8 Free Will, Moral Responsibility, Determinism
many argue, that the ability to do otherwise is not required for moral responsib-
ility but that moral responsibility still requires some sort of control condition. On
our proposal, we call that sort of control “free will.” Ginet, van Inwagen, and
others favoring their terminological approach would need a distinct term.
In our estimation, little rides on this. People can use the term “free will” dif-
ferently, so long as they specify at the outset what they mean by this term. We
believe it best to theorize about free will in terms of a control condition on moral
responsibility. But we grant that reasonable minds can differ. In any event,
regardless of this difference in how we use the term “free will,” philosophers can
still have a substantive—and not merely a terminological—discussion about
what sort of control moral responsibility demands, and about whether it requires
the ability to do otherwise, and whether that ability is compatible or incompati-
ble with determinism, and so on.
Next consider Robert Kane’s (1996) way of defining free will. His idea is that
when we act, we are guided by our ends or purposes. As Kane defines it:
Free will is the power to be the ultimate creator and sustainer of one’s own
ends and purposes. (1996: 4)Interestingly, in contrast with Ginet’s and van Inwagen’s proposal, Kane rejects
the thesis that we should understand free will simply in terms of the ability to do
otherwise, or the ability to will otherwise. Nevertheless, on his developed view,
being an ultimate creator or sustainer of one’s own ends entails that one is able
to do otherwise. So, for Kane, free will does involve the ability to do otherwise,
but it amounts to a substantive and not merely a definitional matter that free will
requires this ability.
In our opinion, what it is for an agent to be a creator or sustainer of her ends
and purposes—even an ultimate creator or sustainer—can be understood in
stronger and weaker terms. At one end of the spectrum, it might come to mean
no more than what is meant when it is remarked that, once the ball was snapped,
it was ultimately up to the quarterback to make the pass. That sense is consistent
with very modest demands on agency. The other end of the spectrum features a
reading of being an ultimate creator that is so demanding that there may be no
metaphysically possible way for such a condition to be satisfied. The virtue of
defining free will in terms of a condition on moral responsibility is that it is then
possible to gain some independent purchase for measuring how strong or weak
the control or freedom must be. Hence, it might very well be, as Pereboom
argues (2001), that free will does require being an ultimate initiator of one’s free
acts. But this will flow as a substantive thesis from examining the conditions on
moral responsibility.
One way to handle these terminological disputes is just to avoid use of the
term “free will” altogether. Better, it might be thought, to adopt more theoreti-
cally innocent terminology. This is how John Martin Fischer and Mark Ravizza
(1998) proceed. Rather than offering a definition of free will, they write instead
in terms of control over action. In setting out their view, they distinguish