Retributivism, Punishment, and the Punitive Emotions

Setting the Stage...

The issue of retributivism and punishment has been a common theme on both the Garden and Forking Paths and Flickers of Freedom.  My present purpose is to try to briefly get clear on the salient terms and themes of the debate.  Because this is the first of what will be a series of posts by me in the coming months on retributivism, I want to define some of the key terms up front to prevent confusion downstream. 

So, for starters, I follow Moore (1997) in defining retributivism in the following way:

“Reributivism…is the view that punishment is justified by the desert of the offender.  The good that is achieved by punishing, on this view, has nothing to do with future states of affairs such as the prevention of crime or the maintenance of social cohesion.  Rather, the good of punishment achieves is that someone who deserves it, gets it.  Punishment for the guilty is thus for the retributivist an intrinsic good, not the merely instrumental good that it may be to the utilitarian or rehabilitative theorist” (p. 87-88).

On this “pure” view, we not only have a right to punish those who cause culpable harm, we actually have a duty to punish them even if punishing them will produce negative consequences.  In short, we have a duty to punish offenders because giving them what they deserve—i.e., making them suffer—is intrinsically valuable.  When I, for one, talk about retributivism, this is the view I have in mind.  

And here are the views I do not have in min

• Weak (or negative) retributivism—i.e., the view that desert is a necessary but not a sufficient condition for punishing offenders. 
• Mixed theories of punishment—i.e., views that provide a consequentialist justification for the institution of punishment but a retributivist justification for the implementation in particular cases.
•  Lex talionis—i.e., a view that attempts to provide a scale for determining how much someone deserves but which is not a theory about why and whether punishment is justified in general.
• The catharsis view—i.e., the view that punishment is justified because it satisfies either the victims’ or society’s thirst for vengeance.
• The denunciatory/expressive view—i.e., the view that justified punishment is a vehicle whereby society can express its moral condemnation.
• The scales of justice view—i.e., the view that offenders deserve to suffer because they attempted to derive benefits from society without at the same time paying the costs associated with being part of society.

I, for one, agree with Moore (1997) that each of these views is either not properly retributivistic or does not capture the distinctive feature of retributivism—which is the sufficiency thesis associated with the aforementioned pure view.  Of course, if readers of Flickers of Freedom want to discuss these alternative views, I am happy to do so.  But when I am criticizing retributivism, I will ordinarily be focused on the pure view.

The second term I would like to define is “retribution”—i.e., making an offender suffer in a manner that is proportionate to the harm he caused because doing so is intrinsically valuable.  So, while retributivism is a theory, seeking retribution is an action whereby one seeks to punish an offender solely because he deserves to suffer for his wrongdoing.  In short, retribution is what the retributivist thinks we have a duty to inflict upon those who culpably violate legal or moral norms.

The third term I would like to define is “legal punishment.”  Following Hart (1968), I am going to define legal punishment such that in order for an act to count as punishment, it must meet the following five conditions:

•  It must involve pain or other consequences normally considered unpleasant.
•  It must be for an offense against a legal rule.
• It must be of an actual or supposed offender for his offense.
• It must be intentionally administered by people other than the offender.
• It must be imposed and administered by an authority constituted by a legal system against which the offense is committed.

With slight modifications, Hart’s definition of legal punishment could be used to cover punishment more generally—for instance, when a parent punishes a child.  But that need not concern us here.  For present purposes, the important point is that a retributivist is someone who justifies punishment—whether it be legal or otherwise—by appealing to its intrinsic value rather than its instrumental value.

Finally, I would like to clarify which reactive attitudes I am interested in when talking about punishment, desert, and retribution.  In this context, I think the desire to “get even”—i.e., the desire to reciprocally harm those who harm others—is a paradigmatic so-called “punitive emotion.”  In short, a punitive emotion is a reactive attitude which leads one to want to punish someone who has violated a legal or moral norm.  Notice one could have a punitive emotion without at the same time seeking retribution.  All that a punitive emotion involves is a desire to punish offenders.  It does not necessarily require the desire to punish offenders for intrinsic rather than merely instrumentalist reasons.  To make this distinction clear, perhaps we should distinguish instrumentally motivated punitive emotions from intrinsically motivated emotions.  For present purposes, I will simply call the former “punitive emotions” and the latter “retributivist emotions.”

In summary, retributivism is a theory of punishment.  Retribution is the purportedly intrinsically valuable act of giving offenders what they deserve.  Punishment is a sanctioning practice whereby individuals are intentionally harmed for violating legal norms.  Finally, a punitive emotion is an affective and reactive response to harm doing whereby the individual experiencing the emotion wants to harm the offender(s) whereas a retributive emotion is a desire not only to punish, but a desire to seek retribution.  With these distinctions in hand, I now want to put some questions for consideration on the table:

• What do you take to be the strongest argument for retributivism?  Conversely, what is the strongest argument against retributivism?
• Is compatibilist control robust enough to undergird retributivism/retribution? 
• Are our punitive emotions justified?  If so, why?  If not, why not? 
• Are our retributive emotions justified?  If so, why?  If not, why not?
• Can our punitive emotions—which were presumably selected for because they helped our ancestors solve the cooperation/free rider problem—be used to motivate and bolster a non-instrumentalist theory of punishment such as retributivism?
• Compatibilists and libertarians both believe in free will—albeit of different sorts—but do they have a shared conception of moral desert?  If so, what is this common conception?  If not, what are the different notions of desert in play in the two respective theories?
• One term I did not define above but which plays an essential role in the debate about retributivism and punishment is moral desert.  I, for one, have no idea how to define it in a non-circular manner.  So, I am hoping folks will give me their own two cents when it comes to the nature and limitations of desert.  Is there a non-circular way of defining desert?  If so, is there as unproblematic way of weighing/scaling/measuring desert thusly defined?

OK, that’s more than enough (if not too many!) questions for now.  Hopefully, the comment thread for this post will set the stage for future discussions about retribution, punishment, and desert. 

Comments

[Neal Tognazzini]

Just a clarification question about the 'sufficiency thesis': which of the following is intended by it?

1) The fact that giving a wrongdoer what he deserves is intrinsically valuable is, by itself, sufficient to justify punishment.

or

2) The fact that giving a wrongdoer what he deserves is intrinsically valuable is, by itself, sufficient to make omitting to punish unjustified. (That is, this fact is sufficient to generate a duty to punish.)

I take it that you mean (2), but your characterization of "weak retributivism" makes it sound like it's only strictly contradictory with (1).

If I'm right that you mean (2), could you point toward some authors who hold that view?

[Thomas Nadelhoffer]

Neal,

I have Moore's views in mind. And here is what he says on this front in *Placing Blame*:

"Moral responsibility ('desert') in such a view is not only necessary for justified punishment, it is also sufficient. Such sufficiency of justification gives society more than merely a right to punish culpable offenders. It does this, making it not unfair to punish them, but retributivism justified more than this. For the retributivist, the moral responsibility of an offender also gives society the duty to punish. Retributivism, in other words, is truly a theory of justice such that, if it is true, we have an obligation to set up institutions so that retribution is achieved" (1997, p.91).

He goes on to state:

"A succinct definition of the retributivist theory of punishment, paralleling that given of the utilitarian theory, is that punishment is justified if and only if the person or persons receiving it deserve it" (1997, p.92)

So, to answer your question, I actually think that by the "sufficiency thesis" I was trying to capture Moore's general view--which, I think, incorporates both 1 and 2. The "weak retributivist," on the other hand, is committed to neither 1 nor 2. The phrase itself comes from H.L.A. Hart, but I take Quinton's "On Punishment" to be the clearest defense of the view. Of course, just because I wasn't focusing on weak retributivism doesn't mean we can't discuss it in the thread--especially if you think there are reasons to prefer it to Moore's purer version!

[Neal Tognazzini]

Thanks, Thomas, that helps. (Though I will point out that Moore's "succinct definition" doesn't on the face of it capture the idea that we are obligated to punish those who deserve it. Nevertheless, as you point out, it looks like we should interpret it as though it does.)

I don't think I really have an *argument* against this strong retributivist view, but it does strike me as extremely implausible. I can see how the intrinsic value of punishing deserving wrongdoers could generate a justification for punishment, but I can't see how it could generate a duty to punish unless we had a general duty to promote whatever is intrinsically valuable. But the suggestion that we have such a general duty strikes me as implausible.

[Josh Shepherd]

Hi Thomas,

I'd like to hear a bit more about how you maintain a distinction between punitive and retributivist emotions. Must one have a theory about the intrinsic value of retribution to experience the latter? Because it sounds as though punitive emotions are what I normally experience when I notice a wrong. What is the nature of the added intrinsic element which makes a punitive emotion also a retributivist emotion? And do they come apart - is it possible to have a retributivist emotion without having a punitive emotion?

[Thomas Nadelhoffer]

Neal,

I take it that the part of Moore's view that you find implausible is also part of Kant's view, no? After all, his "execute the deserving even if the island is disbanding" example makes a similar suggestion--namely, that moral desert makes demands on us such that we have a duty to seek retribution regardless of the consequences.

[Thomas Nadelhoffer]

Josh,

Good question. To be honest, I am unsure it makes sense to talk about retributive emotions. I was simply trying to highlight the fact that punitive emotions are not necessarily desires to seek retribution. When I desire to get even, for instance, my desire could be forward-looking in nature. What I really want is to get even so you don't pull another stunt like that in the future. But my desire to get even could also be non-instrumental--i.e., I want to see you suffer solely because you made me suffer.
If you can think of a better way of drawing the distinction, I welcome your advice. I am certainly not wed to the punitive/retributive labels. I nevertheless think there is an important distinction to be drawn in this vicinity--whatever we decide to label the mental states in question. Would talking in terms of beliefs be more apt than talking about emotions/desires in this context?

[Thomas Nadelhoffer]

Josh,

How about this: Not all desires to punish are desires to seek retribution. But all desires to seek retribution are desires to punish. As such, both are punitive in nature--but there is still an important distinction lurking in this vicinity that needs to be drawn.

[Neal Tognazzini]

Thomas,

It's not that I can't see any reason at all for holding the view that we sometimes are obligated to punish. If, for example, you do something wrong and in the past we've punished all others who have offended similarly, then considerations of fairness might dictate that we punish you, too, even if no good comes of it. But what seems implausible is that the intrinsic value of punishing the deserving could, in itself and apart from other considerations, ground a duty to punish. Insofar as that's Kant's view (I'm not qualified to make that judgment), I think it's implausible.

[Justin Capes]

Thomas,

Great post. In response to question (1), I find W.D. Ross's argument in the Right and the Good appealing, though I think it needs to be worked out more carefully (a project of mine to which I may one day get around to working on again).

In response to question (2): no, of course.

Neal,

Can you say a bit more about why you find the thought that we have a general duty to promote that which is intrinsically good implausible?

Ross claims that we have a prima facie duty to promote good states of affairs, insofar as we can. Would you deny that claim? If so, why? (The claim, as you may be aware, figures importantly into his argument that we have a prima facie duty to punish wrongdoers.)

[Josh Shepherd]

Thomas,

That helps, thanks. I'm certainly no fan of retributivism as a theory of punishment. But it seems that some folks might (let me emphasize MIGHT - I have no axe to grind here) have a type of reactive attitude which they would describe as a desire to dispense retribution, but not a desire to punish. I'm thinking of a situation in which a parent, or a very prudish judge, would say (and believe) something like, 'I don't want to cause you harm, but I have to do this because you deserve it.'

There is a sense in which this is a desire to punish as well (a desire informed by certain beliefs, to be sure) - but not in the sense you discuss, the desire to get even.

But, reading your 5:42 comment, it sounds as though you're thinking of retributivist emotions as desires to get even as well - in which case my example either misses the mark, picks out a type of reactive attitude you didn't have in mind, or picks out nothing of interest.

Regarding your 5:56 comment, I do agree that there is a distinction to be drawn between punitive emotions that involve no thought about retribution and those that do. Perhaps the latter link up with perceived violations of moral or legal norms in some way, whereas the former need reference nothing about normativity -- I can have a punitive emotion towards you if you botch my attempted bank robbery, or whatever.

[Julia Rees]

Josh,

I don't think that's the distinction that Thomas was making. If you botched my bank robbery, I might wish to botch your planned train robbery, just to get even - in which case I would be experiencing a retributive emotion, no?

[Josh Shepherd]

Julia,

Perhaps that's right. Punitive emotions, as Thomas defined them, can be instrumentally motivated, whereas retributive emotions are in some sense intrinsically motivated. I was trying to get clear on the nature of the intrinsicality - is the motivation intrinsic because the offended thinks she has a duty to punish the deserving offender? Or is the motivation intrinsic because the emotion motivates the offended to simply get revenge? I was thinking the former, which is why I was thinking a retributivist emotion involves an extra thought, so to speak. And I was taking punitive emotions to be the more basic and typical responses to harm done - the desire to get even, without the thought about whether the offender deserves, in some moral or legal sense, the punishment.

[Thomas Nadelhoffer]

Josh,

First, the worry I was raising earlier about the intelligibility of retributive emotions was fueled by the following two concerns (a) it is unclear whether we should be talking about punitive emotions rather than desires, and (b) regardless of whether we talk about desires or emotions (or both) it is still unclear whether desires or emotions can intelligently be said to be retributive rather than merely punitive. As for (a), the worry is that while I can imagine cases that involve a desire to punish—which could be motivated by a wide variety of different beliefs—it is unclear what a punitive emotion would be. Moral outrage is an emotional response commonly associated with retributivism, desert, and responsibility. But moral outrage only becomes punitive when coupled with a desire to punish. The emotion itself is not punitive. Revenge, unlike moral outrage, does seem to be punitive—but I think revenge is better construed as a desire to get even than an emotion. Indeed, I now think that the distinction between punitive desires like revenge and moral emotions like outrage makes it clear that I should not have framed all of this in terms of emotions in the original post. I now want to distinguish between punitive desires and retributive desires rather than punitive emotions and retributive emotions. On this modified view, to seek retribution just is to act on punitive desires and beliefs that are coupled with retributive desires and beliefs. However, when someone merely seeks punishment but not retribution, they act merely on punitive desires and beliefs. Moral emotions such as anger, outrage, and disapprobation, on the other hand, may play a similar motivating role in both the retributivist cases and the non-retributivist cases.

I think this issue is related to the case you mention, so let me say a few words about that. I think that it would help if you could construe the punishment as a side effect. If this were the case, you could try to use the doctrine of double effect to suggest that the reluctant judge wants to get even but doesn’t want to punish. But the side effect model is not applicable in this case since the only way to give an offender what he deserves is to punish him. Punishment is the means by which we get even/give someone what they deserve. As such, one can’t both desire to give someone what they deserve—namely, punishment—and yet not desire that the person be punished. Of course, you could have conflicted desires if the offender was a family member or friend. But once you decide to give an offender what he deserves, you must have decided to punish him since the punishment is the means by which you dispense desert. You cannot have a non-punitive punisher although you could have a reluctant/regretful punisher. But in this later case, the punisher desires to punish the offender in order to give him what he deserves, but she wishes nevertheless that she didn’t have to punish him in the first place. I can get my mind around this type of situation. But it is unclear how to make sense of the situation you mentioned.

Finally, I also wanted to say a bit more about the desire to get even—which I think is distinct from both a desire for revenge and a desire for retribution. In the case of the desire to get even, I could want to get even in order to bring about some other consequence—e.g., regain the respect of the community—or I could want to get even solely in order to give the offender what he deserves. Only this latter desire is a retributive desire. Moreover, a retributive desire can be distinguished from a desire for revenge because the former desire is limited by the principle of proportionality and the latter is not—i.e., in order for a desire to be retributive rather than merely vengeful, I can desire to inflict no more suffering on the offender than is deserved. When I have a desire for revenge, on the other hand, I could want to inflict far more suffering on the offender than is deserved. In short, there can be misplaced revenge even if not all revenge is misplaced. That being said, it is an open question whether we should treat the desire for retribution as a moderated species of the desire for revenge. I am tempted to think we should. But I am curious what you and others think…

[Josh Shepherd]

Thomas,

This is all much clearer to me now. I too am curious what others think on these issues, so I'll just say that I like the way you've just carved these notions, I agree that my case makes more sense as an instance of conflicted desires (although might the punisher, in a bout of piety, deny that she really has a desire to punish, feeling instead some desire to fulfill a moral obligation?), and I'm tempted like you to treat a retributive desire as a moderated species of the desire for revenge - although I don't doubt that some would want to resist this move.

[Neal Tognazzini]

Justin,

I suppose putting the claims of retributivism in terms of *prima facie* obligation makes it a bit easier to swallow. But such a prima facie obligation is of course consistent with our not being, all things considered, obligated to punish wrongdoers, and I thought Thomas was after something that wouldn't allow for that weak possibility.

I'm also thinking that the claim about prima facie obligation sounds more plausible in the case of the individual than it does for the state. I'm definitely out of my area of expertise here, but "promoting all intrinsically valuable states of affairs as far as possible" doesn't really sound like a good job description of the state. Given that the institution of punishment is run by the state, then, I doubt that the state is obligated to punish wrongdoers simply because it would be intrinsically valuable to do so.

[Tamler Sommers]

Neal, what do you think of this analogy?

It's intrinscally valuable for children to have health insurance. Does that mean the state is obligated to provide health insurance for children? Not necessarily. But under some theories, the state ought to provide health insurance for the citizens who are unable to attain it for themselves.

Similarly, retributivism sees it as intrinsically valuable for wrongdoers to punished according to their deserts. Can you be a retributivist but believe that the state should stay out of punishing wrongdoers? I think so. Maybe you're a fan of vigilante justice, you think that victims should be the ones that punish--as long as they give the offenders what they deserve. On the other hand, you might think that private citizens are unable to dispense the proper amount of punishment. It's natural for the state to step in when it worries that private citizens are unable to attain an intrinsic good on their own. That is why the state should provide health insurance for children and why it should punish criminals according to their deserts.

[Neal Tognazzini]

Tamler,

I guess I'm also inclined to say that if the state is obligated to provide health insurance for those who are unable to attain it, that obligation can't be grounded in the mere fact that having health insurance is intrinsically valuable. It seems more likely to me that such an obligation is grounded in a corresponding *right*. If we think access to health care is a right of all citizens, then the government ought to provide it to those who can't get it themselves. It's also intrinsically good that those citizens have health care, but that's just bonus, as it were.

Along these lines, if you could make the case that wrongdoers actually have a right to be punished, then I'd be on board with the retributivist's claim that states are obligated to mete it out.

[Justin Capes]

Neal,

For what it's worth, Ross seems to agree with what you said in your response to me. I'm inclined to depart from Ross at this point, though I'm not sure I could defend the departure in any detail at present.

[John Fischer]

Tamler, Neal, and Justin,

Of course, there are lots of cases in which it would be intrinsically valuable for a bad person (or wrongdoer) to suffer, but it would be inappropriate for the state to cause the suffering. So: imagine a rich person who is arrogant, obnoxious, and uncaring--he looks down his nose at people less priviledged than he. Suppose he got his money from an inheritance. Perhaps we would all think it valuable if he lost his money; but it wouldn't be ok for the govt. (or anyone else) simply to confiscate it.

There are other examples in which it is pretty clear that someone deserves something, but it is a separate question whether the govt. may coercively provide it. Here: A person gives all his money (in this trust) to the morally bad child, who does not deserve it. Granted: the bad child doesn't deserve it, and the good child does: but it would be wrong for the govt. simply to transfer the funds to the more deserving child, apart from the parent's wishes.

Actually, I'm not sure if this addresses the specific worries with which you were concerned. But hey: reading this blog can be a full-time job if one is not careful!

[Tamler Sommers]

Well, retributivists like Hegel, Morris, and Primoratz do think that one tenet of the theory is that criminals have a "right" to be punished. Which would fit in with Neal's view about the role of Government--to protect the rights of citizens.

My broader view is that one can be a retributivist without taking a strong stand about the Government's role in applying the principles. Consider many honor cultures in which members of a victim's group the do not want the government and or any third party to dispense any punishment. They want to handle their own business. Still seem like retributivism to me. And looking back at Thomas's point, it still seems to meet his criteria (setting aside the legal punishment aspect). The essential feature is the belief that punishing offenders has intrinsic value. Questions about whether there is a duty to bring about the intrinsic value (and whose duty it is) are external to the question of whether we endorse retributivism.

[Neal Tognazzini]

Tamler,

If we strip retributivism of the thesis that someone (whoever it turns out to be) is obligated to mete out the punishment and emphasize instead the intrinsic value of punishing, then I find the view more plausible. But Thomas made it explicit that the question of duty was part and parcel of his version of retributivism.

As for honor cultures, I'm not sure what to say, except perhaps that there seems to be an intuitive difference between being honor-bound to punish and being duty-bound. And again, I find the honor-bound version more plausible than the duty-bound version.

[Tamler Sommers]

I think we agree, then. Maybe we could say it's an imperfect duty (whatever that means exactly), and not one that the state necessarily has to take on itself.

[Tom Clark]

If retributivism is the thesis that punishment proportionate to the harm inflicted is intrinsically valuable, then the retributivist has to defend the claim of intrinsicality. Consider that a 4 year old might do intentional harm, for instance hit a playmate. I don't think a retributivist would claim that inflicting suffering on that child is intrinsically valuable, even though in correcting his behavior we might have to mete out some sort of punishment. Flash forward to age 16, same scenario. Now the retributivist might well claim the suffering we inflict on the 16 year old harm doer *is* intrinsically valuable, not merely behavior guiding. What explains the shift in the retributivist's rationale for punishment? Why does the intrinsic good of inflicting suffering on an offender suddenly (or gradually) appear at the coming of age of a reasons-responsive agent?

[Tamler Sommers]

Oh oh, Tom--I fear a rehash of our debate in the earlier post. But I can't resist. One response (in fact, the only response that has a hope of working) would ground the difference in our attitudes: we don't feel resentment or moral outrage towards the four year old. Maybe we do at first, but upon due reflection the retributive emotions dissipate. By contrast, we do feel resentment and moral outrage towards the 16 year old, even upon reflection. No further justification required.

After all, you're not asking for further justification for the belief that there's no intrinsic value in punishing the 4 year old...

[Tom Clark]

Tamler,

I’m just sticking with Thomas’ original definition of retributivism and trying to point out what I think is its fundamental weakness in response to his first question.

By your argument, the claim that inflicting suffering is intrinsically good boils down to the fact that we feel resentment and moral outrage when offenders reach a certain age, so want them to suffer. If no one wanted offenders to suffer, it wouldn't be intrinsically good that they do. This seems strange since *intrinsic* goodness usually inheres in the good *thing* itself, which is in this case the suffering, not our desiring it. We should desire it *because* it is intrinsically good. On your view, why shouldn’t retributivists just say we want the suffering, and drop the idea that it’s intrinsically good? Isn’t that latter claim just a spurious justification on your view?

[Tamler Sommers]

Tom, two things.

First, I don't think intrinsic goodness has to 'inhere in the thing itself' such that it makes no difference how anyone is psychologically disposed. That's too metaphysical for my tastes. Nothing would count as intrinsically good. As I see it, intrinsic goodness is simply goodness that isn't solely instrumental--if something is intrinsically good, it's a good even when it doesn't lead to a further good.

Second, I think you're confusing the emotion of resentment or moral outrage with the desire that a person suffers. Although the two often match up, they can come apart as well. I can feel resentment and moral outrage without wanting a person to suffer, and I can want a person to suffer without feeling resentment and moral outrage (though usually that person is a Yankee or Mike Krzyzewski). The sentimentalist retributivist would not call the suffering intrinsically good in the latter cases, but would in the former cases (as long as certain other conditions e.g. involving impartiality or due reflection were met as well). Since moral outrage isn't a desire, it's not a further good that can be achieved. So the suffering doesn't lead to a further good. At the same time, if people weren't morally outraged by the crimes of 16 year olds, then the sentimentalist retributivist will lose his or her justification for the intrinsic value of their punishment.

[Tom Clark]

Thanks Tamler. I’m not sure I buy the claim that resentment and moral outrage doesn’t centrally involve the desire for the offender’s suffering. On the view you’re proposing, moral outrage justifies the judgment that it’s intrinsically good the offender suffer (intrinsically good because moral outrage demands he suffer whether or not it produces any further good). Upon reflection, I want that good to transpire. Doesn’t that necessarily mean I desire that he should suffer? I’m certainly not *indifferent* about whether or not he suffers.

In any case, you’re saying the rightness of retribution doesn't reside in some further fact, for instance a "moral reality" that our emotions track, as Michael Moore claims. He says "The moral fact of the matter often causes our moral beliefs through the intermediate causing of our emotional responses. Our emotions in such case become good evidence of the underlying moral landscape." (p. 181 Placing Blame). You’re saying there is no such landscape, or if there is it’s right on the surface, a direct function of the emotion itself. This seems to me good progress in debunking retributivist theories that try to validate moral outrage by adducing some further non-consequentialist consideration that makes the suffering obligatory. However, when you say that the sentimentalist retributivist requires impartiality and due reflection, as well as emotion, be involved in the justification, it very much sounds as if there *is* some further consideration beyond moral outrage that bears on whether the suffering is intrinsically good.