The Consequence Argument and 'Moral Dilemmas'

I am an incompatibilist. That is, I believe that free will and determinism are logically incompatible. I believe this for various reasons, one of which is the infamous Consequence Argument. Peter van Inwagen is a famous defender of the argument, and here is his articulation of it:

If determinism is true, then our acts are the consequences of the laws of nature and events in the remote past. But it is not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of those things (including our present acts) are not up to us.[1]

I find this argument rationally compelling. I think it is a very good argument; it is valid and has overwhelmingly plausible premises. From what I can tell, plenty of people agree with me. I have never seen anyone reject the Consequence Argument without considerable philosophical reflection and maneuvering. It is a hard argument to defeat.

I am not going to argue for the Consequence Argument. Instead, I am going to use its powerful logical structure to argue for something else, something that those who favor the Consequence Argument should (given the parallel logical structure) favor. In brief, I will argue that the logical structure of the Consequence Argument can be used to shed light on an all-too-common problem in contemporary ethics: "moral dilemmas." The type of dilemmas I have in mind are those in which the agent in question is faced with two options, where these two options exhaust the minimal logical possibility: (i) perform some action A or (ii) fail to perform some action A. Of course, it's logically possible to "tack" things on to (i) and/or (ii). For example, you might go with (i), performing some action A, but also scratching your head while you perform that action. But whatever else you do, you must do either (i) or (ii) at any given time.

But this is not the whole story. The type of moral dilemma I envision is one in which you must do either (i) or (ii) at a given time, where both options are evil. That is, performing some action and refraining from performing that action are, at a given time, objectively wrong.

Here is the Consequence Argument in standard form:

(P1) It is not up to us that p.

(P2) It is not up to us that if p, then q.

(C1) Therefore, it is not up to us that q.

Here is my argument in standard form, which mimics the form of the Consequence Argument. Let "C" stand for some circumstance (i.e., state of affairs) in which either A or not-A must be performed, and let "E" stand for "some evil is done." Thus, to say that "if C, then E" is to say "if some specified circumstance obtains, then some evil is done." Call my argument the "Argument from Blamelessness."

(P3) It is not our fault that C.

(P4) It is not our fault that if C, then E.

(C2) Therefore, it is not our fault that E.

P3 is surely true of many moral dilemmas. It is sometimes not my fault that I am in a situation where all I can do is something evil.[2] Of course, if it is my fault that I am in a position where I cannot help but do something evil, then I am morally responsible and blameworthy for the evil I do as a consequence of being in that position. Although P3 is not true of every moral dilemma, it is nevertheless possible and does happen. And, of course, I am interested in the occasions where P3 is true.

P4 is perhaps even more plausible. For it is surely not my fault that certain moral facts, like "good" or "evil," supervene on certain actions.[3] In other words, I am not responsible for what is right and what is wrong; I do not determine those facts.[4] Because of the nature of C, the only possible actions for me are A and not-A. But both A and not-A are evil, and thus it is impossible that I can do anything good. Put more clearly (according to the wording of P4), it is not my fault that: if C, then E.

C2 follows logically from P3 and P4. Unlike the Consequence Argument, which operates on metaphysical facts such as "it is not up to us that" or "it is not within our power to," my argument uses the somewhat less controversial "it is not our fault that" or "we are not blameworthy for the fact that." But the logic remains the same despite the fact that I have substituted moral facts for facts about the powers of human agency.

This argument has implications for Christian theology. My argument entails that in cases where only A and not-A are possible and both are evil, insofar as one is not responsible for the situation that limits possible action to A and not-A, one is not blameworthy for performing A or for performing not-A. This seems to support a form of Christian inclusivism, since: If it is not your fault that you are in a particular circumstance (say, a circumstance wherein you do not know about Christ), and it is not your fault that the circumstance entails that you can do only evil (you cannot help but fail to exercise saving faith in Christ), then it is not your fault that you do evil (i.e., fail to exercise saving faith in Christ).

The argument also has implications for the ongoing debate over maximizing consequentialism in ethics. According to standard maximizing consequentialism, it is obligatory to perform maximizing acts. In the past, I have offered the following objection to maximizing consequentialism: There are at least some cases in which a maximizing act is intrinsically impermissible, and therefore it should not be performed.

Suppose that, if ten people were treated like slaves, the world would be exceedingly better off than if those ten people were not treated like slaves. Maximizing consequentialism seems to have the implication that we should treat those ten people like slaves. I find, and have always found, this implication unsettling. For it seems to me that it is never permissible to treat anyone in this matter. Of course, someone might subject ten persons to slavery and not know it is wrong, or even not know that they were subjecting ten persons to slavery. These persons, it seems to me, are not blameworthy for their actions on the grounds that they were unaware of the moral implications of their actions. But surely, I have always thought, persons who do know what they are doing could not avoid blame. But if the argument above is correct, then I am wrong: Someone could avoid blame even if they knew that the maximizing action they were performing was evil. It appears, therefore, that my strong objection to maximizing consequentialism fails. This is something of a philosophical "bummer" to me, but I (nevertheless) think it is true.

Then again, maximizing consequentialism "suffer" the same fate. For maximizing consequentialism holds that all maximizing actions are morally obligatory. Thus, when deliberating between two acts, A and B (where A and B are incompatible actions for me at the time), if A is a maximizing act and B isn't, maximizing consequentialism holds that it is obligatory for me to perform A and therefore (at that time) impermissible for me to perform B. Indeed, it is never permissible for me to perform B over A. But as my argument shows, even if it is impermissible for me to perform B, I could be (given the conditions I have described) freely perform B and yet avoid blame. Of course, maximizing consequentialists might not care much about this. They might simply claim that B is impermissible but that the agent who freely performs B isn't blameworthy for doing so, at least in the circumstances I have described. I am doubtful that maximizing consequentialists can make this exception, but I leave it to them to do the work there.

My argument also has implications for dilemmas in abortion. Assuming it is evil to kill the child (or allow it to die) and assuming it is evil to kill the mother (or allow her to die), one could do either and remain blameless. If it is not your fault that these are the only actions open to you (i.e., killing the fetus or killing the mother), and if it is not your fault that both actions entail that some evil is done, then it is not your fault that some evil is done. Thus, either choice is one for which you are not blameworthy.

I suspect this argument has implications far beyond the ones I have mentioned here. For example, suppose that the following two claims are true: (1) If someone is not blameworthy for performing some (allegedly) evil action, even though that person recognizes the action as (allegedly) evil, then that action is not evil in such a circumstance; (2) for any (alleged) moral obligation, it is possible for an agent to be in some circumstance such that the agent would be blameless even if she freely (and knowingly) acted contrary to that moral obligation. This would entail either that there are no universal moral obligations; or that, if there are, they are specified in such a way so as to avoid the problem that my argument, when combined with (1) and (2), creates.[5]

Is there any reason to accept either (1) or (2)? One might think that (1) is a conceptual truth, since "S freely and knowingly performs some act A she knows to be evil but is not blameworthy" seems incoherent.[6] And (2) seems even more defensible, given my argument. But this is far from decisive. I am unconvinced that (1) is a conceptual truth, for example, and I suspect something in my argument works against (1). In any case, I think my argument is worth pondering at greater length and depth.

________________

[1] Van Inwagen, Peter. An Essay on Free Will (Oxford: Clarendon Press), 1983, p.16.
[2] Because of this, some will perhaps prefer a conditional version of the argument. Here it is: If it is not our fault that C and if it is not our fault that if C, then E, then it is not our fault that E.
[3] I am assuming that the moral status of acts is intrinsic to those acts, and thus not determined by the virtues (of lack thereof) in the agents performing those actions.
[4] Counterexample: I promise someone that I will love her forever, and thereby make it true that "Blake keeps his commitment to someone" is morally obligatory. But this is obligatory only because keeping promises, in general, is obligatory. Of course, if I did freely determine which acts were good or evil, then I would be responsible for freely selecting those acts, and thus open for praise or blame.
[5] Perhaps any ethical principle that begins with, "All other things being equal" would be sufficient for this. In fact, I am slightly (and increasingly) inclined to adopt a more Rossian view of ethics in light of this, where only prima facie duties exist.
[6] Or at least false. It's one thing if someone does something evil but (non-culpably) fails to realize it. Ordinarily, we wouldn't blame that person. But when someone freely and knowingly does something evil, it seems that the person is genuinely blameworthy. My suspicion is that the Argument from Blamelessness, together with (1) and (2) above (in the body of my essay), constitutes a plausible argument against the existence of genuine moral dilemmas. Here is how that larger argument might go:

(P1) I am blameless for the fact that C.

(P2) I am blameless for the fact that if C, then E.

(C1) Therefore, I am blameless for the fact that E.

(P3) If I am blameless for the fact that E, then my actions in the context of E (even if I realize that E) are non-blame-conferring.

(P4) If my actions in the context of E (even if I realize that E) are non-blame-conferring, then it is not the case that E.

(C2) Therefore, it is not the case that E.

Although technically it cannot be a fact both that E and that not-E, the argument can be altered in certain ways so as to express epistemic "facts" (or reasons, or probabilities) rather than metaphysical, moral facts. That way, there is no logical inconsistency. Instead, the fact that someone can freely act a certain way, knowing all the relevant moral facts, and still be blameless, gives strong reason to believe that the way the agent freely acted was not evil/wrong/impermissible. Indeed, this conclusion is consistent with an ethic that includes either a single or multiple necessarily consistent principles. This begs for further thought, which I will give it at a later time.