Hi all,
After an agonizing, frustrating, delay (curse you, Typepad!), I hope you'll all join me in welcoming Richard Arneson to PEA Soup! Dick is Distinguished Professor, and Valtz Family Chair of Philosophy at UC San Diego. Without further ado, then, I give you his post, titled "Moral Worth and Moral Luck".
-dd
In a famous essay, Thomas Nagel wrote, “Kant believed that good or bad luck should influence neither our moral judgment of a person and his actions, nor his moral assessment of himself.” He added, “Prior to reflection it is intuitively plausible that people cannot be morally assessed for what is not their fault, or for what is due to factors beyond their control.”
Moral luck occurs when someone is correctly morally assessed for aspects of her agency that lay beyond her power to control. The Control Principle denies that such moral luck occurs. But some moral assessments subject to moral luck seem unproblematic. So consider a Narrowed Control Principle (NCP): we are assessable as moral praiseworthy or blameworthy (worthy or unworthy, deserving or undeserving) only to the extent that the feature of our activity that is the target of these assessments lies within our power to control. Moral praiseworthiness should be distinguished from admirable excellence at moral/practical reasoning and execution of its deliverances. We should also keep in mind a close consort to NCP: the more difficult and painful it would be to do what I see is morally right, and the more cost I see I incur if I do it, other things being equal, the less blameworthy I am if I fail to do it, and the more moral credit I earn if I do it.
The crucial idea here is moral blameworthiness. One can’t be morally blameworthy for what lies beyond one’s power to control. Blaming the person might still make sense, insofar as doing so might bring about good consequences. Criticizing the person’s conduct as morally faulty in various ways might still make perfect sense. But if one doing the evil deed was beyond one’s power to control, the doing cannot be one’s fault. The same goes for moral praiseworthiness when that is specifically the opposite of moral blameworthiness.
In Groundwork section 1, Kant seems to be attracted by the no-moral-luck constraint, but does not accept it. He instead holds that the morally worthy agent does what is morally right for the reason that it is morally right. But consider agents who are innocently ignorant of facts or of moral principles that are needed for choice of right action in their circumstances. For example, consider innocent unjust aggressors. Factors beyond my power to control may bring it about that I am ignorant of correct moral principles or embrace false ones, and so fail to do what is morally right, so it’s plausible to hold that I can qualify as being morally worthy by doing what is morally wrong.
Is there an account of moral worth that satisfies the NCP? Consider the subjective account. Conscientiousness (version 1): The conscientious agent tries wholeheartedly to make her conduct conform to requirements of morality. Her concern is noninstrumental. Her concern is to do what is actually right, not what she happens to think right, so she wholeheartedly tries to discover what is morally right. The more conscientious effort she puts forth, the more morally praiseworthy she is.
Here’s a bad objection against Conscientiousness (version 1): So, there can be conscientious Nazis and Viking marauders, who qualify as morally worthy for their thuggery? This scenario should not faze us. The answer on the view being advanced is, in principle, yes, this is possible, and may occur. If one is tempted to doubt this, one might be holding the background belief that anyone who seeks the moral law will discover it. There is no reason to think this is true. Moral truth might be complicated, hard to discern, and my mental faculties may be not up to the task, try as I may. Anyway even if the optimistic view that anyone who looks for moral truth will find it were true, this would not impugn the subjective account. The upshot would be a denial that as a matter of fact there are morally worthy Nazis and Viking marauders.
Here’s a better objection: Someone knows her search to discover moral truth is going to be fruitless, but keeps trying and trying anyway. Someone else knows she can’t do what she morally ought, but pointlessly keeps trying anyway (resistance to torture example). Someone knows she is going to do the right thing without effort, but pointlessly puts in lots of effort anyway. –This defeats our proposal. We need to go back to the drawing board.
Conscientiousness (version 2). Same as version 1, with this addition. The conscientious agent makes efforts to learn what is morally right to the degree she thinks doing so is appropriate, sensible, and she makes efforts to bring it about that she conforms to what is morally right by her lights to the degree she thinks doing so is sensible, appropriate. Also, she tries (to the extent she thinks appropriate) to bring it about that her beliefs about appropriateness are correct, and so on. (There is a regress here; harmless I hope.)
A worry: Suppose you live in a society, a hunter-gatherer society perhaps, that lacks the concept of morality. There are just rules laid down; no distinctions drawn between types of rules. So it is a matter of luck after all that you do or do not have the opportunity to be conscientious, to become morally praiseworthy according to the subjective account.
Response to worry: Trying hard to obey all social rules does not make you morally praiseworthy, unless you (nonculpably) believed they all had effectively moral status. But I suggest any rational agent to some degree will recognize the moral question, what behavior one owes others, this issue to be assessed from impartial perspective, even if the recognition is inchoate. The rational agent not only can recognize the moral question but does. One might speak of protomoral concern. That one’s culture lacks a full conceptual repertoire in this regard renders it more difficult for one to be conscientious, so not being conscientious is to that degree excusable (affects one’s adjusted conscientiousness score; more on this later).
What if one has views like those of Bernard Williams, so does not think morality is supremely important? I assume in this essay such views are wrong, but an agent might nonculpably embrace them and this will again be an impediment to achieving conscientiousness, so requires adjustment in one’s “score.” If Williams is right, all the questions I consider must be differently posed.
Another worry. Morality might require that one act in a way that corrupts one’s character. Consider cases described by Gregory Kavka. It is morally right, assume, to form the conditional intention to retaliate massively against a nuclear first strike, in stalemate between superpowers. This means one is disposed conditionally to do wrong. So one is in this respect lacking in conscientiousness and morally undeserving. But how can deciding to do what is morally right and doing it make one morally undeserving?
Response: If doing the right thing (as one sees it) sets one’s will, to some extent, against doing what is morally right in future in some range of cases, we can still appraise the person’s overall conscientiousness (version 2) and the assessment might come out positive or negative. First, the agent might succeed in orienting her will toward doing what is right in domains of choice apart from the conditional retaliation intention, and score moral worth points for that. Second, it will continuously be right to renew and strengthen one’s commitment to conditional evildoing, because in these circumstances doing so is necessary to bring about the overwhelmingly good consequences that flow from deterrence. So one will have a continuous excuse, complete or partial, for having a malign will in this limited domain of conditional intention to retaliate. Third, if things go badly and the fatal moment when the disposition to retaliate is triggered, one might fight against it, and get moral worth points for appropriate struggle to quash the urge wrongfully to retaliate, whether or not one’s attempts are successful. The example is disturbing, but so far as I can see does not generate reasons to reject the subjective account of moral worth under review.
There is an obvious rival to the subjective account of moral worth/praiseworthiness/deservingness. This is the objective account: an agent becomes morally worthy (her actions have moral worth) to the degree that she is moved to do what is morally right by perception of the moral reasons that actually are the reasons that make what she does right. Also, the greater the agent’s concern to act on correct reasons (not reasons she merely deems correct), the more morally praiseworthy she is for doing the right thing. Nomy Arpaly has done excellent work that articulates and defends just this view. My discussion here borrows from hers.
Examples: Huck Finn, the character in the Mark Twain novel, who decides not to care about morality, not to follow his conscience, but instead to act as he pleases, and in particular to help his friend Jim escape from slavery rather than do what he thinks morality commands, return property to its owner. But his act here is actually morally right, and he does what is right because he is moved by the reasons that make it right. The objective account gives the right answer here about how to assess Huck and the subjective account gives the wrong answer. Or so one might claim.
Second example, a John LeCarre novel character, son of businessman who turns out to be gangster, working for dad, decides after deliberation to be loyal son and keep working for what he now knows is immoral enterprise. Through weakness of will he impetuously snitches, does the right thing, while trying to do the wrong thing. Arpaly suggests he is morally praiseworthy, because after all, he does the right thing for the right reasons.
The flat-footed but nonetheless correct response to Arpaly is that on the objective view of moral worth, people become morally praiseworthy and blameworthy for what lies beyond their power to control, and this is unacceptable. With this thought in mind let’s return to her examples.
Huck: Huck has attractive qualities, but we should not regard a person as morally praiseworthy for what he is (that smacks of a natural aristocracy view). Huck displays admirable virtue and practical reasoning success, should be praised for this, but this is not specifically moral praiseworthiness in the sense we are trying to delineate.. Regarding the latter, the case is hard to assess. Is Huck blameworthy for not engaging in critical reflection concerning the acceptability of the morality of his society? Probably not, or not much, given his upbringing and circumstances. Does Huck identify conventional morality as what “they” think, not normative for him? Probably not. If he did, then turning against convention would not be turning against morality as he conceives it to be. Subjective account assessment of the case seems OK.
LeCarre: First, one can try to do something without being aware one is trying. So maybe the character is consciously trying to do wrong and unconsciously trying to do right, and the overall verdict on the tryings, according to subjective view, should be mixed. Agent is not very blameworthy for resolving to do wrong, given that circumstances are so adverse for making choice of right. In agent’s circumstances, doing what is right (as he thinks) is hard and painful and likely to be costly, so his doing right is very praiseworthy, on subjective account. Subjective account says we should say same if he decided via weakness of will to do what is actually wrong but believed by him nonculpably to be morally right. Note that whether one actually does the right act, by one’s lights, is not per se important on subjective account. One can try very conscientiously, against big forces impelling one to do wrong, and be more praiseworthy, even though one does wrong, than someone who faces lesser adverse forces, puts forth less effort, but does right anyway (it’s easy).
However, everything said so far in defense of the conscientiousness account of moral praiseworthiness and blameworthiness dances around what many will regard as the decisive, knockdown objection. This criticism holds that anyone’s disposition to put forth wholehearted efforts toward knowing what is morally right and doing it is affected causally by all manner of empirical factors that vary from individual to individual. So the subjective account, even if otherwise plausible, flamboyantly fails to satisfy the no-moral-luck constraint.
Comment: This nonstarter claim is correct on a hard determinist view, which this essay sets aside. If hard determinism is correct—and it may well be—then the idea of moral worth that this essay seeks to uphold is just a lost cause. But even setting hard determinism to the side, the problem persists.
Just suppose we have libertarian free will. (The same point will hold on a soft determinist version of subjective account, which distinguishes among causes, and says that some—abnormal vs. normal, or those external to the agent vs. those internal—render one’s activity in relevant respects beyond one’s power to control and some causes don’t.) Even so, our free will is beset by causal forces, of many sorts, varying from individual to individual. These rule out achieving conscientiousness in some cases and in others render it more or less hard, painful, and costly to achieve conscientiousness. Call this the problem of variable causal entanglement. Whether any agent actually achieves conscientiousness and so becomes morally praiseworthy on subjective account varies with factors that lie beyond her power to control.
Back to the drawing board: Conscientiousness (version 3). Incorporates versions one and two and adds this: the determination of how morally praiseworthy/blameworthy a person is proceeds in two stages. At stage one, the extent to which the person exhibits conscientiousness version 2 is assessed. This yields a raw conscientiousness score. Each person’s raw conscientiousness score is then adjusted to reflect appropriately the ways in which the circumstances of each agent, beyond her power to control, affect her level of cosnscientiousness actually exhibited. This yields true conscientiousness score. Compare to ideal handicapping of golfers for tournament, so any golfer of any ability and whatever particular obstacles to a good game afflict her that day, has same opportunity to win. (Question: does version 3 solve the problem under review or just restate it in other words?)
The revised account of moral worth, conscientiousness version 3, is clearly cut loose from any connection to our actual responsibility practices. In many social settings, associations, and contexts, we hold each other responsible in various ways, attaching negative and positive sanctions to conduct that is desirable or undesirable from the standpoint of association goals. We hold each other’s feet to the fire, and clearly these practices broadly speaking can claim instrumental justification. They advance enterprise goals, and if enterprise is innocent, or just, the responsibility practices serve good goals. Notions of moral blaming and praising are ancillary to these responsibility practices. We can criticize existing responsibility practices—e.g. criminal justice system, at the margins, suggest reforms, but clearly many such practices are instrumentally justified.
The conscientiousness account of moral worth, shaped by the Narrow Control Principle, will clearly lead to very revisionary judgments as to who is morally praiseworthy and morally blameworthy. The serial killer and child molester who fights hard for years against persistent strong urges to do horrible deeds and succumbs to the urges just occasionally might well turn out to be less morally blameworthy than the ordinary middle-class person who faces less severe temptations and succumbs often, committing such evils as getting embarrassingly drunk and spoiling his brother’s wedding party. The serial killer might even turn out to be morally praiseworthy for his failed attempts to suppress his evil urges.
So what confers moral worth on the subjective account is pretty much never knowable in practice and is anyway irrelevant to any sensible responsibility practices. So what is the point?
It is not exactly true that owing to entanglement the subjective account is irrelevant to responsibility practices, since many of these do not rest content with instrumental justification but claim to be regulated by appropriate deservingness/blameworthiness/praiseworthiness norms. This is so also for the responsibility practice of morality as a human institution, on the ground. But the subjective account occupies the space in which these practical rival accounts of moral worth and deservingness claim to be situated. The subjective account stands in the way of our anxious tendency to elaborate weaker accounts that dispense with the Narrow Control Principle and to rely on these weaker accounts to provide less crude rationalizations of our responsibility practices than we can glean from instrumental rationales. But that way lies illusion—comforting illusion perhaps, but still illusion.
Interesting proposal!
I wonder whether the NCP, and its spirit, might be compatible with a hybrid view. Specifically, consider keeping the Conscientiousness3 account of blameworthiness, while adding an objective constraint onto the Conscientiousness3 account of praiseworthiness. The rough idea would be that to get praise you have to pass the Conscientiousness3 test but also be tracking the moral facts.
On this view the conscientious Nazi or Serial Killer is less blameworthy than the lax middle class person but is no more praiseworthy. I take it that this is a bit more intuitively appealing than the full bore Conscientiousness3 view but still has stiff illusion popping implications.
One might worry that an objective restriction on praise conflicts with the spirit of the NCP. But I am not sure. NCP reads: “we are assessable as moral praiseworthy or blameworthy (worthy or unworthy, deserving or undeserving) only to the extent that the feature of our activity that is the target of these assessments lies within our power to control.” As stated the NCP does not obviously apply to assessments of actions as meriting neither praise nor blame, and that is the bin into which the modified account would toss conscientious Nazi acts. Of course we could beef up the NCP to cut out this option, but I am unsure how to motivate it. I would be interested to hear more about what motivates the NCP on your view.
One thought here: It seems intuitively plausible that lack of control undercuts desert for blame because it is unfair to blame people for what is outside their control. But the idea that fairness requires equal access to praise does not seem intuitive to me.
Thanks to Brad Cokelet for this helpful comment.
As he suggests, I resist the hybrid proposal, because I want to explore whether contrary to Nagel we can abide by the control principle and conform to the norm no-moral luck when what is at issue is moral blameworthiness and moral praiseworthiness when that is the opposite of moral blameworthiness.
I agree with Cokelet when he says “the idea that fairness requires equal access to praise does not seem intuitive to me.” We praise people for all kinds of things, and people have vastly unequal access to praise. Being a slow runner, I won’t be eligible for praise for my running ability or achievements. Being morally unperceptive, I won’t be eligible for praise for making correct moral judgments about subtle matters. We accept unequal access to praise and dispraise with equanimity. If I lack the capacity to stand my ground in the presence of the fearful when there are good reasons all things considered to stand my ground, I am cowardly not courageous. Virtue attributions are larded with moral luck–that’s fine.
But along with Nagel and some other readers I detect a flirtation with no moral luck in Kant’s Groundwork, section 1. The idea is that there is a special kind of moral judgment–of moral praiseworthiness and moral blameworthiness or moral worth and moral unworth or moral deservingness and moral deservingness. For this special kind of judgment it does seem intuitive that there should be, must be, equal access to praise. I find this idea appealing, although somewhat weird. My tentative thought is that even though weird in its implications, it’s not incoherent, although if hard determinism is true then no one is ever praiseworthy or blameworthy in this special sense.–Dick Arneson
Thanks for this thought-provoking post. There are several things one might ask about, but I just want to put forward a methodological question. You say, quite provocatively, at the end, this:
“The subjective account stands in the way of our anxious tendency to elaborate weaker accounts that dispense with the Narrow Control Principle and to rely on these weaker accounts to provide less crude rationalizations of our responsibility practices than we can glean from instrumental rationales. But that way lies illusion—comforting illusion perhaps, but still illusion.”
I take it that this is a swipe at Strawson and Strawsonians, who would attempt to build an account of responsibility out of our practices of holding people responsible. Fair — or unfair — enough. But you also say earlier that your view is cut loose from any close connection to our moral practices (and these, I should point out, often go way beyond the merely instrumental). But then what is the source of theorizing about the moral praise- or blameworthiness in question if it is cut loose from our actual responsibility responses? How else are we to determine the “plausibility” of various proposals without consulting our sense of “response” in such cases, which is typically just some sort of actual praise or blame (along with a presumption that a prerequisite of praise- or blameworthiness has been met)? This seems to be what we’re expected to do in your early discussion, and so I’m wondering at the end of the day just how divorced from our actual practices the theorizing (and subsequent conclusions) really is.
David Shoemaker expresses a perfectly reasonable skepticism about my enterprise. I want to lock onto one particular intuition or judgment we seem to have: no moral luck. Or at least, no moral luck with respect to a certain narrow privileged class of judgments, which I don’t claim adequately to have characterized. My question is, is the no moral luck idea coherent and plausible, if we try to elaborate it, and if we help ourselves to generous assumptions regarding free will and responsibility (that is to say, just assume hard determinism is false). My thought is that reflecting in this way, we should hold onto our no moral luck intuition, and let the chips fall where that may so far as out common-sense practices or holding people responsible are concerned. Since I suspect the bulk of these practices can be explained and justified in purely instrumental terms, I’m not so worried about being revisionary with respect to common-sense responsibility judgments and attributions. But you could start by elaborating and developing common-sense responsibility judgments and junk the no moral luck doctrine if our considered responsibility judgments conflict with it. So far I don’t see myself as having provided any argument against locking onto our common-sense responsibility judgments in this way. As Shoemaker in effect notes, I’ve got a hunch here, but no arguments. –Dick Arneson
Thank you for an interesting post, Dick. Needless to say this compatibilist would much rather accept some (constitutive!) moral luck than accept that a fiercely conflicted sadistic killer can be less blameworthy than a wholehearted delinquent borrower of books, and I don’t see why one would regard this last conclusion to be obviously less costly than some moral luck. But never mind that. I’ll only respond to a few points for the moment.
Short comments first. 1) I don’t think your golf-handiccaping analogy works. A person from a certain background is unlikely not just to be conscientious but to ever make any sort of effort to become conscientious. How can this person get an A for for effort if there is has been no effort? “she tried to achieve it but there were difficulties” is different from “it never occurred to him to try it, but it rarely occurs to people in his country”. One can get rid of some moral luck by discounting the failure of effort, but how about a failure of something to occur? 2) I fail to see how the praise many of us give Huck Finn is any other kind than moral. It’s surely not aesthetic, nor does it seem prudential to me.
Interesting that you think of the moral question – or the proto-moral question – as concerning “what we owe others” considered “impartially.” I suspect that if this is what the moral question is, our views have to get closer to each other than they seem, because you have to acknowledge the existence of people who honestly think of themselves as seeking the morally right thing to and who in fact deliberated for hours on the question “what is the moral thing to do?” but who are not praiseworthy for their deliberation or the action it results in because their idea of “the morally right thing to do” has nothing to do with what they owe others and/or has nothing to do with an impartial considering thereof. They SEEM conscientious, but they don’t deserve the compliment.
Here are two examples.
1) The Victorian: Gwendolin, who lives in the 19th century, deliberates as to whether it would be morally wrong for her to have sex with her fiancee, whom she is going to marry in a few months. She gravely and earnestly uses the words “moral” and “wrong” in her head as she deliberate – she is not at all like a contemporary person considering a controversial act wondering what people would say – but the thought “what do I owe others?” or eve “what do I owe him?” never crosses her mind. She simply wonders if having sex with her fiancee would be slutty and impure, and assumes that slutty things are wrong, as that is what her parents told her. If the moral question is “what we owe others”, Gwendolin thinks she engages in moral deliberation, but she isn’t, and so nothing is particularly praiseworthy about the deliberation she engages in.
2)The Ayn Rand Nut: According to Ayn Rand, the moral hero does not care about the interests of others (I am not making this up). As far as she is concerned, asking “what do I owe others” before you act is no less than morally vicious behavior. A morally concerned person asks “what is in it for me?”. Howard’s parents were both Ayn Rand fanatics and moved in Randian circles, and so, (admittedly, like no one I have heard of: pure Randians rarely have kids) Howard grew up in the faith. Before every important decision, Howard asks himself “what is the moral thing to do?”. As he deliberates, the question of what he might owe others never crosses his mind, which is why I think there is nothing praiseworthy about either the deliberation or the selfish action that follows it.
I find Nazis more complicated. Both Tom Hurka (I think!) and I acknowledge that one can imagine a Nazi criminal who is blameless. That Nazi believes, honestly, non-culpably, not as some kind self-deceptive excuse for hatred-fueled behavior, that all Jews, including Jewish children, old people, etc, are effective super-villains who pose a catastrophic threat to the world. She only does what is necessary to neutralize this threat, and that is killing them. Chances are this person never existed (I try to explain why in Unprincipled Virtue). However, if she existed, she would not be blameworthy. It’s not prima facia implausible to suspect that she would deserves praise.
So far I have mentioned two possible Nazis: the theoretical construct who has honestly earned false factual beliefs and the common type who has dubiously motivated false factual beliefs. I imagine we agree on the first one. The second one is under-described – I suspect your view might depend on the role of voluntariness in the formation of self-deception, or the chance that some due diligence can prevent or expose such deception. What about a third Nazi all of whose factual beliefs are true – she thinks Jews are as human as she is, neither vermin nor demons; she thinks most of them pose no threat to the world, and so on – but who still murders and tortures them because she thinks, after deliberation,that she ought to do it? I have a lot of things to say about her, but for the moment I am trying to imagine the deliberation that leads her (a person of a sound mind, let’s imagine) to the conclusion that, say, killing the Jewish child before her is the right thing to do. Where is the sound deliberative route to this conclusion from true factual beliefs plus a consideration, from an impartial perspective, of what behavior she might owe others? Depending on how we interpret “impartial” and “owe”, there might not be such a route – in which case our Nazi’s deliberation wasn’t really a case of seeking the moral thing to do. “Conscientiousness” and “response to the reasons that make actions right” come closer together here in this one respect, I think, at least if the reasons that make actions right have to do with their being actions that we owe people in an impartial way.
By the way, I actually think, though I won’t defend this rough hunch here, that a deliberation that has nothing to do either with other people’s well being/desire satisfaction/something like that or with fairness/respect for persons/something like that is not really about a moral question. If we were to agree on that, it would not get rid of our disagreements about Huck Finn, but would make us give similar verdicts in cases in which a person’s idea of morality, at least when it comes to that particular action, is really… way out there.
I realized I said something unclear at the beginning of my post. I was trying to say, I understand how you can rid of some moral luck by saying that it’s effort that “counts”, not the results, but in the case of the person whose formative background makes it unlikely for her not only to be conscientious but to be, as it were, to be conscientious enough to make an effort to “achieve” conscientiousness, or make an effort to make an effort, it’s not clear to me that you can get rid of the person’s bad moral luck by saying that what matters is the effort it would have occurred to him to make if his background were different.