Introduction to Philosophy


compatibilism - one version
three examples of contrasts
digression: importance of analysis and clarity in philosophy
types of constraint
'free' doesn't mean 'uncaused'
compatibilism related to other views
hard determinism v. compatibilism
examination of compatibilist argument
making a distinction and meaning
argument for 'free' not meaning 'uncaused'.
an argument against compatibilism
hypothetical 'could'
the anti-compatibilist response
introduction to clip from hamlet
a further issue: do causes compel?
review quiz

The attempt to reconcile freewill with scientific determinism

[OH Plan OH3:1]


We have looked at some of the ways in which particular sciences - cognitive science and evolutionism/historicism have been our examples, represent applications to the human being of the scientific assumption that everything is caused.

We have been taking it so far that insofar as these sciences are successful they show that free will is illusory. They are not all successful of course. But there are at the present time, as there always are, one or two candidates that look very promising, that carry the scientific banner high. Cognitive science is perhaps the outstanding prospect now. It is new, carried by the explosive development of cheap and powerful computing, and disillusion has yet to set in. But - early days yet.

Here's an example of the view that scientific psychology leaves no room for 'free-will' proper, dated 1874.

But is it true that the success of cognitive science, or the success of historical materialism, or the success of any scientific perspective on the human being is incompatible with human freedom, human autonomy?

A.J.Ayer pic courtesy CUNY

You are asked to study the defence of compatibilism mounted by A.J. Ayer in the article of his ('Freedom and Necessity') reprinted in the Reader, pp.481-486. Having set the scene, Ayer's defence of compatibilism specifically starts on page 484 ("Let it be granted...") and runs to the end.

Now I want to put to you a line of argument which has allowed people to say No to this question.

- a line which people have argued establishes that you can in fact both have your cake and keep it for another day.

- an argument for the conclusion that just because you have a scientific account of human behaviour it doesn't mean that humans aren't free.

[OH Compatibilism]

return to top

Compatibilism - one version

Summary: 'Free' doesn't mean 'uncaused' - it means 'unconstrained'. There are plenty of acts that are unconstrained and therefore free. Determinism does not undermine our belief in human freedom.

What we mean by 'free' is unconstrained.

It is wrong to assume that 'free' is the opposite of 'determined'.

Think what we mean in ordinary circumstances when we say that an act was performed freely.

Three examples of contrasts

I will give three different examples of cases in ordinary life where we use the contrast between free and not free.


A man is hanging 3 floors up above the street. He is being held by the wrists by a woman who is leaning out of a window. And then you see him fall.

The woman is asked: why did you let the poor fellow go?

'If you had only held on till the fire brigade got there he could have been saved'

She replies:

'I had been hanging on to him for nine minutes and my grip in the end just gave way.'

We would say we had a case where the person's action in letting the guy go was not a free one.

It was constrained, forced.

It would have been different if she had swung the guy a little and then let him go immediately,

saying I could save you, but I won't. You are a stinker and my only regret is that I can do very little

to draw out the length of time it will take for you to die.

This would have been a free action, and one we would hold her responsible for.

But if she holds on and holds on and in the end he slips out of her grasp because she just can't

hang on to him any longer - this then is not a free action on her part.

Pic courtesy

A wealthy magistrate calls into the supermarket. The person has fifty pounds in cash on them, but what he is observed to do is to slip six tins of baby-food into the pockets of his large overcoat and walk out.

He has stolen goods worth a couple of quid.

He had plenty of cash. On him. It turns out that there is no baby in his life, and never has been.

Some people at any rate are likely to think of his action in shoplifting as being irrational.

Shoplifters often seem to be suffering from psychological compulsion.

Pic shows Winona Ryder arriving in Court last year to answer shoplifting charges. Courtesy CNN

They will say he must have been acting under some sort of compulsion, - that he didn't know what he was doing, - had no control over what he was doing, - that he wasn't responsible for what he was doing.

He must have been suffering from some psychological condition which made him do this silly, irrational, senseless, thing.

And they will argue that he must not be treated as fully responsible for it. He is ill, not bad.

Plenty of shoplifting cases will not be thought of in the same way.

Just think of somebody who shoplifts but who actually needs the stolen goods and who has no money to pay for them.

Think of gangs who attack a store in a well-thought-out professional way and make a living out of the technique.

It is argued (eg by Kai Nielsen, article reprinted in White's Introduction to Philosophy) that we use the notion of being free to act to make the contrast between cases of these two kinds.

The professional shoplifter we think of as free to choose that way of making a living, free to choose that line of action; whereas the magistrate is judged not to have been free, to have been under some kind of incapacitating psychological compulsion, forcing him to act in way that he would certainly not freely choose if a choice was left him.

[Darrow case like this.]

Let us take another example.

Before that though:

DIGRESSION: importance of analysis and clarity in philosophy


I am labouring this point because,

(a) a lot has been made of it and

(b) I want you to see it clearly,

see that it is in itself simple,

a claim that you can understand clearly and assess

- once it has been separated out from everything else and put clearly. A large part of philosophy is of course doing just that

- analysing a morass of thought or discussion,

a morass of considerations

and separating out all the different claims that are being run together, separating them out and articulating them clearly

- putting them in clear, straightforward terms that others can understand.

I hope you will have found this first process going on in the seminars a bit - a morass, but some particular points or arguments getting better definition as the argument proceeds.


But I warn you: when you begin to think seriously about a philosophical topic for quite a while the morass, the confusion, gets worse!

You have more thoughts, you read more stuff, and you get more and more depressed.

You just have to trust me when I say: there comes a point, if you persist, when a light appears at the end of the tunnel, you begin to feel some control over the different points and how they relate.

So: a third example of the distinction we make in ordinary life between actions that are free and those that are not free.

[OH Three suggested (OH3:3)]

[First example: you can't hold on any longer

second example: psychological constraint (shoplifting?)

third: action under pressure (duress)]

Bank heist. Pic courtesy Touched by an Angel


A bank manager gives thieves the key to the vault because they have kidnapped his child and threatened to harm the kid if the keys aren't handed over.



See if you can identify one clear case of where a member of your group did something they thought was freely done at the time but which turned out later not to have been.


[OH Identify one clear case OH3:3A]

Identify one clear case within your group of where a member of the group did something they thought was freely done at the time but which turned out later not to have been.

Summary of argument so far: So the argument so far is that in ordinary life we use the word 'free' to mark a contrast between actions which are constrained in some way, and those that aren't.

Types of constraint

The constraints we have considered in our examples are:

physical constraints - her arms were just exhausted , she physically couldn't go on holding on.

psychological constraints - he shoplifted because of a psychological condition which made him do it in spite of his actually not wanting to.

constraints of pressure - he gave them the key because they threatened to harm his child, whom they had kidnapped.

'Free' doesn't mean 'uncaused'

From the examples it is argued that 'free' doesn't mean 'uncaused'.

Think just of the second example:

There is the clever and deliberate shoplifter on the one hand and the kleptomaniac magistrate on the other.

In physiological terms, both actions are determined, both actions are caused.

In both cases, the movements of the hand as it gathers the goods to put them away is brought about by muscles stimulated into action, by discharges in the afferent nerves, which in turn are triggered by patterns of discharge in the wider nervous system, which in turn are brought about by physical promptings in the brain - and so on.

So here you have two actions, one of which we would normally call 'free' and one of which we would regard as not free.

Summary: Free doesn't mean uncaused.

[OH 'Free' doesn't mean 'uncaused' OH3:4]

What these examples show, it is said, is that when you go into it, 'free' doesn't mean 'uncaused'.

'Free' means 'unconstrained'.

And an unconstrained action is not an uncaused one.

In that case there is no incompatibility between determinism and freedom.

Determinism may be true. It may be that everything that happens has a cause. But that doesn't mean actions cannot be free.

Those who mount this argument and defend it are known in the literature as compatibilists.

[Return to OH Compatibilism OH3:zzz]

This sort of compatibilism says that 'free' doesn't mean 'uncaused' - it means 'unconstrained'.

It says there are plenty of acts that are unconstrained and therefore free.

So determinism does not undermine our belief in human freedom.


We can relate compatibilists to other parties like this:

Begin with the argument we started with, which can be written like this:

[OH 1. Determinism (says the scientific OH3:5]

Hard determinism

1. Determinism (says the scientific point of view) is true.

2. If determinism is true, then no human actions are free.


3. No human actions are free

(the scientific view has to conclude).

This is labelled in the literature the argument of the hard determinist.

The argument we have just considered, which says that 'free' means 'unconstrained' (not 'uncaused')attacks hard determinism in its second premise.

Its second premise is wrong, they say.

Determinism's being true has nothing to do with actions not being free. Free actions are unconstrained actions, not undetermined or uncaused actions.

You can believe everything is determined and still believe in human freedom.

These people are labelled in the literature soft determinists or compatibilists.

[OH Compatibilism = Soft Determinism OH3:6]

They are determinists, but they think they can believe that some acts can be free nonetheless.

[Return to OH Compatibilism 3:2]

Summary of compatibilist argument:

'Free' doesn't mean 'uncaused' - it means 'unconstrained'.

There are plenty of acts that are unconstrained and therefore free.

Determinism does not undermine our belief in human freedom.


What do you think?

Do you see difficulties with this argument?


Let me just go over their argument once again, really to convince you that it is as simple as it sounds.

The argument is:

Please notice that in ordinary life we make and use a distinction between actions that are free and those that are not free.

Then please look at the contexts in which this distinction is made, and how it is made.

You will find that the distinction we are trying to make in using the term 'free' is between actions which are constrained in some way and those that are not.

'Free' means unconstrained, therefore. It does not mean uncaused.

You can therefore have free actions which are determined, which have causes.

Examination of compatibilist argument

[OH Plan OH3:6]

I want us now to examine the compatibilist argument.


First, I want to ask about the basis of the compatibilist's claiming that in ordinary circumstances 'free' means unconstrained.

What evidence is adduced? What argument is presented?

I haven't said much in exposition.

I have just said that the compatibilist invites us to consider as follows:

When you say in ordinary circumstances that such and such an action is free, think about it and you will see you are wanting to distinguish between this action and some others that are constrained.

Is this a valid observation?

The compatiblist is saying: we use 'free' to mark off unconstrained from constrained actions, and therefore 'free' means unconstrained.

Is that a fair conclusion to reach?

Let us remember at the same time what the hard determinist said. The hard determinist has taken 'free' to mean 'uncaused'. Why shouldn't he or she be right?

The hard determinist says 'free' means 'uncaused'.

The soft determinist says 'free' means 'unconstrained'.

Who is right?

I can't settle this of course. But I want to draw your attention to an argument mounted on the soft determinist, the compatibilist, side.


The compatibilist thinks he or she can show that 'free' can't mean 'uncaused'.

Everything is caused, the compatibilist says.

We know the compatibilist says this because s/he is a determinist.

So: everything is caused.

If 'free' meant 'uncaused', it would apply to nothing.

But then, could it really mean anything?

[OH 'free' is only meaningful' OH3:7]

'free' is only meaningful if it is used to make a distinction.

If 'free' meant 'uncaused', it would apply to nothing (if determinism is right).


Therefore, 'free' cannot mean 'uncaused'.

If you have word which doesn't apply to anything, what use is it? What meaning can it have?

Descriptive words are surely used to make distinctions, to pick out some things from others, one category of things from another.

'Yellow' picks out one category of things from another. There are plenty of things that aren't yellow.

'Warped' picks out one type of thing from another. There are plenty of things that aren't warped.

'Forged' picks out one type of currency note from plenty of others. There are plenty of notes that aren't forged.

'Could all currency be counterfeit?'

Pic courtesy The Garda Bureau of Fraud Investigation

But if there aren't any uncaused events, 'free' can't pick anything out, can't pick out some events as contrasting with others can't mark any distinction among events.

If everything was yellow, would 'yellow' mean 'yellow'?

If everything were warped, would 'warped' mean 'warped'?

If absolutely every currency note were forged, would forged' mean 'forged'?

Could all currency be counterfeit?

Counterfeit notes are notes passed off as real when they aren't. But if there are no real ones …


What would you say?

Could all currency be counterfeit?


I put this argument before you because it is of a type that has had a certain vogue. You may find it in other guises, applied to other tasks.

For example, life cannot be a dream, it is argued: because you call something a dream to make a contrast between that thing and real life. If everything were a dream, you couldn't make that contrast: so if everything were a dream, the word 'dream' would lose its meaning.


You may be tempted by the following:

By parity of reasoning, everything can't be real.

(Or, might you say, everything can't be caused ....?)

Let me leave that argument there.

The compatibilist's conclusion is that if it meant 'uncaused' free would make no distinction and so would make no distinction and so would in effect be meaningless.


So far: compatibilism argues that 'free' doesn't mean 'uncaused' and so determinism doesn't rule out freedom.

An argument against compatibilism

Now an argument against compatibilism.

What do you think?

What would be your argument against compatibilism?

Big Fig Tree, courtesy Matt Jones


'I care not a fig for what 'free' means or doesn't mean,' this new attack begins. Talk about the meanings of words is just a distraction. The fact is that if determinism is true, a person could not do other than he or she in fact does: and this is what spells the end for human autonomy and responsibility.


So the critic of compatibilism says:

Let us drop this conceptual analysis of 'free'. It doesn't really get us anywhere. We can pose the dilemma in other terms if you like. The essential point is this:

whether a person who has made a decision could have made a different one. The Universe has us by the throat if we can't do anything other than what we do. That is the threat the principle of causality poses.

For surely, if our actions are all caused it is not true that a person who has made a decision could have made a different one?

The compatibilist reply

How does the compatibilist respond to this way of putting the problem? Like this:


The compatibilist says: what exactly do you mean by 'could have done differently'?

One thing you might mean is this:

If circumstances had been different, you would have done differently.

'Let us suppose your teacher said "Go to Cardiff'"'

Cardiff Waterfront, courtesy Each

Think of an example of a decision, eg to come to Lancaster.

Let us suppose


Let us suppose you arrived at your decision when these conditions A B C D E and F obtained.

You then say: I could have made a different decision.

The compatibilist says to this that there is a sense in which this is true, even if determinism is true, even if what you decided was determined, like everything else.

The sense is: if one or more of these circumstances had been different, the outcome would have been different, your decision would have been different.

If you had actually visited Cardiff instead of thinking bad of it just because the teacher recommended it, you would have accepted their offer instead.


when you say

my decision 'could have been different '


if the circumstances that obtained when I made my decision had been different, then my decision would have been different then it is true that you could have made a different decision.


you 'could have made' a different decision.

[OH Hypothetical 'could' OH3:8]


'Charles could have decided differently'

Pic courtesy Channel 4's Picture Gallery

Charles could have decided differently


If circumstances had been different Charles would have acted differently

Hypothetical 'could'

If 'could have made'


'would have made if circumstances had been different'

then it is true,

even though your decision was fully determined by circumstances, that it could have been different.

To say it could have been different is only to say if the causes had been different, it would have been different too.
J..L. Austin, was a philosopher of the mid 20th Century, at the heart of what has become known as Oxford philosophy, and also 'conceptual analysis' (insofar as that is different). He developed the analysis I'm setting out here in terms of 'can' not 'could'. ('Can we do differently?' instead of 'Could we have done differently?') In a memorable phrase, he concluded that 'all cans were iffy'.
Pic courtesy University of Pavia

We can label this the 'hypothetical' sense of 'could'.

'Could' equals 'would if'.

Can I do differently? means 'If things were different would I do differently?'


That is the compatibilist response to the argument put in terms of 'Could'.


Can you think of examples in your own life where (you feel) you made a decision in a particular set of circumstances but where you are convinced an alternative course of action was open to you? Where you chose one thing but could have chosen, in exactly those circumstances, another?


Let us take it just one step further. If this is the compatibilist position, what is the critic's response?

It is to insist that once again the compatibilist has cleverly managed to miss the point.

Hypothetical 'coulds' are not good enough - do not reflect what the critic is trying to say.

When we say a person could have done such and such we do not mean just that if circumstances had been different he would have done something else.

We mean that at the time of choice, in exactly the circumstances that obtained, there were in fact two or more paths open. The person chose one of them.

But in exactly the same circumstances, if the action is to count as something the person really chose the person could

- ABSOLUTELY could -

have done something else.

We are perhaps saying here that there are at least two sorts of 'coulds'.

One is the hypothetical:

a person could have done x


in other circumstances s/he would have done x.

The other is categorical:

a person could have done x


in the very same circumstances x might have been done instead of what was actually done.

So two 'coulds' perhaps, hypothetical and categorical.

[OH Two senses of 'could' OH3.9]

The compatibilist has two possible responses.

One is to say

There actually are not two types of 'could'. All genuine 'coulds' are hypothetical.

When you think about it the idea of a categorical 'could' is vacuous. All you can mean in saying 'He could have done differently' is that he would have done differently if different circumstances had prevailed.

[OH Two senses of 'could' OH3:10]

Introduction to clip from Hamlet

(Lawrence Olivier version)

It is tempting to think the more you think about a decision, the freer it is.

The paradigm of the free action is perhaps the one that is not done quickly or rashly or automatically but calmly, thoughtfully after careful deliberation.

What would you choose as an example from literature of someone making a difficult decision?

Here is my choice.

This is a chap who has quite a bit on his mind. There has been unhappiness in the home, his love-life is not going well and he has begun seeing things.

The decision he is on the verge of is the big one: Is it all worth it? Why not just pack it in?

Just to show I have the vestiges of a classical education, in case you were beginning to have your doubts.

Did anyone have the same idea?


One argument used by compatibilists is that it is a mistake to treat causality as somehow compelling or forcing things to happen. A.J.Ayer is one of these. You will find him putting this point in the article you are asked to read - 'Freedom and Necessity', the Reader, pp 481-486:

"... the use of the very word 'determinism' is in some degree misleading. For it tends to suggest that one event is somehow in the power of another, whereas the truth is merely that they are factually correlated." Ayer, Reader, p. 485.

One way of putting the anti-compatibilist case is to say that the causes of a decision are a sort of constraint. The inner physical state of the body, including for example, the pattern of nervous activity that obtains immediately prior to a decision is an important part of the circumstances, the conditions, which determine what the decision shall be - this is the determinist view. Ayer again is a compatibilist who objects that this is to treat physiological conditions on a par with prison bars or revolvers held against the head. It is to treat them as among the constraints or compulsions or necessities which together determine the outcome of any decision-making process.

But, Ayer says, it is wrong to conceive of causality in this way.

Laws of nature are not prescriptive in the way that laws of the land are prescriptive. They do not place natural objects under obligations. They simply set down established regularities in the behaviour of things.

'... the laws of mechanics do not compel the planets to move in their orbits; they simply describe planetary motion. Acids are not coerced by the laws of chemistry to react with alkalis. The laws simply describe their patterns of reaction.' D.J. O'Connor, Free Will, New York, 1971, Anchor, p.73.

To pursue this line we shall need to think a bit about causality in general, and to this I now turn.


Moore: All that is certain about the matter is: (1) that, if we have Free Will, it must be true, in some sense, that we sometimes could have done what we did not do; and (2) that, if everything is caused, it must be true, in some sense, that we never could have done, what we did not do. What is very uncertain, and what certainly needs to be investigated, is whether these two meanings of the word 'could' are the same. G.E.Moore, Ethics (1912), OUP New York edition p.90.



Question 1

Question 2

Question 3

Question 4



Return to top

Menu of VP's 100/200 presentations