254 Contemporary Incompatibilism: Libertarianism
to stay in Mayberry or move to New York, it seems that it can’t be a desire to act
in accord with the reasons that settles which decision occurs, since in Balaguer’s
(2010) example the reasons are in equipoise. But it’s evident that the agent can
settle which action and decision occur. So it appears that the role of the agent
can’t be played by a desire to act in accord with reasons. And here agent causa-
tion, as in the disappearing agent problem for event- causal accounts of free will,
would seem to yield a solution.
But so far we don’t have a reason for adopting the libertarian version of agent
causation. For agent causation is, arguably, compatible with the causal determi-
nation of action (Markosian, 1999, 2010; Nelkin, 2011; Pereboom, 2015b).
Nida- Rümelin (2007) contends, however, that full- blooded agency, understood
as involving active causation of intention, precludes causal determination.
And Steward (2012) argues that the sort of settling required is incompatible
with causal determination, and that libertarian agent causation is essential to
agency:
If determinism were true, the matters in question would already be settled,
long before it even occurred to me that I might, by acting, come to settle any
of them. And surely it is a condition of being truly able to settle something
that it has not already been settled in advance of one’s potential interven-
tion. If determinism were true, then, I would not be able to settle matters
that it is essential for me to be able to settle, if I am to be an agent. And so,
if determinism were true there could not be agents and there could not be
actions. (Steward, 2012: 39)Steward allows that we have a weak notion of settling that does not require
indeterminism:
One might perhaps speak, for instance, of the fall of the third domino’s
having settled that the fourth would fall, even in a context in which one took
it for granted that the fall of the fourth was already guaranteed by the fall of
the first (or indeed by events and circumstances occurring long before the
fall of the first). (2012: 41)But weak settling is insufficient for action (2012: 41–69), and this claim has
intuitive pull.
However, Steward’s argument restricts the determinist to a state or event-
causal theory of action, and the concern she raises is that such an account cru-
cially features “the disappearance of the agent” (2012: 62–9). But Steward does
not address deterministic agent causation, and it is arguably enough to yield the
requisite notion of settling (Pereboom, 2015b). A profitable way to conceive of
settling is as a kind of difference- making. Carolina Sartorio (2013) proposes a
determinism- friendly event- causal account of the sort of difference- making
required for moral responsibility, and it can be modified to yield a deterministic
agent- causal account of the kind of settling required for agency. On Sartorio’s