This post is a question for those who know more about the debates about moral responsibility. The question is: why is the wrong kind of reasons problem discussed so extensively in the buck-passing/value theory literature but relatively little in the moral responsibility literature? The only discussions I have been able to find are in a couple of Stephen Darwall’s papers where he discusses what we can learn from Strawson. Maybe the issue has been discussed more extensively in which case I would be very thankful for advice…
Fitting attitude accounts of a property have roughly the following structure: ‘An object has a property F iff it is appropriate to have a reaction R (usually a kind of an attitude) towards F’. So, according to fitting attitude account of value, a state of affairs, for example, is good iff it is appropriate to value the state of affairs or to desire for it to obtain or something of the kind.
Pretty much all discussions of moral responsibility begin from the idea that an agent is responsible for an action iff a certain reaction to the agent doing the action is appropriate. Different theorists then disagree about what the relevant reaction is – whether it is the reaction of holding responsible, the reaction of making certain kind of moral evaluations by using moral terms, the reaction of having reactive attitudes (blame, resentment, guilt, praise and the like), the reaction of asking for reasons and explanations, and so on. So, pretty much all accounts of responsibility thus begin from assuming something like the fitting attitudes structure.
The standard problem for the fitting attitude accounts generally is the wrong kind of reasons problem. In the value case, this is often illustrated with the evil demon cases. The saucer of mud is not any more valuable even if it is appropriate for us to value it because the evil demon would punish us otherwise. So, the corresponding worry is that the fitting reaction accounts of moral responsibility too get the extension of morally responsible beings wrong because it is sometimes appropriate to have the relevant reaction towards the agent because of an evil demon’s threat even if the agent isn’t intuitively responsible.
According to Darwall, this is one of the issues Peter Strawson raised in his ‘Freedom and Resentment’. The claim was that, even if there were sometimes utilitarian reasons for having the reactions that are relevant for moral responsibility, the sense in which these reasons make the relevant reactions appropriate is not relevant for moral responsibility. In these situations, the appropriateness of the relevant reactions merely on the grounds of social desirability does not make anyone responsible. So, it seems to be that the people working on moral responsibility have been aware of the wrong kind of reasons problem since Strawson.
What I am less clear about is whether the wrong kind of reasons problem has been seen as a genuine, difficult problem. It seems like, in the responsibility literature, people have assumed that the problem can be easily solved. So, sometimes I’ve seen people add that it’s not only that relevant reactions must be appropriate but also the person must deserve them on the basis of her qualities or that the reasons that make the relevant reaction appropriate must be non-instrumental rather than instrumental reasons. These moves are, of course, familiar from the value context in which they are controversial (I’m thinking of the debates about the object-given reasons in the former case and Philip Stratton-Lake’s solution in the latter).
I can also think of a couple of other moves made in the responsibility literature. Darwall himself suggests that different kinds of attitudes (including the ones related to responsibility) come with their own, distinct norms that are relevant for having those attitudes and so the right kind of reasons are built into these norms whereas the other kind of reasons are external to the norms (this reminds me of Mark Schroeder’s solution to the wrong kind of reasons problem). He also suggests that the right kind of reasons are the kind of reasons that are relevant for deliberation on the basis of which one can come to have the attitudes in question (this reminds me of the wrong kind of reasons scepticism solution). Finally, I also get a feeling that, according to some, when we do enough to describe substantially what makes someone an apt target of the reactive attitudes, the wrong kind of reasons problem falls away. If we know enough of the right kind of reasons and what they are, perhaps we don’t need to worry about the wrong kind of reasons (interestingly, I’m not sure this line of response has been used in the value debates).
Thus, it seems to me that the people working on moral responsibility are aware of the vulnerability to the wrong kind of reasons problem and some solutions have been relied on in passing. However, it seems to me that there is a striking contrast to the value debates. In the latter debate, we get a lot of work discussing the problem explicitly and also in a more general form: people trying to solve the problem in different ways and others trying to explain why the solutions do not work. This seems to suggest that in this context people have thought that this is a pressing and important problem whereas I don’t get the same impression when it comes to the moral responsibility debates. In the latter debates, I haven’t, for example, seen anyone suggest that we should completely abandon fitting attitude/reaction accounts of responsibility because the wrong kind of reasons problem cannot be solved.
So, I would be very interesting to hear from people who work in that area. Is my sense of there being not that much debates about the wrong kind of reasons problem in the moral responsibility literature right? Is there more literature on this and I’ve just missed it? Why is there such a drastic difference? Are the value theorists overreacting to the problem or the moral responsibility theorists underreacting? Why?
Of course–but this isn’t the kind of helpful response you are looking for–many will object off the bat to ‘An object has a property F iff it is appropriate to have a reaction R (usually a kind of an attitude) towards F’. An electron has the property of having negative charge iff it is appropriate to have a certain reaction towards having negative charge, or should it be, a certain attitude to an electron’s having negative charge? This is hard to believe. An electron’s having negative charge is a fact about electrons, not about the appropriateness of our reactions to its having negative charge. I feel the same when the property is being morally responsible but of course many differ.
Jussi: You are generally right that people in the MR literature don’t talk very much about the WKR problem. There are various reasons for this. Until recently, it seems that some people were using the term “apt” to modify “reactive attitudes” in an ambiguous fashion, between right and wrong kinds of reasons. But maybe since Darwall’s book, people have been much more careful about it, more often using “fit” as the relevant modifier.
So why not worry about the WKR problem? There are likely a couple of reasons. First, while value talk may more easily be separated out from practices of valuing, talk of MR can’t so easily be separated from our practices of holding responsible. Indeed, this very close tie is what has generated the ongoing dispute, I think, between those who think being MR is metaphysically prior to holding MR, and those who believe the opposite (see McKenna’s 2012 book for a good discussion). So given that holding responsible is such a deeply human response, and is bound up with our actual practices, it’s natural to think that the kinds of case that generate the WKR problem in the value theory literature — the demon and the saucer of mud — are just not that relevant to the issue, as it’s outside the realm of ordinary human interpersonal exchanges. You need more realistic examples.
Now as it so happens, I think this precise issue HAS come up in a recent and controversial strand of the literature, but no one that I know of has labeled it as such. The strand is on the “ethics of blame,” and it was mostly kicked off by Angela Smith, in her paper “Being and Holding Responsible” (2007), and Tim Scanlon, in Ch. 4 of his 2008 book _Moral Dimensions_. The idea is that there may be reasons to withhold blame from someone, given, say, that she’s beaten herself up a lot already, or what we’ve done in the past to have sort of forced her into doing what she did, or (the sub-strand that has caught fire in the literature) our own hypocrisy. So we may lack standing to blame. The issue is whether this is a WKR or a right kind of reason to suspend blame.
The same issue is relevant for the literature on forgiveness: When you offend against me, but you’ve done all that could be reasonably demanded of you to make it up to me and acknowledge your wrong, is it fitting for me to withdraw blame in forgiveness, or is my withdrawal justified by a WKR?
So there are clearly domains in which the problem is live, but people aren’t really talking about it in those terms. I suggested one reason above: You need much more realistic WKRs to generate the problem. But another is one you touched on, that some people reduce fittingness to desert, and so this, I think, probably does allow them to set aside the problem.
Quick question for Frank: Would you say the same thing about a fitting attitudes analysis of humor, i.e., some object is funny iff it is appropriate for someone to be amused by it? This strikes me as quite plausible, and I would suggest that MR has much more in common with humor than with electrons.
Fwiw, Glen Pettigrove has written an excellent paper on forgiveness and WKR for a new volume on forgiveness I’m editing with McKenna and Nelkin. Highly recommend it.
Frank
you are right to say that the fitting attitudes account is not plausible to all properties such as electronic charge. Whether moral responsibility is one of these properties for which such an account can be given is of course an interesting question. Most have assumed that it is.
David
thanks – that’s very helpful and interesting. It seems like an interesting sociological difference that in the MR context there is a need for examples to be realistic. It seems true that responsibility is tied to the practice of holding responsible but I would like to think that so is value tied to the practices of valuing in a similar fashion. This was one of the lessons from Scanlon’s WWO. If you think of all the things that are valuable and all the ways in which they are appropriately valued, there is huge variation and much of it is based on intricate human practices and conventions. Think of how tea is valued for example in Japanese ceremonies. But, still it seems like it is relevant in this debate what is appropriate when an evil demon threatens.
Also, I think the ethics of blame cases help us to construct more natural WKR cases. Some people, for example, react extremely badly to being blamed or to being questioned or whatever the relevant reaction is. They might harm themselves or withdraw from the social relationships in a way that might be counter-productive. So, in these cases, there seems to be good reasons not to blame (or to have the other relevant reactions) even if these reasons seem to be wrong kind of reasons to undermine the responsibility of the agent.
With the other reason, it is interesting how many resources you can get from deserve. In the value case, it could be argued that some objects deserve to be valued unlike others but I doubt this would convince as a response to the WKR. Also, in the previous type of cases, it does seem that the person who reacts to blame very badly might not deserve to be blamed even if she were responsible.
Brandon,
thanks – looking forward to reading that paper.
Jussi: You’re quite right, of course, that there may be weighty WKRs not to blame someone who will withdraw or harm herself in the way you’ve suggested. This is along the lines of examples Angela Smith gives in “Being and Holding Responsible,” with respect to someone who has beaten herself up enough already. But notice — and this connects up with the naturalistic examples point — when you point to cases like this, it’s much easier to see a stark difference between what’s fitting and what’s (all-things-considered) justified. When someone has beaten herself up enough already, it would be immoral to pile on with blame, even though the relevant blaming emotion (anger, say) would be fitting insofar as it appraises what she did as wrong (or a slight, or a wronging). In other words, the more naturalistic the example, the harder it is (I would think) to gin up the saucer-of-mud kinds of WKR worries.
Thanks David. hmh – there seems to be a drastic difference in how the MR and value debates understand the problem. With the WKR problem in the value context, the point of the artificial cases is to show that we have an intuitive grasp of what the right kind of reasons are and what the wrong kind of reasons are. The problem, however, is then the challenge of how to give a general theory of what the difference between right and wrong kind of reasons is in a way that carves the intuitive joints at the right places without being circular in a way of just restating the intuitions. Now, going back to the MR context, it seems like your suggestion is that we can generate intuitively right and wrong kind of reasons without the funny evil demon cases and I agree with this. But, here too, the problem on the general level would be to come up an account of the right and wrong kind of reasons that would apply generally, carve the joints at the right intuitive places, and not be circular. So, I am curious why people working in the MR context do not seem to have felt the need to find this kind of general theories.
Your answer, however, points to a direction. Perhaps in the MR context, it is not appealing to find a general theory of what distinguishes right kind of reasons from the wrong kind of reasons but rather what we should do is to discuss each reason separately and give an account why these reasons are or are not relevant for responsibility. What is interesting again is why that kind of a strategy has not been pursued in the value theory context.
So, two interesting differences: Value theorists like abstract examples whereas MR theorists do not seem to like them and value theorists seem to like general solutions to the WKR problem whereas MR theorists prefer to discuss the differences between particular right and wrong kind of reasons. The further interesting question is whether these are merely sociological differences (given the structural similarity of the views) or differences justified by the difference in the subject matter.
Jussi: I’m very late to the discussion here, but I’ve been thinking about the ethics of blame quite a bit recently, and, as Dave notes, it’s an area where the WKR might seem to grip.
One prominent aspect of the WKR issue in the responsibility literature concerns how people understand blame. So, take the reasons we might have not to blame someone because of the consequences of our blaming them (e.g., they’ll react very badly). That’s typically a consequence of some overt treatment of them. It’s much harder to make someone feel bad by privately blaming them, rather than expressing it to them. So the natural cases that get taken up usually have the form of some sort of interaction. My sense of the value literature is that the cases more typically focus on holding the attitude itself.
Now, whether this way of thinking about the cases is right — that is, whether we *ought* to think about them in terms of overt treatment — is an open question. For my own part, I don’t think it is the best way to think about them. In fact, I think there are WKR for blaming that are nevertheless the RKR for how we handle our blame (e.g., attending to it; expressing it). But I believe I’m in the minority here.
Jussi: This is very late, but Michael Zimmerman discusses the WKR problem in MR and value theory in “Responsibility, Reaction, and Value”, Journal of Ethics 2010.