Thursday, December 04, 2008
Retribution: duty, or good?
This post falls into the "trumpeting my own work" category (though it might also bear some tangential relation to Ken Simons's interesting series of recent posts on consequentialism). I'm getting it out of the way early so I can move on to asking some questions about legal education, which I'm hoping to spend most of my time here doing.
Anyway, here's the point. Most criminal theorists discuss retributive justice as a deontological duty: a categorical obligation to impose punishment on those who deserve it, and avoid punishing those who don't. (Dan Markel would object to my equation of retribution and desert, but never mind that right now. In fact, I will tend to use the terms "retribution" and "desert" interchangeably, partly for the sake of mixing things up but mainly to annoy Dan.) I think a better way, and perhaps the only realistic way, to view retribution is as a consequentialist good: a positive value we should seek to maximize, but which can (and must) be traded off against other consequentialist goods. (Note: I mean that retribution should be viewed as an intrinsic good, not just as having instrumental value toward achieving some other good such as welfare or utility. Consequentialism can admit of a plurality of goods, and I think retributive justice is one.)
Why does it matter whether we see retribution as duty or good (assuming we care about it at all)? For (at least) two reasons, I think. First, the bad news, at least from the perspective of retribution's status in criminal theory. If retribution is not an affirmative obligation, then it can't provide a (fully) justificatory account of punishment. One explanation of the justification for having a criminal-justice system (i.e., a system of state-imposed punishment) is that our duty to seek retributive justice obliges us to create such a system. But if retribution is "merely" a good rather than a duty, there is no such obligation. Many things are good (e.g., health, fine art, my friend Dave's fried chicken), but that doesn't mean the government is required to provide them all.
But, on the other hand, viewing retribution as a good rather than a duty also provides a means (otherwise missing, I think) of giving it a role as a real-world guide to policy. While the retribution-as-duty view offers a justificatory theory of punishment, it also (and for the same reasons) fails to provide a prescriptive theory of how to design and implement a criminal-justice system. It offers retribution as an ideal, but does not explain how working legal institutions are supposed to seek that ideal in a world where resource constraints and insufficient information make it impossible to impose retributive punishment on everyone who merits it. In the real world, we need to set law-enforcement priorities, and the retribution-as-duty view provides no clear criteria for doing so (or else provides criteria that seem clearly unworkable or silly).
In short, the retribution-as-duty view can support a moral theory, but not a legal system. Hence, being a lawyer rather than a philosopher (as is, perhaps, all too apparent to any philosophers who read my work), I support the perspective of "consequentialist retributivism."
TrackBack URL for this entry:
Listed below are links to weblogs that reference Retribution: duty, or good?:
"I think a better way, and perhaps the only realistic way, to view retribution is as a consequentialist good: a positive value we should seek to maximize, but which can (and must) be traded off against other consequentialist goods. (Note: I mean that retribution should be viewed as an intrinsic good, not just as having instrumental value toward achieving some other good such as welfare or utility. Consequentialism can admit of a plurality of goods, and I think retributive justice is one.)"
Something is a bit funny to me about this way of putting things. I suppose it's true that consequentialism can accept multiple goods, but is it not impossible to try to maximize more than one of them at a time? That seems to be what you're calling for, but it seems that the very notion of trading off is opposed to maximizing. As I've understood things, this is supposed to be the advantage of using some "neutral" value- utility, welfare, or wealth or whatever, is that then you can coherently think in terms of maximization, while if you have multiple incommensurable goods (which is what you seem to imply) you can't maximize more than one. (You could seek some balance, but again, you are then no longer maximizing in the normal sense.)
Perhaps more importantly, doesn't it seem very odd to say that retribution is a good that we might seek to maximize at all? Even if we have a duty to do it, wouldn't we be better off if we didn't have to do it? Thinking of it as a good we might want to maximize seems to imply that at least in some sense we'd be better off if we had an endless stream of small wrongs we could punish, thereby increasing the amount of retribution in the world. But surely a world where we have no need of retribution at all is better than one where we have all the retribution we need. This leads me to think that retribution isn't plausibly an intrinsic good in an important sense. Or am I not getting what you're trying to get at?
Posted by: Matt | Dec 4, 2008 5:36:49 PM
I'm inclined to agree with Matt. Deserved punishment seems more like the negation of an intrinsic evil (perhaps impunity) than an intrinsic good. That's not technically consequentialism, by the way, since deserved punishment is a constitutive rather than a causal means of negating impunity; put a different way, the negation of impunity is a non-causal result and not a causal consequence of deserved punishment. That's how I'd characterize the evaluative side of retributivism, though I actually think the deontic side is more important. More on that later.
Posted by: Adil Haque | Dec 4, 2008 10:51:50 PM
Simply defining something as "good" is not the same as showing that it is. Why is retribution an "ideal"? E.g., Jesus Christ promoted forgiveness over retribution as an ideal; doesn't the moral system centered around his teachings promote a philosophy you've now defined as not "good"? Many seminarians would disagree.
What is "good" about retribution, for example, if it creates negative, unintended but real consequences? How can we measure this "goodness" to know if it's worth the real, tangible harm done in its name? These are the initial questions that come to mind in light of your thesis.
Posted by: Gritsforbreakfast | Dec 5, 2008 8:46:34 AM
In response to Gritsforbreakfast, and as noted in the thread for my other post, it is true that for present purposes, I have asserted (or assumed) retribution to have some value, rather than proving that it does. For anyone who thinks retribution has no value, these posts admittedly fall on deaf ears. (On the particular question of retribution versus forgiveness, you might read Prawf-host Dan Markel's "Against Mercy," available on SSRN, or the symposium in volume 4 issue 2 of the Ohio State Journal of Criminal Law.)
Posted by: Michael Cahill | Dec 5, 2008 10:10:08 AM
In response to Matt (and maybe Adil, a little bit), a couple of thoughts.
First, in one sense, I think the "maximization" problem is even worse than Matt's critique makes out, since I've offered no basis for assessing different "quantities" of desert even if other goods or variables are ignored. If we can spend the same amount of money to punish one murder or two aggravated assaults, which generates more of the retribution "good"? Tough question. (Note, though, that utilitarians face exactly the same tough question in deciding how much of a "harm" or "welfare setback" different crimes generate relative to each other.)
But my point is that the retribution-as-good view at least gives us a frame for asking and answering that question, whereas the retribution-as-duty view doesn't. A big part of the idea here is to present a construct that could conceivably guide law-enforcement authorities who need to make decisions about how to pursue justice on a budget. For those people, the concern is not (necessarily) about balancing retribution against other goals, but about maximizing retribution subject to external constraints, such as limited resources. I think the "good" view does a better job than the "duty" view in setting up a way to approach that problem, though of course it does not itself answer the problem.
At the broader societal level, you're right to say that we're "balancing" multiple goods rather than maximizing any one of them, and then there's a serious commensurability problem. Justice is good, education is good, so how much do we spend on each? But I take it that problem exists in any robust consequentialist scheme, and again, I'm just trying to provide a lens for viewing the issues, not a final solution to the issues.
As to the question in Matt's last paragraph, I don't think seeing retribution as a good leads to the conclusion that having more crimes to punish would also then be a good thing. (Interestingly, in Fairness Versus Welfare, Kaplow and Shavell use a version of this same claim to critique the retribution-as-duty view.) If we view health as an intrinsic good, that's not a reason to make people sick for the sake of making them healthy again, and also not a reason to wish for the development of new viruses for us to cure. As with health (and other goods), what it means to maximize retribution depends on existing circumstances.
Does that make sense?
Posted by: Michael Cahill | Dec 5, 2008 10:47:19 AM
Michael, thanks for posting on this fascinating topic. I wanted to jump in on the analogy you draw between health as an intrinsic good and retribution. You're of course right that the fact that health is an intrinsic good doesn't mean that we want to make people sick in order to bring them back to health. But this is because health describes the state of an entity, not the process of correcting deviations from that state. That process--medicine, let us say--is instrumentally valuable because it fosters the overall state of health. And this is why we don't have any reason to make people sick in order to cure them: for it's not curing them that is intrinsically valuable, it's that they are in the state of good health.
So, mutatis mutandis, the analogy would suggest that retribution qua "correcting" wrongdoing is more like medicine than it is like health: valuable insofar as it promotes some independently desirable end-state. This state could be specified in familiar terms, e.g. the equalization of undeserved advantage, social harmony or whatever. And this is why, as Matt noted, it seems that it would be better if we didn't have to engage in retribution at all--just like it would be better if no one ever got sick, so you never needed medicine.
What might lie at the root of this is an ambiguity in how we understand "retribution." Either, it means the process of correction, in which case "retribution" certainly looks non-intrinsically valuable, since the ideal state towards which it is aimed would presumably explain the value in the retributive process (doling out punishment). Or, it is meant as an oblique description of that ultimate state (e.g., a society in which retribution in this sense is ideally satisfied would be one in which no one ever committed any crimes just as much as one in which every crime is properly punished.) If the second, Matt's objection doesn't take. But retribution understood in this sense would seem to come awfully close to being vacuous, since even a utilitarian could be on board with fostering this kind of "retributivism." For she could say, "correlating punishment with desert is merely one instrumental means (the process view) of achieving a retributivist society (the end-state view); I just happen to think we can better realize that desirable state of affairs by devising a scheme of deterrent costs such that criminal activity is eliminated or brought to its lowest realizable level."
Posted by: vincent | Dec 5, 2008 12:54:12 PM
Thank you, vincent, for that comment, which I think does clarify what I was trying to say, or at least make plain my ambiguities. And, though this may run afoul of blogospheric norms, I think I agree with just about everything you say.
I would tentatively agree with your final paragraph's claim that I mean to refer to the "second" meaning of retribution you describe, and even with the further implication that, as an empirical matter, promoting "desert" in each individual case may or may not be the best way to maximize "desert" in the aggregate. Most of the first half of my book with Paul Robinson, Law Without Justice, considers specific situations where this might be the case. We tend to be skeptical, though open to persuasion, regarding most claims that rules generating clear specific "deviations" from desert can actually promote desert overall.
Anyway, taking retribution/desert/justice as a sought-after good, it's right that the question of how to maximize it is an empirical one, and the best empirical answer might be that maximizing deterrence also maximizes desert, as you (seem to) suggest. Among other things, if there are fewer crimes, there are fewer risks of error in prosecuting those crimes, which can generate failures of justice (in both directions). Maybe there are other positive "spillover" effects of deterrence onto desert, as well.
But I'm not convinced that that makes my version of retribution "vacuous," because it's only possibly and not definitely the case that pursuing deterrence is the best way of pursuing desert as well, and any good empiricist should be open to the idea that other schemes will work as well, or better. Indeed, my collaborator Paul Robinson has a large body of work taking essentially the inverse position: that the best way to promote utilitarian crime-reduction is for the criminal law to pursue (a particular form of) retributive desert. See, for example, "The Utility of Desert," 91 Nw U L Rev 453 (1997).
This is just an off-the-cuff reaction to an interesting line of comments, which I should think about more, but I thought I'd throw it out there.
Posted by: Michael Cahill | Dec 5, 2008 2:17:43 PM
I just found this blog on the internet. There is fascinating discussion going on in here. I find the topic of retribution vs. utilitarianism punishment intriguing and am happy to find others that care about the topic as well.
Michael, you seem quite learned on this subject and I haven’t read all of your work on the topic. Please forgive me if my comments seem remedial. But, do you agree with the premise that retribution stems from consequentialism? That is, does retribution aim to punish based on adverse consequences of a person’s actions while utilitarianism aims to punish based on the malice or ill-intent or a person?
To the extent that the above premise it true, I pose the following question:
To the extent that:
1. Retribution punishes people based on consequences as opposed to malice, and
2. To some extent consequences can sometimes have little or no relationship with intent, and therefore
3. Those consequences do not always accurately represent the intent of the condemned.
Does punishing based on retribution dish out punishment in a way that bears no relationship to the intent of the accused? To some extent, does this mean that people can be punished based on consequences that are the result of chance circumstances leading to adverse consequences? Can this give rise to, in certain situations, innocent (in intent) people being punished based on random circumstantial events that they find themselves in?
This is a totally shameless plug, but I have spent some time examining this on my blog here: LINK
Posted by: Jacob | Dec 12, 2008 2:26:56 PM