Recommended Reading

Dennett, D. C. (1984) Elbow Room Oxford:OUP

Campbell, C. A. (1957) ‘Has the Self Free Will?’ from Of Selfhood and Godhood London:Allen & Unwin. Pp 158-65, 167-79. In Perry and Bratman (1993) (2^nd ed.) Introduction to Philosophy New York:OUP. Pp. 436-45

Hume, D. ‘Freedom and Necessity’ from Enquiries Concerning Human Undersdtanding in Perry and Bratman (op. cit.) pp. 446-56

Thornton, M. (1989) Do We Have Free Will Bristol:Bristol Classical Press.

Watson, G. (2003) Free Will Oxford:OUP

McFee, G. (2000) Free Will Teddington:Acumen

What is Free Will?

When I think about the actions that I take I have the idea that the actions are the consequence of decisions that I make, and that these decisions could very well have turned out other than they did. I decided to have cereal and tea for breakfast this morning, and I did have cereal. But I could have decided to have toast and orange juice. The choice was mine and I made it. This act of making a choice is supposed to be evidence of the exercise of my free will, because there were no constraints forcing me to choose one thing rather than the other. (No constraints = freedom.)

The idea of free will does not deny that there are influences upon our decisions from outside – after all our decisions are made in the context of our desiring to act in the world, and we do not need to deny that the world is a particular way or that the consequences of our actions will be in a certain range of possibilities (if our reasoning does not lead us astray.) We can accept for example, that our decision to drive on the left in Australia is strongly affected by the social conventions of Australians (not to mention their road rules), and, although we think that we could choose to drive on the right, we recognise that the consequences will be other than we actually desire. We might phrase it this way: we use reason to decide what to will (and yes, I realise that this sounds confused, but this is how we often seem to think of it), and yet we don’t think we are really constrained by the discoveries of reason because we don’t have the impression that we couldn’t act ‘unreasonably’ if we wanted to.

Moreover, if we accept this sort of view of the relation between free will and reason and contexts of action, then it shouldn’t really matter to us how powerful these external influences are going to be. If we were living in Afghanistan under the Taliban, say, the consequences of humming a little tune in public would be so serious that we would never decide to indulge ourselves in that way. No woman would ever decide to laugh in public, and no man would ever decide to shave in the morning. That doesn’t diminish in any way our ability to make those decisions. Indeed, the very fact that we have such strong and clear pressures upon us to make certain decisions makes us all the more aware of our unexercised capacity to deny those demands. You may recall that Sartre is famous for having declared that he never felt so free as when France was occupied by the Germans, for at that time he was able to say no to the occupiers. (I don’t have any examples of occasions where he said no, but I’m sure there must have been some.)

One of the reasons why we generally want to have FW is because it allows us to justify moral judgements. After all we can hardly blame someone for their behaviours if their behaviours are not due to actions taken under their own free will. Similarly, for those who have faith in God, there can be no question of God punishing wrongdoers if they are not blameworthy, for that would be unfair, and God cannot be unfair, and they can’t be blameworthy if they haven’t acted from their own FW.

Determinism

There’s a doctrine that has been thought to make real difficulty for the idea of FW. This is the doctrine of determinism. Determinism is the claim that what happens at time t is completely determined by the state of the world at time t-1 and that there is no possibility of things turning out some other way. This doctrine became very popular after Newton, who showed that the actions of the heavenly bodies (so far as they were able to know them) were completely predictable according to his theory of gravitation.

We can see why this would be. Our experience with simple physical systems (don’t ask what makes them simple) is that their behaviour is largely predictable from a knowledge of their initial conditions. Suppose we have some billiard balls sitting on a table and we give the cue ball a whack. Then it will move about the table bumping into cushions and other balls for a certain length of time and the other balls that are set in motion by it will eventually settle down in a new pattern upon the table. Now suppose we set up the table again in just the same way as before and strike the cue ball in exactly the same way as before, then we will expect the final disposition of the balls to be exactly as it was for the first experiment. In fact we take the final state to be completely determined by the conditions of the initial state. It is just this belief that makes us think that we can predict where the balls are going to be after we’ve made our play, because we believe that we also know the rules by which the initial state is transformed into the final state, and that we can apply them. But the theoretical basis for this predictive power is not limited to simple physical systems: it is supposed to apply to all physical systems in the universe. We believe it applies to weather systems, for example, even though we have not the slightest hope of accurately predicting tomorrow’s weather, because we do not have sufficient information to fully describe the initial conditions and the transformation rules are insufficiently well known.

We should emphasise that determinism is a methodological assumption that we make because we think that it is a better template to apply to our studies of the world than other assumptions. By assuming determinism we have achieved a great deal of understanding of how the world works. This understanding has been gained by assuming that determinism is mediated by the causal connections that exist between states, and that the transformations from one state to another are to be explained in terms of causes and effects. In fact, it may be argued that we cannot even conceive of a world in which the states and conditions that are observed are not caused. Whether that means that such a world is impossible is yet another question! However, every state is actually observed to be the result of a transformation from some other state. It does not happen that states simply appear without their being in some way the result of previous states. This is a necessary condition for determinism – which refers to the determination of one state by another.

Determinism versus Free Will

The difficulty for FW arises when we realise that, according to all we know, human beings are also physical systems. This means that if person A is in a certain initial state, in a certain initial context, that this will eventually transform into another state (call it a final state) according to some appropriately complex set of rules. Moreover, if the universe could be rewound so that person A was again in the initial condition, then person A would again transform into that same final state.

But now consider that the initial condition to which I am referring is the situation of a fellow just about to have his breakfast The final situation is one in which he is having cereal and a cup of tea. It looks as if this is going to be completely determined, because the description of the initial conditions are all in physical  terms and the description of the final conditions are likewise, so this person is in a physical system that has transformed according to fixed (possibly knowable) rules, and there is never a possibility that the transformation would have occurred in any other fashion or with any other result. What, then, was free about the will of the fellow as he stood before the open door of his pantry wondering whether to reach for the Nut Feast box or the Brumbies bag? The determinist says: Nothing. The impression we have of the freedom of our will can be no more than an illusion, like the impression that the sun goes around the Earth. Knowledge of the true state of things removes our belief in that illusion, even if the illusion remains – both for FW and for the moving sun.

Note that it is not denied that our choices are made by us. But our choices are a consequence of the physical mechanisms upon which our mental actions supervene. Therefore our choices are quite as much determined as are the transformations of the physical systems. The causal chain that we are assuming is the medium for determinism stretches from the past when we didn’t exist into the present where we are about to make our decisions, and every state along the way is fully determined by the prior state, so our decisions are determined by states that existed before we were in the world; and so, as we cannot claim to be able to control those states, we cannot claim to control any of the successor states.

The Controversy

The determinist argument has the form:

P1.       Our choices are the product of deterministic laws and initial conditions.

P2.       If our choices are the product of deterministic laws and initial conditions then we don’t have Free Will.

----------------------------------------------------------------------

C.         We don’t have Free Will.

This argument is a valid deductive argument, so if we’re going to dispute the conclusion we will have to establish that one or the other of the premisses is incorrect. The types of reaction to this argument can be classified in a rough way by taking note of the attitude taken to the premisses and conclusion of this argument. There are four positions that are best recognised.

Let’s see what these positions involve.

Libertarianism

Libertarians argue that if we can’t have determinism and free will, then we should do without determinism. But can we justify doing without P1? Sure, P1 is only an assumption, but it has plenty of empirical support so we’d really like a proof of its falsity. If C was false and P2 was true then we’d know that P1 was false. But what are our reasons for believing those? So far as the falsity of C goes, that is to say the truth of our having FW, the main argument for that is just that we really, really think that we do, or that we really, really want it to be so for moral reasons. We’ll talk later about arguments for and against P2.

Now, if P1 is false, and especially if our decisions or our will is uncaused, then it seems that we are forced to hold to some sort of indeterminism. Often, these days, they can point to the sorts of hypotheses made in Quantum physics to make indeterminism plausible – such as the Heisenberg Uncertainty Principle. But it is not at all clear that if our decisions are made without cause that this is really the sort of ‘free’ will that we’re interested in having. If our decisions are not caused by ‘us’ then they are not in any way ‘our’ decisions, and therefore if they are what motivate our actions we are the mere puppets of these undetermined events. It seems that we would be equally unlikely to blame someone whose decisions were spontaneously generated ex nihilo as someone whose decisions were fully determined. (In either case we can’t blame – that makes it seems like there’s something wrong with our notion of blame. But pass that by.)

Compatibilism

All soft determinists are compatibilists, but not all compatibilists are soft determinists. The compatibilist claims that determinism and free will could both be true (not P2), and we do have FW (not C), but need take no position on the truth or falsity of P1. The focus of the compatibilist is in opposition to the incompatibilist – the person who believes that P2 holds. Very well, then. Why should we believe P2.

The arguments generally go something like this:

P3.       If we have FW then X

P4.       If determinism is true then not X

-----------------------------------------------------

C2.       If we have FW then determinism is not true, and if determinism is true then we don’t have FW.

The most common argument appealing to this form is one that appeals to our belief that we can do otherwise than we do. It’s also the argument that the compatibilist has most often attacked.

P3₁.      If we have FW then we can do otherwise than we do

P4₁.      If determinism is true then we can not do otherwise than we do

-----------------------------------------------------

C2₁.      If determinism is true then we don’t have FW.

Which we get from replacing X in the argument schema above by ‘we can do otherwise than we do’.

Now the compatibilist wants this conclusion to be false, but the argument given is formally valid, so the only way that the conclusion could fail to be true is if one of the premisses was false. The compatibilist claims that the only way that we can think that P3₁ and P4₁ are both true is for us to understand ‘we can do otherwise than we do’ differently in each instance. They thus claim that the argument is a fallacy of equivocation. (Effectively they are saying that we have replaced X in P3 by something meaning A, and X in P4 by something meaning B.) If we understand ‘we can do otherwise than we do’ in one uniform way then one of the premisses will turn out to be false. For example; if ‘we can do otherwise than we do’ means ‘if we choose otherwise we will do otherwise’ then we get

P3₂.      If we have FW then if we choose otherwise we will do otherwise.                       - true

P4₂.      If determinism is true then if we choose otherwise we will not do otherwise.     - false

And the compatibilist says that this is quite a general result. We can’t make both premisses true with the same substitution for X. So the conclusion doesn’t follow.

And so the argument continues with ever-increasing complexity, but those are the main points to be understood and this claim and counterclaim of equivocation is the main strategy being employed.

Other Arguments Against Free Will

The determinism argument against FW is the most significant, and the one that has been the focus of the greater philosophical strife, but it is far from alone. Here are a couple of others.

The True Future

A rather disturbing argument is ultimately due to Aristotle. If we say that yesterday there was a sea battle, this is a statement about the world which is either true or false. If we say that tomorrow there will be a sea battle, this is also a statement about the world which is either true or false. (Just consider: on the following day we observe a sea battle to occur and we decide that the statement was true.) And because it is either true or false the future is fixed. And if the future is fixed then we cannot do otherwise than we are going to do. And so we do not have FW.

The response to this is that the future statement is not true or false because its truth-maker is non-existent at the time of statement.

Predictability

1.             God knows what we’re going to do – because He’s omniscient – so we have no freedom of action, so we have no FW.

2.             We can tell what people are going to do in most cases. (We rely upon this to a phenomenal degree, just consider what driving would be like if we had no idea what other people were likely to do.) This indicates that their behaviour is constrained and not free, so they don’t have FW.

3.             Psychologists are able to make predictions of how people are likely to behave. Therefore we don’t have FW.