Blog Coordinator

« CFP: 3rd Annual Philosophers' Cocoon Philosophy Conference | Main | Miranda Fricker on Blame and Forgiveness... »

05/20/2015

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Hi Katrina,

Can you say more about why you think the capacity for diachronic self-control (which I agree Bert possesses) grounds moral responsibility? After all, Bert also possessed the capacity "to take care of his kids", "to stay home", "to recall his obligations", and other capacities. If these capacities don't ground his responsibility, then why should this other capacity succeed? And if those capacities do ground his obligations, then why invoke this special capacity?

A quick question in line with Neil's, Katrina: I'm not sure I see what the diachronic self-control aspect is doing for you. That is, why not just go with Smith's rational relations view, which gets direct responsibility for Bert regardless of his past attempts at controlling his future? Indeed, the two forms of control in the two theories are quite different: Roskies is appealing to a kind of *volitional* control, whereas Smith is appealing to *rational* control, which just has to do with the kinds of attitudes an ideally rational agent would bring about in line with her evaluative judgments.

Hi Neil,

A great question.

There are three primary reasons why I think responsibility hangs on the capacity for diachronic self-control.

First, I agree with Adina that diachronic self-inteventions are our best bet for grounding compatibilist free will. I think it is because we are capable of these sorts of self-interventions that we can be praised and blamed for what we do. So the capacity to perform these sorts of interventions seems like a good candidate for the thing moral responsibility tracks.

Second, I think current theories in neuroscience support (in a vague sense) my view. I actually agree with most of your views on the neuroscience of moral responsibility. I believe in something like the global workspace theory of consciousness, and that rational, deliberative thought happens via connectivity between the executive and much of the rest of the brain within the GWS. But as Adina notes, our decisions and thought processes within the GWS have important downstream effects on what we pay attention to (what enters the GWS), implicit attitudes, emotional responses, etc. I think you downplay the importance of agents' ability to perform these sorts of top down manipulations in your book. If we can perform the sort of self-interventions Adina discusses - an empirical question, I suppose - then it seems to me we are responsible for more than just actions that result from deliberative, conscious decisions. We are responsible when we don't notice something, or fly off the handle in a rage.

Third, I think it is important that the criminal law, which is in the business of high-stake responsibility assessments, seems to hang responsibility not just on deliberative, conscious mental states, but minimally conscious, and even implicit, attitudes. I'm just speculating at this point, but there seems to be something about the way in which law, and other moral rules, are internalized by agents so as to be law-abiding that supports my view. Agents very rarely consciously consider rules as a means to be rule-abiding. Instead, these rules are internalized, which may just mean that they are used as reasons to manipulate oneself to be a certain sort of person - one that pays close attention to where their kids are but not necessarily where their keys are, one that notices red octogons, one that learns to diminish their aggressiveness, etc.. I also think a sophisticated analysis of the purposes of punishment may reveal that punishment just ain't aimed at harmful conscious decisions alone, but on harmful agents as well. But now I'm really speculating.

Hi David

Hopefully my response to Neil made clear that I think Adina's volitional account is more in keeping with my views of the neuroscience of decision-making. And I think we only need to look back to Bert's past attempts to control his behavior as a means to determine if he has the capacity for such self-interventions. That is, if Bert hadn't been able to hold down a job, or keep other important appointments, we might think he didn't have full diachronic agency. The sort of control Angie and Adina talk about is different, you're right, but both are diachronic. I brought up Angie's views primarily to show that this sort of view can ground direct responsibility, not just blame for character.

"intervene in our future selves, says Roskies, by manipulating our mental content in ways that have foreseeable consequences...[so] we are in a 'very real sense responsible for who we are' and our behavior."

I'd just reiterate my earlier comment re supererogation - that Bert's lapse is easily understandable but falls outside that of usual behaviour of the average parent. We might, to take a topical example, expect him to read to his children, but won't take him to court.

I agree completely that we can act at t to alter the likelihood that we attend to X at some later time; the contents of the GWS are a function in part of our self-formation. But I allowed for that: my claim always concerned direct moral responsibility. If an agent is responsible for failing to attend to X, then saying that she had the capacity to self-form in such a way that she (most likely) would have doesn't entail that's responsible unless it is the case that she should have so self-formed. And now we can enquire about these past episodes. We need to identify what Holly Smith calls a benighting act to ground her MR. And our questions arise once again with regard to that benighting act.

Hi Katrina,

"Bert was capable of the sort of self-authorship Roskies describes. With regard to his lapse, Bert could have engineered his environment such that he was less likely to forget his parenting schedule: he could have set a reminder or kept a calendar, for example. Bert could have made a conscious commitment to be a reliable parent and set for himself policies to meet this commitment, such that when the topic of his children arose this commitment came to mind."

Is this meant to be a strictly incompatibilist account, or is it intended to work under determinism?

I'm envisioning a Bert advocate disputing the claim that he "could have" engineered his environment so as to preclude the lapse. To do that, the *thought* to do it would have had to have popped up in his mind, as tends to randomly happen with such things ("Ooh, I might forget to pick her up, I better put an outlook reminder in!). But that thought didn't pop up, so what gives? Retributive pain infliction, done for reasons unrelated to the goal of ensuring he remembers next time, seems mean.

I'm also concerned about a Galen-Strawson-type-regress where the final lapse (forgetting to pick up his child) is due to another lapse (forgetting to put a reminder in outlook) is due to another lapse is due to another lapse is due to another lapse and so on, with a regress that ultimately ends in some initial condition that has nothing to do with Bert. In that case, where would the responsibility get off the ground?

Neil, thanks for all your comments. I think I remember reading Smith's argument about benighting acts some time ago -- I'll take another look.

Hi Jesse, Roskies certainly means to provide a compatibilist account. And I agree there may be an issue here, as there is with any capacitarian account of MR, of how we work out what "had the capacity to do" or "could have done" means in a deterministic universe. Usually the move is to give some sort of counterfactual account. I really like Nicole Vincent's 2013 paper on this topic, in which she argues that *any* capacitarian counterfactual account might not be fair, even where the conscious thought to x does pop up, because ought really does imply can, and in a deterministic world, if you didn't, you couldn't have. Specifically, she wonders why it is fair to hold someone who lives in this world responsible for what their mechanism may have done in another world (e.g. responded to the reason that popped up).

From this perspective, it isn't clear why cases where the conscious thought to do x pops up are indicative of "could have done" in a stronger sense than situations where the thought doesn't pop up: neither agent that didn't x really could have x'd. So Vincent comes up with different test (in a paper forthcoming). To decide if a given person should be blamed, we should: ascertain from what sort of mechanism did their action issue, what mental capacities did they have; and ask if it would be a fair system that held responsible this kind of person in this way for that transgression. On this sort of account, it may be just as fair to hold someone like Bert responsible as a person who had the conscious thought to put in a reminder.

Re: the Strawsonian regress worry - I guess one strength of a capacitarian account is that it sort of side-steps this issue because it doesn't have to use tracing. Or, as many of you seem to think, it may be a weakness of the theory that it sidesteps this issue. :)

It doesn't strike me that the capacitarian analysis of control is inconsistent with the truth of determinism. Bert's capacity to obey the law and meet his responsibilities needn't require anything like regulative control or libertarian freedom. The question of whether or not Bert had the ability to prevent the lapse could just be construed in terms of whether or not he had the (morally) relevant psychological equipment to do so.

I assume that when Katrina speaks of diminished cognitive capacities (although please correct me if I'm wrong) she means that most people can internalize the law by (among other things) observing how others have been punished for not obeying it, but there might be a particular agent with something "broken" in their higher-order cognitive abilities such that observation of punishment does not yield secure knowledge. This could be due to something like psychological disorder, the inability to sufficiently store memories, and/or physical damage to their cognitive faculties and executive functions (as in an impaired prefrontal context for instance). That's my takeaway, but Katrina might have something else in mind.

I'm still of the mind Bert's responsibility can be grounded through tracing, however, as it's less clear to me that mere mental capacity alone will secure the conditions for moral blame (although I can see more plausibility for this in terms of legal blame).

Hi Katrina,

Thanks for the clarification. What you say makes sense and appears well thought out.

I have some more thoughts that I think you might find interesting. Before getting into them, let me disclose my own approach to these issues, I am a hard anti-realist about responsibility and desert. I believe that when we put the question to reality, "is X_ responsible for Y_ing, and does X_ therefore deserve Z_?", reality does not have an answer, no matter what X, Y, or Z are chosen to be. This is an unpopular view in philosophy--but to me, it seems obvious and incontrovertibly true.

I acknowledge that we perceive the world in terms of responsibility and desert, but I consider these perceptions to be evolutionary constructs, similar to the constructs of color qualia. Most scientists and philosophers take the view that color qualia--e.g., the "redness" of an apple as we perceive it--does not exist "on" the apple, as appears to be the case. Rather, it exists only in the mind--only in the organism's visual experience of the apple. The organism has evolved a mind/brain that perceives the apple with a quality that doesn't actually exist outside of the mind/brain because the quality consistently maps to crucial pre-phenomenal information that does exist outside of the mind/brain. For an organism to receive that information and act in accordance with it is an adaptation that contributes to survival and reproduction.

Now, suppose that an alternate species, otherwise identical to our own, had evolved alongside ours, but that the evolutionary contingencies underlying the development of that species had been slightly different, such that its perceptions of color ended up entirely inverted across the spectrum relative to our own. Where we see red, this species sees green, and vice-versa. The prevailing view is that our species would not be able to claim that we have more knowledge about reality than this alternate species. If a heated argument were to emerge between members of the two species as to whether apples were truly "red" or "green", reality would not have an answer.

The same holds true, in my view, with respect to resonsibility and desert. We have strong perceptions of their existence, even though they are perceived in a visceral and ontologically-imprecise way, and in some cases are perceived entirely unconsciously. Even though the qualities themselves do not exist outside of the mind, having a mind/brain that perceives them to exist, or rather, a mind/brain that receives the information underlying them, and that is inclined to be strongly viscerally motivated to act in response to them, in the form of punishment and reward (and other types of resource distributions), is very adaptive. The punishment and reward perform critical functions--teaching, incentivizing, and, in the case of punishment that severely harms or kills, disabling and gene-disposing--that ultimately promote the survival and reproduction of individual and group.

Now, punishment has a cost. It uses up time and labor resources, incurs risk of successful retaliation, and inflicts potentially unnecessary damage on entities. It is for this reason, in my view, that the viscaleral motivation that we experience to mete out punishment is highly specific. Generally, we feel it in response to the behaviors of organisms that are reasons-responsive, where the transgressions emerge from reasons, or are chosen because of reasons. Notice that those are precisely the individuals for whom punishment will be the most powerful and effective, in terms of discouraging future transgressions. Evolutionarily, reasons-responsive individuals are the individuals for whom the punishment will be most worth the resource cost, and will deliver the biggest bang for the buck (benefit to cost).

Of course, we can still feel a retributive impulse against entities that are not reasons-responsive. If a snake kills a child, for example, we will want to kill the snake as punishment, even as it is not reasons-responsive, but is just an automaton. This is because the driver behind the evolution of a retributive brain is not *just* teaching/incentivizing, but also disabling and gene-disposing. If we kill the snake, it will never again kill one of our children. Even better, if we view all of the snakes in the batch as "evil" and "disgusting", deserving of being smashed into goo, and we kill them all, we will be that much better off.

The individuals for whom we feel the strongest retributive impulse are those that are fully rational, fully reasons-responsive, but that *still* choose evil, because the only reasons that speak to them are their own selfish reasons. That is, those that are not responsive to reasons associated with the well-being of others, but that are highly responsive to reasons asssociated with their own well-being. All of the worst people in history exhibit such a tendency--entirely rational psychopaths. It is not a coincidence, in my view, that those individuals are the individuals for whom punishment is most effective--gives the biggest bang for the buck--relative to cost/risk. For those individuals, the threat of punishment will produce the greatest behavioral modification.

It is for this reason, in my view, that we experience different levels of retributive impulse in response to different kinds of mental illness and disability. If a person is severely delusional, to the point of not having kind of coherent intelligence--someone who slathers saliva everywhere and screams and flails and has glassy eyes that seem to not register anything, such that we don't perceive the rational, reasons-responsive essence of a person to be "there"--and that individual commits some egregious act, our retributive impulses will soften, even though we still might kill him. But if a person has a different kind of defect or disability, one that allows him to retain reasons-responsiveness--for example, some kind of severe, irrestible sexual perversion--that leads to an egregious act, the retributive impulse will not soften at all.

The pedophile who tells the community, "I absolutely cannot resist. You don't understand, to me, it just feels overwhelmingly, unbelievably good! And since I don't naturally feel much compassion for others, there is nothing to counterbalance the good feeling I get!", is not going to get any mercy for that fact, even as we acknowledge that his mind/brain is somehow "not normal." Now, if we prime the externality of the defect by focusing on its physicality and its status as an intrusion into what would otherwise be a normal brain, such as by pointing to the tumor that is entirely creating the condition by impinging upon specific brain areas, then our retributive impulses might soften--but they will not soften if we leave the individual's mind/brain as a "black box" that is just generically "perverted" and "depraved." In that case we, we will say that he is *intrinsically* evil, and that he deserves to suffer. He may not speak the language of other people's suffering, but he clearly speaks the language of his own suffering, and so we're not going to hesitate to aggressively communicate through that channel, by inflicting severe pain on *him* in retribution for his actions.

Note that the evolutionary benefit to having a mind that optimizes reward and punishment resources in this way, meting them out based on cues associated with their potential effectiveness and utility, is entirely unconscious, just as the evolutionary benefit to sex is unconscious. We harm rational people that intentionally transgress with a mens rea because we feel a deep-seated drive to do so, just like we have sex because we feel a deep-seated drive to do so. In both cases, we are completely unaware of the ways in which what we are doing promotes evolutionary goals.

To summarize, I do not believe that we can make genuinely true and false statements about what a person is "responsible" for, or what a person "deserves." Reality will not come to our defense in those claims, as it would if we were to make the claim that if we hold up a ball and drop it, it will fall to the earth (in some specific amount of time, exhibiting a specific acceleration). To be able to ascribe responsibility and desert, as ontological, extra-mental properties of persons, we would need to establish a privileged carve-out of "person" from the rest of nature. But no such priviliged carve-out exists. Where does the rest of "nature" stop, and "me" begin, such that at that point "nature" stops being to blame for what is happening, and "I" start being to blame for what is happening? We cannot give a true answer, and will *never* be able to, because none exists.

If an alternate species were to evolve alongside ours, and, due to facing different evolutionary pressures, were to evolve in such a way that they lacked the instinctive, intuitive tendency to ascribe responsibility and desert in the way that we do--for example, if this species were to see no one as ever responsible for anything, or of deserving good things or bad things in response to anything, or if this species were to be much harsher than we are, and were to see people as inheriting responsibility and desert from other people--that species would not be "missing" something about reality that we are catching.

Now, with all that said, I do think we can make true and false claims about what our intuitions about responsibility and desert are, even though these intuitions are not objectively true, and even though they are often in conflict with each other. And so I think your project of trying to figure out the "right" response to cases where mens rea is lacking is tractable, provided that by "right" we mean the response that best harmonizes with our (often-conflicted) intuitions, rather than the one that is the philosophical truth.

In the case of Bert, I reach a seemingly weird intuitive conclusion. If I know that he values his kids and always has, and if I know that he feels terrible about his lapse, my sense of his responsibility significantly weakens (or altogether disappears). I'm willing to accept the defense that we can't control these kinds of unfortunate lapses, they just happen, and it's no one's fault. I'm also able to step in his shoes and feel the force of that point--of course it's easy to have the thought "My kids! I have to pick them up!" when you're already in trouble for having forgotten them. But it's very difficult--in fact, it's impossible--to have that thought when the conditions of the environment and of your own mind/brain are such that you DON'T have it, period. Of course, we can say he should have taken precautions to ensure that he not forget (put an outlook note in or a buzzing iphone reminder), but this just pushes the dynamic back one level. He needs to feel the impulse to take those actions at the time that they need to be taken--but he didn't feel that impulse at the requisite time (though he certainly feels it now). We all know what it's like to be the recipient of ex-post monday-morning-quarterbacking. It's not helpful after the fact.

Now, if we switch things up, and assume that he doesn't value his kids, and doesn't feel terrible about his lapse, then my sense of his responsibility, and of his deservingness of punishments, strengthens dramatically. That character flaw almost becomes its own mens rea. That's why I think your introducing Las Vegas as the place where he goes--which implies gambling, vice, flawed characters, shady behaviors, a person who doesn't take things seriously--is doing much of the work here. Instead of having him indulge in some superficiously scandalous activity, let's just suppose he forgets. Instead of picking the kids up, he just continues doing whatever innocent thing he was doing--say, mowing the lawn. Clearly, we are more merciful with him, it seems. To go to the other extreme, suppose he forgets his kids because he is watching porn. As expected, the intuitions shift dramatically towards more responsibility and more punishment, for what is, admittedly, the same underlying lapse.

I'm not sure this result is defensible. Of course, it's defensible in that it squares with the intuition, but it conflicts with other intuitions. What difference does it make whether a man forgets to fulfill his obligations because he's mowing the lawn or because he's watching porn? Both behaviors are fully legal, and both omissions arise from the same underlying type of lapse (forgetting about commitments b/c you are occupied with something else). Yet the superficial scandal of the latter causes us to want to punish more.

I come to another weird conclusion in the hot care death case. Let me start by pointing out that in the hot car death case, we are obviously merciful because nature is already meting out such a harsh and unforgiving punishment on the mother, who fails her own child, and who will have to live with the extreme suffering that will entail. There is no need for us to add more suffering in the form of an incarcerative or corporal punishment--that would just be wrong, mean beyond description. For that reason, I think we should change the example up a bit. That is, posit that the baby that dies is someone else's baby--not hers. Maybe she is a daycare provider, and makes a similar mistake arising from a similar lapse, causing someone else's baby to die.

As before, it seems our sense of what she deserves depends on superficialities that should not matter, but that do. If the caretaker is a 21 year old club-hopping stripper who works in daycare part-time during the day, and the woman who loses the child is a 35 year old homemaker, then the sense is that the caretaker deserves more punishment. But if you flip things, and make the caretaker a 35 year old homemaker working part time as a caretaker, and the mother is a 21 year old club-hopping stripper who needs somewhere to dump her kid, then the sense is that the caretaker deserves less punishment--even if it's the exact same underlying lapse, driven by the exact same underlying mind/brain process.

Also puzzling, and this may be unique to me, is that the very kind of blame and forgiveness logic that we are willing to endorse seems to be a function of what she feels *after the fact.* Suppose that she is a highly self-compassionate, self-understanding person. She quickly absolves herself of guilt and shame for the lapse on the basis that the event was an accident, and that it could have happened to anyone. It seems that our intuitions become more retributive, and that we would be more willing to incarcerate or flog her. If I was the concrete victim, and she responded that way--by being "sorry" for my loss, but by not accepting responsibility for it, and instead aggressively defending herself, even as she is the one who has "caused" all this, from my perspective--I might even want her tortured or killed for it. This holds true even for involuntary things. If she was driving a car and had a heart attack, passed out, and got my child killed, the retributive impulses would still be very hot, even as the state of her "mental equipment" precluded the accident being avoided. It doesn't matter, *she* caused this mess, she (and her health problems) brought it into my world, and so I will see her as a target for retribution (at least initially, pre-reflection). I will gladly come up with ex-post rationalilzatios for her guilt "She shouldn't have been driving a car if she has heart problems, she should have taken care better care of herself, rather than eating unhealthy like so many obese Americans, she should have done whatever she needed to do to not cause so much damage to MY life!"

Now, conversely, suppose that she is an empath or a depressive, and ends up feeling crippling levels of guilt and shame in response to the tragedy. I can easily see us then moving in the *opposite* direction, trying to console her by communicating to her the same responsibility-softening and responsibility-eliminative logic that we would *not* have endorsed if she herself had casually cited it in her own defense. "Don't be so hard on yourself, it was an accident, it could have happened to any of us!" "You can't help that you innocently forgot something." "You can't help that you had a heart attack and passed out!" "It's not your fault" etc.

For me personally, in saying that to her to console her, what I would be offering would *not* be deceptive consolation. To the contrary, I would be making a point that I would actually find compelling, in her specific case. Her lapse was an accident, a failure of a thought/feeling/memory to arise in her mind/brain, the responsibility for which can just as easily be "put" onto nature, physics, God, or any antecedent in the chain, as on to "her" (however we choose to carve out "her" from the environment), and that *any* of us, with a little less luck, could have ended up exactly where she is, having suffered the exact same lapse--or even a worse one.

I find this shift interesting. We seem to be OK with responsibility-softening and responsibility-eliminative logic as long as *we* control its use, and deliver it in our own approved-of doses. If she usurps it to absolve herself, independent of us, then we cease to find it compelling or to endorse it.

Hi Jesse

Well, you are certainly right that there are lots of moving parts with responsibility assessments for harm caused by lapses, different sorts of facts that push and pull our retributive sentiments. This is why I found the Bert case so interesting. And I'm happy enough with your evolutionary story about our moral sentiments and their primary targets (rational agents). Psychopaths are indeed an interesting case from this perspective - seemingly reasons-responsive but still cruel. I have actually argued that psychopaths with normal executive functions ought to be held responsible for their acts because they could correct for their emotional deficits.

But - and this is yet another area I'm not too familiar with - I'm not sure why an understanding of morality as an emotionally-driven mind-dependent property renders it unreal. Many philosophers would agree, I think, that color is mind-dependent, but would still argue that color categories are "real" categories. Our color categories may differ from aliens', or dolphins, but so what? We can still claim people like us are wrong when they call a red apple blue, right?

I think the realism of responsibility matters because the law is in the business of holding people of holding people responsible, and courts can attribute responsibility rightly or wrongly. That retributive sentiments are emotionally-laden makes things more complicated, of course, but I also believe that within broad categories movies can be judged good or bad movies, and meals can be judged good or bad meals.

Lapse cases are especially tricky, which is why it is so much fun to talk about them. They do indeed seem to force us to think about how responsible we are for the selves we have turned out to be (our character) and why. I think we are responsible not just for conscious, reasoned choices, but also for harm flowing from the selves we turned out to be because of the sort of diachronic self-interventions Roskies notes we are capable of. I think the law is correct in holding us responsible for this harm because in an important sense it is under our control.

Many of the facts that cause shifts in responsibility assessments discussed in your comment indicate a person's character, which in turn might manipulate our assessment of how out of character the harm caused was. I think this sort of detailed assessment of character might be helpful when picking a partner or babysitter, but probably isn't appropriate in legal responsibility assessments. Indeed, it is something we might ask the court to guard against. In my view, what we pay attention to and fail to pay attention to is important with regard to certain roles (e.g. parent, teacher, police officer) and within our self-control under normal circumstances. Thus when we lapse with regard to important responsibilities associated with those roles, the law rightly holds us responsible, without needing to go into a detailed character assessment.

Hi Jesse,

Your inverted color spectrum scenario wouldn't seem to result in any apparent disagreement. It seems those aliens wouldn't know that there is a difference in perception, and neither would the humans talking to them, since the aliens would have a word – say, “ared” – that would be applied to the same objects the English word “green” applies, so they would probably think the meaning is the same.
In order to raise an example of what might appear to be a case of disagreement, the aliens (say, species#2) may have perceptions similar to our color perceptions, but associated with very different ranges of frequencies, so that – for example -, the aliens perceive yellow and red traffic lights in the same way, whereas they perceive color-like differences between objects that we humans see as the same, color-wise.

Even so, I would agree with the point Katrina raised (if I'm getting it right): for example, if some individual of species#2 comes to Earth and says that the traffic lights we call “red” and “yellow” are the same color, she is mistaken. The word “color” is a word in English, not in the alien language, and its truth-conditions (whatever they are) are those given by usage in our linguistic community, not that of species#2.

Hi Katrina,

I don't want to veer too far off topic, so please let me know if you prefer to leave objections to realism about responsibility aside, but there variants of the evolutionary argument Jesse raised which would hold that:

- In the case of color, interspecies variation is not a problem because individuals of different species (when they have similar perceptions but associated with different frequencies/wavelengths, as in my example in the reply to Jesse's post) would not disagree, but would be talking past each other: their language, for example, wouldn't be color language, but 2-color language, etc. So, if someone calls a red apple “blue”, they're either making a false claim or just not speaking in English.

- In the case of morality, if some aliens (or advanced dolphins, etc.) have sentiments like those that in the case of humans are normally associated with moral judgments (e.g., blame, guilt, etc.), and also have a language associated with those feelings that plays a role similar to moral language in our society, etc. (these conditions can be specified in more detail if needed), then those aliens have moral language (rather than, say, species#2 has 2-moral language), and individuals of that species and humans could have actual moral disagreements, rather than miscommunication if they encountered each other. A Moral-Twin-Earth Scenario may be used to suggest the semantic analysis in this argument is correct.

Arguments like that might be raised in support of a substantive or an epistemic moral error theory, or some other form of moral antirealism (aside from arguments in philosophy of religion, but that's another issue).

As a realist about moral responsibility, blame, desert, etc., my reply would be that just as in the case of color, humans and species#2 aliens would be talking past each other, but as far as I know, that view is not so common.

I also do not see any difficulty in seeing mind-dependent properties as real. My pain is real, and the fact that I feel pain is real, but it is also a fact that depends on my state of mind. This strikes me as an innocuous claim.

Moral claims rest on these sorts of facts about our mental states. A moral community is one where minds are relevantly similar in such a way that true claims about the mental life of the community can be made. This is why contemporary moral philosophers spend so much time in their painstaking attempts to describe the sophisticated mental architecture of moral agents (reason-responsiveness, capacity to form second-order desires, etc).

I think what causes confusion is that in some contexts we want "real" to signify something that, on principle, can exist independently of our opinions about it, or independently of the way that it appears to us. This is because we want to know whether or not the existence of an object isn't just an accident or illusion of mind, and that we are actually representing an object that exists out in the world rather than just in our heads. In other words, what we really want to know is whether or not the thing exists in the way that we think it does, and in this case, as in regard to physical objects, we just happen to want to know whether or not it exists beyond our minds. Thus, the existence of other minds, for instance, can still be "real" in this very same sense in that my representation of moral agents corresponds to actual conditions out in the world.

This somewhat squares with a non-objective realism or subjective/psychological realism view, but I think these terms are misnomers because I can still maintain that minds are objectively real in a mind-independent way insofar as I am making a claim about minds existing as a feature out in the world independently of my personal opinion about their existence.

Hi Katrina,

"But - and this is yet another area I'm not too familiar with - I'm not sure why an understanding of morality as an emotionally-driven mind-dependent property renders it unreal. Many philosophers would agree, I think, that color is mind-dependent, but would still argue that color categories are "real" categories. Our color categories may differ from aliens', or dolphins, but so what? We can still claim people like us are wrong when they call a red apple blue, right?"

If I say this last sentence in a slightly different way, will it still capture what you mean? As follows:

"We can still claim that people like us are wrong when they say that they see the apple as blue."

They are wrong because, if they are like us, then they *don't* see the apple as blue. They see it as red. Interestingly, if you shine light of a blue wavelength onto the apple, they will see it as grey, not as red.

To extend the point to responsibility and desert, we can certainly make true and false claims about our sensibilities with respect to these concepts, and about the sensibilities of the typical person. Non-realists have no beef with that project.

One of the advantages of a non-realist approach is that it allows us to acknowledge and accept our conflicting experiences of the world, without needing to resolve the "conflict" in favor of one experience or the other. The skin of an apple can be seen as having the color quale red, or the color quale grey, depending on the lighting characteristics of the environment in which it is examined. A color non-realist sees no problem with this result, no need to answer the question: "but what color is the apple, *really*, in reality?" Similarly, an individual's bad action can provoke a reaction of blame, and a strong sense that punishment is warranted in retribution, or not provoke those states of mind, depending on the way in which the action is looked at, on the "light" that is shown upon it. To the moral non-realist, that is not a problem. In fact, it's a result to be expected, given the manner in which the human mind was put together: through a clumsy evolutionary process, whose goal was to maximize the organism's chances of successfully making copies of itself, and certainly not to help the organism uncover deep "metaphysical" truths.

To illustrate the conflict in practice, consider the example of Dzhokhar Tsarnaev. If I think vividly about his victims, about the lifelong suffering and despair that he inflicted on them; if I consider the fact that he *wanted* to inflict that suffering and despair, that he *wilfully* acted to make it happen; if I remind myself that he would have wanted, and would have inflicted, the same suffering and despair on me and my family, were I to have crossed his path; if I envision the external cues of *intentional* evil and wickedness in him--his facial expressions, his gestures, his careful movements, as he sadistically plans the act; if I consider these facts, and then consider the punishment that he has been sentenced to serve out, the punishment sits perfectly well. The idea that he will now have to live the rest of his life in a prison, under conditions that will make it absolutely not worth living, feels "correct", as if justice has been served.

Granted, there's a bit of "OK, now don't go too far, don't do something to him that would be upsetting to watch. Certainly, don't torture him. But certainly, make him suffer for the rest of his life--physically and psychologically. Make his life not worth living, since he has taken away the ability of others to live their lives."

If, however, I force myself to step into his shoes, these reactions shift. Like all rational actions, his actions were chosen on the basis of a set of beliefs and desires. If I had emerged in his circumstances, and had felt the internal force of those beliefs and desires, the sense of direction and resolve that they gave rise to, and had been left to choose in his place, can I confidently say that I would *not* have made the same choice that he made? Hell no! Can you confidently say it?

The truth, of course, is that if I had experienced and embodied the exact psychological and physical contingencies that guided and motivated his behavior, I would have done exactly what he did. When I reinforce this fact in my mind, the sense of blame that I previously felt towards him softens dramatically. That blame shifts onto the psychological and physical contingencies themselves--the confused beliefs and hateful desires, as well as the physical brain states that they mysteriously map to, that are at the root of what happened. If they would be capable of driving me--or anyone like me--into that kind of violence, then *they* are the problem, not "him." He is just the unlucky soul that came to manifest them. Why him, and not us, we will never be able to say.

The assumption behind this exercise may give rise to "metaphysical" quibbles, but clearly, there is nothing incoherent or incorrect in imagining yourself with a set of beliefs and desires whose contents match that of another person, and in coming to an internal appreciation of the fact that, in the presence of those contingencies, you would have chosen the same things that that person chose, the same things that you now blame him for choosing. Appreciating this fact gives rise to a sense that in mysteriously emerging into this world as the people that we are, and not as him, that we are the undeserving recipients of good fortune--the "white privilege" of moral responsibility. We are made aware of the fact that reality did not put us to his test--a test that, given the way that contingencies naturally unfold in this universe, we would have failed too, just as he did. And so we feel humility, relief, and empathy--not blame. The prior reactivity softens dramatically, even though it may not fully disappear.

For as long as he continues to manifest those beliefs and desires, and to defiantly endorse what they motivated him to do, it will be difficult for us not to want to him to experience a life that is awful. But what if we could somehow give him a futuristic pill that changed his brain in a way that caused them to completely dissolve away. What if, in taking the pill, he were to emerge as a person with a kind, compassionate, empathetic disposition, who would never in a million years do what he did, and who, when he looks back at what happened, feels *intense* guilt and shame. Would we retain a visceral desire to make this new version of him suffer in retribution for the tragedies? Would it still feel right and just to see that retribution exacted?

If I want to, I can take this type of thinking all the way to complete exculpation, to a complete drying up and falling away of the reactivity that I experienced when I primed the case in the earlier manner. I acknowledge that others may not react in the same way, that they may not find the thinking to be intuitively compelling. That's fine. My reactions in this arena doesn't have to be "right" or "wrong", and neither do theirs.

Philosophers pose the question: which perception is *right*, that is, which perception accurately represents the agent's *true* responsibility and desert? Is it the perception that arose in response to the first priming, or the perception that arose in response to the second? Asking this question, in this way, is like asking the question: which mental image of the apple--the image in which it appears to be red, or the image in which it appears to be grey--accurately represents its true color, the color that is actually there on its skin?

The obvious answer to both questions, in my view, is: none of the above.

Hi Katrina,

Hey, my apologies for hijacking your thread with the anti-realism stuff. To bring it back, I wanted to let you know that I took a look at the 2013 Vincent paper that you referenced, where she used the maxim, "ought implies can", to attack capacitarian counterfactualism. Excellent paper. Thanks for sharing it.

I wanted to pass on some admittedly unconventional ideas that I had on it. Her argument helped put the finger on an uneasiness that I have with the very concept of exacting retribution on regretful and repentant offenders.

To preface, on Veterans day, I clicked on someone's tweet, and ended up watching depressing interviews of incarcerated former veterans serving life sentences at Angola prison in Louisiana. You've probably heard of the prison, but in case you haven't, it's a huge plantation on the Mississippi, north of Baton Rouge, where prisoners serve out traditional sentences of "hard labor"--picking cotton, okra, tomatoes, and so on. If they refuse, they go to what is called the "dungeon" (solitary), where they spend 23 hours on extended lockdown, and are fed nutra-loaf instead of normal prison food.

Louisiana has some really backwards laws, so most of the inmates at Angola are on life sentences without parole, many for first time felony offenses. I was listening to the prisoners tell the story of how they ended up at Angola (one salient story: an alcohol and cocaine addiction, leading to an intoxicated argument one night, with the result being that the offender killed a good friend--2nd degree murder, LWOP). With defeat and sorrow in their eyes, they described what their lives had been like during the "long, miserable years" of their incarceration, to include the daily regimen, and the experience of losing contact with the world, in particular with their families, most of whom have "moved on" from the ordeal, and rarely visit.

Thinking specifically about the prisoner who killed his friend, which was the case that bothered me the most, I thought to myself: we would be *abhored* by the idea of inflicting this punishment on the man on the basis of some factor that he can't control--the way he appears, his accent, the boogar visible in his nose, that he needs to pick off. Indeed, we inflicted the his punishment on the direct ancestors of the prisoners in that prison hundreds of years ago, purely because of their skin colors, and we rightly consider that to be our greatest national shame. But we quickly accept the notion that it's OK to inflict the punishment on him on the basis of a single choice that he made, in a blurred instant of time, many decades ago.

If he had the ability to go back in time and take that choice back, but refused to do so, then it might make sense to treat him as he is being treated. We could rightly say to him, "Look, you have the ability to undo the great harm that you caused. If you did, your ordeal would end. We would immediately release you, to live the rest of your life as a full-citizen, with full freedom and dignity. But you continue to refuse. How, then, can you fault us for continuing to impose these harms on you?"

But, of course, we can't say that to him. It's not true.

When you look in his eyes and listen to him tell his story, you can see that he deeply regrets his choice, that he would give anything for the opportunity to take it back, and have a life back. And the regret doesn't seem to be purely selfish--it appears to be motivated out of deep guilt and shame at having caused the death of a friend.

We live in a universe that gives a person no more control over the facts of his distant past than over the facts of his appearance. In punishing a person on the basis of his distant past, then, we are punishing him on the basis of something that *might as well be like* his appearance, in terms of the things that he can do to change it.

To bring in the maxim, "ought implies can", what we are saying is that he ought to "not have done" what he did. We're citing his failure to comply with this edict as a justification for rendering the rest of his life not worth living. But the fact is that in this moment, he *can't* comply with the edict. He can't "not have done" what he did. The fact that he did what he did is a sealed fact of history, utterly irreversible.

The diachronics will correctly point out that there was a time in his life when he could have acted so as to have things not turn out the way they did, even though he can't take those actions back now. But so what? Right now, in the world that he must live in, he can't take those actions back. He can't make things to have gone other than the tragic way that they went. Why does it matter that in some distant period of his life, a period that has long since disappeared, he could have?

To use Vincent's line of reasoning against her, I would ask: why is grounding his responsibility in the fact that there was once a time in which he could have made a different choice *any different* from grounding his responsibility in the fact that there is a different possible world in which he could have made a different choice, which is what she criticizes the capacitarian counterfactualists for trying to do?

It's as if we are to believe that Vincent's criticism of responsibility is valid when applied across *space*, but not across *time.* It's wrong to ground a person's desert in what he could have done in some other world, but then it's somehow not wrong to ground his desert in what he could have done in some other era in his life, an era that was long ago sealed into place by the irreversible passage of time.

Thinking about things from this perspective, I find myself uncomfortable with the idea of severely punishing people *even* after they exhibit *full* rehabilitation and *full* repentance. Such punishment certainly isn't reasonable, if one use the "the maxim" as the basis for reasonableness.

To come back to the issue of the diachronic self, I doubt that many philosophers would be impressed with this gimmicky self-help move, but here is what I would say. You are not responsible for your past because you cannot change it. You cannot author any aspect of it--as a book, it is sealed, closed, for the rest of eternity. However, you *can* be responsible for your present and for your future, and you *can* be deserving of retribution for what you are doing now, and will do in the future, because you *can* change those things, right now. You can influence the direction that they will take.

In this context, the term responsibility would become the literal "response-ability." The ability to respond, by doing what ought to be done. The maximum "ought implies can" is preserved because what we are saying *ought* to be done, and what we are threatening punishment for if not done, *can* be done. Right now.

As you mentioned, for Bert, "response-ability" might mean taking a class that will teach him how to be a better parent. Unless and until he complies, we will punish him. And the punishment will be entirely reasonable, because he is response-able for the choice he is making right now, to take it or not take it.

I find the rhetoric behind this approach to be far more satisfying than the diachronic rhetoric that beats up on him because of the things that his past self failed to do, when he has no ability to go back "into" his past self and do those things. He is not "response-able" for his past.

Now, in actual practice, we have to attach negative consequences to recklessness and wrongdoing, for the obvious consequentialist reasons. If we didn't attach negative consequences to it, society would not be able to function properly. The traditional legal concept of responsibility seems to be an appropriate marker to use in deciding who will preferentially bear the unpleasant realities of a consequentialist policy--deciding, for instance, who the "example" will be that we will "make an example" of, in the service of deterrance. And so I take no issue with the conventional concept of punishment as a practice.

However, it seems to me that if we were to drop the saber-rattling, the moralizing, the love-affair with the idea of retribution as an intrinsically good thing, we would emerge with something substantially more uplifting and inspiring. A punishment that is exacted not out of hate--not out of a desire to harm someone as payback--but out of love for *all* of humanity's stakeholders, to include those that have acted negligently and immorally.

When you punish your kids, you punish out of love, not out of hate. You inflict the pain--whether incarcerative (grounding) or corporal (spanking/whipping)--to shape their behaviors so that they don't end up doing things that would be bad for them and bad for others. You make it clear to them, as you inflict this pain, that you don't enjoy it, that if you could get them to behave without punishing them, you would. But since that's not an option, you reluctantly do what you have to do.

Applied to criminality, a similar collective gesture would be, "We don't want to inflict harm on you in this way, but we need a system that attaches negative consequences to the kind of thing that you did. Otherwise, large numbers of people would do what you did, with no fear of reprisal. The order of society would quickly break down. And there is also a victim to remember, who you harmed, that is suffering, and that understandably wants you to feel the same suffering. That victim will suffer even more if we let you off the hook, without imposing consequences on you."

"If we could meet our responsibilities--reforming you, deterring others from engaging in the behavior that you engaged in, and bringing closure to the victim who that behavior harmed--without harming you, we *happily* would. Unfortunately, we can't. So we reluctantly do what must be done, not with a sense of satisfaction and gratification, but with a sense of sadness and concern, not only for everyone else involved, but for you."

Needless to say, punishment in the U.S. prison system does not come from anywhere even remotely close to such an enlightened place. It comes from a place of *hate.* The hate is continually communicated to the prisoner through the harsh, subjugating, humiliating, and spitefully vindictive mechanisms of the institution. One can see it by taking the time to read through any state or federal prisoner's handbook, the hundreds of pages of pointlessly restrictive rules, promulgated with arrogance, condescension, and a tone indicative of preparation and willingness to retaliate if transgressed, as if issued from a king to a peasant.

Angra and Izzy, thanks for helping me better understand Jesse's realism problem, and offering solutions. I'm going to refrain from trying to articulate a strong argument for realism here, mostly because I don't know the literature very well and don't have a strong argument for realism ready-to-hand. But this conversation has peaked my interest, so thank you, Jesse, for that.

Jesse, on *ought implies can*. So glad you looked at Nicole's paper. I think you are asking why it is fair to hold someone responsible for, in Nicole's phrasing, an action that issued from one of their cognitive mechanisms in the past, when she says it seems unfair to hold someone responsible for what their mechanism did in a close possible world. The present agent can’t change or impact the outcome of either mechanisms.

A quick answer is that it is fair to hold agents for past decisions and acts because your past you is *you* in a way that your close possible world you isn’t *you.* You sound like you might be operating with something close to a bundle theory of identity, which would indeed mean you might think your past you isn’t really *you* because there is no *you* at all to pass through time. I think something closer to psychological continuity is correct re: personal identity, so barring amnesia, advanced dementia, etc. that would represent a real break in continuity, I think it is ok to hold agents responsible for their past actions.

Come to think of it, ANY punishment is going to happen in a later time slice than the action for which one is being punished, right? And even forward-looking punishment is a response to a past act, for which the punisher feels the punished may be held responsible. All punishments are backward looking in this thin sense. So it sounds like your position may undermine all punishment, although it might support some minority report-type interventions and quarantines.

In a forthcoming paper, Nicole gets a bit more specific about what she thinks her argument in that earlier 2013 paper really means. (I’m sure she’ll send you the paper if you ask.) She says that to decide if a given person ought to be blamed, we should: (1) ascertain from what sort of mechanism did their action issue, what mental capacities did they have; and (2) ask if it would be a fair system that held responsible this kind of person in this way for that transgression. Thus it seems she may be trying to steer clear of *could have done* capacitarianism completely. Maybe you would approve of this move?

I worry that when we ask if it is fair to hold this person responsible for an act or decision we’re just sneaking the capacitarianism in through the back door. Obviously, my view is that capacitarianism is ok, and we should focus on top-down, executive function-based, diachronic capacities to self-form. This is where the buck stops, and because it gets us out of an infinite regress or question-begging loop trying to find the responsible actor, we can attribute responsibility for this action, attitude, or lapse that caused that harm (barring some sort of excusing condition).

Also, retribution doesn’t have to mean townsfolk with pitchforks. Some of it is based in hate and racism, sure. But when I punish my kids I’m not just trying to shape their future behavior. My 9 year old sometimes does things for which she deserves punishment, even if that punishment won’t have positive future positive effects. If she hurts her brother, she deserves to suffer a bit of unpleasantness by sitting in her room while he plays outside. I deliver this punishment with love (masked by a frown). Certainly, though, retributivism can go awry, and our current system needs a much greater focus on rehabilitation, mercy and forgiveness.

"But when I punish my kids I’m not just trying to shape their future behavior. My 9 year old sometimes does things for which she deserves punishment, even if that punishment won’t have positive future effects."

Hi Katrina,

I'm wondering if there's a further basis for such desert apart from expressing in action (punishment) the retributive emotions you feel, however tempered by love, toward your kids. It certainly feels right to punish retributively, without regard to consequences, but is that intuitive rightness grounded in anything more than the satisfaction of the retributive sentiments, what I see as Tamler's Stawsonian view of respecting reactivity? Given that, on a compatibilist view, offenders couldn’t have done otherwise in their actual (as opposed to counterfactual) opportunities for diachronic self-formation, why do they deserve punishment that does them no good?

From a consequentialist standpoint, retributive emotions can point us in the right general direction since punishment is sometimes necessary to shape behavior going forward. But they are a very crude tool, easily incited and amplified so that punishment sometimes vastly overshoots the consequentialist mark, leaving offenders in much worse shape, as you mention in The Diachronic Self, Incarcerated. So although retribution is often cited as a side constraint to limit consequentialist excess, pointing out the consequentially unnecessary suffering and degradation inflicted by overly harsh punishment can put the brakes on retribution.

Once we see that retributive emotions were installed in us for their natural forward-looking, behavior-guiding function, e.g., to keep cheaters in check and deter aggressors, it isn't clear to me why we any longer need to keep retribution – the infliction of deserved, consequence-independent suffering - as a distinct goal of punishment, one that's often used to justify the worst sorts of treatment imaginable.

Hi Tom,

I think retributivism is a useful constraint on forward-looking considerations, just as forward-looking considerations are an important constraint on retributivism. I agree that retributive sentiments are a fairly crude tool, and that questions of proportionality based upon retribution are often difficult. But as you indicate, purely forward-looking considerations can result in unjust punishment. It sounds like you agree retributive sentiment can serve this limiting purpose, is that right? To my mind, forward-looking considerations just can't allow for a very sophisticated scheme of punishments proportional both to the type of offender and the severity of offense.

To be honest, I'm also worried about all the studies showing that differences in severity of punishment do not have deterrent effects. Homicides are not lower in states that have capital punishment or automatic LWOP sentences, for example. So the idea that a lighter penalty is justified because that is all the penalty that is needed for a deterrent effect just doesn't pan out in practice. Philosophers often abstract away from these considerations, but I'm not sure even in a world of perfect knowledge about the law, and very high certainty of arrest, threat of punishment has such a nuanced effect on potential offenders' behavior, partly because of the way in which law is internalized.

Whether the sort of hybrid theory of justification I am thinking of uses retribution as a "stand alone" justification, I'm not sure. I guess for me neither retribution nor deterrence stand alone.

Hi Katrina,

Although I said retribution is often *cited* as a necessary constraint on consequentialism, I myself don't think it's needed, since proportionality can come from what sort and level of punishment, if any, is needed to reform, deter, incapacitate, etc. Don't inflict more pain and deprivation than necessary is, or should be, the general rule. Given basic humanitarian values of minimizing suffering and maximizing autonomy, why would the consequentialist want or need to overdo punishment? So I guess I don't agree that "forward-looking considerations just can't allow for a very sophisticated scheme of punishments proportional both to the type of offender and the severity of offense."

Punishment sometimes isn't a very effective deterrent, especially for those with little to lose or poor impulse control. So, as you point out, increasing its severity frequently doesn't result in increased deterrence, but of course it does model increasingly punitive behavior, which then feeds back into the culture as a criminogenic factor. I'm not sure how these considerations help make the case for retribution, if that's what you were driving at.

I'm wondering, since you're a retributivist (at least in part, I think), how you'd respond to my query above which I'll post again:

It certainly *feels* right to punish retributively, without regard to consequences, but is that intuitive rightness grounded in anything more than the satisfaction of the retributive sentiments, what I see as Tamler's Stawsonian view of respecting reactivity? Given that, on a compatibilist view, offenders couldn’t have done otherwise in their actual (as opposed to counterfactual) opportunities for diachronic self-formation, why do they deserve punishment that does them no good?

Thanks!

Ok, Tom, let me try again, and I apologize for trying to take the easy way out in my last response. I think you make it sound like a purely forward-looking account of punishment is going to be sort of easy-peasy, or at least won't tend to be overly severe, and I just don't agree. If a person who has poor impulse control flies off the handle and hurts someone badly (maybe paralyzes their victim), how much punishment is going to be necessary to reform, deter, etc on a purely forward-looking account? Given our offender has poor impulse control, we might think no amount of punishment will deter him, and certainly prison is unlikely to reform him. So no prison time might be apt. But no prison time seems a bit lenient in this case, right? Or we might think that due to his poor impulse control, he needs to incapacitated. But for how long? His impulse control is unlikely to get better in prison. Maybe he'll need to be in prison forever if we are trying to stop him from hurting someone else. But life in prison sounds a bit overly harsh. And, we might think a person with poor impulse control will only be swayed by a very severe threat of punishment. From this perspective deterrence requires a very serious punishment, again, maybe life in prison. Then again, maybe we should just incarcerate him until he is rehabilitated. But wait a minute, it hardly seems fair that two guys who commit the exact same crime may serve very different amounts of time in prison based upon their ability to be rehabilitated (say, 2 weeks for one very receptive guy, versus 10 years for the really hardheaded one).

I'm not saying we shouldn't consider general and specific deterrence, incapacitation, and rehabilitation. But (1) I think they provide almost too many considerations about what sorts of punishment are correct in many cases; (2) they can lead to unjust results; and (3) we thus ought to consider them within broad boundaries provided by retributive considerations. In my view a person who writes fraudulent checks or is caught with a small amount of illegal drugs should basically never serve a prison sentence, and a person who intentionally and wrongfully kills another person should probably get a prison sentence. This is the idea of limiting retributivism.

Further, I think agents who have had normal diachronic capacities of self-formation deserve retributive punishment because of those diachronic capacities. They couldn't have done otherwise in a simple synchronic sense, but they do have important powers of self-interventions that shape their selves and actions.

I can't speak specifically to Tamler's Strawsonian view, because I don't know it very well. Sorry about that.

Thanks Katrina. You raise good, tough questions for consequentialists (too many considerations, oh no!), but I don't think retributivism helps to answer them. You suggest that quandaries about fairness can be solved, at least in part, by considering what someone deserves (retributive considerations). But what justifies intuitions about desert? It's more or less assumed that the severity of punishment should track the strength of retributive sentiments, which usually track the harm of the offense (possessing illegal drugs vs. intentional murder), but as I suggested (and as I think you agree), those sentiments themselves are keyed to the natural consequential calculus of what punishment it might take to deter cheaters and aggressors, keep society safe, reinforce moral norms, etc. So why honor those sentiments independently of their forward-looking function when to do so imports a specifically non-consequentialist rationale for punishment that's so often used to justify harsh treatment, something you're clearly not in favor of?

It seems to me that the humanitarian constraints I mentioned - minimizing suffering and maximizing autonomy - are sufficient guides to fair, just and proportionate (that is, minimally necessary) punishment. So yes: two psychologically different offenders who committed the same crime might need different interventions to achieve the same behavior-guiding or public safety objectives. If punishment can't serve any forward-looking purpose, as you suggest in some of your examples, must we punish anyway, and if so, why?

This gets to *your* view on retribution (forget Strawson). What is it about having diachronic capacities for self-formation that makes someone deserving of retributive punishment, punishment that does them or society no future good? Seems to me that such capacities - powers of self-intervention to shape their selves and actions - are exactly those which could justify consequentialist, behavior-guiding punishment, since such capacities make offenders responsive to the prospect of future sanctions. I don't get the inference from diachronic capacities to retribution, so perhaps you could clarify that for me. Unless retribution itself can be justified, it can't serve as a constraint on consequentialism.


Hi Katrina, Jesse, and Tom

I really enjoyed reading through Katrina's post and through the discussion. Since Katrina refers to my work in her post, I thought it might be useful to explain what I see as the attraction of a diachronic approach to capacity.


It seems to me that what makes people whom we might call "capacitous" different from those whom we might call "sub-capacitous" is *how often* the capacitous will recognize and react to reasons in an appropriate manner. Roughly, capacitous people will recognize and react to reasons in an appropriate manner *more often* than sub-capacitous people.

And the reason why this matters, in my view, is because if capacitous and sub-capacitous people were treated the same – that is, if a legal regime did not take this difference into account by cutting sub-capacitous people more slack (perhaps this might gain expression by expecting less of the sub-capacitous, or by treating them less harshly than the capacitous) – then the rules of that legal regime would be such that sub-capacitous people would end up being treated more harshly than capacitous people. Because sub-capacitous people recognize and react to reasons in the appropriate manner less often than capacitous people, they (sub-capacitous people) would get into trouble with the law more often than capacitous people, and so a legal regime which did not recognize this difference would treat sub-capacitous and capacitous people in an unequal manner.

To treat people equally, legal regimes must recognize relevant differences in people's psychology.


I don't know if this is helpful – I hope it is – but rather than writing more TLDR here, I'd rather leave it at that for now and see what questions come up.

Ciao

n

Hi Katrina:

"Jesse, on *ought implies can*. So glad you looked at Nicole's paper. I think you are asking why it is fair to hold someone responsible for, in Nicole's phrasing, an action that issued from one of their cognitive mechanisms in the past, when she says it seems unfair to hold someone responsible for what their mechanism did in a close possible world. The present agent can’t change or impact the outcome of either mechanisms."

Yes, exactly.

"A quick answer is that it is fair to hold agents for past decisions and acts because your past you is *you* in a way that your close possible world you isn’t *you.*"

Makes sense. Very well said.

I acknowledge that what I am offering here is sophistry ;-) But the intention is to express a genuine emotional difficulty that I have with the man's concrete predicament, and the predicament of every person like him. Like all of us, he wants to live a worthwhile life, but the state will only allow him to live a miserable one--a life spent picking cotton day after day in the sweltering heat of a Louisiana prison plantation, earning 5 cents/hour. The state's justification for imposing this useless predicament on him, despite his full rehabilitation, is that he violated a prior ought. But he can't unviolate that ought. He's stuck, and yet the state is beating him for not moving.

This dynamic gets to a larger unreasonableness inherent in life more generally, an unreasonableness that any system that selects by consequence--whether at the level of the species, or at the level of the individual mind itself--will produce. The way he learned that cocaine was something to absolutely, unconditionally avoid was by making the mistake of using it, becoming hopelessly addicted, committing a tragic crime from a state of intoxication, and ending up with an LWOP sentence at Angola. But as soon as that lesson--the most important lesson he will ever learn--was learned, it ceased to be useful to him, because the mistake, with its tragic consequence, was irrevocable.

A teacher that teaches the lesson after the test, rather than before it, is not a reasonable teacher. Unfortunately, in this life, we are dealing with a course that has an incredibly unreasonable teacher. We need to cut the students some slack.

"Come to think of it, ANY punishment is going to happen in a later time slice than the action for which one is being punished, right?"

Usually, but not necessarily. Suppose Bert is ordered to attend a parenting class. The action of not taking it will be punished with arm-twisting. The arm-twisting will begin a few seconds after he has left the room, and will stop the *instant* he chooses to go back in. The punishment will then be perfectly contemporaneous with the omission: "not attending the class." Unlike the rehabilitated prisoner at Angola, Bert can't protest the twisting of his arm on the grounds of not being able to do anything about it. We are right when we say to him: "Look, if you want the arm-twisting to stop, stop the transgression. Go back in the room!" This is reasonable in a way that inflicting a life of unconditional misery on someone for something that happened several decades ago is not.

"So it sounds like your position may undermine all punishment."

All non-consequentialist punishment. I'm OK with that ;-)

To clarify a question I have about your position, suppose that there are two punishments, Punishment A and Punishment B. Punishment A and B are identical in the following relevant respects: (1) rehabilitation, (2) incapacitation, (3) deterrance. But there is a critical difference in: (4) retribution. The difference is that punishment A does not actually inflict any pain or harm. To the observers--the rest of society and the victim--Punishment A *appears* to inflict pain and harm, therefore it deters the rest of society, and provides the victim with maximum retributive delight. But the truth is that the punishment is a sham. The actual pain and harm are not being inflicted.

Punishment B, of course, is the same as Punishment A, but it is not a sham. Its pain and harm that is entirely real.

If you had to make the choice, which punishment would you want for Bert? Which punishment would you want for the man who committed murder-2 and is incarcerated at LSP?

What I'm trying to get at here is, do you want the pain and harm of retribution for its own sake, because it is deserved, or do you want it because you want the *victim* to experience the satisfaction and closure that it will bring?

If the latter, I can stomach that--pain and harm inflicted in the service of love for the victim, his or her well-being. But if the former, that's more difficult to stomach--especially wrt Bert. On my view, you would be punishing him in the service of something empty and invisible--sacrificing a baby to a God that does not exist.

"In a forthcoming paper, Nicole gets a bit more specific about what she thinks her argument in that earlier 2013 paper really means. (I’m sure she’ll send you the paper if you ask.)"

Would love to read it. Email: philosophicalecon at gmail. Twitter: @jesse_livermore

"Also, retribution doesn’t have to mean townsfolk with pitchforks. Some of it is based in hate and racism, sure...I deliver this punishment with love (masked by a frown). Certainly, though, retributivism can go awry, and our current system needs a much greater focus on rehabilitation, mercy and forgiveness."

I find the compassionate strain in your work to be refreshing. But, philosophically, I'm not sure what justifies it ;-) If people can deserve to suffer pain and harm for their actions, why does the level of pain and harm have to stop at some arbitrary point? There is no limit to the evil that a human can produce under the right set of contingencies, so why is there a limit to the pain and harm that a human can deserve in retribution? Why can't sufficiently evil people deserve to be kept alive and tortured for their transgressions?

That said, I acknowledge that taking an officially retributivist stance, but then emphasizing rehabilitation, mercy, and forgiveness in the application to the broken U.S. system, does give you credibility with the folk that you would not have if you were a bleeding-heart that wanted punishment to always be as comfortable as possible.

"But when I punish my kids I’m not just trying to shape their future behavior. My 9 year old sometimes does things for which she deserves punishment, even if that punishment won’t have positive future positive effects. If she hurts her brother, she deserves to suffer a bit of unpleasantness by sitting in her room while he plays outside."

In reading your work, there seems to be a subtle distinction between *pain* and *harm.* You are OK, for example, with your daughter experiencing temporary pain--temporary dissatisfaction--in retribution for being mean to her brother. Pain is a part of life, a part of being human. Deal with it, and if you don't want to deal with it, don't be mean. But obviously you would not be OK with a punishment that harmed her.

Similarly, you are OK with punishments that intentionally impose the pains of incarceration on criminals. But you are not OK with punishments, such as chemical castration, that harm them--particularly in their agential capacities.

Is that an accurate perception?

If so, there is a proposal that you might want to explore. I will raise it in your other thread.

Hi Angra, apologies for missing your earlier comment:

"Your inverted color spectrum scenario wouldn't seem to result in any apparent disagreement. It seems those aliens wouldn't know that there is a difference in perception, and neither would the humans talking to them, since the aliens would have a word – say, “ared” – that would be applied to the same objects the English word “green” applies, so they would probably think the meaning is the same."

Their perception of the apple's color would be different. Now, they might end up using language in a way that causes them to not be *aware* of the difference, but the difference would still exist.

*If* they could accurately communicate with each other, they would realize that their words refer to different quales. The problem would then emerge: when the human makes the claim, "the skin of the apple *is* ", and species #2 makes the [properly translated from alien language into english] claim, "the skin of the apple *is* ", who is right?

If they adjust their statements like this:

Human: "I see the apple as ."

Species #2: "I see the apple as ."

Then, they will *both* be making factually true statements. But if the human is a color realist, she will resist this adjustment. It does not adequately convey what she wants to say, which is that the quale is not just a property of her experience, but a property that also exists on the skin of the apple itself.

Hi Angra, apologies for missing your earlier comment. Snippet to you in the previous post deleted some color words, because I put brackets around them. I show them below in all caps:

"Your inverted color spectrum scenario wouldn't seem to result in any apparent disagreement. It seems those aliens wouldn't know that there is a difference in perception, and neither would the humans talking to them, since the aliens would have a word – say, “ared” – that would be applied to the same objects the English word “green” applies, so they would probably think the meaning is the same."

Their perception of the apple's color would be different. Now, they might end up using language in a way that causes them to not be *aware* of the difference, but the difference would still exist.

*If* they could accurately communicate with each other, they would realize that their words refer to different quales. The problem would then emerge: when the human makes the claim, "the skin of the apple *is* RED", and species #2 makes the [properly translated from alien language into english] claim, "the skin of the apple *is* GREEN", who is right?

If they adjust their statements like this:

Human: "I see the apple as RED."

Species #2: "I see the apple as GREEN."

Then, they will *both* be making factually true statements. But if the human is a color realist, she will resist this adjustment. It does not adequately convey what she wants to say, which is that the quale RED is not just a property of her experience, what she sees, but a property that also exists on the skin of the apple itself.

The comments to this entry are closed.

Books about Agency


3QD Prize 2014: Marcus Arvan