- Albus Dumbledore, Harry Potter and the Philosopher's Stone
In this article I shall provide a summary of Robert Kane's paper Responsibility, Luck and Chance: Reflections on Free Will and determinism (Journal of Philosophy 96 (5), pp217-240; 1999). I shall then present two challenges to the view he elucidates.
Before I get into the serious meat of the paper, allow me to quote its opening words:
Ludwig Wittgenstein once said that "to solve the problems of philosophers, you have to think even more crazily than they do". This task (which became even more difficult after Wittgenstein that it was before him)...Oooh! Burn!
Kane is responding to the arguments of Daniel Dennett. Dennett is a compatibilist of sorts - he argues that we have moral responsibility, and this is frequently taken to entail possession of free will but to me it feels more like a denial that moral responsibiliity requires one to act freely. (Dennett's position is actually very similar to one I very briefly wondered about back when I was a naive fresher who hadn't read much philosophy - see the section titled Moral Identity here.) Furthermore, Dennett argues that libertarian free will is in fact rather unsatisfying: it seems to involve people doing things for no very good reason, as opposed to his conception in which people perform actions in accordance with their character and may be judged for an action in terms of how representative the action is of their character. If under similar but non-identical circumstances the agent would have acted differently, then the action may be seen as an aberration for which the agent should not be held responsible. If large changes to the situation would have been required to change the action, then an action is representative of a wider trait of the agent and is therefore something for which the agent ought to be held responsible.
Kane's aim is to present a view of libertarianism which actually seems worthwhile. He argues that rather than having a character which determines our actions, we form our character through the actions we take. He labels the key decisions we make which determine we shall become "Self-forming actions", and argues contra Dennett that there are good reasons for making these choices, but they are not all immediately visible - indeed, many of them lie in the future.
A particular challenge that Kane aims to deal with is as follows: suppose a man has the choice of going on holiday to either Hawaii or Alaska. He deliberates over this decision, and finds several good reasons for going to Hawaii - it is more pleasant, cheaper, etc - and none for Alaska. At this point, what kind of freedom is it which allows the man to still choose Alaska? This is surely less a case of meaningful choice than of perverse randomness.
Kane's response it that we do not possess free will in that kind of case - it would indeed be perverse to choose Alaska. Instead, we possess free will pretty much entirely in our SFAs, but the preferences which dictate our many other choices stem from SFAs. The man's choosing to go to Hawaii would not be an SFA, and would not of itself be a meaningful choice; however, his preference for hot over cold might stem authentically from his past enjoyment of summers, and so the choice may still be indirectly meaningful.
He also responds to the problem of "moral luck". Suppose a woman is walking to an important interview, when she sees a person being mugged in an alley. She has pepper spray in her handbag, and so could save the person who is being mugged, but this would cause her to be late for her interview. If it is truly indeterminate as to whether or not she does the moral thing by stopping the mugging, then what is there to distinguish it from luck as to whether she saves the person? How, then, can she be either praiseworthy or blameworthy for her action?
Kane responds that, since the businesswoman has good reasons for multiple courses of action, and these courses of action conflict with each other, she is at an SFA. She may be viewed as simultaneously attempting both courses of action - stepping in to stop the mugging, and hurrying along to her interview - and succeeding at one, failing at the other. Suppose that, in the event, she keeps out of the mugging and just rushes along to her interview. Kane would say that she could not control whether or not she succeeded at stepping in, nor could she control whether or not she succeeded at moving along; nevertheless, she could control which one of the two it was that she succeeded at. Hence she is responsible for her decision to move on.
So much for what I intended to be a quick summary. I find his account very appealing, and would very much like to believe it. Unfortunately, I have two key issues with it.
Multiplicity of potential SFAs
Brian is addicted to smoking. He knows it is bad for him, and every single day he swears to himself that he will quit. Yet, every day without fail, he will give in and sooner or later he will pick up the first cigarette of the day.
It seems in this case that each and every one of Brian's attempts to quit smoking has the potential to be an SFA. If he were to succeed, it would be a classic example of an SFA. It also seems strange to claim that certain decisions can be SFAs only if they go in a particular direction. Yet this seems to commit us to the idea that Brian is making an SFA every single morning, in spite of the fact that each and every one of these SFAs is the exact same decision.
If it does not seem strange to classify a decision as an SFA only when it goes a particular way, consider Brian's brother Steve. Steve also smokes, and has been thinking about giving up. However, he decided once and for all that he is approaching retirement and has earned a vice or two to keep him going in his old age. This seems like a very good candidate for an SFA, and does not seem importantly different from the decision made every day by Brian.
Lack of responsibility for failure to act
Let us go back to the case of the businesswoman. She did the presumably immoral thing of moving on and abandoning the mugging victim. This is something for which we want to be able to hold her morally responsible. Unfortunately, according to Kane it seems that we cannot.
Remember, according to Kane the businesswoman was simultaneously trying both to help the person and to move on. She failed at the first and succeeded at the second. According to Kane, then, she was trying to move on and therefore is responsible for doing so; however, she was also trying to help the person, and it was not in her power to succeed at this. Suppose then that we ask her; "Why didn't you help the mugging victim?" She can then honestly respond: "It's not my fault! I was trying to, it's just that I failed at doing so!" I see no reason why this should not generalise across all actions where we wish to hold someone responsible for failing to do something. "I was trying to give money to the poor! I just failed, because I was prevented by buying this shiny new iPhone!" "I was trying to fulfil the terms of the contract! I just failed, because I was prevented by my desire to save money and effort!" "I was trying to resist my urge to do unspeakable things to this person! I just failed, because of my desire to forcibly have sex with them!"
While I would very much like to endorse Kane's account of free will, it has severe problems which seem to vastly exaggerate the importance of certain small decisions, and which prevent us from holding people responsible for failing to act in certain ways.