Skip to content

Moral Luck Revisited

The tragic sinking of the South Korean ferry raises again the problem of moral luck which Bernard Williams did so much to expose in his famous 1976 article on that topic. The South Korean president has now claimed that the captain of the ferry is a murderer, implying that he is subject to the same degree of blame as any other murderer.But consider another case, of a ferry similar in all respects – including the degrees of incompetence and cowardice among the crew – which, through luck, does not sink. There is no doubt that, if the incompetence of the crew were exposed in some way, they would be blamed —  but far less than that the captain of the actual ferry.

This very difference in attitude seems unfair and unreasonable, and Williams’s own solution to the problem is to recommend dropping what he called ‘the morality system’, which insists that blame should be distributed fairly in accordance with a person’s intentional actions and not their outcomes.

This issue is something I have discussed previously on this blog, wondering whether we might avoid the problem of moral luck by seeing morality itself as instrumental to non-moral ends. At an excellent conference on the Moral and Political Legacy of Bernard Williams, organized this week by Lorenzo Greco, at TORCH in Oxford, I outlined another possible solution (not one original to me, I hasten to add): that we follow Kant (and indeed Aristotle) in restricting blame to the voluntary (so both captains would be blamed equally), but permit other assessments of their actions which do not imply moral criticism. So we might say, for example, that the actions of the captain of the actual ferry were extremely unfortunate, regrettable, terrible. We might call his actions shameful, or we might even, as Simon Blackburn suggested in a fascinating paper at the same conference, react to the captain with horror.

But, as emerged in discussion, there are problems with this suggestion. The boundaries of blame are poorly defined, and in practice vague. If we meet the actual captain and tell him he acted shamefully or react with horror (having blamed him fully for his incompetence and cowardice), that is quite likely to encourage him to feel further guilt, or at least to suffer in certain ways, which brings about the kind of unfair moral luck we were trying to avoid. So if we are going to adopt the Aristotelian-Kantian strategy, it may be that we shall also have to recommend a lot more care than we currently take in expressing negative attitudes to the conduct of others even if those attitudes are not paradigmatically blame-involving. It may even be that we should try to avoid taking at least some of those attitudes in the first place.

Share on

6 Comment on this post

  1. Thanks very much for this! The line of thought is tempting but, as you rightly say, it leads to certain difficulties. Actually, I would like to suggest that Williams had already anticipated a closely related problem in “Moral Luck: A Postscript” (1995):

    “…what is the point of insisting that a certain reaction or attitude or judgement is or is not a moral one? What is it that this category is supposed to deliver? Writers on these subjects who agree that in some problematical context an agent may reasonably feel bad about something that he or she has done, and may perhaps reasonably attract negative reactions from others, are still often concerned to insist on the further question whether these various reactions belong to morality. I can only urge again the point I have tried to make in various other places, that invoking this category achieves absolutely nothing, unless one has some account of the singular importance of morality in this restricted sense. I still cannot see what comfort it is supposed to give to me, or what instruction it offers to other people, if I am shunned, hated, unloved and despised, not least by myself, but am told that these reactions do not belong to morality.”

    This is not just the problem of how we will keep moral and non-moral reactions separate, but the question of why we should think it matters that we are keeping them separate. From the point of view of the crew of the Korean ship that makes it into the harbor, it almost seems perverse to insist that while we are directing a series of negative attitudes towards them for their negligence, we are not blaming them for their negligence. The distinction is likely to strike them as somewhat scholastic.

    Furthermore, I am surprised that Blackburn, who has so powerfully illustrated the ways in which morality is primarily a product of sentiment, would suggest that a reaction of horror is somehow categorically distinct from a moral reaction. However, perhaps I misunderstand his suggestion.

  2. Thx Nick. Excellent comment. I did discuss that interesting passage from BW in my lecture, my main point being that he appears not to recognize the special disvalue of the pangs of conscience. But perhaps you’re right that he is implying that feeling shame and so on are just as bad, so we needn’t make a big thing out of what is inside and outside of the morality system. (Elsewhere in that piece he says that the voluntary is ‘essentially superficial’.) But giving up on the idea of fairness behind the morality system seems to me a big step, and close to justifying the punishment of the innocent. So that’s why I think that the implications of the Aristotle-Kant position (Williams of course thinks it’s just Kant’s, but that seems wrong to me) may be more radical than either A. or K. realized. I think you’ve understood Blackburn correctly. But in fact he thinks there is more to be said about the voluntary/involuntary distinction than Hume manages to say in App. 4 of the Enquiry, and develops a more nuanced position using Adam Smith. I think he would perhaps still be happy with the broadly Humean idea that the distinction isn’t terribly important, bec. both moral and non-moral reactions are there to make life generally better for us (a view I’m myself sympathetic to — hence my suspicion of the negative attitudes). If you have written anything on this, I’d be pleased to see it, and please also let me know ( if there’s anything you think it wld be helpful in the literature for me to take a look at. Cheers, Roger

  3. Anthony Drinkwater

    Thank you for your post, Roger.
    Perhaps I am alone, but I question your starting point that different treatment for your imaginary ferry crew would be unfair and unreasonable.
    Imagine the reverse case of “moral good luck” : an assassin fails to kill his victim because of some external event (the trigger of the rifle sticks, a passing truck blocks the line of fire, the potential victim slips on a banana peel and falls just at the moment of the shot….).
    Would it be fair and reasonable to treat the potential assassin exactly the same as if he had succeeded ?

    Your implied view may well work as a rhetoric device to remind those who are quick to condemn others’ behaviour that they themselves could be equally condemnable – such as Christ’s expressed view on adultery in the mind (Mathew 5.28), for example.
    Or else it might work as a personal philosophy of virtue ethics.

    But I suspect that it would be unwise to build a practical moral theory based on judging or condemning others purely on the basis of their intentions (whether this is dressed as moral or some other form of condemnation).

    1. Thanks, Anthony. I like the assassin example. But I’m not claiming that the failed assassin should be blamed as much as we’d now blame a successful one. Rather, what seems fair is that both should be blamed equally. In other words, level of blame is to be assessed from the ex ante perspective. How much we should blame someone for trying to kill someone else is a separate matter. I also agree with you that it might well be unwise for practical reasons to make a serious attempt to reform the morality we have. But we might still think there’s something to be regretted in the way things have turned out. (I say ‘might’. It may be that we’d actually do better with at least some level of reform, though there would almost certainly have to be different costs attached to e.g. successful and merely attempted assassinations so as to encourage potential assassins to turn their guns at the last moment.)

  4. Thank you for this interesting post and for the illuminating discussion. The debate about moral luck is indeed a vexed one, but I cannot help but wonder why the following option is so rarely discussed: define the notion used in the principle thus named (i.e. “people should only be blamed for outcomes within their control”) does not protect from blame agents such that (i) their action or inaction would have had undesirable outcomes, had luck not intervened, and (ii) the explanation of this counterfactual conditional is grounded in the plans governing the activity and the way the agent was performing the activity relative to which the event or absence of event we describe as an action or as an inaction could have been respectively a part/an absence.

    Applying this idea to your example of two negligent crews, both would be equally blameworthy given that the negligent crew whose negligence did not actually ended tragically (i) owes the absence of tragic outcome to moral luck and (ii) moral luck only since it is true that had they undertaken the right plans in the right way, i.e. with the right amount of attention, they would not have owed the absence of the tragic outcome to mere moral luck.

    In other words, it seems to me that moral luck is compatible with blaming people for failing to prevent undesirable outcomes just as it is compatible with blaming people for intentionally bringing about undesirable outcomes as long as moral luck does not affect the plans and the way those plans govern the activity that agents were in position to adjust so as to rule out those outcomes.

    Don’t you think that if this is a workable option, it justifies distributing blame in a way that closely matches the distribution that someone with the intuition expressed by “This very difference in attitude seems unfair and unreasonable” would like to see?

  5. Thanks, Andrews. Yes, I think this is very much in the spirit of the Aristotle-Kant position. There is still the difficulty of working out exactly how much to blame the captains and the crews, but that is a separate issue.

Comments are closed.