Ignorance and Moral Blameworthiness (by Featured Philosopher Julia Markovits)

Thanks so much for inviting me to contribute today!  Recently, I’ve been thinking about the relationship between different kinds of ignorance and moral blameworthiness.   I’m interested in how we should respond to wrong action that is (partly) attributable to moral ignorance, where that ignorance has a cultural dimension. 

If the criteria for the rightness and wrongness of actions is not, fundamentally, culture-relative (and I think it is not), can we make room for the possibility that the appropriateness of praising or blaming a certain right or wrong action is culture-relative?  For example, in virtue of what is 18th century abolitionism more admirable than opposition to slavery today?  In what sense (if any) is 18th century slaveholding less blameworthy that slaveholding is/would be today? 

I.

Gideon Rosen has argued that just as non-moral ignorance can exculpate, so too can moral ignorance (“Culpability and Ignorance,” Proceedings of the Aristotelian Society, 2003).  He labels this the “parity thesis” (p. 64).  In many cases, he notes, the fact that we’re ignorant of some relevant non-moral facts absolves us of blame for the actions we perform out of ignorance.  This is true not only when we couldn’t have known those facts, but also when we could have known them, but our ignorance is non-culpable, because it is not the result of negligence or recklessness or deliberate misconduct in the management of our opinion.  (Say I put salt in your tea, thinking it’s sugar (it’s in a bag marked “Sugar”).  I could easily have discovered that it’s salt, by sampling it first.  But it’s not negligent or reckless to trust that a white crystalline substance found in my kitchen in a bag labeled “Sugar” is, in fact, sugar.  So I’m not blameworthy for my ignorance, or for salting your tea.)

Rosen argues that the same line of thinking exculpates many wrong acts performed out of moral ignorance.  In the moral case, just as in the non-moral case, Rosen thinks, it would be intuitively unfair to blame someone for doing something if he blamelessly believes there’s no compelling moral reason not to do it.  And, Rosen argues, there can be cases of blameless moral ignorance.  Ancient Hittite slaveholders, he thinks, cannot be blamed for failing to recognize the wrongness of slavery; similarly, the sexism of fathers in the 50s, who encouraged their sons but not their daughters to go to college, is non-culpable.  (He assumes that the moral ignorance in each case is not derivative of non-moral ignorance.)  These people are blameless for their ignorance, according to Rosen:  “[t]hat [they have] failed to see through a pervasive and well-protected ideology need not be a sign of culpable recklessness or negligence on [their] part.  It might just be a sign of ordinariness.  And if that’s right we should conclude that [their] ignorance is not [their] fault.” (pp.  67-68)  (The examples suggest, and Rosen suspects, that blameless moral ignorance of this sort will in fact be widespread, and may justify a much more general skepticism about holding people responsible for wrong actions.)

Whatever we think about the culpability of the moral ignorance exhibited by the Hittite slaveholder or the 50s dad, it does seem, intuitively, like their wrong actions are less blameworthy than similar behavior would be today.  It takes a worse person to be pro-slavery or sexist in our culture than it took in their cultures.  Rosen’s parity thesis and his accompanying account of non-culpable ignorance seem to explain those intuitions, and so gains some further support from them directly.

The analogy to non-moral ignorance on which Rosen relies strikes me, however, as troubling.  That’s because non-culpable non-moral ignorance does more than just exculpate:  it changes what we’re obligated to do.  Not only am I not to blame for salting your tea; if you ask for sugar and I, despite my ignorance, refuse you the crystalline substance in my kitchen, I’ve acted wrongly.  Or imagine a doctor who refuses me the penicillin she justifiably believes I need:  she fails in her obligations to me even if, unbeknownst to us both, it turns out I’m allergic.  Moreover, while I may regret putting salt in your tea, and my doctor may regret giving me penicillin if I’m allergic, we shouldn’t feel guilt or remorse for our actions, once we discover the truth.  (I actually think the same thing about agent-regret, as opposed to simple regret, for reasons we can talk more about if you’re interested.) 

But surely we don’t want to reach similar conclusions in Rosen’s cases of moral ignorance!  Surely it’s not the case that the Hittite slaveholder was right (not just blameless) to support slavery, or that the 50s dad was obligated to give preference to his sons’ college ambitions.  And surely they should feel guilt or remorse if they were to learn the error of their ways.  I worry that Rosen’s account entails not just a (plausible) relativism of blame but also a more thoroughgoing (and decidedly less plausible) moral relativism:  Rosen’s argument that the moral ignorance of ancient slaveholders and 1950s sexists is non-culpable worryingly entails that ancient slaveholding and 1950s sexism were not wrong (in the subjective sense of “wrong” that’s relevant to establishing our obligations). 

II.

We might respond to this worry in two ways.

First, we can reject the parity thesis, and insist that non-culpable moral ignorance doesn’t exculpate after all.  The view that non-culpable ignorance exculpates might gain some support from a sort of Control Principle (of the sort Thomas Nagel defends in his paper on moral luck):  I can be blamed for outcomes only if they are, in the relevant sense, in my control (or if my lack of control is my fault).  And control of the outcome of my actions may require conscious awareness of how what I’m doing will effect the world.  Non-moral ignorance undermines this sort of control, but it’s not clear that moral ignorance does:  it doesn’t prevent me from knowing what I am doing, under the relevant description (e.g., putting salt in the tea, shutting down my daughter’s college ambitions, etc), even if it prevents me from knowing the moral status of what I’m doing.

Second, we can accept the parity thesis, but insist that moral ignorance is very rarely non-culpable in the relevant sense.  A better version of the parity thesis might say that epistemically justified ignorance or false belief (of any sort) exculpates.  It’s not clear to me that the false beliefs had by the Hittite Lord or 50s dad are epistemically justified, even if it would be in some sense inappropriate for us to blame them, because of how ordinary these false beliefs were (more on this below).  There are, I think, good reasons for thinking that testimonial evidence in these sorts of cases provides less justification for moral beliefs for non-moral beliefs.  (One good reason:  in the case of recognizing moral experts to rely on, much more than in the case of recognizing non-moral experts, it may take one to know one.)

I won’t say more about these strategies now.  I want, instead, to suggest an alternative account of the relativism of blame – one that accommodates my intuitions about the absolutism of obligation.  Rosen is right that ancient or even 18th century slave-holders seem less culpable that someone with similar attitudes to slavery today.  Furthermore, early abolitionists, or early sex-egalitarians, are more praiseworthy than people with similar attitudes today.  But our explanation of that fact should not entail that slavery or sexism were less wrong in previous times than they are today.

We have the resources to explain this once we recognize that the appropriateness of our praising or blaming an action depends not just on facts about the action, or about the agent’s motivations, but also on facts about us, and our motivations:  we need standing to praise and blame.  In particular, whether or not we can appropriately praise or blame an action depends in part on our judgments about how we would have acted had we been in the agent’s place:  we admire people who act in right ways we think we would not have done had we been in their place, and we lack standing to blame people who do wrong things we think we too would have done in their shoes.  (I’ve defended a view like this in more detail elsewhere, in “Saints, Heroes, Sages, and Villains,” Phil Studies 2012.)  We lack standing to blame the Hittite slaveholder or the 50s dad because we know would almost certainly have believed and done the same in their place.  But it doesn’t follow that their false moral beliefs were justified or exculpating in the same way that false non-moral beliefs can be.

Two follow-up questions I’d like to think more about:

(1) How should the cultural relativism of blame inform our attitudes to widespread or common-place wrongdoing today?

(2) Are there cultural divides so great that some forms of moral assessment across those divides are inappropriate, because we cannot meaningfully ask ourselves how we would have behaved had we been in the agent’s circumstances (because those circumstances are somehow incompatible with our being who we are)?  This would lead to something like Bernard Williams’ relativism of distance…

26 Replies to “Ignorance and Moral Blameworthiness (by Featured Philosopher Julia Markovits)

  1. Hi Julia,
    Thanks for your post, it touches on some issues that I think are really interesting. One quick thing. Michael Zimmerman has been working on these issues for quite some time, defending positions similar to Rosen’s. For reasons that aren’t entirely clear, his papers and books don’t seem to get as much attention as Rosen’s, so I thought I’d take the opportunity to plug his stuff.
    For my own part, I should just say at the outset that I don’t find Rosen’s arguments (or Zimmerman’s arguments, for that matter) all that compelling. (I discuss them here: https://www.academia.edu/2976226/The_Unity_of_Reason). Part of what’s missing from an approach to blame and blameworthiness if you take their arguments at face value, I think, is any interesting positive story about what the grounds could be for blaming someone on these views.
    One of the points you make is, I think quite right, and that is that we tend to think that we need standing to blame someone. It strikes many of us as inappropriate to blame others for the wrongs that we commit, for example. Perhaps a similar mechanism inclines us not to blame the ancient slaveholders or the sexist fathers because we think that we’d be like them if we were in their shoes. While this might explain some of our intuitions about when blame is appropriate and explain our disposition not to blame people for wrongdoing, I think the point points to the importance of recognising a gap between the conditions under which we’d correctly judge someone to be blameworthy and the conditions under which blame is appropriate. Perhaps we lack standing to blame the ancient slaveholders for their nastiness, but can’t we still judge correctly that they’re blameworthy for manifesting de re unresponsiveness?
    At any rate, the quick point is just this. I like your suggestion that we might lack the standing needed to blame the ancient slaveholders, but I worry about any non-skeptical approach towards blameworthiness that regards these characters as not being blameworthy as I don’t see how such accounts can retain much room for Arpaly’s notion of de re unresponsiveness and don’t see any good alternative positive proposal that makes sense of blameworthiness without it.

  2. I’m curious about the claim about standing—and, in particular, about whether you mean it to be a revision of our ordinary practices or an interpretation thereof. (I suppose this is to ask if you’re willing to take more of a position on this than you do in the paper.)
    For example, do you think that the one-must-think-that-one-wouldn’t-do-the-same-thing-in-their-position restriction on blame applies only when we are on relatively equal footing with someone? I guess it seems to me that we can blame people in certain positions over us even if we know that were we in that position, we’d do exactly the same thing. I’m relatively quick to anger; does it follow that were my typically level-headed boss to yell at me for something innocuous, I can’t blame him since I know that I would yell in his position? It seems intuitively like I can blame him here. Maybe it’s because I think I shouldn’t be in his position given my anger problems. Anyways, I’m curious about what you think about these sorts of cases. Maybe you think we shouldn’t do this, but we, in fact, do.
    (I admit that this intuition runs counter to our use of expressions like “I can’t blame them, I’d have done the very same thing”. But this seems like a rather silly thing to say about, say, POTUS when they do something seemingly blameworthy. Like a bizarre form of monday morning quarterbacking.)

  3. Thanks for the very interesting post. I’m inclined to agree with Clayton that the question of whether an individual is blameworthy and the question of whether any person may actually express blame to the individual are separate questions. An appeal to standing can certainly help to answer the latter question, but it doesn’t seem to be able to answer the former.
    Perhaps the way to respond to Rosen, then, is to simply deny the intuition that the Hittite slave owner or the 50’s dad are less blameworthy than someone performing those same actions today. It makes more sense to me to say that their actions are just as blameworthy but that their cultural circumstances mitigate the amount of blame that one can direct toward them. Additionally, appealing to standing in these cases is problematic for a further reason because presumably there are people in each case who have the requisite standing, i.e. the Hittite slaves or the 50’s daughter. Tying blameworthiness to standing only solves these cases from a third party point of view. It strikes me as much less plausible to say that the daughter in the 50’s has less standing to blame her father than a daughter in a similar situation in 2014 would, and a slave in the 1800’s would similarly have just as much standing to blame the slave owner as a slave in 2014 would.

  4. Thank you for your post!
    Like some of the previous commenters, I expected your post to go in a slightly different direction—I had expected a conclusion about the blameworthiness of an agent (as opposed the conditions under which it’s appropriate for one to blame an agent).
    I think you’re probably correct: we have more standing to blame the sexist dads of 2014 than we have to blame the sexist dads of 1950, and this may account for the fact that we’re less inclined to blame the sexist dads of 1950 even though what they did was, objectively, just as wrong. But it seems like there remains a puzzle about the relationship between their wrongdoing and their blameworthiness (as opposed to the appropriateness of our blaming them, given our standing).
    One idea I’m interested in (although I don’t know whether it will ultimately pan out) is the possibility that there are two kinds of obligations, (a) those one has according to the true moral theory and (b) those one has in light of one’s justified beliefs, the former of which determine the wrongness of one’s action while the latter of which determine one’s blameworthiness. This view has the benefit of being able to account for the possibility of blameless objective wrong-doing and blameworthy objective right-doing. (As a result, this view would let people like Rosen avoid the conclusion that the sexist dads of 1950 didn’t do anything wrong, even though the sexist dads of 1950 might be less blameworthy than the sexist dads of 2014.) The drawback of the view is that it divorces one’s fulfillment/violation of objective obligations from one’s blameworthiness, a result many people seem to dislike. I’d be interested in hearing what you think of this ‘double obligation’ view, given that it might also have the ability to account for the intuitions you’re interested in accounting for.

  5. Hi Julia! Thanks for the thought-provoking post. I have three brief comments.
    1. I would like to echo Nate and Clayton’s distinction between blameworthiness and the justification of blame in any individual case, and suggest a relationship between the two. We are only justified in blaming people when we have good evidence that they are blameworthy. But having that evidence is not sufficient by itself for our blame to be justified. Evidence of blameworthiness and standing to blame are both individually necessary and jointly sufficient for blame to be justified.
    2. I think it is easy to conflate intuitions about the role of standing, moral ignorance, and social norms when thinking about the appropriateness of blame across cultures and eras.
    a. I might not have standing (or at least not full standing) to blame the 50s dad who encourages his son, but not his daughter, to go to college, given that I might have been that way too if I were born in that era.
    b. The 50s dad might be less blameworthy than a similarly sexist father today, because the kind of moral education that is easily available today wasn’t as easy to get back then. The 2014 dad is more likely to have culpably failed to understand that sexism is wrong than the 50s dad would have been.
    (Interestingly, Rosen – as far as I remember him – focuses on negligence and recklessness as the paradigmatic types of mens rea in cases of culpable moral ignorance. But I think a lot of moral ignorance is the result of knowingly and intentionally avoiding critical reflection.)
    c. The 50s dad might be less blameworthy than the 2014 dad because supporting his daughter getting a higher education would have been somewhat more difficult than doing so today would be. There was less social and institutional support for women in education in the 50s, and if everyone around the 50s family assumed the daughter would get married young and be a homemaker, then acting contrary to those assumptions might have taken some effort and initiative. The 2014 dad likely wouldn’t have to take as much initiative.
    3. I think the question about whether the justification of blame is culture-relative may depend on what kind of moral psychological account of blame we start with.
    a. If we hold a purely evaluative view where blame is – roughly – just to judge that an action reflects badly on a person’s character, then concerns about standing, and possibly moral ignorance, seem to disappear. (Though I can see how someone might argue against this.) I don’t have to be a feminist to judge that the 50s dad is wrong for being sexist, and that this reflects badly on him. I think I could appropriately make that judgement even if I were just as sexist. Similarly, I might think that the 50s dad is unlucky to have grown up in a less morally enlightened era (if that’s true), but stick to my judgment that his failure to understand that his daughter’s education is just as important as his son’s shows something nasty about him. The fact that the 50s dad would have had less social and institutional support for encouraging his daughter to go to college might mitigate the extent to which his failure really shows something nasty about him, though – at least relative to the 2014 dad.
    b. If we have a thicker moral psychological account of blame, these considerations could change. Maybe blame involves feeling indignant/resentful (Strawson/Wallace), understanding our relationships to be modified (Scanlon), or adopting a disposition to be angry and the like (Sher). If so, then worries about standing and moral ignorance come back into play.

  6. Great post! I wanted to ask about the part where you wrote: “That’s because non-culpable non-moral ignorance does more than just exculpate: it changes what we’re obligated to do. Not only am I not to blame for salting your tea; if you ask for sugar and I, despite my ignorance, refuse you the crystalline substance in my kitchen, I’ve acted wrongly.”
    In the consequentialist literature, for example, the most common line to take is that whether an act is right or wrong depends on whether it in fact maximizes good consequences, not whether the agent had good reason to think it would. And I would have thought that a reasonable way to go. I would say of the person who refuses to give over what they believe to be sugar, when it is in fact salt, has by luck acted rightly but may well be blameworthy for doing so. Or some might want to distinguish between objective rightness and subjective rightness and allow that the person in your example has acted subjectively wrongly. But that would not, on such a view, be to act wrongly full stop. You perhaps hint that the key for you to deciding whether the act is wrong or not is whether the agent should feel guilt (rather than regret) upon learning the truth of the situation. But in situations where I do the best with my information but blamelessly kill the person I am trying to save, I assume the reaction to have in such a situation is to wish one had done otherwise and to feel that one failed to hit the target one was striving to hit. But it feels awkward to me to characterize this reaction on the part of the good but unlucky person as wishing they had acted wrongly rather than rightly. Anyway, just wanted to hear why you think we should tie right and wrong action to the agent’s epistemic situation rather than these other routes.

  7. Hi Julia,
    Nice post. I am sympathetic to the idea that facts about us can affect the appropriateness of our blame and praise. But I think that this idea puts pressure on the part of your strategy which takes the appropriateness of our blame to be indicative of blameworthiness.
    First, consider the Hittite slaveowner. Let’s suppose that we lack standing to blame. But surely his fellow Hittites wouldn’t have lacked the standing to blame. And if they could appropriately blame, then this suggests the Hittite slaveowner was in fact blameworthy. Otherwise, he is blameworthy to some by not to others. I find that a troubling result worth avoiding.
    Also, though you set aside denying the parity thesis, I thought I’d put out a quick rationale for taking that route (since it’s my preferred option). Suppose the daughter had managed to convince the 50’s dad that he has been behaving wrongly and operating with objectionable sexist attitudes. I think he’d be inclined not to let himself off the hook and think, “Wow, thank goodness those objectionable attitudes are so common! Otherwise I’d have been blameworthy for all that stuff.” Instead, when we discover that we’ve been doing something wrong, even if commonly accepted or culturally ingrained or what have you, we feel badly about it in the way that suggests we take ourselves to be blameworthy for having done it.

  8. Clayton –
    Thanks for your comment and especially for the references to Zimmerman’s work and your own paper, which will be very useful to me! I’m basically entirely in agreement with the main thrust of your comment: I agree that there’s an important distinction to be drawn between something like the appropriateness of our blaming someone and some blamer-independent notion of blameworthiness. I’m also extremely sympathetic to accounts of blameworthiness as de re unresponsiveness, like Arpaly’s – in fact I’ve defended a very similar view, inspired by hers, myself. So I guess I think there are (at least) two (probably more!) ‘dimensions’ of blameworthiness, only one of which is varies from appraiser to appraiser. (I explore these two dimensions in more detail in the “Saints, Heroes…” paper I mentioned in my post.) But I still think the appraiser-relative dimension can go a long way towards explaining some of our otherwise puzzling attitudes towards wrong-doing that is ‘ordinary’ or in some sense to-be-expected.

  9. Jack –
    Great point. I think you’re probably right about the boss example, and certainly the POTUS and QB cases! That is to say, I think you’re right that we do make these sorts of judgments, and I think you’re right that we are entitled to them. So can the standing account I endorse accommodate that? I think it can, though I think the details will need some working out. It all depends on what sorts of factors we build into the agent’s circumstances when we ask ourselves if we would have done that if…. I think the account should explain, for example, why we feel differently about a passerby rushing into a burning house (or, for that matter, not rushing in) than we do about a firefighter rushing in (or not). It seems to me we would blame a firefighter for not rushing in even if we would never rush in, and would never choose to become firefighters. But I think that’s what the account should say: we admire firefighters – we would not have had the courage to choose their careers. But we also blame them if they then don’t rush in to perform rescues (at least in ordinary cases), because we think “I wouldn’t have become a firefighter if I wasn’t prepared to take on the associated risks,” or something like that. (This is like your boss example, where (as you suggested) you might think to yourself that, while you would likely have lost your temper were you in his shoes, you wouldn’t have assumed his position given your anger problems.)

  10. Nate,
    Thanks very much for your comment! As you see from my response to Clayton, I’m inclined to agree with Clayton about that, too. But I don’t think the question of blameworthiness can be entirely separated from the question of whether any person has standing to blame. Blame is, after all, a reactive attitude – the question of whether someone acts blameworthily is, I think, a question about how we should respond to their actions. It’s just that I think that question needn’t always have the same answer for every judge.
    I very much like your suggestion that we should think about how the Hittite slaves or the 50’s daughter should react to the wrongdoings in question. It seems to me to provide a possible way of testing our intuitions about the suggestion I made in the post and the suggestion you make here – that the actions in question are blameworthy but blame is mitigated by the circumstances in which the agents acted. If the actions are blameworthy but blame is mitigated (the say it would be, e.g., if someone were acting under duress, or had a lot to lose, personally, by doing the right thing), then it should, I think, be mitigated from the perspective of the daughter or the slave as well. They, too, should not blame as fully as they would otherwise. On the other hand, if the difference in the appropriateness of blame is attributable not to mitigating circumstances but rather to facts about the blamer’s standing, then it might both be true that we lack standing to blame (or to blame as much) but the daughter or the slave (or, for that matter, those at the forefront of the fight for gender equality in the 50s) have full standing to blame. I’d want to think more about it, but I think my intuitions about the cases support the second explanation (the one I’ve floated above).

  11. Amelia and David –
    Thanks for your comments! Since you’re both pushing on the same idea, I thought I’d try to reply to your questions together. Amelia, you suggest that we distinguish between objective and subjective notions of obligation, where rightness/wrongness go along with the objective notion, and blameworthiness with the subjective. David, you’ve similarly suggested that distinguishing between objective and subjective notions of wrongness/rightness, with the subjective notion being the one that’s relevant to blameworthiness (although unlike Amelia, you’ve suggested that what’s subjectively wrong/blameworthy depends on what we believe, not what we’re justified in believing).
    I don’t want to deny that we can and often do use right/wrong (as well as ought) in both objective and subjective ways. I will say that I have a very hard time ‘hearing’ an objective notion of “obligation”: I don’t think there’s any natural sense in which the doctor in my example is obligated not to give me penicillin. And it seems very clear to me, at least, that she is failing in her obligations to me if she refuses me penicillin.
    But in any case, I don’t think that appealing to the objective/subjective distinction can help us with the cases that interested Rosen – the Hittite slaveholder and the 50s dad. The version of the distinction I take Amelia to be suggesting, which is the version I prefer, ties subjective wrongness & blameworthiness to what the agent is justified in believing. My own view is that neither the Hittite slaveholder nor the 50s dad are justified in their beliefs that their actions are unproblematic. I think the view their evidence supports is that their actions are wrong. (This might be so even if, as Rosen suggests, it would have taken a person with unusual moral insight to recognize this fact.) So what the Hittite slaveholder and the 50s dad do is both objectively and subjectively wrong, and so open to blame (unlike, e.g., the doctor who gives me penicillin, despite my hidden allergy). If we want to explain why it may be inappropriate for us to blame them nonetheless, we need to tell a different story (like my story about standing).
    Alternatively, we could adopt something more like the view David suggests – that subjective wrongness and blameworthiness track not the agent’s evidence but her beliefs. But I find that view quite implausible. For one thing, as I mentioned in an earlier comment, I think some people (the slaves, the daughter, the early abolitionist or fighter for gender equality) do have standing to blame (whereas no one, I think, should blame the doctor who gives me penicillin). More generally, I think people have lots of unjustified false beliefs that make them think it’s okay for them to do lots of terrible things. But these people are still blameworthy.
    David – you’re right that it seems awkward for my doctor to think to herself, I wish I’d acted wrongly rather than rightly, and denied Julia the penicillin it seemed she needed. But I think that awkwardness is explained by the fact that the doctor has no aim to do the wrong action, under that description. So it would be weird for her wish to have that content. It’s not weird, of course, for her to wish she’d done the action that was in fact (subjectively) wrong.

  12. Matt –
    I agree with much of what you say here. I don’t know whether the Hittite’s fellow Hittites would have standing to blame (they are also slaveholders, or at least, pro-slavery), but I think his slaves may well have had standing to blame. I certainly think 18th-century abolitionists had standing to blame slaveholders in the colonies. But I don’t find it as troubling as you do that there’s at least a kind of blameworthiness that’s appraiser-relative. After all, blame is a reactive attitude: it doesn’t strike me as that surprising that whether it is merited depends not just on its object but also on who’s attitude it is, so that it can sometimes be merited and sometimes not. (Having said that, I do also think there’s a notion of blameworthiness that doesn’t vary appraiser to appraiser. I just think blame is complicated.)
    I also agree with you that the 50s dad who recognizes the error of his ways should “feel badly” about it. I take that to be an important consideration weighing against Rosen’s take on the matter. As Rosen describes the situation, the dad should feel rather like my doctor might feel if she discovers, after giving me penicillin, that I was allergic all along (though there was no evidence for that). But my view is that the doctor should feel regret, perhaps, or frustration, but nothing like remorse or guilt, or even agent-regret – the special flavor of regret we feel in response to choices we make where we could justifiably have made a different choice. On my view, by contrast, all of these emotions would be appropriate on the part of the father, since he will have recognized that he acted wrongly, not just in some objective sense, but subjectively wrongly. He failed to fulfill his obligations towards his daughter, and her blame or resentment of him would (he should recognize) be entirely appropriate.

  13. Julia,
    I see how what I wrote suggested the interpretation you mentioned, but I would tie blameworthiness to the agent’s available evidence rather than her beliefs, as you suggested would be a better view. Not that I have a worked out view here.
    It seems to me that typically people’s goals in action are not to act well in light of available information but to achieve certain results that are specifiable independently of available information. This seems true of the morally good person as well. Her goal, typically, is not to act well in light of her available information, but to, for example, save the life. She and we would think she failed to achieve that goal even if she acts well in light of available information. A way of seeing this is to adopt a Steve Finlay example where you know that as is I would act well in light of available information but you also have crucial evidence about how to save the person I am trying to save. If you fail to share that evidence I will think you undermined my efforts and were uncooperative (not just that you did not act well in light of your evidence). This despite your action actually helping ensure that I act well in light of my available evidence. Suppose it is granted: people’s goals (including morally good people’s goals) tend not to be to act well in light of available evidence but to achieve certain results, e.g. save a life. If that is granted then it seems awkward to me to say in such cases that success in the agent’s action is not tied to success in that goal but rather to acting well in light of her evidence. And if the master moral assessment is tied to success in light of available evidence rather than success in serving the goal of our action, then good-hearted people will retrospectively (and good-hearted, more informed onlookers prospectively) often be wishing that we did not (or do not) do the morally correct thing.

  14. Hi, Julia
    Thanks for the very interesting post.
    With regard to the second follow-up question, it seems to me it might be meaningful (and coherent), at least if either one of the two is correct:
    1. Mind/body substance dualism is meaningful (and coherent) – even if false.
    2. We can meaningfully and coherently talk about what happened to us as newborns, and we can meaningfully and coherently consider a scenario in which we are born just like we were born, but when we are, say, 2 days old, we’re taken to a place where we are raised in the conditions in which, say, dads from the 50s were raised, or for that matter present-day members of Boko Haram are raised, etc. (perhaps in a parallel universe or something, to avoid some potential difficulties).
    I think we can consider such scenarios, at least when there is no difference is sex/gender; else, things get more complicated, so in a sense at least we would be the ones being raised in that different fashion (though I’m open to the possibility that there is more than one use of identity terms, and the matter of what we mean by “I” needs to be assessed contextually).
    In any case, while at least the scenarios described in 2. seem coherent to me (save for the contextual identity issue), they seem to depart from how we usually assess matters so much that it’s difficult to make any assessments about what we could have done. On that note, it seems to me that when we ask ourselves what we would have done under such-and-such circumstances, we’re assuming at least that we keep at least nearly all of our beliefs, attitudes, character, etc.
    Still, we may be able to use statistical information about what most humans do when raised in such social environments, by means of taking a look at how most people in those social contexts actually behaved or behave (assuming that those social contexts made all the difference, without the addition of something like malnourishment before birth; else, we might perhaps have to consider whether we can extend the identity scenario even before birth, but that’s more difficult), and in absence of any particular reasons to think otherwise, reckon we would probably do the same.
    A problem is that when we do that (and also, with regard to the first follow-up question), in my view the proposed procedure seems to give the intuitively wrong result very often, i.e., that we have no standing to blame many of them, and even that many others do not have such standing.
    For example, we probably wouldn’t have standing to blame the 1950s dad. I’m not sure about the Boko Haram member (what’s the relevant social context), but in any case, it seems problematic when it comes to the standing of others.
    For example, would her daughter have standing to blame him on this account? That’s a tougher one (due to gender and/or sex differences), but on the other hand, the daughter probably wouldn’t have standing to blame her mother, even more worrying slaves usually wouldn’t have standing to blame the slave owners, etc.
    In fact, in order to have standing, it seems they would have to reckon that they would probably be special in the relevant sense. I suppose someone who already is morally much better than nearly all others might suspect that she would be different in such situations too, but even that’s not enough to say it’s probably so: were the conditions that made her much better already present at birth, or two days later? Or did they come later? We don’t know for sure, and neither does the person making the assessment, it seems to me, but she doesn’t have enough knowledge to say it’s likely that she would have been better in the relevant sense.

  15. Julia — I enjoyed the post. However, I’m resistant to the idea that ignorance changes what our obligations are in the way you have in mind. Two things that I hope are helpful:
    First, I like to think of obligation, as among other things, that to which justification in the criminal law pertains, and blameworthiness as, among other things, that to which excuse in the criminal law pertains. Now, prompted by a desire to limn the structure of the criminal law, and by a desire to classify certain problematic cases (e.g. duress), theorists of the criminal law sometimes come up with “tests” of whether an exculpatory condition is a justification or an excuse. Here is one that I remember being associated with Continental legal systems — there can be no conflict between right and right. So, e.g., at most one side of a war can meet the requirements of jus in bello, at most one party to a fight can be justified in her aggression, etc. Here is another that I remember: you’re justified if any old bystander could permissibly help you, you’re excused otherwise. Both of these tests seem to militate in favor of classifying ignorance as affecting excuse rather than justification, and so by extension, blameworthiness but not obligation.
    Now, I can see some reasons why one might discount the probative value of these tests. But then I start to wonder what could militate in favor of your “obligation-changing” view over its negation, or vice versa. I’d need to hear more about what role obligation plays in your broader normative theory.
    Second, it’s not clear to me how much support your examples provide for your view. For yours is the view that ignorance regarding the morally-relevant features of one’s action can change the obligatoriness of that action. Another, far less controversial, view is that ignorance can change the obligatoriness of an action for the rather banal reason that ignorance involves belief, and sometimes belief is among the ordinary morally-relevant features of an act.
    An example may help: The airport is at location A. I mistakenly think i is at location B instead. You ask me where the airport is, and I say it’s at location B. Well, look, my ignorance makes my action permissible because were I not ignorant, my speech-act would have been a lie! But that hardly supports any wholesale position about the relevance of ignorance to obligation.
    I am inclined to think that the way (the imagined) you treated your guest works kind of like lying. You refused your guest something you thought he wanted (namely, sugar) and that it is customary for hosts to provide for their guests. That’s wrong because it expresses disrespect. It’s a diss, a slap in the face. Were you not ignorant, your refusal to put the substance in your guest’s tea would not have been an instance of refusing your guest the substance (sugar) you thought he wanted. (I also imagined it would have been accompanied by your saying, “That’s salt! You don’t want salted tea, do you?”). But that doesn’t show us anything about the general affects of ignorance on obligation. Specifically, it doesn’t show us anything about what you’re obligated or not to put in other people’s drinks, once the social stuff like refusal, etc. are taken out of the picture.

  16. Hi David,
    Thanks for the follow-up. I agree that people (especially good people!) aim to achieve certain results that are specifiable independently of available information, rather than to act well in light of their evidence. And it’s clear that when someone asks for advice about what they should do, at least in the ordinary, prospective case, they aren’t intending you to limit your advice to what would make sense given their evidence. That shows that there must be an objective sense of “should” and “ought” and “right.” (As I said before, it’s much less clear to me that there is an objective sense of “obligation.”) But there must also be a subjective sense of these terms: it’s perfectly fine to tell, e.g., my penicillin-prescribing doctor, who is ruing her decision, “you did exactly as you ought to have done” or “it would have been wrong of you not to proscribe that.” To me, the more morally interesting notion is the subjective one – certainly when it comes to matters of moral assessment of people and actions (even if, as you say, it’s not the one tied to success in achieving the goals of our actions). To some extent, then, the important thing is just to keep our terms straight. But the thing I want to emphasize here is that we can’t simply replace the notion of subjective wrongness with blameworthiness. Something can be subjectively wrong but not blameworthy, or subjectively right but not praiseworthy.
    My sense is that you think it’s at least an awkward consequence of the way of talking that I’ve endorsed, where the subjective notion is the one that’s central to moral assessment, that “good-hearted people will retrospectively (and good-hearted, more informed onlookers prospectively) often be wishing that we did not (or do not) do the morally correct thing.” This doesn’t strike me as so awkward, at least not if we read/hear “the morally correct thing” de re rather than de dicto. (The doctor, e.g., would be more likely to think, “I wish hadn’t prescribed penicillin”, rather than “I wish I’d acted wrongly!”) And even on the de dicto reading, it doesn’t seem so strange for me to think, e.g., of someone who I know will act rightly (in the subjective sense), when I also know that acting rightly will back-fire (for reasons he has no access to), “man, I wish that he’d have a moment of weakness, and do the wrong thing for once!”

  17. A couple of additions to the previous comment:
    On the “50s father” scenario, I said if we were to apply the procedure “what if we had been raised in those circumstances, etc.”. I still think that holds, but I think I should have distinguished (in the “we”) between men and women: for a man, the assessment would be “I shouldn’t blame the father, because I probably would have done the same if I had been raised as he was raised”, for a woman, the assessment would probably be “I shouldn’t blame the father, because I don’t know how I would have behaved if I had been raised as he was raised, including the male gender-specific features of his rearing as a man in that society”. I don’t think those are the right results.
    On the slave scenario (and assuming same sex and gender to simplify), if I try to look at the matter from my perspective, I can tell that if I had been raised as a Hittite slave owner, I probably would have been a Hittite slave owner. But similarly, if a slave of a Hittite slave owner had been raised as a Hittite slave owner, he probably would have been a Hittite slave owner.
    On the other hand, the slave probably may properly say “If I were in his position, I probably would release the slaves”, because in that “if I were in his position”, the slave is considering that he would keep his actual beliefs, attitudes, etc., only thrown into the unlikely situation that he’s gotten slaves somehow.
    But similarly, I may properly make the same assessment: “If I were in the slave owner’s position, I would probably release the slaves” (provided that they wouldn’t be executed if I do or something like that)
    Granted, that sort of thing will not happen to me, but it’s coherent that it might even if extremely improbable, and in any case, the slave also could properly tell that he wouldn’t wake up the next day a slave owner, even if the paths for him to somehow become one are less improbable than they are for me.
    Intuitively, I don’t see any significant difference in this context, so my impression is that we may blame them just as much as the slave may, assuming that we have the same amount of information about their actions (which we probably wouldn’t have, but that’s another matter; in any case, we may well have more info about a Hittite slave holder than about a present-day one), in the sense at least of saying that they are guilty to X degree, etc. (and aside from the issues of standing to forgive, spare or punish).

  18. Hi Chris,
    Thanks for your very helpful comments, which I somehow missed yesterday. In response:
    (1) Sounds good to me: someone’s having evidence of blameworthiness is necessary but not sufficient for it’s being appropriate for them to blame. I’d maybe say a bit more (somewhat tentatively): I think that saying some action is blameworthy need not yet be saying anything about any particular blamer. It may be that an action is blameworthy if there’s a hypothetical person who could appropriately blame the action. But talking about the justification of blaming is always a matter of talking about whether someone in particular is justified in blaming. So now facts about the blamer become relevant.
    (2) I guess I think that facts like those you describe in (b) and (c) – facts about how easy it was to see the moral truth, or how costly it was to perform the right act – can relevant to determining (a) – that is, to determining whether we would have done right in the agent’s circumstances, and so whether we have standing to blame. Are they independently relevant to determining the appropriateness of blame? Well, as I argue in my original post, I’m inclined to think that the 50s dad is, in some sense, culpable for failing to understand the wrongness of sexism, even though we may lack standing to blame him for it. So I think the fact that the truth was a bit harder to get at then (your factor 2(b)) affects the appropriateness of blame via the standing issue. I think facts like the costs associated with acting rightly (such as the factors you mention in 2(c)) can be independently relevant since they can be genuine normative reasons to act differently (and so go some way towards justifying the alternative act). But apart from that, I don’t think the personal difficulty of acting rightly directly affects the appropriateness of blame except to the extent that it affects our standing to blame. Some very praiseworthy actions come very easily to the relevant agent (because she’s so good-natured). Some are not at all costly (consider the soldier who throws himself on the grenade which would have killed him anyway, to smother the blast). Some not so very praiseworthy actions are nonetheless both costly and difficult for the agent (consider everything that goes into being a good parent!). We particularly admire the soldier not because his act was so costly, but because we likely couldn’t have done it. We blame the bad parent not because it would have been easy to do right in that case, but because we would have done better.
    (3) I like what you say here, and again, I agree. I have a thicker psychological account of blame of the sort you canvas in 3(b) in mind.

  19. Hi Angra Mainyu,
    These are important worries you’re raising, and I’m not sure I have a good answer to them – I’m still trying to work out what I think. I talked a bit in my “Saints, Heroes…” paper about the difficult matter of deciding what to include in the agent’s “circumstances” when we ask ourselves what we would have done had we been in their circumstances. It seems like we want to hold fixed the “external” circumstances in which the agent acts; but of course, we don’t want to build in everything about the agent – as your examples help bring out, that would make the question of what we’d do were we in his circumstances trivial (and the requirement that we ask how WE would have behaved in the agent’s circumstances empty). Nor can we simply say that that we hold all our “internal” qualities fixed – when asking whether I would, e.g., have performed some rescue if I were in the rescuer’s shoes, it surely matters if the agent is taller or stronger than I am. An initially appealing thought is that we hold fixed our own moral character, including things like our moral beliefs and attitudes, when imagining ourselves into the agent’s situation. This would mean that we certainly could blame the Hittite slaveholder or the 50s sexist, just as much as we could blame such behavior today. Your comments suggest you have some sympathy for this response. I’m reluctant to go that far, because I think the thought “would I have done better in their shoes?” helps explain why we think, e.g., that a bad upbringing mitigates blame to some extent (and also that some kinds of special training, including moral training, also lessen the amount praise merited by admirable actions). Of course, I also share (to some extent) Rosen’s intuitions about his examples: that we should blame the Hittite slaveholder and the 50s dad less than we should blame similar behavior today, and I think that’s because of what we think about how we would have acted in their circumstances. These intuitions suggest that we should, when fixing the agent’s “circumstances”, also hold fixed moral features of the agent that directly result (or are heavily shaped by) her circumstances. But here’s where we start to run up against my worry: certainly in the case of the Hittite, and maybe also in the case of the 50s dad, there’s a question about whether it’s even possible to really imagine ourselves in the agent’s circumstances in this extended sense without giving up on so much of ourselves that the question “how would *I* have acted?” becomes an empty one. I’m really not at all sure where to draw the line. But, going back to your examples, my instinct is that thinking about the appropriateness of blame along these lines will mean that blame is maybe less often appropriate than we might pre-theoretically have thought it was. (To soften that blow a bit: that isn’t, of course, to say that these agents’ actions aren’t wrong, or even blameworthy in some more abstract sense – a point that I hope has come out a bit more clearly in my back-and-forth with other commenters than it did in my initial post.)
    One other important point you raise that I completely agree follows from my account: very often, it will be very difficult for us to know what we would have don in the agent’s shoes, and in these cases, we won’t know whether our blame or resentment is appropriate or not. Also, we can easily make mistakes about this sort of thing. For example, most people presumably think, of themselves, that they would not have gone as far in compliance with authority in Stanley Milgram’s experiments as most of his experimental subjects did. But presumably, most of us would have gone that far. All this goes to show, in my opinion, that we should be very cautious about blaming people when we haven’t been in their situation ourselves.

  20. Hi Andrew,
    Thanks for these helpful comments. I guess I’m not sure what weight to give to the way the distinction between justification and excuse is drawn in some criminal law scholarship. I’m certainly not inclined to think, together with the Continental criminal law scholars you reference, that “there can be no conflict between right and right.” It seems intuitively clear to me that there can be such conflict (and I’ll note here that I don’t believe in moral dilemmas!). There can be such conflict, e.g., in cases involving special or agent-relative obligations: say, both our children are at risk, but only one can be rescued. I ought to push to save mine, and you ought to push to save yours. If there can be such conflict as a result of agent-relative obligations, I don’t see why there couldn’t also be such conflict as a result of differences in what two agents know. The agent-relative obligations case also highlights a problem with the second test you mention – the bystander test. Just because I’m justified in taking certain measures to save my child, doesn’t mean some bystander would be justified in helping me do that (if, e.g., more lives could be saved elsewhere). Again, if that’s possible because of special obligations, I don’t see why it shouldn’t be possible because of special access to evidence. (Of course, in the latter case, the bystander might instead simply share his evidence with me, bringing our obligations in line!)
    You asked why I want to hold onto the view of obligation according to which it is evidence-sensitive – what role it plays in my broader normative theory. The short answer is that my attraction to the view is largely independent of it’s role in my broader normative theory – I guess I just think that’s what the English word “obligation” means! It seems obvious to me that the doctor in my imagined example is obligated to prescribe penicillin for me, that she is justified in doing so, etc. It seems obvious to me that if she refused, she’d be violating her obligations to me. (Right and wrong are trickier, because they have objective and subjective senses both of which are salient.) But my linguistic intuitions here may be more pronounced than they should be! Having said that, I do also think there’s some theoretical pay-off to drawing the lines where I want to draw them, as came out a bit in my earlier exchange with David Sobel. I don’t think we can just drop talk of subjective wrongness and replace it with talk of blameworthiness (or excusability), because I think whether a subjectively wrong act is blameworthy is a much more complicated question! (But for the most part, as long as we are clear about what we mean by right/wrong/obligatory/justified, etc, it may be possible just to translate between these different uses.)
    As for your second point: I have to think more about it. I don’t think all the cases that interest me can simply be captured by noting the way in which what we believe (including what we believe falsely) can be morally relevant, and affect what our obligations are. That’s because, on my view, our obligations depend not (or not just) on what we believe, but rather on our evidence. The doctor in my penicillin example does fail to fulfill her obligations to me when she prescribes me penicillin if she didn’t check with me first about my drug allergies, even if it doesn’t occur to her that I might be allergic. But what you say about the salt case also sounds right to me. Perhaps your explanation isn’t incompatible with mine – what makes it wrong to refuse a guest the apparently harmless substance it seems he wants is that doing soe is disrespectful, etc. But as I say, I’ll need to think more about it. Thanks again!

  21. Hi Julia,
    Thanks for your reply, and you’re right that I think we may properly blame the father and the slave owner, though in my assessment, though I would say probably to a lesser extent than we may blame today’s counterparts (“probably”, because I think it depends on the case, but given only the info that he’s a Hittite slave owner/father in the 50s vs. present-day counterparts, the probabilistic assessment would be warranted), though I don’t have the impression that it’s because of how we would have acted.
    Rather, my impression is that in actual cases, there is a difference in degree of immorality of the behavior.
    For example, it may well be that the 50s father (if he deliberately puts more effort in helping out his son) believes that her daughter would not suffer nearly as much as his son if she doesn’t graduate, or even that she’s going to be better off if she doesn’t – he believes she just doesn’t know it yet, perhaps because in his mind she has a whim and women tend to be like that (this might also involve false moral beliefs about her character, but it doesn’t need to), etc.
    If he held those non-moral beliefs without epistemic culpability, his greater support of his son’s studies would be less immoral than it would otherwise be (or not at all, depending on the amount of suffering, etc.; as you point out, that kind of thing may change one’s obligations).
    Now, I think he is epistemically culpable for having those false non-moral beliefs, but less so than a present-day father with similar beliefs would be, which makes (all other things equal) his behavior less immoral.
    In those cases, though, moral ignorance is at least partially derivative of non-moral errors, so Rosen’s stipulation would rule them out.
    Also, I think there is at least one kind of false moral beliefs that lessen the immorality of some behavior, namely in some cases when the moral error is not about the morality of one’s moral obligations under such-and-such circumstances, but about the moral character of a person.
    For example: Alice’s mother, before she dies, tells Alice “Don’t trust Jones; he’s not a good man”; one may tweak the scenario so that Alice’s false moral belief is non-culpable, and would affect what obligations she has and their extent, making some behavior towards Jones less immoral than it would otherwise be, or not at all.
    In real cases, there may be culturally supported false moral beliefs about the moral character of all members of a group, and while holding those beliefs was epistemically culpable then, I think it was usually less so than it would be to hold them today, all other things equal (The false moral beliefs in question perhaps historically originated on false non-moral ones, but perhaps this isn’t always so, and in any case, the person holding those false moral beliefs needn’t have been told about the specific non-moral beliefs that gave rise to them, and maybe now holds false non-moral beliefs (e.g., about how to expect some people to behave) derivative from false moral ones, rather than the other way around).
    Still, it seems to me that Rosen’s point requires us to exclude such cases too, so leaving them aside, we get cases in which the only mistake or ignorance is about moral obligations, not derivative from non-moral errors, or from moral errors of the kind I mentioned above. In such cases, my views are more tentative.
    For example, let’s say that in the 1950s, two fathers are told that their behavior is immoral, that they’re being insensitive to their respective daughters, etc., that they’re hurting them.
    One of them reflects on the matter and changes his ways. The other one doubles down. Wouldn’t the latter’s behavior be even worse than it was before, when he was accepting the socially prevalent view without having ever confronted a suggestion that such behavior was immoral?
    It’s not entirely clear to me that this is a case in which immorality is lessened by non-derivative moral ignorance of the kind Rosen has in mind, though, since I might inadvertently be considering non-moral errors that usually happen in real cases, or maybe the difference here results from the (correct) non-moral belief that the two fathers have gained, namely that they have been informed that they’re hurting their respective daughters with their behavior, etc.
    I’ll have to give it more thought, and consider your hypothesis in more detail as well, though my intuition is that blameworthiness and immorality either both decrease or neither does.

  22. Thanks so much for your helpful replies and for being willing to do this PEA Soup thing! I see what you are saying in your replies to me and just need to think about it all much more. I find these topics interesting but have not managed to focus on them. But you are completely right that what caught my eye was not just there being a subjective (moral) ought, which I totally buy, but it being in some way the central moral ought. Thanks again!

  23. Thanks for your follow-up, Angra Mainyu. I totally agree with you – it’s very hard, especially in the case of the 50s dad, to imagine that the moral ignorance involved is in no way derivative from non-moral ignorance, and to some extent the non-moral ignorance might be less culpable than similar ignorance would be today. And that would change my assessment, not just of the appropriateness of blame, but also of how wrong the respective acts are. Rosen’s examples are supposed to exclude that possibility, I believe, but it’s hard to tell to what extent our intuitions about his cases are colored by the fact that the real-life versions of these cases are probably much messier in that respect.

  24. It might be helpful just to spend a moment considering what we mean by moral ignorance. With facts about the world, our success in science has convinced us that we do indeed know a lot of facts that people in the past did not know. We know that the evening star is the morning star. We know why people in the past thought they were different. We know (in outline) how it is that we came to see that they are the same thing. I assume most of us would be happy with calling the identity a fact about the world, and with labelling the state of those in the past “ignorance”.
    With morals, it’s different. In order to clearly identify a case of moral ignorance, we have to find a moral fact that we now know for certain to be true (and are there any of those?!); we have to identify a group of people either in the present or in the past who do not know that moral fact; and ideally, for extra clarity, we should have a decent historical story about why the ignorant group does not know the fact, and why we do know the fact.
    As it happens, I do believe in such things. I’m basically a Rawlsian, and I think that the implications of Rawls’ enlightenment-style equality of persons are pretty radical, and are still being worked out. (Cf. rich guys getting off jail sentences and “affluenza”.)
    In a policy debate about 2,000 years ago, Sima Qian said something with which the emperor disagreed. He was thrown in jail. Fifty years ago, Mao Zedong, who certainly knew the arguments for free speech, threw in jail those who commented critically on his presidency. In modern America, you can’t get thrown in jail for political commentary, however much the president dislikes it.
    It seems that Mao was certainly culpable; was Emperor Wudi? Certainly no-one expected him to behave other than as he did. Emperors who accepted the expression of dissenting opinions were lionised. It was possible, but there was not assumed to be any moral obligation to do so. It did not affect his strength or standing.
    Thinking about it now, I’m inclined to go with Rosen. The social system in which Wudi and Sima were embedded was so alien – so immoral, in our terms – that I find it hard to attach moral blame to a single act. In my mind, I’m struggling to disentangle the act from the system. Not sure if this is just a failure of my imagination/analysis, though!

Comments are closed.