Monday, August 11, 2008

Moral Demands and Compliance Effects

The concept of 'moral demands' is a familiar one: it concerns the sacrifice a moral theory asks us to make, i.e. the expected loss of welfare from conforming to its requirements, compared to how well-off we might otherwise expect to become. But Liam Murphy, in his Moral Demands in Nonideal Theory, argues that we should also consider the costs and benefits that accrue to us from others' compliance with the moral theory. This yields a broader concept -- that of "compliance effects" (p.52).

Murphy points out that for pretty much any (remotely plausible) moral theory, we'd be a whole lot worse off if no-one was complying with its requirements. It starts to look odd to condemn a theory as 'too demanding', then, given that the compliance effect on us is a net positive. One might turn to a comparative conception of demands, according to which utilitarianism demands too much of the wealthy because they would be so much better off under libertarian norms. But this objection may be turned on its head: "commonsense morality is very demanding on [a poor man] in virtue of the fact that he would do so much better under utilitarianism." (p.55)

P.S. Here's a delightful quote from p.60:

Our individual and social lives are so thoroughly structured by moral and political concerns that we apparently lack any independent perspective from which to examine the impact of those concerns on what they structure.

Though I'm even more struck by the point that our lives are "so thoroughly structured" by moral norms that most of the time we don't even realize it. (Consider, for example, the implicit interference that property gives rise to.)


  1. Very interesting idea, I look forward to reading that book.

    Since there is a small number of takers on this thread so far, perhaps a slightly tangential thought will be forgiven?

    I wonder whether there could be a distinction between a reason to be moral and a reason not to not be moral. I have in mind the value that accrues to one's narrow self-interest from moral action. This happens in a number of ways: happier conscience, better social status, more friends, better treatment, etc. I don't want to say that these effects are reasons to be moral. The reasons to be moral concern the advantages to everyone, not merely to oneself. One should help another for that other person's sake, not for one's own. However these effects may play a legitimate motivating role in ruling out reasons *not* to be moral. Morality as an institution may well be justified on the grounds that it would be better for us all, even though the same doesn't hold in the single person case. Now imagine an individual who really does care about helping other people for their sakes, but is also worried about the cost to himself. We can reassure that person with the effects described above. This is to rule out a reason not to be moral, rather than provide one to be moral. Does this make sense?

  2. Hi Barry, I take it your suggestion is that prudential reasons will (when all is considered) not be in conflict with moral ends anyway. That may be true in some cases, or to some degree, but I doubt the correlation is perfect. (Cf. the classist amoralist.) Suppose we are morally obligated to work long hours and donate all we can afford (i.e. beyond what's necessary for our own bare survival) to charity. It would be incredible to deny that this is, in fact, a significant cost to one's self-interest.

  3. Richard,

    “The concept of 'moral demands' is a familiar one: it concerns the sacrifice a moral theory asks us to make, i.e. the expected loss of welfare from conforming to its requirements, compared to how well-off we might otherwise expect to become.”

    I thought that moral demandingness was something different. Moral demandingness concerns the sacrifice a moral theory asks us to make, sure enough, but the sacrifice of self-direction in life, not of expected welfare. A moral theory is demanding when it demands of us that we deviate from the way we wish to lead our life for the sake of morality. (Compare what makes for a demanding training regimen. It's not the expected loss of welfare from following the regimen.)

    I don't find it odd at all to condemn a theory as overly demanding despite that compliance with its demands is a net positive for all. Compare: many demanding basketball coaches are just this way, compliance with their demands is a net positive for all, but the coaches are made no less demanding for this fact. Indeed, it is frequently the paternalistic demands of a moral theory that are the most demanding.

    Maybe there are simply two concepts here: the expected loss of welfare a moral theory imposes on someone, and the degree to which a moral theory requires one to deviate from how he wishes to lead his life. I submit, as a side point, that the latter better fits with what we mean when we say that utilitarianism is too demanding on us academicians.

    The main upshot: commonsense morality could not possibly be demanding-in-my-sense on the poor. Commonsense morality does not require of the poor that they deviate from how they wish to live their life.

  4. Jack, that's an interesting suggestion, but I'm not sure it has the implications you suggest. (Indeed, if we are preferentists about welfare, the distinction would seem to collapse: to be well off just is to live the life you wish to lead. If I were obliged to become a high-powered lawyer to farm my labour for charity, this not only is contrary to how I want to live my life, but - for this very reason - is bad for me. At any rate, I don't think it's the paternalistic element of utilitarianism that people are lamenting when they call it "too demanding". It's the sacrificing your own interests for others bit.)

    Most importantly, your alleged "upshot" is way off. You think a life of poverty and deprivation is "how they wish to live their life"!? The poor suffer from a gross lack of freedom (or "self-direction"), and commonsense morality just reinforces this.

    (The problem is that you are comparing commonsense morality using the status quo -- i.e. commonsense morality -- as a baseline. A less question-begging approach is the 'comparative conception of demands' discussed in my main post, whereby we compare against other moral theories. And obviously the poor would have far more autonomy under an a more egalitarian moral system than under commonsense morality. That is just to say that commonsense morality is very demanding for them, in terms of freedom no less than welfare.)

  5. Richard,

    We need to keep two things separate: (i) given compliance with a certain moral theory, the degree to which one lives a life one wants to lead, (ii) the degree to which the demands of a moral theory permit one to live the life one wants to lead. I am suggesting that it is the latter only that is relevant to how demanding a moral theory is.

    Let me thus reissue the upshot. The demands of commonsense morality permit anyone, affluent or penurious, to live the life they wish to lead (more or less). The moral demands of commonsense morality, themselves, are not forcing people to deviate from their preferred life course. It is for precisely this reason that I stand by my claim: commonsense morality couldn’t possibly be demanding on the poor.

    Here is a different question, though I fear you took the two questions to be one and the same. Given compliance with commonsense morality on the left hand and compliance with utilitarianism on the right, on which hand do people tend to lead more of the lives they wish to lead, or on which hand do people have greater ability to pursue the kind of life they want to lead? Probably the right hand, sure enough. (I think it is more than a stretch for you to call this ‘autonomy’ and ‘self-direction,’ but lets leave that issue aside). Compliance with utilitarianism enables the penurious. Compliance with commonsense morality delimits the affluent. I contend, however, that it is only the delimiting aspects of a moral theory, and not the enabling aspects, that are relevant to how demanding a moral theory is.

    Another way to see my point. Imagine that an evil sorcerer has decided that if everyone complies with a particular athletic training regimen he will enact great suffering on the world, greatly reducing our ability to satisfy our preferences. It would be bizarre, and in my view plainly false, to say that the training regimen is morally demanding, even though the result of compliance with the training regimen is reduced welfare.

  6. Jack, I see two issues here:

    (1) Should we look at only the 'active demands' of a moral theory, or its broader 'compliance effects' (i.e. including passive costs and benefits to oneself resulting from others' fulfilling their obligations under the theory), in assessing how burdensome it is? This suggested shift was the point of my original post.

    (2) Should these costs and benefits be assessed in terms of welfare or something more along the lines of 'freedom'?

    I argued in my previous comment that the answer to (2) makes no significant difference. In your latest comment, it looks like you are instead addressing issue #1. (You could restate everything you want to say in terms of welfare rather than 'the degree to which one lives a life one wants to lead'.)

    You write: "The moral demands of commonsense morality, themselves, are not forcing people to [sacrifice anything]... [thus] commonsense morality couldn’t possibly be demanding on the poor."

    That's certainly the standard way to look at things. The point of my post was to suggest an improvement. One problem with the standard view, which Liam Murphy points out, is that "a moral scheme that is perfectly enforced through law can impose no demands. This seems absurd." (p.48)

    Consider 'Enforced Utilitarianism', according to which the police must punish us if we fail to become high-powered lawyers and relinquish our income to the needy. Do you want to say that Enforced Utilitarianism "couldn't possibly be demanding on" the affluent? After all, it's not "the moral demands of [E.U.], themselves," which are forcing us to do anything. It's the police, i.e. other people, doing to us what the theory tells them to.

    Obviously, E.U. would impose massive [albeit 'passive'] burdens on us. There is no principled basis for ignoring this just because the effect is mediated through others' actions. That's not a relevant or important difference. (As Murphy says, "if we ignored the burdens flowing to our target agent from other agents complying with a given theory, we would end up with an artificially low assessment of overall demands.") Likewise when it comes to the passive demands of commonsense morality on the poor -- and, more importantly, the passive net effect -- for as Murphy says, "There is surely no reason to be interested in the measure of how other people complying with a theory will burden our agent without also being interested in how this will benefit him."

    I agree that this shift from active demands to compliance effects is conceptually revisionary, and at odds with some prima facie intuitions. It's not how people are initially inclined to think about the issue. But the revision is necessary, because the way people are initially inclined to think about the issue is incoherent.

  7. Richard,

    I think this will end in a difference of opinion. That’s okay. I feel like I learned a lot from hearing what you had to say (as usual). But let me say a few last words. Your breakdown into (1) and (2) is helpful. My view is that the demandingness of a moral theory is determined by the active demands of the moral theory that delimit individuals’ self-direction.

    Perhaps I need to read Murphy’s book. But by all initial appearances, it is Murphy’s reversionary view, and I take it your’s now too, that is riddled with incoherence. Consider again the evil sorcerer and the training regimen. Are we really to say that the training regimen is morally demanding? Really? There are plenty of truth things that can be said about the training regimen: that we hope there isn’t compliance with it, that compliance with it is bad, that compliance with it has unexpected negative consequences, that as compared to compliance with other training regimens compliance with this one is morally bad for all walks of life. Why go to back-bending lengths, distorting our conceptual practice in counter-intuitive ways, to claim that in addition, the training regimen is morally demanding? It is such an unintuitive consequence. After all, the demands of training regimens are practical, not moral.

    As for Enforced Utilitarianism, I don’t get the argument. I ask: The enforcers are enforcing what moral theory? Utilitarianism? Well, utilitarianism is morally demanding on the affluent (in my sense of morally demanding). The demands of utilitarianism, given a poverty stricken population, permit the affluent very little self-direction. So, part of the intuition that Enforced Utilitarianism is demanding is that the enforcer’s are enforcing a morally demanding moral theory. Also, the enforcer’s, themselves, are being demanding, not morally demanding but demanding nonetheless, the way that micromanagers and attention-to-detail coaches are demanding. An E.U. society is doubly demanding. There are the moral demands of the moral theory there being enforced. Then, there are the practical demands of the enforecer’s actually enforcing—the prodding, the threats and so on. So, I don’t get the argument. I have the intuition that in an E.U. society there is an awful lot of demanding, and indeed there is.

    (If it turns out that utilitarianism is false, then we tell a different story. People in an E.U. society face enormous practical demands only, for the true morality actually does not require that they act in the way that their local enforcer’s are making them act.)

    What seems to me incoherent it to lump together what a moral theory says, what it actually demands of people, and what results, as a causal and contingent matter, from compliance with the theory. I could understand arguing that compliance effects are more important than demandingness—on my consequentiualism days I feel exactly thus—but trying to yoke the two together seems to me philosophically confused. I don’t get the upshot.

  8. Also, the difference between assessing in welfare and assessing in freedom does dissolve when we adopt a compliance effect view, but it is absolutely apparent when we adopt an active demands view. On the active demands view, freedom is a matter of being free from active demands that force you to deviate from how you wish to live your life.

  9. Jack, I was thinking of E.U. as its own theory, which doesn't demand anything directly of us. (That is, we may stipulate that we have no moral obligations.) Instead, it obliges the police to force us to do all the things that standard utilitarianism would normally ask of us. Your view implies that standard utilitarianism is objectionably demanding on us, but E.U. isn't. I respond that there is no reason to prefer E.U. to utilitarianism. In any sensible or philosophically interesting way of assessing how burdensome a moral theory is for us, E.U. is no better.

    I suspect our disagreement stems from the different theoretical roles we see for talk of 'demandingness'. You want to accommodate prima facie intuitions and standard ways of describing things. I can't imagine why you'd bother, but hey, whatever. I'm more interested in the broader philosophical/ethical issues, e.g. assessing and objecting to moral theories, so I'm especially sensitive to the theoretical role 'demandingness' is supposed to play in objections to utilitarianism. Active demands just don't have any ethical interest taken on their own, and can't ground a coherent or sensible objection here. It's compliance effects which ethicists should really be talking about. So I am.

  10. P.S. the measure of moral demandingness ranges over moral theories, not training regimens or other things which may or may not have bad consequences. It sounds "bizarre" to call the regimen morally demanding because it's a category error.

  11. Of course, the training regimen objection is easily parlayed into a moral theory objection. Same sorcerer, this time if people comply with the demands of some gerrymandered moral theory, then the bad news. Still the gerrymandered moral theory is not, itself, demanding. It is certainly not made demanding by the extrinsic factor of the sorcerer, or so it seems to me. How could a sorcerer's whim make a moral theory more or less demanding?

    We just disagree about whether active demands have ethical interest taken on their own. Speaking my own mind: one of my deepest suspicions against utilitarianism is its many active demands; that is, its demandingness.

  12. If the moral theory requires you to do X even though this will foreseeably lead you to suffer under the sorceror's retribution, then the moral theory is clearly demanding/burdensome in this respect. More generally, any moral theory which requires you to do things regardless of the consequences is clearly going to be very burdensome when those consequences are bad for you.

    "We just disagree about whether active demands have ethical interest taken on their own."

    Really? You mean you actually endorse the claim that E.U. is better (less objectionably burdensome) than utilitarianism?


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)

Note: only a member of this blog may post a comment.