I should like to have put this post off until I had read Sam Harris’s new book on free will, as well as re-reading Dennett’s Freedom Evolves, but it seems that it cannot wait, since Jerry Coyne has just put up another brief post on the subject of free will, and I need to claim some elbow room at this point which will allow me to develop things later at more leisure. I do have to say at the outset that I still cannot for the life of me think why it is thought so important to take a very strong stand on this issue where so much intellectual as well as material complexity is involved. Nor do I think we can really take seriously Jerry’s point that philosophers have already done as much as they can on this issue and that it is now time to take it away from philosophers and to turn it over entirely to scientists and experiments. This is not to deny the importance of science in this area; it is just to suggest that jumping into experiments at this point is a bit premature.
For instance, Jerry provides five different areas in which he thinks progress can be made. Let me list them here for the sake of reference, though I will not refer to all of them in what follows:
- Find out how people really conceive of free will, and whether they’re dualists who believe one can actually choose freely among alternatives at any given moment. Yes, I know some studies seem to show that many aren’t dualists, but I’ve read those studies and haven’t found them very convincing.
- If people are dualists, we need to tell them that there is no free will in the contracausal sense. This is what I have been doing, and, to a large extent, what Sam’s book does. Many here seem loath to do that; indeed, some have said that we have to keep the precious knowledge of determinism to ourselves lest it discombobulate the “masses”. I find that condescending and invidious: above all, we must speak the truth. After all, rejecting contracausal free will does have practical implications, at least for the justice system, as well as for people’s scientific view of how their brains work.
- We need more psychological experiments like those described in Daniel Wegner’s book, The Illusion of Conscious Will (I like that book though it’s a bit overwritten), and like the Libet and Soon et al. experiments.These studies—criticize them as you will—tell us something about how “decisions” are formed in our minds, where the neurons for those decisions lie, and about the time course of how we act. Other studies like those Wegner describes tell us how people conceive of their agency when acting, and how notions of personal agency can be either increased or deceased by experimental manipulation.
- And finally, we need more studies of the brain. How do diseases, injuries, or manipulations affect our notion of agency? What parts are responsible for the idea that we can really make choices? And, most important, what actually goes into play, neurologically, when we are faced with alternative actions and “choose” one?
He adds a fifth later on, but we can ignore that for the present.
Let’s start with number 1. This seems to me to be particularly unimportant. It doesn’t matter whether people are intuitive dualists or not, and while it might be interesting to find out, it shouldn’t affect the truth about “free will.” It is very likely, at least in areas deeply influenced by Christianity, where personal responsibility for belief or unbelief, as well as for sinfulness or righteousness, are central to the understanding of the human person, that people will tend to be dualists with respect to mind and matter, and, in those terms, will most likely think of persons as causal agents in the strong sense of being the ultimate originators of decisions and acts. But this is by no means certain, since Calvinism (following Augustine) seems to think of the only ultimate agent as God, and the lives and decisions of human beings are determined far in advance — or, as Calvinists would say, predestined — by God. But naive ideas about free will are not part of the “problem” of free will, and, interesting as it may be to find out what most people think, has very little to contribute to the solution of the problem.
Jerry’s second point is a curious one, because it is not altogether clear what is meant by “contracausal free will.” And telling people that there is no such thing, unless we are very clear as to what we are denying, is not only not required by our commitment to truth, but may in fact be simply a confusion. In his book Straw Dogs John Gray speaks about free will in these terms. His example is Joseph Conrad’s Lord Jim. “Did [Jim] jump? Or was he pushed by events?”, Gray asks (65). And then he goes on:
If Jim is to be held accountable for his jump he must have been able to act otherwise than he did. That is what free will means — if it means anything. 
And this, if I remember rightly, is what Jerry means when he speaks of free will, but how important is this sense of having been able to do otherwise to the question about free will? This may be what people mean, but is it important? Dennett explicitly denies its importance, as does Flanagan; and until this matter is settled it makes very little sense to continue dealing with it as though this were the only sense of free will worth mentioning or having.
It is unclear what is meant by the ability to act otherwise. Indeed, many of the things that we say regarding free will turn out, on examination, to be unclear. Watch this little clip of Stephen Pinker talking about free will:
The distinction being made here is a fairly commonplace one. There are reflexes that can be tested with a bright light or a hammer to the knee, and then there are things that we do. Obviously we do not speak of free will in respect of the first category. But what about the second? In Elbow Room Daniel Dennett describes behaviour of the Sphex ichneumoneus (see 10-11), and how, in laying its eggs, it builds a burrow, seeks out a cricket, which it then paralyses, but does not kill, drags it to the burrow, checks the burrow to make sure that all is in order, and then drags the cricket into the burrow where it will be food for the grubs which hatch from the eggs which Sphex then lays alongside it. However, if, while the Sphex is checking the burrow the cricket is dragged away from the burrow, Sphex will drag it back, check the burrow again, and then, if the cricket is dragged away again, will drag it back and check the burrow as before, and so on, indefinitely — or at least until the biologist has tired of play this interesting game. Sphex behaviour is determined in quite a strong sense of the word.
The fear, says Dennett, is that we are all sphexish in this sense, at least a little bit. His book Elbow Room is an exploration of all the different ways in which human beings might be thought to lack the ability to act freely (whatever is meant by that). These fears and worries are the “bugbears” which, he suggests, bedevil our attempts to talk about the idea of free will intelligibly, and his book is a warning not to feed the bugbears. It is simply not clear to me that we have got beyond the point of feeding these bugbears, and this is why philosophy still has a role to play here, since, if we are going to examine something empirically, at least we must have a clear sense of what it is that we are examining. I am not convinced that we have such clarity.
However, for the nonce, simply notice that this sort of clarity is needed before we continue with the discussion. We really do need some clarity about what free will might be thought to be, and we must show why some compatibilist approaches to the issue are unjustified, before we set off confidently after empirical snarks. As Pinker notes, the brain is so immeasurably, almost incomprehensibly complex, that the actual causal connexions between brain events and the “decisions” that we make are almost impossible to discern — as in fact Dawkins recently acknowledged in his conversation with Anthony Kenny and Rowan Williams. First, consider the following exchange:
Of course, the question here is whether we can reduce the problem in the way that Dawkins suggests without ending up examining something quite other than the context in which it makes sense to speak about free will. As Williams points out, dealing with small-scale, uninteresting and quite limited types of “action” (such as raising one’s hand) may be very different to dealing with large-scale, extraordinarily complex “actions” which is really what concerns about free will and determination are all about.
But there is a deeper problem involved here, which the Dawkins, Kenny, Williams conversation brings out quite clearly. Here is the last part of the segment of the conversation dealing with free will, continuing on from the point at which the last clip ends.
Reflect on Kenny’s statement that “the whole idea of constructing body and mind like that is quite wrong,” and consider how much the interpretation of the empirical evidence depends upon that way of constructing body and mind. When we consider the brain events, and the consciousness of having made a decision, the suggestion seems to be that the decision is merely the illusion of action thrown up by brain events. The illusion is, in a sense, the ghost in the machine, and plays no role in decision-making. The “decision” is just the output of brain events that have dealt with all the causal inputs, but it is not something that is done by, or the result of a decision of, an agent. But are we so sure that we can divide the “decision” up into these independent parts, as though part were the ghost in the machine, and the rest consisted merely of the nuts and bolts? Indeed, the assumption underlying this is that the “agent” or “self” is simply a bit like God in relation to the universe, a metaphysical presupposition of a pre-scientific way of thinking about human beings that was brought into question some time ago, and in particular by Gilbert Ryle in his The Concept of Mind.. Recall, too, Dennett’s concern about “sphexishness,” and ask what is the difference between Sphex and human beings. Are human beings just more complex automatons? Or is it possible that the very complexity of the human brain provides the elbow room necessary for making choices, so that we do actually choose what we choose? I see forbidding problems ahead about the ability to do things for reasons — which would call the whole project of science into question, as Plantinga has already helpfully suggested in another connexion.
The question will arise here at once whether, complexity or no, the human being could have done otherwise; but it is not entirely obvious that this is the most important question that we can ask. Given the complexity of the brain and its interaction with the environment, the probability that environmental factors and brain states will ever again be in precisely this relation is very improbable (as Dennett points out). If this is the case, then the intelligibility of wondering, after the even, about the ability to have done otherwise, is called into question. Moreover, this is not something that could be capable of empirical confirmation or disconfirmatiion, and reducing it to simple cases like “deciding to raise one’s hand,” in the way that Dawkins suggests, and neurological experimentation has done, will not do as a way of reducing the question to experimentally manageable proportions, for the question is precisely one about complexity. And this is why (as it seems to me) the issue of free will cannot be “considered the bailiwick of neuroscientists and psychologists rather than philosophers,” as Jerry Coyne puts it, for, whoever does it, there is still philosophical work left to be done.
One of the points that Dennett (as well as Pinker) makes is that the brain is so enormously complex it is hard to see why we should conclude that “could have done otherwise” is in any sense a definition of what we mean by free will. For, “could not have done otherwise” is a definition, not of determinism, but of fatalism — a very different thing. Determined we may be; indeed, this is a presupposition of the naturalist world view; but it does not follow that there is no room for choice — Dennett’s “elbow room.” We cannot think that it is necessary that, in order to have a range of choices, and having chosen one, that that choice was, in some sense, fated to happen, as it is, for example, in the case of Sphex. For we can, as Williams points out, merely say to the person choosing that his decision is determined, and that he cannot do otherwise, and immediately the person, if he so chooses, can falsify our prediction by choosing to do something quite different. And in response to the claim that the difference is made by the introduction of another causal feature, namely, the claim, made within his hearing, that his decision is determined, all that we need say is that, while that is true, this is in fact a technique that can be used, quite successfully, by someone who has learned (for example) to control his anger, and so the recurrent claim that (say) I could not have done otherwise may be recurrently thwarted.
One thing that I need to say in conclusion is to raise once more the question about the role of philosophy in all this. It is true that, in one sense, philosophy cannot come to empirical conclusions, although there is now a brand of “experimental philosophy” which is trying to overcome that limitation. However, the idea that philosophy, in not coming to empirical conclusions, is merely like theology, a matter of propping up ”cherished notions, the production of many conflicting and irreconcilable “solutions,” and so on,” as Jerry says, is not, in any sense, a response to what philosophers have to say about free will, for philosophy, in its disagreements, is dealing with the lack of clarity in our concepts, and these are still concepts which scientists have to use in exploring things empirically. Yes, philosophers disagree, but that does not mean that, in areas like free will, one can simply bypass philosophy and explore things experimentally, without dealing with philosophy’s concerns. Philosophy’s concerns have to do with the nature of what, in fact, is being investigated by science, and, so far, it is not clear to me that sufficient clarity has been demonstrated to show that there is anything here for science to explore.
However, as I say, this post is in the nature of a placeholder, since this is by no means all that can be said; nor would I like it to be thought that this is all that I will say about free will. I do have a question, however. Why is this issue considered to be so important, in particular, for atheists or anti-theists to resolve? And what do they think has been resolved when, as some people tend to think, it has been resolved?