Ryle on the will

I’m going to take a look at Ryle’s comments on the will from chapter 3 of The Concept of Mind (one of the better chapters of that book in my opinion, largely for being one of the few in which Ryle manages to remain reasonably concise and to the point). Ryle’s general project, of course, is to discredit the “official doctrine” that minds are separate from bodies, and are constituted by events taking place in a sort of special internal theatre. The will, says Ryle, forms an important part of this doctrine.

On the official doctrine, the will is a central faculty of the mind, and the exercise of the will consists in acts of volitions. Volitions are essentially mental acts whereby we direct our bodies to do things. They’re posited to explain how our ideas become actions. It’s one thing to desire an ice cream, but I will only actually get an ice cream if I execute a volition that causes my muscles to move. In this way, volitions account for the distinction between voluntary and involuntary acts. The movement of Frank’s arm was caused by a muscular twitch. The movement of Vincent’s arm was caused by Vincent willing that his arm move, i.e. he executed a volition that caused his arm to move.

Ryle offers four arguments against volitions. We will see that the first argument, though overstated, contains a damaging problem for the volition theorist; the second and third arguments fail; and the fourth argument is inconclusive. (Note that although Ryle rejects volitions, he doesn’t reject the voluntary/involuntary distinction. I’ll examine Ryle’s account of that distinction in a later post.)

1. Ryle’s first objection to volitions is simply that nobody ever actually talks about volitions in day-to-day life, and when we try to describe volitions in any detail, we don’t know what to say about them. There are two threads to this argument, and it’s worth separating them. First, nobody ever uses the language of the will and volitions; i.e. “no one ever says such things as that at 10a.m. he was occupied in willing this or that, or that he performed five quick and easy volitions and two slow and difficult volitions between midday and lunch-time” (Ryle 2000: 63).

This argument is weak. The obvious response is that since volitions are a central part of our minds, to the extent that volitions underly every voluntary action, there’s simply no need to speak about them. In the same way, we don’t talk about moving our muscles – we don’t say things like “the bicep muscle of my right arm contracted” – though muscular movements occur every time we perform actions. Indeed, the ubiquity of muscular movements is precisely the reason why we don’t bother to talk about them. Only in special circumstances, such as learning human anatomy or performing exercises, need we direct our attention towards them. The same, the volition theorist could argue, is the case for volitions; their very ubiquity is why we never talk about them.

Second, when we try to analyse volitions in any detail, we run into a host of problems with spelling out exactly how they work:

By what sorts of predicates should they be described? Can they be sudden or gradual, strong or weak, difficult or easy, enjoyable or disagreeable? Can they accelerated, decelerated, interrupted, or suspended? Can people be efficient or inefficient at them? Can we take lessons in executing them? Are they fatiguing or distracting?... (Ryle 2000: 63).

I’m not sure how much force this argument has. In the first place, I doubt that all of those questions are really as difficult for the volition theorist as Ryle supposes. Surely, for instance, it’s clear enough that if there are volitions, then they can be difficult or easy and enjoyable or disagreeable. You want to approach a woman at the bar, but it’s difficult to push yourself to do; you want to buy a sandwich, and you have no trouble executing the volition to do that. Similarly, it’s easy to make sense of a volition’s being interrupted or suspended: you want to approach a woman at the bar, but you’re hesitating, finally you pluck up the courage and start to walk towards her only to change your mind a split-second later.

Still, I suppose we can grant Ryle’s basic point that there are various respects in which volitions are quite odd; there are many questions we might ask about volitions that are difficult to answer. Does this spell trouble for volitions? Whenever I see arguments of this sort, I’m reminded of this passage by Richard Routley about the ontological troubles posed by clouds (Routley was parodying a passage in Quine’s famous paper On What There Is, a subject for another blog post; but I think it also serves as a response to Ryle):

Take, for instance, the cloud in the sky above; and again, the adjacent cloud in the sky. Are they the same cloud or two clouds? How are we to decide? How many clouds are there in the sky? Are there more cumulus than nimbus? How many of them are alike? Or would their being alike make them one? … is the concept of identity simply inapplicable to clouds? But what sense can be found in talking of entities which cannot meaningfully be said to be identical with themselves and distinct from one another? These elements are well-nigh incorrigible… (Routley 1982)

Ryle overstates his argument. There are indeed some perplexing questions about volitions; but there are also perplexing questions about clouds, and nobody doubts that there are clouds. However, this isn’t to say that Ryle’s argument has nothing to offer. There’s a question lurking in there, which he hints at but doesn’t state explicitly, that the volition theorist should find genuinely worrying, namely: how do we individuate volitions? I want some ice cream, so I go to the kitchen and get some. The official doctrine says that I have the volition to eat ice cream. But must I also have the volition to get up off my chair? To open the door to the kitchen? To put one leg in front of the other as I walk? When I walk to the kitchen to get some ice cream, how many volitions are involved? Speaking for myself, I have absolutely no idea. It could be one or it could be ten. This, I think, poses a rather more serious problem for the volition theorist. Volitions are supposed to be conscious acts of the mind, and it is by executing a volition that I engage in a voluntary act; surely then I should be able to tell how many volitions I’ve executed.

2. No evidence can ever be given that anybody ever actually has volitions. This is a consequence of the fact that volitions are conceived as fundamentally inner events: so when considering other people, we can only ever infer from behaviour to the volition that supposedly caused it. But this can be no better than a guess. The problem is that any behaviour that might have been caused by volitions could equally have been by caused nonconscious processes, and we can never get inside other people’s minds to check which is the case. Even if a person explicitly states “yes, I willed to open the door” or “I executed a volition to open the door,” we have no reason to think that these statements were caused by conscious volitions rather than nonconscious processes.

It might be objected that we know other people have volitions by analogy with ourselves. This is, of course, a classic solution to the problem of other minds, and it’s easy enough to see how it’s supposed to apply here. Put simply: (1) I know directly that my volitions cause my own behaviour; (2) I see that other people’s behaviour is very similar to mine, that they have similar developmental histories, are made of similar materials, etc; (3) I can infer that other people’s behaviour has similar causes as my own. This argument is, of course, very controversial. Ryle bypasses much of the controversy by simply denying premise (1). I don’t in fact know that my behaviour is the effect of my volitions. Ryle’s argument for this is that even if we do have volitions, the connection between volition and action is completely mysterious. I may have a volition to open the door, and then open the door, but my opening the door could have been caused by something else. The volition could have been causally inert. In other words, nothing I experience allows me to distinguish the case where I have a volition to open the door, which causes me to open the door, from the case where I have a volition to open the door which has no effect on my behaviour, but I open the door anyway as a result of some hidden cause.

Ryle goes too far here. If we frequently have volitions and our actions match these volitions, surely the simplest explanation is that our volitions cause our actions. Now to be fair, Ryle does raise an interesting problem. However, it’s a problem that applies to causation in general, and was raised by Hume several centuries ago: we can never observe event A causing event B, à la a necessary connection between A and B; we can only observe a constant conjunction between A and B. Yes, the causal connection between my volition to raise my arm and my arm’s subsequently raising is mysterious; I never observe the former causing the latter. But the causal connection between any two events is mysterious. Nothing I experience allows me to distinguish the case where the ball hits the window and causes the window to shatter, from the case where the ball hits the window and has no effect on the window, but the window shatters anyway as a result of some hidden cause. It’s hardly fair to saddle the volition theorist with a problem that applies to literally every causal claim.

3. Third, Ryle appeals to the classic interaction problem: causal interaction between volitions and bodies is impossible, since the mind, which is what executes volitions, is defined as a different kind of substance than the body. This is a standard argument against Cartesian dualism in general, of course. A couple of comments. First, I’m not sure how serious a problem interaction really is for the Cartesian dualist. It’s a textbook argument, I know, but as noted above causation in general is odd, and it’s not obvious that two radically different substances can’t interact. Second, more importantly, one doesn’t have to be a dualist to posit volitions: one could simply identify volitions with certain kinds of brain processes, for example. The interaction problem has no bite whatsoever in this case.

4. Finally, Ryle poses a dilemma for the volition theorist. Since volitions are themselves acts, specifically acts of the mind, we can ask: are they voluntary or involuntary? If the former, we face a regress. Volitions are what make acts voluntary, so if volitions are voluntary, they must be preceded by earlier volitions, and these earlier volitions must be preceded by still earlier volition, and so on. But if volitions are involuntary, then we have the odd result that acts count as voluntary by being caused by certain kinds of involuntary acts.

It seems to me that the force of this objection depends on what we want a theory of volitions to do. If the volition theorist’s goal is to account for free will, so that volitions are somehow supposed to confer on us something like the philosopher’s notion of free will, then Ryle’s dilemma poses a serious problem. The volition theorist can’t say that volitions are voluntary or he lands himself in a regress; but if he says that volitions are involuntary, then he’s trying to account for free will be appealing to involuntary acts. Putting aside worries about whether or not this latter option is even coherent, it makes the appeal to volitions superfluous. If free will can be grounded on involuntary acts, why even bother postulating volitions?

However, the volition theorist need not have such lofty goals. He might, for instance, merely be attempting to account for the distinction between voluntary and involuntary acts – this is in fact what Ryle takes the point of the volition theory to be. (Doesn’t the distinction between voluntary and involuntary imply free will? No. Whether or not we have free will, there’s a significant distinction between my arm’s moving as a result of a muscular twitch and my arm’s moving because I consciously directed it to move. In everyday discourse, we use the terms “voluntary” and “involuntary” to name this distinction. The volition theorist might simply be trying to explain this everyday distinction, rather than concerning himself with the deeper question of free will.) In this case, I don’t see why it would be a problem that volitions are involuntary. All the volition theorist is doing here is saying that behaviours of a certain type, namely voluntary behaviours, are those that are caused by a certain type of inner event, a volition. If there’s a volition in the causal chain, then the behaviour is voluntary. Compare, for instance, homicide – to say that a death is a homicide is just to say that the death was caused by another human. Obviously, the man who commits a homicide is not himself a homicide. Why then should we require that the volition that causes voluntary behaviour itself be voluntary?

To conclude, while I tend to share Ryle’s skepticism of the notion of volitions, I’m not sure how successful his arguments against volitions are. Anyway, that’s all for now.

Routley, R. (1982) “On What There Is Not”, Philosophy and Phenomenological Research, vol. 43, no. 2, December, pp. 151-177.

Ryle, G. (2000) The Concept of Mind, London: Penguin Classics.