Friday, 24 January 2014

Double Effect and "Purely Positive" Training

Some people argue that there is no such thing as �purely positive� training, because even 100% R+ trainers �use� aversives.  This makes them no better than P+ trainers because in both cases, the dog is suffering during training.  They give examples of dogs �in drive� and so desperate for a reward that withholding it is aversive.  A dog with an incredibly high prey drive and low self-control, for example, might shake, cry, and show classic signs of stress and frustration when deprived of the chance to chase a toy.  If chasing a toy is the reward in an R+ training session, then the dog is not rewarded for non-performance of a desired behavior, so for these kinds of dogs are arguably going to suffer during such a session.  

There are a few things wrong with the practicalities of this scenario, but let�s go with it as a way to test whether the conclusion - that the R+ trainer�s use of aversives makes them �no better� than a P+ trainer - is justifiable.  Although there have already been some excellent discussions of this issue, I want to show how academic ethics can give dog trainers a different set of resources to make the argument. 

Trolleys and Bridges

The Doctrine of Double Effect is a popular idea in academic ethics.  It states that it is permissible to allow harm to happen if it is an unavoidable side-effect of a good action.  The classic example goes something like this:

You are standing by a train track, next to a lever that switches the points from the track on the left to the track on the right.  You notice that a train is coming, and that there are five workmen on the right track, but also one workman on the left track.  Do you switch the points?

Philippa Foot argues that it is permissible to switch the points, because five people dying is worse than one, and - crucially - you�re not killing one person in order to save five people (Foot 1967).  Judith Jarvis Thompson contrasts Foot�s example with this one:

You are standing on a bridge, next to a very overweight man.  You notice that a train is coming, and that there are five workmen on the track in front of you.  There is no way to warn them, but you know that pushing the overweight man onto the track will stop the train and save the workmen.  Do you push the man?

Thompson argues that it is not permissible to kill a man deliberately in order to save others; it isn�t right to use someone�s body like this.  The difference is in whether the death (a moral wrong) is an intentional, or integral part of the action.  In the Trolley Problem, the deaths of the one workman wouldn�t change the fact that five workmen would be saved by switching the points.  In the Bridge Problem, the death of the overweight man is an integral part of saving the five workmen.  Double effect states that it is permissible to do an action with harm as an unintended consequence, but not to do harm as an intentional and integral part of an action.  So, it is okay to flip the switch in the Trolley Problem, but not to push the man in the Bridge Problem. 

How This Applies to R+

The mechanism of operant conditioning that underlies R+ training works whether the dog becomes distressed when the reward is withheld or not.  Something positive being added to a dog who is already happy will work as a reinforcer.  The circuit doesn�t depend on there being any stress or frustration; if it happens that the dog does suffer from a bad lack of self control that causes him to get upset at the absence of a reward, this is something the R+ trainer should try to minimize.  Note that I'm not suggesting that all frustration should be completely avoided; some is unavoidable, especially when we're working with high drive dogs doing the thing they love.  Frustration, if kept to a minimal level, can actually provide additional motivation (Amsel and Roussel, 1952) and removing it entirely from the picture, whilst ethically desirable, is often practically unworkable.

The critic of the R+ trainer is assuming that the R+ trainer�s methods work like the Bridge Problem, where the death of the overweight man is an intended part of saving the five workmen.  If the stress was needed for the training to work, then this would be true.  But in fact, the R+ trainer�s methods work like the Trolley Problem, where the death of the one workman is neither intended or necessary for saving the five.  The R+ trainer�s ideal situation would be a dog who didn�t feel stressed at the prospect of his reinforcement being withheld, since the stress isn�t necessary or intentional.  

So, there is a moral difference between training methods that intentionally use aversives, and methods that have stress or frustration as an unintended, unnecessary side-effect.  It�s not true to say that there is no such thing as a purely positive training method, even though it might be true that some real-life training situations are less than purely positive. This is something R+ trainers should be working to avoid.  

References and Notes

  • Foot, Philippa (1967). "The problem of abortion and the doctrine of double effect". In B. Steinbock and A. Norcross (eds), Killing and Letting Die. NY: Fordham University Press. 
  • Thompson, Judith Jarvis (1976) "Killing, letting die and the trolley problem". The Monist 59, pp 204-17.
  • Amsel, A and Roussel, J (1952) "Motivational properties of frustration: effect on a running response of the addition of frustration to the motivational complex". Journal of Experimental Psychology 43(5), pp. 363-368
  • General notes on drive and frustration in dogs:
  • Another way to approach the fallacy of "purely positive training":


Thursday, 16 January 2014

"Results Based Training" - Does the End Justify the Means?

The Question:

I sometimes hear it as a rebuttal against trainers who represent themselves as purely positive, or as adherents to a particular theory. 

�I�m not a positive dog trainer, I�m a results-based trainer.  I do what works.� 

What does this mean?  

Means and Ends

One way to see the results-based trainer is as making use of the old adage, �the ends justify the means�.  At first glance, this looks like a way to bypass ethics in favor of pure pragmatism, as if it could be rephrased as �Never mind what�s right or wrong, what matters is that I achieve my aim in training�. Seeing the results-based trainer that way would be unfair, however, it would characterize her as amoral or concerned with suffering. 

Even Machiavelli, to whom the phrase is most commonly attributed, claimed that if cruel acts were ever necessary to preserve a State, they ought to be as swift and minimal as possible, meaning that he was committed to the idea that these acts did need some external justification.  So, the principle of charity dictates that we can't see the results-based trainer as trying to sidestep the demands of morality by appealing to pragmatism.  Instead, we have to understand her as making a specific claim about morality.  

Purely in terms of ethical theory, �the end justifies the means� can refer to consequentialism, the theory that what happens as the result of an action, rather than its motive, determines whether an action can be called good.  Practically, �the ends justifies the means� can refer to the claim that it is permissible to do something a little bad, in order to achieve a much greater good.  This is the most pertinent application for dog trainers.  In particular, it informs the rationale of many P+ trainers.

Or - and here we return to the theoretical understanding of the phrase -  it can mean that the use of an aversive like a shock collar cannot be classed as harm because it is necessary to achieve the goal. Something harmful is only morally wrong when it�s true that we shouldn�t do it.*  This is the most problematic interpretation, but there is some precedent.  We don�t see the vet as harming our dogs when she operates on them, even though she is causing pain.  The operation is for the dog�s greater good, and this takes away the moral badness of the pain.  Of course, we expect the vet to do everything she can to minimize any pain - unnecessary suffering is a morally wrong harm, but necessary suffering is not.  If the pain of using an aversive is both unavoidable and in the dog�s best interest, then the pain cannot be classed as a morally wrong harm.  

The results-based trainer could be seen as claiming that causing pain to a dog is permissible only so long as it successfully teaches the dog to do what the trainer wants, because what matters in training is results first. This runs into the requirement to justify the necessity of the use of aversives.  

The Burden of Proof

Assuming that an R+ and a P+ method are equally effective, there is a burden of proof on the P+ trainer to demonstrate that the harm is a necessary element of the training.  She must show that there is no way to avoid a small amount of harm if we are to achieve the end we are seeking.  Like the vet, the P+ trainer must show that whilst she is doing everything she can to minimize suffering, some aversion is necessary. 

No such immediate burden of proof rests on the R+ trainer, because positive methods are not harmful in themselves.  Assuming equal efficacy, there is no demand to justify the use of R+ rather than P+ methods.  This is the kind of reasoning the LIMA scale makes use of; if we�re not using the least invasive, minimally aversive method for training, we need to give a reason why.  

It�s not enough to claim �I just do whatever gets results�, because in most cases the same results can be obtained through a variety of training approaches.  We therefore have to ask, what does the results-based trainer class as the best result, and is this compatible with what�s best for the dog? 


*This is why an earthquake is harmful but not a moral wrong - morality only applies to the actions of people. It is also why surgical procedures are painful but not wrong - it would be wrong for a surgeon not to operate to save a patient�s life because it would hurt them. So when we say an action is wrong, we are saying that it should not be done.

Saturday, 11 January 2014

Proving the Rules

If I were to punch triple heavyweight champion boxer Wladimir Klitschko in the face, several things could happen.  I could end up on the floor.  I would almost certainly break my fingers.  I may end up spending a night in jail.  Lots of potential outcomes, but one thing is nearly certain: I wouldn�t hurt him.  

Does my inability to injure Klitschko mean that the claim �punching people in the face hurts� is untrue?  Logically, yes.  But not practically.  The rule �face-punching = pain� is still a good enough guide that we would rely on it to guide our actions.  We can still say with enough certainty, �since I don�t want to hurt this person, I should refrain from punching him in the face.�  Klichko�s iron jaw is a corner case, an exception that �proves� the rule, it�s not a reason to reject the rule outright.  

Getting on to dog training, I�ve sometimes seen it claimed that because pain is subjective and one dog will feel something as painful when another won�t, we should dismiss phrases like �shock collars are painful�.  I can easily believe that a Klitschkoesque mastiff would experience the same force applied from a shock collar differently from a antelope-necked Italian Greyhound, but the individual corner cases once again don�t obviate the usefulness of the rule in general.  Shock collars might not hurt every dog at every level, but shock collars still hurt.  When we decide whether training a dog using shock is okay, we can still take "shock collars = pain" as useful data in most cases. 

Why might there be resistance to the idea that we can talk about variation at the level of individual dogs whilst still maintaining the integrity of general rules at the more abstract level?  One reason is the way that some of these rules have become entrenched to the point where even one deviation is considered heretical. 

To give a non-P+ example, every so often someone reports their frustrations when R+ training a dog who is not food motivated at all.  A dog who just doesn�t want to work for treats, or even toys.  Online at least, reports of these dogs sometimes come up against R+ trainers who simply refuse to believe in their existence, claiming that �dogs are food motivated� must mean that every dog, everywhere, forever must be interested in working for treats.  They�re wedded to the rule to the point that any exception creates cognitive dissonance, which gives rise to creative attempts to avoid accepting the account, often phrased as �you don�t have high enough value treats� or, �your timing must be wrong�.  

The answer to escaping from this nest of dissonance and dogma is practically simple, but unfortunately nearly impossible to implement in the environments in which these rules are discussed.  It�s just to know the dog in front of you; his physiology, temperament, what interests him and what he doesn�t like.  Remembering that the individual level of interaction and the abstract level of theory demand different types of observation and reasoning ultimately strengthens the theory that grounds the rules, by giving theorists the chance to introduce some flexibility and exceptions to their techniques whilst retaining their fundamental insights.

Monday, 6 January 2014

Call Your Dog!: Recognizing the right to make demands

In an earlier post, I mentioned Stephen Darwall�s definition of respect and how it�s related to recognizing other people as a legitimate source of authority.  One area that this kind of respect for authority is lacking is at the dog park (or anywhere dogs are allowed to meet each other at close range). There are lots of good reasons why you ought to call your dog if someone asks you to:
  • Because the other dog might be dangerous;
  • Because the other dog might be contagious;
  • Because your dog might inadvertently scare the other dog;
  • Because your dog might not behave like you expect him to;
  • Because I told you to.  
Of all of these, only the last one really matters. You might not always understand why someone is asking you to call your dog, but you ought to do it anyway.  

Last week, I was walking my dog in a park, in an area where many people let their dogs run free.  A lady with a stately Golden Retriever, off-leash, wandered into our range.  I noticed that the dog was getting quite close to my reactive dog Edie, who was on a long line. I called over, �Hey, could you call your dog?�.  The lady replied, �Oh, my dog doesn�t care about little dogs like that�, and she did not call her dog.  I said, �Well, mine�s in training and can be aggressive, so could you call your dog?�, but again the lady just smiled and repeated herself. 

Then, of course, the elderly Golden did notice Edie and began to wander towards her (there�s always a first time, right?).  Thankfully, we were able to exit the scene before Edie started to panic, so no harm was done, but this interaction is typical of the failure to recognize other humans as a legitimate source of demands when it comes to dogs. 

It shouldn�t have mattered whether the lady believed her dog wouldn�t �bother� my dog, she should have recognized that I was within my rights to demand that she call her dog.  The reason for this is that she should have realized that I was a morally legitimate source of demands in this situation.  

The legitimacy of moral demands is not conditional on the other person agreeing with or even understanding them, they rest on the recognition that the person doing the demanding has the authority to make such demands. Darwall uses the example of someone standing on your foot on the train.  You ask him to get off your foot, not whether he thinks it�s a good idea to be standing there.  That doesn�t matter.  You don�t explain that he�s hurting you, or that you need to move your foot to get off the train.  That doesn�t matter either.  What matters is that we all have a personal space that we can demand others not interfere with, without needing to supply an external reason.  Our bodies, possessions and the dependents in our families - including our dogs - are all within this space.

Moral authority is mutual; I recgonize you as a person who can make certain demands, and you recognize the same authority in me.  Darwall calls this �the second-person standpoint�; the understanding that you and I are equally able to claim our rights from each other.  Therefore, I shouldn�t refuse demands from you that I would resent you refusing from myself.

When I ask you to call your dog, I�m asking you to recognize me as able to make the demand that you respect my sphere of authority, to enter into a mutual relationship based on reciprocal respect.  Refusing to do so doesn�t just put our dogs in danger, it�s disrespectful on a more fundamental level.  So please, call your dog.   Even if you think I�m overreacting.