Act-Utilitarianism and the Contrary-to-Reason Objection

The following argument was inspired by the excellent work of two fellow PEA Brains: David Sobel’s “The Impotence of the Demandingness Objection” and Dale Dorsey’s “Weak Anti-Rationalism and the Demandingness of Morality”:

  1. If a normative requirement dictates that one should perform a certain act even though one has decisive reason, all things considered, not to do so, then it is what I will call a contrary-to-reason requirement — a CTR requirement for short.
  2. If a normative requirement is a CTR requirement, then it is to be rebuffed in the sense that one should give it no heed and should, in fact, act contrary to its dictates.
  3. Many of act-utilitarianism’s moral requirements are CTR requirements.
  4. Therefore, either many moral requirements are to be rebuffed or act-utilitarianism is false.
  5. Moral requirements are never to be rebuffed, for if S is morally required to do x, then S has sufficient reason to do x.
  6. Therefore, act-utilitarianism is false.

Note that this objection is not identical to the too-demanding objection, for this objection applies not only to theories whose requirements demand more from us than we have sufficient reason to give, but also to theories whose requirements demand less from us than we have decisive reason to give. To illustrate the latter, consider the legal requirement that one not commit suicide. This requirement is, I believe, a CTR requirement, for I think that one can have decisive reason (even decisive moral reason) to commit suicide even when it might be less costly for oneself to refrain from committing suicide. Suppose, for instance, that the only way to save my family from some unbearable burden is to commit suicide. In such a case, I might have most reason (and even most moral reason) to commit suicide even though I might be better off, self-interestedly speaking, not doing so. If this is right, then we can object to this legal requirement on the same grounds that I would object to the requirements of act-utilitarianism: they both require us to act in ways that we ought not to act.

Although this is not the too demanding objection itself, it is I think what leads many to reject act-utilitarianism on account of its demandingness: act-utilitarianism, it seems, demands that we make sacrifices that we have decisive reason not to make. The thought is, for instance, that agents have decisive reason to pursue their own projects and self-interest to an extent that is not permitted by act-utilitarianism.

Now, if I understand them correctly, both Dorsey and Sobel endorse (3), but would counter against this sort of argument by rejecting (5). But I think that this is too little too late. In rejecting (5), you may save act-utilitarianism as the correct moral theory, but you do so at the price of depriving it of much of its putative import. So I think that act-utilitarians would do better to reject (3), although my own view is that (3) is true. I think that act-utilitarians should reject (3), because I think that they will be unhappy with the implications of (4). For instance, I think that Peter Singer would be very unhappy if the only way to salvage his view about what we are morally obligated to give to the needy was to concede that we ought not, all things considered, give as much as we are morally obligated to give to the needy. I think that Peter Singer really wants to convince us that we ought (and note the lack of the qualifier ‘morally’) to give more to the needy and not just that we are, on some normative standard, required to give more. But I’m interested in what others think. Is the argument sound? If not, which premise should be rejected? Or, at least, which premise should the act-utilitarian reject?

58 Replies to “Act-Utilitarianism and the Contrary-to-Reason Objection

  1. Hi Doug –
    Great post! As you say, I wanna reject 5. (Actually, here’s a blatant appeal to authority: Singer rejects 5, too. Check out his “Replies” in *Singer and His Critics*. For another accusation that rejecting 5 is a phyrric victory for AC, check out Hurley’s “Does Consequentialism Make Too Many Demands, or None At All?” in Ethics a few issues ago.) Here’s a quick reply (don’t take it too seriously):
    We should not think that because some AC demands are CTR demands that all or most such demands are. This would depend on an inquiry into the nature of practical reason. But I think that most people would agree that we have practical reason to give away more than we do, and likely for consequentialist reasons. Hence in these cases, AC can influence our considered judgments about practical reasons, though it cannot, likely, push the pedal to the metal (and defend A-T-C rationality of self-sacrifice in order to avoid the deaths of, say, two others).
    The above is true even if we regard all rational choice as choice in which there are other obligations pulling against morality. But not all choice is like this. Consider, for instance, systems of moral education. If AC is the true moral view, it will have a huge role in attempting to influence the projects and desires of the next generation. I actually think this is a robust an attractive role, something that corresponds to Mill’s view expressed at U iii 10-11. Of course, that’s not going to get us all the way to AC, nor will it get us expansive rational authority now. But I think there’s much to be said for it’s importance nonetheless. At least, I certainly think it shows that rejecting 5 does not destroy the importance of AC.
    Small point. “Give no heed to” is not the same as “act contrary to”. A CTR requirement is, as you say, not rationally decisive. But that doesn’t mean that there’s no reason at all to conform to it. There may be a reason, but one that’s outweighed by a weighter reason. But if so, I am not required to give it no heed; I am surely at least permitted to give it heed appropriate to the weightiness of the associated reason.

  2. Hi Dale,
    Three quick points:
    (1) You say,

    If AC is the true moral view, it will have a huge role in attempting to influence the projects and desires of the next generation.

    You mean that, if AC is the correct moral theory, then it should play a huge role in our attempts to influence the projects and desires of the next generation, right? And I take that the ‘should’ here is the ‘should’ of morality, not the all-things-considered should, right? So even if AC is true, it could be that we should, all things considered, attempt to influence the desires and projects of the next generation so that they are very unlikely to comply with the demands of AC. Isn’t that right?
    (2) You also say,

    A CTR requirement is, as you say, not rationally decisive. But that doesn’t mean that there’s no reason at all to conform to it. There may be a reason, but one that’s outweighed by a weighter reason. But if so, I am not required to give it no heed; I am surely at least permitted to give it heed appropriate to the weightiness of the associated reason.

    Note that I didn’t say that you should give no heed to the reasons that count in favor of your acting as the CTR requirement dictates. You should, of course, give heed to all your reasons for action. What I said, though, was that you should give no heed to the requirement if it is a CTR requirement. CTR requirements should be given no heed. Moreover, you should not comply with them.
    (3) I’m aware of the Hurley article; I’m sympathetic, as you know. I’m surprised, though, about Singer’s position. I’ll have to look that up. If you have specific page numbers let me know. In any case, I’m sure that my students will be relieved to hear that Singer isn’t claiming that they ought to do all the demanding things that he claims that they are morally required to do. I’m relieved too.

  3. In any case, I’m sure that my students will be relieved to hear that Singer isn’t claiming that they ought to do all the demanding things that he claims that they are morally required to do.
    Singer’s position is more subtle than this paraphrase suggests. He believes morality gives us reasons for action, but thinks self-interest is a source of reasons as well. When morality and self-interest come into conflict, there is no higher court of appeal to adjudicate the dispute. There are things people ought to do from the point of view of the universe, things they ought to do from their own points of view, and no way of deciding between the two. Your students, I’m afraid, won’t be relieved to hear that reason is divided against itself.

  4. Doug,
    Thanks for the cool post and the shout-out. I think Paul Hurley’s somewhat recent paper in Ethics argues in something like the way you are arguing. It is coincidental that you picked Peter Singer to make your point, as he explicitly (at least at times–e.g. in the Singer and Critics book) allows that one may not have most reason to act as morality requires. Hurley puts this point, in “Does Consequentialism Make Too Many Demands or None at All?” Ethics 2006, by saying that such a move wins the battle but loses the war (the war being about what the agent has most reason to do). But I think that Railton, Singer, and I would have been tempted to say Sidgwick are consequentialists about morality without thinking that forces them to a view of reasons according to which we have most reason to be moral.
    It is a pity that we don’t always have most reason to do what is morally correct, but if that is the case I don’t think we should call morality whatever it is that we have most reason to do.

  5. This is a minor quibble but I’m no longer sure about what is meant by ‘normative’ and ‘requirement’ in connection to CTRs if there is *decisive reason not* to act in these ways. To avoid this problem, maybe a better language would be to say that according to the argument AU would false on the overall level but could be true on the contributory, pro tanto level on which its normative requirements stand.

  6. Hi David,
    You write, “But I think that Railton, Singer, and I would have been tempted to say Sidgwick are consequentialists about morality without thinking that forces them to a view of reasons according to which we have most reason to be moral.” Note that I didn’t claim that we always have decisive reason to be moral, but rather that we never have decisive reason to be immoral. And, as I understand his view, Sidgwick accepts (5). He accepts (5), because he accepts that there is always sufficient to act as one is morally required.
    You also write: “It is coincidental that you picked Peter Singer to make your point, as he explicitly (at least at times–e.g. in the Singer and Critics book) allows that one may not have most reason to act as morality requires.” But is this because he thinks that there is sometimes just as good a reason to do something other that what morality requires, or is this because he thinks that there is sometimes decisive reason to do something other than what morality requires? If only the former, then his position is consistent with (5). Do you deny (5)?
    Hi Pablo,
    Thanks for this. I was just going off what Dale said that Singer had said. I haven’t read Singer on this. But if you’re right, then Dale (and, perhaps, David too — although David was a bit more careful) are misrepresenting Singer’s position. According to you, Singer accepts something like Sidgwick’s dualism of practical reason, where there is always sufficient reason to act as morality dictates as well as always sufficient reason to act as egoism dictates. But that is perfectly consistent with (5).
    Hi Jussi,
    By ‘normative requirement’, I only mean ‘a requirement according to some norm/standard of behavior’. Given this definition, it seems that there are many normative requirements that are CTR requirements. For instance, many requirements of law and etiquette (which are norms/standards of behavior) are CTR requirements.

  7. One more thing: So Sidgwick would deny (3), not (5). And if Pablo is right about what Singer’s position is, then Singer would also deny (3), not (5). And this is exactly what I said that utilitarians ought to do. That, it seems to me, is the most plausible tact to take. But if Dale and David think that it would be more plausible to deny (3), I wonder if they might say why.

  8. Nice post! Help me out with what it means to have a CTR requirement. You give a sufficient condition in (1) here,
    If a normative requirement dictates that one should perform a certain act even though one has decisive reason, all things considered, not to do so, then it is what I will call a contrary-to-reason requirement — a CTR requirement for short.
    Are CTR requirements supposed to be normative requirements of moral theories (or other theories) that are true? Or are CTR requirements supposed to be normative requirements of moral theories (or other normative theories) simpliciter? I can see how AU considered not as a true moral theory might issue in a CTR requirement. But it is difficult for me to imagine a world (assuming there is one) in which (i) AU is true, (ii) AU issues a requirement and (iii) that requirement is a CTR. What I am getting at is that, under the assumption that AU is true, I’m not at all tempted to reject (5). Under that assumption, I’m tempted rather to urge people to readjust their expectations from morality.

  9. Hi Mike,
    I’m with you. I also find it difficult to “imagine a world (assuming there is one) in which (i) AU is true, (ii) AU issues a requirement and (iii) that requirement is a CTR,” for I’m inclined to think that (5) is a conceptual truth. So CTR requirements are supposed to be normative requirements of moral theories (or other normative theories), whether they be true theories or not.

  10. Doug, perhaps you could say more about what you mean by ‘requirement’.
    On one natural understanding, a requirement is a thing which is required. But the things which are required by AU, presumably, are acts, and I’m not sure how one ‘rebuffs’ or ‘pays heed to’ an act (unless it is the act of someone else giving you some advice).
    I suppose what you have in mind is that requirements are instructions or commands. But then it’s not clear to me how AU gets to be false on account of its issuing instructions which ‘are to be rebuffed’. Presumably instructions are neither true nor false. (Also, I’m not sure whether AU really issues instructions or commands. One might think, only people can do that.)

  11. Hi Campbell,
    This sort of language was only meant for rhetorical effect. But none of it is essential to the argument. So let (2) = “If a normative requirement is a CTR requirement, then one should act contrary to its dictates.” And let (5) = “If S is morally required to do x, then S has sufficient reason to do x, and S should do x only if S has sufficient reason to do x.”

  12. oh, I see (4) will need to be changed as well. Let me, then, just stipulate that ‘rebuffed’ is being used as a technical term such that a requirement is to be rebuffed if and only if one does not have sufficient reason to act as it directs one to act.

  13. I think I understand your reformulated (5), perhaps because you’ve dropped ‘requirements’ altogether. But I’m still unsure about (2). What is a requirement, and how does it have dictates?

  14. Campbell,
    A requirement is anything of the following form: “S is f-required to do x” — where ‘f’ stands for something like ‘morally’, ‘prudentially’, ‘legally’, etc. This requirement dictates/requires/prescribes that S do x.

  15. Ahhh, I see. So a requirement is a sentence.
    In that case, I take it, there can be both true and false requirements. So I think you need to revise your (4) and (5) as follows:
    4′. Either many true moral requirements are to be rebuffed or AU is false (because it implies false requirements).
    5′. True moral requirements are never to be rebuffed, for if S is morally required to do x, then S has sufficient reason to do x.
    Perhaps this deals with the worries raised by Jussi and Mike too.

  16. Hi Doug –
    Re: Singer. I’m not as up on my Singeriana as some others, perhaps. I’m thinking of his reported conversation with Nagel on “F, A, and M”. Anyway, I was reading this as a denial of 5, but it might be that he’s gesturing at a dualism view.
    Re: next generations. No–should of A-T-C obligations. It seems to me that moral reasons get defeated for A-T-C obligations only when there are competing systems of norms. In this sort of a case, there are no such systems of norms. Now, this might be avoided if one is, say, a hedonist about prudence. If so, prudence isn’t malleable in the way I took it to be. But I hereby flag my assumption that prudence is malleable–that a desiderative, or achievement, or some other such view is plausible. Do you have something else in mind?
    Re: why deny 5? You’re right that I give (in that paper) short shrift to the “dualism” view. I think a dualism is a position of last resort, and one that we needn’t accept. We can and do treat reasons as fungible, and we can and do issue A-T-C judgments in most cases. Anyway, I’m prepared to believe that there are some cases where there is sufficient reason to act both morally and, say, prudentially, but I think these cases just are when moral reasons and prudential reasons are equally weighty; I don’t believe we lack a perspective from which to judge A-T-C reasons.
    And if so, it seems to me that 5 is an incredibly strong view. Perhaps too strong to be plausible. There are all sorts of norms out there: moral norms, prudential norms, associative norms, professional norms, etc. It implies that in every case moral reasons are as weighty as, or weightier than, prudential reasons. This, I think, is difficult to accept no matter what your view of moral norms is. And I think it’s also difficult to defend on recognizable views of practical reason. But this is just burden shifting.
    Incidentally, my goal in the paper was to defend impartial views against demandingness objections. This is compatible with accepting a “dualism” view; you’ve now convinced me that I need to make this more clear.

  17. Doug,
    Sorry, my previous substantive post was, in parts, incoherent.
    You are thinking that it is seriously counterintuitive to allow that morality requires X but rationality requires not X. You think it not so worryingly counterintuitive if morality requires X but rationality permits not X. Do I have you right to there?
    I take your point about Sidgwick–that he is only committed to the latter. My hunch is that Singer and Railton would accept the stronger claim, but I can’t back that up with a passage.
    Clearly the stronger claim is more hostile to the authority of morality. I’d like to think more (and get more help thinking about) the significance between these two positions. Maybe part of the thought is that if morality has no special status then it is a waste of time to get serious in developing a moral theory whereas if morality has the “gives undefeated reasons” status, then developing a moral theory is still directly relevant to what to do. I’d be interested to hear more at this point.
    I have made my peace with the rejection of your 5 above. Thus, because I am not working to deveop a conception of morality that one always has an undefeated reason to obey, I find thoughts like those pressed by the Demandingness Objection less telling. I see the Objection as just noticiing the strain in trying to keep morality’s instructions and rationality’s instructions from coming apart.

  18. Doug,
    Sorry, my previous substantive post was, in parts, incoherent.
    You are thinking that it is seriously counterintuitive to allow that morality requires X but rationality requires not X. You think it not so worryingly counterintuitive if morality requires X but rationality permits not X. Do I have you right to there?
    I take your point about Sidgwick–that he is only committed to the latter. My hunch is that Singer and Railton would accept the stronger claim, but I can’t back that up with a passage.
    Clearly the stronger claim is more hostile to the authority of morality. I’d like to think more (and get more help thinking about) the significance between these two positions. Maybe part of the thought is that if morality has no special status then it is a waste of time to get serious in developing a moral theory whereas if morality has the “gives undefeated reasons” status, then developing a moral theory is still directly relevant to what to do. I’d be interested to hear more at this point.
    I have made my peace with the rejection of your 5 above. Thus, because I am not working to deveop a conception of morality that one always has an undefeated reason to obey, I find thoughts like those pressed by the Demandingness Objection less telling. I see the Objection as just noticiing the strain in trying to keep morality’s instructions and rationality’s instructions from coming apart.

  19. Doug,
    Sorry, my previous substantive post was, in parts, incoherent.
    You are thinking that it is seriously counterintuitive to allow that morality requires X but rationality requires not X. You think it not so worryingly counterintuitive if morality requires X but rationality permits not X. Do I have you right to there?
    I take your point about Sidgwick–that he is only committed to the latter. My hunch is that Singer and Railton would accept the stronger claim, but I can’t back that up with a passage.
    Clearly the stronger claim is more hostile to the authority of morality. I’d like to think more (and get more help thinking about) the significance between these two positions. Maybe part of the thought is that if morality has no special status then it is a waste of time to get serious in developing a moral theory whereas if morality has the “gives undefeated reasons” status, then developing a moral theory is still directly relevant to what to do. I’d be interested to hear more at this point.
    I have made my peace with the rejection of your 5 above. Thus, because I am not working to deveop a conception of morality that one always has an undefeated reason to obey, I find thoughts like those pressed by the Demandingness Objection less telling. I see the Objection as just noticiing the strain in trying to keep morality’s instructions and rationality’s instructions from coming apart.

  20. Dale,
    Re: what you said about next generations: I don’t understand. Are you saying: “If AC is the true moral view, then we should, all things considered, act in ways that will make it more likely that the desires and projects of the next generation are in line with the demands of AC? I don’t see how the consequent follows from the antecedent. Also, why don’t you think that there can be competing non-moral reasons to act in ways that will make it less likely that the desires and projects of the next generation are in line with the demands of AC?
    I also don’t understand your reasons for rejecting (5). (5) does not imply “that in every case moral reasons are as weighty as, or weightier than, prudential reasons.” You can accept (5) and hold that many prudential reasons are weightier than moral reasons; it’s just if you accept (5), then you think that when prudential reasons are weightier than non-moral reasons, those moral reasons don’t issue in a moral requirement. For instance, I think that I have a moral reason to keep my promise to hold office hours. But if I’m given an opportunity to make a million dollars and if my doing so requires me to miss my office hours, then this, I think, is a case where the prudential reason I have to make a million dollars trumps the moral reason that I have to keep my promise. But I also think, in accordance with (5), that I’m not morally obligated to keep my promise under such circumstances, although I may very well be morally required to make it up to those students who waited in vain for me to show up to them.
    By the way, I’m also no fan of dualism. But one can reject dualism and accept (5).

  21. David,
    You have me right.
    You’re interested in hearing more as to why I accept (5). Well, my main reason is that I just find it intuitively compelling. I wish that I had something more helpful to say. Perhaps, I could also point to some of the considerations that Shafer-Landau offers in favor of the view that if S is moral required to do x, then S has a pro tanto reason to do x. I happen to think that these sorts of considerations count just as much in favor of (5): the view that if S is morally required to do x, then S has sufficient reason to do x.
    I’m curious. Do you not find (5) intuitive at all? I know that you have to reject it given your other commitments (e.g., to subjectivism about reasons and to a non-subjectivist moral theory). But do you reject it because you find these other views more intuitively plausible than (5), or do reject it independently of these other commitments?

  22. Consider the following claims:
    1) If morality requires X, then one has most reason to do X.
    2) Morality can require X, yet one might not have most reason to X. However, when morality requires X, one’s reasons always permit X.
    3) Morality can require X yet we may have most reason to not X.
    2 says one’s reasons can permit one to defy morality. 3 says one’s reasons can require you to defy morality.
    What I am not yet feeling in my bones is what the complaint is against 3 that 2 is sufficient to meet (rather than requiring a move all the way to 1). I don’t mean to be saying that I have an argument against that thought. Just that I don’t yet feel what would tempt one away from 1 and 3.
    Insofar as 3 seems counterintuitive to me, it seems (or seems to seem) that this is due to the attractions of 1.
    I suppose one thing that moves me here are cases where it feels to me like anything worth calling morality will require tremendous sacrifice on my part (e.g. if you don’t press this button we will torture you to death, if you do press this button we will destroy Ecuador). In such cases it feels to me like there is a conflict between what anything worth calling morality would tell me to do and what makes sense from my own vantage point. I have the sense that the resonance of “Why be moral” suggests that this apparent conflict is widely felt.
    Likely I am too theory infused to serve as a good detector of pre-theoretic intuitions, but I am tempted to say that the falsity of 1 seems unsurprising to me, but sad. 1 seem to me to have to-be-rooted-forness to them, but that does not make me think that it is intuitively correct.

  23. Consider the following claims:
    1) If morality requires X, then one has most reason to do X.
    2) Morality can require X, yet one might not have most reason to X. However, when morality requires X, one’s reasons always permit X.
    3) Morality can require X yet we may have most reason to not X.
    2 says one’s reasons can permit one to defy morality. 3 says one’s reasons can require you to defy morality.
    What I am not yet feeling in my bones is what the complaint is against 3 that 2 is sufficient to meet (rather than requiring a move all the way to 1). I don’t mean to be saying that I have an argument against that thought. Just that I don’t yet feel what would tempt one away from 1 and 3.
    Insofar as 3 seems counterintuitive to me, it seems (or seems to seem) that this is due to the attractions of 1.
    I suppose one thing that moves me here are cases where it feels to me like anything worth calling morality will require tremendous sacrifice on my part (e.g. if you don’t press this button we will torture you to death, if you do press this button we will destroy Ecuador). In such cases it feels to me like there is a conflict between what anything worth calling morality would tell me to do and what makes sense from my own vantage point. I have the sense that the resonance of “Why be moral” suggests that this apparent conflict is widely felt.
    Likely I am too theory infused to serve as a good detector of pre-theoretic intuitions, but I am tempted to say that the falsity of 1 seems unsurprising to me, but sad. 1 seem to me to have to-be-rooted-forness to them, but that does not make me think that it is intuitively correct.

  24. Consider the following claims:
    1) If morality requires X, then one has most reason to do X.
    2) Morality can require X, yet one might not have most reason to X. However, when morality requires X, one’s reasons always permit X.
    3) Morality can require X yet we may have most reason to not X.
    2 says one’s reasons can permit one to defy morality. 3 says one’s reasons can require you to defy morality.
    What I am not yet feeling in my bones is what the complaint is against 3 that 2 is sufficient to meet (rather than requiring a move all the way to 1). I don’t mean to be saying that I have an argument against that thought. Just that I don’t yet feel what would tempt one away from 1 and 3.
    Insofar as 3 seems counterintuitive to me, it seems (or seems to seem) that this is due to the attractions of 1.
    I suppose one thing that moves me here are cases where it feels to me like anything worth calling morality will require tremendous sacrifice on my part (e.g. if you don’t press this button we will torture you to death, if you do press this button we will destroy Ecuador). In such cases it feels to me like there is a conflict between what anything worth calling morality would tell me to do and what makes sense from my own vantage point. I have the sense that the resonance of “Why be moral” suggests that this apparent conflict is widely felt.
    Likely I am too theory infused to serve as a good detector of pre-theoretic intuitions, but I am tempted to say that the falsity of 1 seems unsurprising to me, but sad. 1 seem to me to have to-be-rooted-forness to them, but that does not make me think that it is intuitively correct.

  25. David,
    Isn’t the complaint just Sidgwick’s idea that there is no perspective from which to adjudicate between the demands of morality and the demands of prudence? If you think (a) that moral reasons and prudential reasons compete but are incomparable, (b) that, when competing reasons are incomparable, they do not defeat one another, and (c) that, when competing reasons fail to defeat one another, it accords with reason to act as either dictates, then you’ll see your 2 as being more plausible than your 3, for 3 wrongly presupposes that you can come to some judgment as to what you have most reason to do in cases where moral reasons and prudential reasons conflict.
    That said, I do find (1) plausible. My point was only that you don’t need anything that strong to make the case against act-utilitarianism.
    I have a feeling that the “Why be moral?”-question resonates more with those who accept a theory such as act-utilitarianism than it does with those who accept a less demanding moral theory.
    In your specific case, I think that it can make good sense, and from my own vantage point, to refuse to press the button. It seems to me that, when sufficiently weighty, moral reasons will carry more weight than even certain weighty prudential reasons will, and they will do so from my own vantage point.

  26. David,
    You have me right.
    You’re interested in hearing more as to why I accept (5). Well, my main reason is that I just find it intuitively compelling. I wish that I had something more helpful to say. Perhaps, I could also point to some of the considerations that Shafer-Landau offers in favor of the view that if S is moral required to do x, then S has a pro tanto reason to do x. I happen to think that these sorts of considerations count just as much in favor of (5): the view that if S is morally required to do x, then S has sufficient reason to do x.
    I’m curious. Do you not find (5) intuitive at all? I know that you have to reject it given your other commitments (e.g., to subjectivism about reasons and to a non-subjectivist moral theory). But do you reject it because you find these other views more intuitively plausible than (5), or do reject it independently of these other commitments?

  27. Doug –
    I keep biffing the “moral reasons/moral requirements” terminology. My bad. Anyway, what I meant by “moral reason” was “whatever is spit out by the moral point of view”, which I take it you’re calling a “moral requirement.” I think 5 entails that “moral requirements” are always at least equally weighty with prudential and other norms. And that, as I said earlier, seems too strong to me. You appear to defend its strength by appealing to an “architectonic” view of morality (cf. Robert Louden, “Can We Be Too Moral?”). Anyway, I’m not sure I see the plausibility of such an account independent of an antecedent commitment to 5. Seems more plausible to me to say that ATC moral “requirements” are just whatever the most weighty moral reasons are, not the sum total of the weighed-up moral reasons plus the weighed-up prudential reasons, plus the weighed-up associative reasons, plus the … . Offhand, this view seems conceptually cluttered. But I’m sure you’re “Are Moral Reasons” paper gives me a thorough thumping on this.
    On AC and future generations. Here’s what I’m thinking. Sorry I spit this out so quickly last night. I was really thinking of two different roles here. First, given the assumption that other systems of norms rarely conflict with putting into place a system of external sanctions, AC determines external sanctions, whatever it would say about those external sanctions. Call this “role 1”. I was also thinking that on not particularly controversial assumptions that if AC is in charge of cultivating the next generation, that one thing AC would recommend is, as Mill says, to cultivate the next generation to adopt projects that are consistent with AC’s demands. Call this “role 2”. Doing so will produce the best overall consequences than, say, allowing the next generation’s prudential interests to diverge from the overall good. Role 2 won’t go all the way to aligning prudence and morality for the next generation, but it will go some way. This further assumption is, of course, subject to qualification in various ways based on, e.g., archangel/prole concerns and other sorts of concerns. I’m just riffing on the following thoughts, from U iii 10:

    Not only does all strengthening of social ties, and all healthy growth of society, give to each individual a stronger personal interest in practically consulting the welfare of others; it also leads him to identify his feelings more and more with their good, or at least with an even greater degree of practical consideration for it. … Consequently the smallest germs of the feeling are laid hold of and nourished by the contagion of sympathy and the influences of education; and a complete web of corroborative association is woven round it, by the powerful agency of the external sanctions.

    Seems to me that if AC is in charge of the “powerful agency of external sanctions”, it will try to encourage this sort of unity btwn one’s own welfare and the welfare of others. It may not get all the way there, but it is still a powerful role for AC morality even if we deny 5.
    You worry about my totally undefended assumption that no systems of norms conflict when deciding upon systems of external sanctions. I admit the possibility of occasional conflict–I should have qualified. There may be exceptions, but not robust exceptions. All I wanted to show is that the denial of 5 still allows a robust role for AC morality. Is there still some reason for thinking that denying 5 is a phyrric victory only?

  28. Hi Dale,
    You write, “I think 5 entails that ‘moral requirements’ are always at least equally weighty with prudential and other norms.” But what does it mean for one type of requirement to be more, less, or equally weighty than another? Reasons have a certain weight to them, but I didn’t think that requirements had any weight to them. Since I don’t understand the claim that you say is entailed by 5, I find it difficult to assess whether it is, in fact, entailed by 5 as well as difficult to assess whether it is, as you allege, implausibly strong. Could please explain what you mean.
    Perhaps, you mean something along the following lines. Some philosophers do talk about one prima facie duty being more stringent than another, and by this they mean, I take it, that it takes more countervailing considerations to prevent this prima facie duty from becoming an actual requirement than it does to prevent the other prima facie duty from becoming an actual requirement. So maybe you think that (5) entails that it takes weightier countervailing considerations to prevent a prima facie moral duty from becoming an actual all-things-considered requirement than it does to prevent a prima facie prudential duty from becoming an actual all-things-considered requirement. But I don’t see how (5) entails this. Again, consider the fact that I have a prima facie moral duty to keep my promise to hold office hours as well as a prima facie prudential duty to do what is needed to make a million bucks. But I can accept 5 and believe that it takes more to prevent the latter from generating an all-things-considered requirement than it does to prevent the former from generating an all-things-considered requirement.
    Regarding the next generation stuff: As I understand the view, AC is a criterion of rightness. That is, it is of the form: S’s doing x (a concrete act) is morally permissible just when, and because, it meets condition C. But you suggest that AC not only determines whether concrete acts are morally permissible, but also “determines external sanctions.” And you say that it can be put “in charge of cultivating the next generation.” I know this only loose talk, and a certain amount of loose talk is fine. But, in this case, I really don’t understand what you’re trying to say. What exactly is it that you think follows from the claim that “AC is the true moral view”?

  29. Here’s an intuition-tester that I just love. It’s from Marcia Baron. Imagine that you are a mother whose child is so ill that she might soon die without a certain lifesaving operation. There is a waiting-list at the hospital for similar cases. You work at the hospital so you could ‘pull springs’ behind the scenes and get you child to the front of the list. This would give her increased odds to survive.
    Now, the thought is that morality and moral reasons require you not to pull strings (pretty much on any theory of morality). It is not justifiable on impartial grounds (the other children and parents are in the same situation) and neither does it maximise any impartial value. However, it does seem to be true in the case that the mother has most reason to pull strings to save her child – that is what she ought to do. So, I find this case supporting David’s 3 against his 1 and 2.

  30. Jussi,
    I don’t see how this helps much. If David’s 1 is a conceptual claim, then we should just come to the conclusion either that it is false that the mother is morally required to refrain from pulling strings or that it is false that the mother is rationally required to pull strings. And even if David’s 1 is not a conceptual claim and so depends on what the correct substantive accounts of rationality and morality are, then it’s still not clear that it helps, because I suspect that anyone who accepts David’s 1 is likely either to reject a theory of morality that according to which the mother is morally required to refrain from pulling strings or reject a theory of rationality according to which the mother is rationally required to pull strings. I, for instance, don’t think that the mother is rationally required to pull strings. Would you think that if the mother was fully informed and yet refused to pull strings that she would be guilty of irrationality?

  31. Note also that David formulated his claims in terms of ‘most reason’, whereas I’ve been using the phrase ‘decisive reason’. They are not necessarily equivalent. It could be that the mother has most reason to pull strings, but, for some reason, those reasons don’t generate a rational requirement. In that case, she would have most reason, but not decisive reason, to pull strings.

  32. I had another post disappear in to the aether. Doggone it. Anyway, two points in response.
    1. I think we’re having a Cool Hand Luke moment on the distinction btwn “moral reasons” and “moral requirements”. I thought I understood how you were using it, but I guess I didn’t. I want to leave conceptual room for moral anti-rationalism–you appear not to, so maybe that’s one obstacle. So I’ll try one more time, in language I’m comfortable with. Morality issues requirements. So does prudence. So does the “personal point of view,” and various other systems. So do, for that matter, debutante norms, norms of Victorian Midshipmen, etc., etc. Some of these requirements provide practical reasons. Others don’t. Could be that morality is like the latter category, more like debutante norms or “Klingon Honor Norms”. (I take it that this is Foot’s now repudiated position in “Hypothetical Imperatives”.) Could be that morality is more like prudence, which clearly does provide practical reasons. What I was suggesting was that (5) requires that the practical reasons generated by moral requirements are always as weighty as the practical reasons generated by every other system of norms from the point of view of all-things-considered rationality; what we have, all-things-considered, most reason to do. And that’s too strong. Am I still being unclear? I’m not sure I know how to state it in a way that’s any plainer.
    2. I think the above helps to explain unclarity on my other proposal. Sorry for the “loose talk”; I was using shorthand in order to reduce the length of the post. Here’s the long version. I think moral requirements always provide practical reasons (morality isn’t like the Klingon norms), but that those reasons can be outweighed. I was interpreting your original rejection of this move as claiming that if the reasons generated by morality can be outweighed, then, even if AC is the right account of morality, this is a phyrric victory, because in so many of its important demands, morality will get outweighed by other systems of norms. Do I have you right? However, I deny that this is fully phyrric for AC. I deny it for other reasons (see my first comment), but I also deny it because there are choice scenarios in which other systems of norms conflict only rarely. In particular, I was thinking of decisions to impart widespread systems of external sanctions, especially those that dealt with the cultivation of the next generation. This, I claim, is an important area in which the reasons generated by morality are not outweighed by the reasons generated by other systems of norms, or are outweighed only rarely.
    I fully understand that AC is a criterion of rightness and not a decision procedure. My further claim was that given that, in most cases, AC morality does not get outweighed at the level of practical reason in circumstances involving the implementation of widespread external sanctions, morality (and hence all-things-considered practical reason) will issue in demands for policies that will cultivate the next generation to adopt prudential desires and projects, and relationships with others, that are compatible with the moral requirements of AC. Why? Because those are the external sanctions that produce the best consequences. Better consequences result when agents are not rationally required to do that which produces non-optimal consequences. But this is also imperfect.
    Adding this all up, even if morality can be occasionally outweighed by other systems of norms, if AC is true, it is not robbed of all its force when it comes to all-things-considered rationality. Am I still speaking Klingon?

  33. Doug,
    I don’t know. It seems helpful to me. I like these kinds of real life cases because they are good in pulling out our pre-theoretical intuitions. Of course, 1 can be presented as a conceptual truth. But, our pre-reflective semantic intuitions seem not to support this claim if we find intuitive cases where there is more reason to take the non-moral option.
    On the non-conceptual side, of course the defender of 1 can do what you say. But, that is not going to move us much before she gives strong grounds for her substantial views about morality and rationality. Unless such grounds are provided, I rather hold onto my intuitions about the case than accept the substantial views that support 1.
    I don’t think the problem is that much the most reason/decisive reason distinction. I’m much more worried about the slide from talking about reasons to talking about rationality and rational requirements. You started from reasons and I thought that was good. But, then we got onto rationality from that. I think this is a mistake. I don’t think it is always irrational to do the act for which there are less reasons. So, even if there is most reason or decisive reason to act in one way, it is not necessarily the case that the agent who does something else can be said so be irrational. But, that is another story I guess.

  34. Hi Dale,
    Let’s stick to the first issue for now. I may get back to you on the next generation stuff. Now, on the first issue, you write: “(5) requires that the practical reasons generated by moral requirements are always as weighty as the practical reasons generated by every other system of norms from the point of view of all-things-considered rationality.” I find this curious. I don’t think that moral requirements generate reasons, but rather that it’s the other way around: reasons (specifically, moral reasons) generate moral requirements. To illustrate, assume that, in Singer’s famous case, you’re morally required to pull the drowning kid out of the shallow pond given that it only means that you’ll get your clothes dirty and be late for your lecture. Now I think that the fact that your pulling the drowning kid out would save his life is a reason (more specifically, a moral reason) for you to do so. You wouldn’t deny this, would you? Is it your view, though, that the fact that you’re morally required to pull the kid out provides you with an additional reason to pull the kid out? If yes, then we can talk about that, for I would disagree. If not, then we’re just talking about the moral reason you have to pull the kid out. But note that (5) does not entail that this moral reason overrides all non-moral reasons (e.g., prudential reasons) in determining the all-things-considered rational status of your act. (5) entails only that when the moral reason is overridden by non-moral reason, the moral reason doesn’t generate a moral requirement. That is, if you had a much stronger prudential reason to not pull the kid out such that is would be, all things considered, objectively irrational for you to pull the kid out, then you would not be, in such circumstances, morally required to pull the kid out.

  35. Hi Jussi,
    I agree that “even if there is…decisive reason to act in one way, it is not necessarily the case that the agent who does something else can be said so be irrational.” But isn’t this only because one could have false beliefs? Note, then, that I asked, “Would you think that if the mother was fully informed and yet refused to pull strings that she would be guilty of irrationality? [emphasis added]”
    Maybe we just disagree in our pre-theoretical intuitions. But there are a number of people who would agree with me about the pre-theoretical intuitions are here–see, for instance, Josh Gert.

  36. Hi Doug –
    I am ecstatic to leave the other issue behind! You use “moral reason” to refer to something I would call a “moral consideration”. And here’s why. When I read “reason” I immediately think: ooh–something that counts in favor of my doing something, something that’s normative. But I want to allow the conceptual possibility of strong moral anti-rationalism (the claim that a moral verdict that x is required never counts, non-derivatively, in favor of my performance of x, the claim that morality never provides non-derivative practical reasons). I want to be able to ask whether moral considerations or moral requirements are things that count in favor of doing something; whether they provide “reasons”. (I think if anything provides practical reasons, it’s going to be moral requirements; I don’t see that we have any reason to act on moral considerations that are outweighed.) Hence, I don’t want to say that what you’re referring to counts as a “moral reason,” but rather as a consideration that helps us determine what morality requires. Whether “moral requirements” provide reasons will depend on a substantive inquiry into the nature of practical reason and the relationship between morality’s verdicts and rationality’s verdicts. Now you might be interpreting “moral reason” as neutral between moral anti-rationalism and moral rationalism. If so, that’s fine, but it just needs to be made clear in a way that I think the terminology doesn’t make clear. If we interpret “moral reason” as something that genuinely counts in favor of my doing something, this seems to rule out strong moral anti-rationalism on a conceptual level. I don’t want to do that. Like Jussi, I think, though strong anti-rationalism is false, it’s not conceptually false. Sorry for so many scare quotes.

  37. That depends. Does fully informed mean all factual information or factual information and all the information about her reasons? The first doesn’t get us anywhere as the rationality would depend on much of the reason-judgments she has made. If the latter is meant, then presumably she has judged that she has most reason to pull the strings (assuming she has most reason to act in this way). If she has made this judgment, then I do think rationality requires her to act accordingly. Rationality does require you to do what you have judged you have most reason to do (I’d like to add something about wide-scopes here).
    Right, on Gert’s view rationality can require only not bring about harm to yourself (if remember this right) when there are no relevant justifications around. I have yet to find a normal person who finds the consequences of this view of what rationality requires intuitive.

  38. Jussi,
    Yes, I find that implication of Gert’s view implausible as well, but the one that is relevant to our discussion is that it is always rational to cause harm to oneself when there is some compensating benefit to another. I’ve asked a few people and I have yet to find someone who holds both that morality forbids pulling strings and that rationality requires pulling strings. Those that I’ve asked think that it’s rationally permissible to refuse to pull strings given that this would involve taking someone else rightful place in the queue for the life-saving operation, and this is after I make clear to them that I’m concerned about objective, not subjective, rationality. Perhaps, the experimental philosophers can determine whether I’m right about this.
    So is it your view Jussi that the mother is rationally required to pull strings? Do you think, then, that it would be irrational for the mother to refuse to pull strings even if she both knew all the relevant factual information and had no judgment of the following sort “I have decisive reason to pull string”?

  39. Dale,
    I’m really confused now. As you do, I take moral reasons to be a genuine reasons for action. What led you to think otherwise? So can we agree that a moral reason is something that is both a moral consideration and a genuine reason for action? If not, what exactly is a moral reason on your view?
    Now you claimed that (5) implies “that in every case moral reasons are as weighty as, or weightier than, prudential reasons.” This is false, but you then retracted this, and claimed instead “(5) requires that the practical reasons generated by moral requirements are always as weighty as the practical reasons generated by every other system of norms from the point of view of all-things-considered rationality.” I then asked what you meant by ‘a reason that’s generated by a moral requirement’. But you never answered the question. Is the fact that x is morally required itself a reason to perform x? Or are you referring to the reasons that we have to perform the act that is morally required (e.g., the reason you have to pull the kid out of the pond: that your doing so will save his life). If it’s the latter, (5) does not entail that these reasons are always at least as weighty as the reasons that we have to do what prudence requires.

  40. Hi Doug –
    I’m sorry I apparently haven’t latched on to your wavelength. I hope I’m not losing my mind, here. That’s always a possibility, I suppose. You say “moral reasons generate moral requirements”. The direction of this relationship seems backwards to me. It appears, on the assumption that “moral reasons” are “genuine reasons for action”, to rule out on a conceptual level the possibility of strong moral anti-rationalism–the claim that the fact that x is morally required is no (non-derivative) genuine reason for action to x. I was never assuming that you had a different interpretation of “moral reason”. But I think maintaining the conceptual coherence of an alive-and-well position seems to me sensible. I wanted to allow you the possibility that you could accommodate its coherence (i.e., by offering an alternative understanding of “moral reason”–a non-normative understanding), a possibility you seem to resist.
    When I say “moral requirements generate reasons” I just meant exactly what you mean by the same talk in your previous post, only with the explanatory relationship reversed. I mean the first option: “the fact that x is morally required itself a reason to perform x”. It seems to me that this is the only sensible way to characterize the relationship between practical reason and morality. And, by the way, you’re mistaken in thinking that my earlier statement is false: it’s only false if your explanatory direction holds; that is, if moral reasons generate moral requirements. But I want to resist this. I want to say that “moral requirements generate (or don’t generate) moral reasons”, not the other way around, as you say. I don’t see any other way of accommodating the conceptual coherence of a conceptually coherent position. The two claims are identical on this interpretation.

  41. Hi Dale,
    Hey, it takes two to be on different wavelengths. So maybe I’m the problem.
    In any case, I thought that strong moral anti-rationalism was the denial of weak moral rationalism, which I take to be:
    WMR: If S is morally required to do x, then S has some reason to do x.
    I didn’t think that WMR involved a commitment to the view that the fact that x is morally required is itself a reason to do x, as you seem to assume. You say, for instance, “strong moral anti-rationalism–the claim that the fact that x is morally required is no (non-derivative) genuine reason for action to x.”
    Also, I don’t see why the assumption that moral reasons (for action) are, by definition, a species of reasons for action in conjunction with the claim that moral reasons, if there are any, are the sorts of things that can generate moral requirements “rules out on a conceptual level the possibility of strong moral anti-rationalism,” i.e., ~WMR. Strong anti-rationalism is still a conceptual possibility, for it may be that there are no moral reasons (that is, no moral considerations that are genuine reasons for action).
    In any case, I now see the source of our disagreement: You assert and I deny that the fact that x is morally required is itself a reason to do x. Is the fact that x is required by etiquette itself a reason to do x apart from whatever good consequences abiding by that requirement might have? If not, why the difference? And doesn’t your view rule out the conceptual possibility of ~WMR — at least, it does if there are any moral requirements? On your view, if there is a moral requirement to do x, then there is, necessarily, a reason to do x, for the fact that x is morally required is itself a reason to do x. This seems like bootstrapping to me. Someone asks you “Why be moral?” and you answer: “Because it is moral, and that is itself a reason to be moral.”

  42. Doug,
    I agree with the people you’ve asked. But, I think the rationality question is misplaced here (ought questions and reason questions would be better). So, what I want to say about the case is:
    1. It would be morally wrong to pull the strings (this is because it is unjustifiable impartially).
    2. The mother has most reason to pull the strings – this is what she ought to do. (I’ll trust mothers I’ve talked to about this one).
    3. The mother is not rationally required to pull strings – she is not irrational if she doesn’t (especially given that, as you say, she hasn’t judged that this is what she has most reason to do).
    This set of claims is consistent because I, unlike Gert (and maybe Parfit), think that rational requirements are mainly requirements of consistency internal to the agent’s psychology. And, one can be coherent and still act against the best reasons. Gert’s view about rationality is in contrast externalist and I think that creates major problems.
    Requiments of reasons, as I see them, on the other hand, are not internal to the psychology of the agents. So they are requirements made by reasons – things in the world. Rationality in part helps us to respond to such things but it does not consist of doing what there is most reason to do.

  43. Jussi,
    Fair enough. I’m with Parfit and Gert, and you’re with Scanlon (or, at least, Scanlon of WWOtEO). I wonder whether people have pre-theoretical intuitions regarding 2 (about what they ought to do) as distinct from 1 and 3.

  44. Dale,
    On more thing, I though that we could distinguish between weak, moderate, and strong moral rationalism as follows:
    WMR: If S is morally required to do x, then S has some reason to do x.
    MMR: If S is morally required to do x, then S has sufficient reason to do x.
    SMR: If S is morally required to do x, then S has decisive reason to do x.
    But how do you lay out the landscape in terms of a commitment to the “claim that the fact that x is morally required is…[a] (non-derivative) genuine reason for action to x”?

  45. Doug,
    that’s good. I would guess that people would more probably have intuitions about what ought to be done or should be done in the situation than about what would be the rational course of action. I’ve thought that ‘rational’ is philosophical term of art that isn’t really home at people’s everyday practical thinking.

  46. Doug –
    I think I misspoke. When I said “the fact that x is morally required is itself a reason to perform x” I meant that to be conditional on accepting a form of moral rationalism: if some form of moral rationalism is true, this would be the relationship between morality and reasons. Could be that the fact that x is morally required is no reason at all to perform x. Could be that morality is like the norms of the Debutante Ball Society, which don’t provide reasons at all. Could be that morality is like the Klingon Norms. So I didn’t mean that to be a conceptual claim. But I think that claim could be false.
    You define Weak Moral Rationalism as follows, and Strong Moral Anti-Rationalism as the denial of this claim: “If S is morally required to do x, then S has some reason to do x”. Here’s why I was interpreting your view as committed, conceptually, to WMR. Because WMR is a conditional, the denail of WMR is a conjunction (call this the SMAR conjunction): “S is morally required to do x” AND “it is not the case that S has some reason to do x”. But, I thought that, on your view, we arrive at moral requirements through an examination into moral reasons; as you say: “moral reasons generate moral requirements”. So if that’s the case, it sounds ot me like you cannot accommodate both conjuncts: that an agent S would be morally required to x, but also not have any reasons. This is because moral requirements are generated by moral reasons.
    As I’m reading this over, there may be another possibility. Your term “generation” might have been meant as a sufficient, but not necessary, condition. You might have said that moral reasons will generate moral requirements, but that moral requirements can be generated by other things than moral reasons. (This sounds odd to me, but maybe.) That would allow the conjunction: the existence of moral requirements, but the non-existence of moral reasons.
    Re: second question. I meant WMR. Does that get me into trouble? It might get me into trouble.

  47. Hi Dale,
    As you point out, my view does not entail that ~WMR is conceptually impossible, for my view is that moral reasons (if there are any) can and do generate moral requirements under certain circumstances, not that there can be no moral requirements absent moral reasons.
    Now could you explain why you think that “(5) requires that the practical reasons generated by moral requirements are always as weighty as the practical reasons generated by every other system of norms from the point of view of all-things-considered rationality.” (5) doesn’t even imply that there are practical reasons generated by moral requirements. Of course, your view is that moral requirements generate practical reasons, but (5) doesn’t imply this. So if (5) doesn’t even imply that there are practical reasons generated by moral requirements, then it certainly doesn’t imply that there are practical reasons generated by moral requirements that are always as weighty as those generated by every other system of norms from the point of view of all-things-considered rationality.
    Here’s what (5) does imply: (5) implies that whenever S is morally required to do x but prudentially required to do y, then there will be at least as much reason, all things considered, to do x as to do y. But if what one is morally required to do is limited to only those acts that one has sufficient reason to do (as it would be if (5) were true), then I don’t see why this should be surprising or implausible. If you can’t be morally required to do something you don’t have sufficient reason to do but you can, let us suppose, be prudentially required to do something that you don’t have sufficient reason to do, then there will be some cases where these two types of requirements conflict and one has sufficient reason to do what’s morally required but insufficient reason to do what’s prudentially required, and there will be no cases where these two types of requirements conflict and one has sufficient reason to do what’s prudentially required but insufficient reason to do what’s morally required. But what’s problematic or implausible about this given our assumption that one moral requirements, but not prudential requirements, are constrained by what there is sufficient reason to do?

  48. Dale,
    One more thing: Even if we assume that moral (and other — such as, prudential) requirements generate reasons to comply with them, (5) wouldn’t entail, as you claim, that the “practical reasons generated by moral requirements are always as weighty as the practical reasons generated by” prudential requirement “from the point of view of all-things-considered rationality.” (5) implies only that the practical reasons that count in favor of complying with some moral requirement (where this includes both reasons that are and are not generated by the moral requirement itself) are in total just as weighty, all things considered, as the practical reasons that count in favor of complying with some conflicting prudential requirement. But, again, I don’t see why you would think this is too strong if what one is morally required to do is limited to only those acts that one has sufficient reason to do (as it would be if (5) were true).

  49. Hi Doug –
    What you said. I meant my claim to be equivalent to yours; I should have added the “If moral requirements generate practical reasons, then” to the beginning of the phrase. And if that’s right, my claim and your version of it are equivalent. Anyway, I just meant exactly this: “whenever S is morally required to do x but prudentially required to do y, then there will be at least as much reason, all things considered, to do x as to do y.”
    You’re right that on your view this shouldn’t be surprising, because prudential norms, associative norms, all other norms can influence the extent to which we are morally required to do something. So, in essence, I wonder if you might agree that our disagreement is, at base, terminological. You want to say no to AC, because AC offers CTR requirements, and moral requirements are never CTR. If I have a moral requirement, that just entails that I have sufficient reason to act. I want to say that AC is the true account of moral requirements, but it’s moral requirements that generate practical reasons (if, in fact, they do), and other sorts of norms influence the extent to which these moral reasons generate sufficient reasons to act. Isn’t this just a dispute between two different understandings of the term “moral requirement”? And I don’t, I guess, have a particularly satisfactory defense of my understanding of this debate, it’s just my hunch that the only thing that determines the moral requirements we have are moral considerations, not moral considerations (or reasons) plus associative, prudential, and other considerations. It’s an attempt at conceptual tidiness, and, I think, it conforms better to our ordinary understanding of the terms in play. Moral requirements are determined solely moral considerations, not by moral considerations and a host of non-moral considerations including prudence and associative considerations. Just identifying moral requirements with that which we have most reason to do appears to capture one intuition (i.e., 5) in a way that’s a bit too easy: by just changing what we mean by moral requirement. Anyway, this is just a hunch. (p.s. Have you looked at the Louden article? You might find that congenial.)

  50. Sorry – I just saw your newest post. I’m a bit confused, but I *think* this is what you mean, no?
    If moral requirements generate practical reasons, (5) implies that ALL the practical reasons to act on moral requirements (including derivative AND non-derivative reasons) will be at least as weighty as the practical reasons not to act in accordance with moral requirements (including prudential reasons, associative reasons, etc.).
    Is that right? Anyway, to be honest, that sounds too strong to me. I would think that at least in some cases, you might have a moral requirement to x, and a variety of (derivative and non-derivative) reasons to x, but a wide variety of reasons to act against x (let’s say prudential norms, associative norms, professional norms, etc., all point to y, rather than x). Sometimes the other side could be stronger. But, as I say, this shouldn’t be surprising on your understanding of the terms in play. And I think that might be a reason to prefer my account. I think this result is prima facie suprising. Again, a bare appeal to intuition.

  51. Sorry – I just saw your newest post. I’m a bit confused, but I *think* this is what you mean, no?
    If moral requirements generate practical reasons, (5) implies that ALL the practical reasons to act on moral requirements (including derivative AND non-derivative reasons) will be at least as weighty as the practical reasons not to act in accordance with moral requirements (including prudential reasons, associative reasons, etc.).
    Is that right? (Good point about the non-derivative reasons, by the way.) Anyway, to be honest, that sounds too strong to me. I would think that at least in some cases, you might have a moral requirement to x, and a variety of (derivative and non-derivative) reasons to x, but a wide variety of reasons to act against x (let’s say prudential norms, associative norms, professional norms, etc., all point to y, rather than x). Sometimes the other side could be stronger. But, as I say, this shouldn’t be surprising on your understanding of the terms in play. And I think that might be a reason to prefer my account. I think this result is prima facie suprising. Again, a bare appeal to intuition.

  52. Hi Dale,
    I think that there are good reasons to reject the claim that “[m]oral requirements are determined solely [by] moral considerations, not by moral considerations and a host of non-moral considerations including prudence and associative considerations.” If you’re interested, my thoughts on the matter can be found in my “Are Moral Reasons Morally Overriding?”.
    We certainly agree that (5) entails that “whenever S is morally required to do x but prudentially required to do y, then there will be at least as much reason, all things considered, to do x as to do y.” I’m just not sure why you thought that this entailment “is difficult to accept no matter what your view of moral norms is.” If your view is that moral norms are not very demanding and are, in fact, limited to requiring only acts that there is sufficient reason, all things considered, to perform, then this entailment does not seem “too strong to be plausible.”
    What is the Louden piece?
    I wouldn’t agree that our dispute is just terminological. For one, our dispute about whether the fact that x is morally required is a reason to perform x doesn’t seem to be merely terminological.
    I’m pretty sure that I meant what I said. But, in any case, why don’t we just stick with this: (5) entails that “whenever S is morally required to do x but prudentially required to do y, then there will be at least as much reason, all things considered, to do x as to do y.” My point was that (5) doesn’t entail that moral reasons override non-moral reasons or that the reasons that moral requirements generate (if they do generate reasons) are always just as weighty as the reasons that prudential requirements generate. As far as (5) is concerned, it could be that the reasons that the prudential requirement to do y generates are stronger than the reasons that the moral requirement to do x generates so long as there are other reasons to do x that ensure that there is at least as much reason, all things considered, to do x as to do y.

  53. Yeah, looking back on it, I think it was a mistake to call it “terminological.” That seems to trivialize what I, too, think is a real disagreement. I certainly do need to take a closer look at your paper–I scanned it while preparing mine, but I need to dig deeper.
    I think we’re on the same wavelength when it comes to the last point. All I meant by “derivative and non-derivative” was that “there [may be] other [possibly non-moral] reasons that ensure that there is at least as much reason” to do the morally required thing. Anyway, I think we’ve reached intuitive bedrock here; I still think that sounds surprising.
    Re: Louden. It’s called “Can We Be Too Moral?” I think it’s from Ethics in ’88 or something. It’s basically in response to Wolf and some others of her stripe, people who take up something like her position (like me). He’s suggesting that we shouldn’t think that we can be too moral, because we should think of “morality” as an all-encompassing practical inquiry, one that takes reasons of all sorts into consideration. He calls this an “architectonic” view of morality. I was reading your position as being similar to Louden’s, but you might want to distance yourself.

  54. Hi Dale,
    I don’t think that my view has much in common with Louden’s. In contrast to Louden, I do think that we can be too moral. And, contrary to Louden, I think that we shouldn’t always strive to be as morally good as possible. To be as morally good as possible, one must, I assume, not only always perform one’s moral duty, but also always perform supererogatory acts when possible. Moreover, when there is spectrum of supererogatory acts, where some go further beyond what duty requires than others, one must perform the morally best supererogatory alternative if one is to be as morally good as possible. But I think that sometimes there are decisive reasons, all things considered, not to perform the morally best act. Thus, I think that it is possible to be too moral. My view is not that one always has sufficient reason to perform the morally best act, but only that one always has sufficient reason to perform acts that are morally required. Morally required acts are not co-extensional with the morally best acts, except on implausible versions of maximizing act-consequentialism, which I reject.
    On my view, morality doesn’t encompass the whole of our practical reasons, for, on my view, what I have most moral reason to do is not always what I have most reason to do, all things considered. I don’t think that moral considerations override, or are always more important than, non-moral considerations.
    It seems to me that you have, at least, once above and, if I recall correctly, a couple times in your paper conflated what are, in fact, two separate issues:
    (1) Do moral considerations (and/or reasons) override non-moral considerations (and/or reasons)?
    and
    (2) Is MMR (the view that if S is morally required to do x, then S has sufficient reason to do x) true?
    One can, as I do, answer “no” to the first and “yes” to the second. And the fact that (1) seems implausible does not in the least count against the plausibility of (2). Of course, it sounds like you just don’t find (2) very plausible in itself. Fair enough. I was only objecting to your claim that (5) implies such things as “that in every case moral reasons are as weighty as, or weightier than, prudential reasons” or even that “practical reasons generated by moral requirements are always as weighty as the practical reasons generated by every other system of norms from the point of view of all-things-considered rationality.”

Comments are closed.