Indifference as excuse

ABSTRACT

According to an influential view, ‘the amount of blame people deserve varies with the extent of their indifference’. That is, the more wrongdoers act from a lack of moral concern, the more they would be blameworthy. This paper argues for the exact opposite claim: the more wrongdoers act from indifference, the less they are blameworthy – that is, in a properly interpersonal way.

KEYWORDS:

• Blameworthiness
• ignorance
• degrees
• moral concern
• interpersonal blame
• detached blame

1. Introduction

Solomon is a professor of philosophy in 2023. He believes that women are less competent abstract thinkers. As a result, he does not take his brilliant female colleagues and students seriously: he does not read their work, teach their work, hire them, invite them to meetings, listen to them in meetings – the list goes on.¹

Is Solomon blameworthy for his sexist conduct? The reaction of virtually everyone: ‘Yes, of course he’s blameworthy. He’s even quite a bit blameworthy’. We think this is mistaken. Solomon may be a horrible person, but he is not blameworthy in an important – to be specified – sense.

To begin: our story won’t be a volitionist one. Volitionists are interested in Solomon’s epistemic condition. What does Solomon know? If he is unaware that he is acting wrongly, volitionists say, he is off the hook.

Attributionists look elsewhere. Attributionists are interested in Solomon’s quality of will. What does Solomon care about? If he acts from a lack of moral concern, then – even if he does not know that his conduct is problematic – he would be on the hook.

Attributionists, then, seem to underwrite people’s widespread verdict about Solomon. The question, of course, is whether that verdict is correct. We do not think it is. Instead, we consider Solomon’s indifference a serious handicap for him. The more indifferent he is, we will say, the less it makes sense to expect better of him. To be sure, we may react to him in some way, but our reaction will be ‘detached’ rather than ‘properly interpersonal’.

In saying this, we are inspired by Mason’s (2019) account. According to Mason, the key function of blame is to remind wrongdoers of their own values, and to get them to acknowledge their fault. It is pointless to start a ‘blame conversation’ with Solomon and get him to acknowledge his fault as long as he does not agree that his conduct is problematic.

As we see it, Mason is on the right track. We would not engage with Solomon as we would engage with others who do see that sexism is morally problematic. Yet Mason’s account faces problems. Getting people to acknowledge their mistakes is not so easy, and it is questionable, as we will explain, whether we should take this as blame’s main function.

We will present an alternative analysis. The plan is this. In §2, we summarize the volitionism versus attributionism debate. In §3, we present Mason’s view and address our worries. In §§4–5, we set forth our alternative view.

Before beginning, a comment on what we take the phrases ‘yes, he’s blameworthy’ or ‘no, he’s excused’ to capture. These may capture: (i) what reaction we normally have, or are naturally inclined to have, to the wrongdoer, (ii) what reaction is useful in some way, e.g. to encourage the wrongdoer to do better in the future, or (iii) what reaction is appropriate, given what the wrongdoer knew or cared about. These distinctions are all too familiar, but they can’t be emphasized enough. Let us just note that – following volitionists and attributionists – we’ll be interested in (iii) rather than (i) or (ii).

2. The debate

According to volitionism, S is blameworthy for A only if S was aware, at the time of acting, that A was wrong (or, was aware that she should have informed herself better about A’s permissibility, but didn’t do that against her better judgment).² According to this view, Solomon is only blameworthy if he himself sees that what he is doing is morally problematic. As we imagine the case, this condition is not satisfied. Solomon acts from both factual and moral ignorance. He does not know that women can be excellent abstract thinkers, nor that it is wrong to discriminate against women in the ways described in the case – nor, indeed, that he should be more attentive and reflect more on these matters.

In one version, Solomon does not even question his belief that women are less competent abstract thinkers, and thinks to see all sorts of evidence for it every day. In another version, he is not fully convinced that women are less competent, though he would surely deny that he is a sexist himself. In yet another version, he knows full well that he is discriminating against women, but he also thinks that his conduct is not too bad and, overall, permissible. As long as he is unaware, at the time of acting, that he is acting wrongly, he is off the hook according to this approach.³

For attributionists, it does not matter much what the agent herself is aware of. More important, for them, is what the agent cares about. And people might not know things because they don’t care too much. According to attributionism, S is blameworthy for A iff S did A from a lack of moral concern (i.e. for features that make A in fact problematic).⁴ According to this view, Solomon is blameworthy for his sexist conduct when he acts from a lack of concern for the intellectual skills of his female colleagues and students.

How to determine Solomon’s degree of moral concern? To do so, we may ignore his self-reported endorsements. Solomon might think and claim he cares a lot about morality, but this does not make him adequately concerned in the right way. Instead, we should look at his perceptual and motivational dispositions. Is he disposed to notice the achievements of women in meetings, and disposed to change his conduct in light of that? Solomon seems to lack these dispositions. He gets numerous opportunities to update his view and behaviour, and yet he fails to take any of them.

For our purposes, it is important to note that attributionists also have a story about degrees of blameworthiness. Here’s Arpaly: ‘the amount of … blame [wrongdoers] deserve varies with the depth of their motivation or the extent of their indifference’ (2003, 115).⁵ Let us state it thus:

Traditional degrees

The greater the lack of concern from which S does A, the greater S’s degree of blameworthiness for A, and vice versa (keeping other relevant factors fixed).

For example, acting wrongly because one received grave news is less blameworthy than if one’s indifference is greater and one would do it even when one had not received the bad news (cf. Arpaly 2003, 69). One promising (though not altogether uncontroversial) strategy is to explicate this modally. Namely, S does A from a greater lack of concern to the extent that S still does wrongful A across nearby worlds where (keeping the rest fixed) it is easier for S not to do A. For example, go to worlds where the wrongdoer does not receive the grave news (keeping the rest fixed), and check if she still acts wrongly. The more such worlds in which she acts wrongly, basically, the greater her lack of concern (cf. Wieland 2019).

In Solomon’s case, we may ask: what if he had a female colleague who was also a friend? What if he had a female supervisor? What if he were strong enough to resist the influence of his male friends? Would he then still be a sexist? If so, his lack of moral concern is greater – and he would be more blameworthy according to Traditional Degrees – than if he wouldn’t.⁶

Of course, in all these cases Solomon is unaware that he is acting wrongly. In contrast to volitionists, however, attributionists deny that such awareness is required. Acting from a lack of moral concern would suffice for blameworthiness. Generally, the two views diverge in cases where agents are unaware that they are acting wrongly, but where morally concerned agents would not be unaware in this way and indeed perform better.⁷

Some have suggested these views don’t really disagree, but describe different forms of blameworthiness. For example, Mason (2019: ch. 4) suggests that volitionism is concerned with ‘conscience’ blameworthiness (i.e. about whether agents act wrongly by their own lights), while attributionism would be about ‘motivation’ blameworthiness (i.e. about whether agents act on flawed motives).

Zimmerman (2008, 193, cf. 2017) suggested that volitionism is about whether wrongdoers deserve punishment, and indeed to suffer for what they have done, while attributionism offers merely a negative moral evaluation, with no such implications regarding punishment.

In addition, Levy (2005) argued that attributionists can’t distinguish people’s blameworthiness for specific acts and their overall badness. Appealing to the fact that someone’s wrongdoing is robust across possible worlds, the concern is, is just saying something about the agent’s (bad) character.⁸

To some extent, attributionism – the ‘fine-grained’ interpretation of the view just outlined – can accommodate this concern. It can still say that someone is very blameworthy for some particular act, even though that person is no bad person overall (or vice versa). For example, imagine that Solomon failed to hire the best candidate, who was female. Moreover, he does so in most nearby worlds where hiring this person is much easier (e.g. where she has hardly any competition). But now also suppose that he treats his female friends and family members always fairly, and acts rightly in plenty of other respects (and robustly so across possible worlds). In such a case, he would be very blameworthy for not hiring the best candidate, even though he is no bad or sexist agent overall.

Still, the question remains: what sort of blameworthiness do attributionists attribute? If they say that a wrongdoer acts from a great deal of indifference, and so is quite blameworthy for some piece of conduct, do they say more than ‘you acted from a great deal of indifference’? This seems to be a negative moral evaluation, but what further content or function does it have? Before presenting our account, we will review Mason’s recent proposal.

3. Mason’s account

Mason makes a key distinction between ‘ordinary’ and ‘detached’ blameworthiness. To illustrate, contrast two variants of Solomon’s case:

1. Solomon knows he should change his life. When confronted, he would say: ‘sorry, I know’.

2. Solomon genuinely believes that women are less competent abstract thinkers, and does not see that ignoring their work is problematic. When confronted, he would happily admit: ‘of course I am not reading female authors’.

As Mason suggests, in (A) Solomon is ordinarily blameworthy, while in (B) his blameworthiness is detached. Let us take both in turn.

In (A), Solomon shares values according to which sexism is wrong. If we blame him, our blame is communicative. That is, we start a so-called ‘blame conversation’, where we expect that he will acknowledge his fault:

Ordinary blame:

You know A is wrong! Acknowledge your fault.

Response wrongdoer:

Yes, sorry, I know.

In response, we expect that he feels guilt or remorse and apologizes. At the end of the conversation, we may forgive him.⁹ The function of this type of blame is to remind the wrongdoer of his own values. Here’s Mason:

In blaming her, I am acting as a proxy for her own conscience: I am simply reminding her of what she already knows to be the case, and making vivid that she has failed. (2019, 105, cf. also 122)

In (B), in contrast, Solomon does not share values according to which his sexism is wrong. This time, we will not start a blame conversation with him. It would be pointless to start such a conversation because he will not see that he acted wrongly. In response to his conduct, we will only express emotions of repugnance, anger, hurt, disdain, or contempt, and do not expect any response. The function of this type of blame – detached blame¹⁰ – is not to remind the wrongdoer of anything, but to remind others in our moral community of our values. As Mason puts it:

Individuals sometimes … end up with no recognizable morality at all. There is no point in trying to communicate with such people, but we are bound to react to their trampling on the values we hold dear. … Our reactions are more for the benefit of each other than for the benefit of the people blamed. We demonstrate to each other our commitment to our values. (2019, 122)¹¹

The difference between Mason and attributionism may already be clear. For Mason, acting from indifference – what Solomon does in (B) – is not sufficient for (ordinary) blameworthiness. Some sort of epistemic condition is required.¹² As we said, if Solomon does not believe that sexism is morally problematic, it makes no sense to start a blame conversation with him and get him to acknowledge his fault.

Mason also differs from volitionism. To explain the disagreement, consider:

(C)	Solomon does know sexism exists in society, and that it is bad. Yet the problem is too abstract for him, and he feels it does not occur in the academic context in which he works.

In this case, Solomon does not believe, at the time of acting, that he is acting wrongly. Thus, volitionism’s epistemic condition is not fulfilled, and he would be off the hook according to it. Mason offers an alternative analysis. Solomon does not classify his own conduct as such, though he does share values that render his sexist conduct morally problematic. As a result, he may still be in a position to acknowledge his fault, and in such a case an ordinary blame conversation won’t be pointless. Fully spelled out, then, an excuse for ordinary blame, on Mason’s account, does not take the form ‘I didn’t know A was wrong’ but rather ‘I don’t endorse values according to which A is wrong’.

To summarize:

1. The function of ordinary blame is to remind the wrongdoer of her values (i.e. values that she did not adequately act on).

2. One can remind the wrongdoer only if she actually shares these values.

3. Hence, if the wrongdoer does not share them, she is excused from ordinary blame.

There is much to learn from this argument (and we will defend our own version in §4). Yet we would like to address some concerns about it, based on further variants of our case.

(D)	We have no clue about Solomon’s values and whether he knows that his sexist conduct is wrong.

What to do in this case? If we have no access to Solomon values, we don’t know whether it makes sense to start an ordinary blame conversation. Paradoxically enough, the simplest way to figure out what he believes would be to start an ordinary blame conversation, and see whether he responds with ‘yes, sorry, I know’ or rather with ‘of course I am not reading female authors’.

(E)	Currently Solomon does not see that sexism is problematic, but he’s a sensible guy and talking with him will open his eyes.

In this case, we know that Solomon does not believe that sexism exists, or at least that his conduct can be classified as such. However, we also know that he tends to learn fast, and gradually grows as a moral agent. So, when we start an ordinary blame conversation with him, he responds: ‘Ah seriously? Let me think about this’, and after some reflection, he is prepared to acknowledge his fault, and promises to do better.

These cases complicate Mason’s approach, yet there are things one could say in response. We may accommodate (E) by building a diachronic element into the account. We should not just say that someone is ordinarily blameworthy only if she already shares the relevant values (i.e. according to which her conduct is morally problematic), but also if she is prepared to adopt them – i.e. within the timeframe of a blame conversation.¹³

We may accommodate (D) by basing blame attributions on your knowledge about the wrongdoer, and by allowing that these may be mistaken. We may temporarily assume that Solomon is ordinarily blameworthy, and then revise our verdict if he responds with ‘of course I am not reading female authors’. Note that we could not do the same for (presumed) detached blameworthiness. Given that the latter is not communicative, and we don’t expect a response by the wrongdoer, we wouldn’t get any opportunity to revise our presumption. Given that often we do not know what people exactly believe, starting an ordinary blame conversation may then well be the default response.

However, there are also considerations that point into the exact opposite direction, that is, according to which ordinary blame should not be the default response. Consider:

(F)	Solomon knows that his conduct is wrong, but he hates it when others blame him. When confronted, he would not acknowledge but rather deny his fault.

This case features a typical human being. People really don’t like it to be blamed. Psychologically speaking, it’s hard to see oneself as a wrongdoer, and acknowledging one’s faults is pretty hard (cf. Tavris and Aronson 2007). As a result, blame conversations that have this aim will be tough, sometimes a lengthy process, and something one may rather avoid doing. Moreover, even if one does start a difficult conversation, they might well be a mere waste.

Mason’s account rests on the assumption that people are excused from ordinary blame if such a conversation is pointless.¹⁴ In (F), too, the conversation is pointless, albeit for a different reason: it’s pointless because the wrongdoer is not able to acknowledge his fault, and not because he would not share the relevant values. Human beings often face these psychological obstacles, and (F) is not an exclusive case. Are we very often excused from ordinary blame? It can’t be that we are excused only because the conversation is pointless. If we are excused, the explanation should go deeper.

Of course, it is an empirical question which means are effective towards some goal. Similarly, it is an empirical question which means are effective to getting people to acknowledge their fault and reminding them of their values. Some people might be like Solomon in (E). They are sensible people who are willing to listen to others as long as they are being approached in a non-confrontational manner.¹⁵ Others, however, will need a harsher approach:

(G)	Solomon does not see that sexism is problematic, but he’s the kind of person that loves being challenged. Only if you yell at him, will he start listening and possibly revise his beliefs.16

Indeed, for some people blame conversations (confrontational or non-confrontational) are very effective. For others, non-verbal emotional responses work better (e.g. anger, disappointment). There may be numerous strategies to remind people of their values – let them watch a documentary on sexism, or offer them friendship and spend time with them. However, much of this has nothing to do with blame. We should keep two questions clearly separate. Firstly, what (outward) reaction to the wrongdoer is effective (or pointless) towards some goal? Secondly, what (possibly private) reaction to the wrongdoer is appropriate, given what she knew or cared about?

In light of this, we would like to press two points about Mason’s account. First, it is problematic to let the appropriateness of blame depend on the effectiveness of the expression of blame (blame’s uptake). Whether it would be appropriate to express blame, should not depend on how the wrongdoer will respond (e.g. on whether or not she will acknowledge her fault). In response, one may weaken the view and suggest that the point of blame is not to remind the wrongdoer, but to raise one’s concerns about her wrongdoing regardless of uptake (along the lines of e.g. Springer 2013). Such an account would not face the problem cases (D) to (G). For example, you can raise your concerns in (F) (where Solomon hates it when others blame him) even though Solomon would deny his fault.

Even so, we do not think that that’s enough (our second point). The point of blame is not just to raise one’s concerns about an episode of wrongdoing. As we will propose it, ordinary blame – blame that addresses the wrongdoer – involves more than that.¹⁷

4. Our account

Although we agree with Mason that it is instructive to think of attributions of blameworthiness in terms of blame conversations, they do not always take such an outward form. When we consider people blameworthy, then what content do such attributions have (i.e. other than ‘you acted from a great deal of indifference’)? Here is what we want to say: to consider someone blameworthy for wrongful A, is to think (privately) or say (publicly):

You shouldn’t have done A! I expected more of you.

For blame to be appropriate, then, is for the judgment (‘you shouldn’t have done A’) to be correct, and for the expectation (‘I expected more of you’) to be fitting. These expectations, as we understand them, have four key features: they are backward-looking, normative, based on the wrongdoer’s quality of will, and agent-relative.

Firstly, they are backward-looking because they are about wrongdoing that already happened. Looking back on the action, you think that, given your estimation of the wrongdoer, more could be expected of them than what they did. You think they performed worse than they should have. (You may also expect her not to commit the same wrong again, i.e. in the future, but that is a separate expectation.)

Secondly, the expectations have normative input. It is not just that you predicted that the wrongdoer wouldn’t do A, that you considered it likely that they wouldn’t do it – for example, given their track record, or given what the average person would do. It is also not that you were merely hoping that they wouldn’t do A, believing that there was a chance they wouldn’t do A. It is also that you think that they should have taken that chance.¹⁸

Thirdly, these expectations are based not on what the wrongdoer herself thought she had to do, but on what she cared about (i.e. her perceptual and motivational dispositions).¹⁹ Solomon did not think he should change his conduct. To some extent, though, he may be disposed to know more and do better (these dispositions come in degrees, as we’ll explain in a minute). The expectations we have in mind are based on that.

What would it matter if Solomon was disposed to do better? Even if he wasn’t, you might think, we expected more of Solomon. We expected him to care more in the first place! We agree. Sexism is wrong, and, in a sense, we expect people not to behave like Solomon. He should care more and do better. Even so, these ‘agent-neutral’ expectations do not amount to blame. Blame, on our proposal, involves ‘agent-relative’ expectations that people should have done better than they did given what they cared about. These agent-relative expectations are based on the wrongdoer’s quality of will. We do not base them on the quality of will she should have had. Of course, we may be tempted to do that, i.e. to expect people to care more and blame them for failing to do so. Yet, such agent-neutral expectations cannot be used to back up our (agent-relative) attributions of blameworthiness.

Where Mason suggested that blame involves the (forward-looking) expectation to see one’s mistake (given what one knows), we suggest that it involves the (backward-looking) expectation to have done better (given what one cared about). Like attributionists, we focus on the wrongdoer’s quality of will rather than her beliefs. Unlike attributionists, though, we think that indifference does not ground blame, but rather excuses. Contrast:

I cared, but it was too difficult to do better

I don’t care.

Where the traditional attributionist excuse takes the former form (cf. Robichaud and Wieland 2017), we are suggesting the latter. Of course, merely saying that one doesn’t care shouldn’t excuse. That would be too easy. What matters is whether the wrongdoer in fact lacks the relevant perceptual and motivational dispositions. To what extent did Solomon indeed fail to notice the achievements of his female colleagues, and fail to question his opinion about them at all? These dispositions come in degrees, i.e. he may act from a greater or lesser degree of indifference. To illustrate, consider these further variants of Solomon:²⁰

(H)	He truly enjoys the old boys network, feels it’s superior, and does not see any alternatives.
(I)	He feels competition, and wants the professorship himself.
(J)	He feels threatened, and afraid that he has to work in ways, or take into account perspectives, he is unfamiliar with.
(K)	He cannot handle all the changes, and is tired of coping with the perplexingly sophisticated world.
(L)	He has discriminated against women throughout his whole life, and is not strong enough to face this fact about himself.

In these cases, Solomon acts, respectively, from superiority, egoism, fear, status quo bias or helplessness, and self-image protection. His motives differ, and so do his dispositions. In (H), Solomon acts from the greatest degree of indifference. In modal terms, his wrongdoing is very robust: he would even act wrongly in nearby worlds where it is much easier for him not to be a sexist (e.g. where he has no competition, or where he does not feel threatened).

The wrongdoing of the others is less robust. For example, in (L) Solomon is concerned about his self-image, but also partly sensitive to features that make his conduct wrong. In nearby worlds where he could change his conduct in ways that wouldn’t conflict with his self-image, he might well do better. Think, for instance, about worlds where people tell him this:²¹

I’m afraid to tell you have treated your female colleagues and students unfairly. You have not noticed it, but Naamah and Zillah have actually been working on the same topic as you, and you will learn a lot from their contributions. By the way, you were not the only one: others in the field committed the same wrongs. Morally speaking, you should do better. It is normal if you feel surprised now. If you want, we will be available to help and brainstorm about how you and your colleagues can do better.

Such a message would be irrelevant to Solomon in (H). In that case, he is not concerned about protecting his self-image. He just enjoys the world as it is, and wants to keep it that way.

We want to say that Solomon is less blameworthy in (H) than in (L) – even though he is clearly a worse person in (H) than in (L). We expected less of the Solomon motivated by superiority than of the Solomon motivated by self-image protection. The latter was at least partially concerned about what makes his conduct problematic, and he could more easily have done better (i.e. his wrongdoing is less robust across nearby worlds where it is easier for him to do better). Since he could more easily have done better, our expectations of him were higher. However, if he was not even partially concerned about what makes his conduct problematic, and could not easily have done better, we drop such expectations, and stop considering him blameworthy (in this sense).²²

To summarize:

1. S is blameworthy for wrongful A only if (and to the extent that) it makes sense to expected better of her regarding A.

2. It makes sense to expected better of S only if she acted from at least partial concern for what makes A wrong.

3. Hence, if S acted from indifference, she is not blameworthy for A.

Moreover, one might want to go further and suggest that (this type of) blameworthiness and expectations are proportionate to each other. The degree of blameworthiness, then, would depend on the degree of the corresponding expectations. Someone who is fully indifferent, cannot be expected to do better, and would be fully excused. Someone who acts from slight concern, can slightly be expected to do better, and is slightly blameworthy. Someone who acts from much concern, can be expected to do better to a high degree, and would be blameworthy to that high degree. Generally put:

Reversed degrees

The greater the lack of concern from which S does A, the lower S’s degree of blameworthiness for A, and vice versa (keeping other relevant factors fixed).

This is attributionism turned on its head. We think it follows from what we have said, but do not want to defend this general claim here.²³ Our central concern is that indifference excuses, in some way. We are well aware that this thesis is counterintuitive. Let us briefly respond to some worries, before addressing the final one in some detail.

Worry 1: Why not also blame fully indifferent agents, and expect more of them?

Reply: You can expect more, but your expectations would not be justified. Given their dispositions, fully indifferent agents could not easily have done better, and it is not appropriate to expect this of them.

Importantly, as long as most wrongdoers are not fully indifferent (i.e. lack any relevant perceptual and motivational dispositions whatsoever), one might think that we should not fully excuse them, and that our form of blame is often appropriate – at least to some degree.

Worry 2: Doesn’t your account give the wrong analysis of Frankfurt cases? If Jones wants to assassinate the mayor, and ultimately decides to pull the trigger, most people consider him blameworthy even if Black was at the ready to ensure he committed the crime (cf. Frankfurt 1969, 835–836). In such cases, though, we cannot say to Jones ‘I expected more of you’ since he would have done what he did regardless. Your account seems to imply that Jones is excused.

Reply: These are different expectations. In Frankfurt cases, agents cannot be expected to do better in the sense that circumstances external to them are such that they cannot act otherwise. Our account, however, asks whether we can expect better of agents given their own dispositions. Even when we could not expect Jones to do better given Black’s plan, we could still expect him to do better given what he himself cared about. Whether such expectations are justified will depend on his exact dispositions. Thus, we need to know more about Jones to determine his blameworthiness. For example, it may still be that he is conflicted about the crime (as we may well imagine). In that case, he may still be considered blameworthy on our account.²⁴

Worry 3: This seems like a nice proposal of what some form of blame looks like, but is this intended to be the full picture? Specifically, shouldn’t we leave room for blame that does not entail any expectations?

Reply: What we have described so far is supposed to be a form of interpersonal blame. We, too, think that, if someone is excused by indifference, she may still be blameworthy in a more detached way. We may not expect more of the wrongdoer, though we may still express our frustration and signal to others in our moral community that their conduct is problematic.

Worry 4: When you say that Solomon is excused, you take blame to be very different from how blame is normally understood. Why care about this sort of blame?

Reply: Regardless of whether our blame is ‘niche’ or revisionary, we still believe, as we’ll discuss next, that looking at blame this way is valuable.

5. Interpersonal blame

In principle, blame can be valuable for different parties: the wrongdoer, the person wronged (the victim), the moral community at large, the blamer, or a combination of these. For each of these parties, blame can have various functions:²⁵

The victim: by blaming, you may stand up for her, protest on her behalf, acknowledge that she was indeed wronged, and/or that you care about this and don’t condone it.

The moral community at large: by blaming, you may make others aware a wrong has been committed, remind them of their values, and/or signal your commitments to others.

The blamer: by blaming, you may remind yourself that this was wrong, let go of your anger and frustration, remind yourself of your commitments, and/or reconsider your relationship to the wrongdoer.

The wrongdoer: by blaming, you may make the wrongdoer aware she acted wrongly, invite her to respond in some way – e.g. to feel guilt or remorse, to offer a justification for what she did, or walk her through Mason’s blame conversation – and/or encourage her to care more and do better in the future.

Some of these functions are ‘detached’ and do not really engage the wrongdoer, while others are more ‘interpersonal’ (we will drop Mason’s term ‘ordinary’²⁶). According to Mason, interpersonal blame takes the form: ‘You know A is wrong. Acknowledge your fault’. Moreover, the point of these attributions, Mason proposed, is to remind the wrongdoer of her own values. On our account, in contrast, interpersonal blame takes the form: ‘You shouldn’t have done A. I expected more of you’. What might be the point of such attributions?

The point, as we see it, is to appeal to the wrongdoer’s intrinsic motivation to do better. Intrinsic motivation, in this context, is the motivation to do better simply because doing so is the right thing. You might not get anything out of it over and above acting morally.²⁷ Extrinsic motivations to do better, then, are about obtaining further benefits: to feel good about oneself, to avoid condemnation by others, to look good in the eyes of others, to get their credits, to protect one’s reputation, to relieve one’s own distress, to avoid guilt feelings, or to avoid legal punishment.

To blame people in a detached way offers them at most an extrinsic reason to do better. You express your frustration or disappointment, the wrongdoer may notice that you didn’t like it what she did, and decide to change her conduct because of this. People don’t like it to be blamed – recall case (F) – and don’t want to be seen as bad agents, i.e. not by themselves and not by others. She improves only because she wants to avoid your condemnation, but not because morality demands it. Hence, detached blame can still serve the purpose of keeping people in check. People may want to abide by social rules to maintain their standing in the community, even if they are not intrinsically motivated to do the right thing.²⁸

Detached blame may even offer the wrongdoer no reason at all, not even an extrinsic one. As Mason pointed out, you may also only signal to others in your moral community that a wrong has been committed, and fully ignore the wrongdoer herself. For Mason (2019, 121), detached blame functions as a relationship modifier.²⁹ If Solomon does not share values according to which his sexism is problematic, and you blame him in a detached way, you may also try to avoid him in the future – not only to reduce the ways in which the wrongdoer may harm you in the future, Mason adds, but also to avoid condoning his behaviour. One might worry that we shouldn’t exclude people from the moral community in this way, and ‘write them off’ as moral agents.³⁰

Does the same worry apply to our account? If Solomon acts from indifference (i.e. lacks the relevant perceptual and motivational dispositions), we don’t expect more of him and don’t appeal to his intrinsic motivation to do better. Even so, this doesn’t necessarily imply that we write him off, for at least two reasons.

Firstly, avoidance can be partial (as Mason 2019, 139 also points out). Solomon is a sexist at work, but he may still act rightly, and be considered part of the moral community, on other fronts. He does not take his female colleagues seriously, though he may still, say, be a great friend or responsible citizen in other ways. If that is so, we won’t avoid him entirely: we might not invite him to certain work meetings, but still engage with him elsewhere.

Secondly, we agree that there is a risk of writing off people, and of reserving blame only for an exclusive inner circle. However, it is important to see that exclusion is not essential to our account. In order to remind the wrongdoer, Mason argued, she needs to satisfy an epistemic condition. In order to appeal to the wrongdoer’s intrinsic motivation, we say, she needs to satisfy a quality of will condition. If people don’t satisfy these conditions, they are no candidates for these uses of blame. Even so, withholding such interpersonal blame is still compatible with interacting with them in other, positive ways: informing them, or making them care in the first place (i.e. help them satisfy the relevant conditions). That does not sound like writing them off.³¹

What is more, even if we do ‘write them off’ in the sense that we consider them in part to stand outside of the moral community, this may also help open up ways to engage with them, rather than only shut doors. If we conclude that we cannot expect more of Solomon, morally, we may give up on appealing to his intrinsic motivation, and instead try more strategic ways to arrive at a modus vivendi that keeps the community intact. We may make our peace with the fact that Solomon is morally underdeveloped in this sense, and decide how best to live alongside him given this fact. This may strengthen a peaceful community, rather than weaken it.

What’s the value of appealing to one another’s intrinsic motivation to do better? The basic idea is that we shouldn’t only be interested in keeping one another on track on a purely cognitive level, but also motivationally.³² Reminding one another of our values is valuable, but so is expecting one another to act on them. In both of these cases, interpersonal blame is supposed to be of primary value for the wrongdoer (rather than for the blamer, the victim, or the community at large³³). In our case, if we expect better of the latter, and appeal to her intrinsic motivation, she is being respected as a moral agent. Specifically, we see her as someone who can act for more than merely extrinsic reasons, and does the right thing simply because it is the right thing. We think it is valuable if others see us like this.³⁴

It is worth pointing out that, in contrast to blame of the form ‘acknowledge your fault’, blame of the form ‘I expected more of you’ does not necessarily invite or require a response. Of course, the wrongdoer may still respond in some way – for example, by saying ‘I see, I promise to do better’ – yet there need not be any blame conversation. While ‘I expected more of you’ is backward-looking, the corresponding invitation to improve is forward-looking. Sometimes wrongs are one-shot moral mistakes, in which case we just invite the wrongdoer not to do that again. More typically, the wrongdoing forms a pattern of repeated behaviour – Solomon’s sexism – in which case we invite them to improve.

Despite this forward-looking aspect, our account need not run into the type of problems that afflicted Mason’s account. Appealing to people’s intrinsic motivation to improve has value regardless of the actual uptake of doing this. This has value whether or not people actually succeed in doing better. For, in such a case, wrongdoers are still being respected as moral agents (that is, as agents who act for more than merely extrinsic reasons, as described earlier). In contrast, the value of reminding people of their own values lies exactly in the actual uptake of doing this.³⁵ If wrongdoers are not being reminded as a result of the blame conversation, Mason’s assumption was, the latter is pointless. Consider these cases:

(M)	Solomon hates it when others blame him. When confronted, he won’t be encouraged to do better, but sticks to his guns instead.
(N)	You hate it to confront Solomon. You fear the conversation will be painful, and that he might avoid you, i.e. to be rid of the moral pressure.

In such all too familiar cases, having a blame conversation with Solomon might not be worth it. It is counterproductive or even painful. However, this does not mean he is excused. As we suggested, Solomon is blameworthy if, and to the extent that, we could expect better of him (given his degree of concern, which these case descriptions do not say much about). By doing this, we appeal to his intrinsic motivation to do better. This has value even if when, on balance, it is not worth the costs.

What if we blame Solomon privately in these cases? In such a situation, you might say, we don’t really engage with him (or appeal to anything). We see at least two possible analyses. Firstly, we could say that blaming others privately – merely thinking, but not saying, ‘I expected more of you’ – does not have much value. It may be valuable instrumentally, i.e. insofar as it helps us to also say these things on other occasions, but that would be it. Secondly, and more promisingly, we could say that by blaming wrongdoers privately, they are still being respected as moral agents (as just described). You might not actually invite them to do better for intrinsic reasons, but you still consider them capable of doing so. That still has value.

6. Conclusion

What goes for sexism, goes for other cases. Compare someone who simply wants to enjoy himself, and feels the right to fly, eat meat, etc. versus someone – with the same major carbon footprint – who suspects that something is not quite right, though is not strong enough to face this fact about himself and manages not to think about it, or even confabulates elaborated stories on why his conduct would be permitted. The former agent cares only about himself, while the latter is concerned about his footprint, to some extent, though not enough to change his life.³⁶Are these agents blameworthy? According to Traditional Degrees, the former type of agent is more blameworthy as he acts from a greater deal of indifference. We say the exact opposite: the former acts from more indifference, and is therefore less blameworthy. Given what he cares about – hardly anything but himself – we expect less of that agent. In this sense, his indifference excuses.

Acknowledgements

For comments and discussion, we would like to thank: Anna Hartford, Dominik Boll, Lisa Bastian, Gunnar Björnsson, audiences in Milan, Vienna, and Amsterdam, and a referee of the journal.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 This case is adapted from Arpaly (2003, 104). In our case, Solomon does not live in an isolated community.

2 In this paper, ‘S’ stands for an agent, and ‘A’ for an action of hers.

3 Why do volitionists think this? The idea is that only if Solomon himself knew he had to reflect more, he could have acted on that knowledge and do better, i.e. in a rational way (cf. Zimmerman 2008: ch. 4; Rosen 2003; Levy 2009).

4 The locus classicus: Arpaly (2003), which we summarize here.

5 Arpaly and Schroeder (2014: ch. 7) add some qualifications regarding manifestation, but the basic idea remains the same: the greater his manifested lack of concern, the greater his degree of blameworthiness. Cf. Björnsson (2017): the greater the lack of concern required to explain the wrongdoing, the greater his degree of blameworthiness. Sarch (2019), too, defends such a principle, albeit in the context of criminal, rather than moral, culpability.

6 In Arpaly’s original story (2003, 104), the question is: what does Solomon notice and do in nearby worlds where his community is less isolated? If, in many of those worlds, he is not ignorant (and, indeed, not a sexist), then he does possess the relevant dispositions.

7 cf. FitzPatrick (2008), Harman (2011), Wieland (2017), Robichaud (2017), among others. There are also ‘complex akrasia’ cases where agents are blameworthy according to volitionism, but excused according to attributionism (cf. Hartford 2020). In these latter cases, S does A in awareness of what she is doing is problematic, though on the basis of okay motives, and non-robustly across nearby worlds.

8 For similar reasons, Markovits (2010) argued that praiseworthiness for rightful conduct does not depend on whether the latter is modally robust.

9 Mason’s blame conversation is an extended version of McKenna’s (2013) account, where the opening move is an expression of indifference or ill will by the wrongdoer, and blame a response of disapproval.

10 In an earlier paper, Mason (2015) used the term ‘objectively’ blameworthy rather than ‘detached’. The term ‘detached’ is particularly fitting: in blaming someone in this way, you ‘distance’ yourself from the wrongdoer. Note that this is not detached in the sense that it is non-emotional (for this latter use, cf. Pickard 2013).

11 Shoemaker and Vargas (2021) explicate this in terms of ‘costly signalling.'

12 Talbert (2012) and Smith (2013) suggest that blame can function as protest, and this function requires no such epistemic condition. You can stand up for the victim even when the wrongdoer does not know her behaviour is problematic. Mason agrees that blame can have this function, though classifies it as detached blame.

13 Not if it takes, say, six years of reflection and weekly ethics classes. Following Fricker (2016), Mason (2019, 123–124) allows for such a diachronic element, when she suggests that blame may also be used ‘proleptically’, i.e. as a way to educate the wrongdoer. Note that this admits paradoxical implications. If S does wrongful A at t₁, and learns at a later time t₂ that A is wrong, then at t₂ S becomes retroactively (ordinarily) blameworthy for A at t₁.

14 Thus Mason: ‘because she is not acting badly by her own lights, an attempt at communicative blame would be pointless – we would not be latching onto any receptors or recognition in her.’ (2019, 114)

15 Experimental data of Scaife et al. (2020) suggests that blaming people for their implicit biases need not be counterproductive (as some may think), but rather strengthens people’s intention to do better – at least, as long as the blame is non-confrontational and presented in a low emotional tone.

16 This case is inspired by Upadhyaya’s comment: ‘beyond solidifying views people already hold, we seek to give people new information and invite them to reconsider even their deepest convictions’ (2020, 485).

17 Globally, we will follow Mason’s focus on blameworthiness as accountability (and maintaining the moral community) rather than, e.g. retribution (giving wrongdoers what they deserve) or remediation (acting in the interests of the victims).

18 Rudy-Hiller (2020) argues that it’s not so easy to cleanly separate normative and predictive expectations, as the latter may recalibrate the former.

19 For the paper’s purpose of targeting Traditional Degrees, we take this attributionist assumption on board, though this, to be sure, is the point where volitionists and attributionists part ways (cf. FitzPatrick 2008; Levy 2009).

20 Moody-Adams (1994) describes some of these in terms of ‘affected ignorance’. Arpaly (2003, 108) describes a version of (K).

21 Inspired by the ‘blame condition’ in Scaife et al. (2020, 4–5).

22 The idea is that we do not only lower any predictive expectations we might have of him, but also, and especially, our normative expectations. As suggested, the latter are based on the quality of will the agent had and not should have had.

23 An alternative would be a threshold view, according to which the wrongdoer needs a certain degree of moral concern, in order to be blameworthy at all.

24 Acting for reasons of one’s own, then, is not sufficient for blameworthiness. All depends on further details about the agent – cf. cases (H) to (L) above.

25 Cf. the contributions in Coates and Tognazzini (2013), e.g. Franklin (2012, 216–218) and Bell (2013, 267–269). The point of this brief overview is not to be exhaustive, but to situate our account in conceptual space.

26 The term ‘ordinary’ has the connotation of ‘being the most typical’ sort of blame, while this meaning is not what is intended. Indeed, it is questionable whether, as a matter of fact, interpersonal blame is more frequent than detached blame.

27 Contrast Darwall’s (2006) view that blame offers the wrongdoer a ‘second personal’ reason to do better, i.e. a reason to do better given by the blamer (rather than, as we say, by morality).

28 Cf. Calhoun (1989) on moral reproach as a ‘tool for effecting social change’.

29 Mason borrows this from Scanlon (2013), though applies it to detached blame only. If you ordinarily blame the wrongdoer, and the latter acknowledges her fault, then typically you continue on the same terms as before.

30 For versions of this worry, cf. Korsgaard (1996, ch. 7), Smith (2005, 269).

31 Zheng (2021) argues that, rather than blaming wrongdoers, we may be interested in formative moral criticism which aims to improve rather than assess, and that such formative responses are virtually always justifiable. We agree that constructive feedback can often be valuable, but our account is more restricted: interpersonal blame (of the sort described) is appropriate only if wrongdoers satisfy a quality of will condition.

32 Fricker (2016), too, suggests that the point of blame is to promote the alignment of moral understanding, as well as corresponding motivation.

33 Yet, there may also be (derived) value for the blamer. If you expect more of someone, while that person could not care less, your expectations will likely be frustrated. In such a case, it is less frustrating to push extrinsic buttons.

34 Rather than seen as some someone who can only be triggered on the basis of extrinsic buttons. cf. Strawson’s objective attitudes, according to which we regard the wrongdoer ‘as something to be managed or handled or cured or trained’ (1962).

35 The same problem afflicts the classical forward-looking view that the function of blame would be to improve wrongdoers or to encourage them to do better (cf. Smart 1961). The value of improving or encouraging others lies, too, in the actual uptake of this.

36 Plausibly, many people are of the second type. Only a few of us care only about themselves (cf. Wieland 2017). We recommend Hartford (2020), for further pairs of cases.