Four Thoughts on Free Will

I am lecturing on free will this week.  This is one topic that is logically well organized in the history of modern philosophy.  There are two fundamental questions: are human choices determined by the past, and can human beings be morally responsible for their choices?  These give rise to four possible positions. 

Hard Determinism.  This position may be described by the following argument:

1.       If human choices are determined by the state of the Kosmos (D) prior to those choices, then free will is impossible (~F). 

2.      D. 

3.      Therefore: ~F. 

That is a straight forward modus ponens argument, and so it is valid.  If you do not want to accept the conclusion, then you have to reject one of the two premises.  Here F can be understood as metaphysical freedom (uncaused action) or moral freedom (an action for which the actor is morally responsible).  The hard determinist rejects both kinds of freedom. 

The strength of this view is that it preserves the principle of the uniformity of nature, if that principle implies determinism as many thinkers suppose. 

Libertarianism.  This position may be defined by a modus tollens which begins with the same premise:

1.       If human choices are determined by the state of the Kosmos (D) prior to those choices, then free will is impossible (~F). 

2.      F. 

3.      Therefore: ~D.

Again, this is a valid argument.  As hard determinists boldly state D, libertarians boldly state F.  Libertarians conclude that determinism is false, at least as applied to human beings, who they believe to be agents capable of uncaused action. 

The strength of libertarianism is that it preserves our common sense experience of choosing.  My sense of having choices is as vivid as my sense that the world is solid underneath my feet. 

Because they share premise 1, the two positions are both called incompatibilist.  Hard Determinism and Libertarianism both regard determinism and free will to be incompatible. 

Compatibilism.  The third view, perhaps the dominate one in philosophy, rejects premise 1 above and may be expressed in a simpler argument. 

1.       F and D. 

2.      Therefore F.

3.      Therefore D. 

Again that is valid, if tautological.  Compatibilists believe that free will is not only logically compatible with determinism, but that it actually requires it.  Why do we praise a person as good because of her good actions if not because her good character determined those actions?  It is likewise with blameworthy actions. 

Compatibilists think that hard determinists and libertarians are focusing on the wrong things.  Am I free to leave the room I am sitting in?  To answer that question we don’t need to ask whether my decision to leave is determined by the state of the cells in my brain prior to my decision.  We need to ask whether the door is locked or not, or whether I know that someone is waiting outside to do me harm. 

Human actions are free, in this view, if at least two conditions are met:

1.       The agent is not compelled to act or not act (either by physical constraints or compelling threats) and

2.      If the agent had acted otherwise, she would have done otherwise. 

I suspect that condition one is contained in condition two.  The question is not whether my choices are caused by prior states of the Kosmos (they are, according to the compatibilist) but whether my choices are causes of my actions.  To the extent that the latter is true, I am free.

The strength of Compatibilism is that it saves moral responsibility by immunizing the same against determinism.  Why do we seem largely agree to hold sane people responsible for the crimes that they commit while holding that insane people are not responsible?  This is because sane people respond in more or less predictable ways to moral sanctions.  They understand that they will be punished for infractions and so are less likely to commit them.  Insane people are incapable of a rational response to such sanctions.  Precisely because sane people are more or less responsive to legal sanctions (along with moral opprobrium) such moral sanctions make sense.  There is no point in applying them to a full tilt loon. 

It occurs to me, however, that there is a fourth position.  This is in fact my position.  This position is compatibilist in so far as it denies that determinism and free will are incompatible.  I think that what matters is that my choice determine my action, regardless of whether my choice is determined by the past.  In that respect, I agree with the compatibilists. 

However, I don’t see any reason to accept determinism.  No reasonable person would deny that the past influences the future.  If yesterday you offered me a lot of money to write an essay, I am more likely to write it today.  However, influenced and determined are never the same.  If X influences Y in a very robust way, then Y is predictable from X.  That is the kind of relationship that science mines.  It is enough to calculate the influence.  Does X result in Y 99% of the time?  That is a robust finding.  Science rarely if ever gets to %100.  All or almost all science is probabilistic rather than determinist. 

It is common to assume that this is because of limits on the data or apparatus.  We can only measure anything to within some margin of error.  But why assume that?  The assumption is unnecessary.  This state of X makes the subsequent state of X more or less probable.  That is the best we can ever do. 

Determinism is one of the great myths of modern science.  It is an attractive idea, useful to some degree in thought, and altogether unfounded in reality.