In looking at the current landscape of dog training it seems there are multiple training paradigms, all claiming to be the best or most effective at getting your pooch to behave.
In the olden days when a person depended on his dog(s) for hunting assistance, protection, assistance with herding and flock protection, or in some cases even transportation a person’s life might quite literally depend of how well trained their dog was. What training techniques did they use? Were they very different than those we use today?
wild guess, but I think training a working dog today and 2000 years ago are close to the same. This is of course depending on what work the dog is supposed to be doing.
the nature of people and animals don´t usually change that much in such short time.
In the past dogs weren’t trained much at all. Working dogs were always raised with other working dogs, usually their parents. They learned how to work by working alongside the other dogs. Mistakes were met with punishment, good behaviour with reward. It was really that simple. Most people never trained their dogs at all to any great degree.
The idea that a person’s life depended on having a dog trained to the Nth degree doesn’t have much basis in reality. A good dog certainly made life much easier, but life really wasn’t that knife-edge. A herd dog that was 10% less effective didn’t really cost anybody anything, it just meant it took 2 hours to finish the job rather than 1 hour and 50 minutes. A sled dog that was 10% slower had much the same effect. A hunting dog that was 10% less effective meant that prey escaped 88% of the time rather than 80%, but how would the owner ever measure that?
In most cases dogs were expected to fulfill multiple roles anyway. A sheepdog didn’t just have to pen sheep, it also had to pull down a lambing ewe out in the paddock, act as a guard dog and occasionally function as a hunting dog. Hunting dogs were expected to track, point, bail, bite and guard and do all that on a dozen different prey animals. Because of that there was no single “best” dog or best training technique. One dog could be proficient in all those things, another expert in one and marginal in the rest and both dogs would be highly useful.
Additionally every owner had different requirements. A good sheepdog working just sheep on flat ground had very different requirements from a dog that was expected to work sheep and occasionally work cattle in hill country.
As such there was there was no real way of knowing that a dog wasn’t working at peak. It was only with the introduction of competitions that people were able to judge which dogs were working the best. However competitions are not the real world, and the best competition animals are rarely, if ever, the best real world animals.
Which is why training wasn’t a big part of a dog’s life in times past. People wanted dogs that got the job done. Dogs learned from other dogs. Dogs that wouldn’t or couldn’t learned were killed and replaced and so the breed improved. This modern penchant for training dogs to a fine degree is very much a function of the modern world where people have lots of leisure time and dogs have no practical use.
When could we neuter dogs? Wouldn’t that have an impact on trainability–I know now when a dog is problematic, trainers often suggest neutering. But what happened before you could? Or did people just…um, cut them off?
First off, neutering dogs only addresses those problems directly associated with sexuality. Things like wandering or aggression in males. It doesn’t make them more tractable, in fact because it decreases sex drive it makes many breeds less tractable. With no drive for assertiveness or aggression many castrated males become effectively impossible to train well for many uses.
Secondly, castrating animals as an ancient training technique obviously never occurred, since it makes it impossible to breed form the animal.
romans seemed to train dogs and even wild animals for show in the colliseum.
the thing with limited markets (trained dogs being one such) is the preponderance of trade secrecy. that could go both way: methods change every year or onnce every thousand years.
For a while I was collecting books on dog training - old Will Judy books, hunting and herding books from the early 1900s.
Most common approach (and this mirrors what I learned in the early-mid 1980s) was very aversive techniques coupled with positive reinforcement to clearly illustrate this = good and this = bad.
The pendulum has swung recently to “purely positive” reinforcement which doesn’t work well with all dogs. Main thing to keep in mind: there is no such thing as one training paradigm that fits all dogs. Dogs are not a single mass, like a fungus. They’re individuals. What works for one won"t necessarily work for the next.
I’m confused. Dogs “learned how to work by working alongside other dogs,” such as their parents? Who trained the parents? I"m assuming SOME training had to take place somewhere along the line, right? It couldn’t all be instinct, could it? Even shepherding dog breeds need to be trained in the basics of what to do, especially if there were different requirements based on topography, etc. Of the shepherd breed dogs I’ve known, I could see a definite herding instinct, but none of the dogs I’ve known would instinctively pull down a lambing ewe in a paddock. But maybe I’ve been running with the wrong pack, so to speak.
From a handful of recent sources I’ve read on training: negative reinforcement can be used to stop unwanted behavior, but the downside is that, in a situation where stress overtakes their fear of punishment, they’ll revert. You can keep a dog from biting by punishing, but that dog will be more likely to bite when stressed than a dog who’s been positively taught other stress responses.
Negative reinforcement doesn’t have to be a kick in the rear, though – more “mild”
negatives like a spritz with a squirt bottle or shaking a can of pennies are also negative reinforcement. The latter was really, really useful in helping my wife and I curb our puppy of a nasty barking habit. Once we got him to stop barking constantly, we were able to redirect him to better behaviors.
Actually, that’s a positive punishment. Negative reinforcement refers to when you try to encourage a behavior by taking away something that the subject views as bad.
Their parents. And they were trained by their parents and so forth, right back to the first commensal wild wolves.
Almost all the behaviours we use in working dogs are simply wolf instincts that have been cut short, usually curtailed at the point of actually killing the animal.
The thing is that there have been at least 5, 000 generations of training occurring, simultaneously with very strong selective (albeit unintentional) breeding.
As I noted earlier, the training largely consisted of praise when a dog did the right thing, and punishment for doing the wrong thing. Amongst most herd dogs even today this is the standard. If a young dog bites an animal at the wrong time or turns them in the wrong direction, somebody, often several people, throw rocks at it. That is about the extent of the training the receive. If a dog persists in that behaviour it is shot, and a replacement is found from the next litter.
If you multiply that over 5, 000 or more generations of dog it is quite effective. But the amount of training that any individual dog received was maybe one hour in its entire life.
The reason why that worked is because dogs were never expected to do more than they could do. Wolves were never originally expected to herd sheep. They were at best expected to make them run in a straight line. With generations of training they became capable of grabbing and holding sheep, and that trait was then taught to the following generation, which in turn meant that over time the breed became capable turning sheep and so forth. But nobody every woke up one morning and taught a dog to do this. Dogs largely taught other dogs. Humans mostly just threw rocks at the dogs doing the wrong thing.
You must have been. Grabbing sheep and holding them down is almost universal in all sheepdog breeds. It’s something that many, possibly most, dogs have to be actively trained out of doing. That’s hardly surprising, since it’s only been in the last few hundred years that that stopped being their primary purpose. Until a few hundred years ago people didn’t yard sheep to shear them. They simply had the dog pull them down one by one on the paddock, and sheared them where they fell. That’s still the standard practice over much of the world.
Training a dog to pull a sheep down isn’t hard. It’s training it to only do it on command that’s tricky, but even that is simple if the animal has another trained dog to learn from. Dogs learn really fast from other dogs.
Perhaps it is a mistake to bring up Ceasar Milan in GQ. Leaving out my opinion. He is not part of the trend toward positive methods. He is roundly disliked by those that do use positive methods. I don’t want to discuss him methods, just point out an exception.