I suspect I’ve just wandered into a longstanding dispute, but I’m curious what people think about the nature of wrong acts, and in particular about whether an agent incapable of perceiving the relevant wrongness/rightness reasons can nevertheless perform wrong acts. Here’s a representative quote from someone who does think this, Gideon Rosen, in his 2004 paper “Skepticism About Moral Responsibility”:
When you pull my chair out from under me just for laughs, there is no doubt that the act is wrong. But if you’re only five years old, or if you mistakenly believed that I wouldn’t mind, then even though the act is wrong, it may be a mistake for me to blame you for it.
I agree with the blame claim, of course, but I don’t know what to make of the claim that the act is wrong. Would we say of a bear who did the same thing that the act was wrong? Interested to hear considerations for and against this claim.
I don’t like Gideon moving to the less responsible agent to make his point. Suppose it is the completely evil but otherwise rational person who does it because it will hurt you and fully understands how long the process of recovery will take and how painful it will all be and does it for that reason. He understands that some take that to be a reason not to do it but thinks to himself that so long as this advances his aims he has a reason to do it. I see such an act as quite wrong. I don’t know exactly what I am signing on for in saying not only that he does not see such considerations as reason-providing against such action but also that he is “incapable” of doing so. But so long as incapable keeps a procedurals gloss of “would not after procedurally rational deliberation” or some such, then I think I am ok with that.
I take ‘wrong’ to be ambiguous. When I wonder to myself whether it would be wrong for me to x, I probably have the evidence-relative sense of ‘wrong’ in mind. But if, by contrast, I ask God whether it would be wrong for me to x, I probably have the fact-relative sense of ‘wrong’ in mind. So if you pull my chair out from me while being unable to appreciate any the reasons against your doing so, then you’ll lack evidence for its being wrong for you to do so. Thus, it would not be wrong in the evidence-relative sense for you to do so. But if I would mind your doing so, then it would be wrong in the fact-relative sense for you to do so. And blame tracks evidence-relative wrongness, not fact-relative wrongness. Thus, if an agent is incapable of perceiving the relevant wrongness reasons for refraining from x, then her x-ing won’t be wrong in the evidence-relative sense. But it could still be wrong in the fact-relative sense.
Doug: So bears, babies, and mosquitoes could perform wrong actions (in the fact-relative sense)?
Sobel: So psychopaths can perform wrong actions, as many think. The follow-up, then, is what it means to “have a reason.” Seemingly, the bear who swipes the chair out (so as to remove a non-tasty obstacle to eating me?) has a kind of reason too. So the bear acts wrongly?
“I take ‘wrong’ to be ambiguous. When I wonder to myself whether it would be wrong for me to x, I probably have the evidence-relative sense of ‘wrong’ in mind. But if, by contrast, I ask God whether it would be wrong for me to x, I probably have the fact-relative sense of ‘wrong’ in mind. So if you pull my chair out from me while being unable to appreciate any the reasons against your doing so, then you’ll lack evidence for its being wrong for you to do so. Thus, it would not be wrong in the evidence-relative sense for you to do so.”
Hi Doug,
That’s very interesting, but I thought that the ‘wrong’ of deliberation -was- the fact-relative sense of wrong. (This explains my anxiety that I might have overlooked something important or might be ignorant of some relevant fact.) Setting that aside, though, I wonder about this reading of the evidence-relative sense of ‘wrong’. If I’m aware of various non-moral facts, they should be part of my evidence. If those facts constitute reasons not to do something, couldn’t they make it ‘wrong’ (in the evidence-relative sense) for me to do something even if I don’t see their force? I always thought that the idea was that the facts that determined whether an act is wrong (in the evidence-relative) sense supervened upon a subject’s evidence but wouldn’t necessarily put that subject in any position to see that the relevant acts were wrong. (Compare this to the situation where I’m flooded with evidence that bears on some hypothesis but cannot work out the relevant connections. It’s wrong (in the evidence-relative sense) to draw fallacious inferences from the evidence even if I couldn’t see these inferences for what they were.)
I myself am happy to say that something would be wrong for X to do even tho X has no reason to not do it.
Dave: Nothing I said implies that bears, babies, and mosquitoes could perform wrong actions (in the fact-relative sense). It may be that only rational agents can perform wrong acts (in the fact-relative sense) and these beings are not rational agents.
Clayton: I agree that I can be worried about doing what’s wrong in the fact-relative sense, but I don’t think that the fact-relative sense of ‘wrong’ is the one that’s most central to deliberation. It seems to me that the sense of wrong that’s most central to deliberation is the one where, on pain of irrationality, I intend to refrain from doing what I believe to be wrong in that sense. And, as Jackson-style cases illustrate, that’s not the fact-relative sense of wrong. And, on your other point, I took Dave to be talking about cases where the agent cannot appreciate the reason-giving force of certain facts. That’s distinct from just not appreciating (or seeing) their reason-giving force. And I’m not sure that I agree that it’s wrong (in the evidence-relative sense) to draw fallacious inferences if I can appreciate some reason for making that inference but cannot appreciate any reasons not to. Perhaps, for some fact F to count as evidence for S to believe that P, it must be not only that F counts in favor of P but also that S is the sort of agent that’s capable of responding to her awareness of F by believing that P.
That’s part of what I’m asking, Doug, namely, whether rational agency is required to perform wrong acts, and if so, why, if that rational agent lacks the capacity to recognize/respond to the non-wrong reasons?
Sobel: What work does your hard-core view here do for you? I’m assuming you divorce wrongness from blameworthiness. So what bonus does it have to be SO divorced? Deliberative advantage? Something else?
Dave: Okay, but then I don’t think that the quote that you give demonstrates that Rosen thinks that an agent who lacks the capacity to recognize/respond to the non-wrong reasons can do wrong. After all, the typical five-year-old has the capacity to recognize/respond to the reasons not to pull the chair out from me and ditto for the person who mistakenly believes that that I wouldn’t mind having the chair pulled out from me. And is the question whether the agent has the capacity to recognize/respond to the types of reasons at issue or to the particular reasons at issue?
In any case, I think that an agent must have the capacity to recognize and respond to the relevant sorts of reasons (i.e., the relevant sorts of moral reasons) in order to do wrong in acting in a way that fails to respond appropriately to those reasons. So a bear doesn’t do wrong in tearing apart your tent to get the food that you left inside if the bear doesn’t have the capacity to recognize and respond to the relevant moral reasons (say, those having to do with autonomy). I would be interested to hear it if you have any textual evidence that Rosen would deny this. The quote that you give seems to me insufficient evidence for this.
I think it is just that I think the best theory of reasons has it that agents need not necessarily have strong reasons to do what the best theory of morality says is morally required. So no further work being done than that. But I may well have misinterpreted your question. But am I right that folks who like the framework you offered about a tight connection between being capable of seeing a reason against something and the capacity to do wrong assume that agents must have a reason to avoid doing wrong?
From Amelie Rorty:
I don’t think there is an all-context reply to your PEA Soup Question about ‘wrong acts.’ If we’re trying to teach a 5-year old child to stop pulling out chairs from under people, it may be important to label the act as wrong, and give a long palaver about why it is wrong. Or: If pulling out a chair from under someone would save them from some other mishap, the act wouldn’t be wrong. You’ll of course want to distinguish classifying token and type attributions of wrongful acts.
Doug: The Rosen quote was just to jumpstart things and get feedback from those smarter than me about this issue. (I’m still waiting, by the way. Boom!)
But at any rate, from what I can tell, the implication of your view is actually that psychopaths don’t perform wrong acts, is that correct? (Assuming that they lack the capacity for recognizing moral reasons, and even though they can act on a variety of other sorts of reasons.)
Sobel: It’s the nature of “have reason” that I’m not yet clear enough on to answer your question. I think their thought is that the relevant fact must be “perceivable” in a reasonish way (i.e., recognizable as at least a putative reason).
Amelie: This sounds right to me. It was calling the act “wrong” with respect to the five-year-old that gave me great pause, and so got me thinking about the function that calling an act wrong might play in various theories (or in life, for that matter). This came out in one of my questions to Sobel above. So I’m wondering if calling such acts wrong, where the agent really does lack access to the relevant reasons of rightness, might actually still be connected to thoughts about blame in virtue of the labeling’s serving a kind of proleptic function, i.e., it is targeted to agents in a way that could help bring them into the community of morally responsible agents, even thought they’re not in it yet. But otherwise, I don’t see the point of it.
You might be interested in philosophic reflection on the case of Adam and Eve. The story implies that they did something wrong when they ate from the tree of good and evil, but that they were innocent before they took some bites; so it naturally raises questions both about the fitness of God’s anger and punishment and about how their acts could be wrong (or rebellious or what have you) before they knew the difference between good and evil.
There are two very interesting papers on this by good philosophers: “Lost Innocence” by Herbert Morris and “The Judgment of Adam” by Wayne Martin.
Ahh, yes, that’s very helpful, Brad, thanks! Although from the description, my anger sounds more fitting than God’s.
Another thought on bears.
Mark Rowlands argues (roughly) that animals can be moral subjects but not moral agents because they can act out of empathy, but they lack reflective/rational capacities to take justifications into account. This is an interesting view, and it raises the possibility that animals can act wrongly.
It seems absurd to claim that an ant can act wrongly because they lack reasons responsiveness when it comes to moral reasons. But if Rowlands it right, we might say that a normally empathic dog acts wrongly in some case and this would be grounded on the assumption that the dog is normally responsive to some relevant stretch of moral reasons (roughly ones grounded in welfare considerations and not respect/dignity considerations).
So when a psychopath performs an act that’s harmful to others as a result of being unable to recognize or respond to the reasons for refraining from harming others, I think it would be mistaken to say that she ought to have refrained from performing that act. Now whether we say that the psychopath’s act was wrong just depends on whether we are using ‘wrong’ to mean ‘that which the agent ought to have refrained from performing’ or ‘that which a rational agent ought to refrain from performing’.
I find appealing, David, your thought about the proleptic function of calling “wrong” the ‘morally blind’ agent’s actions, although I suspect that ‘ease of reference’ (to actions-types that generally figure in attributions of blame/praise) will also be part of the explanation (though the latter presumably will enable(…) the proleptic function).
Considering “right” actions momentarily, I suspect the reason that the psychopath, but not the porpoise, can do (or be said to have done) *the right thing for the wrong reason*– even if both are ‘morally blind’–is that the psychopath, but not the purpoise, can e.g., “intend to save the drowning child”. That is, the psychopath can act on the basis of intentions that would normally be present among the attitudes of the morally-sensitive person (i.e., the intention to save), but the psychopath’s ‘intention in’ (or motive for) saving the child is not, we can suppose, morally praiseworthy.
Again, that’s very helpful, Brad, thanks! I’m a big fan of empathy doing heavy lifting in moral theorizing, so I’ll check it out.
Yes, Doug, I’m thinking that there’s indeed an ambiguity along these lines, perhaps.
Dan: Thinking about doing the right thing for the wrong reason is helpful, thanks.
FWIW, I think that Jackson medicine example (one drug likely cures but could kill, the other partly cures but doesn’t kill), mine shaft examples and three envelope examples show that the deliberative ought is not about what one would want someone to do given all the facts (contra Clayton if I read him correctly). In each of those cases you have to take into account the costs of being wrong in figuring out what to do. I see no reason to treat right and wrong differently than ought in this respect. (This doesn’t deny that one can say that one should have done something else given the actual facts or that you did the wrong thing in retrospect. But those aren’t deliberative judgements.)
Errol Lord’s Sander’s prize paper had some good stuff on this as I recall.
It indeed seems to me that the five-year-old would act wrongly in knocking you down, but the bear wouldn’t. But what about (a) a highly intelligent dog (or dolphin, or elephant), or (b) an adult human who was permanently insane in some really radical way (i.e. maybe they thought they were the victim of a Cartesian demon), and as such incapable of even marginal participation in normal interpersonal relationships?
I ask because the things that initially occur to me as potentially explaining why the child acts wrongly, but the bear doesn’t, apply to the dog but not the severely developmentally disabled adult. In particular, I’m inclined to think that in order to act wrongly in X-ing, you need to be such that it would be appropriate for others to hold you to the expectation–in a very loosely Strawsonian sense–that you not X. So long as we take the “very loosely” part seriously, it’s clear enough that the child meets this condition, along (though naturally I don’t need to tell you this) with more-or-less realistic versions of psychopaths or people with dementia. And so too does the dog, I’d think–but the radically insane person doesn’t, even as a borderline case.
And yet I’m inclined to say the radically insane person can still (blamelessly) act wrongly. I don’t know what to say about the dog. I’d be curious to hear what others think.
oops: “severely developmentally disabled” in the first sentence of the second paragraph should read “radically insane”–the former was an artifact of an earlier version of a comment.
Ben, I like the connection between wrongness and blameworthiness, or at least some kind of interpersonal “demanding.” My thoughts about “proleptic blaming” earlier can capture some cases one might think aren’t captured by the claim initially. But then I’m unsure why you think the radically insane person can still blamelessly act wrongly. My intuitions are pretty stable on that one that such a person couldn’t act wrongly. But those intuitions could easily be theoretically tainted. Do you have any more to say on that point?
My thought is that an action’s being wrong has to do with it expressing/being evidence for a bad will, or not-good-enough will, toward a person; wrongness violates relationships.
Bears and mosquitoes have no wills, hence no relationships with humans, of the right sort, so they can’t do anything wrong.
Five-year-olds have wills or some approximation of it, and are capable of the right kind of relationship but often have to be trained into it, or trained into what sorts of behavior would be appropriate for the relationship. So they can do something wrong, though our response to their wrongdoing is going to have a heavy element of moral education.
People who swipe chairs out from under you with the motive of doing you a favor are demonstrating good will, so they are not doing anything wrong. They may be doing something unwise.
I think of psychopaths as people who have wills but are more-or-less psychologically determined in ways that prevent them from using those wills to form morally appropriate relationships. They do wrong, often. Whether it makes sense to get too upset about this, or to take more of a quarantine approach, is a separate question.
Like Ben I think that the best way to understand wrongness is in terms of the appropriateness of the “punitive” moral reactive attitudes, e.g., compunction, guilt, blame. I think what Strawson suggests about the case of the five-year old child is that the action may be “sort of” wrong, inasmuch as these attitudes would be appropriate to some degree, but not the same degree as they would be toward an adult: “Thus parents and others concerned with the care and upbringing of young children cannot have to their charges either kind of attitude in a pure or unqualified form. They are dealing with creatures who are potentially and increasingly capable both of holding, and being objects of, the full range of human and moral attitudes, but are not yet truly capable of either. The treatment of such creatures must therefore represent a kind of compromise, constantly shifting in one direction, between objectivity of attitude and developed human attitudes. Rehearsals insensibly modulate towards true performances.”
The psychopath is a tricky one for me. Like Mill, as I read him, I’m inclined that think that it’s really the “self-reactive” attitudes of guilt and compunction that have the closest conceptual connection to wrongness. If the psychopath really is incapable of these attitudes, then maybe her actions cannot be morally evaluated.
“FWIW, I think that Jackson medicine example (one drug likely cures but could kill, the other partly cures but doesn’t kill), mine shaft examples and three envelope examples show that the deliberative ought is not about what one would want someone to do given all the facts (contra Clayton if I read him correctly).”
Hi Mark, I think that’s a popular line but it would be nice if the people who pushed this line said more about things like reparative duty and the difficult cases like compensation for injuries caused by imperfect self-defense, wrongful conviction, etc.. (To his credit, Michael Zimmerman tries to tackle this. In some ways, so does Barbara Hermann. I just didn’t find their treatment persuasive.) One thing I like about the line that Zimmerman takes is that there’s no need to distinguish the senses of ‘ought’ to handle the cases like the mineshaft case. The case shows that there’s no sense of ‘ought’ (or ‘right’, ‘wrong’) that people like Moore got right. I guess I don’t see the attraction of the view that says that cases like this show that there are two senses of ‘ought’, one that turns on all the facts, one that turns on just the evidence that the subject can use (or however we fill in the details) and deliberation is just concerned with the one that doesn’t depend upon the facts. It doesn’t create the burden of explaining which ‘ought’ is more central to deliberation, how the various notions of ‘ought’ relate to one another, etc. (As the different notions proliferate, you start to wonder what theoretical gain could be had.)
Dale: As I recall, Strawson doesn’t speak at all about “wrongness.” (I don’t have a searchable PDF of the article right now, so I can’t check for sure.) Indeed, the connection between wrongness and blameworthiness (where that just refers to “angersome”) may not be that tight insofar as the angersome could include a much wider domain than the merely wrong. For example, suppose I’m picnicking alone and drop my lone plastic fork in the mud. You (a stranger) are at the next table, and you’ve got a whole bag of plastic forks. I ask you for one, and you say no. Seems to me it’s apt to be angry at you even though you’ve done nothing wrong.
Clayton — I’ve taken the view that when you’re certain about the relevant propositions, the objective ought is the deliberative one, but when you’re uncertain about the relevant propositions, the epistemic-probability-relative ought is the deliberative. (Hence my exasperation with this move by opponents of the “there are multiple oughts” approach away from talking about ought simpliciter to talking about THE deliberative ought.) And I think we can characterize the non-objective oughts in terms of objective reasons and the notion of a try. Ellie Mason has also developed a trying-type view. What’s wrong with that?