The same personality in two different dogs can be made up of completely different puzzle pieces. But every time you work out a piece you have another tool in the training toolbox that you can use to communicate and work better with the dog and so hopefully help the dog achieve a higher level of discipline in training.

I used to sit in my Sam's Foundation course and wonder why oh why were we asking our dogs to 'watch me' and have them gaze into our eyes for minutes on end, or why it was important to see how many 'downs' our dogs could do in 60 seconds, I really couldn't see the point but accepted at face value that it was important to do, and filed it away to see if I could work out what it was about later on. With the 'watch me' command Sam's attention span started at under 3 seconds, but it's worked gradually up to a reasonably reliable minute. The 60 second 'downs' forced me to become an effective motivator, something I wasn't at all good at before.

In reflecting on why such exercises were important, eventually two things about generalised dog training really clicked:

- Training is almost all about keeping the dog's attention. If my dog doesn't come when he's called its because something more interesting distracted him from my call. That's one way to look at it anyway.

- 'Shaping' is the best word I've come across to describe the way the training process is supposed to work, the idea of raising criteria in achievable steps, using intermediate behaviours as steps along the way, working at the learner's pace etc. I think it's so important to follow the principles of shaping, no dog can learn a complex or demanding behaviour in one step, but the behaviour can be worked-towards over time and it's amazing what can be done. Sometimes the steps along the way seem very far removed from the goal, but they're just used as an appropriate stepping stone.

I think most people have experienced the really gratifying and directly tangible aspects of training when they get their dogs to do 'tricks' like sit, shake, down, stand and so on. At those times it seems so direct and the training process is so clear - stimulus, response, reward. It took me a very long time to see that through training I could actually have an effect on much less tangible aspects of my dog's behaviour, like emotional responses and general personality. The method for training these less tangible behaviours is still the same, but the exercises to train for are often very obscure, not at all obvious.

I don't know if everyone would agree that personality can be trained (partly or at all?), but that aside I describe the intangible parts of the training process this way: To teach a dog to heel, the best way isn't to start by going for a walk and saying 'heel' and rewarding the dog whenever they're in the correct position (though that is the ultimate goal), mainly for the obvious reason that heeling and sniffing-exploring are mutually exclusive so of course the dog chooses not to heel. Heeling is too demanding a behaviour to teach in one step, so most training courses teach behaviours like getting and keeping the dogs attention first and step-sit, heel-turn, sit-stay, 60 second sit, 60 second down, sit-in-front, sit-by-side, weaving and general targeting exercises as ways to shape individual behaviours that form part of heeling. At the end owners are taught to lure their dogs to a heel position by holding food in front of their nose to position properly by their side, but as far as I can see this is really just an operant conditiong step so that a cue can be associated with the behaviour that is already occurring. Once the cue is learned ('heel') the original method of rewarding when the dog is in the right position can then be used to shape the position more precisely. And it's all done off leash. Again, this is all just one way to do it, but who'd have thought that training for something as non-obvious as attention span or weaving between the legs could pay big dividends when trying to teach heel. A lot of training seems to be like that - taking very obscure steps to help rapidly produce a final behaviour.

To try to manage Sam's intense and domineering personality my aim was to build up an arsenal of training tactics I could use to ultimately eliminate his predisposition to dog fighting. The squeaky ball I use is great for his recall, bar-open/bar-closed is great for shaping his competitive anxiety, and a good attention span is great for pre-empting low grade conflicts. The interesting thing to me is that none of the tactics actually address the act of fighting itself. (It's also important to say that if all else fails - and it occasionally will - it's nice to have the fallback tactic of hauling Sam out of a fight by the tail or hind legs. )

In a nutshell, I don't believe dog-dog aggression can be trained out / eliminated just by giving out treats in a park, or even by removing treats from the park situation entirely, or for that matter by any one solution. To my mind there's much more to it, and like all training, working up to eliminating all dog-dog aggression is a process of shaping that starts by working on exercises that may seem a long way removed from the park. It's also worth saying that there is no single cause of dog-dog aggression, or even a finite list, and to this day I still haven't pinpointed all of Sam's myriad potential triggers. (The ones I have found I have had a lot of success changing though, I can detail some of them.) The number of behaviours involved, and the complex mechanics of each individual training exercise, is really what's so hugely challenging and problematic about it all!

I still say though it's worth trying - and should be, in fact, fun to try. I love dog training!