Compatibilism and moral responsibility
At a key point in the argument for incompatibilism, I asked this rhetorical question: “If I wasn’t responsible for the intention, why am I responsible for the act?” One powerful contemporary form of compatibilism argues that the right answer here is just “Why not?” Why should anyone think that the fact that I’m not responsible for having the mental states I do means that I’m not responsible for the acts I perform as a result of my mental states?
One reason, of course, is that sometimes we say people aren’t responsible for what they did because we know someone else has manipulated their mental states.
Suppose, for example, a hypnotist gives me a posthypnotic suggestion that I will have an irresistible desire to close my hand when she flicks her fingers. Knowing this, you might want to hold her and not me responsible for the flooding of the valley, when—after her flicking her fingers—I close my hand. This form of intervention, early in the causal chain that leads from my interior states to the contraction of my muscles, seems as exculpating as the more straightforward intervention of making my muscles contract by an electric impulse—or, for that matter, just squeezing my hand closed by force. So it's certainly true that sometimes the fact that others produce our mental states provides an excuse. And this is true sometimes even when it is not someone else but something else that does the work. Suppose that the desire to close my hand was caused by a brain tumor. Wouldn't that excuse me, too?But, the compatibilist will say, from the fact that some causes of our mental states relieve us of moral responsibility it doesn't follow that all causes of our mental states relieve us of moral responsibility. If you aren't hypnotized, don't have a brain tumor, and so on, then you are responsible for what you do (or fail to do).
This claim might even be made consistent with some version of the principle of moral responsibility.To see how, consider again the example of my closing my hand and flooding the valley. Suppose I close my hand of my own volition. Then you might rightly say that I ought not to have done it. In defending myself, I might draw on the widely accepted idea, which is one version of the principle of moral responsibility, that:
OC: Someone ought to do X only if he or she can do X.
(This is sometimes abbreviated as “ ‘Ought' implies ‘can.' ”) So I could say, “Well, if I ought not to have done it, then, according to OC, I must have been able not to do it. But, surely, if determinism is true, I couldn't have done otherwise.” But a compatibilist could reply: “That doesn't follow. You certainly could have done otherwise; in fact, you would have done otherwise if you'd had a different intention. The sense of ‘can,' in which ‘ “ought” implies “can” '—and ‘ “ought to have” implies “could have” '—is only that: you would have done otherwise if you had chosen to. And the truth of determinism gives us no reason to doubt that. For there are surely many possible worlds where you chose otherwise and acted differently.” So far, I think, an incompatibilist is likely to find this rather unconvincing. For how does it help that I would have acted differently had I chosen to if I couldn't have chosen to?
Rather than answering this question directly, let me reframe the challenge of determinism in a different way. (I will get back to the question I just asked eventually!) The basic idea of incompatibilism is something like this: we are responsible only for what is under our control, and determinism shows that we don't control anything. But, as Robert Nozick once pointed out, nobody ever argued that because determinism is true, thermostats don't control temperature.
If a thermostat is working—controlling the temperature—and the heat is off, it's still true that if the temperature had been below its set point (the temperature it is designed to maintain), it would have switched on the heater. If determinism is true, the temperature couldn’t have been lower. Nevertheless, the thermostat would have turned on the heater if it had been. Suppose that a thermostat is indeed working in this sense and the temperature drops below the set point. It will turn on the heater. And the heater will have been turned on under the thermostat's control, even though the thermostat is a deterministic system. Analogously, then, I am in control of whether the valley floods if, if I were to choose, I would close my hand and set in motion the process that releases the water. So, if I choose to close my hand, then the flooding is under my control. If determinism is true, I could not have chosen otherwise: but that doesn't mean that the flooding isn't under my control.To see why this might help motivate the compatibilist's response, let's fill in the story of the closing hand and the flooding valley a little more. Suppose the reason I have my hand round the detonator switch is that I work for a hydroelectric company. The dam I can blow up, if I choose, is one of two through which the water drops through two turbines into two valleys. One valley is highly populated; the other is not. There has been a great deal more rainfall than any of the engineers predicted when the dam was designed, and the result is that the overflow pipes are not sufficient to carry away the excess water that is surging down river toward the dam. If I blow up this dam, then the water level will fall fast enough to stop water flowing over the other dam. And the dam with the dynamite is the one that drops into the less populated valley. So if I do nothing, water will flood both valleys, drowning many people; but if I blow up this one, only the less populated valley will be flooded, and very many fewer people—perhaps, if I am lucky, none—will die.
I have done everything I can to warn people in the less-populated valley to prepare.
Now I must take responsibility for risking the lives of a few people in order to save many. If there were no people in the other valley (and I had been aware of the fact), I would have blown up the other dam. So there's at least one circumstance in which I would have chosen otherwise. True, if determinism is correct, there couldn’t have been fewer people in the other valley. But surely, the compatibilist will say, the fact that I chose as I did because of what I knew and because I was trying to minimize loss of life makes me responsible for what happened. And, in fact, I should be praised for having made the correct, if tragic, choice. What matters, in other words, is that I formed my intentions and acted in response to my understanding of the facts and my aims. Let us call my understanding of the facts and my aims, taken together, “my reasons.” It is simply irrelevant whether those reasons were the result of inexorable causal processes. If I had made my decision as a result of a hypnotist's flicking her fingers or of a brain tumor, my act would not have been responsive to my reasons. What makes me responsible, in short, is that I acted on my reasons.So we can return to the question I left hanging a while ago: How does it help that I would have acted differently, had I chosen to, if I couldn't have chosen to? It helps because the reason I couldn't have chosen otherwise is sometimes that what I chose to do was required by the reasons I had. When that happens, when what necessitates my action is my reasons, then I am responsible. My acts are under the control of my reasons. And that is very different from the case where what necessitates my action is force, or a tumor, or hypnosis.
Notice that, on this view, if I am responsible only where I act for reasons, then the practice of holding people responsible—of blaming and praising them for what they do, and punishing sometimes what is blameworthy and rewarding sometimes what is praisewor- thy—is appropriate only in cases where they are acting for reasons.
And that makes the practice of holding people responsible one that will be appropriate only in cases where the fact that the agent will be held responsible might have an effect: in the cases, that is, where fear of blame and punishment, or anticipation of praise or reward, might make a difference by adding to the reasons that the agent is responding to. If I am responding to an electric impulse or a tumor, there's no role for anticipation of reward or punishment in shaping my action.This is only, of course, a beginning of an exploration of how com- patibilists seek to make space for free will—understood as having your actions under the control of your reasons—in a world in which what we do is ultimately caused by events outside us. There are contemporary incompatibilists who are skeptical of this solution and who believe that our ways of ascribing moral responsibility should be abandoned or, at least, revised. (Of course, if they are determin- ists, they should presumably think that we can't help ascribing moral responsibility, even if we shouldn't!) So the debate goes on. But I hope this preliminary introduction to the debate between compati- bilism and incompatibilism confirms what I said at the start of 9.10: the problem of free will exemplifies the way in which some philosophical questions belong not to the specialized subfields—episte- mology, philosophical psychology, philosophy of language, metaphysics, ethics, and so on—but bring them all together. I think it is because it requires all the intellectual resources of the subject that the problem of free will is so challenging.
9.12