dogs communicate to each other, is it?

It may come as a shock to some people to learn that Karen Pryor did not invent clicker training. It was invented by the late wife of legendary animal trainer Bob Bailey. Bailey was on an animal training list-serv that I was on. Both of us were quiet lurkers, but one day Bailey popped up to set one "pure positive" person right on his ass. Mr. Bailey wanted it known to the list that he himself was not a "clicker trainer." He used clickers, sure, but he also felt there was a place for mild coercion. He even thought there was a right time and place to "shoot the dog" when dealing with extremely dangerous animals. Not a peep was heard after that.

Back to Cesar Millan. One of Millan's core points is that people with dogs are confused about what they have at the other end of the leash.

Yes, a dog is not a wolf (even if both animals DO organize themselves on a pack model), but a dog is also not a CHILD.

A dog is a dog.

Millan's core training message is that we must communicate with dogs on their level and with their values, and not our own.

Yes, dogs value food and affection.

But dogs also understands teeth, power and dominance.

Mr. Millan's point is NOT that dogs need to be bitten or beaten to get them to behave. Mr. Millan is a very civilized trainer and not a violent man. His point is simply this: almost every dog has within it the capacity to be submissive to a true leader.

Puppies are submissive to larger dogs, and smaller dogs are submissive to larger dogs. In a pack, there is only one Alpha dog and one Beta dog, and there is a "pecking order" all the way down to the Omega.

This is NOT news to anyone with a few free-range dogs. The order I release my own dogs in the morning is determined by pack order. If I do it wrong, the dogs squabble. If I do it right, all is harmony. The same is true for feeding.

Is dominance and coercion the way you train a "sit-stay"? Well, actually, yes and no. It's not quite that simple. Click-and-reward is the core method of training a dog to do a trick, but a dog that has been over-fed is not very attentive is it? A dog that has been caged or crated and not given much excercise is a bit too bouncy to focus on you, isn't it? And if a dog is going to learn anything it needs a calm, assertive and not-too-verbal person who consistently does the same thing over and over again. In fact, this is exactly what Cesar Millan offers and when he teaches -- along with a good dose of "Your dog is not your child," and "this is a choke chain -- learn how to use it."

And so I come back to a question I had after reading Mr. Derr's piece. What was Mr. Derr's intention? Did he have a point?

In fact, I think he did. I think his point is that he hopes New York Times readers will read his own book. There is nothing wrong with that, but it is not a good reason to slap someone around with poor thoughts and thin facts. I do not know much about Mark Derr, but he does not seem to know much about dog training. Even less than me, and I am a basket case.

And, for the record, I have read Mark Derr's book. It is a canine history book, and what I remember is that it was not terrible, but not great. The most memorable bit of information I got out of it is that Lewis and Clark ate a lot of dog meat on their expedition, and one of them (I can't remember which one) actually preferred dog flesh to elk or deer.

Interesting, but not quite as useful as knowing how to put a choke collar on a dog properly. Barbara Woodhouse taught that useful lesson to a generation of public televisions viewers some 30 years ago. Cesar Milan is now teaching it to a similar psycho-demographic on The National Geographic Channel. Some things will always be needed.

.