Mesh, Reasons-Responsive, Leeway Theories 221due to certain especially strong reasons not to take it. He would refrain because
he would get more “energy or focus” in the face of these strong reasons (1998:
74). The drug user, the objection contends, is not reactive to most reasons not to
take the drug, and so is not morally responsible for taking it, despite the fact that
he would respond to an especially strong incentive. This would then suggest—
contrary to Fischer and Ravizza—that reactivity is not all of a piece, and thus
that reactivity to some reasons to do otherwise does not confirm a mechanism’s
general capacity to react to all reasons to do otherwise. Fischer and Ravizza’s
reply is to claim that, while such cases are possible, in the scenario in which
such an agent gets more energy or focus, a different mechanism is at work (74).
However, the worry is that they have no principled basis for mechanism individ-
uation, and thus it is hard to see what could license their contention (McKenna,
2001). Why couldn’t an agent act from the same mechanism and get more
energy or focus only in the presence of (for example) especially strong reasons?
Indeed, this seems to describe Mele’s case of the agoraphobic.^14
On the matter of mechanism individuation, Ishtiyaque Haji’s similar reasons-
responsive compatibilist theory compares favorably (1998). Haji also develops
an actual sequence, reasons- responsive view, and considers which aspects of an
agent are to be held fixed in settling whether, in acting as she did, she was appro-
priately reasons- responsive. He proposes that what we hold fixed is an agent’s
proximal desire, the one immediately preceding action, and also the motivational
base from which this desire gains its relative strength (1998: 75–8). In the case
of Mele’s agoraphobic, we would get the right result: Set the house on fire and
you’re bound to alter his motivation base. Nevertheless, when evaluating
whether he is reasons- responsive in remaining in his house and not attending his
daughter’s wedding, holding fixed his actual desires and motivational base (and
thinking about this as at least partially constitutive of his mechanism), if no
reasons would persuade him to leave, it seems his phobia can be regarded as
freedom- defeating, which yields the proper result.
Some might see haggling over the details of mechanism individuation as
involving minutia with little bearing on Fischer and Ravizza’s larger enterprise.
That would be a mistake. The viability of their view depends on this notion.
Without it, they cannot give a mechanism- based account of reasons-
responsiveness. But by their own lights, they cannot, as some have advised they
should (e.g., Ginet, 2006), fall back on an agent- based account without having to
concede that agents in Frankfurt examples are not reasons- responsive. And
without the contention that agents in Frankfurt examples are reasons- responsive,
they cannot analyze guidance control (source freedom, as we would put it) just
in terms of reasons- responsiveness.
There are two options for Fischer and Ravizza and other reasons- responsive
theorists, if the costs of retaining a mechanism- based theory are too high. One is
to agree to an agent- based reasons- responsive theory and also to commit to
leeway freedom as a requirement on moral responsibility. Nelkin’s and Wolf ’s
positions (Section 8.5) count as examples of such a view. More recently, in
coauthored work with David Brink, Nelkin has developed in more detail the