Some Of Our Books


« An Argument Against Subjectivism about Moral Wrongness | Main | Deontic and instantive morality »

April 06, 2015


Feed You can follow this conversation by subscribing to the comment feed for this post.


to my eyes you have a very promising idea. I'm excited to read your papers and see what else you have to say about this. It seems that love as you have it and other non-deontic second-personal reactive attitudes give very interesting colors to moral taxonomy.

Having said this, I have an initial wonder. How would you characterize the threshold between deontic reactive attitudes and attitudes of the heart? Is it a set of merging and unclear cases or would you posit a defining (set of) condition(s)?

My concern comes with what you have to say about trust, especially in cases of testimony. True, in some cases you are not to resent someone who breaks the trust you had placed in him. On the other hand, there seems to be a lot of cases where feelings of betrayal are quite appropriate. Take the feeling of betrayal I could feel when I find that my friend (who recognizes and ‘accepts’ the trust I place in him, and in similar conditions he trusts me) has lied to me about how well he has been feeding my dog while I’m on vacation –I would certainly expect some kind of reciprocation.

When a trusted person lies to me there is a clear-cut case of broken trust. However, in some cases this breaking can well be received with attitudes of the heart, and in other (very common and pervasive) cases it is received with deontic reactive attitudes. Even more, cases where we could talk about straightforward reactive attitudes of the heart can easily transform in clear cases where deontic reactions seem accurate.
Do you have something to say about this? How would you characterize the cases in which lying calls for reactive attitudes of the heart?

Again, great entry.

Thanks very much for doing this, Steve. I'm glad to see you turning your attention to the wider range of interpersonal attitudes. I have many, many questions about various details of your view, but I will try to rein it in to focus on a central issue. You've long taken the standard "blaming" reactive attitudes -- resentment, indignation, and guilt -- to implicate and/or carve out the realm of the deontic, and so to involve demands that the offender hold himself accountable for a violation of the relevant obligation (a violation that the resenter has the authority to demand he not perform). (By the way, you used the term "answerable" first above where you normally use "accountable" -- do you take them to be interchangeable?) This is meant to capture a central Strawsonian feature. The question is what that feature is. You take it to be a "second-personal, reciprocating structure." This is what you take to unify the accountability reactive attitudes with the more positive attitudes of the heart you're talking about today, despite only the former being deontic.

I wonder about the extent to which this captures the heart of the Strawsonian stance. While Strawson does use the language of "demands" regularly, he talks almost as much about "expectations." And more importantly, he emphasizes repeatedly how our reactive attitudes (all of them, negative and positive) reflect these expectations and demands with respect to (others' and our own) quality of will. While it's not entirely clear what he had in mind, we can roughly capture it by the other term he used, "regard." And, crucially, the kind or degree of regard we expect or demand from others is a function of the "type of relationship in which we stand to another human being."

This suggests a different way of unifying the reactive attitudes (both accountability and of the heart), as the kinds of attitudes that reflect our concerns with regard, as a function of specific relationships. In negative cases, regard includes respect, of course, and so includes the deontic. But it doesn't seem wholly captured by the deontic. I may respond with shame (a reactive attitude many don't talk about) in response to my failures of my own expectations with respect to some (perfectionist/societal) ideal, a failure of self-regard. Alternatively, I may resent someone for doing something suberogatory (she wouldn't let me borrow a few bucks when I really needed it!). Or I may resent someone's purely emotional insensitivity (he was amused when my daughter fell and broke her arm!). These are all arguably failures of regard of various kinds, albeit not deontic failures.

And on the positive side, gratitude, trust, love, etc. are unified as well in virtue of their relation to regard, the manifestations of which count as such in light of the terms of a specific (antecedent) relationship. There's a kind of "above and beyond," or unexpected degree of, regard involved in the kinds of benefits that render gratitude fitting, for instance. Trust as well involves a certain degree of regard, as does love (and its own expectations and demands).

Sorry, I've gone on too long, as I'm kind of thinking out loud here. But to boil it all down, why not think of the perhaps more fundamental unifier of all the reactive attitudes as regard (defined in light of the terms of antecedent relationships), of which respect/deontic demands are just one subset, rather than -- or perhaps in addition to? -- the second-personal, reciprocating structure? I hope that makes sense.

Hi Steve,

great post. Really looking forward to reading your papers on this.

A little bit in the direction of the stuff about others' "quality of will": do you have thoughts on how commitments (of the sorts related to love and friendships) are implicated in the second-personal attitudes of the heart?

Commitments are a little bit similar to promises. And in earlier work (if I recall correctly), you've treated promises as a key example of the more deontic type of second-personal attitudes/duties.

It is a common thought that true love and friendships require our being steadfast and committed - rather than our being there for each other only when it is fun and convenient.

So I wonder if there are some elements of the second personal attitudes of the heart (such as love and friendship) that point a little bit in the direction of the deontic second-personal relations -- even if other aspects of the second personal attitudes of the heart point in different directions.

Again, the aspect that would point in a similar direction would be the element of commitment, which is often thought to be a crucial part of desirable forms of love and friendship.

So, it would be interesting to hear about how you think about the importance of committments here - and how your thoughts would relate to your earlier work on the second-personal aspects of promises.

Great post, Steve! As you know, you and I agree on a lot of this. The biggest question for me is still what the second-personal is, once all of the deontic concepts and attitudes are treated as a just a region of the second-personal realm. It is something about answerability (which must be broader than accountability) and seeking a reciprocating form of uptake? I, too, am curious about what you think of David S's suggestion regarding regard. :)

As you know (again), I don't think interpersonal--nee "normative"--hope is just one of several non-deontic interpersonal attitudes; rather I think hope is the non-deontic analogue of demand. So, as demand is manifest in deontic attitudes like resentment, hope is manifest in what I'm presently calling "instansic" attitudes like disappointment ("instansic" from the Latin root "instans," meaning "urgent" or "pressing"). Deontic and instansic morality together constitute the interpersonal realm. I have been thinking of the interpersonal realm as roughly the space of reasons, and I wonder about thinking of the second-personal realm in this way, too. I know you believe there are reasons that are not second-personal, but couldn't it be that, whenever we relate to each other *as reasoners* we relate the relevant reasons second-personally (even when said reasons are themselves not second-personal)?

Thanks for your post, Alejandro. First, I am not denying that moral reactive attitudes are not sometimes fitting responses to failing to live up to our trust, especially when these are the cases we would classify as betrayals. I am just saying that it is not essential to trust, unlike legitimate demands, that failing to conform makes reactive attitudes like resentment and moral blame fitting.

Also, I'm not advancing any particular normative claims about when we have a case that is apt for holding someone accountable with (deontic) reactive attitudes like resentment and blame is fitting and when we do not. I am making claims about the nature of the attitudes themselves and the ethical ideas to which they are tied conceptually.

So take your case of the friend whom you trust to feed your dog. My claim here is that the trust itself is non-deontic; unlike a case where you have a legitimate demand on your friend to feed your dog, as when she promises to do so, if you trust your friend to feed your dog then your trust does not necessarily give you any authority to demand anything just in itself. Of course, if we build in various facts that trigger an obligation on your friend's part, e.g., reliance, or an implicit understanding or agreement, then there will be a legitimate expectation that would make blame for failure fitting. The point is that these are not intrinsic to trust in, for example, they are intrinsic to promise.

David, thanks so much for these great questions and points.

For many purposes, I think we can stipulate that in a suitably broad sense, "quality of will" and "regard" are what reactive attitudes (deontic and non-deontic) are concerned with. I doubt though that romantic love or even all of friendship has to do with the will, even broadly construed. Much of what we want from lovers and even friends has to do with forms of attention, appreciation, regard, etc., that are not under our voluntary control.

Also, I doubt that what makes an attitude reactive can be just that its object is regard, since there are third-personal responses to lack of regard, e.g., annoyance or disgust that do not presuppose implied relationship (and second-personal address) and so cannot be reactive in Strawson's sense. Of course, you add that these are within the context of relationship. But even here, there are attitudes that occur within relationship (annoyance at a friend or loved one) that are not themselves reactive, since it is not true of the attitude that it implies or presupposes the relationship.

Despite the fact that Strawson includes shame, I actually don't think he should have included it since shame does not carry the presuppositions of the participant stance. As I see it, shame is a reciprocal attitude to contempt or disdain, which also do not imply relationship or interpersonal address. They do not have an implicit RSVP in the way reactive attitudes do. They do not therefore call for a reciprocating attitude in the way blame calls for its object to take responsibility for and acknowledge his wrongdoing, hold himself answerable etc. (See, e.g., the discussion of guilt and shame in Brendan Dill and my "Moral Psychology as Accountability" (

Here I am interested in the difference between deontic and non-deontic reactive attitudes.

Finally, take your cases of resentment of the suberogatory and amusement. For resentment to be fitting, I take it that these must be cases not just of things that fall short of virtue, there must be something like a wronging. I can well imagine thinking on some such cases that resentment would be fitting, but then I find myself thinking that the person does wrong. It is wrong to be amused by someone's falling and breaking her arm. If you resent someone for not loaning you a few bucks, then it seems to me that you are seeing her as having wronged you.

Thanks a lot for the reply, Steve. There are many things to follow up on, but I don't want to distract from your discussion with others. So just in response to your first paragraph, I don't think the will in "quality of will" need refer to voluntary control (so it's not a Kantian "good will"). (Michael McKenna's book has an excellent discussion of this point.) I think you're exactly right that expectations in friendship and love include various forms of attention, appreciation, and regard (and I would want to add emotional sensitivity and simpatico emotional responses). But these all quite plausibly fall under the rubric of an attitude or perceptual stance we have toward each other, which I think is quite in line with what Strawson had in mind for "quality of will." Indeed, I think there are plenty of expectations and even demands we have within various interpersonal relationships that don't implicate voluntary control in the least, despite the fact that they are matters of regard.

Great stuff! This helps me formulate something I've been claiming for a while, namely that there's two kinds of forgiveness. Very roughly, one consists in ceasing to hold negative reactive attitudes towards the agent. This is familiar, but the other form is not equally well-recognized. To put it in your apt terms, it consists in opening up to positive second-personal attitudes of the heart. Crudely, if someone has something that hurt you deeply, you can cease to resent them (and thus forgive them in the first sense) without yet being open to love and trust toward them (so that in the second sense, you haven't yet forgiven them). Perhaps you could also come to trust someone while still resenting them for what they did. So I think recognizing the diversity of second-personal attitudes helps with understanding the phenomenon of forgiveness (and relatedly, the different forms of blame) as well.

Thanks so much for your question, Sven. I grant that commitments and, of course, mutual commitments, generate obligations (in this case, bipolar obligations TO one another). That's central to my arguments in "Responsibility Within Relations" in MORALITY, AUTHORITY, AND LAW: ESSAYS IN SECOND-PERSONAL ETHICS I. And these obligations entail legitimate demands, accountability, and justification for deontic reactive attitudes. I don't think, however, that commitment is essential to second-personal attitudes of the heart like love, trust, etc. It is consistent with that that the most excellent forms of these are steadfast, but that doesn't entail any claim or demand to their steadfastness. Think of God's love, which is certainly supposed to be steadfast but which we can neither claim nor demand.

Thanks so much, Adrienne. There is so much that we agree on here. It may be that personal hope plays something like the role in second-personal attitudes of the heart that demand plays in deontic reactive attitudes, though I'm not sure how to fit that into the case of gratitude. Any thoughts on that?

I'd also like to hear more about what you mean by calling the interpersonal realm the space of reasons. You're right that I think there are all sorts of reasons that aren't second personal, and you don't seem to be denying that. I agree that when we reason together (as you and I are right now) we presuppose a common second-personal authority with respect to our reasoning, but I'm not sure how that affects what we are reasoning ABOUT.

Thanks so much, Antti, for your observation about the connection between two kinds of reactive (reciprocating) attitudes and two kinds (or perhaps stages) of forgiveness--withdrawal of deontic reactive attitudes and reopening oneself to second-personal attitudes of the heart like love and trust. In the first stage, one lays off the resentment. In the second, one lays oneself open.

This seems entirely correct to me.

That's a great point, Steve. Many thanks for your reply.

Excellent post, Steve! I look forward to digging into the papers you mentioned above. I do have a few comments/questions if I may.

I'm curious. Given that trusting and loving relationships often give rise to obligations (which you admit) can the separation between deontic and non-deontic hold in practice? There seem to be cases that speak in favor of the separation, for instance cases of parental love don't seem to give rise to obligations in young children (or God's love as you mentioned in your reply to Sven). That said, it seems that one loses love over time if the obligations that arise from the trusting or loving relationship are not met (one falls out of love for different reasons as well). This suggests that the attitudes are intertwined or deeply connected.

Now, you will want to resist the connection but there seems to be *at least* a weak connection. If the obligations that arise from trusting and loving relationships are NEVER upheld, then the love or trust would not seem fitting. I grant that this is not the case in parental loving relationships but this doesn't seem so apparent in reciprocal non-parental relationships.

So, I think I agree that they are different initially, but as the relationship matures they seem to be essential. I would want to argue that feeling the attitude and directing the attitude is quite different so maybe the connection I am alluding to is connected to the fittingness of directing the attitude (of trust or love) and not about the feeling itself.

Hopefully this was coherent. In any event I look forward to hearing what you think and what others have to say about this. Fascinating stuff!

Thanks so much for your thoughtful comment, Justin. I certainly agree that love and respect are intertwined in all sorts of ways, and that when obligations of relationship are not met this can tend to undermine or lessen love. But there is no necessary connection, so far as I can see. That is part of what makes human relationships so wonderfully complex. It is an interesting question whether love can ever be unfitting. I am not sure it can be.

"It is an interesting question whether love can ever be unfitting. I am not sure it can be."
--sure it can. :)
[and thanks for posting this--such interesting material!]

The comments to this entry are closed.

Ethics at PEA Soup

PPE at PEA Soup

Like PEA Soup

Search PEA Soup


  • Unless otherwise indicated, the views expressed in any given post reflect the opinion of only that individual who posted the particular entry or comment.