Debt and Desert

According to what may be called the Debt Model, blameworthiness is defined in terms of deserved suffering. The Debt Model has a significant implication: one is less blameworthy if one has experienced some of the suffering one deserves, and no longer blameworthy once one has experienced the full amount of suffering one deserves. Blameworthiness, according to the Debt model, is not forever. In recent papers, Clarke (2022) and Howard (2022) independently criticize the Debt Mod - el and argue for the opposite conclusion: if one is blameworthy, one will remain blameworthy forever. In this paper, I respond to this criticism as well as a recent argument against the Debt Model from Tierney (2022). I then present a prima facie case for the Debt Model. I argue that Clarke’s attempt to accommodate the intu - itions elicited by this and similar cases has unwelcome implications. I will end the paper with some remarks about what roles we want our conceptions of blamewor - thiness to play. If Clarke’s account were right, the fact that an agent is blameworthy would play a far less significant role in our moral life than we often tend to assume. The Debt Model, I will suggest, makes better sense of our blaming practices.

Blameworthiness is sometimes defined in terms of deserved suffering.According to Gideon Rosen (2015), an agent is blameworthy just in case he deserves to suffer for what he has done (Rosen 2015).On other views an agent is blameworthy just in case he deserves to feel guilty -to a certain degree, for a certain duration, where the feeling of guilt is in part constituted by suffering (Carlsson 2017, 2019, 2022, Port-more 2019a, b, 2022). 1 Such views have a significant implication.When we deserve some suffering, it is of a certain amount.After all, if it is at all plausible that we can deserve to suffer, the suffering needs to be proportional to our wrongdoing.If 5 years imprisonment is what I deserve for my crime, I have gotten what I deserved after 5 years in prison and I no longer deserve to be in prison.Similarly, if an agent deserves to feel a certain amount of guilt for his offense, he no longer deserves to feel guilt once he has experienced the amount of guilt that he deserves.If blameworthiness is defined in terms of deserved suffering, it follows that one is less blameworthy if one has experienced some of the suffering one deserves, and no longer blameworthy once one has experienced the full amount of suffering one deserves.Blameworthinessis not forever. 2On such views, deserved suffering is akin to a debt that the wrongdoer needs to pay down for blameworthiness to diminish or disappear.It will be useful to have a label; so let us say that these views are examples of a Debt Model. 3  In two recent papers, Clarke (2022) and Howard (2022) argue for the opposite conclusion: if one is blameworthy, one will remain blameworthy forever. 4Clarke calls this view PERMANENT.The aim of this paper is to defend the Debt Model and to provide some arguments against PERMANENT.The paper is structured as follows.In Sect. 1, I present the Debt Model.In Sect. 2 I respond to recent arguments against the Debt Model from Howard (2022) and Tierney (2022).In Sects. 3 and 4, I discuss Clarke's positive and negative arguments for PERMANENT.Section 5 provides a case against PERMANENT.Sections 6 and 7 look at Clarke's attempt to accommodate the intuitions elicited by cases like the one presented in Sect. 5 and argue that this solution has some unwelcome implications.It will become apparent that there is more substantial agreement between Clarke's account and the Debt Model than one might expect.One might worry that the disagreement just stems from different conceptions of blameworthiness.Considering this worry, I will end the paper with some remarks about what roles we want such conceptions to play.If Clarke's account were right, the fact that an agent is blameworthy would play a far less significant role in our moral life than we often tend to assume.The Debt Model, I will suggest, makes better sense of our blaming practices.

Debt Models
The connection between moral wrongdoing and incurring a moral debt is both familiar and complex.As Linda Radzik points out, both religious and secular ideas of atonement are rooted in the idea that "to wrong another person is to incur a moral debt, and so a way correct the wrongdoing is to repay that debt." (2009: 25). 5 According to Nelkin (2013b) to forgive a culpable wrongdoer is to release her from the debt she incurred by acting wrongly.The idea of a debt is also often taken to be an important part of the phenomenology of guilt.Herbert Morris writes: "When we think of what it is to feel guilty…we think not only of painful feelings but of something that is owed; and pain is somehow connected with paying what one owes" (1976: 90). 6ne specific way of developing this theme is by considering what it is to be blameworthy.A complete theory of blameworthiness must answer two questions: what makes an agent blameworthy at the time of the action and what makes an agent blameworthy after the action?The first question is about synchronic blameworthiness, or what it takes to become blameworthy.The second is about diachronic blameworthiness, or what it takes to remain blameworthy (Matheson 2024: 2-3).According to some philosophers the answer to these two questions is the same.If the agent satisfies the conditions that make her blameworthy at the time of the action, she will necessarily remain blameworthy after the action.Blameworthiness, on this view, is interminable.On other views, the replies will diverge; it is possible for diachronic blameworthiness to diminish or cease depending on what happens after the culpable wrongdoing.Whether an agent was blameworthy at the time of the action will depend on whether she acted wrongly, with control and satisfied the epistemic conditions.Whether an agent is blameworthy at a later time will of course depend on whether the agent was blameworthy at the time of action.But it will also, on these views, depend on other factors.
One such factor could be psychological connectedness.Khoury and Matheson (2018) argue that a person who was synchronically blameworthy for a past misdeed is not diachronically blameworthy if he no longer bears any psychological connections to his past self who committed the wrongful action.
A Debt Model of blameworthiness, by contrast, says that whether an agent is diachronically blameworthy will depend on whether and to what extent she has repaid the moral debt that she incurred by her culpable wrongdoing.Stated in such general terms, the model leaves it open what the moral debt consists in and how it should be repaid.Different theories will fill out these details in different ways.In this paper I will focus on accounts that take the wrongdoer's suffering of guilt to be a central part of what needs to be repaid for the agent's diachronic blameworthiness to change.
In order to introduce these views, which I will simply call the Debt Model from now, it is useful to start with the definition of blameworthiness.To be blameworthy is normally defined in terms of the following biconditional: An agent S is blameworthy for X if and only if it is appropriate to blame S for X.
For this generic biconditional to be informative, we need to specify what we mean by "blame" and "appropriate".According to Rosen (2015) and Portmore (2019bPortmore ( , 2022) ) an agent S is blameworthy for X just in case it is fitting to blame S for X.Fittingness, on their accounts, is understood in terms of accurate representations. 7An emotion is thus fitting just in case what that emotion represents is accurate or correct.Both take deserved suffering to be a part of what the blaming emotions represent.On Rosen's view, one of the constitutive thoughts of the blaming emotions, which he identifies with resentment, is what he calls "the retributive thought": the wrongdoer deserves to suffer for what he has done (Rosen 2015: 83).An implication of this view is that once the wrongdoer has suffered what she deserved to suffer, resentment will no longer be fitting, and the wrongdoer will thus no longer be blameworthy.On Portmore's account, a part of the representational content of blame is that the wrongdoer has "not suffered all the guilt, regret, and remorse that she deserves to suffer in recognition that she has violated [a] legitimate demand' ' (2022: 50).Once the wrongdoer has suffered all the guilt, regret, and remorse she deserved, this representation will no longer be accurate, and blame will thus no longer be fitting.As a result, the wrongdoer is no longer blameworthy.My own account (2017,2019,2022) differs from Rosen's and Portmore's in two respects.First, instead of focusing on otherdirected blame, I take self-directed blame to be the fundamental notion in the definition of blameworthiness.To blame oneself, in my view, is to feel guilty.Second, I understand appropriateness in terms of desert rather than fittingness.On this view, desert is not part of what the blaming emotion represents but is rather the relevant norm of propriety.On this account, an agent S is blameworthy for X if and only if she deserves to feel guilty for X (Carlsson 2017(Carlsson , 2019(Carlsson , 2022)).It follows that if an agent who did deserve to feel guilty for some reason no longer deserves to feel guilty, she is no longer blameworthy.Moreover, if someone has come to deserve less guilt than she initially did, she will have become less blameworthy than she initially was.
There are differences between these views, but they also share important similarities.On these views blameworthiness is desert-entailing.And what the blameworthy agent deserves is a form of suffering.But the amount of suffering an agent deserves is not constant.It can be influenced by what the wrongdoer feels or does.These views thus share a crucial implication: Blameworthiness is not forever.It can diminish and be extinguished.Of course, this is compatible with the fact that the wrongdoer was blameworthy. 8Nor does it mean that blameworthiness has to be extinguished.Both Portmore (2022: 71, note 30) andI (2022: 194) allow for the possibility that some wrongs are so serious that it would be impossible for the wrongdoer to experience the amount of guilt he deserved.
7 This is by no means the only way of understanding fittingness.For a general overview, see Howard (2018).For discussions of fittingness related to blameworthiness specifically, see Macnamara (2020); Clarke and Rawling (2022b);McKenna (2022b).Clarke and Rawling (2022a) argue against the alethic view of fittingness, according to which is understood in terms of true representations. 8Both proponents of the Debt model and proponents of PERMANENT can agree that there is one sense in which blameworthiness is permanent: no one can change that fact that an agent was blameworthy at a certain time.The debate is about whether an agent who was blameworthy at a certain time will remain blameworthy (to the same degree).See Carlsson (2022), Tierney (2022) and Matheson (2024).

Wrong Extension Arguments
According to the Debt Model, a blameworthy agent will deserve a certain amount of guilt.It might be natural to understand amounts of guilt as we understand amounts of pain: as a multiplicative function of intensity and duration, such that one could experience the guilt one deserves either by experiencing a more intense feeling of guilt for a short duration or a less intense feeling for a longer duration.But as Howard (2022: 6) notes, this interpretation has counterintuitive implications.First, someone who did something seriously wrong might experience a spike of guilt -a feeling of guilt so intense that all the guilt he deserved is exhausted over the course of a few hours.Second, someone who unjustly killed an innocent person might experience guilt of the intensity one might expect after stealing a candy bar.However, if he experiences this low-grade emotion for long enough, it seems that he will eventually have experienced all the guilt he deserved.Howard takes these examples to show that guilt remains fitting even after people have experienced the amount of guilt they deserve.As a result, they remain blameworthy, even though they no longer deserve to feel guilty. 9This would mean that the Debt Model is mistaken.
Howard's interpretation of these cases, however, has an implausible implication.Consider the person who experienced a spike of guilt, so that he, per stipulation, experienced all the guilt he deserved to feel for a serious wrongdoing over the course of a few hours.On this interpretation, the justification for his feeling of guilt changes from desert to fittingness.Before the spike he deserves to feel guilty; afterwards he no longer deserves to feel guilty, but it is nevertheless fitting that he feels guilt.Although clearly related, fittingness and desert are not identical relations. 10There are different accounts of what distinguishes fittingness from desert.Clarke (2016) argues that desert but not fittingness is a consideration of justice.Others have argued that we have pro tanto reason to induce the deserved state (like guilt) in the deserving but not to induce fitting states (like grief) (Pereboom 2014, 2021, Carlsson 2017).McKenna (2022) argues that desert, but not fittingness, discounts, or silences -and not merely outweighs -moral reasons that may otherwise apply.These distinctions are of course not mutually exclusive.If Howard were right, we would expect that some of these hallmarks of desert were present before agent got what he deserved and that they would disappear afterwards.But that does not seem to be the case: the kind and strength of the wrongdoer's reason to feel guilty appears to be the same both before and after the spike of guilt and the long duration of low-intensity guilt.For example, it still seems just and non-instrumentally good that the wrongdoer experiences guilt after the spike.This suggests an alternative and more appealing explanation of the cases: these two agents did not get what they deserve, and this explains why we believe they remain blameworthy even after the spike and the long duration of lowintensity guilt.
If so, we must understand the amount of guilt wrongdoers deserve differently than the way Howard suggests.Proponents of the Debt Model argue that agents are blameworthy just in case they deserve to feel guilty-to a certain degree, for a certain duration.They emphasize that to feel guilty is to painfully acknowledge one's wrongdoing (Carlsson 2017;Portmore 2019a; see also Clarke 2013;2016 andMacnamara 2020).What does this mean?Guilt is painful, but what the wrongdoer is pained by is something specific: the wrong one has committed and its moral significance.What one deserves, on this picture, is neither merely coming to believe that one has acted wrongly and that one's wrongdoing carries a certain moral significance, nor it is merely to feel a certain amount of pain.It is rather to painfully appreciate one's culpability as well as the moral significance of one's act by experiencing the emotion of guilt (Portmore ms).This painful emotional appreciation involves thoughts (about what one could and should have done differently); patterns of attention (focused on the victim and the effects of one's wrongdoing), and motivational tendencies (to apologize, confess and make amends) (Macnamara 2020;Portmore Ms;Achs 2023).
Guilt, like other emotions, is thus a distinct mode of apprehending or appreciating value or disvalue.11For an emotion to be an appropriate appreciation of some value or disvalue it need to be proportionate to the evaluative features of its object, or in the words of D' Arms and Jacobson (2000: 73), it needs to have the right "size".Someone experiencing extreme fear of their neighbour's moderately dangerous dog would not appropriately appreciate its dangerousness.Nor can the emotion's short duration compensate for its disproportionate size.Suppose that the dog would remain moderately dangerous for the next six months.The fact that the extreme fear only lasted a short while, does not make it any more appropriate as an appreciation of its dangerousness.
We can now see why Howard's counterexamples don't work: they are not illustrations of blameworthy wrongdoers getting what they deserve.What the blameworthy deserve, according to the Debt Model, is the painful appreciation of one's culpability and the moral significance of one's action.An accurate appreciation requires that the emotion is proportionate to one's wrongdoing.This is one important respect in which guilt differs from mere pain, and which makes it implausible to understand the amount of guilt in terms of multiplicative functions of intensity and duration.If a murderer feels the intensity of guilt one might expect after stealing a candy bar for a very long time he is not experiencing the guilt he deserves because this guilt does not amount to a proper appreciation of his culpability and the moral significance of his wrongdoing.The fact that he experiences it for a long time does not make that emotional reaction any more deserved.Similarly, if one feels an extremely intense but very brief spike of guilt for a serious wrongdoing, it would not amount to an appropriate appreciation of one's wrongdoing.This will take time and cannot be achieved in a matter of hours.Moreover, even in the time one experiences it, the emotion would have the wrong size; it would be disproportionate to the disvalue of one's wrongdoing.Tierney (2022) agrees with the Debt Model that blameworthiness is not forever.But like Howard, she thinks that the Debt Model has counterintuitive implications.
To illustrate, she asks us to consider Jackie who promises to take care for Edie's beloved houseplants, but who overwaters them because she fails to read Edie's carefully prepared instructions.Jackie realizes her error and feels incredibly guilty for being so careless.When Edie comes home to their withered flowers, they blame Jackie, and this blame seems appropriate.12Because, Tierney stipulates, Jackie has experienced the exact amount of guilt she deserves to feel, she is not blameworthy at the time when Edie blames her, according to the Debt Model.But Edie's blame seems appropriate, both to Edie and to Jackie.Something has gone wrong.
The problem, according to Tierney, is that the Debt Model only focuses "on what occurs inside the blameworthy agents' heads." (2022: 391).She notes that "acknowledgement, apologies and restitution are all things we typically think blameworthy agents owe to those they have wronged.
[…] And just as many theorists take blameworthy agents to owe something to their victims, they also take blame to express a demand or call for these obligations to be fulfilled.But these familiar features of both blameworthiness and blame are missing from [the Debt Model]" (ibid.). 13, 14  I believe that this objection fails to appreciate the resources of the Debt Model.Both Portmore and I explicitly deny that diachronic blameworthiness only depends on what goes on inside the wrongdoer's head: Plausibly, when one acts wrongfully, with knowledge and control, one will not only deserve to feel guilty, but also incur certain duties towards one's victim.These can come in various forms, but at the very least it seems that such wrongdoers often will have a duty to apologize, compensate and make amends.It seems plausible that an agent will continue to deserve to feel guilty for his wrongful action until such duties are fulfilled.(Carlsson 2022: 195).
Similarly, Portmore (2022: 69-70) asks us to compare two agents, one of which has experienced guilt, but who also has done much to make amends (apologizing, paying reparations), and one of which who has done neither.Portmore argues that the second wrongdoer is more blameworthy than the former, because she deserves more guilt.
According to the Debt Model, an agent's blameworthiness at a given time is determined by how much guilt she deserves to feel at that time.But as the above quotes bring out, the amount of guilt one deserves to experience can be influenced by several factors, including the fulfilling of one's reparative duties.Blameworthiness, on these views, is not just about what goes on in the blameworthy agent's head.Moreover, this is not an ad hoc amendment to the Debt Model.To feel guilty, according to the Debt Model, is to blame oneself. 15Whether and to what degree one deserves to blame oneself will plausibly depend not only on how much one has blamed oneself, but also on whether one has fulfilled one's reparative duties. 16The Debt Model does not imply that experiencing guilt is the only route to becoming less blameworthy.Exactly how the experience of guilt should be weighed against fulfilling reparative duties in determining how much self-blame the agent deserves is a complicated issue, beyond the scope of this paper.But it seems clear that fulfilling reparative duties, pace Tierney, can play an important role for how the Debt Model understands diachronic blameworthiness.
Let us now return to Jackie and Edie.If, as Tierney stipulates, Jackie has experienced all the guilt she deserves, it seems to follow that it would no longer be appropriate for Jackie to blame herself.Edie's blame on the other hand is appropriate.But this would be a strange asymmetry which is not borne out by the description of the case.When we accept the blame of others, apologize, and offer reparations, we typically also think that we deserve to blame ourselves.The natural conclusion to draw from the description of this case would therefore be that Jackie still deserves to blame herself, i.e. to feel guilty for what she has done, in part because she hasn't yet fulfilled her reparative duties. 17The Debt Model, I have argued, can accommodate this insight.
3 The Prima Facie Case for PERMANENT Clarke (2022) defends PERMANENT, the view that if one becomes blameworthy, one will remain blameworthy forever.In this section I will discuss his positive argument for that view.Clarke understands blameworthiness as follows: An agent S is blameworthy for X just in case some possible instance of blame of S for X, by someone, would be fitting (Clarke 2022(Clarke : 2580)).
According to Clarke, blame is a stance of holding the wrongdoing against the wrongdoer.This might take the form of an emotional reaction, like resentment or indignation, but it need not.Blame is multifarious, but it is important to note that for Clarke, blame is something more than a judgment of blameworthiness.Clarke argues for PERMANENT: Once you are blameworthy for something, you are always blameworthy for it.Even if blame by this or that person can cease to be appropriate-perhaps because that person has blamed you enough-and even if, at some point, all things considered no one should blame you any longer, you remain worthy of blame; some possible instance of blame of you, by someone, for that offense would be fitting (2022: 2579-2580).
Clarke's positive argument for PERMANENT is as follows: If one is guilty of a moral offense, one remains so without end […] If one is guilty of a moral offense, then one is culpable for it.And one who is culpable for an offense is to blame for it.To be to blame for something is to be worthy of blame, or blameworthy, for it.Hence, blameworthiness is forever: once blameworthy for an offense, always blameworthy for that offense.And since one is worthy of blame just in case blame would be fitting, we may add that once one is blameworthy for an offense, it will always be the case that some possible instance of blame of one for that offense would be fitting.This I take to be a prima facie case for PERMANENT.(Clarke 2022(Clarke : 2582) ) This argument relies on conceptual entailments between the notions of "being guilty", "culpability", and "blameworthiness".Given this argumentative structure, it's important to know exactly what "being guilty" means in this context.Clarke does not define it.Suppose first that it is used as a synonym for "blameworthiness".If this is how we understand being guilty, the argument would be question-begging.Whether blameworthiness is forever, after all, is exactly what this debate is about.As Clarke himself notes, it is not trivially true that blameworthiness is forever, so if being guilty simply means being blameworthy, it cannot be trivially true that being guilty is forever either.If being guilty is not a synonym for blameworthiness, we have several options.Legal guilt is determined by a wrongful act (actus reus) and a guilty mind (mens rea).Analogously, we could perhaps understand being guilty as meaning that the agent has acted wrongly with a sufficient degree of control, knowledge, and freedom.When this fact obtains, it is true forever.Hence, on this reading, once an agent is guilty, she is guilty forever.But it does not follow that the agent is blameworthy forever.Presumably, an agent is blameworthy (now) if it is appropriate that she is blamed for her action (now). 18Participants of this debate all agree that it was appropriate to blame the wrongdoer at some point.The question is whether it still is.However, the appropriateness of blame now does not simply follow from the fact that the agent acted wrongly with a sufficient degree of control, knowledge, and freedom at some point in the past.After all, it might be that the wrongdoer experienced guilt and atoned for his action, which might render it inappropriate to blame him, although it was appropriate before.Again, this is exactly what the debate is about.Alternatively, we can understand being guilty as the claim that it was appropriate to blame the agent at some point t.But again, this does not mean that it is appropriate to do so now.For Clarke's argument to be successful, the concepts "being guilty", "culpability", and "being blameworthy" need to entail each other.However, they only do so if they are defined in a way that presupposes what the debate is about.Because of this, Clarke's argument does not provide a prima facie case for PERMANENT. 19  4 The Living and the Dead Clarke argues that the Debt Model will struggle to make sense of the blameworthiness of the dead.Many of the dead are blameworthy, according to Clarke, and it is fitting to blame them.Yet the dead do not deserve to feel guilty: If the dead no longer exist, they are no longer capable of feeling guilt or experiencing any form of suffering.They then no longer deserve to undergo those experiences.One who deserves to feel guilt has a reason to feel guilty; one who deserves to experience suffering has a reason to experience-or a reason to bring it about that she experience, or a reason to permit it to be brought about that she experience-that suffering.But the no-longer-existing dead have no such reasons.Hence, they no longer deserve to suffer.(Clarke 2022(Clarke : 2591) ) It is not obvious that the dead are blameworthy, given that they no longer exist.It seems plausible that when we blame the dead, we're blaming the person who once was.Compare our fond feelings for the dead.It's not that we have fond feelings for something that no longer exists; it's that we have fond feelings for that person who once was. 20If this is right, we might say that many of the dead were blameworthy because they didn't experience in life the suffering the deserved, and it is fitting to blame them now in virtue of their past blameworthiness.This doesn't pose a prob-lem for the Debt Model, since the dead who were blameworthy also deserved to feel guilty while they were alive. 21 The Debt Model also has the resources to reply to the challenge even while admitting that the dead are blameworthy.The solution is to make a distinction within that Model between the living and the dead.A living person is blameworthy for an action to the extent that she deserves to feel guilty for that action.A dead person, I suggest, is blameworthy for an action to the extent she deserved to feel guilty for that action when she died.A dead person can be blameworthy now and forever in virtue of the possession of a particular past property, namely the property of being deserving of guilt at the moment of death. 22This would explain how the dead can be blameworthy and why it is appropriate to blame (some of) the dead (to some degree).Moreover, this explanation can also make sense of the idea that that it appropriate to blame a fully unrepentant dead wrongdoer more than someone who experienced guilt and atoned for his or her action while she was alive.This would be a modification of my version of the Debt Model, but it doesn't seem problematically ad hoc, since it retains the central features of the Debt Model: An agent's degree of blameworthiness is a function of her deserved guilt.The blameworthiness of the dead is still a matter of unpaid debts.The account of blameworthiness for the living stays the same while the account of blameworthiness for the dead is fixed by how blameworthy the agent was at the time of death.The different treatment of the living and the dead is licensed by a crucial difference between them, namely, if Clarke is right, the fact that the dead cannot deserve to feel guilty, whereas the living can.
Clarke criticizes a suggestion that is in some respects similar to the one above.Portmore (2019aPortmore ( , b, 2022) ) takes an agent to be blameworthy only if what blame represents is correct.According to Portmore, blaming emotions represent the wrongdoer as "not having suffered all that she deserves (or deserved) to suffer."(2022: 73, Portmore's emphasis).This representation will be true about dead wrongdoers who in fact did not suffer all that they deserved to suffer.The dead can thus be blameworthy.Clarke argues that this is an implausibly sophisticated account of what blaming emotions represent: it seems to violate what Rosen (2015: 75) calls the naivety constraint: the representational content of blaming emotions must be framed in terms that everyone capable of such emotions understands.However, Portmore's account might have the resources to respond to this objection.On his account, the representational content need not be an occurrent belief or thought.We discover the representational content of emotions by considering "what typically elicits mental states of this kind, what normally attenuates them, what their phenomenology is like, what interpretation of their representational content rings true to those who possess them, and what sorts of act tendencies and patterns of attention are generally associated with them" (2022: 54).In light of this empirical data, we are then to give an interpretation of how the emotion appraises its object. 23The question is thus whether Portmore's interpretation of what guilt represents makes most sense in light of such data. 21For a similar argument, see Matheson (2024: 11). 22Or perhaps at the moment where it becomes impossible for the wrongdoer to experience any more guilt. 23See D' Arms and Jacobson (2000) for the same methodological approach.
However, a proponent of the Debt Model doesn't need to take a stance in this debate.My suggestion above is not about the content of blaming emotions.It is rather a theoretical proposal for how to understand the blameworthiness of the dead.As such, it is not subject to Clarke's objection.
One could, however, make a normative objection.It might be that this account would be implausibly unfair.Suppose that someone died in a car accident before they had a chance to experience the guilt they deserved, apologize, and atone for a wrongdoing.This person might remain more blameworthy -forever -than a person who did something morally worse, but lived a long time, experienced guilt and made up for their wrongdoings.However, this seems to be an acceptable form of circumstantial moral luck.Moreover, this account makes sense of the moral urgency of atonement.It seems to be a serious matter when people are not able to make up for their wrongdoings.Deathbed confessions and expressions of repentance are important parts of our moral life, and this view can account for that.I thus believe that Clarke's negative argument against the Debt Model is unsuccessful.I will now present a prima facie case against PERMANENT.I will then discuss Clarke's responses to cases like this.

The Prima Facie Case Against PERMANENT
If PERMANENT were true, there is nothing a wrongdoer or victim could do that will change the wrongdoer's status as blameworthy.But PERMANENT also seems to imply a stronger claim: there is nothing that can change the wrongdoer's degree of blameworthiness.If experiences of guilt, apologies or atonement could change the degree to which a possible instance of blame, by someone, would be fitting, there would be no principled reason for why more guilt, apologies or atonement could not render blame unfitting altogether.I think both of these implications of PERMANENT are implausible.There seem to be certain things a victim or a wrongdoer can feel or do that might change the wrongdoer's status as blameworthy or at least change their degree of blameworthiness.Consider the following case: Two Wrongdoers: Suppose that Paul and Peter both did something wrong towards a friend.They did so with the same quality of will, with the same level of knowledge and control.At the moment of action they are equally blameworthy.Then their trajectories split.Peter acknowledges that what he did was wrong but decides not to think more about it.He isn't particularly bothered by his action, and he doesn't feel guilty.He does not apologize or seek forgiveness.Paul on the other hand recognizes that he has committed a serious wrong.He experiences prolonged and intense guilt because of what he done.He apologizes and tries to atone for his wrongdoing.After doing so, he asks his friend for forgiveness. 24he question now is whether Peter and Paul remain equally blameworthy in light of these different trajectories.An implication of PERMANENT is that they are both blameworthy, to exactly the same degree.I submit that they are not.This is of course a rather compressed example and its various elements (feeling of guilt, expressions of guilt, and atonement) could be pulled apart and considered separately.But for now, it is sufficient to note three things.First, we do blame people who apologize, atone, and express what we take to be genuine feelings of guilt less than people who do not.This practice seems perfectly appropriate.Second, the Debt Model, as we have seen, can explain why this is the case: if a wrongdoer has experienced some of the guilt he deserves, he will be less blameworthy than he otherwise would have been.Moreover, it seems plausible that an agent will continue to be deserving of guilt until he has fulfilled his reparative duties.
A proponent of might reply as follows: Granted, PERMENANT entails that Peter and Paul are equally blameworthy for the initial wrong.But it doesn't entail that they are equally blameworthy, full stop.Peter, in addition to being blameworthy for the initial wrong, is also blameworthy for his failure to make amends, whereas Paul is not.This fact, it could be argued, explains why the two men aren't equally blameworthy.But this is perfectly compatible with PERMANENT: Peter and Paul are, and will remain, equally blameworthy for the initial wrong. 25 I think this reply is unconvincing, for two reasons.First, consider a version of the case where it is impossible to make amends.Suppose Peter and Paul got shipwrecked on a remote island, with no way of contacting their friend, or that their friend died.Intuitively, the fact that Paul feels guilty for what he has done, whereas Peter does not, seems relevant for their relative degree of blameworthiness.But PERMANENT entails that they are blameworthy to the same degree.
Second, suppose it is appropriate for Paul's friend to blame Paul just after he has committed his wrongdoing.Then Paul goes through the entire process of apology, experiencing, and expressing guilt, compensation, and atonement.According to PERMANENT, it would still be fitting to blame Paul, to the exact same extent, even after this process has occurred, since nothing that happens after the wrongdoing can change one's degree of blameworthiness.This, I think, is implausible.It seems true that a failure to make amends can constitute an additional wrongdoing.But it also seems plausible that experiencing guilt and making amends for one's wrongdoing will influence one's degree of blameworthiness for the initial wrongdoing.
According to PERMANENT neither experiencing guilt, apologizing or other forms of atonement can influence the wrongdoer's degree of blameworthiness.Given this, Clarke must explain why it nevertheless appears appropriate for guilt and resentment to diminish.This will be the topic of the remainder of the paper.

Blameworthiness and the Fitting Duration of Blame
Clarke accepts that both resentment and guilt can reasonably diminish.However: 25 Thanks to Justin Capes for raising this objection.
The reasonable diminishment of a wronged person's resentment does not imply that blame by others must diminish if it is to be fitting.The reasonable diminishment of feelings of guilt likewise lacks that implication.In neither case, then, is it implied that the offender has become less blameworthy.(Clarke 2022(Clarke : 2593) ) There are two ways of explaining how resentment and guilt, i.e. both other-directed and self-directed blame, can reasonably diminish without thereby making the wrongdoer any less blameworthy. 26First, resentment and guilt might remain fitting (at the same level of intensity) forever, but there might be moral and prudential reasons for these emotions to diminish.I will return to this strategy in the next section.
Second, the fittingness of resentment or guilt might change, without a corresponding change in the object of these emotions.Drawing on recent work by Na'aman (2021), Clarke suggests that emotions can be rationally self-consuming: the mere fact that one has experienced an emotion can make it fitting for it to diminish.27Na'aman does not argue that the duration of an emotion itself provides reason for that emotion to diminish: "The reason for one's resentment, whether intense or mild, is the injustice done, not facts about the duration of one's resentment; similarly, a year after the death of one's beloved, the reason for one's lingering grief remains the death of the beloved and does not include facts about the process of grieving one has undergone" (2021: 251).But this, Na'aman argues, is compatible with the claim that the fittingness of an emotion can depend on its duration.On Na'aman's view, the duration of an emotion is a background condition.Background conditions are part of the explanation of one's reason to ɸ without itself being a reason to ɸ (Dancy 2004;Schroeder 2007).To use Na'aman's illustration: The fact that I love someone is part of the explanation for why I have reasons to grieve their death, but it is not itself a reason to grieve.On this view, the duration of an emotion serves as a background condition that is part of the explanation for why the emotion can fittingly diminish without itself being a reason for why the emotion should diminish.Thus, the fact that you have already experienced resentment is part of the explanation for why it is no longer fitting to resent a wrongdoer to the same extent and the fact that I have already experienced guilt is part of the explanation for why it is no longer fitting that experience guilt to the same extent.
Importantly, Na'aman does not believe that the duration of regret affects the regrettability of what is regretted, or that the duration of shame affects the shamefulness of what one is ashamed of."Generally, the mere fact that a person has been amused, ashamed, or disgusted by something does not imply that the object has ceased to be amusing, shameful, or disgusting" (2021: 254-255).It is this feature of Na'aman's account that allows Clarke to maintain that although it may be fitting for blame to diminish or cease, this has no implication whatsoever for the agent's blameworthiness.
In adopting Na'aman's account, Clarke must reject a widely held understanding of fittingness.According to the standard view, the relation between the fittingness of a response and the evaluative quality of the object of the response is one of equivalence.Something is admirable if and only if it is fitting to admire, regrettable if and only if it is fitting to regret and amusing if and only if it fitting to be amused by.These biconditionals have been taken by many in this debate to express semantic, conceptual and/ or metaphysical truths (Brandt 1946;D'Arms and Jacobson 2000;Howard 2018Howard , 2022;;Berker 2022).On this view, there cannot be a change in the fittingness of regret without a change in the regrettability in the thing regretted; there cannot be a change in the fittingness of shame without a change in the shamefulness of an act, or a change in the fittingness of blame without a change of blameworthiness of the agent (Howard 2022).Clarke must deny this equivalence.He nevertheless defines blameworthiness in terms of fitting blame.Recall his definition: An agent S is blameworthy for X just in case some possible instance of blame of S for X, by someone, would be fitting.(Clarke 2022(Clarke : 2580) ) This definition retains a necessary connection between blameworthiness and fitting blame, while at the same time allowing for a fitting diminution of an emotion without a corresponding change in the object of the emotion.Yet this definition, combined with the idea of rationally self-consuming emotions, also raises new problems.
Suppose everyone affected by a wrongdoing has experienced the appropriate blaming emotions for a certain duration of time.The victim has experienced resentment, the wrongdoer guilt, and third parties have experienced indignation.If blaming emotions are rationally self-consuming, it follows that it would no longer be fitting for any of the affected people to experience blame, or that it would be fitting for them to feel less blame than they initially felt.However, according to Clarke, some possible instance of blame, say by a person who reads about the incident in a newspaper, would still be fitting.So on Clarke's view, the wrongdoer is still blameworthy -even though it would no longer be fitting for the victim to blame him, or for the wrongdoer to blame himself, or for the affected third parties to blame him.In this case, it will also just be the third party's initial fitting blame that matters for the wrongdoer's blameworthiness.For as soon as the newspaper reader's indignation has lasted for a while, it will also cease to be fitting for her to blame the wrongdoer.But other people might learn about the wrongdoing and -at least for a while-fittingly blame the wrongdoer.On Clarke's view, if we know that this kind of blaming is fitting, we would also know that that the agent is blameworthy.
Yet, it is not clear why our analysis of blameworthiness should privilege the fitting blame of possible observer rather than the fitting blame of people directly affected by the wrongdoing.Nor is it clear why the fittingness of an initial reaction should be privileged over the fittingness of an emotion that has persisted for some time.These questions are particularly pressing, given the dialectical situation.The question Clarke wants to answer is whether an agent remains blameworthy, to the same extent, after he has blamed himself and has been blamed by others.Clarke also accepts that an agent's blameworthiness implies the fittingness of blame.Given this dialectical situation, we need a principled reason for why our analysis of blameworthiness should prioritize the fitting reaction of people who haven't blamed the wrongdoer over the fitting reactions of victims and wrongdoers after they gone through a process of blame and self-blame.Clarke's account does not seem to provide one.
Another problem with applying the idea of rationally self-consuming emotions to the case of blame is that the mechanism is strictly intrapersonal.28On Clarke's account of fittingness, the victim's own resentment can become less appropriate as a result of his having experienced that feeling.However, the fact that the wrongdoer has experienced guilt, or atoned for her action, plays no normative role in the diminution of victim's fitting resentment.But this is not how appropriate diminution of blame tends to work.As a result of a wrongdoer expressing sincere guilt, resentment, and anger towards them will often diminish or disappear, and this will often strike us as appropriate.The mechanism of rationally self-consuming emotions does not, as Clarke himself (2022: 2594) notes, explain this.This, however, is the main phenomenon that Clarke needs to explain away.

Desert and the Ethics of Blame
Clarke's solution is to appeal to the ethics of blame.Atonement and expressions of sincere guilt do provide reasons for others to blame the wrongdoer less, but these reasons do not affect the fittingness of blame and are thus not relevant for the wrongdoer's blameworthiness.Clarke (2022Clarke ( : 2595) ) writes: A wrongdoer's suffering of guilt can evoke our sympathy and love.[…] recognizing the real deal can soften our hearts, give us hope that the agent can be counted on to do better in the future, and leave us willing to reengage in relationship with the offender.

Moreover:
That someone is worthy of blame is one consideration bearing on whether to blame her.But it is rarely if ever the sole consideration bearing on that question; indeed, it is rarely if ever the only moral consideration.There is a meanness of spirit in undiminished blame of someone who has experienced remorse, apologized, made amends, resolved to do better, carried out that commitment, and been forgiven by her victims (2595).
When faced with a repentant wrongdoer, we will, according to Clarke, have conflicting reasons.On the one hand, facts in virtue of which the agent is blameworthy provide us with reasons to blame him.Such reasons render (possible instances of) blame (by someone) fitting.They are the right kind of reasons. 29Indeed, since the wrongdoer's atonement and suffering of guilt, on Clarke's view, does not have any bearing on the degree of his blameworthiness, we have the right kind of reasons to blame him to exactly the same degree as before he experienced guilt and atoned for his action.On the other hand, the wrongdoer's sincere suffering of guilt provides us with reasons of sympathy and love not to blame him, or to blame him less.These are reasons that do not have any bearing on his blameworthiness, so they are the wrong kind of reasons for blaming emotions.According, to Clarke, guilt and atonement thus provide us with reasons not to resent a wrongdoer, but these are the wrong kind of reasons for blame. 30 It is certainly true that we can have reasons of love and sympathy that count against continuing blaming a blameworthy wrongdoer.But this does not seem to be all that is going on in cases where an agent has experienced guilt and atoned for his wrongdoing.Love and sympathy provide reasons of compassion.Such reasons are compatible with the claim that the agent deserves to be blamed.However, to continue blaming a repentant wrongdoer, to the exactly same degree as before she experienced guilt and tried to atone for her wrongdoing, does not only seem to indicate a lack of compassion or sympathy.It also seems unjust.Again, compare Paul and Peter.If the victim continues to blame them to the same degree, and for the same duration, despite Paul's guilt and atonement, this seems to be unjust towards Paul.The Debt Model can explain why this is the case, since it follows from this view that Paul is less blameworthy than Peter.Clarke (2022) accepts that agents deserve to feel guilty and will deserve less guilt depending on how much they have already experienced.But he denies that the fact that an agent deserves to blame herself is relevant to her blameworthiness: it is possible that an agent does not deserve to blame herself at all, but that she nevertheless is fully blameworthy.Clarke thus seems committed to the claim that reasons that are provided by facts about how and when the agent deserves to blame herself are the wrong kind of reasons, i.e. reasons that do not have a bearing on the agent's blameworthiness.But it is not straightforward to how one can defend the claim that the amount of self-blame that a wrongdoer deserves, is irrelevant to an agent's blameworthiness.In order to argue for this claim, one might adopt two different strategies.I will try to show that both are unpromising.One possibility, not embraced by Clarke, would be to reject the general claim that desert is relevant to blameworthiness. 31However, there are good reason to believe that facts that ground deserved blame provide right kind of reasons for blaming.Desert, just as fit, picks out a relation between a response and an object, where the object is worthy of, or merits that response. 32Desert-relations thus differ from considerations concerning whether it is valuable or good to have a certain response to an object.An agent's deserved blame is not sensi-30 I understand wrong kind of reasons, as considerations in favor of an attitude that do not have a bearing on whether something is blameworthy, shameful, regrettable and so on.Such considerations can nevertheless be very good reasons to form or have an attitude. 31For authors who have defined blameworthiness in terms of desert, see e.g.Feinberg (1963); Pereboom (2001Pereboom ( , 2014Pereboom ( , 2021););McKenna (2012McKenna ( , 2019McKenna ( , 2020)); Clarke (2013Clarke ( , 2016)); Clarke and Rawling (2022b); Nelkin (2013aNelkin ( , 2019)), Rosen (2015) Carlsson 2017), (2019, 2022); Portmore (2019aPortmore ( , b, 2022)). 32Clarke and Rawling (2022b) argue that fittingness and desert of blame are grounded in the same facts.Feinberg (1963);McKenna (2022a, b) understand desert as a species of fittingness.tive to instrumental reasons. 33Consider standard cases of wrong kind of reasons to blame: a demon offered me a million dollar to blame myself; I lack the standing to blame you; blaming you would have disastrous consequences.Such reasons against blaming are clearly compatible with the fact that the agent is blameworthy.Similarly, it makes sense to think someone fully blameworthy, but to decide to blame them less out of love and sympathy.But it is far less plausible to say that they deserve less blame now than they did two years ago, yet they are blameworthy to exactly the same extent.Desert of blame appears relevant to an agent's blameworthiness in a way that one's love or sympathy are not.The explanation of this intuition is that the facts that ground desert, as opposed to facts that grounds compassion and sympathy, provides right kind of reasons for blaming.
Another option would be to claim that it is only deserved self-blame that is irrelevant to moral blameworthiness.But this does not work either.First, it is hard to provide any principled reason for why only deserved other-directed blame should be relevant the agent's blameworthiness.Moreover, if what a wrongdoer deserves is fundamentally the blame of others, this will also presumably be a certain amount of blame.So, this version of the view would also imply that once the wrongdoer has received the amount of other-directed blame, he would no longer be blameworthy.
A more promising solution to this problem, suggested in another recent paper by Clarke and Rawling (2022b: 235), is that an agent remains blameworthy just in case she deserves to be blamed by someone.They write: One remains to blame; blame by others can still be deserved, even if one no longer deserves to feel guilty.But neither do we tie blameworthiness to desert of blame by any particular other.Just as one's guilt can become unfitting over time, so, we think, another's resentment can become unfitting.Neither change renders one no longer blameworthy.A third person's proportionate blame might still be fitting and deserved.
We could thus modify Clarke's (2022) definition of blameworthiness to incorporate desert: An agent S is blameworthy for X just in case some possible instance of blame of S for X, by someone, would be fitting and deserved.
This would allow Clarke to retain a necessary connection between desert and blame, without giving up the claim that an agent is blameworthy to exactly the same extent even if he no longer deserves to blame himself (or deserve less blame than he used to).But this suggestion faces a parallel problem to the one I raised against Clarke's official formulation (2580).Suppose the victim and relevant third parties have blamed a wrongdoer to the extent that he has gotten what he deserved from them.Suppose also that he has experienced all the self-blame he deserves.It follows from the modified version of Clarke's account, that the wrongdoer is nevertheless blameworthy, since an agent is blameworthy just in case he would deserve blame from a potential 33 For on more on this, as well as on the similarities between fittingness and desert, see Clarke (2024: 64).third-party blamer.Again, we face the problem of why our analysis of blameworthiness should epistemically privilege the deserved, initial blame of an unaffected third party rather than blame he deserves to receive from the victim, affected third parties, and himself.
This account also leads to some striking asymmetries.Suppose that Paul has experienced all the guilt he deserves.In that case, he no longer deserves to blame himself.Yet, on Clarke's account, he may still deserve to be blamed by others.Suppose, on the other hand, that Peter has not blamed himself, but has received all the blame he deserved from the victim.If so, it follows that he deserves to blame himself, without deserving any more blame from the victim.I think this is an unfortunate result.When we believe that we deserve to blame ourselves, we will experience the blame of the victim as pro tanto justified.Similarly, when we are victims, we will believe that the wrongdoers self-blame is pro tanto justified.
Finally, is it true that possible instances of deserved blame, by someone, are independent of the amount of blame the wrongdoer has already received, by himself and others?I don't think so.Suppose first that what the wrongdoer deserves is expressions of blame.Imagine an office employee in a large company did something wrong.News of his wrongdoing are slowly disseminated through the company, so that every day new people will learn about it.His co-workers will thus one by one have gain reasons to express their blame towards the employee.Given that what he deserves is to be blamed by each of them, on this view it does not seem that the cumulative effect of this blaming would provide desert-based reason against continued blaming.Of course, desert-based reasons are always pro tanto, so co-worker 469 would not have an all things considered reason to blame the employee.But he would nevertheless deserve the expressed blame of co-worker 469.This seems implausibly harsh.
Clarke, however, takes blame to be an attitude of holding the wrongdoing against the wrongdoer.He mentions angry indignation as one example.This attitude need not be expressed at all.This avoids some of the harshness of the above example. 34 But even this kind of blame yields implausibly harsh results.Again, compare the unrepentant Peter, with Paul who has experienced severe guilt for what he has done, apologized, and atoned for his wrongdoing.It follows from Clarke's account that Peter and Paul would deserve exactly the same amount of angry indignation.This strikes me as implausible.

Conceptions of Blameworthiness and the Drama of Atonement
The Debt Model understands blameworthiness in terms of deserved guilt, whereas Clarke rejects this connection. 35But Clarke and the proponents of the Debt Model would agree on a number of important claims: that the fact that one has done something wrong with the appropriate degree of control and knowledge will remain true 34 But not all, in particular if we allow for non-experiential harms.If one is harmed by another's unexpressed indignation or resent, a wrongdoer will deserve this harm, forever, from a potentially endless list of people, and there is nothing she can do about it. 35As does Howard (2022;footnote 11).
forever; that a wrongdoer deserves to feel guilty, and that she will no longer deserve to feel guilty once she has experienced the amount of guilt she deserves.
At this point one may reasonably wonder whether all of this isn't just a verbal debate.Clarke defines blameworthiness in terms of the fitting initial blame of a possible blamer; the Debt Model defines it in terms of the wrongdoer's deserved guilt.Whether PERMANENT is plausible may seem to follow directly from one's conception of blameworthiness.I don't think this is quite right.As we have seen, there is also substantial disagreement about whether guilt and atonement matter for potential third-party blame.However, I also believe that something important is at stake in the debate about which conception of blameworthiness we adopt.If Clarke's account were right, the fact that an agent is blameworthy would play a far less significant role in our moral life than we often tend to assume.After all, on his account it is possible to be fully blameworthy, even though it would not be appropriate for anyone affected by the wrongdoing to blame the wrongdoer, and even though the wrongdoer no longer deserves to blame himself, or no longer deserves to be blamed by the victim of his wrongdoing blame.We may know that an agent is blameworthy, without knowing whether it is fitting for us to blame him, whether he deserves our blame, or whether it is pro tanto justified for us to express our blame. 36I think this would be a disappointing result.The question of whether an agent is blameworthy would lose much of its moral urgency and it would fail to make sense of important part of our blaming practices.On the Debt Model, by contrast, the question of blameworthiness retains its importance.Whether an agent is blameworthy depends on whether the wrongdoer deserves to suffer the pain of blaming herself, and what she can do to be released from this suffering.One of the main attractions of the Debt Model, as far as I can see, is that it makes sense of this drama of atonement. .