[To Hell With] the TNR Principle!

Here’s the latest in my series of increasingly dangerous concessions to Strawson-style compatibilism.

Assume (plausibly) that theories of moral responsibility hinge on our all-things-considered intuitions or judgments about the proper conditions for deserving blame, praise, punishment and reward.Now imagine a person, call him Jack, who finds hard incompatibilist arguments intuitively compelling all things considered, and therefore accepts that no one can be morally responsible for their behavior. (He even publishes a couple of articles defending that view.) But then Jack thinks of a scenario in which someone deliberately, willfully harms his daughter.The offender in the imagined scenario meets all normal compatibilist conditions, but not the incompatibilist conditions that Jack believes are necessary for desert.

Were this scenario to occur, Jack would feel that the offender deserved to suffer for the act— he would feel this more strongly than just about anything else. In fact, he thinks it would be obscene of him to apply incompatibilist principles to this case, to worry about whether the guy was causa sui, or whether the factors that trace back beyond his control led him to harm his daughter. If someone said "but what about the TNR principle?" he would say “F*ck the TNR Principle!” More importantly, even now, just imagining the scenario, Jack believes that this is the right or appropriate response—that it would be wrong not to feel this way. Lastly, Jack realizes that he wouldn’t feel that response to be appropriate if the offender didn't meet certain basic compatibilist conditions.If the man was completely crazy, or manipulated into performing the act, he doesn’t think that it's wrong not to feel retributive towards that person. (He might feel retributive anyway, but now, upon reflection, he sees that as more irrational.)

Now Jack is fully aware of his incompatibilist commitments.He finds the four-case argument and the basic argument no less compelling.He is also aware that his retributive attitude in this case has deep psychological roots that trace back to his evolutionary history, kin selection etc. But he doesn’t care. He still thinks it would be deeply inappropriate to favor his incompatibilist intuitions over the intuition that the guy who harmed his daughter deserves to suffer for that crime. And he cannot see why the same reasoning shouldn’t apply to other people’s daughters (and relatives, friends etc.) as well.Since intuitions are the ultimate arbiters for theories of responsibility, and the force of this latter intuition is greater that the intuitions that favors incompatibilist principles, does Jack has reason to rethink his all-things-considered rejection of compatibilism?

Comments

[Justin Capes]

One also wonder's about Jack's anti-libertarian commitments. If he were to reject those commitments, then he could retain both his incompatibilist views and his belief in moral responsibility. Of course Jack is no doubt aware of this, but you've said nothing about what would prompt him to reject incompatibilism in favor of responsibility as opposed to rejecting both compatibilism and determinism (or responsibility-undermining indeterminism). Many of Jack's colleauges, I expect, find both of the latter theses objectionable. Why shouldn't he be tempted to become a libertarian? Is compatibilism more plausible to him, than responsibility-allowing indeterminism?

[Kip]

Et tu, Tamler?

Didn't Nichols and Knobe suggest that affect biases us (irrationally) towards compatibilism?

Kip

[Tamler Sommers]

Justin, right, good point. I'm assuming Jack finds libertarian accounts to be just as lacking when it comes to satisfying the conditions of desert. (As Galen Strawson does, for example.) So there's no reason I can see for Jack to opt for libertarianism since it has the additional disadvantage of requiring indeterminism to be true. And when it comes to the particular case with his daughter, Jack is satisfied with the compatibilist conditions.

[Tamler Sommers]

I'm not proud of it, Kip!

Yes, that paper does suggest an affective bias for compatibilism. But in this case, Jack is aware of that bias, aware even of the origins of the bias, and still, upon reflection, maintain that it's the appropriate attitude to hold.

[Michael McKenna]

Tamler,

You are beginning to see the truth, though perhaps still only dimly. Walk towards the light Tamler, the light!

[Eddy Nahmias]

Let's set up your case as a Moorean dilemma.

Suppose Jack now, after deep philosophical reflection, accepts this proposition:
1. Either (a) TNR is false or (b) I cannot justifiably hold my daughter's murderer morally responsible.

Without a strong, independent argument for TNR, it's hard to see why Jack should (much less would) accept b over a.

It seems to me that every argument for incompatibilism or free will skepticism relies on some TNR principle rather than offering a strong, independent argument for TNR. What would such an argument look like?

[Tamler Sommers]

Eddy,

Isn't the four case argument an independent argument for TNR?

[Eddy Nahmias]

How would it go? Something like this?
1. The best explanation for why Plum in case 1 (or case 2 or 3? and hence 4?) is not responsible for doing A is that there are sufficient conditions for his doing A for which Plum is not responsible in any way (i.e., TNR).
2. So, we have good reason to accept TNR.

I don't accept premise 1, but rather than digressing here, is this the sort of argument you have in mind for TNR?

[Tamler Sommers]

That's it, right. Of course, I find TNR to be intuitive on its own. But as Michael McKenna reminded me a while back, Derk's argument doesn't ride on that.

[Justin Capes]

Tamler,

Does Jack think responsibility would require determinism, or would it be compatible with indeterminism? And don't listen to Michael. Isn't it said somewhere that devils can manifest as angels of light!?

[Cihan]

Tamler,

What do you think of Shaun Nichols' argument about being a non-realist and keeping your reactive attitudes too? See "After incompatibilism: A naturalistic defense of the reactive attitudes"

[Eddy Nahmias]

Well, if the argument is supposed to look like the one I reconstructed, then Jack is going to face a Moorean dilemma that looks like this:
1. Either (a) premise 1 (above) is false or (b) I cannot justifiably hold my daughter's murderer morally responsible.

And premise 1 could be false because the best explanation for Plum's not being responsible in case 1 is something other than TNR (e.g., the neuroscientists control Plum in a way that determinism does not). Or Jack could respond that Plum is responsible after all.

What if Plum in case 1 is the murderer of Jack's daughter? Presumably Jack will blame the neuroscientists who control the outcome (even if the neuroscientists are deterministically caused to bring about that outcome).

[Alan White]

Tamler, I for one think you've hit on an axiological theme that I have always thought was underplayed in FW debate. At least one associated reason that incompatibilism has had on its side is its presumed fundamental value in making us moral beings (positively so if we have IFW; by hard incompatibilism a supposed irreplaceable loss of value). Kane speaks of this (e.g.) in terms of self-forming actions and ultimacy of responsibility--both irreducibly valuable things if the underlying metaphysical underpinnings exist. But the compatibilist viewpoint, besides being motivated by metaphysical enchantment with determinism and semantic problems with freedom, also has been accompanied by an emphasis on the pragmatic values of being able to distinguish some kind of useful repsonsibility that retains a core of the basic moral precepts of justice and punishment. It could be that Jack has developed enough doubt about the usefulness of the metaphysical roots of incompatibilism to cast further doubt about the wisdom of basing our most basic moral and justice-oriented judgments on them.

[Tamler Sommers]

Justin,

No, Jack certainly doesn't think responsibility requires determinism (in fact, part of thinks they're incompatible). But he doesn't think indeterminism is of any help when it comes to responsibility either. He would feel just as retributive, but no more, to the criminal if persuaded that indeterminism was true. And thanks for the warning about Michael. It's a constant worry for me that I'm succumbing to his wiles.

Eddy, I'm sympathetic to what you're saying. That's the problem!

Cihan,

Yeah, I see this as completely in line with (and inspired in part by) Shaun's argument in that paper. But I take his argument to be one for (post-incompatibilism) compatibilism. The view sketched here also comes more from the inside, my (I mean Jack's) own convictions, rather than from an analysis of the empirical implications of embracing hard incompatibilism.

Alan, thanks, I'm inclined to agree--to an extent anyway, but to an alarming extent. And all of that seems very much in line with Strawson's original aim in Freedom and Resentment, although he didn't frame the argument there in terms of a battle of competing intuitions and values.

[tnadelhoffer]

Tamler,

Why not just say that under certain tragic circumstances, even you could be induced to have retributivist intuitions of such force that they would effectively undermine your ability to think clearly about the issues at hand? I, for one, would certainly feel furious anger if someone hurt my loved ones. I would even be strangely comfortable with making the perpetrator suffer myself if the chance arose. But what I would NOT conclude is that all of this knuckle-dragging eye for an eye business was justified--just deeply ingrained. Since when did having a strong feeling or inclination to do x = a justification to x anyway? For instance, you say "more importantly, even now, just imagining the scenario, Jack believes that this is the right or appropriate response—that it would be wrong not to feel this way." But what's the argument here for the appropriateness of the intuitive/affective response? It can't simply be the fact that your emotional response is strong, no?

[Tamler Sommers]

Thomas, thanks, finally someone coming to the defense of hard incompatibilism! Here's the problem I see with your reply though: In what sense am I not thinking clearly about this issue? It hasn't happened (thank goodness), and upon reflection, I endorse it, I still think it's the right response.

Not only that, I wouldn't make the same claim if compatibilist conditions weren't met. (This, to me, is one of the most damaging features of this line of argument.) In that case, I'd agree with your interpretation. I'd likely feel retributive anyway, maybe even wanting to inflict the suffering myself as you say, but now, in a calmer moment, I don't endorse that response.

The contrast between these two cases, it seems to me, speaks in favor of compatibilism somehow. Why should an reflective judgment about a general principle (TNR) override a reflective judgment of equal or greater strength about the right way to view a particular case? As Eddy says, it seems like one of those Moorean dilemmas...

Of course, I'm more than open to being persuaded otherwise.

[Manuel Vargas]

Jack, I have a proposal for you . . .

[Cihan]

Tamler,

I don't see Nichols' position as compatibilist. I see it more as a (practical) justification for why one might want to hold onto reactive attitudes even though one thinks there is no such thing as moral responsibility.

For me, compatibilism consists in thinking that certain theses are true. What sorts of attitudes one should have is a completely different matter.

[Tamler Sommers]

Manuel, even worse, I was deciding between two posts today. The other one--"Is moral responsibility a whale or a witch"--is a dangerous concession to the revisionist.

Cihan, I guess that's where we disagree, I think the two are related...

[Thomas Nadelhoffer]

Tamler,

Here are three considered judgments you could have:

(a) If someone hurt my loved one and satisfied the basic compatibilist conditions, I would intuitively feel that it would be instrumentally valuable for that person to suffer.

(b) If someone hurt my loved one and satisfied the basic compatibilist conditions, I would intuitively feel that it would be intrinsically valuable for that person to suffer.

(c) If someone hurt my loved one but did not satisfy the basic compatibilist conditions, I would intuitively feel that it would be neither instrumentally nor intrinsically valuable for that person to suffer.

You seem to assume that because you believe (c), this somehow provides you with evidence that (b) is no less appropriate than (a). But what is the argument for this move? Of course, you would understandably be filled with outrage if any of these situations came to fruition. Moreover, this outrage would dissipate much more readily in the event that (c) came to pass. But why think this fact alone somehow makes (b) acceptable? Once you strip away all of the benefits that both you and society would derive from punishing the person who harmed your loved one—e.g., catharsis, restorative justice, prevention, and rehabilitation—what value is left for the suffering? The issue is not whether you would intuitively judge that the person who hurt your loved one should suffer—for you could think this even if you believed (a) but not (b). Rather, the issue is what justifies your intuition that the person who hurt your loved one ought to suffer purely for suffering’s sake? In answering this question, I think you have to make sure you keep a bright line between (a) and (b)—a difference that I believe too many compatibilists blur when doing so helps their case (unwittingly or not).

Cihan,

In personal correspondence, Shaun has suggested that he does not mean for the view he developed in “After Incompatibilism” to be taken as form of fictionalism about free will and moral responsibility. In suggesting that we “relinquish” the incompatibilist intuition, I think he really is suggesting that we give up incompatibilism. As such, I think his view is closer to Tamler’s than you suggest—even if the kinds of reasons they offer are different.

[Tom Clark]

Tamler,

That Jack's retributivist response depends on the offender meeting compatibilist conditions makes perfect sense from an evolutionary standpoint: only offenders who act intentionally, sanely, without coercion, etc. are proper targets of punitive responses since otherwise those responses are wasted on a being for whom they have no (future) deterrent effect (the insane, the very young) or who doesn't need such deterrence (the coerced).

But of course this points up the original, natural *consequentialist* rationale of the retributivist response: it worked to deter intentional aggressors. Does this mean you are justified in inflicting suffering on the offender even if no good consequence ensues (the definition of retribution), just because you reactively *want* to punish him, whatever the outcome? Of course not. Or if it does, I want to hear your justification.

Perhaps Jack feels that it's "obscene" and "deeply inappropriate" to give any weight to incompatibilist intuitions because he thinks that being swayed by them would show (to himself or others) that he doesn't *really* love his daughter. But this isn't the case. He could love his daughter madly, really *want* to kill/dismember/torture the culprit, but reflectively decide that acting on this desire isn't justifiable, all things considered (such as the kind of culture we want to live in). And remembering that the offender doesn't bear ultimate responsibility for his character and actions helps to diffuse retributive rage by distributing responsibility to factors outside him - factors that, had they been different, might have prevented the offense in the first place.

During his presidential campaign, Mike Dukakis was pilloried for not responding in the right (retributive) way about how he'd feel if his wife were raped (or something like that, can't remember). He was way too dispassionate. So yes, it's naturally *right*, that is, understandable and expectable, for Jack to have retributive desires. But whether it's right for him or the state to *act* on them is an entirely different matter.

[Tamler Sommers]

Thomas, that's really interesting but I'm not sure why you're bringing instrumental value into the picture. I have no intuition one way or the other about whether the retributive attitude is instrumentally valuable in any of the cases. Do you? The crucial issue, as far as I can see, is that Jack thinks it's RIGHT (intrisically I guess) for the person to suffer in the compatibilist condition case but not in the non-compatibilist condition case (although he might still want the person to suffer in the latter case if it happened). What does instrumental value have to do with any of this?

You ask: "Once you strip away all of the benefits that both you and society would derive from punishing the person who harmed your loved one—e.g., catharsis, restorative justice, prevention, and rehabilitation—what value is left for the suffering?"

And Jack's answer is: "The motherfucker deliberately, intentionally, cold-bloodedly harmed my daugther!" If there's no value in him suffering, then there's no value in anything. (I'm not endorsing that view necessarily, but I have hard time figuring out what's wrong with it.)

Tom,

I think that might be a really promising reply. Although I don't think you even need to bring in the consequentialist value of the retributive attitude which, as I note above, I'm not sure is relevant. (Again, I'm not sure if there is a consequentialist value to retributive attitudes in these case, I have no intuition about that one way or another.)

I would put the reply like this: the reason Jack feels it would be deeply inappropriate to bring in causa sui etc considerations into this case is because, given our psychology, anyone who did that would be, perhaps unknowingly, showing that he doesn't love his daughter enough. And however strongly Jack feels about the basic argument, he feels more strongly that he loves his daughter. So if he thinks that accepting the basic argument under those circumstances means implicitly that he doesn't sufficiently love his daughter, he'll say farewell to the basic argument. In a weird sense, this would explain away the intuition that it's right to feel that the offender in that case deserves to suffer.

The Dukakis analogy occured to me, but I've resisted it until now. Thinking it over though, I'm inclined to say that to the extent that this argument works for compatibilism, it might work as a defense for capital punishment as well.

[Mark Smeltzer]

Tamler,

Interesting post!

You said in response to Eddy that Jack finds TNR plausible because of some sort of abduction from the 4 cases argument.

I'm curious though... if the crime in the 4 cases argument were about his daughter getting harmed, would he still agree that the person in the final, compatiblist condition satisying case does not deserve punishment?

If he does believe that the perpetrator deserves punishment in the final case, would it be fair to say he wouldn't find the 4 cases argument to be a compelling reason to warrant his belief in TNR?

[Michael Drake]

Suppose Jack's affective response to his daughter's vividly imagined murder morally, situationally justifies his retributive thirst (and perhaps would morally, situationally justify his actually taking revenge), all things considered; why should this fact foreclose Jack's desire to reform himself so that those same conditions elicit instead a pattern of affectivity that conforms to the correct (which is to say, incompatibilist ;-) picture of action, all things considered?

[Kip]

Like Cihan, I read Nichols' paper as standing for the following:

1. Incompatibilism, and no-free-willism, are technically right.
2. But we should (irrationally; instrumentally?) hold onto reactive attitudes to avoid the danger of learned helplessness.

In view of Thomas N's post above, I'm calling out Shaun to find out where he really stands.

[Mike Patterson]

Tamler,
Thanks for introducing this blog to us this semester; I hope I can add a little to this great discussion.

I think Tom was onto something with this debate. I understand when you say you would not want your reactions to be mistaken as cold or unloving. However, in your paragraph "given our psychology..." are you endorsing the idea that the amount of retribution demanded or obtained is directly proportional to the amount of love felt? It seems to me that you are trying to use your moral anger as proof about your relationship with your daughter-both to yourself and to others. I hear Jack saying, “Unless I cause the unmitigated suffering of this person, I do not really love my daughter.” The stance that one should, or even could, use a person’s anger (or how much suffering he extracts) as an even halfway effective measurement for love does not sit well with me. There is a serious tension in the idea that to properly show our love of someone, we must be driven to anger and can't act according to our informed beliefs.

I’m curious, without this fear of being labeled as unloving, would this dilemma of yours go away?

As an aside, if someone was able to control their emotions even in a horrific situation such as this (with serious effort over time), wouldn’t people find them praiseworthy for still being able to place the blame correctly(i.e. using TNR)?

[tnadelhoffer]

Kip,

Since I don't suspect Shaun will know that he has been "called out"--i.e., since I am unsure he will know about this post--I will say a little more in defense of my interpretation of his view. Here are the background assumptions he makes at the beginning of the paper:

1) He has always found incompatibilism to be intuitive.
2) The folk find incompatibilism to be intuitive.
3) Determinism is true.*

He then goes on to try to show that the incompatibilist intuition--especially when conjoined with the assumption of the truth of determinism--would lead to socially maladaptive behavior. On his view, moral anger--which purportedly does not make sense if both the incompatibilist intuition and determinism are true--leads to very important pro-social behavior.

In light of the gathering data on moral anger and cooperation, Shaun concludes that "we are right to ignore or relinquish the commitment to incompatibilism"--since doing so will make our interpersonal lives better. You appear to be focusing on the "ignore" option--which makes it seem like his view amounts to some form of fictionalism--whereas I really think Shaun wants us to tentatively relinquish it, at least until a more powerful argument has been made for free will revolutionism.

p.s. I am working on this kind of argument now--which is apparently even more pressing now given that Tamler appears to be falling off the wagon of free will nihilism!

*What he actually says in a footnote is "The broader assumption is really that we lack libertarian free will, and various indeterminist theses are also incompatible with free will. I focus on determinism merely to ease exposition."

[Tamler Sommers]

Mark,

Interesting question--I wonder how the four case argument would be affected if instead of Clue characters, the victims were portrayed as people we knew.

Michael D, that might be a way to go. The thing that might be hard to justify, then, is why incompatibilism is the correct notion all things considered. Why isn't the situational justification of this retributive feeling something important to be considered.

Mike P, welcome to the garden! (Mike is a Master's student here at Houston, carrying the the MR skepticism torch--he wrote a really good paper critiquing Shaun's argument in that article.) You're right. I didn't mean to suggest that our psychology really is such that if you don't feel retributive, you don't love your daughter. The important thing is that Jack believes this to be the case, maybe unconsciously, which is why, upon reflection, he endorses his retributivism. That's the sense in which the endorsement could be explained away, as it doesn't really pertain to his belief about the correctness of the retributive attitude.

Thomas, I haven't fallen off the wagon yet! It's just a bumpier road than I once thought.

[tnadelhoffer]

Tamler,

I brought the instrumental value of retributive emotions into the picture because I think the benefits of these emotions that I mentioned are the actual reasons why our deep seated retributivistic intuitions were adaptive in the first place. The claim that making people suffer for their "internal wickedness" is intrinsically valuable is something we cooked up after the fact as a post hoc rationalization for our retributivistic inclinations. But given that our suite of retributive emotions could be viewed as either instrumentally or intrinsically valuable (or both), when you say "Jack thinks it's RIGHT…for the person to suffer in the compatibilist condition case but not in the non-compatibilist condition case," the "right" could (and I would argue should) simply mean that you think it is appropriate for the perpetrator to suffer given the intrapersonal and interpersonal benefits that would result. These same forward-looking justifications explain why you wouldn't think it was right to make the perpetrator suffer in the non-compatibilist case. But notice that by framing it this way, desert simply drops out—especially if desert entails the intrinsic value of making people suffer for their wrong-doing. This doesn't mean that there's no value in making the "cold-blooded mother fucker" suffer. It simply means that the suffering is not valuable in and of itself as the retributivist would suggest. You seem to be (I believe mistakenly) assuming that one could not be a free will skeptic and think that there is value in making wrong-doers suffer for the harms they create. I am simply trying to point out that this need not be the case. To reject retributivism is simply to reject the idea that making people suffer for the harms they cause is intrinsically valuable. This need not lead you to conclude that making wrong-doers suffer couldn't be valuable in other ways. But so long as you can keep a hold on both your desire to see the perpetrator suffer and your belief in free will skepticism, I see no reason to give an inch to the free willers!

[Rob]

Tamler,

There seem to be two distinguishable components to Jack's stance, as you've characterized it: his view of the offender as deserving to suffer, and his attitude toward this view as as appropriate, all things considered. The second-order character of the second component of his stance, it seems to me, leaves open the possibility that Jack could nevertheless still hold that the actual punitive institutions charged with handling the perpetrator should reflect his (Jack's) more impersonal incompatibilist commitments, without shaking his conviction in the appropriateness of his view toward the offender. It might be an uncommon psychological feat, registering a rather complex value set, but not, I think, impossible.

[Michael Drake]

Tamler, on the view I'm sketching the retributive feeling remains something important to be considered. But now it stands as just one amid other important, competing considerations. Tom Clark's invitation to reflect on the contingent evolutionary pathways that led to our standard affective responses opens up one avenue for second order reflection. Our admiration for those who have apparently mastered, at least to a very high degree, their retributive impulses (and, one suspects to some extent, their retributive instincts) opens another.

One possibility is to think of "morality" as operating in two registers: The first is that implicated by actions in response to immediate, situational pressures, where such actions are understood to flow from (and whose justification at least partially* flows from) the kind of character we happen to have at a given time; the second is that implicated by our higher order reflections about what kind of character we ought to work toward developing.

*I'm hedging a bit against an objection from the following kind of case: on the two-register view sketched, a blood-thirsty murderer, having the *bad* character commensurate with that occupation, would arguably be "morally justified" in murdering (namely, by virtue of his murderous affectivity). I wouldn't necessarily resist this conclusion, since I reject moral universality. (To paraphrase Max Stirner, the murderer who assails me is in the "right," and I who strike him down am also in the right: for what I defend against him is not my right, but myself.) But if we wanted to situate the issue within a universalist framework, I think the thing to say would be that the only sort of "moral justification" that should matter is the sort that attaches to "good actions," which we can call just those actions that would flow from a desirable (or at least acceptable) character under the relevant conditions. On this view, if we believe that vengefulness is a defective character trait, then acts of revenge are not morally justified, no matter how intense their felt-appropriateness.

[Rob]

But Michael, as a considered result of an imagined scenario, isn't Jack's attitude towards the offender already engaging both of the registers you describe? And as a normative claim about what it is right to feel in such circumstances, isn't his attitude more than a mere feeling at his reflective disposal to be placed amid competing considerations? By characterizing his attitude as a "feeling" or a species of "vengfulness", I wonder if you haven't rendered it something more manageable than it in fact is.

[Tamler Sommers]

Thomas,

Yeah, I figured that. Of course, just because these atiitudes were adaptive for my ancestors, that doesn't mean they have instrumental value now. As I said, I have no idea whether they do or not. But your idea, I take it (I've defended this view myself) is that my belief about the intrinsic value of retribution can be explained away as a post-hoc justification of a (once) adaptive attitude. Is that right? Maybe. The problem is that Jack knows that origin story, he wrote it up himself one time, and he STILL believes that the offender ought to suffer regardless of any consequentialist benefits that might arise from it.
You write:

"You seem to be (I believe mistakenly) assuming that one could not be a free will skeptic and think that there is value in making wrong-doers suffer for the harms they create."

But that's not the case. I think that the free will or MR skeptic cannot believe that there is value in that suffering over and above consequentialist considerations. You seem to agree. My point is that Jack does feel the offender ought to suffer independent of consequentialist considerations (again, upon reflection, fully cognizant of why he might feel this) and so, on that point at least, he's being inconsistent with his MR skepticism.

Rob, you're right, Jack MIGHT do this. And he's aware that this is an option. I would never say that's impossible. But in this case, upon reflection, Jack has endorsed both the first and second order attitudes. And he doesn't (yet) see any principled reason for embracing your alternative.

Michael,

Maybe I'm misunderstanding, but why should Jack favor the general principle "vengeance is a defective character trait" over the intuition "revenge in this case is morally justified"? Assuming the two judgments conflict (I'm not sure they do), why couldn't he just revise the general principle (a la Rawls)?

[Eddy Nahmias]

Thanks for planting a seed in the Garden, Tamler. Clearly, things have been a bit barren recently and we needed a good cold-blooded murder to talk about. It seems like we're trying to decide between Jesus and Jack Bauer here, and both have their attractions.

Is there room here for a position which allows Jack to say:
1. my daughter's murderer is morally responsible for what he did (having the compatibilist capacities), he is a truly bad person, and he fully deserves to be the target of my and my community's reactive attitudes, such as resentment and moral outrage.
2. it is necessary to lock him away to protect the community (and perhaps to satisfy some important needs of the victim's family, etc.)
3. causing him suffering for the sake of suffering is not really justified, even if Jack feels it is.

I'm not clear why the whole MR debate hinges on the intrinsic value of causing suffering to wrongdoers.

[tnadelhoffer]

Eddy,

The issue concerning the intrinsic value of suffering is relevant because there is a principled difference between saying that someone deserves to suffer for wrong-doing and saying that it would be appropriate for someone to suffer for wrongdoing. When you say "causing him suffering for the sake of suffering is not really justified" I believe this is tantamount to saying that he doesn't deserve it--at least as far as desert has traditionally been used in the literature on punishment (think of Kant here). At the end of the day, I think compatibilists can deliver the goods on 2, I just don't think they can deliver the goods on 1. That is precisely why I would prefer that compatibilists not use the term 'desert' so expansively. To the extent that you think the murderer in Tamler's example deserves to be the target of our resentment and outrage, you must think it is acceptable for us to make him suffer for his wrong doing. But either the desert you mention in 1. just collapses into 2.--which is to say, when you say he deserves it, you mean their are forward-looking reasons that justify the suffering--or you are using desert in the more robust sense that you then turn around and deny in 3. Everything here hinges on what you mean by desert in 1.

[Tom Clark]

Tamler, you wrote:

"The problem is that Jack knows that origin story, he wrote it up himself one time, and he STILL believes that the offender ought to suffer regardless of any consequentialist benefits that might arise from it."

What are Jack's reasons for this judgment?

and similarly you wrote:

"My point is that Jack does feel the offender ought to suffer independent of consequentialist considerations (again, upon reflection, fully cognizant of why he might feel this) and so, on that point at least, he's being inconsistent with his MR skepticism."

Is there anything operating here besides Jack's natural reactivity commandeering his value hierarchy? It would seem so, because you say that he feels this way *upon reflection*. How do you see these reflections proceeding such that Jack ends up acting inconsistently with his well thought through MR skepticism?

This is the story that needs to be filled in to make compatibilist retributivist (non-consequentialist) desert coherent, not the scandal it manifestly appears to be.

[Neil]

Shorter Tamler:

You could use facts to prove anything that's even remotely true.

[Michael Drake]

Tamler, I'm not sure based on the facts I can say anything about what Jack *should* do; as the saying goes, I don't know Jack.*

But if the question were put to me, I would say the felt-appropriateness of vengeance triggered by vividly imagined paradigmatic injustices is outweighed by the following: the (now) environmental inappropriateness of the response; the danger that our passions -- particularly our violent passions -- can lead to rash acts (including rash acts of reasoning); the bad faith of all the standard rationalizations for retribution (and the nagging suspicion that should therefore afflict us when we find ourselves happily finding a path strewn with "reasons" that leads neatly to the dissipation of our occurrent urges); and by its tension with the near universal admiration for those reputed to meet violence and indignity with equanimity (and Rob, this example goes to your point about manageability).

*Though (to echo Tom's comment above) one thing I've learned about Jack is that it seems likely, for any n, that Jack's response to further incompatibilist considerations c1, c2,..., and cn will be "I don't care"!

[Eddy Nahmias]

Let me help Jack restate his case. He's thought about this a lot. He used to think that TNR was so intuitive that no one could deserve moral condemnation (Thomas, I can't see why the reactive attitudes *require* the belief that their target deserves to suffer just for the sake of suffering). Now, Jack thinks it is more intuitive that the killer deserves moral condemnation.

There are lots of arguments for free will/MR nihilism, but Tamler has correctly pointed out that they all seem to rely on a principle like TNR.

I ask again: what is the reason to accept TNR (preferably an argument, not just an appeal to intuition)? (just a reminder: TNR is a principle that says something like: an agent cannot be responsible for doing A if he is not at least partially responsible for the conditions that bring about his doing A.)

[tnadelhoffer]

Eddy,

Reactive attitudes are just that--attitudes. They need not necessarily translate into actions. I could very well be upset or off put by your behavior but not do anything about it. If you take what Strawson says seriously, then it doesn't make sense to fuss about whether our attitudes are justified. I, do, however think it makes sense to talk about our actions needing to be justified--especially when these actions are aimed at intentionally harming someone. So, there are reactive attitudes on the one hand, and re-actions on the other hand. Presumably, if you have a sufficiently negative reactive attitude such as moral anger, it will lead you to want to do something about it. That's where the suffering comes in.

Keep in mind that when Tamler started this thread, he was not simply talking about attitudes. He specifically said that the he would judge that the wrong-doer deserves to suffer in this case. So, let's stick to the original framework of the discussion.

Typically, when people say that someone deserves to suffer that means that they believe the appropriate authority ought to make that individual suffer (sometimes this might be the immediate victim, sometimes it might be the state). Moreover, at least according to the way the term desert is normally used in the literature on responsibility and punishment, to the extent that someone deserves to suffer, there is intrinsic value in making them suffer. Now you can use the term however you'd like, but you need to be careful that you use it in a way that enables you to properly distinguish between your aforementioned 1. and 2. My response to your earlier comment was supposed to present you with a dilemma.

When you say that P deserves to suffer for x-ing, you either mean that making P suffer is intrinsically valuable or you mean that it is merely instrumentally valuable.* If you mean the latter, then the distinction you tried to draw earlier between 1. and 2. collapses. That's fine by me, but that's a very weak notion of desert with which you're operating. If, on the other hand, you mean the former, then I would like to see the argument for the intrinsic value of making someone suffer for their wrong-doing that doesn't just come down to the fact that you happen to have a perhaps unshakably strong intuition that they deserve it.

[Robin Aldridge-Sutton]

Hey Tamler

What if Jack had independent reasons to doubt the reliability of the intuitions grounding his belief in retribution justified by compatibilist conditions? Say he read some empirical work that showed that such intuitions in such cases are distorted by the emotion involved. They don't reflect a balanced cognition of the relevant facts, they tend to dissipate over time as the emotion fades, and so on. In that case might he decide not to trust those intuitions and prioritise those grounding his belief in hard incompatibilism?

[Tamler Sommers]

Whoa, a lot responses (on behalf of poor Jack):

Eddy,

Unlike Michael, I do know Jack, and he wants to say more than: "the guy who harmed my daughter deserves to be the target of resentment and outrage." He wants to say that the guy deserves punishment, hard treatment as Feinberg would say, at the very least. He agrees with Thomas that it's hard to call anything else full-blooded compatibilism in this kind of circumstance. But thanks for restating his case nicely in the next post.

Tom, I'm wondering if it begs the question to say that Jack's new intuition, upon reflection, when the offense hasn't happened yet, is "commandeering his value hierarchy. Why should his intuition about Professor Plum or premise 2 in the Basic Argument be higher up in the hierarchy than his all-things-considered intuition about what he thinks is appropriate in this scenario? I think there might be a response but it's harder than I once thought (in say, my Illusion of Freedom evolves paper) to develop it. And it still might ultimately be unsuccessful.

Shorter Neil:

My comment makes no sense but at least it's short.

(Seriously, I have no idea what you're getting at. As I sometimes say in Apples to Apples: if it's a joke, I don't get it.)

Michael (and Tom),

You're right, he shouldn't have said: "I don't care." He doesn't mean that exactly. What he means is: "I believe more strongly that the guy deserves punishment than I believe the key premises in the skeptical arguments. And I'm aware of the origin of this feeling, but I don't see why that should completely strip it of reliability."

Robin,

Yeah, I think that's the way to go, but I'd need to know more about those independent reasons. Your example doesn't seem to apply because the offense hasn't even occurred, so what is there to dissipate over time? If he found out he'd been hypnotized to feel retributive every time he contemplates harm to his daughter, he'd certainly have to take that into account. You might claim that some combination of biological and sociological forces are the equivalent of this kind of hypnosis, but that's a tough case to make given how reflective he's at least trying to be about this whole matter. I do think though that this is the only way out of the deadlock that Eddy described in his last comment--a kind of balancing act that takes into account the reliability and strength of the competing intuitions, as well, perhaps, as how well each coheres with his other moral and non-moral beliefs.

I want to say that I made Jack his own person on purpose. Although I obviously have some sympathy with his perspective, that allows me to press his argument farther than I might otherwise. I'm not necessarily Jack, although I know him well.

(That's it, sorry, Neil might be right: there needs to be a shorter Tamler.)

[Tom Clark]

Tamler wrote:

"Why should his intuition about Professor Plum or premise 2 in the Basic Argument be higher up in the hierarchy than his all-things-considered intuition about what he thinks is appropriate in this scenario?"

An all-things-considered intuition seems a bit strange, since when considering all things we usually take our (often inchoate, emotionally driven) intuitions as one element that feeds into our considerations. For instance, the intuition that those who harm our daughters should suffer non-consequentially might be considered in the larger context of what sort of culture we want, the virtue of restraining one's reactive impulses, the humanitarian principle of minimizing unnecessary suffering, etc. So when you say Jack, upon deliberate reflection, stands by the intuition that anyone who harmed his daughter should be made to non-consequentially suffer, isn't this the sort of mental calculus he's carried out, or should have carried out?

If so, I'd like to hear from compatibilists how Jack reaches the reflective conclusion about the appropriateness of non-consequential suffering. His intuition can't be all-things-considered without having undergone this sort of contextual evaluation.

Jack says

"I believe more strongly that the guy deserves punishment than I believe the key premises in the skeptical arguments. And I'm aware of the origin of this feeling, but I don't see why that should completely strip it of reliability."

Here Jack seems to be going with his feeling (gut intuition) as a reliable indicator of a truth about the world: that retributive desert is just. Presumably there are independent means of confirming that truth, otherwise he couldn’t reasonably conclude that the feeling is reliable. So, what are those means? What’s the argument that backs up the intuition that non-consequentialist desert is just? Again, I’m simply highlighting the fact that feelings and intuitions aren’t justifications in and of themselves, but only play a role in justifications that consider all things, or at least several, beyond the intuition itself. I guess this is something like what you said in response to Robin:

"a kind of balancing act that takes into account the reliability and strength of the competing intuitions, as well, perhaps, as how well each coheres with his other moral and non-moral beliefs."

[Tamler Sommers]

One thing I just realized, my original post reverted back to an earlier draft, and didn't include a passage I added later--something that's somewhat relevant to Tom's post. After saying that Jack believes the retributive attitude to be the "right or appropriate response," I added "he thinks it would be obscene to worry about whether the offender was causa sui, or to excuse him because the causes of the act trace back to factors beyond his control."

This seems important in the context of why Jack sees this judgment not merely as an irrational impulse, but something that he has, on balance, considered to be the right judgment. Now, again, someone could try to explain away the "obscenity" of worrying about incompatibilist conditions at a time like this. But that's more work (for the hard incompatibilist).

[Robin Aldridge-Sutton]

Yeah that bit's really important. It is actually in there in the third paragraph though.

Thanks for replying to my comment. You're right the emotion wouldn't fade since it's provoked by reflection rather then an event. I still feel like this emotion somehow lowers the reliability of the judgement, but I know plenty of people think emotion is necessary for good morality. This is why I'm really interested in the theme in experimental philosophy of tracing the psychological processes that produce these kinds of conflicting intuitions. But really I think understanding them enough to be able to confidently reject one set for the other may be a long way off yet though.

I think the answer to your original question must be yes, Jack must rethink his position on the compatibility question. The odd thing is that positing moral responsibility is the default setting, so it seems like something must've changed since he chucked that out for some incompatibilist premises. Maybe he had never considered this kind of case (of harm to his daughter) before, or he was very unusual in never being much committed to attributing responsibility? Or has something else changed - has he had kids maybe?

[Tamler Sommers]

I don't know about Jack, but the first time I had inklings of this kind of objection was in Spring, 2004, when I was teaching a seminar on free will in grad school. My daughter was born two weeks earlier...

[Cihan]

Tamler,

What happened to unreliability of intuitions? What happened to meta-skepticism? I guess this Jack character isn't as like you as our commentators have interpreted.

"Now Jack is fully aware of his incompatibilist commitments.He finds the four-case argument and the basic argument no less compelling."

That's the pre-metaskepticism Tamler.

Well you might say that even though Jack still agrees with the arguments of metaskepticism, Jack doesn't even care about those arguments - the unreliability of intuitions or what not!

But look what happened. Due to his understandably emotional response to a hypothetical case involving his daughter, Jack got anchored to a position so tightly that no reasoned argumentation is going to have any effect on him.

That seems a little worrying. One should always have a door for reason - I'm not saying that there is anything wrong with Jack's response - it's just that you always have to have the "fuse" of reason.

[Paul Torek]

Eddy,

There's a good reason why you're "not clear why the whole MR debate hinges on the intrinsic value of causing suffering to wrongdoers." Because it doesn't. Let's throw that red herring back in the water. Talk of intrinsic value makes this issue out to be axiological, when it is actually about right and wrong. Maybe we should blame Kant, whose status as exemplar of deontology is perhaps debatable.

What ought to be contrasted with the instrumental value of punishment is not the alleged intrinsic value of the perpetrator's suffering, but the fairness of the proposed actions. I.e., the fairness of the institution of punishment including its application to this perp. And along this route we can find plenty of flesh on the bones of Jack's claim to have an all-things-considered judgment - not just a feeling - that this hypothetical perp would deserve punishment. We can even trot out our favored conceptions of fairness and justice and of how to reason about them.

[tnadelhoffer]

Paul,

First, that you happen not to be fishing for the intrinsic value of suffering doesn't mean that others aren't. One person's "throw back" is another's "catch of the day."

Second, it is unclear how your shift to fairness/justice is supposed to avoid the problem if you're simply assigning these intrinsic value. On your view, if it is just/fair for me to x, does that make x-ing intrinsically valuable? Conversely, if it would be unjust for me to fail to x, does that make my non-xing intrinsically bad?

If so, then to say that P deserves to suffer for x-ing is just to say that not making P suffer would be intrinsically bad and making him suffer would be intrinsically good.

If, on the other hand, you don't think the intrinsic value of making wrong-doers suffer has anything to do with the fairness/justice of punishment, then perhaps you could say a bit more about how fairness/justice helps in this context that doesn't just collapse into the sort of forward-looking considerations I mentioned above.

In short, I would like for you to "trot out" some of the conceptions of fairness and justice you think would help here that (a) supports Jack's change of heart, and (b) avoids the objections that have appeared thus far in the thread.

[Paul Torek]

My own favored conception of justice and justification owes a lot to Jürgen Habermas and Tim Scanlon. Both philosophers focus on norms and principles that those who discuss morality with the aim of agreement could reasonably accept or reject. Axiology is still in the picture, in my view, because what you could reasonably accept or reject depends on what you value - but we needn't value the same things to agree on norms. You can value dancing more than astronomy, and I the opposite, yet we can both agree on the social-political norm that each of us should be free to do what we love. People who view the suffering of perpetrators as intrinsically valuable, and those who don't, can both agree that we need to be protected from murderers. Even those who worry that they may some day, through a series of unfortunate events, turn out to be murderers themselves, can agree to this.

You don't have to be a "contractualist" (bit of a misnomer IMHO) or discourse theorist, though, to throw away the red herring of intrinsically valuable suffering. Just about any kind of deontology will do. So too would many forms of virtue theory, I think. Any of these approaches can ask whether punishment is right, without either reducing the question to a utilitarian calculus or hinging on intrinsically valuable suffering.