Towards the end of his book "On Becoming A Person", person-centred psychotherapist Carl Rogers writes:
"I feel a deep concern that the developing behavioural sciences may be used to control the individual and to rob him of his personhood. I believe, however, that these sciences might be used to enhance the person."
That was in the 1950's; the relatively new research fields, Rogers' "person-centred" psychotherapy and B.F. Skinner's (and others') behavioural sciences were progressing in parallel and, with their very different goals, it was inevitable that the two approaches would provide much intertwined debate.
A common perception is that the use of positive reinforcement (whether or not involving clicker training) can only be a good thing. As much as I support the use of positive reinforcement and clicker training, I would argue that this is a misconception. For some time I have been concerned that many people have started to learn clicker training in order to obtain instant co-operation from their horse, bypassing the need to search for the reasons for any perceived lack of co-operation. The clicker provides a clear and potentially unambiguous marker to a horse and desired behaviours can be taught relatively quickly. Students might be taught to put a behaviour "on cue", meaning that a behaviour should always be performed in response to the cue and never in the absence of that cue. But is this really any different from, or more enlightened than, the old "teach it who's boss" mentality which some of us have tried to leave behind? Does the horse feel better for our new-fangled thinking or does it just ease our consciences to claim that we are working in the positive?
When I read Rogers' book recently, I realised that he had come to similar conclusions - almost 50 years ago. He describes how the behavioural sciences had led to great steps forward in man's ability to predict human behaviour and personality - which members of society might be successful, troublesome, potentially prejudiced, who might bow to pressure. It was possible to provide conditions under which workers might increase their productivity or reduce their psychological flexibility, or under which consumer responses and public opinion might be altered. Furthermore, it was possible to set up conditions which caused individuals to make judgements contradictory to their true opinions - without ever becoming aware of the stimuli which influenced those changes. It was possible to completely destroy someone's opinion of himself, making him dependent on another.
I could go on and on, but think it is sufficient to say that the behavioural sciences can provide a rather scary outlook on life - think of Aldous Huxley's "Brave New World" or George Orwell's "1984", depending on whether a positive or negative approach is adopted. Or think of the blatant manipulation of the participants of television's "Big Brother". Regardless of the example used, the effect is to reduce the free will inherent in an individual. I would guess that anyone who watches the old footage of Marion and Keller Breland's "Animal Behavior Enterprises" might feel themselves torn between common sense saying that it was 100% positive reinforcement and choice whether or not to perform, yet the heart saying that there is something wrong about hundreds of animals shut up in Skinner boxes, performing repetitive behaviours in order to earn treats. Oh yes, and to entertain the general public.....
So if not operant conditioning, then what......?
Returning to the second half of Rogers' statement - how can we use the behavioural sciences to enhance the personality of the horse, without resorting to the "tell it who's boss" mentality and without resulting in a horse who tells us who's boss instead?
As I started to use clicker training, I was lucky in that I learnt from trainers who wanted to encourage the free will of the horse, working with behaviour instead of against it - isn't so much of horse training about pre-emptive correction or punishment of unwanted behaviours? I learnt that free-shaping (with or without a clicker) is a very powerful means by which the real horse can be released. A free-shaping session is one in which there are no wrong answers, no negative consequences. The horse can offer behaviours in return for rewards, or it can decide not to participate and go off to eat grass. There is still a certain amount of manipulation - only the behaviours I desire can get rewarded (so is Brave New World unavoidable?) - but this is minimised if the horse can still benefit through not participating and if I am ethical in my choice of behaviours.
These sort of training sessions - which can last from a few seconds to over an hour, the horse decides - are invaluable for allowing a horse to be himself. My interpretation is that he can learn to trust that wrong answers won't be punished. I can use his reactions as a guide to my behaviour and adjust it accordingly. The trust escalates on both sides - the human learns to trust that it's not in a horse's nature to take advantage or "take the mickey", the horse learns that he doesn't need to fight because the restrictions on him are reduced. The relationship becomes one of conversation and mutual give and take, not control - but if the human takes enough time to learn about behaviour, the balance is found so that safety is not compromised. Neither is there any question of doing less with the horse - I've spent the last couple of years doing sponsored rides, various competitions and mock hunt, barefoot and bitless - and I've never felt more sure of what my horse is thinking. It never ceases to amaze me how much more safe and willing he feels.
Other than my speculations, I had no formal understanding of why building a relationship like this works, at least, not until I read Rogers' book. His work has been providing both experiential anecdote and scientific corroboration of what makes a constructive, and in his case therapeutic, relationship for decades. It is difficult - and inadvisable - to describe in a nutshell why Rogers comes to his conclusions, but he summarises that the most successful relationship can grow when the therapist, and in my opinion the trainer, can be
"a) genuine, integrated, transparently real in the relationship; b) acceptant of the client as a separate, different person and acceptant of each fluctuating aspect of the client as it comes to expression; and c) sensitively empathic in his understanding; seeing the world through his client's eyes".
If he can behave in such a way, the client is able to feel less defensive and more open to being himself - my belief is that this is how a horse can feel in a free-shaping session. Gradually this mutual understanding and acceptance can expand to the world outside that free-shaping session and become a normal way of life.
Returning to Rogers' initial statement above - we can use the behavioural sciences to help create our relationship. But there is more to life than physical reinforcement. What does your horse feel? It is no good me claiming that I have "positively reinforced" my horse because I have just given him a treat - it is down to what he feels and experiences. And I can't know that unless I trust and accept him for who he is.
Copyright Catherine Bell 2006