Retributivism and the Problem of Psychopathy

 Psychopathy is a developmental disorder that confers greatly increased risk to life-long behavioral problems.  First, psychopaths are notoriously domineering, exploitative of others, and deficient (or entirely lacking) in proper moral emotions such as guilt, remorse, and empathy.  They are also stunningly hyper-aggressive, predatory, and recidivistic.  Despite the fact that only about .05 to 1% of the population is thought to be afflicted with psychopathy, some estimates suggest that psychopathic individuals could nevertheless be responsible for as much as 30%-40% of all violent crime.  Furthermore, psychopaths are prone to a number of interpersonal problems such as glibness, self-grandiosity, pathological lying, manipulativeness, remorselessness, shallow affect, callousness, and they usually refuse to take responsibility for their actions, often choosing to blame their victims instead.  Finally, psychopaths lead a "chronically unstable and antisocial lifestyle" (Hare et al. 1990, p. 340)—e.g., poor behavioral control, sexual promiscuity, lack of realistic, long-term goals, impulsivity, early behavior problems, and juvenile delinquency. 

In addition to these aforementioned behavioral impairments, psychopaths also appear to have impairments in moral cognition.  For instance, there is evidence that both adult psychopaths and children with psychopathic tendencies perform differently than normals on the so-called “moral/conventional” task (Blair 1996; 1997)—a task whereby participants are asked to distinguish authority-dependent conventional norms from authority-independent moral norms.  Whereas even normal three year old children are able to navigate the task, psychopaths seem to explain all norms in terms of their authority-dependence.  Second, psychopaths tend to consider themselves to be “above the rules”—i.e., they tend not to think that moral and legal norms really apply to them (Chandler & Moran 1990).  Third, when reasoning about real-world rather than merely hypothetical moral dilemmas, they appeal far more often than normals to the moral legitimacy of their own needs and interests (Trevethan & Walker 1989).  Finally, psychopaths seem to misapply and misunderstand moral concepts.  For instance, Hare (1993) provides the following example: “When asked if he experienced remorse over a murder he’d committed, one young inmate told us, “Yeah, sure, I feel remorse.”  Pressed further, he said that he “didn’t feel bad inside about it” (p.41). 

The issue I now want to address is a potential problem psychopaths pose for retributivist theories of punishment.  In short, I think the disorder of psychopathy presents the retributivist with a dilemma—namely, either the retributivist must (a) weaken the conditions for full blown moral desert sufficiently to accommodate the psychopath, or (b) concede that psychopaths have diminished responsibility in light of their cognitive, affective, and behavioral impairments.  If the retributivist opts for (a), then she must set the bar for responsibility so low that she runs the risk of opening the door for full blown responsibility to other groups of individuals we usually don’t hold fully responsible—e.g., children.  If, on the other hand, the retributivist opts for (b), then she is committed to punishing psychopaths less harshly than their non-psychopathic counterparts.  However, given that psychopaths are highly recidivistic and prone to instrumental and not merely impulsive aggression, this means that the retributivist is committed to releasing markedly more dangerous individuals into society while detaining less dangerous non-psychopathic offenders who committed the same kind of crime.  I find (b) especially problematic for the retributivist since it commits her to mitigating the responsibility of a group of offenders who are arguably the apex predators of the world of crime.  

As we continue to learn more about the developmental, genetic, and neural underpinnings of psychopathy, I believe that the problem of psychopathy will put increasing pressure on retributive theories of punishment.  Any thoughts?

p.s. I think the same type of dilemma applies when it comes to the practical implications of compatibilist theories of responsibility.  So, if you’d like to comment on that instead, I would be happy to discuss that in the thread.

Comments

[Joe Campbell]

Nice post, Thomas! I'm teaching Philosophy in Film this summer and we just saw A Clockwork Orange. After doing a bit of research, it struck me that the main character in that film, Alex (Malcolm McDowell), is clearly a psychopath. I tried to motivate the idea that Alex might not be fully responsible for his actions, given his psychological disorder, but most the class did not buy it. We also saw a clip from Fritz Lang's "M," where Peter Lorre plays a child killer and the class was split about whether or not to assign him full responsibility. Thus, while much of the class can find room for thinking that child killers have diminished responsibility there doesn't seem to be the same latitude wrt psychopaths. I wonder if there is any experimental research supporting my informal conclusion.

I'm not sure I see the problem your suggesting, though, since it seems to me that it is open to the retributivist, as well as the compatibilist, to hold that it is appropriate to incarcerate psychopaths on the basis of treatment as opposed to punishment. Why wouldn't that response work?

[Chris Franklin]

Very interesting stuff Thomas. I would be inclined to think that retributivists should opt for (b). You claim that a problem for this option is that it commits retributivists to “releasing markedly more dangerous individuals into society while detaining less dangerous non-psychopathic offenders who committed the same kind of crime.”

But I don’t think result is a commitment of (b). All that follows is that the view entails that we punish psychopaths less harshly then non-psychopaths who perform the same crimes (all else being equal). But we can lock psychopaths up in other institutions that do not amount to punishment. That is, we can protect ourselves without necessarily having to punish.

But even if this result did follow, I am not sure it would be problematic. We need to be very careful here on what we are punishing people for. If some awful person responsibly performs some action A and some really good persons responsibly performs the same action, then I do not think that one person deserves more punishment than the other for this action. Thus, in determining punishment for the psychopath it does not seem his being “highly recidivistic and prone to instrumental and not merely impulsive aggression” is relevant to the degree of punishment (unless of course it is relevant to his performing the crime, i.e. it shows grave indifference, or something like that). This suggests to me that the psychopath’s general character is not relevant to determine degree of punishment for a certain action (set of actions).

[Noah Te Stroete]

When you characterize our correctional system as retributive, only then do you encounter the dilemma you delineate. I don't think the only purpose of the correctional system is punishment. It also serves to keep violent criminals safely away from the general public. This is for the public's good as well as the criminal's. Yes, keeping a psychopath locked away is also for his or her own good. How can giving a psychopath free reign to cause mayhem possibly be good for him or her (in the Aristotelian sense)?

If you consider this aspect of the correctional system, it seems like less of a moral problem and more like a practical solution (in my opinion).

[Philoponus]

Thomas,

Allow me to distinguish between juvenile and adult psychopaths. Why do you think a retributivist would need to concede diminished responsibility to an adult psychopath? Deviant thinking and discontrol mitigate serious criminal behavior only when these conditions are not self-inflicted or the perpetrator has not willfully neglected to treat or remedy his detect. Take the case of Michael Ross that Fischer writes about. Before Ross finally started raping and killing women, he knew for years that he was obsessed with violent sexual fantasies and had increasing trouble controlling himself around women. Yet he sought no professional help and allowed himself to develop into a serial sexual murderer. I agree with the jury that rejected his psychopathology defense and found him deserving of the death penalty.

Children with emerging psycho-pathologies are perhaps a more difficult case for the retributivist. We now have national initiatives to screen children for MDD and OCD. We ought also to be screening for psycho-pathologies, in order to catch the Rosses and Ted Bundies before they blossom. We obviously can’t punish very young children who show only bad tendencies, but by the time these children hit grade school they are often bullying and assaulting classmates, vandalizing property, killing pets, etc. There is the criminal behavior that would justify our punishing these junior psychopaths with appropriate confinement until they commit themselves to treatment to acquire self-control. It could be argued that juveniles who decline to be treated for their diagnosed pathologies should face aggravated, not mitigated, punishment. I don’t see that a retribuvist who takes this line encounters your dilemma, but maybe I’ve missed something.

[Thomas Nadelhoffer]

Joe, Chris, and Noah,

Since you all made similar points concerning treatment/prevention, etc., I thought I would first say something on this front. For starters, treatment is certainly not a retributivist justification for punishment. After all, rehabilitation, just like deterrence and prevention, is a forward looking justification. And given that some retributivists think giving people what they deserve is the *only* legitimate justification for punishment, they simply do not have the avenue you suggest available to them. So, if we are talking about punishment (i.e., incarceration in jail or prison), the retributivist really does seem to be stuck mitigating sentencing for psychopathy if she opts for (b) rather than (a).

That being said, it is worth pointing out that since the Sentencing Reform Act of 1984, at least at the federal level, only retributivist/just deserts considerations are supposed to be considered at the sentencing phase of criminal trials. At the time, the Sentencing Commission distinguished two main views concerning why we punish:

"Some argue that appropriate punishment should be defined primarily on the basis of the principle of “just deserts.” Under this principle, punishment should be scaled to the offender’s culpability and the resulting harms. Others argue that punishment should be imposed primarily on the basis of practical “crime control” considerations. This theory calls for sentences that most effectively lessen the likelihood of future crime, either by deterring others or incapacitating the defendant."

At the end of the day, the Commission ultimately came down on the side of those who favored a just deserts approach to criminal punishment. On this view, the likelihood of future dangerousness is largely irrelevant to sentencing, which instead ought to focus on features of the past action that affect what the agent deserves. In short, the Commission decided that punishment should be backward-looking rather than forward-looking.

As a result, since the mid-1980s, “virtually all of the variables that potentially could be used as scientifically valid risk factors for violence under a forward-looking consequentialist ‘crime control’ theory of punishment are explicitly excluded from consideration in federal sentencing procedures” (Monahan 2006, p. 391). The situation at the state level, on the other hand, is less clear, partly because a number of states don’t have sentencing guidelines at all. But, at least when it comes to federal sentencing, the strategy you’re suggesting with psychopaths—namely, detain them longer despite the fact that they are, ceteris paribus, less deserving than non-psychopaths—is (a) not open to retributivists qua retributivists, and (b) not open to legal decision makers more generally.

Now, Chris suggests we could lock psychopaths up in “other institutions” instead. Unfortunately, as things presently stand, the disorder of psychopathy is neither sufficient for civil commitment nor is it sufficient for the insanity defense. Because psychopaths are not delusional—i.e., because they know, in some sense, what society permits and what it forbids—they are only targets for criminal prosecution. Moreover, psychopathy is an insufficient insanity defense for the same reasons. As such, under most existing criminal and civil statutory schemes in this country, psychopaths can only be punished/incarcerated for their crimes. Sexual predator statutes open up another avenue—which we can talk more about if anyone’s interested—but I will set that aside for now since that would only apply to psychopaths who are also sexual offenders.

Finally, since Joe mentioned treatment, it is worth pointing out that not only are psychopaths resistant to treatment, but most treatment modalities actually make them more likely to recidivate! So, whether you give them group therapy or individualized cognitive behavioral therapy, they end up being more rather than less likely to reoffend. There are several explanations for this puzzling phenomenon, but the one that makes the most sense to me is that the therapeutic experience inadvertently teaches psychopaths how to better understand and manipulate people. That being said, there is one exception to this finding—namely, juveniles who have psychopathic tendencies. Michael Caldwell is presently doing some truly ground breaking work with juveniles in Wisconsin whereby he has shown that the treatment program he has developed has a rehabilitative effect even on the most violent teens with psychopathic tendencies. But as things presently stand, nothing similar has been shown to be possible in adults. As such, not only can the retributivist qua retributivist not use treatment as a justification for detaining/punishing psychopaths, but even a rehabilitative theorist cannot use this as justification.

[Thomas Nadelhoffer]

Philoponus,

First, technically, juveniles cannot be psychopaths—they can only display varying degrees of psychopathic tendencies. For instance, the PCL-R—which is the most widely used diagnostic tool for psychopathy—is not used on children or juveniles. Instead, researches use the PCL: YV (i.e., Youth Version) which cannot be used to diagnose psychopathy proper.

Second, I find your claims about individuals “allowing” themselves to turn into violent sexual predators (a) odd, (b) empirically strained, and (c) socio-economically myopic. Imagine a 10 (or 12 0r 15) year old who realizes he has violent sexual fantasies but who doesn’t have the available resources—emotional, familial, and/or financial—which would enable him to seek treatment. Are you really suggesting that when this child grows up, he is fully responsible for his disorder? That strikes me as, well, odd and myopic.

Now, as for Ross—who was an adult—if he was indeed a sexual predator *and* a psychopath and given that there are no existing treatment modalities that could have helped him qua adult psychopath, your suggestion that he should have done something is empirically strained. After all, what are you suggesting he should have done to stave off the disorder?

Third, as for your suggestion concerning identification, detainment, and prevention, let me just point out that Caldwell’s aforementioned research aside, given that (a) there are virtually no effective treatments for psychopathy (especially for adults), and (b) as we better understand the genetic and neural underpinnings of the disorder we will be able to identify (potential) psychopaths at increasingly earlier ages, your suggestion could potentially lead to our identifying budding psychopaths at age 5-6 and removing them permanently from society. That’s fine as far as it goes, I suppose, but notice there is no way of motivating that on retributivist grounds. As such, your point seems irrelevant to my earlier suggestion concerning the problem the disorder of psychopathy poses for retributivism.

That being said, I think the question you raise is both interesting and important more generally and one I am presently working on in my research on neuroprediction and the law. But for now, I would prefer to limit our attention to the problem of psychopathy specifically as it relates to retributivism. Perhaps in a future post I will focus on psychopathy and prediction.

[AA]

Thomas,

Perhaps neither (a) nor (b) is necessary and retributivism can handle the case as is. We might simply say that we don't punish children retributively not because they are not just as culpable as adults, but because they have a much greater ability to change and improve. With children, there is a future potential "mature, innocent self" that is being foreclosed on and punished unjustly in order to punish the current "immature, guilty" self. That this radically different future self will come into being is far more uncertain in the case of adults and even more so in the case of adult psychopaths.

Well, then, one might ask, what about a paranoid schizophrenic, who murders people with wanton abandon? Well, if he has significant moments of lucidity which are at a discontinuous break with the paranoid self, that might again be a reason not to apply retributive punishment. But lacking that and supposing he has little hope of ever changing for the better, I'm not sure retributivists should not subject such a person to harsh punishment.

[Joe Campbell]

Thomas,

Again, very interesting stuff! A few responses to your comments above. This is a bit out of my area, so forgive me if I say something stupid.

You wrote, in response to me, Chris, and Noah: “For starters, treatment is certainly not a retributivist justification for punishment. After all, rehabilitation, just like deterrence and prevention, is a forward looking justification. And given that some retributivists think giving people what they deserve is the *only* legitimate justification for punishment, they simply do not have the avenue you suggest available to them.”

Can’t speak for the others but I was not offering treatment as a retributivist justification for punishment, but a justification for incarceration. Certainly one could hold that punishment is justified for retributivist reasons only BUT that we can incarcerate folks for practical reasons, as well, under the guise of treatment, as “enemy combatants”, etc.

When you add that “since the Sentencing Reform Act of 1984 (SRA), at least at the federal level, only retributivist/just deserts considerations are supposed to be considered at the sentencing phase of criminal trials” this complicates the matter. I gather that your point is that there is no basis, on retributivist grounds alone, for the long-term incarceration of psychopaths. But whatever problems one has in justifying the SRA, one could still be a retributivist about PUNISHMENT and provide a justification for long-term incarceration on, what you call, “forward-looking” considerations – so long as the incarceration is not a form of punishment. This says nothing about your criticism of the SRA, which is spot-on. It is just a logical point about the retributivist view of punishment.

Note also this quote from above: “Some argue that appropriate punishment should be defined primarily on the basis of the principle of ‘just deserts.’ Under this principle, punishment should be scaled to the offender’s culpability and the resulting harms.” Thus, it is not just the offender’s culpability that matters. One could admit that psychopaths have diminished responsibility but that they still deserve long-term sentences (given these guidelines), provided that their crimes are particularly severe, which might be the norm for psychopaths.

But I take it that your main point is that sentencing guidelines cannot be based on purely retributivist considerations, and this seems right to me. The SRA is problematic. But a retributivist about punishment can agree with this, as well.

Two last points. First, I don’t think it matters that our current forms of treatment for psychopaths don’t work. Of course, on a social level it matters a lot but if we’re just talking about the logic of retributivism about punishment, it doesn’t matter.

Second, one of the assignments for my Philosophy in Film course was Darrow’s defense of Leopold and Loeb. I’m embarrassed that I had never looked at it before! My impression was that Darrow gave a kind of skepticism about moral responsibility argument – L&L were not fully responsible because, given the conditions of their circumstances, they had no choice about their actions, their actions were not up to them. There is some of that, of course, but it is particularized to L&L in the sense that he thinks that their age (17/18) is a mitigating factor. If it is sound, the argument could be used to show that no one of that age is fully responsible for his actions. Likely this is true.

[Eddy Nahmias]

Thomas, thanks for the very useful discussion of some of the problems psychopaths pose to punishment theorists. I would like to hear more about why psychopaths pose problems for compatibilists (other than the problem that one might kill us and not even feel bad about it!).

This mostly summarizes earlier comments, but (1) I think there may be some responses to the first horn of your dilemma (i.e., there may be reasons to hold some psychopaths responsible for not using their capacities for self-control to avoid committing crimes), and (2) I don't understand at all why retributivists are committed to "releasing markedly more dangerous individuals into society while detaining less dangerous non-psychopathic offenders who committed the same kind of crime." Being a retributivist does not force one to abandon all forward-looking considerations (see below on the Sentencing Act). The retributivist may have to say that the psychopath deserves less *punitive* punishment than others, which is counter-intuitive given that we tend to feel more severe retributive impulses towards the nastiest and least repentant psychopaths, but it's much less counter-intuitive than the theorist who is a skeptic about retribution and desert across the board.

Nonetheless, the retributivist can, like any other punishment theorist, use scientific information about psychopaths to suggest that some of them need to be quarantined, perhaps for life. If we assume that the psychopath does not deserve (as much) retributive punishment, such quarantine should not be designed to be painful, though to make it safe, it may look a lot like solitary confinement.

Finally, I think you've oversimplified the way our legal system thinks of punishment. State systems certainly consider factors relevant to general and specific deterrence (and in more limited cases, rehabilitation), though I certainly agree with you that the system is out of whack in how it punishes (in theory and especially in practice). And the 1984 Sentencing Act does include forward-looking features. It eliminated "rehabilitation" as a factor to consider in sentencing, but it did include specific and general deterrence along with incapacitation and retribution as relevant factors, and even some specified forms of life-improvement (see (D) below):

"18 United States Code Section 3553(a):
(a) Factors To Be Considered in Imposing a Sentence.— The court shall impose a sentence sufficient, but not greater than necessary, to comply with the purposes set forth in paragraph (2) of this subsection. The court, in determining the particular sentence to be imposed, shall consider—
(1) the nature and circumstances of the offense and the history and characteristics of the defendant;
(2) the need for the sentence imposed—
(A) to reflect the seriousness of the offense, to promote respect for the law, and to provide just punishment for the offense;
(B) to afford adequate deterrence to criminal conduct;
(C) to protect the public from further crimes of the defendant; and
(D) to provide the defendant with needed educational or vocational training, medical care, or other correctional treatment in the most effective manner..."

[Thomas Nadelhoffer]

AA,
I am unsure what you mean when you state that the retributivist (R for short) can “handle the case as it is.” The case I raised involved neither children nor paranoid schizophrenics—each of whom already have legal excuses/defenses under the criminal law. Rather, the case I raised was psychopathy. Given that psychopaths are not delusional, they do not fit your case of the schizophrenic. Given psychopaths' moral impairments, however, you could view them as “child-like”—but then if you go on to hold psychopaths fully responsible, you run into the first horn of the dilemma since you would presumably be committed to holding children fully responsible as well. I suspect you were trying to address this in your comments about children, but to be honest, I didn’t really understand your point. Normally, we don’t punish them as harshly because they don’t have fully developed cognitive and conative moral psychologies and hence don’t deserve to be treated on par with adults who do. Instead, you are suggesting that they really do deserve to be punished just as much as adults but we chose not to do so in light of their potential future moral selves. Is this your view? If so, you would need to say more. As it stands, I either disagree with what you’re saying or I don’t understand what you’re saying.

Joe,

First, it sounds like you had a really interesting course! That being said, I wanted to say a bit about your comments about incarceration vs. punishment. For clarity’s sake, let’s distinguish incarceration from commitment/detention. In this country, incarceration is the primary form of punishment for criminal offenses. It is intended to be unpleasant and this unpleasantness is partly why we consider it punishment. Indeed, the more we think someone deserves, the longer the sentence—i.e., sentences/incarceration can be scaled to accommodate desert. Civil commitment/detention, on the other hand, is used for dangerous offenders and non-offenders who do not satisfy the normal standards for criminal responsibility—e.g., delusional schizophrenics, bi-polar individuals suffering manic episodes, etc. With this distinction in hand, I think it is clear that the R qua R must argue for commitment/detention and not incarceration for psychopaths (assuming the R opts for the second horn of my aforementioned dilemma). After all, to sentence the psychopath qua enemy combatant rather than qua offender to a lengthier sentence that he deserves is to run afoul of the R’s maxim that we should only punish offenders according to their desert. In a recent issue of Neuroethics, Stephen Morse—a compatibilist and a retributivist—argued that psychopaths should be committed rather than incarcerated even if the law presently doesn't afford this option to legal decision makers. So, perhaps you have something like that in mind. But what the R cannot say is that even though the psychopath deserves less in light of his moral impairments, we should nevertheless punish him more—i.e., that we should give him a longer sentence than he deserves in the name of the greater good. Indeed, if the R opts for the second horn of the dilemma, it looks like he is forced to opt for something like the view Morse has developed lest the R ends up being saddled with the view that because the psychopath is less deserving he should receive a lesser sentence than non-psychopaths who committed the same kind of offense.

Now, you go on to state that “whatever problems one has in justifying the SRA, one could still be a retributivist about PUNISHMENT and provide a justification for long-term incarceration on, what you call, “forward-looking” considerations – so long as the incarceration is not a form of punishment.” As a conceptual point, this is surely right as far as it goes. But notice, that (a) if we commit psychopaths rather than incarcerating them, and (b) there are no treatment modalities that we can offer them while they are indefinitely (if not permanently) committed, then the distinction between commitment and incarceration runs the risk of collapsing. After all, the primary difference in a prison and a civil commitment facility is that the former is designed for punitive purposes and the latter is inherently rehabilitative. It is the rehabilitative component that justified committing people—especially when they have not broken any laws. So, I am unsure the R qua R can adopt the strategy you outline here lest he blurs the boundary between incarceration and commitment (or between the criminal law and civil law). This is partly why most legal scholars and behavioral scientists have nearly universally rejected the Supreme Court’s decisions concerning sexual predator statutes in Kansas v. Hendricks and Kansas v. Crane. In writing for the majority, Justice Thomas suggested that these statutes are civil and not criminal and hence the appellants did not have legitimate due process claims, etc. Most commentators call the laws “quasi-criminal.” I think the same could be said for the scheme you’re suggesting. As such, I think it settles awkwardly with the basic R maxim I mentioned earlier. If non-rehabilitative commitment is just punishment in disguise, then the R simply doesn't have that option available to him.

That being said, I also want to address you suggestion that “One could admit that psychopaths have diminished responsibility but that they still deserve long-term sentences (given these guidelines), provided that their crimes are particularly severe, which might be the norm for psychopaths.” It’s true that even with diminished responsibility, if the offense was severe enough, a psychopath could still deserve a long sentence. That’s true as far as it goes. But notice that if we imagine two offenders who commit the same type of crime under similar circumstances, but one is psychopathic and the other is not, the R qua R will have to judge that the psychopath deserves a less lengthy sentence that his non-psychopathic counter-part—which would mean than we would essentially be freeing a more dangerous offender before freeing a less dangerous offender even though they committed roughly the same crime.

Finally, I just wanted to point out that Darrow is a favorite figure of mine and arguably the most important and influential lawyer in this country’s history. To my knowledge, his “innocent by reason of determinism defense” in L & L’s trial had never been used before and has not been used since. However, the “youth defense” component of his defense was very prescient indeed! Of course, there are still a number who are unpersuaded by even this and some of these individuals are presently sitting on the Supreme Court!

Eddy,

Thanks, as always, for your feedback. I will provide a more detailed response later. For now, I just wanted to point out that what I posted earlier in the thread were comments from the Sentencing Commission's discussion concerning their justification for drafting the guidelines as they did. So, while you're right that forward-looking considerations seem permissible as stated by the SRA itself, these considerations still only become legally relevant by first passing through a retributivistic hurdle. But more on that later. On the one hand, the issue of sentencing is really a complicated mess (and an ever-evolving one, especially as of late). On the other hand, I really need to finish something else today. So, I will have to issue a promissory note until tomorrow!

[Matt King]

Thomas,

You write of the difference between civil commitment and criminal punishment that, "[it] is the rehabilitative component that justified committing people", having already suggested that rehabilitation is nearly (at least at present) impossible for adult psychopaths. Committing such folks would amount to, as you put it, "quasi-criminal" treatment, which I take to imply that it would be quasi-punishment. But maybe you can say a little bit more about why.

For suppose a relatively small segment of the population developed a virus that, while only causing mild flu-like symptoms for them, was seriously fatal to all others (and terribly contagious). Now, I would think that quarantining such individuals may be justified, and, if justified, would likely be a result of protecting others from potential harm. Now, let us suppose further that the virus is (at present) pretty much incurable.

What we have, then, is a segment of the population which is isolated and greatly restricted in its freedoms. This would be a hardship and looks very much like the treatment criminals receive. But we wouldn't justify it in terms of desert, and it wouldn't, so far as I see, count as punishment.

Would such treatment count as quasi-punishment in your view? And would it pose similar problems for retributivists? This would be an interesting result, insofar as I think scenarios such as this one are familiar and haven't been thought to pose any trouble for retributivists before.

[Tamler Sommers]

Hi Thomas, good discussion--I just want to hop on board with the other commentators and argue that you can be a retributivist without being an absolutist and excluding all forward-looking considerations. In other words, you can see value in giving people the punishment they deserve but at the same time recognize that this isn't the only value or the only reason to punish. Prevention, deterrence, and rehabilitation can have their place in a moderate retribitivist theory of punishment. I don't even think the retributivists have to call the punishment something else (like incarceration or quarantine) in these cases. If anything, this just adds insult to injury if it amounts to the same thing.

People think of Texas as the paradigmatic retributivist state when it comes to punishment, but unlike many states with capital punishment, Texas law requires that the criminal pose a future threat in order to be given the death sentence. Without that, no matter how atrocious the crime, he or she can't get the death sentence. This seems like a prime example of forward-looking and backward looking considerations blending together in criminal law.

[Noah Te Stroete]

AA said:

"Well, then, one might ask, what about a paranoid schizophrenic, who murders people with wanton abandon? Well, if he has significant moments of lucidity which are at a discontinuous break with the paranoid self, that might again be a reason not to apply retributive punishment. But lacking that and **supposing he has little hope of ever changing for the better**, I'm not sure retributivists should not subject such a person to harsh punishment."

It is anti-empirical to suppose that any paranoid schizophrenic has little hope of ever changing for the better. This supposition is simply contrary to the evidence. Paranoid schizophrenics have the best prognosis (and a very good one at that) of any subtype of schizophrenic given they receive the right medication and treatment. Better yet is the prognosis for people with schizoaffective disorder (which is what I have).

Furthermore, I don't know if your assumption is also that paranoid schizophrenics are murderers or if you were just giving an example of a paranoid schizophrenic who just happened to murder repeatedly. If it was the former, then I must say that I am utterly disgusted with that comment. There is nothing in the DSM-IV that says committing murder is a symptom of paranoid schizophrenia. To even be ambiguous with your wording that someone reading this discussion thread could construe what you said in this way is reckless, ignorant, and irresponsible.

Also, one of the points of this discussion is that psychopaths cannot be rehabilitated *and* have debilitated moral cognition which *does* pose a problem for the retributivist. The case of a paranoid schizophrenic is not even analogous.

[AA]

Thomas,

Yes, I am suggesting that Rs could hold psychopaths fully responsible for their crimes, and that children need not be so held because of their potential future moral selves. What makes someone deserving of punishment may be the commission of a bad act with some de minimis degree of volition. However, perhaps retribution is only appropriate against the immoral self, and if there is a large chance that punishment will fall on a current or future moral self, retribution may be inappropriate or at least need to be carefully weighed against its inevitable overbreadth.

[Thomas Nadelhoffer]

Tamler,

A detailed reply to you, Eddy, and others will have to wait a bit since I am laboring under looming or already passed deadlines. But for now I just wanted to point out that you are simply incorrect to suggest a retributivist can favor deterrence or prevention over desert. In short, to be a retributivist of any stripe--whether it be stringent, moderate, or meek--is to believe that we ought never give people more punishment than they deserve. To give people more punishment than they deserve is to run afoul of one of the tenants that holds all forms of retributivism together. So, while deterrence and prevention may indeed be viewed as positive side effects of punishing people, it is not (and cannot be) part of the justification of punishment. Indeed, allowing these competing justifications enter the picture just makes it all the more likely that you will punish the offender more than he deserves.

Let's imagine, for the sake of argument, that the retributivist can establish that P deserves x amount of punishment. Let x be 1 year in jail. Now, let's imagine that punishing P x + 1 year will maximize deterrence (and add an additional year of prevention). The retributivist qua retributivist has to be committed to punishing x rather than x + 1. Why? Because x + 1 is more than P deserves and the basic retributivist intuition is that you cannot punish people more than they deserve. Indeed, one of the main selling points of the theory is that unlike consequentialist justifications, retributivism can purportedly prevent unjust punishments--i.e., punishing people more than they deserve.

On the view you and others are pretending is retributivistic, you could punish people more than they deserve. Call that desert-based consequentialism if you'd like. But retributivism it ain't. But more on that later. This short (and admittedly inadequate) comment will have to be enough grist for the mill for the make believe retributivists for now :)

[Justin Capes]

Thomas,

Why can't retributivist principles be understood merely as prima facie duties? We have a prima facie duty, a retributivist might think, to punish responsible wrongdoers, and we have a prima facie duty not to punish them more than they deserve. But all prima facie duties can--in principle, at least--be overridden by other stronger duties. The position I've just described could fairly be described as retributivist, I would think, but it leaves room for other forward looking considerations.

[Thomas Nadelhoffer]

Here is an example for Justin, Tamler, and others:

Imagine a philosopher who claims to believe that punishing culpable wrong-doers in order to give them what they deserve is a prima facie duty. Now, imagine he also happens to believe that he has a prima facie duty to punish culpable wrong-doers in order to deter and prevent crime. Finally, imagine he believes that, in practice, the latter duty always has more weight/overrides the former duty. As a result, he always thinks you should punish offenders as much as is needed to deter and prevent even if it means punishing them more that they deserve.

On your view, this philosopher counts as retributivist even though he never actually thinks actual offenders should ever be punished solely as much they deserve. Obviously, you are free to define "retributivism" or any other term however you see fit. But, I would simply say that in this case, the person is a consequentialist who sees the theoretical and potential practical value of desert and retribution but who never thinks these considerations ought to carry the day when it comes to actual punishment. At best, I would say this philosopher is a retributivist-sympathizer, in principle, but not a retributivist in practice.

Perhaps the mistake is that y'all seem to think that if someone believes in desert at all that makes them a retributivist. I think this is incorrect. What makes someone a retributivist is the role desert plays in justifying and/or implementing punishment. The view you guys are trying to carve out makes it possible for desert to play little to no role at all for the so-called retributivist. That seems like an unmotivated and untenable view to me. But here again, different stipulative strokes for different folks.

[Matt King]

I hope I am excluded the general comment about make believe retributivists. I agree with Thomas that retributivism should be characterized by the role desert plays in the justification of punishment (though I'm not as convinced as he that we can so easily dismiss the suggestions in this thread).

But I disagree, or at least fail to see, the precise problem posed by psychopaths. Not all ill-treatment counts as punishment, and the retributivist can make that distinction more easily than, say, consequentialists punishers.

The parents of a teenager who withhold the car keys and thus prevent him from attending an out-of-town party perhaps harm him, but they aren't punishing him if their justification is to keep him out of harm's way (they believe he is likely to get into some sort of trouble or other if allowed to go). But if they keep the car because he's been truant at school and staying out all night, then the harm visited by keeping him from the party now looks like punishment.

The deprivation is the same - but only one counts as punishment. Retributivist parents wouldn't justify the first denial by an appeal to desert, nor would they have to.

And for my money, I don't see why there couldn't be perfectly good justifications for visiting deprivations upon psychopaths without them being punishments (as the quarantine case suggests).

[Thomas Nadelhoffer]

Matt,

No, that sarcastic remark was aimed at Tamler's flirtation with his retributivist feelings! :) I am sorry I didn't reply to your comment earlier--but I was in a rush and just responded to the most recent comment by Tamler. As for your quarantine example/question, here's a scenario:

Let's imagine that the disease confers a greatly increased risk for extreme acts of violence, deception, remorselessness, etc. such that any individuals who contract the disease pose a serious threat to society. So, we quarantine them. Now, ordinarily, when we talk about quarantine, we have non-violent-inducing illnesses in mind. As such, while we do have to detain the infected, we are obligated to do everything possible to make their "stay" pleasant. After all, we are depriving them of their freedom even though they did nothing culpable. As such, they can freely mingle with one another, move around freely, play games, watch tv, eat decent food, etc. Plus, there is no reason to put them in cells, constantly invade whatever privacy they have, etc. Now, this is not the case with our violent "patients." Instead, they have to be closely and constantly monitored since they pose a potential threat both to other "detainees" and the workers in the detainment facility. As such, their freedom of movement, their activities, etc. within the facility must be watched and curtailed.

Now what is the practical difference in how we are treating these detainees and how we treat garden variety offenders who are incarcerated? If the answer is nothing, then what we are doing to the "violent ill" is de facto punishment. So, while your quarantine example poses no problems for retributivists, that's only because typical cases involving quarantine share two properties: (a) the quarantined did nothing wrong, and (b) even though we have to detain them, it is possible for us to make the detainment fairly pleasant. However, (a) and (b) do not apply in the case of psychopaths--who share much more in common with my hypothetical detainees with the violence-inducing disease. Once you flesh out what quarantine would like in the case of psychopaths, it starts looking a lot more like punishment--if not in intention, then in implementation. Or, so it seems to me.

[Thomas Nadelhoffer]

Tamler,

Texas is a tricky example since the "dangerousness" threshold you mention--which was deemed constitutional by the Supreme Court in both Jurek v. Texas (1976) and Barefoot v. Estelle (1983)--requires virtually no real evidence of future dangerousness. In Barefoot's original trial, Dr. Grigson--aka "Dr. Death"--testified that he was *certain* Barefoot would kill again despite the fact that (a) Grigson never met Barefoot, (b) never actually reviewed his file, and (c) arrived at his certain judgment based on a hypothetical scenario. The Court found that the statutory requirement of a finding of future dangerousness in capital sentencing was acceptable even though the APA filed an amicus brief that pointed out that clinical predictions on this front are typically only right 1/3 times--a frightening fact that the Court explicitly acknowledged and explained away in their decision. Based partly on the Court's earlier ruling in Jurek that it couldn't "disinvent the wheel" when it comes to predictions of dangerousness, in Barefoot the Court held that the Texas statutory scheme and the State's reliance on unreliable "experts" are both constitutionally acceptable.

So, while you are right that Texas includes this forward looking element, the bar is so low that it seems like a not so sneaky way of pretending to care about social safety while dispensing retributivist justice to the evil bastards they think deserve to die. The ruling in Barefoot is still the law of the land when it comes to predictions in the law. But that is a story for another day. For now, I just wanted to point out that Texas is not *really* as progressive--i.e., forward looking--as you suggest since they could just as well be flipping coins.

[Justin Capes]

Thomas,

Imagine a Rossian deontologist who believes that one has a prima facie duty to keep promises. Suppose, however, that every time he actually makes a promise circumstances arise in which he could bring about an astronomical amount of good by breaking his promise. Would this person simply be a consequentialist who sees the practical value of promising keeping?

I would say no. I would call this guy a deontologist. Similarly, I'm inclined to call the guy in your story a retributivist. Indeed, I've always understood Ross as a retributivist. But he allows that other considerations can (indeed, on his view, must) play a role in the real world.

In any event, suppose we accept that retribitivism has to do with the role desert plays in justifying punishment and not merely with desert itself. Still, the view I (and perhaps others) are attempting to cut out isn't unmotivated. It might be motivated by the sorts of considerations that Ross raises, for example. Is it untenable? One day I hope to get around to answering that question.

[Thomas Nadelhoffer]

Justin,

Here is a question for you: What do you take to be the necessary conditions for a theory of punishment to count as retributivist such that the hypothetical scenario I provided earlier counts? I have suggested that the following is a necessary condition: To be a retributivist, you must minimally be committed to the idea that it is never acceptable to punish people more than they deserve in the name of the greater good. On my view, if you give this up, you end up peddling a consequentialist theory of punishment in retributivist garb.

Also, it's worth pointing out that in the real world, we can always prevent more crime by punishing offenders more than they deserve and often (if not always) deter more crime by punishing offenders more than they deserve. So, the Rossean you mention is arguably in the same position as the hypothetical philosopher I mentioned above.

[Matt King]

Ok, Thomas, I guess my response to the hypotheticals is that, to the extent that the case mirrors a quarantine scenario, I'm not inclined to think of the *harshness* of the situation as punishment.

I'm not clear on what de facto punishment is. Painful or harmful treatment? I think it must be more to be punishment - indeed, I'm inclined to think punishment must be a redressing of some kinds (which may be objectionable as it makes punishment retributivistic by definition - but no matter here).

You write, "it starts looking a lot more like punishment--if not in intention, then in implementation." But I should think that things like intentions make all the difference here. The same treatment can be or not be punishment depending on the stance of the would-be punishers (as in my parents example).

Sometimes, whether some treatment is justified may turn on whether it is punishment or not - so, say, imprisoning an individual may be justified if it's in response to his responsibly killing someone, but not if he's just a general jerk. But that imprisonment of this sort is justified when punishment doesn't rule out other potential justifications when it isn't punishment. And the quarantine model is just one example.

[Tamler Sommers]

"To be a retributivist, you must minimally be committed to the idea that it is never acceptable to punish people more than they deserve in the name of the greater good. On my view, if you give this up, you end up peddling a consequentialist theory of punishment in retributivist garb."

That's simply not true. What you're describing as a "minimal" commitment is an extreme and absolutist form of retributivism that few retributivists have accepted. On this view, it's not acceptable to punish one person more than they deserve even if the whole world would be destroyed if you don't. Kant may have subscribed to this kind of insanity, Hegel too for all I know. But contemporary retributivists like Primoratz (and not so contemporary ones like Ross--as Justin notes) explicitly reject absolutist forms of their position. According to a reasonable moderate form of retributivism, you should not punish people more than they deserve unless there are overwhelming or at least very significant consequentialist reasons to do so. And when you are compelled to punish people more than they deserve, you try to make it up to them as best you can. What makes them retributivists and not consequentialists is that that they would not punish people more than they deserve for miminal or moderate consequentialist benefits. The consequentialist is committed to punishing people even if there is just a slight net gain in welfare or whatever.

Preventing psychopaths from killing innocent people would arguably be an overwhelming consequentialist reason to keep psychopaths locked up--one that overrides considerations of just-deserts from the (reasonable) retributivist's perspective.

[Thomas Nadelhoffer]

Tamler,

No, all retributivists have adopted the "never punish more than deserved" principle. To the extent that the other folks you're talking about give that up, they aren't really retributivists! :) Once you give up the absolutist part, at the level of implementation the view you mention will nearly always cave in to consequentialist pressures for the reasons I mentioned above in my response to Justin:

"Also, it's worth pointing out that in the real world, we can always prevent more crime by punishing offenders more than they deserve and often (if not always) deter more crime by punishing offenders more than they deserve. So, the Rossean you mention is arguably in the same position as the hypothetical philosopher I mentioned above."

It's not just that non-absolutist retributivists will have to punish people more than they deserve in some "sky is falling" and other doomsday cases, it's that in nearly all garden variety cases, the consequentialist considerations will (and arguably should) carry the day. So, what's the difference between this kind of view and a straight forward consequentialist view--especially at the level of implementation?

That being said, I will pose the same question to you I posed to Justin: Just how little consideration to moral desert can a theorist give and still count as a retributivist on your view? After all, I have already tried to show that on the sort of view you're arguing for, someone could be a retributivist who nevertheless always judged that we ought to punish offenders more than they deserve in the real world. If you want to call that retributivism, that's fine as far as it goes. If, on the other hand, you don't think your view commits you to this, why not?

[Thomas Nadelhoffer]

I just wanted to thank everyone for their feedback thus far. I have thoroughly enjoyed and benefited from the discussion! Obviously, I think this is an interesting topic and I am delighted others agree--even if they disagree with my (admittedly idiosyncratic) views concerning the nuts and bolts of retributivism. That being said, I wanted to let everyone know that owing to looming deadlines and a recent elbow injury (damn you, jiu jitsu!)--which limits how much and how fast I can type in any one sitting--I am unfortunately going to be unable to reply to comments for a few days. So, feel free to continue discussing amongst yourselves (as I will be approving comments as they're submitted). But it will be a few days before I can rejoin the discussion. I nevertheless wanted to thank you for playing along and encourage you to keep it going!

[Justin Capes]

Thomas,

In answer to your question, I'd simply repeat some of what Tamler said, though I'll add a few remarks of my own. Following Tamler, I would say that part of what makes someone a retributivist and not consequentialist is that they would not punish people more than they deserve for mininimal or moderate consequentialist benefits.

Practically speaking this may not look very different from consequentialism for precisely the reasons you mentioned earlier, but whether it does or not depends on the circumstances. Does this make the view consequentialist in retributivist garb? I'm inclined to think not.

Again, I think the comparison to normative theory is instructive. Practically speaking, the judgments that many deontologists will make about what we ought to do in actual circumstances and the justifications they offer for these judgments mayn't differ at all from the judgments and justifications offered by consequentialists, but that doesn't make them consequentialists. They are deontologists (in part) because they would not break promises, or kill, or lie, or steal, or betray a friendship, or what have you for minimal or moderate consequentialist gains. Whether their judgments about what we ought to do and the reasoning behind these judgments mirror consequetialists' will depend on the circumstances.

And sorry to hear about your injury. It sounds painful. That said, however, if it happened as a result of your letting yourself get trapped in an arm bar, I'll be a little disappointed.

[Thomas Nadelhoffer]

Justin,

Two very quick responses:

First, and this applies to Tamler as well, I think you need to say something a bit more concrete when it comes to the minimal or moderate consequentialist benefits that you don't think trump the weakened retributivist standards you are taking as distinguishing features of your view. For instance, are you only going to flip flop to consequentialism in doomsday cases? I, for one, think you owe an account on this front to the pure retributivists like Kant (and Moore, more recently) no less than the consequentialist like me who think you're really just desert-sensitive consequentialists. To establish your retributivist bona fides, you really do need to say something more substantive about where your retributivism ends and your consequentialism begins--not necessarily in this blog post, of course, but somewhere down the line!

Second, it was not an arm bar--but an elbow lock I had never seen. Perhaps that's because the guy was a second-degree black belt from Brazil who trained under Carlos Graice Jr. So, continue feeling sorry for me and don't be disappointed! :) Had I known better, I could have avoided tearing tendons and ligaments! But alas...

[Dan Speak]

Just coming back to the discussion after a few days. I have nothing to add but wanted to note that I really have learned a lot from it. Thanks people. Now THAT's a blog!

[Tamler Sommers]

Thomas,

What exactly is a 'desert-sensitive' consequentialist? Isn't that a contradiction in terms (since desert is a backwards-looking concept)?

[Thomas Nadelhoffer]

Tamler,

You tell me, it's your crazy view. Seriously though, I take it mixed theorists like Rawls and Hart are desert-sensitive consequentialists. When it comes to justifying the institution of punishment they're consequentialists, but at the level of implementation, they think desert must at least partially constrain how much we punish. I prefer not to call mixed theories retributivist for the reason Moore spells out in his seminal work on retributivism. Indeed, even though I don't agree at all with Moore about the truth of retributivism, I think his account of retributivism really is the gold standard.

As for your suggestion that "desert-sensitive consequentialism" is a contradiction, I see no reason to think it is. Indeed, I think there are probably lots of compatibilists who believe in moral desert and in giving people what they deserve, but who are nevertheless consequentialist at the end of the day when it comes to punishment. My terminology was really just meant to highlight that one need not be a desert skeptic (nor a skeptic about the importance of giving people what they deserve) to be a consequentialist about punishment.

That being said, it's ironic that you raise the contradiction objection since on the watered down versions of so-called retributivism that you and Justin are pushing in this thread, you are committed to giving people what they deserve unless and until not doing so produces more than moderate--whatever that it--positive consequences. In short, you are going to need a consequentialist calculus to figure out when your retribution stops and your consequentialism begins--i.e., you are a forward-looking retributivist. That's just the converse way of describing my purportedly contradictory desert-sensitive consequentialism, no?

Dan,

I am glad you are enjoying the discussion. Hopefully, I will post something similar soon about MAOA, compatibilism, and retributivism. So, stay tuned. Perhaps you'll enjoy that discussion as well.

[Eddy Nahmias]

Maybe this will help (or maybe it will just muddle things up!):

One can be a Monist or a Pluralist about the criteria of punishment--that is, about what justifies punishment (and degrees of punishment)--just as one might be about morality more generally--that is, one might think there is just one criterion to determine when something is right or wrong, or one might think there are more than one criteria, and they sometimes conflict.

Thomas seems to be saying that if one is a Retributivist about punishment, one must be a Monist Retributivist. Others seem to be saying that one could be a Pluralist Retributivist, who says that the criteria for punishment involves (1) (and perhaps first and primarily?) looking at what the criminal deserves in a backwards-looking manner, but also (2) looking at what forward-looking outcomes will likely result from treating the criminal in certain ways (e.g., incapacitating her, deterring her and others, rehabilitating her). The Pluralist then puts 1 and 2 together to come up with the best sentence, all things considered. The problems arise when 1 and 2 conflict (though they do not always conflict). Perhaps Ross counts as a Pluralist of this sort. I would argue that our legal system is Pluralistic about punishment in just this way (even if it may overemphasize 1 at the expense of doing 2 better).

Another possibility is that one is a Monist about the *word* "punishment", arguing that it should only apply to, say, retributive responses to a criminal (e.g., causing them pain, executing them, forcing them to pay reparations to victims--though this sort of "restorative" punishment gets tricky to interpret as only retributive). But even so, one might be a Pluralist about how to treat criminals, arguing that in addition to considering how to "punish" them (retributively), we need to consider how to treat them in other ways in order to account for other moral, social, and political demands (where these other treatments are not called "punishment"). So, we might need to "punish" the psychopath less than a normal criminal because we determine that the psychopath lacked certain capacities, but then we might, as Pluralists, decide that we need to quarantine (and try, perhaps in vain, to rehabilitate) the psychopath more than other criminals. As in most tough moral decisions, we may have to sort out conflicts between "punishment" so defined and other treatments, but I don't see that as a reason to abandon Pluralism.

Personally, I favor Pluralism, and if there are any Monists (Kant and Michael Moore) who would reject the very possibility of these other treatments of criminals, then I find their view, well, daffy.

[Thomas Nadelhoffer]

First, it is clear that I can't keep away from this discussion despite the fact that I should, all things considered, do just that. As such, my continued participation in this thread disproves skepticism about weakness of will! :) But I digress...

Second, as usual, Eddy has shed some clarificatory light on the discussion, but I nevertheless think that retributivists cannot be pluralists without giving up on being retributivists for the reasons I have already spelled out ad nauseum. That being said, one obviously could (and I believe should) be a Pluralist about punishment.

However, unless and until someone spells out how they want to draw the distinction between "minimum to moderate non-retributivist-trumping-consequences" and the "more than moderate retributivist-trumping consequences" it will continue to seem to me that the related views being put forward by Tamler, Justin, Eddy, and others, are simply forms of consequentialism whereby desert nevertheless has to play some potentially limiting role--however small--in the implementation of punishment. Given that figuring out when this limiting role should be trumped seems like a cost-benefit analysis to me, I am sticking to my guns when it comes to the status of these views as non-retributivist.

That being said, rather than further beating this dying horse in the thread, I suggest everyone read Moore's Placing Blame (especially Chapters 3 & 4). By my lights, he makes it clear--in a way I have not in this post--that the sorts of Pluralist/Prima Facie/Sometimes-Forward-Looking views you guys insist on calling retributivism aren't properly retributivistic.

In the meantime, I also invite Eddy and anyone else who's playing along to respond to my earlier claim that one can both believe in desert (and giving the deserving their due) while being a consequentialist through and through. Of course, I don't believe in desert, so this isn't *my* view. But it seems that one potential misunderstanding that may be driving this entire discussion is the following: P believes in desert and giving people what they deserve when it comes to punishment--at least to some minimal extent. Therefore, P *must* be a retributivist of some stripe. I think this is simply incorrect. While desert is a necessary element of any retributivist theory, P's merely believing in desert is not sufficient for establishing that P's a retributivist. Or so me thinks...

p.s. We could also get back to talking about psychopathy as well, no? :) Plus, I still owe Eddy a response to his claims about the Sentencing Reform Act of 1984!

[Tamler Sommers]

Thomas,

But the pluralist retributivist (to use Eddy's term) I'm describing rejects the view that consequentialism justifies the institution of punishment. They believe that both consequentialism and retributivism justify the institution.

And when you write:

"Indeed, I think there are probably lots of compatibilists who believe in moral desert and in giving people what they deserve, but who are nevertheless consequentialist at the end of the day when it comes to punishment"...

again, this doesn't apply to the position I'm describing. Pluralist retributivists are not consequentialists at the end of the day because they don't punish exclusively to bring about better consequences. In fact, in many cases, they'll sacrifice good consequences in order to give the criminal what he or she deserves. I always thought that to establish your consequentialist bonafides, you weren't supposed to find this acceptable.

In other words, the irony you point out cuts both ways. The consequentialists you imagine have to take into account something that by its very nature is not tied to future consequences (desert) in order to determine the proper punishment.

[Tamler Sommers]

Just to finish up in light of the last post, Thomas writes:

"However, unless and until someone spells out how they want to draw the distinction between "minimum to moderate non-retributivist-trumping-consequences" and the "more than moderate retributivist-trumping consequences" it will continue to seem to me that the related views being put forward by Tamler, Justin, Eddy, and others, are simply forms of consequentialism whereby desert nevertheless has to play some potentially limiting role--however small--in the implementation of punishment."

Why doesn't the "desert-sensitive consequentialist" face the spelling out requirement? Why can't I say:

Unless and until someone spells out how they want to draw the distinction between "minimum to moderate retributivist considerations consequences" that do not constrain consequentialist implementation and the "more than moderate consequentialist-trumping considerations" it will continue to seem to me that the views being put forward by Thomas (and Moore?) are simply forms of retributivism whereby future consequences nevertheless has to play some role--however small--in the implementation of punishment."

[Thomas Nadelhoffer]

Tamler,

It's going to depend on how one defines desert and what one counts as backward looking. That someone committed a similar offense in the past and that someone acted in an especially malicious way in this particular case are both (a) entirely backward looking, and (b) entirely relevant to any forward-looking theory of punishment (especially rehabilitative and preventive models). Notice, these are the very same backward looking features that retributivists focus on when applying their moral theories and their beliefs about free will to the problem of punishment.

Any consequentialist worth her salt will allow these kinds of backward looking considerations to play an important role in figuring out how much to punish. The difference is that the consequentialist isn't interested in using these features to ensure offenders suffer for the sake of suffering--which is the hallmark of retributivism--but rather to use this information as part of the cost-benefit analysis one has to undertake to figure out how much punishment will maximize the good, happiness, welfare, safety, or whatever else they take to be the consequentialist goal. I take that to be one of the key differences between consequentialist theories of punishment and retributivist theories. The consequentialist has to pay attention to both the relevant backward looking and forward looking considerations. The retributivist, on the other hand, is limited to the backward looking desert-entailing features of the case. Or, here again, so me thinks.

p.s. I wrote this before Tamler's most recent follow up comment. So, I will have to say something in response to his most recent comment later as I am off to physical therapy--where they will tell me I should be typing less. Surprise, surprise! Who woulda thunk I'm as stubborn as a mule?! :) That being said, I just want to point out that what I say in this comment speaks to Tamler's attempt to flip the script! I can tell a coherent story about how and why "desert-relevant" features matter on a desert-sensitive consequentialist view. Check (or checkmate)?

[Tamler Sommers]

This reminds me of the great Woody Allen piece about the two guys playing chess by mail, both thinking they mated each other:

(http://maxxwolf.tripod.com/woody.html).

[Justin Capes]

I agree with much of what Eddy and Tamler have said in their last few posts. So I won't say anything more on whether the pluralist view I've been pushing really counts as retributivism.

About spelling out where backwards looking considerations end and forward considerations begin, I suspect the best we can do (or perhaps I should say, the best I can do, at least at this point) is to say that there are clear cases on either end and the rest are hard. If Jones deserves life in prison (or some equivalent sentence) but executing him would perhaps deter one or two crimes similar to the one Jones commited, then we shouldn't execute him. Similarly, if executing him would save a million lives, then given that he is responsible for his crime, perhaps we should execute him, even though doing so is strictly speaking more than he deserves. Most cases will not be as obvious as these, of course. Most cases will be quite difficult and will require a degree of practical wisdom. This is frustrating for practical purposes, but theoretically I don't find it all that disturbing, given my commitment to (a moderate form of) moral particularism.

[Thomas Nadelhoffer]

Tamler,

That's funny as hell (although you have to remove the last parenthesis in order to get the link to work). Of course, there should have been an all-important emoticon of some sort after my check/mate comment but I had just used one a few lines earlier and I didn't want to drive Eddy crazy (since, if I remember correctly, he is no fan of emoticons--or was that you?!)

Justin,

Like I said earlier, I am well aware that this comment thread isn't the place for you to think all the way through and flesh out the issue. It's just a place to birth future discussions. As such, I look forward to seeing the future fruits of your (non) retributivist labors downstream! :) The good news about our line of work is that we are priviledged to be able to carry out these type of discussions over the course of our careers. The cool things about blogs--at least on my view--is that they help us figure out where we need and want to plant and then nurture our own theoretical and practical seeds.

That being said, thanks again to everyone for playing along in this thread and patiently responding to my attempts to paint you all consequentialist! In the next few weeks I will upload a follow-up post about Kant's island--which this thread has led me to revisit. But more on that later!

[Philoponus]

Thomas,

Thank you so much for your kind response. I’m sorry that you found my comments on psychopaths and diminished responsibility so odd and implausible, but may I share something I have from prosecutors who actually face diminished responsibility pleas from serial rapists and murderers. A key issue is always whether the perpetrator ever sought help for his pathology, or instead, devoted his time and resources to cleverly concealing his growing pathology. It is important trial fact that people like Ross and Gacy and Bundy never sought any psychiatric help (though they had access to and could have afforded it).

You asked me skeptically, “what should he (Ross) have done to stave off the disorder?” Forget about his access to and the likely effectiveness of professional psychiatric care. There is one simple, no-cost remedy by which he could have stopped his criminal career at the start. Suppose Ross had presented himself at the local Connecticut DA and said “I have knowledge of serious crime that is about to occur. I find myself in the grip of homicidal sexual fantasies that I believe I can no longer control and will shortly be played out for real. In other words, I’m about to start raping and killing women.” No DA is ever going to send that man home and say “call us when you do your first one”.

Depending on the state’s mental health statutes, someone making sort of declaration can be ordered confined for psychiatric evaluation and police investigation. He may end up being released in 30 days, but ordered to report to a psychiatric case worker, who can thereafter petition for his arrest if he misbehaves in certain ways or misses appointments. More importantly, he has become a permanent person of special interest to the police in the event of any rape/murder. The police will routinely keep track of him. His opportunity to commit even one such crime without being immediately arrested is now very limited. A string of such crimes is a virtual impossibility.

That is what Ross should done to prevent or abort his heinous criminal career. If you cannot control yourself, the state is ready and able to control you.

Please feel free to leave this late comment unpublished—I would have preferred to sent it to you privately—but I wanted to say something on the factual issue of whether adult psychopaths, or adolescent soon-to-be psychopaths, cannot get treatment that will protect themselves and especially their would-be victims, despite the fact they cannot afford expensive psychiatric care. The same statement as above, made by an adolescent to a school counselor or in a walk-in mental health clinic, will draw attention from the authorities. Suicidal or homicidal declarations are taken very serious, so help is a few words away if someone wants it.

[Tom Clark]

Philoponus, you say

"A key issue is always whether the perpetrator ever sought help for his pathology, or instead, devoted his time and resources to cleverly concealing his growing pathology. It is [an] important trial fact that people like Ross and Gacy and Bundy never sought any psychiatric help (though they had access to and could have afforded it)...That is what Ross should [have] done to prevent or abort his heinous criminal career. If you cannot control yourself, the state is ready and able to control you."

Indeed, Ross/Gacy/Bundy likely lacked certain control capacities, so could not control themselves in normally available ways, which is why they didn't seek help. Are they therefore blameworthy for not having sought help, for lacking those capacities? We can only answer yes if we suppose they deliberately engineered their lack of control, which seems far-fetched given what we know about the genesis of psychopathy.

The "clever concealment of a growing pathology" is the pathology itself, not something that the psychopath could have chosen not to undertake given his physical and psychological circumstances. So there was no "simple, no cost remedy" available to Ross, nor is there to anyone in the grip of such psychology, unless they're lucky enough to reveal their impulses to someone in a position to take appropriate action.

[Philoponus]

Tom,

Suppose (God forbid) you or I woke up tomorrow and started having very powerful fantasies about committing sexual assaults. There are a few cases like this in the literature. Normal people suddenly and inexplicably turn into sexual predators, sometimes pedophiles. A forty year married dentist suddenly finds himself arouses by the sight of ten year olds. Within a few weeks the dentist is hanging around schoolyards, beginning to plan his first attack. After an attack or two, he is caught and incarcerated and medically examined, and sure enough, we find a big prefrontal tumor impinging on various control areas. The tumor is removed and the pedophilia disappears. Familiar story, right?

OK, so imagine this same thing starts to happen to me or you. Suddenly we find we are sexually attracted to children and every day the feelings are becoming stronger. You and I, I predict, would very soon be camped out at our psychiatrist’s office demanding tests, scans, medications, etc. We know how fragile self-control is when the brain goes awry, even for people with Ph.D’s in philosophy. We would feel, wouldn't we, a very strong responsibility to do whatever is necessary to stop ourselves from turning into the sexual predators that we could unwillingly become. The main point is that we can take effective action now. Our ability to seek help is not compromised by our growing sexual dyscontrol, at least if we act now before we let things get much worse.

Go back to the case of Michael Ross, who had no tumor but a pathology that developed over many years. Do I understand you to be saying that the fantasies that pushed him increasingly out of control with women also always prevented him from seeking psychiatric help or just turning himself in? But people with very serious psychiatric problems do seek help and turn themselves in, fearing for their own safety and the safety of others. Ross never did, and it was never shown that he had some special irresistible inhibition from seeking help. This is the gap that sank his defense. The jury did not believe that Ross could not have at least tried to stop himself, especially early on, or that he ever wanted to stop himself. If Ross had sought help for his growing sexual dyscontrol even once, the capital case against him would have been much harder, because there would have been the evidence he was trying but failing to control himself.

Ross never claimed that he didn’t know what he doing or didn’t know it was wrong. He claimed he could not stop himself, but that defense only gets traction with a jury when you can also show that the perpetrator actually wanted to stop and ever tried to stop himself. I don't see anything odd about holding someone responsible for doing whatever it takes to try to stay in control of himself when the consequences of dyscontrol are so terrible.

[John Fischer]

Well, this doesn't exactly have to do specifically with the relationship between retributivism and psychopathy, but people might be interested in the work of UC Irvine neuroscientist James Fallon: http://www.npr.org/templates/story/story.php?storyId=127888976

It turns out he has genetic signature of the psychopath; it raises some interesting issues. (Perhaps the signature predicts success in academics, as well as psychopathy??? Or perhaps just sucess at Irvine :))

[Tom Clark]

Philoponus,

Thanks for the additional remarks. The issue boils down to whether Ross could have at least tried to get help at some point in his career - you think he could have and I don't. I think we agree that *something* (maybe a combination of things) prevented him from trying to get help along the way. What was it? Perhaps it was lack of a normal capacity to fear for himself and others, perhaps the lack of a normal sense of responsibility and/or empathy, perhaps the sheer strength of his impulses, but it was certainly *something*, not the willful refusal to exert a normal, sufficient control capacity that was ready and able to be engaged to steer him towards help. For how would we plausibly explain such a failure of control? Only in terms of there being a sufficiently strong impulse that *overrode* the actual capacity, that incapacitated it - that's the real story behind a seeming refusal to engage the capacity. We shouldn't buy into the myth of the unresisted but resistible impulse. So the way I see it, it isn't the case that given the conditions in play he could have tried to seek help. In which case, did he deserve to die?