We believe that we have free will, and this belief plays a central role in how we think about ourselves and about our lives. Free will matters to us, and not just because we think that having free will is a necessary condition of being morally responsible. Free will matters to us because having some say in the overall shape of our lives matters to us. We want to develop our skills and talents as best we can, make the most of our opportunities, and live the best lives it is in our power to live.
We know that there are severe limits to the control that we have over our lives. What we are able to achieve depends not only on us, but on our surroundings. We know that luck plays an enormous role, and we know that the playing field is not level. And even if we are fortunate in our starting point -- in our native endowments, our parents, social class, and so on -- we know that unforeseen events and the actions of other people can frustrate our plans and thwart our hopes. Knowing this, we might wonder how far our abilities extend, how much we are truly able to do. We might wonder whether we are really able to do, or even try to do, all the things we think we are able to do. We might worry that our powers of imagination are too limited, preventing us from even considering some of the things we might otherwise be able to choose and do. And so on.
But we don't doubt that we have free will. What we do is not the only thing we can do. The choices we make are not the only choices in our power to make. We are able to think for ourselves, whether or not we actually do so. We are able to try to do lots of things, whether or not we actually do try. We have abilities we don't exercise, perhaps some abilities we never exercise. We don't have to do what we do. We are able to do otherwise. Or so we believe.
We also believe that we often use or exercise our free will -- we take a moment to think before we act; we decide what to do, and then we act. And while we are sometimes in a position where we make a choice while being mistaken about what our options are, this is not always the case. Ordinarily, when we make a choice we really do have the choice we think we have. You are, right now, able to continue reading this essay, and you are also, right now, able to stop reading. Take a moment to consider these options. Then decide. Welcome back!
There is more to our view of ourselves as free agents than the belief that we have real choices and genuine alternatives, but this belief is at the core of our commonsense view of ourselves as persons with free will. If we were somehow convinced that we never really have a choice about what we do, we would conclude that our belief that we have free will is either outright false, or, at best, that we are radically mistaken about the nature of our free will.
We can agree about this while leaving other questions up for grabs. Do we have a choice only on those occasions when we deliberate, weigh our reasons, make a decision, and then act, or do we also have a choice when we act spontaneously, or on the basis of a hunch, or on the basis of what feels right or seems obvious? Do we have a choice in cases of coercion, and other cases when we have "no real choice" or "no reasonable choice"? What about cases where making a particular choice would go so deeply against our values that it is, in some sense, unthinkable for us? These are questions about which reasonable people may disagree.
We can also leave open questions about the exact relationship between choice and moral responsibility. We can all agree that moral responsibility doesn't require having a choice at the time of action -- we blame a drunk driver for the accident she causes, even if she wasn't able, at the time of the accident, to do otherwise. But does it require having a choice, at some earlier time, between sets of reasonably foreseeable outcomes of one's action? Or is it enough, so far as moral responsibility is concerned, that a person had a choice, at some time in the past, about the actions which led, eventually, to the formation of her present character? Again, these are questions about which reasonable people may disagree.
The answers to these questions are important, but irrelevant so far as the problem of free will and determinism is concerned. For determinism appears to have the consequence that it is never true that anyone has a choice of any kind -- big or small, important or trivial. If we choose to lie, we could not have chosen to tell the truth. If we choose the scenic mountain road, we could not have chosen the freeway. If we choose the apple, we could not have chosen the orange.
Roughly stated, determinism is the conjunction of two claims: i) that we are no exception to the fundamental laws that govern everything else in the universe; ii) that these laws state sufficient (as opposed to merely necessary or probabilistic) conditions for the occurrence of every event. More precisely, determinism is the thesis that, for every time t, the laws together with some proposition about the state of the universe at that time, entail the state of the universe at every later time.
Determinism is not part of our commonsense view.1 But as soon as we understand what the thesis of determinism says (and why it is a live possibility that determinism, or something close enough, is true), we see the problem. For if determinism is true, it seems that whatever happens must happen. And if that's true, then it seems that it must also be true, on each and every occasion, that what we do is the only thing wecan do. We are never able to do otherwise, or even choose or try otherwise. And if that's so, then our belief that we have free will is an illusion.
The free will/determinism problem is the problem of deciding whether this apparent bad consequence of determinism really is a consequence. Suppose that scientists announce that they have discovered irrefutable evidence that the thesis of determinism is true. Would this mean that our belief that we have free will is false?
The commonsense compatibilist2 says "no". The incompatibilist says 'yes'.
The libertarian incompatibilist goes a stepfurther and says that we don't need to worry about this hypothetical scenario because we have good reason to believe two things: First, we really do have free will. Second, it's not possible both that we have free will and that determinism is true. Therefore, we've got good reason to believe that determinism is false, and, moreover, false in just the right way to provide the kind of "open" future needed for free will.
Which of these two competing views -- libertarian incompatibilism or commonsense compatibilism3 -- should we believe?
We might start by noting that the two views are not on a par. The libertarian view has commitments that the compatibilist view does not. Both sides agree that we have free will, and both sides agree that we often also have what is sometimes called freedom of action; we are able to choose otherwise and also to act on the basis of that choice. But the compatibilist says that the truth of this commonsense belief is not hostage to what future science might tell us about the fundamental laws, whereas the libertarian says that it is. And the libertarian also claims that our reasons for believing that free will is incompatible with determinism are so good that we are justified, on the basis of our current evidence, in believing that determinism is false in just the right way. This seems dubious. So it seems that the commonsense compatibilist starts off with an advantage, at least against the libertarian. The commonsense compatibilist might insist, therefore, that since she and the libertarian agree about the free will claim, the libertarian needs an argument to defend her incompatibility claim.
The Dispositional Compatibilist provides a further, positive argument for commonsense compatibilism by showing how our commonsense beliefs about free will are compatible with our beliefs about the natural world and our place in it, in a way that is not contingent on the falsity of determinism.
There are different ways of being a Dispositional Compatibilis 4, but what all varieties of Dispositional Compatibilism (DC) have in common is that they claim that the most fundamental free will facts are facts about our causal powers (for instance, our power to decide on the basis of deliberation) and that our causal powers differ in complexity but not in kind from dispositions like fragility, elasticity, and flammability. In the older literature a distinction was sometimes drawn between "active" and "passive" powers -- the active power "exerted" by the rock when it breaks a window, the passive power manifested by the window when it is broken. The contemporary literature does not recognize this distinction, and neither does DC. The claim is that all dispositions -- of natural objects, artifacts, and human beings -- are causal powers with the same kind of causal structure. How exactly that structure should be understood is the subject of much metaphysical debate. But the main point, so far as DC is concerned, is simple. We think that determinism is incompatible with free will because determinism seems to have the consequence that we have no power over anything, not even our own choices. But if our causal powers are dispositions, this is not true. For dispositions are uncontroversially compatible with determinism. And dispositions are real properties of their bearers; they don't cease to exist simply because they are not being manifested. A rock doesn't lose its window-breaking power just because it isn't currently breaking windows. A person doesn't lose her decision-making power just because she isn't currently making a decision. Nor does she lose her power to decide to do one thing just because she makes another decision instead.
Can the solution to the free will/determinism problem really be so simple? Let’s take a closer look.
Dispositions and Abilities
Let's begin by considering some features of dispositions.
When we attribute a disposition to an object, we do so on the basis of a counterfactual property that we associate with objects which have that disposition. For instance, the counterfactual property that we associate with fragile objects is the property of being such that if the object were dropped or struck, it would break. But an object might have the counterfactual property associated with a disposition without having the disposition. A sturdy concrete block on the edge of the roof of a twenty story building has the property of being such that if it were dropped, it would break. But the concrete block is not fragile. And an object may have a disposition without having the counterfactual property that we associate with that disposition. A crystal glass doesn't cease to be fragile when it is carefully wrapped in styrofoam, but once it is thus protected it no longer has the property of being such that if it were dropped, it would break.
Why do we say that the protected glass is fragile but the precariously perched concrete block is not?
First approximation: Because the glass is the kind of thing that breaks easily, in a wide range of circumstances; the concrete block is not. A bit more precisely, the glass shares some intrinsic property with other fragile objects (glasses not wrapped in styrofoam), many of which have the counterfactual property. By contrast, the concrete block's possession of the counterfactual property is a highly extrinsic property; if the block is moved closer to the ground it loses this property.
It turns out to be surprisingly difficult to make these intuitive ideas more precise5, but here is a rough characterization of what it is for something to be fragile: An object is fragile just in case it has some intrinsic property or set of properties P such that if the object were dropped while in the right surroundings (intuitively, those surroundings that count as providing a test for the disposition: no protective packing, from a moderate height, onto a hard surface, and so on), its being dropped and its having P would together cause its breaking. Because this intrinsic property figures in this way in the manifestation of the disposition, we call it the causal basis of the disposition.
More generally, to have a disposition is to have a certain kind of causal power -- a causal power which an object has by having some intrinsic property that serves as the causal basis of the disposition.
The key claim of DC is that our abilities, including the abilities in virtue of which we have free will, are dispositions, with the same kind of causal structure. But before we get to free will, let's take a look at some less controversial examples of human abilities: the ability to walk, walk a straight line, play piano, fall asleep as soon as your head hits the pillow. Let’s call these kinds of abilities 'ordinary abilities'; think of them as abilities to move your body in certain kinds of ways or abilities to cause certain kinds of changes in your body. These abilities are structurally like dispositions in the following ways: We have them by having certain intrinsic properties that are the causal basis of the ability (e.g. we have the ability to walk by having unbroken legs and certain other properties of our brain and nervous system) and we don't lose them simply because our surroundings have changed. (You might object that the shackled prisoner is no longer able to walk, and there is a sense in which this is true. But the sense in which it's true is the sense in which the styrofoam-packed glass is no longer able to break.) Finally, while our abilities are relatively stable properties, they can be lost (and sometimes regained) in the same way that an object can lose (and sometimes regain) a disposition. A wet match is no longer flammable; a melted glass no longer fragile. You lose your ability to walk when you break your legs, your ability to walk a straight line when you drink too much, your ability to fall asleep as soon as your head hits the pillow when you drink a cup of coffee just before going to bed.
Some philosophers think that our powers and abilities are fundamentally different from dispositions and capacities because they think that the manifestations of dispositions and capacities are caused "willy-nilly"6 by environmental stimuli whereas it is up to us whether we exercise our abilities. There is some truth to this, but we need to be careful. First, note that one of the abilities I listed -- the ability to fall asleep as soon as your head hits the pillow -- is exactly like the dispositions I've discussed insofar as its manifestation is caused, "willy-nilly" by the environmental stimuli of your head being on the pillow. Despite this, we call it an ability, and it’s a valuable ability to have. Second, it isn't true that the manifestations of dispositions are always caused by environmental stimuli. An animal's disposition to search for food is manifested in response to the internal stimulus of the animal's hunger. The ability to play piano is an ability that we exercise in response to the internal stimulus of trying to play. And other abilities have other kinds of internal stimuli. The ability to fall asleep by doing long division with your eyes closed is an ability you exercise in response to the internal stimulus of doing long division with your eyes closed.
To sum up: Our ordinary abilities are dispositions. These abilities are intrinsic in the same way that dispositions are intrinsic: to have one of these abilities is to have some intrinsic properties that are the causal basis of the ability; so long as you retain these properties, you retain the ability. Sometimes our surroundings prevent us from exercising one of our abilities in the way that styrofoam packing prevents a fragile glass from breaking. But this isn't always or even typically the case. Often we fail to do something we have the ability to do simply because we don't try to do it. On these occasions we have both the ability and the opportunity to do other than what we actually do.
Dispositional Compatibilism is the name for a family of views. I will give a brief sketch of how I understand the view.
The most fundamental free will facts are facts about some bundle of abilities in virtue of which we think and deliberate and make decisions and choices and form intentions about what to do. Let's call this bundle of abilities 'the core free will bundle'. So long as we retain this core bundle, we retain free will, even if we are prevented, by unfortunate surroundings or loss of ordinary abilities, from moving our bodies in the ways we try to move them.
There is lots of room for argument about what abilities are required in order to have the core free will bundle, but the following are relatively uncontroversial parts of the package: the ability to form and revise beliefs on the basis of evidence and argument; the ability to deliberate for the purpose of deciding what to do in response to one's intention to so deliberate; the ability to decide what to do in response to one's trying, by deliberating, to do so.
The abilities that are part of this bundle are all dispositions.
In addition to the abilities which constitute the core free will bundle, we have various other abilities, including the ability to act on the basis of our decisions and the ability to act on the basis of intentions which are simply acquired,withoutbeing the product of any prior mental action. These other abilities are part of the free will we think we have (when we think, for instance, that having free will gives us some say in the overall shape of our lives).
These abilities are also dispositions.
We have the free will we think we have, including the freedom of action we think we have in situations where we make a choice between courses of action that are genuine alternatives for us, by having some bundle of abilities and being in the right kind of surroundings -- the surroundings that are test cases for the relevant dispositions.
For instance, while typing this paragraph, I considered whether to stop and go to the kitchen for a cup of coffee. After thinking about it briefly, I decided not to, and went back to typing. Suppose that this is the kind of normal case where I really did have the choice I thought I had. I was free to decide and act as I did and also to decide and act otherwise. It was up to me what I did.
On my view, these commonsense beliefs are true in virtue of the facts about my abilities and my surroundings. I was free to decide otherwise because I had and exercised the ability to decide, on the basis of deliberation, whether to get the coffee or continue typing. (I wasn't in the grip of any compulsive desire or subject to a post-hypnotic command, or anything else that deprived me of this ability to decide, nor was there anything in my surroundings that would have somehow prevented me from exercising this ability.) I was free to act otherwise - to get the coffee -- because I had the ability to walk and because my surroundings were favorable. (I wasn't shackled, the door wasn't locked, there really was coffee in the kitchen, and so on.)
Having some such bundle of dispositions, together with suitably friendly surroundings, suffices for our having the free will we think we have. Perhaps something else might also suffice -- some power of agent-causation that's not a disposition. But such agent-causal powers, if they exist, are not needed for free will. And the falsity of determinism is also not needed, for it is no part of this story. If you think it is needed, you need an argument. It's hard to see what could count as an argument, since dispositions don't cease to exist simply because they aren't being manifested, and since having the ability to decide whether to do something is an ability that, by its very nature, is exercised either by deciding to do that thing orby deciding not to do that thing.
Could it be argued that if determinism is true, then our surroundings are always unfavorable, robbing us of the opportunity to exercise any unexercised ability? I don't think so. While it is possible that a person is always in surroundings which prevent her from exercising one or more of her abilities in the way that the styrofoam prevents the fragile glass from breaking, determinism doesn't have this consequence. What determinism does entail is that if we don't exercise one of our abilities there was some sufficient cause of our not doing so. But it doesn't follow that our surroundings always fail to provide what's needed for a test of one of our unexercised abilities. Sometimes we don't exercise an ability simply because the stimulus event does not occur. To claim that the non-occurrence of the trying (or other stimulus) counts as a deprivation of opportunity is to reject the claim that our abilities are dispositions. Perhaps that's right, but in the absence of any other account of what our free will powers are, DC stands unrefuted.
In this section I will respond to three objections to DC.
First objection: Having free will can't be a matter of having any disposition(s). Even if there is a sense in which a lump of sugar is able to dissolve, it isn't up to the sugar whether it dissolves. Nor is it up to a car whether it starts or up to a computer whether it runs one program or another.
Reply: Having a mind is a necessary condition of having free will. Lumps of sugar, cars, and computers don't have minds.
Second objection: A dog has a mind and has dispositions, not only to behave in various ways, but also to make certain kinds of choices. But a dog doesn't have free will.
Reply: The case of dogs is less clear than it seems. If we deny free will to dogs, we must also deny it to babies and young children, and commonsense is conflicted on this point. But I'm inclined to agree that dogs and babies lack free will. DC has the resources to say the right thing, which is that we acquire free will gradually as we acquire more sophisticated abilities, including abilities to make increasingly complex choices. We learn to do things with our bodies before learning to do things with our minds, and we acquire the abilities that ground our freedom of action before we acquire the abilities that ground our free will.
The third objection is one that was widely accepted as a refutation of an older dispositional account - the Conditional Analysis.7 According to the Conditional Analysis, a person is able to do X just in case it's true that if she tried (chose, decided, etc.) to do X, she would. Here's an example of the kind of case that, everyone agreed, was a decisive counterexample to the Conditional Analysis.
Clea is an excellent cyclist, but she had a bad accident and now has a pathological fear of bike-riding, so she isn't able to try to ride her bike. But since she is in perfect physical condition she would succeed in riding if she tried. (Or so it is claimed.) We are supposed to agree that Clea isn't able to ride her bike. But the Conditional Analysis says that she can, so the Conditional Analysis is false.
Does this case refute DC?
I don't think so. The case is underdescribed, but on two plausible versions of the story, DC can say the right thing.
First version. Clea's phobia is so extreme that she won't calm down even if she somehow manages to get on her bike (perhaps with a little help from her friends). She'll panic and lose control of the bike. In this case, she lacks two different abilities -- a volitional ability and also an ordinary ability. She lacks the ability to try to ride and she also lacks the ability to ride. It isn't up to her whether she rides.8
Second version. Clea's accident left her with a phobia that is purely volitional in this sense -- she isn't able to try to ride. But if she somehow did try -- if she found herself in the saddle with her legs making, or beginning to make, the first bike-pedaling motions, her fears would melt away and she would be her old bike-riding self again.
In this case, we should say that she has the kind of ability to ride that a puppy has to chase its tail and a baby has to crawl in the direction of a bright and shiny object. She's got the ordinary ability to ride, but she lacks the second order ability to cause herself to exercise this ability (by reminding herself of her resolution to overcome her fears, taking a deep breath to calm herself, and so on). She can't make herself ride; it isn't up to her whether she rides.
We've got the free will we think we have, including real choices and genuine alternatives, by having a certain bundle of abilities (dispositions) while being reasonably fortunate with respect to our surroundings.We may continue to have free will despite losing many abilities and despite being very unfortunate in our surroundings. But determinism robs us neither of free will nor freedom of action.
Davidson, Donald, 1973. "Freedom to Act". In Essays on Freedom of Action, ed. by Ted Honderich. Routledge and Kegan Paul.
Fara, Michael, 2008. "Masked Abilities and Compatibilism". Mind 117: 843-865.
Frankfurt, Harry, 1971. "Freedom of the Will and the Concept of a Person". Journal of Philosophy 68: 5-20.
Graham, Keith, 1972. "Ifs, Cans, and Dispositions". Mind 14: 186-199.
Hobart, R.E, 1934, "Free Will as Involving Determination and Inconceivable Without It". Mind 63: 1-27.
Lewis, David, 1997. "Finkish Dispositions". Philosophical Quarterly 47: 143-158.
Manley, David, and Wasserman, Ryan, 2008. "On Linking Dispositions and Conditionals". Mind 117: 59-84.
G. E Moore, 1912. "Free Will". In Ethics. Oxford: Oxford University Press.
Perry, John, 2004. "Compatibilist Options". In Freedom and Determinism, ed. by J. Campbell, M. O'Rourke, and D. Shier. Cambridge, MA: MIT Press.
Smith, Michael, 1997. "A Theory of Freedom and Responsibility." In Ethics and Practical Reason, ed. by G. Cullity. New York, NY: Clarendon Press.
Smith, Michael, 2004. "Rational Capacities". In Ethics and the A Priori: Selected Essays on Moral Psychology and Meta-Ethics. New York, NY: Cambridge University Press.
van Inwagen, Peter, 1983. An Essay on Free Will. Oxford: Clarendon Press.
Vihvelin, Kadri, 2004, "Free Will Demystified: A Dispositional Account". Philosophical Topics 32: 427-450.
Vihvelin, Kadri, 2008. "Compatibilism, Incompatibilism, and Impossibilism", in Contemporary Debates in Metaphysics, ed. by Sider, Hawthorne, and Zimmerman, Malden, MA: Blackwell Publishing.
Vihvelin, Kadri, 2011. "How to Think about the Free Will/Determinism Problem". In Carving Nature at its Joints. Contemporary Philosophy Series, vol. 8, ed. by Campbell and O'Rourke, Cambridge, MA: MIT Press.
Vihvelin, Kadri, 2013. Causes, Laws, and Free Will: Why Determinism Doesn't Matter, New York, NY: Oxford University Press.
 Our commonsense beliefs about causation and law are vague, as are our beliefs about truth and time. We think of the future as "open" and "up to us" in some sense that contrasts with the "necessity" or "fixity" of the past, and determinism seems to be incompatible with this. But commonsense does not recognize some key distinctions -- in particular, the distinction between the claim that the future is determinate, in the sense that there are true propositions about the future, and the claim that the future is determined, in the sense that there are, in the present, sufficient causes of future events.
 Not all compatibilists are commonsense compatibilists. Some are "revisionary compatibilists". Others are "semicompatibilists" who say that determinism rules out free will but is compatible with moral responsibility. For discussion of semicompatibilism, see Chapter .
 These are not the only possible views. The hard determinist says that determinism is true, and because of this we have no free will; the impossibilist says that free will is impossible, at least for human beings, regardless of the truth or falsity of determinism. But hard determinism and impossibilism are almost impossible to believe, except when doing philosophy. (Vihvelin 2008.)
 Moore 1912, Hobart 1934, Graham 1972, Davidson 1973, Smith 1997 and 2004, Fara 2008, Perry 2004, and Vihvelin 2004, 2011, and 2013. Frankfurt famously denies the relevance of alternatives to moral responsibility, but he also claimed, in Frankfurt 1971, that freedom of will is logically independent of freedom of action and made some remarks which suggest that this is because he thinks that the disposition which grounds freedom of action is distinct from the disposition which grounds freedom of will.
 For discussion of some of the difficulties, and two very different accounts, see Lewis 1997 and Manley and Wasserman 2008.
 van Inwagen 1983, pp. 10-11.
 Moore 1912, Graham 1972, Davidson 1973.
 On this version of the case, the Conditional Answer gives the right answer since if Clea tried to ride, she would fail.