Since the publication of An Essay on Free Will, it has
become increasingly to me that free will is a philosophical mystery—something
that philosophers do not understand at all. (it is not the only one. For
example, no philosopher understands conscious experience of the apparent
“passage” of time.) I do not mean to imply that imply that free will is a
mystery in the theological sense: something that is beyond all possibility of
human comprehension. That may or may not be the case. I contend only that as of
this date, no philosopher has achieved an understanding of free will. That may
be because free will is indeed something that human beings are incapable of
understanding, but it may be because we human beings have not yet discovered
the right way to think about free will. I will lay out the essence of this
mystery in four fairly simple statements—labelled ‘first’, ‘secondly’,
‘thirdly’, and ‘fourthly’. First, there are excellent arguments for each of the
following three propositions:
If antecedent conditions are the laws of nature determine the way
in which a human being shall act at a certain time, then the person’s act at
that time is not free. (This proposition, of course, is the proposition
commonly called ‘incompatibilism’.)
If antecedent conditions and the laws of nature do not determine
the way in which a human being shall act at a certain time, then that person’s
act at that time is not free.
If a human being’s acts are never free, then the consequence of
those acts are not the fault of that human being.
Secondly, the following fourth proposition seems to be true beyond
all possibility of dispute:
Some of the consequences of some of the acts of some human beings
are their fault.
Thirdly, these four propositions form a logically inconsistent
set, and, therefore, either the excellent arguments for at least one of the
first three propositions must contain some flaw or else it must be that (to
take one example among many millions of compelling examples) the deaths of six
million Jews in the extermination camps were not anyone’s fault. Fourthly, no
one knows of even a plausible candidate for a flaw in any of the arguments for
the first three of the four propositions (not, at any rate, in the arguments I’m
thinking of), and to den the fourth would be simply bizarre. I doubt whether
many philosophers will agree with my statement that “no one knows of even a
plausible candidate for a flaw in any of the arguments for the first three of
the four propositions,” but it represents my considered judgment. My own view
is that there is a flaw in the argument (the argument that I think is
the best argument for this conclusion) for the proposition ‘If antecedent
conditions and the laws of nature do not determine the way in which a
human being shall act at a certain time, then that person’s act at that time is
not free’. But I haven’t any idea what this flaw might be.
I would recommend that the “problem of free will” be understood as
follows: it is the problem of discovering a flaw in at least one of the
arguments for the first three propositions—or else, of explaining how the
seemingly self-evident fourth proposition, could, despite all appearances, be
false. In my judgment, no one has the least idea how to solve this problem.
That is what I mean by saying that “free will is a mystery” (Peter Van Inwagen,
“Author’s Preface to the French Translation of An Essay on Free Will,” in Thinking
about Free Will [Cambridge: Cambridge University Press, 2017], 181-83)