1 Schoeman, Ferdinand, “Responsibility and the Problem of Induced Desires,” Philosophical Studies, 34 (1978): 293–301, p. 293.CrossRefGoogle Scholar

2 Kane, Robert, Free Will and Values (Albany: SUNY Press, 1985), p. 37.Google Scholar

3 The theory developed by John Fischer and Mark Ravizza is history-sensitive. See Fischer, John Martin, “Responsiveness and Moral Responsibility,” in Responsibility, Character, and the Emotions, edited by Schoeman, Ferdinand (Cambridge: Cambridge University Press, 1987), pp. 81–106Google Scholar, and Fischer, John Martin and Ravizza, Mark, “Responsibility and History,” in Midwest Studies in Philosophy, 19 (1994): 430–51.Google Scholar Alfred Mele defends a theory of individual autonomy that is history-sensitive in Autonomous Agents, forthcoming from Oxford University Press.

4 Frankfurt, Harry G., The Importance of What We Care About (Cambridge: Cambridge University Press, 1988), p. 54.CrossRefGoogle Scholar

5 See Schoeman's “Responsibility and the Problem of Induced Desires.” References to Schoeman in the text are to this work.

6 Such cases are also extensively discussed by Mele in Autonomous Agents. Mele relies on such cases, among other things, to argue for a history-sensitive theory of individual autonomy.

7 I am assuming that, although normal (and autonomous) agents are sometimes unable to shed or attenuate cherished values that issue in intentional conduct, it is implausible that this inability precludes their being morally responsible for that conduct. Some support for this assumption derives from reflection on “Frankfurt-like cases” discussed in the text below. The crux of the idea here is this: Assume S's value, V, frequently issues in intentional action. Assume, also, that, were S to show on any occasion any inclination whatsoever to shed V, an irresistible imp would prevent S from shedding V and would compel S to act on V. It so happens that S never attempts (on S's own) to shed V (and so the imp never intervenes). I submit that (other conditions for moral responsibility being satisfied) S is morally responsible for conduct deriving from V although V is unshedable. S acts on V just as S would have had no imp been around.

8 Frankfurt, Harry G. develops such an example in “Alternate Possibilities and Moral Responsibility,” Journal of Philosophy, 45 (1969): 839–45.Google Scholar The second disjunct needs at least this qualification: the agent did not (willingly) see to any such desire being instilled in her.

9 Alfred Mele calls such Frankfurt-type cases “expanded Frankfurt-style scenario.” See Autonomous Agents, chap. 8, part 4.

10 Double, Richard, “Puppeteers, Hypnotists, and Neurosurgeons,” Philosophical Studies, 56 (1989): 163–73. References in the text are to this work.CrossRefGoogle Scholar

11 Slote, Michael, “Understanding Free Will,” in Moral Responsibility, edited by Fischer, John Martin (Ithaca: Cornell University Press, 1986), pp. 124–39, p. 137.Google Scholar

12 Watson, Gary, review of Dennett 1984, in the Journal of Philosophy (1986): 517–22.Google Scholar

13 Robert Kane, Free Will and Values, p. 37.

14 (ii) has been questioned by Slote in “Understanding Free Will” and by Audi, Robert in Action, Intention, and Reason (Ithaca: Cornell University Press, 1993), pp. 184–86.Google Scholar

15 The account is also to be found in Double, Richard's The Non-Reality of Free Will (New York: Oxford University Press, 1991), chap. 2.Google Scholar

16 See Mele, Alfred, Autonomous Agents and Springs of Action (New York: Oxford University Press, 1992).Google Scholar

17 This point is controversial. For considerations in its favour see, e.g., Fischer, John Martin, The Metaphysics of Free Will (Cambridge, MA: Blackwell Publishers, 1994), chap. 7.Google Scholar For considerations against it see, e.g., Inwagen, Peter van, An Essay on Free Will (Oxford: Clarendon Press, 1983), chaps. 3 and 4.Google Scholar

18 For further discussion on the notion of volition control, see my “Autonomy and Blameworthiness,” Canadian Journal of Philosophy, 24 (1994): 593–612.CrossRefGoogle Scholar My construction of the concept of volitional control has been influenced by the discussion of reason's responsive mechanisms in Fischer's “Responsiveness and Moral Responsibility.”

19 The notion of psychological continuity originates with Derek Parfit. See Reasons and Persons (New York: Oxford University Press, 1986), pp. 205–206.Google Scholar

20 Parfit, Reasons and Persons, p. 206. Parfit suggests that strong psychological connectedness is the preservation over time of at least half the number of psychological connections that normally hold, over every day, in the lives of actual persons.

21 For discussions on this point, see Strawson, P. F., “Freedom and Resentment,” Freedom and Resentment and Other Essays (rpt. London: Methuen, 1974), pp. 1–25;Google ScholarRussell, Paul, “Strawson's Way of Naturalizing Responsibility,” Ethics, 102 (1992): 287–302;CrossRefGoogle Scholar and Fischer, John and Ravizza, Mark's introduction to Perspectives on Moral Responsibility (Ithaca: Cornell University Press, 1993).Google Scholar

22 I emphasize that I am using a technical notion of ‘normative agent’. One should not confuse, say, Clinton with a normative agent and then balk at the (implausible) suggestion that Clinton has his evaluative scheme essentially. Seemingly, Clinton could have had a very different evaluative scheme than he in fact has.

23 I am indebted to Alfred Mele for the suggestion that IPA2 may have to be supplemented with the following: “First, A is not causally over-determined-it is false that S would have done A even in the absence of its cause in PA. Second, A's cause in PA is not morally irrelevant. It is not the case, for example, that supposing that although a part of A's (nonover-determining) cause is in PA, that cause plays a very minor role—it influences only slightly the speed of S's A-ing in a case in which A's speed is morally irrelevant.”

24 The example is Alfred Mele's; see Autonomous Agents, chap. 10, part 3, for an insightful discussion of such examples.

25 Alfred Mele proposes (Ibid.) that to the extent to which an “engineered” deliberative habit is at work in a deliberative process, the agent is not deliberating autonomously, where engineered deliberative habits amount, roughly, to the purposeful instilling of a habit in an agent in such a way that her relevant capacities for control over her mental life are bypassed in the process. Since there may not be a neat overlap between an agent's autonomously A-ing and an agent's being morally responsible for A-ing (see my “Autonomy and Blameworthiness”), one might well accept Mele's proposal without endorsing the further view that the agent with engineered deliberative habits cannot be morally responsible for an action that issues from “engineered deliberations.”

26 Some might raise this objection: why can we not consider scenarios in which Ernie has this desire (the motivational precursor of A) but was never brainwashed (in some such scenarios he decides not to A)? After all, regarding Frankfurt-type scenarios, you allow us to consider alternative scenarios in which the interveners are absent. Why, in our search for scenarios, are we allowed to subtract the Frankfurt interveners but not the pertinent aspect of the agent's history? Briefly, my answer amounts to this: clearly, our own histories exert heavy leverage on shaping our characters, personalities, and psychologies. I propose that we hold constant in eligible alternative scenarios the historical factors that contribute to the formation of the agent's character and psychological constitution, and those that give rise to the conative elements on which the agent acts in the actual scenario.

27 I am grateful to Alfred Mele for his comments and suggestions.