MANIPULATION AND INDEPENDENCE 1

Transcription

1 MANIPULATION AND INDEPENDENCE 1 D. JUSTIN COATES UNIVERSITY OF CHICAGO DRAFT AUGUST 3, Recently, many incompatibilists have argued that moral responsibility is incompatible with causal determinism because causal determination by blind physical processes is relevantly similar to causal determination via neurological manipulation. 2 And because no one, the thought goes, can be responsible for their actions if they are subjected to neurological manipulation, neither can anyone be morally responsible for their actions if they are causally determined by blind physical processes. Developing this idea, Derk Pereboom argues that in cases of neurological manipulation, the fact that an agent is not morally responsible for her action is best explained by the fact that she is not the causal-historical source of her action. 3 But Pereboom goes on to suggest that causal determination by blind physical processes similarly undermines an agent s standing as the causal-historical source of her action. After all, if causal determinism obtains, then any action that the agent performs is logically entailed by the state of the world in the distant past together with the laws of nature. So, if determinism obtains, what an agent does ultimately depends on the state of the world in the distant past together with the laws of nature. Accordingly, Pereboom and other incompatibilists conclude that moral responsibility is not compatible with causal determinism of any sort, whether it is the product of covert neurological manipulation or the product of blind physical processes. Thus, compatibilism the thesis that causal determinism and moral responsibility are compatible is putatively false. Notice that Pereboom s argument rests on his claim that the features that undermine responsibility in cases of neurological manipulation are also present (and relevant) to attributions of moral responsibility in cases of mere causal determination by blind physical processes. This, I will suggest, is no 1 For helpful comments on these ideas and earlier drafts of this paper I d like to thank John Martin Fischer, Philip Swenson, Patrick Todd, Neal Tognazzini, and Robyn Repko Waller. 2 See Derk Pereboom, Living Without Free Will, 2001; Al Mele, Free Will and Luck, 2006; Patrick Todd, Defending (a Modified Version of) the Zygote Argument, forthcoming. 3 Pereboom, Of course, what I say here in response to Pereboom will apply mutatis mutandis to Al Mele s Zygote Argument. Indeed, as far as I can tell, the approach I develop in this paper constitutes a reply to any extant version of the so called manipulation argument.

2 small assumption. Indeed, I will argue that there are important differences between cases of neurological manipulation and causal determination by physical processes. And these differences make a difference with respect to moral responsibility as well as to attributions of praise and blame. Thus in light of these differences, manipulation arguments of the sort offered by incompatibilists like Pereboom fail to undermine the truth of compatibilism. But first, we must fill in the relevant details of Pereboom s manipulation argument the so called Four Case argument against compatibilism So to begin, suppose that there is a team of neuroscientists who decide to construct a human agent. In every respect possible, they engineer this agent to be identical to a human who has developed through the normal means of conception, upbringing, etc. Like any other parents, the neuroscientists want their child, who they call Plum, to grow up with their values. Unfortunately for Plum s moral character, the neuroscientists value murder and mayhem. And doubly unfortunate for Plum, they have more effective methods for shaping Plum s values and actions than do normal parents. So Plum is neurologically manipulated to reason egoistically in situations in which violence is an effective means to achieving his ends. And together with the other character traits that the neuroscientists have induced in Plum, his egoistic deliberations guarantee that he will seek to achieve those ends violently. Of course, his parents take care not to turn Plum into some mindless automaton. Specifically, they construct him in such a way that he possesses the sort of agential capacities that compatibilists claim to be sufficient for moral responsibility. Thus, when he acts, his actions issue from his own moderately reasons-responsive mechanism, 5 there is a harmony between his higher-order volitions and his effective first-order desires, 6 he possesses the capacity to grasp and act on specifically moral reasons, 7 etc. 4 For expository purposes, my discussion of Pereboom s argument is brief. However, it adequately characterizes the argument against compatibilism. 5 John Martin Fischer and Mark Ravizza, Responsibility and Control: A Theory of Moral Responsibility, Fischer and Ravizza argue that acting from one s own moderately reasons-responsive mechanism is necessary and sufficient for the freedom-relevant or control condition on moral responsibility (Fischer and Ravizza also believe there to be an epistemic condition on moral responsibility). 6 Harry Frankfurt, Freedom of the Will and the Concept of a Person, Frankfurt argues that a mesh between an agent s effective first order desire (i.e., her will ) and her second-order volition (i.e., her second-order desire about which first-order desire she wants to be her will), is sufficient for moral responsibility. 7 R. Jay Wallace, Responsibility and the Moral Sentiments, Wallace ties moral responsibility to a capacity for recognizing, applying, and regulating our behavior in light of specifically moral reasons for action. 2

3 Therefore, as much as it is possible consistent with the fact that he was created in a laboratory by a team of neuroscientists Plum grows into an apparently normally functioning adult agent. With this description of Plum in the background, Pereboom constructs the following case Case 2. 8 Case 2. Plum is like an ordinary human being, except that he was created by neuroscientists, who, although they cannot control him directly, have programmed him to weigh reasons for action so that he is often but not exclusively rationally egoistic, with the result that in the circumstances in which he now finds himself, he is causally determined to undertake the moderately reasons-responsive process and to possess the set of first- and second-order desires that results in his killing Ms. White. He has the general ability to regulate his behavior by moral reasons, but in these circumstances, the egoist reasons are very powerful, and accordingly he is causally determined to kill for these reasons. Nevertheless, he does not act because of an irresistible desire. 9 Regarding Case 2, Pereboom claims that despite the fact that Plum possesses the agential capacities that compatibilists take to be sufficient for moral responsibility, Plum is not morally responsible. Some compatibilists have argued that this is false. 10 These compatibilists claim that neither causal determination (say, by blind physical processes), nor neurological manipulation of the sort described undermine moral responsibility. After all, these hard compatibilists claim, just as our own parents perhaps deterministic role in our upbringing doesn t threaten our moral responsibility, neither does Plum s parents deterministic role in his upbringing threaten Plum s moral responsibility. Despite the potential appeal of such a solution, I want to say, in agreement with incompatibilists like Pereboom, that there is something 8 Though it s the first case that I consider, this is the second case Pereboom offers as part of his Four-case Argument. However, for my purposes, it is the most pertinent. But just in case you re interested, Case 1 is exactly like Case 2 in every way except that in Case 1, the neuroscientists exercise direct control, moment by moment, over Plum s activities. I omit this case because it clearly seems to be different than mere causal determination by blind physical processes, since the moment by moment manipulation doesn t allow for the kind of continuity that seems essential for agency a kind of continuity that is not ruled out by causal determinism as such. 9 Pereboom, 2001: See e.g., Gary Watson, Soft Libertarianism and Hard Compatibilism, 1999; Michael McKenna, A Hard-line Reply to Pereboom s Four-case Argument, 2007; John Martin Fischer, The Zygote Argument Remixed, 2011, etc. 3

4 deeply worrisome about the idea that we might be neurologically manipulated via the same sorts of manipulation that Plum undergoes and nevertheless morally responsible for our actions. 11 That is, it is natural to think that this sort of manipulation does undermine (or at least mitigate) our moral responsibility i.e., it s intuitive to think that if we were neurologically manipulated in the same way that the team of neuroscientists manipulates Plum, we would not be morally responsible for our actions (or we would be less morally responsible for our actions). 12 However, I am less sanguine than Pereboom that this judgment vitiates the truth of compatibilism. Indeed, I am inclined to think that there is some important difference between the neurological manipulation that Plum experiences and causal determination by blind physical processes. And this difference, I will suggest, makes a difference for attributions of praise and blame. Thus, this difference is relevant to moral responsibility. Further, given this difference, I will suggest that Pereboom s analogy between neurological manipulation and mere causal determinism is too loose to warrant incompatibilist conclusions. Mere causal determination by blind physical forces is sufficiently different from covert manipulation, and as a result, compatibilism remains a viable option. More perspicuously, in what follows I will argue that the causal histories of actions performed by manipulated agents have different modal properties than the causal histories of action performed by agents who are merely causally determined. And as we will see below, in some ordinary cases of praise and blame, a related kind of difference in the modal properties of an action s etiology can affect the degree to which the agent is praiseworthy or blameworthy for that action. Thus, these modal properties are relevant for the degree to which an agent is morally responsible for her action(s). As a result, the mere fact that manipulated agents are not responsible for their actions does not license the conclusion that causal determinism similarly undermines moral responsibility. 3. To get to this conclusion however, we need to momentarily step away from issues relating to compatibilism, manipulation, and determinism. So first, I want you to consider the following two cases (A1 and A2), specifically with 11 It s worth pointing out however, that some hard compatibilists notably John Martin Fischer (personal correspondence) do appreciate the worrisome nature of neurological manipulation of the sort described in Case 2. Fischer thinks that manipulation of this sort undermines, or perhaps mitigates the degree to which an agent is blameworthy for her actions even though it does not undermine her moral responsibility for such an action. 12 For a recent account of degrees of responsibility that might be able to explain why manipulation would mitigate our responsibility, see D. Justin Coates and Philip Swenson, Reasons-Responsiveness and Degrees of Responsibility, forthcoming. 4

5 an eye towards the question, does the degree to which these agents are praiseworthy differ? A1. Win is a considerate partner, and sometimes just because he knows it makes her happy on his way home from work he buys his partner Pearl a bouquet of tulips. Today, as he approaches the florist on his way home, it pops into his head that he should buy Pearl flowers. Yet, just as he turns to go into the shop, Pearl calls Win to tell him that she had a rough day, and that if he brought home a bouquet of tulips, the otherwise terrible day would be, in some small way, salvaged. Win doesn t mention to Pearl that he was already planning on buying her tulips, and when he brings the bouquet home, she is very grateful. Before continuing, fix in your mind how praiseworthy (if at all) you take Win to be in A1. Once you have that worked out, I invite you to consider A2. A2. Win is a considerate partner, and sometimes just because he knows it makes her happy on his way home from work, he buys his partner Pearl a bouquet of tulips. Today, as he approaches the florist on his way home, Win is so tired from work that that it doesn t even occur to him, as it sometimes does, to stop in at the florist and buy Pearl some tulips. Yet, just as he passes the shop, Pearl calls Win to tell him that she had a rough day, and that if he brought home a bouquet of tulips, the otherwise terrible day would be, in some small way, salvaged. Win forgets about his long day, and starts thinking about how he can make Pearl s day better, and when he brings the bouquet home, she is very grateful. Again, before you go on, make a judgment of how praiseworthy (if at all) you take Win to be in A2. Once you have made such a judgment, let s compare these cases more carefully. To begin, note that in each of these cases, Win brings Pearl a bouquet of tulips. And plausibly, in neither of these cases is he obligated to do so at least not morally. As a result, Pearl s gratitude in A1 and A2 seems entirely appropriate since intuitively, in each case Win is praiseworthy (to some degree or other) for his actions. Moreover, if gratitude is (i) responsive to and (ii) takes as its object the concern others have for us manifested in their agency, it seems that Win genuinely deserves Pearl s gratitude in such cases. But despite the fact that Win is praiseworthy (to some degree or other) in both scenarios, I am inclined to think that Win A1 is more praiseworthy 5

6 than Win A2. 13 After all, in A1 his decision to bring Pearl flowers is independent of her request. Though spontaneous, his thought to buy her flowers reveals something of his devotion to her something especially admirable. Of course, this isn t to say that Win A2 isn t also admirably devoted to Pearl. But it does suggest to me at least that Win A1 is more praiseworthy for this particular action because of how it expressed that admirable devotion. But what exactly does it mean to say that in one case his decision is more independent of her request? And why exactly would that be relevant to the degree to which he is praiseworthy in A1 and A2? Before I turn to these questions, I want to first consider a parallel set of cases. 4. Now let s consider two more cases (B1 and B2), again with an eye towards the question, does the degree to which these agents are blameworthy differ? B1. Win is a neighborhood tough. He regularly breaks into cars and steals any valuable electronics. Today, as he s walking along the boulevard, he notices an expensive car. The owner is not around, and he thinks as he sometimes does when he sees expensive cars that he should break into it. Just after he makes this decision, his buddy Alex approaches from the other direction, also eyeing the car. Before Win can say anything, Alex suggests that they break into the car and steal any valuable electronics. Win doesn t mention to Alex that he was already planning on doing so, and together, Win and Alex break into the expensive car. Again, make a judgment about the degree to which Win is blameworthy (if at all) in B1 before continuing on to B2. Once you have that in your mind, consider the following: B2. Win is a neighborhood tough. He regularly breaks into cars and steals any valuable electronics. Today, as he s walking along the boulevard, he notices an expensive car. The owner is not around, and he thinks as he sometimes does when he sees expensive cars that he should stop to take a look at it and admire it (Win appreciates good cars). Just after he makes the decision to stop and admire it, his buddy Alex approaches from the other direction, also eyeing the car. 13 A note on nomenclature: when I refer to Win x I simply mean to refer to Win, with the subscript referring which scenario is being picked out. Thus, I mean for Win A1 to be transworld identical to Win A2. 6

7 However, in this case Alex has more malevolent intentions, and he suggests that they break into the car and steal any valuable electronics. Although Win hadn t been considering doing so, he sees Alex s suggestion as attractive, and together, Win and Alex break in the expensive car. Now, judge the degree to which Win is blameworthy (if at all) in this scenario. And once you have settled on that, compare your judgment in B1 to your response in B2. If you re like me, there will be some difference. Of course, in each of these cases, it s true that Win breaks into a car. And in each of these cases, I am inclined to think that he is blameworthy for doing so. Moreover, it seems clear that he is sufficiently blameworthy in each of these cases to warrant some severe responses. However, it also seems to me that Win B1 is more blameworthy than Win B2. Perhaps a natural explanation of this intuition is that moral luck plays a bigger role in Win B2 s committing the crime. After all, had Alex not been waltzing down the same street (at the exact moment that Win passed the car), Win B2 would not have broken into the car. But this is not the case for Win B1. After all, even if Alex had not been present, Win B1 would have broken into the car. And again, this seems to point to the idea that in B1, Win s decision to break into the car is independent of Alex s presence in a way that it is not independent of Alex s presence in B2. With this in mind, I want to say more about the relevant notion of independence. 5. Now, as I ve said, in A1 and A2, Win is clearly praiseworthy for his considerate deed. And in B1 and B2, Win is clearly blameworthy for his malicious deed. But in A1 and B1, Win seems to be more praiseworthy and blameworthy (respectively) for his actions than he is in A2 and B2. After all, in A2 and B2, had Pearl and Alex not intervened, Win would not have bought the flowers or broken into the car. This suggests that the degree to which an agent is praiseworthy or blameworthy for her action depends (in part) on what they would have done had their action been independent from the other agents interventions. And on the natural assumption that the degree to which an agent is praiseworthy or blameworthy is a function of the degree to which she is morally responsible, this suggests that the degree to which an agent is morally responsible for her action depends (in part) on what they would have done had their action been independent from other agents interventions. In other words, when (e.g.,) Win B1 decided to break into the car, his decision depended on fewer extrinsic features of the scenario than was the case when Win B2 decided to break into the car. And because it 7

8 depended on fewer extrinsic features, we might think that it more accurately reflected Win s character. Accordingly, Win B1 is more responsible and more blameworthy (to some degree or other) for breaking into the car than is Win B2. To put the point a bit differently, in A1 and B1, Win would have performed the praiseworthy and blameworthy actions irrespective of Pearl or Alex s presence. But in A2 and B2, Win would not have performed the praiseworthy and blameworthy actions independently of Pearl and Alex s intervention. And although this isn t a full analysis of the sorts of conditions that underwrite an agent s moral responsibility, I think this does provide (the beginnings of) a suggestive account of why, in A1 and B1, Win is more praiseworthy and blameworthy than he is in the alternative scenarios. After all, this kind of counterfactual independence plausibly reflects that the act in question is, to some greater degree, the agent s own. And the degree to which an action is an agent s own certainly seems relevant to the degree to which the agent is morally responsible for her actions. Now, if this independence between an agent and the set of facts that actually contribute to the actual production of her action does in fact reveal something about our attributions of praise and blame, then there is good reason to suspect that neurologically manipulated agents like Plum are less blameworthy for their actions than those agents who are merely causally determined by blind physical processes. After all, independently of the specific manipulative inputs given to him by the team of neuroscientists, there is no reason to suspect that Plum would have murdered Ms. White. That is, in the nearest world in which Plum is not neurologically manipulated to kill Ms. White, it s not plausible to think that he would have killed Ms. White. After all, however we construe Plum in this alternative scenario, it seems clear that his killing of Ms. White in the actual world is not independent in the sense of independence present in A1 and B1 of the wishes of the neuroscientists. That is, the etiology of his murdering of Ms. White would be tied to a specific causal history that accompanies the neurological manipulation a causal history that is not present in the absence of the particular sort of neurological manipulation that Plum actually experiences. We can express the counterfactual dependence of Plum s action on the manipulation as follows: (1) If Plum were not neurologically manipulated to murder Ms. White, it would not have been the case that Plum would have murdered Ms. White. 8

9 We have reason to accept (1) because of standard counterfactual semantics. 14 After all, the nearest world in which the antecedent of (1) obtains plausibly isn t a world in which the consequent obtains, either because Plum is not created, or because the neuroscientists manipulate him to act in some other way. In either case, the truth of (1) strongly suggests that Plum s decision to murder depends on neurological manipulation in some strong sense. 15 And as we ll see, the fact that Plum s act of murder supports the truth of this counterfactual will be relevant to the question of whether there are important differences between causal determination by covert neurological manipulation and causal determination by blind physical processes. 6. Moreover, notice that something like (1) is also true of Win in the situation described in cases A2 and B2. For example, consider: (1*) If Win A2 had not been called by Pearl, it would not have been the case that Win bought Pearl flowers. In other words, Win wouldn t have bought the flowers if Pearl had not asked for them. The same can be said of Win B2. (1**) If Win B2 had not seen Alex, it would not have been the case that Win broke into the expensive car. That is, in B2 Win wouldn t have broken into the car if Alex hadn t first suggested it. Recall that these counterfactual properties of the causal history of Win s action plausibly reflect a difference in the degree to which Win was praiseworthy and blameworthy. Similarly, we should expect that because in Case 2, Plum s action supports the truth of a relevantly similar counterfactual, the degree to which Plum is responsible and blameworthy for his action is also affected. In other words, just as the sort of dependence present in (1*) and (1**) reflects a modal property of the causal history of Win s action that is relevant to the degree to which he is responsible, the presence of the same sort of dependence in (1) suggests that in cases of 14 See David Lewis, Counterfactuals Note that I am not reductively analyzing the notoriously vexed notion of dependence in terms of counterfactual dependence here (such reduction is undoubtedly doomed to failure). However, I am suggesting that (at least in this kind of case) the counterfactual dependence of Plum s decision to murder White on the fact that he is neurologically manipulated gives us good reason to think that this decision depends on the neurological manipulation in some deeper sense. Thus on this picture, counterfactual dependence of the sort expressed by (1) is a manifestation of a deeper, perhaps unanalyzable kind of dependence. 9

10 neurological manipulation, the degree to which an agent is responsible is affected. However, despite the fact that neurological manipulation supports (1), we shouldn t think that something like (1) is true in the situation in which Plum is merely causally determined by blind physical processes to murder Ms. White. In might be that even if our actions are not independent from neurological manipulation that is present in their causal history, what we do might be independent from the presence of mere causal determinism in the causal history of our actions. I turn to this possibility below. 7. Given what we ve just seen, I want to argue that in the case of causal determination by blind physical processes, it is plausible to think that Plum s decision and subsequent murder of Ms. White is importantly independent from the fact that he was causally determined (by blind physical processes) to do so. To see this, consider Case 4. Case 4. Physicalist determinism is true, and Plum is an ordinary human being generated and raised under normal circumstances, who is often but not exclusively rationally egoistic (exactly as egoistic as in Case [2]). Plum s killing of White comes about as a result of his undertaking the moderately reasons-responsive processes of deliberation, he exhibits the specified organization of first- and second-order desires, and he does not act because of an irresistible desire. He has the general ability to grasp, apply, and regulate his behavior by moral reasons, but in these circumstances the egoistic reasons are very powerful, and together with background circumstances they deterministically result in his act of murder. 16 Notice that in this case the fact that causal determinism obtained in the actual causal sequence of Plum s action is independent (in the sense at stake) from his act of murder. This is, in this case, whether Plum is causally determined to murder Ms. White doesn t bear on the actual-sequence etiology of his decision to do so. If it did, then it would be true of Plum that had he not been causally determined to murder Ms. White, he would not have done so. But this is not the case. After all, in the closest world or set of worlds in which causal indeterminism obtains (i.e., the closest world in which he is not 16 Pereboom, 2001: 115. As I did with Case 1, I have omitted Case 3 since it does not directly bear on the point I wish to make here. For the curious, Case 3 is identical to Case 2 except that instead of the team of neuroscientists causally determining Plum to murder Ms. White, Plum is causally determined to murder White by his parents and environment in the course of an ordinary upbringing. 10

11 causally determined to murder Ms. White), it is plausible to think that the same prior events would still have occurred and indeterministically issued in Plum s murdering Ms. White, since he would still reason egoistically, it would still be the case that murdering White would further his aims, etc. Indeed, in this set of worlds, we would still have the causal chain that leads to Ms. White s murder. 17 In other words, the mere introduction of causal indeterminism to the actual-sequence etiology of Plum s decision doesn t, by itself, seem relevant to whether a type-identical causal sequence in fact obtains. 18 And although the evaluation of such counterfactuals is admittedly delicate and contentious, it is thereby plausible to think that: (2) If Plum were not causally determined (by blind physical processes) to murder Ms. White, it would have nevertheless been the case that Plum would have murdered Ms. White. As I see it, (2) simply codifies the intuition that even if the causal sequence leading up to his decision had been indeterministic, Plum would have chosen, for the same egoistic reasons that drove his choice in Case 4, to murder White. 19 But notice, something like (2) is also true of Win in the situation described in cases A1 and B1. Consider, for example (2*): (2*) If Win A1 had not been called by Pearl, it would have nevertheless been the case that Win would have bought Pearl flowers. And likewise, consider (2**): (2**) If Win B1 had not been provoked by Alex, it would have nevertheless been the case that Win would have broken into the 17 Though in such a case, this causal chain would be such that the probability of the causal antecedents producing the relevant outcome (i.e., Plum murdering White) would be less than 1.0 (as would be the case given the truth of causal determinism). Of course, in the nearest world to the world in which Plum is causally determined, the relevant probabilities would undoubtedly be high perhaps.999. And it is in light of this fact that we have good reason to accept (2), which I argue for below. 18 Of course, the introduction of causal indeterminism does affect whether the antecedent causes are sufficient for Plum s murdering White, and this might affect whether Plum has the ability to do otherwise (it certainly affects whether holding fixed the past and the laws Plum has the ability do otherwise). But this is not germane to the issue at hand. 19 Of course, the mere fact that on the assumption of causal indeterminism it s true that Plum would choose murder White doesn t undermine the further fact that some incompatibilists (though not Pereboom) take to be paramount for responsibility namely, that Plum could have chosen to do otherwise, even holding fixed the past and the laws of nature. 11

12 car. In A1 and B1, I claimed that Win was more praiseworthy and blameworthy (respectively) for his actions than he was in the alternative scenarios of A2 and B2. If this is true, then we should think that Plum is more blameworthy for murdering White assuming causal determination by blind physical processes (i.e. Case 4) than he is as described by Pereboom in Case 2 (i.e., the case of neurological manipulation). After all, the difference between (1) and (2) is the same as the difference between (1*) and (2*), and as we saw from the cases involving Win, that difference, while apparently only minor, is relevant to Win s moral responsibility because it is relevant to the degree to which it is appropriate for Pearl to be grateful for the flowers. Likewise, the difference between (1) and (2) is the same as the difference between (1**) and (2**). And as we saw, that difference is similarly relevant to Win s moral responsibility because it is relevant to the degree to which it is appropriate for us to resent or otherwise blame Win for breaking into the car. 8. Thus, there is an important difference between the neurological manipulation as described in Case 2 and mere causal determination by blind physical processes as described in Case 4. And further, as I argued earlier, this difference is present in at least some ordinary cases in which we alter judgments of praise and blame, gratitude and resentment. So, it is, as they say, a difference that makes a difference. And while the actions of causally determined agents are in some real sense independent of the causal determinism itself, the actions of the manipulated agents are not independent of the manipulation at least if the manipulation looks anything like the sort described by Pereboom and other incompatibilists. Of course, I haven t hereby shown that compatibilism is true, but I do think we are in a good position to defend compatibilism from this sort of manipulation argument. The cases involving Win help to show that there is a relevant difference (with respect to moral responsibility, attributions of praise and blame, the fittingness of gratitude and resentment, etc.) between neurological manipulation and causal determination by blind physical processes. And this is all that is required to undermine the incompatibilist argument. After all, because the causal histories of Plum s actions support importantly different counterfactuals in Case 2 and Case 4 counterfactuals that we know (from considering Win) to be relevant to moral responsibility, praiseworthiness, and blameworthiness it is possible that the best explanation for the fact that Plum is not responsible in Case 2 does not generalize to Case 4. Quite simply, Plum s murder of White in Case 2 12

13 depends on the manipulators in a way that it would not, were Plum merely causally determined by blind physical processes. This is because in the latter case, the causal history that produces Plum s action possesses importantly different modal properties. As a result, we have little reason to think that manipulation cases of the sort Pereboom develops successfully show causal determinism to wholly undermine moral responsibility and its attendant practices of praise and blame. 13

14 Works Cited Coates, D. Justin and Philip Swenson, Reasons-Responsiveness and Degrees of Responsibility, Philosophical Studies. forthcoming. Fischer, John Martin, The Zygote Argument Remixed, Analysis, Fischer, John Martin and Mark Ravizza. Responsibility and Control: A Theory of Moral Responsibility. Cambridge: Cambridge University Press Frankfurt, Harry. Freedom of the Will and the Concept of a Person. Journal of Philosophy Volume LXVIII Lewis, David. Counterfactuals. Malden: Wiley-Blackwell McKenna, Michael. A Hard-line Reply to Pereboom s Four-case Argument. Philosophy and Phenomenological Research Mele, Al. Free Will and Luck. New York: Oxford University Press Pereboom, Derk. Living Without Free Will. Cambridge: Cambridge University Press Todd, Patrick. Defending (a Modified Version of) the Zygote Argument, Philosophical Studies. forthcoming. Wallace, R. Jay. Responsibility and the Moral Sentiments. Cambridge: Harvard University Press Watson, Gary. Soft Libertarianism and Hard Compatibilism. Journal of Ethics