1. Introduction
Skepticism about blameworthiness says that there is good reason to doubt that, in our world, humans are ever blameworthy or—I will use these expressions interchangeably—that humans are ever appropriate targets of blame for what they do. Skeptical arguments are typically based in considerations about how blameworthiness fits into a world that is deterministic or in which human actions and intentions are the results of luck beyond the agents’ control. Because of factors such as these, the arguments go, humans are never worthy of blame. But how, exactly, should we understand “blame” here? In other words, what kind of blame should skeptics be skeptical about and, correspondingly, what kind of blame should those who want to tackle skeptical arguments be concerned with? This is the main question of the paper.
Typical accounts of the nature of blame aim at capturing the core of our everyday blame practice and at identifying every blame response as such. This is not the aim of this paper. The reason for this is that there are responses that are called blame—at least by some—and that are obviously appropriate even if our world is structured by luck or determinism. Skeptical arguments should not be concerned with these responses. This paper aims at identifying a specific kind of blame. The goal is to characterize the responses that should be at issue between those philosophers who argue that, in our world, humans are never appropriate targets of blame and those who deny this. Achieving this goal is important because, first, it will make sure that skeptics and their opponents are talking about the same thing when they talk about blameworthiness. Second, it will identify those aspects of our practice that would be inappropriate if the skeptical arguments are sound.
In section 2, I will present three constraints for an account of the kind of blame skeptics should be skeptical about. Section 3 argues that there is reason to doubt that prominent views on blame meet them. In section 4, I will make a new proposal. The basic idea is that the kind of blame skeptics should be skeptical about is to be understood in terms of claim forfeiture.
Let me begin with a closer look at what I call skepticism about blameworthiness. Most authors in the debate focus on moral responsibility and take being blameworthy to be one way of being morally responsible. Derk Pereboom, for example, says:
For an agent to be morally responsible for an action in this [basic desert] sense is for it to be hers in such a way that she would deserve to be blamed if she understood that it was morally wrong, and she would deserve to be praised if she understood that it was morally exemplary. The desert at issue here is basic in the sense that the agent would deserve to be blamed or praised just because she has performed the action, given an understanding of its moral status, and not, for example, merely by virtue of consequentialist or contractualist considerations. (Reference Pereboom2014, 2)
Following Pereboom, what I call skepticism about blameworthiness understands the relevant kind of worthiness in terms of basic desert (see also Levy Reference Levy2011, 2; Waller Reference Waller2011, 2–3; Caruso and Morris Reference Caruso and Morris2017, 839; for an overview, see Caruso Reference Caruso and Zalta2018). The sense of desert at issue does not depend on positive effects of blaming, such as deterring people from future wrongdoings, or on whether the agents have consented to being blamed in advance. The relevant sense of desert is such that whether agents deserve blame for their actions only depends on the negative moral status of what they did and on how they were when they did it, especially the kind of agency and knowledge they had. Thus, this paper is concerned with the claim that there is good reason to believe that, in our world, no human basically deserves blame. Given this understanding of worthiness, the main question is: how to understand the relevant kind of blame?
Three further notes are called for. First, skepticism about blameworthiness is narrower than standard skepticism about moral responsibility. This is because most skeptics also doubt that humans deserve praise in our world. As I will not say anything about praise in this paper, I will stick with skepticism about blameworthiness. Second, I will stay neutral toward whether skepticism about blameworthiness is true. The goal is to clarify how to understand this view, not to say whether it is correct. Third, I will focus on blame for actions. But what I will say should also work when framed in terms of blame for character traits, attitudes, or something else.
2. Three constraints
Let us adopt the skeptic’s perspective and ask: What can we expect from an account of the kind of blame we should be skeptical about? Of the many possible desiderata and constraints, I will pick out three and then show in the next section that some famous accounts of blame have problems meeting them.
Consider, first, the neutrality constraint. It says that an account of the kind of blame skeptics should be skeptical about should not make skepticism trivially false. To illustrate, take the dummy view that blaming someone is constituted by the belief that this person did something bad. On this view, it is trivially true that blaming people is sometimes appropriate because some people, surely, do something bad. Skeptics do not and should not deny this.
An account of the kind of blame skeptics should be skeptical about should, second, meet a revisionary constraint. It demands that the account deliver a picture of blame that makes skepticism a thesis that is relevant for our everyday lives. More precisely, the account of blame should make sense of the idea that skepticism is revisionary because it implies that an important part of our everyday practice is inappropriate. The motivation for accepting this constraint is that most skeptics, in fact, present their views as essentially revisionary. The aim of this paper is to make a suggestion to skeptics about what they should say about blame. Therefore, the resulting account should fit their revisionary outlook. To illustrate, consider the following statements (see also Vilhauer Reference Vilhauer2012 for a discussion of why free will–skepticism is revisionary):
This characterization [of responsibility in the basic desert sense] leaves room for an agent’s being morally responsible for an action even if she does not deserve to be blamed or praised for it—if, for example, the action is morally indifferent. Since much actual human blaming involves the basic desert presupposition, any skeptical account of blame will be revisionary. (Pereboom Reference Pereboom2014, 127–28; italics added).
Most would agree that if no one is responsible, then there are many ways in which we must act toward and conceive of one another differently, including adapting existing social practices to fit this reality. One such practice is taking reactive attitudes toward one another, in particular blaming attitudes like resentment, indignation, and guilt. (Milam Reference Milam2016, 102)
This is just one example of the practical importance of the debate, and retributivist desert moral responsibility captures it since it maintains that free will is inextricably connected to the justification of certain kinds of judgments, attitudes, and treatments that are important to criminal law, social policy, and our interpersonal relationships. (Caruso and Morris Reference Caruso and Morris2017, 844)
Since disbelief in moral responsibility would clearly have profound consequences for our interpersonal relationships, society, morality, meaning, and the law, it’s important to question whether these consequences would be (on the whole) good or bad. (Caruso Reference Caruso and Zalta2018, sec. 3)
These citations show that many important authors with skeptical leanings take skepticism to be revisionary. An account of the kind of blame skeptics should be skeptical about should make sense of this.
Third, an account of the kind of blame skeptics should be skeptical about should meet the justice constraint. It should make sense of the idea that blaming those who are not blameworthy is ceteris paribus morally problematic in the sense of unjust. To be clear, the account should not imply that blaming those who are not blameworthy is necessarily all things considered wrong. One reason for this is that sometimes there may be overwhelming forward-looking reasons for blaming those who are not blameworthy. Then, other things are not equal. The account only needs to make sense of the intuition that blaming those who are not blameworthy is unjust when other things are equal.
Again, this is an essential part of standard skeptical theories. Recall Pereboom’s characterization of blameworthiness in terms of desert. In a more recent paper, he makes it explicit that he takes the relevant notion of desert to be a moral one. He specifies that he is concerned with “what might be fair or deserved in the domain of moral responsibility: blame and praise, penalties and rewards” (Reference Pereboom2019, 261). And he contends: “I propose that, in this domain, what’s basically deserved and basically fair coincide” (262). As fairness is a moral notion, essentially connected with justice, we should conclude that Pereboom is committed to the claim that blaming those who are not blameworthy is ceteris paribus unjust.
Similarly, Gregg Caruso and Stephen Morris argue that debates about skepticism should be concerned with a notion of responsibility such that saying “that one is morally responsible for a good or bad act (in the sense relevant to free will) is to say that one justly deserves to be rewarded or punished, praised or blamed, for that act” (Caruso and Morris Reference Caruso and Morris2017, 839; italics in original). An especially clear expression of the idea that skepticism is a view about fairness and justice comes from Neil Levy:
But people are also strongly committed to the simple principle of fairness that underlies and motivates a great many of the arguments (including mine) for the conclusion that agents are not morally responsible, a principle that states that agents do not deserve to be treated differently unless there is a desert entailing difference between them. (Reference Levy2011, 9)
Thus, many authors with skeptical leanings take skepticism about blameworthiness to be the view that there is good reason to believe that it is, in our world, ceteris paribus unjust to blame people in a certain way because they are not worthy of this kind of blame (thanks to an anonymous reviewer for proposing a similar formulation). In what follows, I will make a suggestion to these authors about how to understand blame. Therefore, the picture of blame should fit their general moral outlook.
To sum up, an account of the kind of blame skeptics should be skeptical about should not make skepticism trivially false; instead, it should make sense of the idea that skepticism is revisionary regarding our everyday lives, and that it implies that blaming those who are not blameworthy is, other things being equal, unjust. In what follows I will argue that some standard accounts of blame violate at least one of these constraints.
3. What skeptics should not be skeptical about
3.a Judgment, conative, and functional views on blame
I will begin with some prominent accounts of blame, developed by nonskeptics about blameworthiness.
Some authors argue that blaming people is nothing more than holding certain judgments about them. Some say that the judgment that others have ill will or that they have committed a fault is a form of blaming them (e.g., Hieronymi Reference Hieronymi2004; Fricker Reference Fricker2016). As far as I see, no skeptical argument implies that these judgments are always incorrect.Footnote 1 Agents can have ill will and commit faults even if determinism is true and if what they do and how they are is a matter of luck. Thus, these views do not identify the kind of blame skeptics should be skeptical about.
Other views say that blame is a combination of judgments with some nonjudgmental element such as a desire (e.g., Sher Reference Sher2006, chap. 6; Arpaly and Schroeder Reference Arpaly and Schroeder2014, chap. 7), or the revision of the relationship that one has with the target of blame (e.g., Scanlon Reference Scanlon2008, chap. 4). Again, skeptical arguments should not imply that these responses are always inappropriate. If your friends repeatedly and without justification did not do what they promised to do, then you can appropriately judge that they are untrustworthy, that they lack good will, committed a fault, did something bad, or something similar. You can also appropriately desire that they had acted differently, and revise your relationship with them by, for example, giving up the intention to invite them to your birthday party. These responses seem appropriate just because of how your friends were and what they did. Considerations about luck or determinism do not change this. Skeptics should not be skeptical about the appropriateness of these attitudes. Therefore, the kind of blame skeptics should be skeptical about should not be understood in terms of them.
Some accounts analyze blame in terms of the function it fulfills. Depending on which version one accepts, a response that expresses protest (e.g., Talbert Reference Talbert2012; Smith Reference Smith, Coates and Tognazzini2013), initiates a conversation with its targets about their bad actions (e.g., McKenna Reference McKenna, Coates and Tognazzini2013), communicates that they are at fault (e.g., Fricker Reference Fricker2016), or signals commitment to norms that forbid their actions (Shoemaker and Vargas, Reference Shoemaker and Vargas2021) is a form of blame. There are many different ways to fulfill these functions. Imagine again that your friends repeatedly broke their promises to you. In a quiet moment in which you are cool, calm, and collected you tranquilly tell them: “I don’t accept that you broke the promises that you gave me, and I want to talk to you about it.” Plausibly, this expresses a kind of protest, aims at initiating a conversation, tells your friends that they were at fault, and signals a commitment to the norm to keep promises. Thus, it is a form of blame, according to these views. Is there reason to have doubts about the appropriateness of this response in a world that is ruled by determinism or luck? I don’t see one; your friends in fact broke their promises, you are affected by their conduct, and they violated norms that you accept. It seems appropriate to express this in a calm way and to have a conversation with your friends about it in a quiet moment even if the circumstances hold that typically trigger skeptical intuitions (see Pereboom Reference Pereboom and Shoemaker2017 for a similar reply).
To sum up, there is reason to doubt that typical judgment, conative, or functionalist accounts of blame identify the kind of blame skeptics should be skeptical about. They deliver a kind of blame that suggests that skepticism is trivially false. This is hardly surprising if one keeps in mind that these accounts of blame have been developed by nonskeptics about blameworthiness. Therefore, it appears fruitful to take a closer look at what those who lean toward skepticism say about blame. I will do this in the next two subsections.
3.b The punishment view
Some of those with skeptical inclinations characterize the kind of blameworthiness (or responsibility) they are concerned with in terms of appropriate punishment. Galen Strawson, for example, presents his main thesis by saying:
We are what we are, and we cannot be thought to have made ourselves in such a way that we can be held to be free in our actions in such a way that we can be held to be morally responsible for our actions in such a way that any punishment or reward for our actions is ultimately just or fair. (Reference Strawson and Watson1994, 15; all italics in original).
Thus, worries about just punishment are essential for G. Strawson’s version of skepticism. Caruso and Morris hold a similar view. They discuss Peter Strawson’s proposal to understand the sense of responsibility that is at issue in debates about (in)compatibilism in terms of appropriate emotions. Caruso and Morris reply that “it seems appropriate in the context of the free will debate to reject the understanding of moral responsibility favored by P. F. Strawson in favor of one in which the propriety of reward and punishment plays a central role” (Caruso and Morris Reference Caruso and Morris2017, 843–44).Footnote 2
These statements suggest what I call the
Punishment View: Skeptics about blameworthiness should argue that, in our world, humans are never worthy of punishment.
Let us see how this account deals with the three constraints. First, the Punishment View characterizes blame in terms of punishment and punishment necessarily involves intentionally imposing a burden on others (see Bedau and Kelly Reference Bedau and Kelly2015 for an overview). Intentionally imposing a burden on others is a paradigmatic example of something that is in need of a justification in terms of justice. For example, intentionally imposing burdens on random people is, other things being equal, unjust. This explains why the Punishment View meets the justice constraint, which says that the account of blame must make sense of the idea that blaming those who are not blameworthy is ceteris paribus unjust.
Moreover, it is not trivially true that humans in our world basically deserve that someone intentionally imposes burdens on them because they did something bad. This is a substantial claim that needs to be argued for and defended. Therefore, the Punishment View also meets the neutrality constraint.
However, this view has problems meeting the revisionary constraint. Let me elaborate (for similar replies see Smith Reference Smith and Shoemaker2019, sec. 4.3; McKenna Reference McKenna2019, 267).
Punishment is typically characterized as a response by people who intend it to be harmful and who take themselves to have the standing to respond in this way to what they perceive as someone’s objectionable behavior (see Bedau and Kelly Reference Bedau and Kelly2015). The revisionary constraint demands that an account of the kind of blame skeptics should be skeptical about should make sense of the idea that skepticism is revisionary because it implies that an important part of our everyday practice is inappropriate. Now, how often do you intend to harm people in response to what you take to be their bad conduct?Footnote 3
This is an empirical question. But I hope and believe that this intention is rarely part of my everyday life even though I believe that I blame others and myself quite regularly. I resent others, feel guilt, express protest, and so on. However, my everyday blame responses do not seem to involve the intention to harm people. At most, they involve accepting harm as a side effect. If this is so, then the Punishment View implies that I and, hopefully, many of us blame each other rarely in the sense skeptics should be skeptical about. Accepting skepticism would not imply that we need to rethink much of our everyday practice. Thus, this view has problems meeting the revisionary constraint.
Of course, the view still is revisionary insofar as it suggests that our institutional punishment systems need to be reformed. Indeed, many skeptics are concerned with our criminal justice systems (see, e.g., Kelly Reference Kelly2018; Shaw, Pereboom, and Caruso Reference Shaw, Pereboom and Caruso2019; Caruso Reference Caruso2021). However, the authors with skeptical inclinations cited in section 2 also think that skepticism is essentially revisionary with regard to our everyday lives. The revisionary constraint demands that the account of blame make sense of this idea. But if the intention to harm is not an important part of our everyday lives, then skepticism about blameworthiness understood as skepticism about deserved punishment is not important for our everyday practices.
Even though the Punishment View has problems, we can learn something positive from it about how skeptics should think about blame. The view easily meets the neutrality and the justice constraint because it depicts the kind of blame at issue as something that clearly needs a justification in terms of justice. A promising account of the kind of blame skeptics should be skeptical about may take the Punishment View as a model in this respect. In the next subsection, I will discuss a view that has problems because it does not depict the relevant kind of blame as something that clearly needs a justification in terms of justice.
3.c The emotion belief view
A famous view says that blame necessarily involves emotions such as forms of anger—especially resentment and indignation—when blaming another person, and guilt in the case of self-blame (e.g., Watson Reference Watson1987; Wallace Reference Wallace1994, chaps. 2, 3; Wolf Reference Wolf, Wallace, Kumar and Freeman2011; Menges Reference Menges2017; the inspiration is, of course, P. F. Strawson Reference Strawson and Watson1962). There are many different ways to spell out this idea. As the goal of this paper is to understand how skeptics should think about blame, I will focus on a version that has been presented by a skeptic, namely Pereboom.
Pereboom characterizes the paradigm of the kind of blame he focuses on as a complex attitude that contains at least an emotion and a belief:
[T]he attitudes of moral resentment and indignation include the following two components: anger targeted at an agent because of what he’s done or failed to do, and a belief that the agent deserves to be the target of that very anger just because of what he has done or failed to do. (Reference Pereboom2014, 128; for similar views, see Pickard Reference Pickard2013; Rosen Reference Rosen, Clarke, McKenna and Smith2015)
This suggests an account of the kind of blame skeptics should be skeptical about, namely the
Emotion Belief View: Skeptics about blameworthiness should argue that, in our world, humans are never worthy of a complex attitude that involves anger targeted at them because of what they did and how they were when they did it and the belief that they deserve to be the target of that very anger just because of what they did and how they were.
Let us see how this account deals with the three constraints. First, the Emotion Belief View seems to meet the neutrality constraint because it is an open question whether the belief component is ever correct in our world. Skeptics may contend that no human basically deserves anger in our world; nonskeptics may argue that some humans do. Therefore, the Emotion Belief View does not make skepticism trivially false.
Concerning the revisionary constraint, one may wonder how often people have the belief that others basically deserve certain emotional responses. Pereboom contends that the belief is very common (see Pereboom Reference Pereboom2014, 128). In what follows, I will grant the Emotion Belief View that this is so. Then, skepticism implies that an important part of our everyday blame practice is undermined such that the revisionary constraint is met.
The real problem for the Emotion Belief View comes with the justice constraint (I discuss another problem for Pereboom’s and similar views in Menges [Reference Menges2020]). It demands that the account of blame make sense of the idea that blaming those who are not blameworthy is ceteris paribus unjust. In order to test if the Emotion Belief View meets it, imagine the following case: your friend is angry with you and she believes that you deserve this anger because you freely and knowingly broke a promise to help her move. Thus, she blames you, according to the Emotion Belief View. But as she sits in her car and does not let you know about her anger, you do not learn anything about it. Moreover, you did not know that she was moving today and you never promised to help. Thus, you are not blameworthy. A bit later your friend realizes that you are not blameworthy because she forgot to tell you about her moving today. She is not angry with you anymore, soon forgets the whole affair, and it never plays a role in your friendship again.
An account of blame that meets the justice constraint makes sense of the idea that blaming those who are not blameworthy is ceteris paribus unjust. Intuitively, however, you are not a victim of injustice in the case sketched above. The belief is incorrect and the emotion unfitting. But, intuitively, nothing seems to be unjust here. This is a problem for the Emotion Belief View.
The intuition can be bolstered by theoretical considerations. A strong claim says that beliefs and emotions are not even among the things that can be unjust (see Sher Reference Sher2019 for a detailed discussion; see also Carlsson Reference Carlsson2017, sec. 3). This claim can be supported by the idea that only things that can harm people can be unjust. Emotions regarding others and beliefs about deserved emotions, however, never harm their targets. Expressions of these attitudes can be harmful and, thereby, unjust, but not the unexpressed attitudes. Another way to support the strong claim that emotions and beliefs cannot be unjust starts by assuming that the only point of justice is to govern how we interact with each other. Therefore, justice requirements are only relevant for interactions, but not for private attitudes. Therefore, private attitudes like unexpressed beliefs and emotions cannot violate a requirement of justice and cannot be unjust.
Even though I find these lines of reasoning plausible, they are controversial (see, e.g., Basu Reference Basu2019 for wrongful beliefs). I do not need to rely on them because there is another problem for the Emotion Belief View. It starts with the intuitive idea that not every unfitting-incorrect emotion-belief pair is unjust for the target of the attitude. Imagine that someone admires me for having written On What Matters and believes that I deserve this admiration. However, I am not Derek Parfit, the book’s author. Therefore, the emotion-belief pair is unfitting-incorrect. Perhaps, this would be unjust for Parfit. But it is not unjust for me, the target of the emotion-belief pair. Thus, there are unfitting-incorrect emotion-belief pairs that are not unjust for their targets. Proponents of the Emotion Belief View are committed to saying that anger and the belief that this anger is basically deserved is unjust when the target is not blameworthy. But why? What explains that this unfitting-incorrect emotion-belief pair is unjust, while other unfitting-incorrect emotion-belief pairs are not unjust for their targets? I do not see an answer and, as far as I know, no author has developed one. This is a problem for the Emotion Belief View.
As an intermediate conclusion, it is not intuitively plausible and there is theoretical reason to doubt that the relevant emotion-belief pairs are ceteris paribus unjust for their targets if they are not blameworthy. Thus, the view has problems meeting the justice constraint. As a reply, one may modify the view and focus on the expression of the emotion-belief pair instead of the attitude itself. The result would be the
Expression View: Skeptics about blameworthiness should argue that, in our world, humans are never worthy of expressions of the complex attitude characterized by the Emotion Belief View. (Pereboom [Reference Pereboom2014, 129] suggests such a view.)
Does the Expression View meet the justice constraint? Plausibly, some expressions of the emotion-belief pairs are unjust if their targets are not blameworthy, such as aggressively reproaching or physically punishing them. However, skeptics need an account of blame which makes sense of the idea that every instance of this kind of blame is unjust when the target is not blameworthy and when other things are equal. And there is reason to doubt that the Expression View meets this constraint. (Carlsson [Reference Carlsson2017, 99–101] develops an argument to the same conclusion.)
To see this, consider people who have learned to express their blame attitudes in neutral or productive ways. Imagine a variant of the moving case in which your friend expresses her blame attitudes toward you by producing a Jackson Pollock–style picture. Intuitively, her act of painting is in no way unjust for you.
Michael Jordan seems to be a real-life example. In interviews after his career, he says that when he felt offended by opponent players or coaches, he often “took it personal” and, as a result, played especially dominantly and successfully (see, e.g., Nonstop Sports 2020). It is plausible to understand his playing as an expression of the blame emotion-belief pair: the interviews suggest that he was angry when he felt offended and that he believed that his opponents deserved this anger. But it is not very plausible that his playing is unjust for the target of the attitude when the target is not blameworthy. You are not a victim of injustice when Jordan plays better than you even if this expresses his unfitting anger or his mistaken belief that you deserve his anger.
More generally, there are expressions of the relevant emotion-belief pairs that are not even ceteris paribus unjust for their innocent targets. But the justice constraint demands that the account should make sense of the idea that blaming those who are not blameworthy is, other things being equal, unjust.
To sum up, the Pereboom-inspired views identify the relevant kind of blame with emotion-belief pairs or with expressions of emotion-belief pairs. They have problems meeting the justice constraint. This becomes clear when one compares these views with the Punishment View. An important feature of the Punishment View is that it characterizes the relevant kind of blame as something that is clearly in need of a justification in terms of justice. However, this is not true of the Emotion Belief and the Expression View. It is far from clear that the responses characterized by these accounts are always in need of a moral justification.
Let me take a step back. An account of the kind of blame skeptics should be skeptical about should, first, deliver a picture of blame that does not make skepticism trivially false. Second, the account should make sense of the intuition that something important for our everyday lives is at stake in the discussion of skeptical arguments. Third, it should make sense of the idea that blaming those who are not blameworthy is, other things being equal, unjust. I have presented reason to doubt that some famous accounts of blame meet these constraints.Footnote 4 It seems worthwhile to look for a better alternative.
4. The Claim Forfeiture View
4.a A fresh start
Let us start by considering another version of the moving case discussed above. In this variant, your friend aggressively confronts you for breaking your promise to help her move. You tell her that you never gave a promise, but your friend does not listen and continues reproaching you.
Your friend’s conduct puts you in a special normative situation: first, you are in the position to demand that she stop responding this way. You could say: “There is absolutely no reason to talk to me like that. Stop it.” Correspondingly, your friend acquires the duty to see to it that she stops. Second, if she continues confronting you, you are in a position to demand an apology and (symbolic) compensation, and your friend acquires the corresponding duty to act accordingly. Your friend may fulfill this duty by sending you chocolate tomorrow with a note saying she feels sorry.
This web of normative relations can be summarized by saying that you have a claim against your friend that she not respond this way to what you did. Having a claim against people that they not do something just is being in a position to legitimately demand that they not do it. Moreover, having a claim against others that they not do something corresponds to their directed duty to see to it that they don’t do it. If they still do it, then you are in a position to demand an apology or (symbolic) compensation (see, e.g., Wallace Reference Wallace2019, 6–9; for an overview see Wenar Reference Wenar and Zalta2020).
Claims of this kind are conditional. Imagine that crazy kidnappers told your friend sincerely that they will kill her children if she does not aggressively confront you for breaking a promise. Intuitively, your friend’s response is justified in this case because your claim is overridden (e.g., Thomson Reference Thomson1976, 10–11) or because this constellation explains that you did not have the claim in the first place (e.g., Wallace Reference Wallace2019, 201). You would then acquire secondary claims against your friend that she later explain herself. But her aggressive confrontation does not wrong you.
You may also waive your claim. If you tell your friend in advance that she may confront you as much as she wants if this makes her feel better after a stressful day, you thereby give up your claim against this response. Then, her confrontation does not wrong you.
Intuitively, your claim against being the target of such a response can also be forfeited. Imagine a variation of the case in which you did promise that you would help her move, that there is no justification for your not helping, you knew exactly what you were doing, you were not under severe stress, on drugs, directly manipulated, and so on, and you decided to stay on the couch out of your own will. Then, it seems you are not in a position to demand that your friend stop confronting you, your friend does not have the duty to see to it that she stops, and you cannot reasonably demand an apology or compensation. Your claim seems to be extinguished because of your morally problematic conduct and how you were when you performed it.
What has been said so far generalizes. We seem to have standing—though in most cases overridable and waivable—claims against others that they not respond to what we do in ways that are nontrivially negative for us.Footnote 5 It seems that some of these claims can be forfeited. Facts about our objectionable conduct and about how we are when we perform it sometimes seem to explain that claims against negative responses can be extinguished. Then, other things being equal, others do not wrong us when they respond in this way toward our deeds.
All this should be relatively uncontroversial if one accepts claim talk at all. In what follows, I will argue that this is all we need to account for the kind of blame skeptics should be skeptical about.
4.b The main idea
The main idea is that the kind of blame skeptics should be skeptical about is to be analyzed in terms of claim forfeiture. More specifically, I propose the Claim Forfeiture View on Blame or, for short:
CFB: Skeptics about blameworthiness should argue that, in our world, humans are never worthy of responses that
-
(i) would violate a claim of theirs if it is not waived, overridden, or forfeited and
-
(ii) involve the thought that they forfeited the claim against this response because they did something morally objectionable and because of how they were when they did it.
It is important to note that there may be instances of blame that do not fulfill the two conditions. Proponents of CFB would then say that skeptics do not need to be skeptical about them.
Let me take a close look at the two conditions of CFB. The first is open insofar as many different responses can violate a claim of others. Correspondingly, many different responses are such that skeptics should be skeptical about them. These responses have in common that they nontrivially negatively affect their targets. There is an intense debate about the justification and function of claims (see Wenar Reference Wenar and Zalta2020 for an overview). But all parties seem to agree that an agent only has a claim against some x if x nontrivially affects the agent. When I walk down a street, I do not have a claim against you that you get out of my way even though you getting out of my way is in my interest (I would not need to take the additional step to go around you) and even if you not getting out of my way diminishes my freedom (I cannot go straight). It may be nice of you to get out of my way. But you not getting out of my way affects my interests and freedom too insignificantly to ground a claim of mine.
When does something affect us nontrivially? I do not have an answer. But to avoid misunderstandings, CFB is not committed to the idea that this something must have the potential to cause nontrivial harm. Suppose I am asleep and a stranger carefully touches me without waking me up. Perhaps I have a claim against the stranger’s touching me even if it does not nontrivially harm me. One could say that the touching would nontrivially affect my physical integrity or personal autonomy which grounds the claim.
That claims are only concerned with nontrivial aspects of our lives has implications for CFB. This view identifies only those responses as falling under the kind of blame skeptics should be skeptical about that can violate claims. As claims are only concerned with things that significantly affect the claim holder, CFB says that only such responses can fall under the kind of blame skeptics should be skeptical about.
The second condition identifies a mental aspect of blamers. CFB requires that they think that their targets forfeited a claim because they did something morally objectionable and because of how they were when they did it. This thought does not need to be a deliberate, explicit, or fully endorsed judgment. Typical theories of the emotions, for example, say that emotions involve appraising an object (for an overview, see Scarantino and de Sousa Reference Scarantino, de Sousa and Zalta2018). This appraisal should not be understood as a deliberate, explicit, or fully endorsed judgment, but as a less demanding kind of representation. Condition (ii) of CFB should be understood analogously.
Whether blamers fulfill condition (ii) is an empirical question. But it seems plausible that the thought is common. We often need to ascribe such a thought to blamers in order to explain their behavior. To illustrate, imagine that Sirius confronts Peter for cowardly betraying their common friend. Peter replies: “Stop talking to me in this way!” Plausibly, Sirius would blame Peter even more aggressively. A good explanation for this is that Sirius thinks that Peter is not in the position to demand that he stop. Now imagine that Sirius continues blaming Peter, who then says: “You should say you’re sorry for being so rude.” Again, this would probably intensify Sirius’s blame. A straightforward explanation for this is that Sirius thinks of Peter as not being in the position to demand apologies or symbolic compensations for being confronted in this way.
More generally, explaining Sirius’s behavior involves ascribing the thought to him that his response to Peter’s betrayal does not unjustifiably violate one of Peter’s claims. This may be so because the response is trivial and not in the business of claim violations or because the response is nontrivial but the claim is overridden, or Peter has waived or forfeited it. As long as Sirius has no reason to think that his confrontation is trivial, that Peter has waived his claim, or that responding in this way has very good consequences, it seems most plausible to ascribe the thought to Sirius that Peter has forfeited his claim because his betrayal was morally objectionable and because of how Peter was when he did it.
My empirical speculation is that many everyday blame responses are relevantly similar. In order to make sense of how blamers behave we often need to ascribe the—typically nonexplicit—thought to them that the targets of their responses forfeited their claim against these responses because of their morally objectionable actions and how they were when they performed them.
Let me illustrate CFB with some examples. Take the case in which your friend has repeatedly and without justification broken their promise to you. After the last incident, you step on their foot only with the intention to hurt them. Intuitively, we all have a standing conditional claim against others that they not hurt us by stepping on our feet. In this situation, you think that your friend has forfeited this claim because they have lied too often to you and because of how they were when they did it—because they knew what they were doing, they were not under severe stress, and so on. According to CFB, your stepping on their foot belongs to the kind of blame skeptics should be skeptical about. For this to be so, it is irrelevant whether you feel or express an emotion when you step on their foot. This is one respect in which CFB differs from the Emotion Belief View discussed above.
Imagine, second, a similar situation in which you do not step on your friend’s foot but aggressively reproach them: “Who do you think you are that you can treat me like this?” Intuitively, we have a standing conditional claim against others that they not aggressively confront us in such a way for what we do. In this case, you think that your friend has forfeited this claim against you because of their morally objectionable breaking of promises and because of how they were when they did it. Therefore, CFB says that your aggressive reproach falls under the kind of blame skeptics should be skeptical about. For this to be true, you do not need to intend to harm your friend. Perhaps you only accept harming them as a side effect of expressing that you do not accept their behavior. This is one respect in which CFB differs from the Punishment View that requires that blamers intend to harm the blamees.
Finally, imagine a variant of the case in which you form the judgment that your friend lacks good will toward you and does not behave the way a friend should. You desire that they had acted differently and revise your intention to invite them to your birthday party. Moreover, you wait for a quiet moment in order to tell them calmly that you do not accept this behavior and that you want to talk to them about it. Standard protest, judgment, conative, and functional views imply that this is a form of blame. Importantly, proponents of CFB can accept this. However, CFB suggests that this response does not belong to the kind of blame skeptics should be skeptical about. This is because we plausibly do not have standing claims against others that they not respond in these ways to us. For example, we have no claim against others that they invite us to their parties when other things are equal. Therefore, CFB says that skeptics should not be skeptical about such a response.
To sum up, the basic idea of CFB is to account for the kind of blame skeptics should be skeptical about in terms of claim forfeiture. What should be at issue between skeptics and nonskeptics are responses that can violate claims of the targets and that involve the thought that the targets have forfeited their claims because of what they did and how they were when they did it.
4.c Testing the Claim Forfeiture View
Let us see if CFB meets the constraints for an account of the kind of blame skeptics should be skeptical about.
The neutrality constraint demands that such an account should not make skepticism trivially false. One may worry that combining CFB with skepticism has implausible implications in other domains, for example with regard to self-defense. When people unjustifiably attack you then it seems that they forfeit their claim against your using proportionate harm to defend yourself. Combining CFB with skepticism, one may worry, implies that using harm in self-defense always infringes a claim because the unjust attackers never forfeit their claim against it. The worry concludes that this would make skepticism obviously false such that CFB violates the neutrality constraint.
Note first that many instances of self-defense can be justified even if those who pose the threat have not forfeited their claim. This justification points to the bad consequences one avoids by using self-defense (e.g., McMahan Reference McMahan2005, 386). Thus, combining CFB with skepticism leaves room for justified self-defense.
Moreover, according to one picture, attackers forfeit claims against proportionate harm in self-defense just because they pose an unjustified threat (e.g., Thomson Reference Thomson1991). For this to be true, it does not matter how they are when they pose the threat, for example what control they had. Combining CFB with skepticism leaves room for this. These views together imply that agents never forfeit a claim against negative responses because of what they did and because of how they were when they did it. The way an agent is plays an essential role in explaining the kind of claim forfeiture that is crucial for CFB, but it plays no role in the kind of claim forfeiture that is relevant for self-defense, according to the view under consideration. Thus, even if no human has the properties that ground the first kind of claim forfeiture, unjustified attackers may still forfeit their claim against proportionate harm.
More generally, I see no reason to doubt that CFB meets the neutrality constraint. It is an open question whether, in our world, humans forfeit their claims against negative responses just because of what they did and how they were when they did it.
The revisionary constraint demands that the account should make sense of the idea that something important is at play because skeptical arguments imply that a significant part of our everyday practice is inappropriate. The cases discussed above suggest that our everyday practice involves many responses that fulfill the conditions identified by CFB. Recall your friends’ aggressively confronting you for breaking a promise and Sirius’s harshly reproaching Peter for betraying their friend. As these responses plausibly fall under CFB, many of our everyday responses fall under it too. If the skeptical arguments are sound, then these responses would be inappropriate.
Here it becomes clear that CFB has an advantage over the Punishment View. The latter has problems meeting the revisionary constraint because the kind of blame depicted by this view does not seem to play an important role in our everyday lives. CFB fares better in this respect.
The justice constraint demands that the account should make sense of the idea that blaming those who are not blameworthy is ceteris paribus unjust. CFB meets this constraint because, just as the Punishment View, it characterizes the relevant kind of blame as something that clearly needs a justification in terms of justice. It says that the relevant responses violate claims if they are not waived, overridden, or forfeited. A moral justification of such a response is called for. And violating an innocent’s claim is clearly unjust when other things are equal.
Recall that the Emotion Belief View has problems meeting the justice constraint because there is good reason to doubt that having or expressing the relevant emotion-belief pair is ceteris paribus unjust if the target is not blameworthy. Therefore, CFB has an advantage over the Emotion Belief View.
Let me come to a possible independent objection. Recall that CFB does not need to identify every blame response as such. However, some may worry that CFB identifies some responses as blame responses that, intuitively, are not.
Imagine that Bellatrix kills Sirius because she does not like him. Some may worry that CFB implies that Bellatrix’s killing is a form of blaming Sirius. But this is not so. Her killing him fulfills condition (i) of CFB because it violates Sirius’s claims if he has not waived or forfeited them and if they are not overridden. However, her killing Sirius, plausibly, does not fulfill condition (ii), which holds that responders think that their targets have forfeited their claim because of their morally objectionable conduct and because of how they were when they performed it. There is no reason to ascribe this thought to Bellatrix when she kills Sirius because she does not like him. Therefore, CFB does not imply that this and similar potential wrongings are blame responses.
Some may object that CFB implausibly implies that every instance of self-defense is a form of blame. My response is, again, that there are instances of self-defense that do not involve the thought that the targets forfeited their claims because of something morally objectionable and because of how they were. Take the sad case of child soldiers. Imagine that the only way to protect your children from being unjustifiably killed by child soldiers is to harm the child soldiers. When you do it, you may think that because the soldiers are children, brainwashed, or on drugs, no aspect of the way they are explains their claim forfeiture. Rather, you may think that the fact that they pose an unjustifiable threat to your children’s lives explains that your using proportionate harm is justified (I am not claiming that your using harm in this case is justified; I imagine that you think that it is justified). Then your harming them is a form of self-defense, but it does not fulfill condition (ii) of CFB. Thus, CFB does not imply that every instance of self-defense is a blame response.
This reply can be generalized. Many seeming false positives of CFB do not fulfill condition (ii). And the responses that fulfill conditions (i) and (ii) are not false positives.
To sum up, CFB meets the three constraints for an account of the kind of blame skeptics should be skeptical about. Therefore, it has important advantages over the other views discussed above. Moreover, CFB can be defended against the objection that it implies false positives.
4.d Being skeptical
An especially interesting upshot of CFB is that it frames the debate about skepticism in a new way. It proposes that the core question between skeptics and nonskeptics should be: When humans perform a misconduct in our world, do at least some of them forfeit their claim against a negative response just because of what they did and how they were when they did it?
Take the case in which you broke your promise to help your friend move out of your own will, without justification, and with full knowledge. From an everyday perspective, it seems plausible to think that some of these facts about how you were when you broke the promise (partly) explain why you do not have a claim against your friend’s confronting you. According to CFB, skeptics should doubt that this is so. Skeptics should argue that because of determinism or luck, human misconduct and the way humans are when they perform it never explains why they forfeit claims. In our world, humans do not have the properties that could play this explanatory role.
Skeptics should continue, stating that blaming humans in the way identified by CFB is problematic in two different ways. First, every instance of this kind of blame involves the incorrect thought that the way the targets were when they did something morally objectionable (partly) explains why they forfeit a claim. Second, many instances of this kind of blame wrong their targets. If the targets have not waived their claim and if it is not overridden, then the responses unjustifiably violate the claim. Thus, skeptics conclude, this blame practice is epistemically and morally problematic. Nonskeptics should deny this. They should argue that our world is such that at least some humans forfeit a claim against negative responses because of what they do and how they are when they do it.
As claim forfeiture plays no role in current discussions about skepticism, CFB provides a new way of framing the debate. From this perspective, manipulation, rollback, replication, and luck arguments try to show that, under certain conditions, agents do not forfeit their claims because of their objectionable conduct and how they were when they performed it. Spelling out the implications of CFB for the arguments for and against skepticism is a job for future research.
To sum up, CFB accounts for the kind of blame skeptics should be skeptical about in terms of claim forfeiture. To blame people in this sense is to respond to what they did in a way that can violate a claim of theirs and to think that they have forfeited their claim because they did something morally objectionable and because of how they were when they did it. This view frames the debate about skepticism in a new way.
5. Conclusion
A significant problem for the discussion about whether humans are blameworthy for their deeds is that it is unclear how to understand the kind of blame the discussion should be concerned with. Some accounts of blame make it too plausible that humans are blameworthy without dealing with skeptical arguments. Some accounts depict a blame practice that is not an essential part of our lives such that skepticism loses much of its bite. This paper makes a new proposal. The basic idea is that the kind of blame skeptics should be skeptical about is constituted by responses that can violate the targets’ claims and by the responders’ thought that the targets have forfeited these claims because of their morally objectionable actions and because of how they were when they performed them. This view identifies an important part of our everyday lives and frames discussions about whether humans are blameworthy in a new way.
Acknowledgments
I’m very grateful for helpful comments on earlier versions of the paper by Andreas Brekke Carlsson, Simon-Pierre Chevarie-Cossette, Nora Heinzelmann, Max Kiener, Erasmus Mayr, Stefan Riedener, Daniel Telech, an anonymous reviewer for the Australasian Journal of Philosophy, and an anonymous reviewer for this journal. I discussed ideas from the paper online “in” Vienna in November 2020, “in” London with the team of Roots of Responsibility in March 2021, and “in” Erlangen in April 2021. I wish to thank the audiences for comments and suggestions. Thanks also to Claire Davis for proofreading.
Funding Statement
Work on this paper was supported by the Austrian Science Fund (FWF): P 34851-G and is part of the research project The Sense of Responsibility Worth Worrying About.
Leonhard Menges is assistant professor of practical philosophy at the University of Salzburg.
Open access
