Nathan Nobis nobs@mail.rochester.edu http://mail.rochester.edu/~nobs/papers/det.pdf ABSTRACT: What would be so bad about not having libertarian free will? Peter van Inwagen argues that unattractive consequences follow from rejecting belief in libertarian free will, namely, first, that those who believe in determinism could not consistently deliberate about what to do and, second, that we could never justifiably believe that anyone is morally responsible for his or her actions. I argue that his first argument is unsound and that determinists can consistently deliberate, and I show that his second argument is invalid, given his own proposed ways to understand the notion of moral responsibility. Thus, van Inwagen has not shown that these unattractive consequences follow from rejecting libertarian free will and, thereby, compatibilism remains untouched by his arguments. Peter van Inwagen (and others) hold that determinism is incompatible with being free: either we re determined or we re free (or neither), but not both. 1 Assuming neither is an incorrect answer, which is it: are we determined or free? As an initial, indirect response to his own question, Which, if either [i.e. freedom or determinism], ought we to accept? van Inwagen asks a further question, What would it mean to reject free will? (19, emphasis mine). He answers that rejecting free will has some interesting consequences: first, that anyone who reject[s] free will could not consistently deliberate about future courses of action and, second, and much more importantly, that to deny the existence of free will commits one to denying the existence of moral responsibility (19). 2 van Inwagen s main premise is, roughly, that if determinism is true, then we are not free, and if we are not free then we do not have various important goods, such as the capacity to consistently deliberate and, especially, be moral responsible. van Inwagen is confident that we are sometimes morally responsible 3 and that this shows that determinism is false and that we are free in a sense that requires the falsity of determinism. But if his arguments for the importance and necessity of incompatibilist freedom for moral responsibility are not compelling (because dire consequences would not follow if we rejected these notions), this gives us reason to doubt that a strong case has been against compatibilism, at least on this front, and thus that, for all we know, compatiblist conceptions of deliberation, moral responsibility, and freedom are not the wretched subterfuges or quagmires of evasion that Kant and William James thought they were. 4 Rather, 1 See Peter van Inwagen, An Essay on Free Will, Oxford UP, 1983 (unless otherwise noted, all further references to work by van Inwagen are cited by a page number in parentheses in the main body of the text). Similar arguments are found in his The Incompatibility of Responsibility and Determinism which appeared in M. Brand and M. Bradie, eds., Action and Reponsibility, Bowling Green State University, 1980, 30-37. Also see his more recent Fischer on Moral Responsibility, The Philosophical Quarterly, 47, 1997, 373-381. As far as I can tell, this recent essay s arguments for the incompatibility of moral responsibility and determinism are not significantly different from those presented in An Essay on Free Will. Also see further references and discussion of van Inwagen s other writings in the footnotes below. 2 Similarly, Robert Kane, an incompatibilist- free-willist, asks what he calls The Significance Question : Why do we, or should we, want to possess a free will that is incompatible with determinism? Is it a kind of freedom worth wanting.. and, if so, why? (13) His reply is that.. free will is significant and worth wanting because it is a pre-requisite for other goods that humans highly value (15). The suggestion is that we have a stake in being very critical of any arguments against free will because if we conclude that these arguments are sound, and we believe that free will is necessary for other goods that we highly value, then we must conclude that either we do not have these other goods, or that we have them somehow illegitimately. See his The Significance of Free Will, Oxford UP, 1996. 3 van Inwagen asks the rhetorical question,.. surely we cannot doubt the reality of moral responsibility? (206). For doubts, see Saul Smilansky s van Inwagen on the Obviousness of Libertarian Moral Responsibility, Analysis 50, 1990, 29-33. 4 Quoted in Kane, op. cit., 12.
compatibilistic conceptions of freedom might be good enough to do the jobs that van Inwagen says can be done only by incompatiblistic conceptions of freedom. 1. On the Alleged Inconsistency and Incoherence of a Deliberating Determinist van Inwagen says that rejecting free will has interesting consequences. But what does he think free will is? An informal statement of his view is that a person has free will (or acts freely ) when very often, if not always, when he has to chose between two or more mutually incompatible courses of action.. each of these courses of action is such that he can, or is able to, or has it within his power to carry it out (8). We might characterize free will or of acting freely by stating a free will thesis (FWT): (FWT): Someone is free with respect to doing A if, and only, if she can do A and she can refrain from doing A. 5 van Inwagen says that he will first examine the possibility of rejecting the free will thesis,.. without attempting to answer the question What does the denial of free will logically entail? [emphasis mine] That is, [he] shall not ask what we should logically or rationally be committed to believing if we gave up a belief in free will, but, rather, what effects such a rejection would have on our lives whether or not we accepted all those propositions to which the rejection of free will committed us. [He] shall then go on to ask what rejecting the freewill thesis would logically commit us to (153). van Inwagen says that it can be shown that we all believe in free will (154) but his first question concerns how our lives would change if we rejected belief in free will, as he construes it. He claims that those who profess that we are not free are either inconsistent or do not mean what they say (154). 6 van Inwagen s argument is unsound; its unsoundness seems to depend on not listening carefully to what some people who reject free will actually say regarding these matters. Some determinists do mean and understand what they say and are not inconsistent when they reject the kind of free will that van Inwagen says we sometimes have. Consider someone, the determinist, who denies that we are free in van Inwagen s sense above. She says she is not free. She sincerely holds that: (DET): With respect to any action A, either she can do A or she can not do A, but it is not the case that she can do A and that she can refrain from doing A. 7 Our determinist, however, sometimes deliberates about what to do, i.e., decides or chooses between various incompatible courses of action. She considers, e.g., whether to go to the store or to stay home and, thereby, not go to the store. It seems to her that she can do either action: it seems to her that she can go to the store and it seems to her that she can stay home. These both seem to be possible courses of action for her. But she seems to lead herself into self-contradiction here because the fact that she deliberates and, ipso facto, consults her beliefs about the range of divergent possibilities before her, seems to show that she believes here that: (DEL): With respect to some particular action of going to the store, it is the case that she can go to the store and that she she can refrain from going to the store. 8 5 (FWT): ( A & ~A), [not (A & ~A)!] 6 Cf. van Inwagen s writing, But aren t there people who believe that no one has free will, including themselves? Well,. there are certainly people who say that they believe this, but I suspect that they are not describing their own beliefs correctly. But even if there are people who believe that no one has free will, it does not follow that these people do not believe in free will, for people do have contradictory beliefs. If may be that on one level the abstract and theoretical certain people believe that no one has free will, although on another level the concrete and everyday they believe that people have free will (197, Metaphysics, Westview Press, 1993). 7 (DET): ( A or ~A) & ~( A & ~A) 2
(DEL) implies the falsity of (DET): if the determinist believes (DEL), the cannot consistently believe (DET). 9 So, van Inwagen concludes that the determinist has inconsistent beliefs here or else either she does not mean what she says when she says she believes (DET) (i.e. she doesn t believe [DET]) or she does not mean what she says when she says that she believes (DEL) or (DET). But van Inwagen is too quick here. Following all philosophers who have thought about deliberation, he states that, (ALL) one cannot deliberate about whether to perform a certain act unless one believes it is possible for one to perform it (154, emphasis mine). 10 Since the determinist deliberates about doing A or not doing A, van Inwagen concludes that the determinist believes that it is possible for her to do A and is possible for her to not do A. But the deliberating determinist does not believe this, i.e. she does not believe (DEL), and if we pay careful attention to the words she uses to describe her deliberation this becomes clear (see the store example above). Her words show that the principle that she does accept is something like (DEL ) or (DEL ): or, (DEL ): With respect to some particular action of going to the store, it seems to her that she can go to the store and it seems to her that she can refrain from going to the store. (DEL ): With respect to some particular action of going to the store, for all she justifiably believes she can go to the store and for all she justifiably believes she can refrain from going to the store. So van Inwagen s and all philosopher s principle (ALL) is false: determinists can (and do) deliberate about whether to perform a certain act and not believe that it is possible for them to perform it, because their beliefs concern what seems possible to them, not what they believe is possible. Note carefully that to not believe that an act is possible is not to believe that the act is impossible: determinists do not deliberate over whether they should draw a round square or not, or ride a bike to Mars or not (however, if they did falsely believe these were [or even seemed] possible, they might somehow try to deliberate about them, as well). There is second argument to show that the one cannot deliberate about whether to perform a certain act unless one believes it is possible for one to perform it principle is false. Consider this dialogue. We ask a thoughtful determinist, What is possible, relative to the laws and past? She replies, Well, we know that the past was possible and, furthermore, it was necessary, given the past and the laws. So we can identify some concrete possibilities by of pointing at things that are and have been actual. But about the future, we really don t know what will turn out to be possible whatever the laws and the past or present necessitate or churn out will be what s possible. We can make some reasonable guesses as to what is possible in general, but we really can t give much in the way of details as to what, in particular, is possible relative to the laws and the past. So, back when I was deliberating about going to the store or not, to be honest, I did not believe that I knew both that I could go to the store and that I could refrain from going to the store. In fact, I really had no beliefs about what was possible, relative to the past and the laws, at all. If I was asked, 8 A = Going to the store : S believes ( A & ~A). 9 If S believes (DEL), i.e., ( A & ~A), then S can t consistently believe (DET): ( A or ~A) & ~( A & ~A). 10 Cf. when one deliberates, one s behavior manifests a belief that what one is deliberating about is possible (156, An Essay on Free Will). Below I show that this is ambiguous and false, most obviously because of the many meanings of possible. Cf. also van Inwagen s more recent claim that.. I cannot try to decide whether to do A or B unless I believe that doing A and doing B are both possible for me. And, therefore, I am convinced that I could not try to decide what to do unless I believed that sometimes more than one course of action was open to me (196, Metaphysics, Westview Press, 1993). Below I will argue that the openness that van Inwagen speaks of is ambiguous and that the sense or modality that some determinist accept does not generate inconsistency or incoherence. 3
Is your going to store consistent with the laws and the past? I would have replied, I have no idea. I have no beliefs about this because I have no idea what particular events or actions the laws and the past entail (and thus make possible); I m only trying decide whether to go to the store on the basis of how things seem to me, my hunches, and what doesn t seem obviously false to me. But if you see me going to the store, you ll get your answer. So, this shows that one can deliberate about whether to perform a particular act and completely lack beliefs about that particular act s modal status relative to the laws and past. Again, van Inwagen s and all philosopher s principle (ALL) is false. A determinist, especially one who accepts van Inwagen s framework and definitions, does not believe (DEL) because she does not believe that her doing A and her doing not A are both consistent with the laws of nature and the past. The laws and the past imply only one possible future, so either doing A or not doing A is possible, but each is not possible. 11 Thus the laws and the past do not allow for both A and not A to be the kind of possibilities that van Inwagen claims that we must believe exist in order to deliberate. The sense of possibility that the determinist speaks of here is not that of relative to the past and the laws, rather it is something internal, her beliefs or the way things seem to her, and this neither generates any inconsistency for the determinist nor does it show that she not mean, understand, and believe what she says: she does understand how it seems that she can go to the store and it seems that she can refrain, she means what she says, and she believes this as well. So van Inwagen s claim that to reject free will is to condemn oneself to a life of perpetual logical inconsistency (160) is false. 12 2. van Inwagen s argument for the incompatibility of moral responsibility and determinism van Inwagen claims that it is a philosophical commonplace that If we do not have free will, then there is no such thing as moral responsibility (161). Recall that if someone is free with respect to doing A, then she can do A and she can refrain from doing A. It intuitively seems that moral responsibility requires free will: if someone is morally responsible for doing bad deed A, then it is the case that she could have not done A; otherwise she had to do A, was somehow forced to do A, and thereby is not morally responsible because her act was unavoidable. Thus it seems that moral responsibility requires the ability to act freely, which van Inwagen argues is incompatible with determinism, which, if he is correct, generates the conclusion that moral responsibility requires the falsity of determinism and a kind of free will that is incompatible with determinism. 13 11 Put somewhat formally, it is that case that: [{(Laws & Past)! doing A} or {(Laws & Past)! not doing A}], but it is not the case that: [{(Laws & Past)! (doing A and doing not A)]. Doing A and not doing A are not both consistent with the (Laws & Past). 12 For more on this topic, see Tomis Kapitan s Deliberation and the Presumption of Open Alternatives, The Philosophical Quarterly 36 (1986): 230-51, and Philip Pettit s Determinism without Deliberation, Analysis, 1989, 49,42-44. 13 In a recent essay, van Inwagen writes, Why should anyone care whether we have free will (in this special sense)? The answer is this: We care about morality, or many of us do, and according to the classical tradition, there is an intimate connection between free will and morality.... But.. if there were no free will if no one were able to act otherwise then no state of affairs would be anyone s fault. No one would ever be morally accountable for anything... very bad states of affairs would.. never be their fault. For example, suppose a father has raped his nine-year old daughter and, as a result, she has suffered immediate physical pain and terror and has experience life-long psychological and emotional disorders. Unless the father had at least some measure of free will, the pain and terror and the rest are not his fault. He cannot be blamed for them. They are not something for which he can be held to account (221-222, emphasis mine. From When Is the Will Free? in Philosophical Perspectives, 3, Philosophy and Mind and Action Theory, 1989, reprinted in Timothy O Connor, ed., Agents, Causes and Events: Essays on Indeterminism and Free Will, Oxford UP, 1995, 219-238. This quotation is from the reprint). Below I argue that van Inwagen s arguments that, e.g., the father here cannot be blamed or held to account are unsound. Concerning what is required for something to be anyone s fault, van Inwagen does not explain his concept. I discuss van Inwagen s analogously undefined concept of desert below and suggest that that, based what little he says about what is required for desert, it is an open question whether indeterminism is required for it. I suggest the same conclusion applies concerning actions or events being anyone s fault. Until van Inwagen analyzes his concept, we cannot judge whether it is or is not compatible with determinism. 4
An informal reply can be given: first, van Inwagen s argument here has some intuitive appeal since the inability to do otherwise is often a legitimate excuse that absolves one of moral responsibility. But, second, cases from Locke 14 and Frankfurt, 15 cases where someone does something A for which she seems morally responsible for even though she lacks the ability to do otherwise because of the presence of a covert agent who will intervene and cause her to do A if she does not do A on her own (but yet does not because the person does A on her own), seem to give some good reasons to doubt that the ability to do otherwise is necessary for moral responsibility. van Inwagen discusses Frankfurt-style cases and principles (162-180) and concludes that they do not refute the incompatibilist s belief that being able to do otherwise is necessary for moral responsibility. I will not address the Frankfurt-style cases or van Inwagen s conclusions about them because it seems to me that there is a simpler and more direct strategy available to argue that, contrary to van Inwagen, moral responsibility does not require the kind of freedom that involves indeterminism, i.e., that moral responsibility is compatible with determinism. Before I present this strategy, I will briefly and informally state two more of van Inwagen s more sophisticated arguments for the incompatibility of moral responsibility and determinism: (1) If determinism is true, then our acts are the consequences of the laws of nature and events in the remote past. But we are not morally responsible for what went on before we were born, and neither are we morally responsible for the laws of nature. Therefore, we are not morally responsible for the consequences of these things, including our own present acts (parallel to argument on 16, 56). (2) No one is morally responsible for facts about the past before she was born. No one is morally responsible for the laws of nature. No one is morally responsible for the fact that the laws of nature and with the past before she was born entail present events and actions. Therefore, no one is morally responsible for present events and actions (185) It should be noted that arguments (1) and (2) are instances of argument forms that are valid under some, perhaps many, modal or sentential operators. A question is whether they are valid with an S is morally responsible for p operator, i.e. whether there is a plausible interpretation of morally responsible such that there is a set of possible worlds where S is not morally responsible for the past and the laws, but S is morally responsible for some present events or actions. If there is such an interpretation, then the arguments are invalid. One way to show that the argument form of (1) and (2) is invalid, van Inwagen suggests, is for a compatibilist to devise a counterexample.. that can be evaluated independently of the question whether moral responsibility and determinism are compatible (188). He challenges a compatibilist to develop such a counterexample. I accept the challenge. My strategy to defuse van Inwagen s arguments for incompatibilism begins with a question for him: You argue that moral responsibility is incompatible with determinism. We understand what determinism and incompatibility are, but what do you mean by moral responsibility? van Inwagen does not provide an analysis or definition of moral responsibility. He says little to explain his understanding of the concept. This is surprising, since the term moral responsibility is not one of ordinary conversation. Ordinary, non-metaphysical folks talk about people being responsible and being held responsible for their actions, but they never talk about people being morally responsible or being held morally responsible for their actions. Similarly, they ask, Who is responsible for this? (sometimes asking merely whose actions or intentions caused something to be so), or Who should be held responsible for this? Who, if anyone, should be punished (or praised) for this being so? But ordinary folk never preface their use of responsible with moral, as far as I have heard. 14 John Locke, Essay Concerning Human Understanding, 1690. 15 Alternate Possibilities and Moral Responsibility, Journal of Philosophy 66, 1969, 829-839. 5
So, moral responsibility is a technical term from philosophy. An intelligent layperson who heard a philosopher using the term could rightfully ask, What is this moral responsibility you are talking about? I ve never heard of the term. When describing the sense of free will he is interested in, van Inwagen said that he s interested in free will in just that sense of free will that is relevant to questions of moral responsibility (20). This was unhelpful then and does not help us now either. Unlike most of his other technical terms (which he rigorously defines), van Inwagen does not give even an informal general definition of moral responsibility. But he claims that he has listened to philosophers who deny the existence of moral responsibility [but] cannot take them seriously (207), because they are inconsistent when they, e.g., say that a certain act was a shoddy thing to do and say that its agent was not morally responsible when he performed it: those who are not morally responsible for what they do may perhaps deserve our pity; they certainly do not deserve our censure.... The philosopher who denies free will continually contradicts himself because his nonverbal behavior continually manifests a belief in free will. But, I would suggest, the philosopher who denies moral responsibility speaks words that contradict his theories, words like That was a shoddy thing to do. It is not only that his deeds belie his words (though of course that is true too), but that his words belie his words (207). In addition to his claims about the incompatibility of shoddy actions and not being morally responsible for them, van Inwagen claims that one cannot say of a person that he is not morally responsible for what he did and also exclaim What a perfectly despicable way for him to behave or You d think a person with her advantages would know better than that or I can never think of what [he] did without feeling sick (207-208). From this we can extract the beginnings of a something like a definition of (or, at least, a truth about) moral responsibility: (MR) If S s doing A was shoddy, despicable, corrupt, or sickening, then S is morally responsible for doing A. van Inwagen notes that these kinds of judgments are often right and they are extremely important for getting along in the world (208). Everyone, with the possible exception of, say, logical positivists, emotivists, and some moral anti-realists, agrees with van Inwagen that there are truths about these matters and that these truths are morally and socially important. Some would be willing to accept this kind of definition (MR): it s not complete, but maybe we couldn t do much better than it. But how does anyone s believing that (MR) is true commit her to the belief that we are free in an incompatiblist sense (and, thus, show that if she is a determinist, then she is inconsistent if she believes [MR])? The general argument here seems to be this: (1) If S s doing A is shoddy, despicable, corrupt, or sickening, then S is morally responsible for doing A. (2) If S is morally responsible for doing A, then S could have not done A. (3) If S could have not done A, then S was not determined to do A. (4) If S was not determined to do A, then determinism is false. (5) Therefore, if S is morally responsible for doing A, then determinism is false. 16 To explain these premises, we have assumed that, especially given the stipulated quasi-definition of moral responsibility, we all agree about premise (1). Although common sense suggests that premise (2) is true, Locke and Frankfurt s examples show that premise (2) might very well be false; I have said that I will not say anything against premise (2) along Lockean or Frankfurtian lines. Premise (3) follows from the fact 16 A longer version of this argument could be given that takes a detour through the (FWP), linking moral responsibility to the ability to do otherwise, doing otherwise with free will, and free will to indeterminism. 6
that if S was determined to do A, then S s doing anything other than A was incompatible with the past and the laws of nature, but since S could have done something other than A, S was not determined to do A. Premise (4) is obvious, and line (5) is the conclusion which follows validly from the premises. Locke and Frankfurt deny premise (2) for somewhat complicated and dubious reasons. I wish to deny premise (2) for a far simpler reason: suppose we believe that someone S has done something A that is shoddy, despicable, corrupt, or sickening. By our quasi-definition of moral responsibility (MR), S is morally responsible for doing A. But if this definition is correct we seem to be making claims either about the moral status of S s act or claims about our own psychological reactions to it. If we are making moral claims when we say S did something despicable and corrupt, and we are, say, consequentialists, then we are saying, roughly, that, In doing A, S has done something that brings badness into the world but no greater good that outweigh the bad, which is wrong If we are Kantians and talk about S s despicable actions, we are saying, roughly, that, S did something she could not will that everyone in like circumstances do, which is wrong, or that S has used someone as a mere means, which is wrong. If we hold a different moral theory we will give a different moral explanation. If our reaction is that S s action sickens us, we are making claims about our psychological or physiological reactions. We have assumed that we strongly believe that S s doing A is shoddy, despicable, corrupt, or sickening, (i.e. S is morally responsible for doing A), but the fact (granting van Inwagen s argument here for the incompatibility of determinism and doing otherwise) that S could have not done A is simply not relevant to the truth about our moral evaluations of S s act: it is shoddy, despicable and sickening that S did bad action A, even if S could not have refrained from doing A. This inability is especially irrelevant to our making true observations of our own feelings. So even if we realized that S could not have done anything but A, that S was unfortunately determined to do A, this would not change that fact that S did a bad thing and the reasonableness of our belief that this is so. So premise (2) is false, given the understanding of moral responsibility that van Inwagen offers, and our words here do not belie our words, contrary to van Inwagen s judgment. van inwagen develops a case of someone being drugged so that he is determined to something bad: Suppose we.. discovered that he did that thing shortly after he had been given, without his knowledge, or consent, a drug that is known to alter human behavior in radical and unpredictable ways. Suppose this discovery led us to decide that he had not been responsible for what he was doing at the times he performed the act. It seems to me that we could not then go on saying, That was a perfectly despicable thing for him to do, not even if we qualified that assertion by adding, though he wasn t responsible for his acts when he was doing it. that additional clause, in fact, does not seem to me to be a coherent qualification of the original assertion.. (208). Let s make this case more concrete: suppose we know that a loving, devoted mother accidentally takes drug that caused her to brutally and gleefully murder her child. All would agree that she has done something truly despicable. If she did not actually perform an action here, then certainly her behavior was despicable. She, or her behavior, is a cause (or is responsible for) her child s death. But she shouldn t be blamed or punished, if the causal story here is true (although the distraught mother will probably partly blame herself here, even if this blame is wholly unjustified and does no good). If anyone, the person who left out the drug should be blamed or punished here. But, van Inwagen s quasi-definition of moral responsibility implies that the mother here is morally responsible, even though, as stipulated, she could have not refrained from murdering her child. But, again, her inability to do otherwise is simply irrelevant to the fact that she did something (or, at least, something happened) that is despicable and sickening, and for which she is morally responsible for, according to the quasi-definition (MR). So premise (2), that if S is morally responsible for doing A, then S could have not done A, is, again, false, as is van Inwagen s claim that one can t be determined to do something despicable and not be morally responsible for it (again, as he construes moral responsibility ). van Inwagen s concludes about these kinds of cases that: 7
To call and act despicable is to censure its agent for performing it, while to say of an agent that he was not responsible for what he was doing when he performed an act is to excuse him for performing it; and one cannot simultaneously excuse and censure (208, emphasis mine). To censure is to express disapproval or condemnation. But one can justifiably censure some action or event that seems to one to have been inevitable or necessary (e.g., some political event or scientific achievement). And one can excuse someone for doing something that one censures (e.g. acquitting, pardoning or forgiving someone that it known to have done a thing that one disapproves of and condemns). So one can reasonably simultaneously both excuse and censure. In defense of premise (2), an incompatibilist might offer a new interpretation of it, linking moral responsibility with punishment in premise (1) 17, such that it reads: (2 ) If S s doing A is an act that should be punished, then S could have not done A. She might defend (2 ) by an argument like this: (3) No one should be punished for facts about the past before she was born. No one should be punished for the laws of nature. No one should be punished for the fact that the laws of nature and with the past before she was born entail present events and actions. Therefore, no one should be punished for present events and actions (parallel to argument on 185) This argument is unsound. At least sometimes, questions about whether some one should be punished are decided purely on moral grounds. 18 We lock-up serial killers because we can reasonably expect that, if we don t, they will kill again. That would be bad, and since we can prevent this, we do (and we should). Even if this punishment does not rehabilitate the serial killer, we still believe that we should keep him locked up so society is safe from him. And here the truth that the killer shouldn t be punished for the past and the laws of nature is irrelevant: even if the killer was wholly determined to kill, it just does not affect the justification for punishment here (although it might affect our sympathy for the killer 19 ). This example shows that, at least sometimes, contrary to van Inwagen a determinists deeds do not belie their words, and that argument (3) is unsound. An incompatibilist might reply with a new version of premise (2): 20 (2 ) If S s doing A an act that deserves to be punished or justice requires that S be punished, then S could have not done A. 21 17 Revised premise (1): (1 ) If S s doing A is shoddy, despicable, corrupt, or sickening, then S is morally responsible for doing A and S should be punished for doing A. 18 Galen Strawson notes that pessimists.. conclude that strong free will is not possible, and that ultimate responsibility is not possible either. So no punishment or reward is truly just or fair, when it comes to moral matters.. [and].. ultimate responsibility exists if and only if punishment and reward can be fair without having any pragmatic justification, Free Will entry in Routledge Encylopedia of Philosophy, p 749. One should ask what the difference is here between a moral and pragmatic justification. Also, Strawson s notions of what is truly just or fair are left unanalyzed here and, furthermore, that it s an open question whether these notions can be rendered compatible with determinism, if our ordinary of justice and fairness (as if we had an ordinary concepts here) are incompatibilist. 19 E.g., see the discussions of the case of murderer and child abuse victim Robert Alton Harris in Gary Watson s Responsibility and the Limits of Evil: Variations on a Strawsonian Theme, and Fischer and Ravizza s Introduction, both in Fischer and Ravizza s Perspectives on Moral Responsibility, Cornell, 1993. 20 Again, a revised premise (1) is needed : (1 ) If S s doing A is shoddy, despicable, corrupt, or sickening, then S is morally responsible for doing A and S deserves to be punished for doing A. 21 Note however that van Inwagen s above-discussed claim was that those who are not morally responsible for what they do may perhaps deserve our pity; they certainly do not deserve our censure (207, emphasis mine). He uses the term deserve, but does not explain what he means. So it is unclear if he has in mind a thick, probably incompabilist, conception of desert, or a more mundane (and compatiblist) one (e.g. That child deserves a toy for her good 8
(2 ) brings in new moral concepts of desert and justice. It is unclear whether these moral evaluations require indeterminism or not, and whether or not they can be compatibilized. In order to evaluate (2 ), the person who offers it has a lot of explaining to do. Thus, I have shown that premises (2) and (2 ) are false. I have shown that van Inwagen s own proposed senses of moral responsibility (definition [MR]) generate possible worlds where S is not morally responsible for the past and the laws, but is morally responsible for some present events or actions. Furthermore, the propriety of the application of these sense of moral responsibility can be evaluated independently of the question whether moral responsibility and determinism are compatible, one of van Inwagen s own proposed constraints for adequacy in criticizing his arguments (188). Premise (2) or (2 ) was required for van Inwagen s arguments that moral responsibility is incompatible with determinism and that moral responsibility requires incompatiblist freedom. Since (2) and (2 ) are false, his arguments are unsound. 3. Conclusion I have shown that van Inwagen has not shown that rejecting free will has the interesting consequences that he says it does. First I have shown someone who rejects free will can consistently deliberate about future courses of action and, second, and that to deny the existence of free will does not commit one to denying the existence of moral responsibility, especially if moral responsibility is explicated in the concrete terms of definition (MR). Thus, van Inwagen s arguments here against those who hold that moral responsibility is compatible with determinism and deliberation are unsound and, for all we know, some kind of compatibilism is correct. 22 behavior! ). But if he is referring to the thicker kind of desert, it would seem that if pity could, perhaps, be deserved, then censure could be deserved as well, perhaps. Similar consideration apply to his discussion of actions and events being anyone s fault in When Is the Will Free?, op. cit. For discussion of 'desert' and determinism, see Ralph Ellis's "Ethical Consequences of Recent Work on Incompatibilism,' Philosophical Inquiry, 8, 1991, 22-42. 22 For helpful comments and discussion on this paper, I am grateful to Richard Feldman, John Fischer, Tom Kapitan, Eddy Nahmias, and Saul Smilansky. 9