I’ve long been a believer in the principle that if you ought to do something, then it has to be the case that you are able to do that thing (that is, ought-implies-can), though I realize that this principle has problems with respect to issues of agency (assuming determinism). I’ve got a different sort of question about it, though, looking at it from the perspective of dirty hands cases. Let me get at my question with an example.
The evil empire known as the Los Angeles Lakers was dismissed from its championship run by this year’s blue-collar wonder team, the Detroit Pistons. A friend recently told me that he’d given up being a lifelong Laker fan because of way the L.A. media and the team have packaged the Kobe Bryant situation as “Kobe’s Ordeal,” rather than as a serious ordeal for his alleged victim, or even merely as Kobe himself has tried to package it, namely as a failure of fidelity. They’ve turned it into another obstacle for him to face (“From the court of law to the basketball court! Amazing!”), rather than as a moral failing on his part.
Now it seems to make sense for my friend to say “I think we Laker fans ought to rescind our allegiance.” That is, it seems natural to say we’ve got a legitimate ought-claim here. But it also seems that the ought-claim holds only if there’s a point to rescinding one’s allegiance. But what would be the point of my friend taking a stand here? Either it’s because without making his stand, my friend couldn’t live with himself; or it’s because he wants to put pressure on the organization and media to take the allegations against Kobe more seriously. If it’s the first, no problem. But what if it’s the second? (That is, what if he could live with himself? After all, he knows that the allegations against Kobe are serious; he just wants to put pressure on others to take the matter seriously.) Then we’ve got a problem. Knowing the irrationality of fandom, my friend realizes that his stand won’t effect any real social change. So this seems like one version of a dirty hands problem – if compliance with your apparent obligation is rendered pointless because of widespread noncompliance by others, then you might as well not comply yourself, it seems.
So, to get to my question, my impression is that discussions of dirty hands cases tend to focus on just that question: whether there is an obligation to do act A in these kinds of dirty hands cases (where the point of A requires widespread compliance with A, and where everyone else is, wrongly, doing ~A, in which case doing A yourself is pointless). I’m wondering, however, whether one couldn’t take the other side, and assume that there is, in fact, an obligation even in cases of widespread noncompliance, and instead deny that ought implies can. That is, my friend can’t do what (by hypothesis) he ought to do, namely make his stand to the end of producing social change, because the noncompliance of others makes it impossible for him to achieve his end. But if ought doesn’t imply can, then we could still say that he ought to make his stand. And this would be true of any principled stand where one must either play along with convention and get one’s hands dirty or take a principled stand to no good effect. I’m not too familiar with the literature here, so if anyone out there in the blogosphere is more familiar and can help me on this one, please chime in.
The “ought implies can” rule, it seems to me, is something like “If you ought to X, then you can do X.” A counterexample would come in the form of a case where someone ought to X but cannot do X. The first X and the second X have to be the same action. Your case does not seem to me to represent a counterexample to this rule. Your case is more like the case of a person who ought to X but cannot Y. Specifically, in your case, your friend ought to make his stand even though doing so does not have the effect it is designed to have. Call this action “X.” You say that your friend cannot make his stand in such a way that it DOES have its intended effect. But this is another action altogether: It is one action to “take a stand” without its having a certain effect (i.e. to “X”), and it is another action to “take a stand” WITH its having a certain effect. Call the latter action “Y.” Thus your friend ought to X, but he cannot do Y. This may be an interesting case for other purposes, but I do not think it represents a counterexample to the “ought implies can” rule.
Or am I way off?
Let’s formulate the principle that “ought implies can” as follows.
Ought Implies Can:
For any proposition P, if an agent A *ought* to see to it — bring it about, make it the case, etc. — that P, then A *can* see to it that P.
Now, consider the following propositions.
(P) Your friend withdraws his support for the Lakers.
(Q) The Lakers take the allegations against Kobe more seriously.
(R) Your friend withdraws his support for the Lakers with the aim of encouraging the Lakers to take the allegations against Kobe more seriously.
I don’t see that any of these presents a counterexample to Ought Implies Can. The first two are easy: your friend *can* see to it that P; and, although he can’t see to it that Q, it’s implausible that he *ought* to do so. The third may be a little more tricky. If your friend is rational, well-informed, and so on, then it may be impossible for him to see to it that R. But then I don’t really see why he ought to see to it that R.
Another point: I don’t see how the behaviour of other Lakers fans has any bearing on the possibility of your friend’s influencing the attitudes of the Lakers’ administration. No matter how many of the other fans withdraw their support, it will make no difference whether or not your friend also withdraws his support. The support of a single fan is just too insignificant for the Lakers to take any notice, irrespective of what all the other fans are doing.
Thanks to David and Campbell for the comments.
As to David’s comment, I actually had in mind your “Y,” rather than your “X.” So, my friend (it seems) has an obligation to Y, but can at best X. So, in these terms, the counter-example takes the form of: my friend ought to Y, but can’t Y.
This Y is roughly Campbell’s “R”, which (rather than P or Q) is what I’m concerned about. The reply here is that “If your friend is rational, well-informed, and so on, then it may be impossible for him to see to it that R. But then I don’t really see why he ought to see to it that R.” Two points seem pertinent here. First, *if* we suppose the hypothesis that my friend has an obligation to R, then that’s where the problem for ought-implies-can would enter in, and I’ll grant that the problem hinges on that supposition. Second, though, I think that (as I mentioned) it is pre-theoretically, intuitively plausible that there’s a (defeasible) obligation to R. So the question then becomes one of making sense of our pre-theoretical moral judgments, which might require abandoning ought-implies-can (or so I’m arguing). Of course, your intuitions and mine might not be seeing eye to eye on this one, so I might be cut off in the argument if the intuitive appeal of R isn’t so appealing after all. I would, however, argue that rejecting the intuitive appeal of R requires an *argument*, which can’t be had merely by stating that R is incompatible with ought-implies-can, since that, of course, is the whole question.
But I am open to discussion on this issue, that perhaps there is no obligation here. In this sense, your “other point” is actually very relevant. It raises a question that I’m not sure how to evaluate, namely whether we have an obligation to take a stand for Z, where Z requires widespread compliance, but where one individual’s compliance will make no difference to Z. Obviously, if no one complies, then Z won’t happen, so *some* sufficient number of people need to comply; but it’s unclear that that obligates any one person to comply.
I case like this was brought up in an Ethics class I took with Phillip Montague, whom, I believe has a paper on the following case.
Joe falls in a ravine. It takes three people to get Joe out. I can’t do it alone, but me and my two buddies can. I ought to do something, but can’t get Joe out of the ravine alone. Suppose that I am obligated to get Joe out, not to just try to get Joe out. This is a substantive thesis, but is plausible for other reasons. I take it that this kind of obligation is like (R) and ‘Y’ above.
So, it looks like I ought to do something that I can’t do. A counterexample to the ought implies can thesis.
Reply: This type of obligation is special. The obligation is one the group has, not me. The group is obligated to get Joe out of the ravine. The only sense in which I am obligated to get Joe out of the ravine is that I am a member of a group that has an obligation to get out of the ravine. But really, I am not obligated simpliciter to get Joe out of the ravine. So, no counterexample. (R) and ‘Y’ are like this. So, it is not the case that S is obligated in those cases either, at least, not obligated simpliciter.
There are semantic issues surrounding the notion of obligations had in ivrtue of being a member of a group that has an obligation versus having an obligation simpliciter that are troubling. But, insofar as there is some way to understand the distinction, there is no problem with this response that I can see yet.
First of all, my Y is not the same as Campbell’s R. Your friend is perfectly capable of taking his stand (i.e. doing Campbell’s P) with the AIM of bringing about Q. That is, he is perfectly capable of doing R. What he is not able to do is to actually ACHIEVE this aim (in other words, he is not capable of doing P in such a way that Q actually does result). That is, he is not capable of doing what I called “Y.” Thus, if you are saying that your friend ought to R, then you are not presenting a counterexample to “ought implies can.” Only if you are saying your friend ought to Y are you presenting such a counterexample.
Generally speaking: If I believe I ought to ___ even though I cannot ____, then this belief about what I ought to do cannot provide, on its own, any guidance about what to do in the “real world,” i.e. in the circumstances which actually obtain. For instance: Suppose I cannot swim. Suppose I ought to rescue the drowning man. Fine; but I CAN’T rescue the drowning man. My “ought” does not provide me any guidance, on its own, about what I REALLY ought to do. In the circumstances which actually obtain, I may as well stand around whistling, or throw rocks at the drowning man, or run off to find a lifeguard, or anything; my “ought” does not give me any immediate way to deliberate among the options which actually present themselves to me, because no matter what I do I will still NOT be doing what I ought to do.
In your example: If you are saying that your friend ought to take his stand with the result that Q (i.e. he ought to “Y”), then this “ought” does not provide your friend any practical guidance about what to do in the real world. He can agree with you, saying “I ought to Y,” and then – what should he do? Whether your friend decides to R, or decides to not-R, he is still not doing what he ought to do (i.e. Y). The prescription that your friend do Y does not enable your friend to deliberate among the competing courses of action which are available to him.
I do not know whether you are saying that your friend ought to Y or that he ought to R. You seem to think that your case provides your friend guidance about what he actually ought to do in the real world; this leads me to believe you are saying he ought to R. But if that is the case, then you are not presenting a counterexample to “ought implies can.”
Thanks again to David, and also to Christian, for helping me think through these issues. I’m still not sure, though, that there’s no problem here.
Let’s start with Christian’s solution, which is to say that my friend is only obligated qua a member of a group, rather than as an individual, and it’s the group, properly speaking, that is obligated. The reason I’m not confident in this solution is that it leaves my friend’s dilemma intact. That is, the individual member of the group still has to either fulfill his portion of the obligation fruitlessly (assuming the noncompliance of the rest of the group) or neglect his apparent obligation as a member of the relevant group. Perhaps there’s a way out with the group/individual duty distinction that I’m not seeing here, but it seems to me the dilemma still stands.
As to David’s reply, it is true that R and Y are not *exactly* the same. However, they are the same in the relevant way, namely that my friend can’t do either one. You grant that he can’t do R, but say that he is capable of doing Y, because he can *aim* at a result regardless of the compliance of others. This seems dubious however, under Campbell’s condition that my friend is rational, well-informed, etc. At least this seems true if we accept the premises that a rational, well-informed person could see the pointlessness of her endeavors and that a rational, well-informed person cannot, strictly speaking, aim at a goal which it is impossible to realize. (To borrow Kantian terminology, I can wish for everyone to be in perfect health in one hour, but I can’t will it [or aim for it] since I know this is impossible.)
The argument from practical deliberation is a different issue, though. The argument, if I am reading it right, which I may not be, seems to be:
(1) Counterexamples to ought-implies-can of this sort entail that those ought-claims provide no practical guidance.
(2) [Implicit?] A theory implying that ought-claims provide no practical guidance is problematic.
(3) [Implicit?] Thus, a theory like the proposed one, which includes (1), is problematic.
First let’s focus on (1). Even if my friend can’t achieve Y, he could still be guided towards getting as close to Y as possible, which would have him do P, rather than ~P. Second, let’s grant (1) for the sake of focusing on (2). My initial worry was that our intuitive judgments in cases like these might mean we have to abandon ought-implies-can. It might also turn out that we have to abandon the Assumption that every ought-claim provides determinate practical guidance as well. This is a common move made by those working on “unresolvable, tragic” dilemmas, where there appears to be no right thing to do (think, e.g., of Sophie’s choice). As I said in my reply to Campbell, we could of course doubt the intuitive plausibility of the ought-claim my friend made, but I don’t see the Assumption as being strong enough to overturn the ubiquity of my friend’s kind of ought-claim. For we often make those types of ought-claims when grand issues of social justice are at stake. To take another example, it seems intuitive to say that the ordinary German shouldn’t have condoned Nazi practices even though those practices would have continued regardless of whether or not she condoned them.
Correction: I should have said that David grants that my friend can’t do Y, rather than R (I got ’em reversed in the third sentence of the third paragrapn of my last comment).
Josh,
I can’t see how the dilemma remains. You wrote:
“The reason I’m not confident in this solution is that [1] it leaves my friend’s dilemma intact. That is, [2] the individual member of the group still has to either fulfill his portion of the obligation fruitlessly (assuming the noncompliance of the rest of the group) or [3] neglect his apparent obligation as a member of the relevant group.”
I suppose Joe “can” fulfill his portion of the obligation that the group has. A fan “can” recind their allegiance. Assuming the other group members don’t fulfill their portion of the obligation just means the group obligation isn’t satisfied and the compliant and complacient are deserving of blame. But, I don’t see how this makes Joe’s action fruitless or the fan’s action fruitless. It seems quite the opposite. There is a point to what they do, for if the others had done their part, the obligation would have been satisfied and the world better.
Two Points:
1. Is not the problem here that you are inadequately defining the scope of the “ought”? In a group situation, should not the ought encompass all intermediate steps, such as actually assembling the group? Thus, the ought should be described as “I ought to maximize the number of people who withdraw their support from the Lakers,” for which the individual’s own support is merely one instance. The moral obligation would extend to writing letters to the editor, calling sports-talk radio show, cajoling associates. Similarly, in the man-falls-down a ravine example, the moral “ought” is not to help the guy up (an impossible task), but to round up the posse.
2. The more direct counterexample to ought-implies-can is the fact that — to a certain extent — team allegiance is very emotion. “I ought to stop loving you, but I can’t,” says the woman in the movie. “I ought to stop routing for the Lakers, but I can’t turn it off like a switch. I stop buying tickets and watching the games, but every time I hear they lose a game on the news or the papers, I get sad inside.” When the “can” involves not merely a physical action but an emotional response, it is less obvious that we “can” do what we “ought.”
In thinking about Christian’s most recent comments, it seems that I should clarify what the obligation facing my friend is supposed to be. It’s not merely ‘stop being a fan.’ Instead, it’s ‘stop being a fan as a means of effecting social change.’ The problem is that this is something he *cannot* do (assuming the noncompliance of the other fans). Of course, there is a “point” to his action in the sense of a “purpose.” But his action, given the noncompliance of others in the group, is pointless in the sense that it can’t effect what it’s supposed to effect, namely social change.
Regarding Richard’s comment, I think you are right that those subordinate obligations would be implied by the primary obligation to effect social change. But the issue, as I see it, is focused on one of those subordinate obligations, that of ceasing one’s fan-hood to effect social change. And it is *this* that is pointless. And, again, when we look at it from our pre-theoretical judgments, this does seem like an “ought”, even though it is impossible. (As for the issue of ability to muster emotions, this seems to me to fall under the usual objections to ought-implies-can, having to do with agency and determinism, that I mentioned in the opening of the original post.)
Thanks, Josh, for your reply.
Let me try to put my objection in what I hope are clearer terms, now that I’ve had some time to think it over.
My argument is this:
If an obligation is capable of offering a person real-world, practical guidance about what she actually ought to do, then this obligation is not a counterexample to the ought-implies-can rule.
My reason for thinking this is as follows. Suppose a person has four alternative courses of action available to him: A, B, and C. Suppose the person ought to D. But the person cannot D. From the perspective of the person’s obligation to D, all of his possible courses of action are on the same “moral plane”: They are all not D. Thus from the perspective of his obligation, the person has no way to decide whether to A, B or C. Any one of these would equally be a violation of his obligation to D. Thus, if a person has an obligation to do something which he cannot do, then this obligation cannot offer him any practical guidance – any practical way to decide which of the various courses of action actually open to him are “right.”
Now, one might reply to this that, if the person in question cannot D, he ought to get “as close as possible” to doing D. If C is, for instance, closer to D than A or B, then given the obligation to D, coupled with the impossibility of doing D, our person ought to C. In that case our person’s obligation to D really does give him some practical guidance. There are many problems with this move, including the problem that it is unclear (to me, anyway) how to determine the “distance” between one action and another. But the biggest problem is that it adds another obligation to the person’s original obligation. The original obligation was to D; the new, and different, obligation is to “get as close as possible to D.” The person may indeed have this new obligation; and if he does, then he certainly ought to fulfill it. But we cannot pretend it is not a new, separate obligation. And notice that once we add this new obligation, we no longer have a case in which our person cannot fulfill his obligations. With this new addition, it seems that (stated in full) our person’s obligation is to “do D if possible, and to get as close to doing D as he can if D is impossible.” This our person is perfectly capable of doing; therefore the modification we have introduced no longer permits us to say we have found an exception to ought-implies-can.
It seems to me, then, that if an obligation can provide a person real-world, practical guidance about what the person really ought to do, then it is not an exception to ought-implies-can.
Now, I am not entirely clear whether Josh thinks the obligation he is ascribing to his friend can provide to his friend real-world, practical guidance. But there are some reasons to think Josh thinks the obligation can provide such guidance.
A. From the original post: “That is, my friend can’t do what (by hypothesis) he ought to do, namely make his stand to the end of producing social change, because the noncompliance of others makes it impossible for him to achieve his end. But if ought doesn’t imply can, then we could still say that he ought to make his stand.”
The reasoning required to get to the conclusion that Josh’s friend “ought to make his stand” is (for me) somewhat hard to follow, but the conclusion is clear: Josh is saying that his friend ought to make his stand. This is something which, as I have argued before, Josh’s friend is perfectly capable of doing.
B. From Josh’s July 7, 2004 11:58 AM comment: “Even if my friend can’t achieve Y, he could still be guided towards getting as close to Y as possible, which would have him do P, rather than ~P.”
So here Josh is suggesting his friend’s obligation is to get as close to doing Y as possible (recall that Y=”make his stand such that it has the intended social-change effect”); and, according to Josh, to “get as close as possible” to Y would be to do P (P=”make his stand,” whether it has its intended effect or not). I am not clear on how, exactly, an obligation to Y, combined with an inability to do Y, should imply an obligation to P, but that is irrelevant. The important point is that, in the end, Josh is attributing to his friend an obligation which his friend is perfectly capable of fulfilling: An obligation to P.
In both passage A and passage B, Josh’s conclusion is that his friend ought to “P”: ought to make his stand, whether there is any “point” in doing so or not. As I have argued before, I think Josh’s friend is perfectly capable of doing this.
On the other hand, there are some places where this interpretation of Josh’s remarks seems misguided. Consider, for instance:
C. There is an assumption running through much of Josh’s discussion here which I am not sure I agree with. That assumption is that a person is incapable of doing what is pointless. Josh seems to think this assumption is warranted if it is accompanied by the additional assumption that the person in question is rational, well-informed, and so on. Thus Josh says, in his most recent comment, regarding what exactly the fan is obligated to do:
“It’s not merely ‘stop being a fan.’ Instead, it’s ‘stop being a fan as a means of effecting social change.’ The problem is that this is something he *cannot* do (assuming the noncompliance of the other fans).”
Once again, I simply do not see how this is the case. The idea seems to be that if a person is rational, well-informed, etc., then the person will never do something which is pointless; everything the person will do will have a reason. But isn’t “to fulfill a moral obligation” a good enough reason? If the friend has an obligation to “stop being a fan as a means of effecting social change,” and if rational, well-informed people have reason to fulfill their obligations, then I just don’t see why the friend wouldn’t have reason to stop being a fan as a means of…etc. However, this again is irrelevant. The point is that Josh thinks his friend has an obligation to “take his stand” with the AIM of effecting real-world change, and Josh thinks that, under the assumptions of rationality, etc., his friend is unable to do this.
I apologize for this rambling, disorganized comment, which has gotten too long. The point I am trying to make is that I think Josh is trying to have his cake and eat it too. Josh is trying to ascribe to his friend an obligation which his friend is incapable of fulfilling, but which is still capable of offering to his friend real-world, practical guidance. If Josh is willing to let go either of the having or the eating of his cake, I do not think I will have any problems with Josh’s argument. Until then, I think Josh may not be being as consistent as he ideally would be.
But the obligations in these cases are not real obligations. Obligations come in the form of a action X, not an action X with a qualifier on other agents mental states. I’m not obligated morally to change someone’s mind about something. I may be morally obligated to try to change their mind through reasoning with them and whatever various actions I can take to that effect. But in the end, their failing to change their opinion is not my fault, but theirs.
The phrase “to effect social change” is considerably similar. It involves, changing other peoples mind’s. Since agents, and agency, depends on an agent’s ability to decide for themselves what is right, there seems to be something discordant with pinning obligations on people that involve the negation of another agency. Ought-implies-can is coupled with Kant’s other mandates, like “Never use an agent as merely a means to an end.” The obligation to effect social change in this case seems to couch other’s agency as merely a means to an end, which is why we get this inconsistency in the ought-implies-can rule for this particular case.
Hence, I don’t think this case is a bona fide exception to the rule, since no one can be obligated to alter another’s agency…
I’m not clear about what is supposed to be new here. Sure, concern about ‘dirty hands’ can pose a contributors’ dilemma. But we already know that there is a set of moral problems concerning aggregation and collective action which arise all over the place.
(See Ch.3 in Parfit’s Reasons and Persons; sec.25 seems to me to set out the issues here).
I’ve been away from this post for awhile, but let me try to respond to the recent comments, in reverse order.
First, on the comment from “adm,” the original issue, recall, was not about whether there are dilemmas about collective action. Rather, and this was originally a question, the issue was whether, first, it seems like this is a different approach than usual to dirty hands cases (that is, the approach would be to challenge ought-implies-can rather than what I take to be the usual approach of questioning whether there is an obligation), second, if it is already out there, where I might be pointed for more on the topic, given my relative lack of familiarity with the scholarship on dirty hands cases, and, third, whether those out there reading this (in what I like to think of as the Soup Bowl) considered it a viable alternative.
On Chris’s comment, I don’t know whether trying to create social change involves changing minds, rather than merely changing behavior (my guess, given analyses of such phenomena as racism as not merely behavioral, is that both would be involved, but that’s just an off-the-cuff guess). In any case, though, I was (again) simply granting the intuitive plausibility of my friend’s perceived obligation, and (again) I’m willing to hear arguments that it’s not actually an obligation despite its initial plausibility, though (again) such arguments, to not be question-begging, would have to assert more than ought-implies-can. And I found the Kant reference a bit misdirected: there’s nothing in effecting social change that necessarily requires treating others as mere means. Presumably, if you do it right, you won’t treat others as mere means. Incidentally, none of this is to say (as Chris seemed to suggest) that others’ noncompliance doesn’t also fall under their own responsibility.
As to David’s comment, my post might be a lot of things, but I’d hate to think it’s inconsistent! All along I’ve intended to be talking about an obligation for my friend to rescind his fanhood towards the end of effecting social change. Sometimes, merely as a form of shorthand, I’ve simply referred to this as an obligation to “take his stand.” I hope that shorthand hasn’t caused too much confusion. Regarding whether or not it is pointless, I’m not sure how else I can put my claim after what I’ve said above, other than to note that I’ve never claimed that ceasing his fanhood is pointless in the sense of “having no reason.” Rather, it’s pointless in the fairly ordinary sense that “it can’t effect what it’s supposed to effect,” just as it’s pointless for me to effect the complete eradication of world hunger (which, suppose, is my goal) by donating my rather meager savings to Oxfam. Furthermore, if I’m rational, well-informed, etc., then it seems fairly ordinary to say that completely eradicating world hunger will *not* be the goal of my donation.