Once in every while an ethical dilemma will swim across the horizon, a dilemma whose wake will induce in me a bout of moral seasickness. My compass spins, my bearings wheel and lurch. One such is the reappearance of “facilitated self-harm”. I am not over-fond of the word “facilitate”. It drips with the oil of evasion. It sits too easily on the smoothest tongues. What is meant here is helping people to hurt themselves. It refers to the practice of providing people who self-harm the sterile means to accomplish it.

There are those who work in ethics, robust souls on the whole, who don’t give much house room to moral feeling. They will tend to snicker at the “yuk factor”, at squeamishness, at moral disgust, arguing instead that only reasons count, that there is no place for sentiments in ethical debate. Being weaker minded, I am not of their party. Moral seasickness is my Davy lamp. It tells me when a dangerous dilemma is drifting my way. It tells me when it is time to start thinking. Yuk! can be the beginning of argument, though it cannot be its end.

So what to make of helping people to harm themselves? A good consequentialist, someone who holds that the consequences of an action are what matter morally, could well say that if the provision of sterile blades leads to less overall harm then it could be justified. The argument is a strong one. Looking behind the stories, it seems to be the case that self-harm is a strategy used by some to cope with terrible emotional conflict, that it helps them to manage otherwise unmanageable misery. If self- harming cannot be stopped, then it is surely right to limit the damage it causes. As an aside, it is interesting that many of the calls for the provision of safer methods of self-harm come from nurses, from those who work more intimately and daily with people in emotional extremity.

Try as I might though I cannot entirely rest with this consequentialist conclusion. If self-harm is the condition requiring treatment, can it be right to provide a cleaner knife? There are some possible parallels, some near analogies. I have worked with medical charities being asked to provide sterile equipment for female genital mutilation. There is also body dysmorphic disorder, where someone so takes against a healthy limb that they will chop it off themselves unless a surgeon is willing to assist. In all these cases an argument runs that because they will do it anyway, medicine should intervene to minimise the harm.

Thinking about my unease, I suspect it has something to do with the consequentialist conclusions running up against a deep-lying principle about the purposes of medicine, a principle so deeply digested as to form part of our ordinary moral sentiments: the principle that medicine should do no unnecessary harm. The weighing of such a principle against the measurable consequences of assisting or not assisting can never be easy. I suspect these cases will always present a dilemma and any decision will always feel approximate and imperfect. But if we decide to go with “facilitation”, if we accept the consequentialist arguments, then I think we accept an attendant duty: to undertake rigorous audit and assessment. If we are going to argue from consequences then we have to have a pretty good idea what those consequences might be.

Julian Sheather is ethics manager, BMA. The views he expresses in his blog posts are entirely his own.