We are pleased to present the latest installment of Ethics at PEA Soup, in which we host a discussion of one article from each issue of Ethics. The article selected from Volume 122, issue 4, is Sarah Buss's "Autonomous Action: Self-Determination in the Passive Mode." Ethics has kindly provided free access to the article here. We are also extremely grateful to Hilary Bok, who has agreed to provide the critical précis, which begins below the fold.
Sarah Buss’ ‘Autonomous Action: Self-Determination in the Passive Mode’ is a wonderful paper: rich, subtle, and thought-provoking. Buss’ goal is to elucidate the type of autonomy needed for moral responsibility. This sort of autonomy requires more than mere voluntary or intentional action: we can act intentionally even if we are in the grip of an apparently irresistible compulsion. But it is much less demanding than perfect or ideal autonomy.
Buss first argues that we should not conceive of this sort of autonomy as involving a sort of super-agency, in which the agent does something special that secures her autonomy, like deliberating, choosing, or endorsing her action. This seems right: on the one hand, as Buss notes, we are responsible for our thoughtless actions as well as for those that we have chosen; on the other, our choices themselves might not be autonomous.
The key to this sort of autonomy, Buss argues, lies in our passive contributions to our choices. Buss notes that our mental, emotional, and physical states can affect our actions in two ways: first, as things we think about when deciding what to do, and second, as things that color what facts we notice, what we make of them, and which we take to be reasons for action. With respect to influences of the second type, Buss writes, “agents are necessarily passive.” (658) And our autonomy depends on our relation to them: “whether someone acts autonomously depends on whether she can be identified with the direct, purely causal, nonrational influences on the formation of her intentions.” (658)
We are humans, members of a species. Buss argues that for this reason, any condition that is incompatible with minimal human flourishing, like a serious illness or disability, is “external to a human agent’s identity.” When our actions are caused by such a condition, they are caused by something that is, in an important sense, not an expression of us. “Sickness is a hostile takeover. It is an attack on a person by something external to the human being she truly is.” (668)
We are not, in this sense, less autonomous when the fact that we have a broken leg leads us to decide not to go running. In this case, our broken leg figures among our reasons for acting, and we decide what to make of it. But some illnesses do not simply present us with difficulties to be adjusted to; they affect which facts we notice, and what we take to constitute reasons for action. For instance, anorexia does not keep its victims from eating by making it difficult for them to swallow; it makes them think that they should starve themselves. When some illness affects us in this way, the actions that result are in an important sense not fully our own.
I am generally in sympathy with Buss’ arguments against the idea that the kind of autonomy she’s talking about requires some special exercise of agency. Most of the time, we act without deliberating or making conscious choices. This is as it should be: the fact that I do not have to deliberate about whether or not to show up for class every day, still less about every turn I make when driving to work, is a blessing, and one that allows me to think about more interesting things. Sometimes my thoughtless actions reflect previous choices, but often they do not: when someone strikes up a conversation with me, for instance, my response does reflect my character, and I am responsible for it, but normally it does not (as far as I can tell) reflect deliberation, past or present.
I think that Buss overstates our passivity with respect to what we notice, and which facts we see as reasons for action. We are passive with respect to these things at the moment of choice: as with all traits of character, by the time we arrive at a moment of decision, the time to work on them is past. But we can certainly affect them beforehand. We can train ourselves to notice the effects of our actions on others or to be indifferent to them; to notice where all the interesting people are hanging out at a party, or to look and see whether there is someone who no one seems interested in talking to; to appreciate problems with which we are not personally familiar or to ignore them. If we have not done so, on many accounts we can be held responsible for that fact.
Nonetheless, I think Buss is right to argue that the sources of autonomy are not some special exercise of agency, but some more general state that allows us to assume, absent evidence to the contrary, that someone who is in that state acts autonomously. That said, I have several questions about Buss’ position.
First, a central point in Buss’ argument is that because we are human, any trait that is inconsistent with minimal human flourishing is external to us. I am not sure why our humanity implies that we should identify with traits that are required for minimal human flourishing, as opposed to (for instance) traits that are broadly characteristic of human beings. I find Buss’ view more attractive than this one, but I am not sure how Buss would support it. In particular, the idea that as humans we must value traits that promote human flourishing seems to be vulnerable to some of the arguments Buss directs against the claim that as rational agents we must want to do what we really have reason to do. (666-667)
Second, the claim that we must regard traits that are incompatible with minimal human flourishing as in some sense external to us seems to me false. I assume that any account of autonomy must allow for some way in which I can decide for myself what is central to my identity. Suppose that I decide in this way that my flourishing is not central to me. Perhaps I care more about the flourishing of others, and believe that whenever I could promote their flourishing by sacrificing my own, I should. Suppose further that in the world as I find it, opportunities to promote others’ flourishing by sacrificing my own are frequent, and that the sacrifices I might be called upon to make are significant enough that were I to make them, I would not be capable even of minimal flourishing. I might, for instance, have to sacrifice my life for the sake of others.
In this case, my disposition to sacrifice my own interests might well be incompatible with minimal human flourishing. If, despite this fact, I choose to cultivate this disposition, would the actions that followed from it necessarily be alien? If so, why? In some cases, there might be an answer: I might, for instance, be so deeply depressed that I cannot imagine how I could possibly value my life. But surely it is possible that I might not value my own flourishing as more than a means to some other goal without some story like this being true of me. In such a case, why should I accept as a criterion of alien-ness something that I have decided not to value?
Third, the traits and dispositions that are inconsistent with some minimal degree of human flourishing are not all illnesses. Being unlikable, for instance, prevents you from forming close relationships. Having a terrible temper can get you into all kinds of trouble. Depending on one’s circumstances, being a person of principle who refuses to countenance genuine injustice can be incompatible with life. While each of these traits might result from a mental illness of some sort, each of them could also be traits that an agent is genuinely responsible for. I am not sure how Buss would distinguish them from illnesses; if she does not, her view would have very counterintuitive consequences.
Fourth, illnesses are not always unchosen. This is obvious in the case of addiction, though it is not always clear that people who are beginning to use addictive drugs fully appreciate how bad addiction is, and how likely they are to become addicts. But sometimes, surely, they might understand the dangers they run quite clearly. Besides, there are other cases. Consider a person with a serious mental illness who had worked her way through to some sort of fragile equilibrium, and who, presented with s
omething that seemed likely to cure her illness, or at least to greatly alleviate it, chose not to. One can easily imagine the details of this in such a way that the decision looks like a product of the disease, but I think one can also imagine it not being one. She might think that a hard-won equilibrium should not be tampered with lightly, or she might have come to identify herself as the person who fought her way through to that equilibrium, and not want to change. In that case, it is still a disease, and thus on Buss’ account things one does as a result of it are things one is not responsible for. It is not clear to me that this is right.
Most of my questions center around the tangled relationship between mental illness and choice. I am not satisfied with my own understanding of these issues, and I thank Sarah not only for a really illuminating paper, but for providing the occasion for this discussion of them.
–Hilary Bok
“(W) e determine our actions in the way necessary and sufficient to render us accountable for what we do when and only when our intentions reflect the decisive influence of our nonagential identity as human beings; and this condition is satisfied when and only when our intentions are determined by physiological and psychological states that do not typically prevent human beings from functioning minimally well.”
Professor Buss,
So, God forbid, should I become seriously ill, I would no longer be accountable for my conduct? Worse, I’d no longer be myself, nay, less than fully human? Putting aside questions pertaining to Incompatibilism, it still seems possible to act autonomously in the face of one’s afflictions. Let us suppose that I begin to act boorishly towards my caregivers, offering as my excuse the fact that I’m in considerable pain. I cannot meet my basic needs, so my illness qualifies as debilitating, but I am being provided for by others. My hitherto fulfilling activities are also out of the question, plunging me into despair. Meanwhile, the elderly lady down the hall, hurting just as much if not more than I am, is at least not taking it out on the nurses. A lifetime of patiently, prayerfully, enduring lesser discomforts has prepared her to bear the cross at hand. Her serenity, the fruit of that perseverance, is admired by all those with whom she comes into contact. Do you mean to tell me that she should not be commended and I, well, pitied, but deprecated? I have heard it said that we find out what a person is truly like when things aren’t going his way. Didn’t my illness merely bring out, make manifest, my misanthropic side? Ditto for the ‘good patient’ and her serenity? Catholics consider suffering not only meaningful, but potentially redemptive. There appears to be no room for this notion within your philosophy.
I simply do not share your “asymmetrical” intuitions. Regarding the 3 paradigms, I would say that the agents involved allowed their emotions to get in the way of doing the right thing. People manage all the time to overcome anger, fear, and depression in the pursuit of virtue. They turn the other cheek, ignore their fears, and try to look on the bright side. (Depression is one of the 7 deadly sins.) You make things much too easy. Where is the incentive to develop traits such as forbearance, courage, cheerfulness if excuses of the sort you would accept are readily available? Why shouldn’t we instead hold up as role models those who do manage to act conscientiously in the face of adversity? According to your philosophy, all bets are off as soon as my normally fulfilling activities become too much for me. But, again, why shouldn’t we take how one bears up under severe trials as the true measure of a person? As long as one is in possession of one’s faculties, isn’t one responsible for how one treats others, regardless of one’s physical condition?
What of akrasia? The above conditions are often taken as causes of weakness of the will- which many take to be perfectly consistent with responsibility. Am I to be let off the hook for my bad behavior once illness induced psychological resistance to doing the right thing appears? Or am I not then required to strengthen my resolve? If I know that I will not be held accountable for wrongdoing attendant upon an illness, where is the incentive to develop self-control? On the flip side, what do I have to live for, in terms of self-actualization, if sickness means I am no longer myself?
Your thesis yields another asymmetry, curious at least to my mind. When circumstances conspire against me, I’m still rightly expected to do my best: I’m supposed to soldier on, rise to the occasion, pray for greater strength, etc.. But, according to you, when my body begins to fail me, my responsibility diminishes. We hold people accountable in the former instance, even if the difficulty appears insurmountable, why not latter? Why are less than optimal circumstances “not truly me,” especially those I have come to count on? Where is the boundary here? I don’t think you can say one’s skin, since; again, there are calamities that are at least as detrimental to flourishing as any illness. Wars, natural disasters, economic depressions all make it extremely difficult to go about one’s business. Yet praise and blame don’t seem out of place in such exigencies.
In sum, pain, fear, and suffering, are a part of the human condition. If they become so extreme that I cannot ‘think straight’, then my loss of autonomy is best explained by the Deliberation Account. If not, then since I remain in possession of my faculties, I am still responsible for my conduct, especially how I treat others.
Smaller points:
‘Necessarily, we lack direct control over the direction of our own thinking, no matter which factors influence this direction.’
As I’m sure you know, Duns Scotus maintained that our ability to summon and dismiss reasons at will was the surest mark of free will. I am inclined to agree.
‘This happens, for example, when overpowering fear or pain so “governs” our practical reasoning that we find it impossible to take seriously any alternative to obeying the robber or the rapist or taking pain-numbing drugs. In such cases we may be appropriately responsive to the best reasons we have; yet this responsiveness does not suffice to render us accountable for the effects of our fear or pain or drug-induced fog.’
Frankfurt once told me at a colloquium that he thought that the prudent mugging victim acted perfectly freely.
Thanks so much to Sarah for this great paper and to Hilary Bok for a precis asking several excellent questions.
I have many questions myself. Let me say from the get-go that I’m fully in agreement with the main early point about passivity: one has to move to the passive in order to make sense of (what I call) attributability, of an action’s being truly mine. I also find the connection you want to draw to minimal human flourishing to be very promising and intriguing. I have several questions about specific cases, though. First, a somewhat glib case: is what prevents children from being accountable that they suffer from the debilitating condition known as childhood? Or are they just not at the age at which they have yet formed a nonagential identity for which there can be debilitations pertaining to minimal human flourishing?
Second, and more seriously, I’m quite interested in the case of depression. But I think determinations of the right responsibility assessments of such folks vary, depending on whether we take up the first- or the third-person perspective on them. For many clinically depressed patients (based on what I’ve read, as well as private discussions), what they do in the depressed state feels to them as if it’s very much their own, a reflection of who they are when depressed. And this point may be buttressed by thinking about how, for some, the transition into clinical depression is gradual. (Self-ascriptions may vary, of course.) But they are certainly not immune to guilt; they hold themselves accountable. Where we tend to get suspensions of accountability are in third-person cases, where we (the non-depressed) cut the depressed serious slack. We are doing so, I think, by making a direct time-slice contrast between who the depressed person is now and who they were back when. Thus the “he’s not himself” plea. So here’s a case in which first- and third-person determinations of accountability may conflict. You seem to favor the third-person determination. Why?
Furthermore, there are some accountability predicates that obtain even for the depressed, namely those regarding how they handle being depressed. Some are heroic about it, some are pitiable. Why think these sorts of intentions are different from those for which they aren’t accountable?
Finally, while your account really helps, I think, to understand the “he’s not himself” sort of Strawsonian exempting plea, what of the “he was raised in unfortunate circumstances” plea? In particular, consider a case that Susan Wolf thinks is clearly exempting, that of JoJo, the son of an evil dictator who embraces his dad’s evil values when he is an adult. Wolf thinks our intuitions are that he’s not responsible, but it’s hard to see why not in terms of some human malfunction, or some disability pertaining to his flourishing (he really does flourish in many respects, after all). Other less dramatic cases are those racists who were raised in racist households. IF we are to suspend accountability assessments of these folks, it is hard to see why on your analysis.
Now one possible reply is to deny that these agents aren’t accountable after all. Indeed, there is some strong empirical evidence (that David Faraci and I have gathered) to suggest that people still do find JoJo accountable when he tortures peasants on a whim (or when the racist acts as a racist). But even so, they find him less accountable (less blameworthy, in our language) than the unmitigated dictator (or racist who embraces racism after looking unblinkered at all the options). It is hard to find a reason based on dysfunction or flourishing to explain this response, however. (You touch on the case of sexism briefly at the end of your article but don’t address this sort of worry yet, I think.)
My comment is related to one of Hilary’s (above). Sarah suggests that autonomy is compromised when an intention is determined by a pathological trait that non-rationally influences “what things [someone] takes into account and what significance she attributes to them.” Sarah also emphasizes that there is “no determinate, specifiable point at which a personality trait becomes just extreme enough to qualify as a trait of ‘pathology.’” This raises the question of whether a trait could count as a defect relative to one’s identity as a representative of a certain kind of being without being “extreme enough” to count as pathological (or border-line pathological). If so, then it’s not clear why a trait must be pathological to compromise autonomy; why couldn’t merely defective traits compromise autonomy, since they too fail to match one’s identity as a representative of a certain kind of being? If, alternatively, all defective personality traits count as pathological, and talk of extreme traits is not doing any real work here, then it seems that all defective personality traits with non-rational influence (of the relevant sort) will compromise autonomy when they play a determining role; but my sense is that Sarah talks about extreme traits precisely because she wants to avoid this counter-intuitive (overly-inclusive) conclusion.
Thanks for a terrific paper. I have a quick question about your view’s implications for animal autonomy. You write that an agent’s autonomy depends on “whether she can be identified with the direct, purely causal, nonrational influences on the formation of her intentions.” You then argue that she can thus be identified if these influences aren’t “external” to her, where this externality condition is understood in terms of what promotes minimal human flourishing. But if that’s your view then I’m not sure how you rule out the autonomy, say, of dogs, whose intentions, presumably, are also influenced by various causal, nonrational factors and for whom a flourishing life is also, presumably, possible. I trust that we don’t want to endorse the view that dogs are accountable for their actions so long as their intentions aren’t shaped by causal, nonrational influences that impede doggie flourishing – so long as they aren’t severely depressed, etc. But how do we avoid that result on your view? Is it that dogs lack intentions? Is it that there’s something special about human flourishing?
Very thought-provoking paper! I have a minor doubt about negative states excusing or exculpating actions in a way that (non-dramatic) positive states do not. You mention someone saying, “She was so depressed that she forgot to wish him a happy birthday,” and claim this explanation suggests, “that the agent may not be accountable for what she did”. Insofar as I go along with this tempting view about the case, I am also drawn to thinking the same can be true of a positive case – albeit not the positive cases you mention.
Here is my line of thinking. It is commonly thought that being happy and optimistic makes one prone to overlook or misinterpret features of situations.
Imagine someone saying, for example, “Because she was so happy and optimistic she did not notice he was depressed and missed a chance to help prevent his suicide.” I can see that in a case like that the agent should feel bad about the failure, but I am less confident she can rightly be held responsible for the failure to help.
I will be curious to hear what you think about this case and others in which positive emotions, moods, etc impede an agent’s grasp of the situation or the normative relevant of its features.
I guess my main point is that, if I just think about the cases, I am inclined to say the same thing here as in your case of the depressed guy who forgets a birthday.
Professor Buss, this was a very persuasive paper and I think the connection you draw between human flourishing and autonomy is an important one. However, I’m not completely convinced that your account should be seen as necessarily in competition with other models. Although you do a great job drawing attention to the flaws of other models and the merits of your own approach, I’m inclined to see the other models as capturing just some segment of the concept of autonomy and needing of supplement, rather than replacement. And this goes for yours as well because I’m not sure the human flourishing condition can do all the work you want it to do. I think this comes out best in your response to the mad scientist case beginning on p. 686. You argue that the reason why the mad scientist case should be regarded as an example of non-autonomy is because having one’s stable psychological states exercise influence over one’s desires and behavior is necessary to normal human functioning, thus by blocking this influence one’s autonomy is compromised. However, this doesn’t seem to capture what is driving our intuition in the mad scientist case. What does seem to be driving our intuition is that one’s sovereignty is being interfered with by the mad scientist because one is effectively being used as an instrument for accomplishing the will of another (i.e. whatever goal the mad scientist has in altering his subject’s desires) without one’s consent. And that’s true regardless of the changes the mad scientist makes. In contrast, you allow that the mad scientist’s interference does not compromise autonomy where he “merely remove[s] the impediment of pathological influences.” But I think even in such cases we would still hold that the autonomy of the target of the mad scientist’s efforts has been compromised. Otherwise, I see no reason to distinguish this case from one in which, for instance, a suicidal person is coerced or tricked into taking anti-depressant medication. It seems uncontroversial to hold that such a person does not act autonomously with respect to taking the medication, though we might say that the compromise in autonomy was justified or even that such a person becomes more autonomous once the medication is working. Likewise with the mad scientist’s target: it is sufficient that because the target did not consent to the mad scientist’s experiment, he or she should not be considered autonomous, regardless of whether the experiment hinders his or her ability to achieve human flourishing. This suggests to me that your account is subject to a similar worry as the one involved in rationalist accounts of autonomy – that it gives good meaning to the ‘government’ prong of self-government but too little content to the ‘self’ prong. Perhaps this is where the deliberation/endorsement models can supplement your account? Thank you in advance for any thoughts you may be able to offer on this.
This is a response to Hilary’s comment.
I would like to thank Hilary for her very thoughtful and thought-provoking comments.
I first read through these comments about three hours ago. Here is what they provoked in me:
I unreservedly endorse her claim that “we can train ourselves to notice the effects of our actions on others or to be indifferent to them.” I agree, too, that the failure to make whatever efforts are required to notice the things we ought to notice is itself something for which we can be held responsible.
I am also very sympathetic with the perplexity she expresses in the first of her questions. The issues this question raises seem very important, and I have certainly not given them the attention they deserve. Pondering these issues quickly takes us far beyond worries about the conditions of moral responsibility. What should we say, for example, about a human being who was born with mental capacities that are far more limited than ours, and who, for this reason, has a very different way of being in the world — where this includes a different set of preferences, a different range of emotional responses, etc.? On the one hand, this human being is “retarded” – that is he has been slowed, prevented from realizing his potential. To the extent that this is the right thing to say, this is because his species membership is an important aspect of his identity, and as such it provides a standard for determining whether he is thriving. (A member of any other species with the same mental capacities would not qualify as “slow,” or in any way “handicapped”/ “dis-abled.”) Looked at this way, his mental deficiency is an unfortunate circumstance; anyone who cares about him would “cure” him, if this were possible. But what if this human being doesn’t see things this way himself? He could, of course, be mistaken. But — and here we come to the other hand – I don’t think he need be. In fact, it seems to me that such a human being might reasonably believe that his mental capacities, and everything about him that is closely connected to these capacities, are such an important aspect of his identity that he might as well have died had he been cured of them. Even, moreover, if he were not prepared to go quite this far, he could reasonably believe that the conditions under which he can thrive are determined by the mental capacities he actually has.
As I say, I am not sure how to think about this. I do, however, want to make a few observations. First, it seems more plausible to say that someone is the victim of bad luck if she *loses* certain normal human capacities than if she was born without them. Second, as fuzzy as the line is between conditions that enable someone to live a good human life and conditions that enable her to address her most basic needs, this is an important distinction. Third, it seems possible to describe our basic needs in a way that does not distinguish among species (e.g., they include the need to for some nourishment, the need to avoid pain (physical or psychic) that renders it extremely difficult to participate in the activities that are typical of one’s species, etc.). In any case, fourth, we seem to need this distinction to make sense of the distinction between illnesses/disabilities and other impediments to living a good human life. Fifth, we seem to rely on this further distinction in determining whether it is appropriate to regard someone as accountable for her action.
Since this last point is central to my paper, let me say something about how it relates to my tentative remarks about human beings who lack the mental capacities that distinguish most adult human beings from other animals. So as not to be distracted by the possibility that such human beings are not capable of using their reason in whatever way is essential to self-government, let us imagine a case in which someone who has the requisite rational capacities admits that there is nonetheless something about her that makes it extremely difficult for her to satisfy certain basic needs. (Perhaps, her condition makes it extremely difficult for her to care about the effects her choices will have beyond the next few days. Or perhaps it simply makes it extremely difficult for her to make friends and influence people – so difficult that no one wants to hire her.) Surely, it is possible for such a person to appreciate this fact about herself and yet resist any attempt to “cure” her. Indeed, it could be reasonable for her to believe that what she would lose from such a cure would be greater than anything she could possibly gain. And she could reasonably believe this, even though she concedes that were she to undergo such a profound change, she would almost surely see things differently. She might not know how best to formulate her conviction: Is it the belief that because she is different, she has a distinctive way of thriving? Or is it the belief that there is something more important than thriving? — that she would have to be alienated from herself to want to be cured of what is so central to her way of being in the world, and that she is not alienated from herself in this way? Whatever the proper interpretation of her conviction may be, we can ask: Are there circumstances under which the abnormal condition prevents her from being accountable for what she does? My account does not yield a definite answer to this question. Rather, it implies that we will regard her condition as an impediment to self-government if and only if we believe that — despite her attitude toward this aspect of her psyche, and despite the fact that this attitude is not unreasonable — it is external in the sense that an illness is external.
This is a *diagnosis* of our moral responsibility assessments. In the paper, however, I make a further claim: this diagnosis allows us to see when moral responsibility assessments are *justified*. For this to be right, it has to be reasonable for us to regard a human being’s species identity as an essential part of her identity. I think this is reasonable. But, as I stress in the paper – and as my remarks about the person with less-than-human mental capacities suggest – there are certainly other sorts of assessments in relation to which this aspect of a person’s identity is not what matters.
Indeed, my reflections here suggest that we need to distinguish conditions of accountability from what we might call conditions of “authenticity.” Someone like the person I imagine above is likely to insist that these two conditions do not come apart in her own case; she is likely to insist that she is fully accountable for what she does when she acts under the influence of the condition that makes it so difficult for her to satisfy some of her most basic needs. But precisely because of the way that accountability-undermining influences work – precisely because they are causal influences on our reasoning — it is rare that someone who is subject to such influences is in a position to appreciate that something is preventing her from playing a decisive role in forming her intentions. In any case, it seems to me that the person I imagined could grant that her condition rendered it more difficult for her to be accountable for her actions under certain circumstances. She could grant that her resistance to being “cured” is a resistance to undergoing a change that would make it easier for her to govern herself. What matters to her most, she could explain, is maintaining her integrity: it is more important to her to be “true to herself” in the sense that involves being true to her perspective on the world than it is to be free from causal influences that are external to her identity as a human being. [Note that on my account, insofar as this preference reflects the influence of the very accountability-undermining condition she prefers to preserve, she is not accountable for her refusal to take steps to make it more likely that she will be accountable for what she does. Is this a problematic implication? I don’t think so. But perhaps others will help me to change my mind.]
I take these musings to be a response to all of Hilary’s questions. But let me conclude by saying an additional word or two in response to questions 2 and 3.
I hope it is clear that I endorse her claim that there is a very important (though hard to pin down) sense in which “I can decide for myself what is central to my identity,” and that this means that I can identify with certain aspects of my psychic life that not only render me an abnormal/unusual human being, but also make it difficult for me to thrive, and even to function minimally well. Do I agree that “any account of autonomy” must do justice to this aspect of an agent’s identity? Yes, I do: Because any account of autonomy must be compatible with an adequate account of intentional action, and because I believe that we cannot form the intention to pursue a given end if we think we have overriding reason not to pursue it, any account of autonomy must acknowledge that a person fails to act autonomously if her behavior fails to reflect her own beliefs about what is important and valuable. As my comments about the relationship between authenticity and autonomy suggest, however, I also believe that we can be in the dark about what underlies our evaluative and normative judgments. And this means that we can be in the dark about whether the causal influences on these judgments are impediments to self-government. And this means that we can be in the dark about whether we are accountable for the intentions that reflect these judgments. Again, this does not mean that it is possible for us to initiate an action without taking responsibility for what we are about to do. It simply means that *taking* responsibility for your actions does not suffice for *being* responsible in the sense that makes it appropriate for someone to blame you if what you did was wrong. It means that we can be mistaken in thinking that we are not operating under exculpating conditions. So, at any rate, it seems to me. In taking this position, I do not mean to deny that someone can have very good reason for making choices as a result of which she becomes so deeply depressed that she can no longer see any reason to go on. Again, my point is simply that this person has chosen (perhaps wittingly) to do things that compromise her capacity for autonomous choice.
As for question 3: As I note in the paper — and note briefly above – I do not feel qualified to say anything definitive about just which psychic conditions fall on which side of the fuzzy line between instances of pathology and character flaws. Can irascibility take a form that makes it appropriate to regard it as a mental illness? I don’t know. My aim in the paper was simply (i) to call attention to the fact that we do seem to think that the distinction between pathology and character is an important distinction, and (ii) to argue that this assumption is crucial to our assessments of moral responsibility (and that this explains why we find some of these assessments so difficult). I should, perhaps, add, that certain character traits may qualify as character flaws (vices) because they dispose people to do things that will lead them to acquire other traits that make it impossible for them to satisfy their basic needs. In this case, the choices influenced by the character flaws would themselves be autonomous choices.
This is the first time I have ever participated in a blog. I look forward to reading the comments that follow my exchange with Hilary (as well as those that have already been posted), and I will do my best to reply to these comments in the next couple days when I think I have something useful to say.
Sarah
Here are a few rather quick responses to the other comments on my paper. I will take them in the order in which the comments appear:
*Regarding the worry that on my account, a person cannot be blamed for what she does if her intention to do it can be attributed to the decisive influence of pain or fear. First, I want to stress that even if a given condition excuses our behavior, we cannot treat this fact as a reason to, e.g., yell at the nurse who is trying to help us. This would be to treat the fact that I have an excuse as a consideration in support of yelling. But this would be to misunderstand what excuses are. So, too, even if suffering intense pain excuses my yelling at the nurse is not a reason for me to yell at her. This means that, whatever one may think about my account of accountability, it does not imply that people are justified in cultivating — or failing to change — whatever psychological conditions make it more difficult for them to govern themselves. [This remark is, in effect, a reply to Dave’s second comment. I think that assessments of accountability are third-personal. But I also believe that no one can do anything intentionally without taking responsibility for what she does. This is for deep reasons concerning the relationship between ‘is’ and ‘ought’: no one can set herself an end without determining what to make of the facts of which she is aware; every rational agent has to draw her own conclusions about what she has reason to do. If some rational agents are nonetheless not accountable for the conclusions they draw, this must be a fact of the sort that is not accessible from the deliberative perspective.]
Second, as I note in the paper, when we say that someone did something because she was scared, we do not generally distinguish between the following two thoughts (this is necessarily rough): (i) her fear played a decisive causal role in which facts she noticed, and what “weight” she attributed to these facts; (ii) her action reflected her belief that she was in danger and her desire to avoid this danger if possible. From the mere fact that her action can be explained in terms of the sort of belief/desire pair in (ii), it does not follow that she has an excuse for, e.g., passing names to the secret police who are working for the corrupt dictator.
But (third) what if we do believe that her fear distorted her evaluation of her options? On my account, even in this case, we do not necessarily take it to be exculpating. This depends on at least two things. (1) Do we believe that the fear was of such an intensity that a human being who generally suffered such fear would find it very difficult to function? For all I say in the paper, fear may never satisfy this condition. (2) Even if the answer to (1) is “yes,” was there something about the agent’s character that explains why she made certain choices that rendered her vulnerable to this sort of fear? If so, then though she was not a self-governing agent when she passed on the names, she may well be accountable for doing so — since she may be accountable for the fact that she was “in the grip” of this fear.
*I have already responded to Dave’s second point above, and my response to his first point (about the status of children) is below. As for the third point, I do not think that “pitiable” is an “accountability predicate.” I do mention the case of the person who makes a heroic effort to overcome the influence of her depression. Here is what I say (in note 62):
Finally (at least for now), my answer to the last point can be found in “Justified Wrongdoing.” There I argue that if we are disinclined to blame JoJo and other people who take themselves to have sufficient reason to do what we believe they have overriding reason not to do, this is because (to make a long story short!) we take them their ignorance to be justified, given the unusual set of experiences they have had as children.
*Chrisoula’s question probably deserves a paper-length response. Perhaps some of the observations I made in response to Hilary are relevant. For now, I will limit myself to supplementing these observations with a citation from the passage in the paper where I most directly address the issue:
*As I note more than once in the paper, autonomous agents are necessarily agents who do things for reasons. This disqualifies any creature who acts from instinct, or on impulse. What interests me in the paper is the fact that agents who are in a position to give (justifying) reasons for why they did what they did are not necessarily accountable for acting for these reasons.
And what about children? Do they belong with the creatures of instinct or impulse? The all-too brief answer is that many young children do belong with these nonrational agents; and there may well be some circumstances in which teenagers do too. Of course, at a pretty young age, children learn to think of what they do as making sense, as appropriate, justified, reasonable, etc. If we truly believe that their actions are motivated by such normative assumptions, then if we do not believe that anything prevented a particular child from forming her own intentions on a particular occasion, it will be appropriate for us to hold her accountable for what she did — even though her relatively limited knowledge of all sorts of things will often make it inappropriate for us to blame her for going astray.
*Clearly, I do not share the intuition that someone’s happiness can excuse her negligence. (Nor, by the way, do I think that ordinary sadness qualifies — though my account does not depend on any such particular intuitions.) I would like to say something more than “I disagree!” For now, I will confine myself to wondering what more general assumption underlies the particular intuitions that differ from mine. What justifies these intuitions? In what sense is the happy agent less self-determining than others? (It can’t just be that her happiness plays a decisive role in determining her negligence. Whenever someone acts, aspects of her psyche play a decisive causal role in the formation of her intention. So the question is: in what accountability-undermining sense could her happiness qualify as “external” to her?)
*Perhaps there is an ambiguity in speaking of what “compromises” someone’s autonomy. An intervention fails to respect an agent’s autonomy when it is of a sort that is not permissible without the agent’s consent. But such an intervention need not have any impact whatsoever on the agent’s capacity to govern the formation of her own intentions — nor need it prevent her from exercising this capacity.
Thanks for the response to my happiness example. It makes me doubt whether we need to establish that the happiness is “external” in order to justify letting the person off the hook. Professor Buss’s paper has me imagining, instead, that all we would need to establish is that a *lack* of happiness would be very bad for the agent & that therefore being in that state would be “alienating” in the sense Professor Buss identifies.
On this view we should not hold the person responsible for the negligence caused by happiness because, and insofar as, it would be seriously bad for the person to not have been happy. Imagine that Jim is on cloud nine because he has recently fallen in love and gotten engaged. I am suggesting that his negligence is excused (or mitigated) because if he were not happy in this circumstance that would be seriously bad for him. If you are imagining that the lack of happiness would not be bad enough for Jim to excuse his negligence, you can simply re-imagine the case with a ramped up level of harm, to see that the strategy for explaining the excusing power of positive states could work.
The general idea is that positive states excuse or mitigate insofar as the agent would be alienated if she lacked that state & the sense of alienation is interpreted along the lines Professor Buss suggests. So although the reason against holding the person responsible would not depend on establishing that the positive state is “external”, it would still be in the spirit of Professor Buss’s account. One could then explain why actual positive states are normally not excusing by appeal to the fact that their absence is rarely bad enough for the agent to be alienating.
Just thought I would mention the idea in case it is of interest.
Sarah,
I’m not sure that, on your passive model, you can so easily dismiss worries about animal autonomy (and hence animal accountability). You say that a necessary condition for being autonomous is acting for reasons. But on your passive model what is it to act for reasons other than to act on the basis of facts that happen to accord in the right way with psychological elements with respect to which their bearer is ultimately passive? Perhaps I’ve misread your paper, but my understanding was that your model looks something like this: there are lots of facts out there, only some of which strike us as reasons for action. Those that strike us as reasons for action do so because they tap into various elements of our psychology with respect to which we’re ultimately passive and of which we may be unaware. And we are accountable for acting on those reasons so long as those psychological elements are conducive to flourishing.
That’s just a rough outline, of course, but if that’s your model then it seems to me that it’s going to apply to at least some non-human animals. And one of my concerns is that if you try to build more into the notion of “acting on a reason” so that non-human animals are excluded, you might wind up embracing elements of the super-agent model.
‘This means that, whatever one may think about my account of accountability, it does not imply that people are justified in cultivating — or failing to change — whatever psychological conditions make it more difficult for them to govern themselves.’
What incentive do I have for moderating my behavior if I have a get out of jail for free card in the form of my intense suffering? Why bother making the attempt? The next time it happens; I’ll simply apologize afterward or even Dostoevsky style during the boorishness, knowing full well that I will be met with understanding. I’ll tell you what would at least get me thinking a change is in order: someone dismissing my proffered excuse and admonishing me. The fact is my suffering is not even a good excuse for mistreating others. My faculties are still intact; I know that what I’m doing is wrong- why excuse me, especially since the patient down the hall isn’t acting that way? For children, suffering would be exculpating and we cut adults slack for a while, but at some point expect them to act right despite their pain. How long would Nurse Buss be willing to excuse pain related verbal abuse? Why not, moreover, hold the depressed mother accountable when she fails to fix lunch? We know that she is capable of doing it, albeit with some effort. When she doesn’t perform that simple task, we should instead ask her why she didn’t summon the strength exerted on other occasions.
‘As my comments about the relationship between authenticity and autonomy suggest, however, I also believe that we can be in the dark about what underlies our evaluative and normative judgments.’
Ever since reading Being and Nothingness, I have had a hard time wrapping my head around the notion of hidden springs of action. How autonomous can one be if such conatus exist?
Great paper.
I agree with Hillary that many mental states that are incompatible with flourishing are not disorders. This discussion echoes old problems with the DSM definition of “mental disorder” which is based on the idea of causing distress or disfunction. It has been pointed out that if we think of all mental syndromes that cause distress as disorders we would have to regard as disorders states that happen, given the way society is, to make other people hate you and cause you significant distress (in ways that Sarah seems to think of as compromising flourishing: you lose jobs, you end up in prison, etc). Homosexuality is an obvious example, Hillary’s unwillingness to condone injustice is an even better example, and even simply being a bad person or a criminal can be a case in point. Being a criminal or a serious contrarian can truly ruin your life. Does it mean it’s an illness? The writers of DSM try to deal with the problem by declaring ad hoc that a mere conflict between individual and society is not the same as a disorder, as the latter has to involve something wrong with the individual herself. Sarah hints at a solution by discussing the possibility that a society might be sick. But we need a way to tell the difference between cases in which the individual is sick and cases in which there is only a societal conflict.
But Hillary has explained the problem perfectly, so let me focus on an intuition that Sarah seems to rely on: the intuition that we treat negative emotions as excusing from blame (or praise) but we do not treat positive emotions that way. I think this intuition dissolves if we think of some more examples.
Studies show that good mood makes people more likely to help other people. Thus, people who have found change in a phone booth are more likely to help others than people who have not. When I tell my students about this their reaction suggests that they would be disappointed to know that a person who did them a good turn only did so because she had found change in a phone booth. To give a more ordinary example, consider a normally nasty boss who, on a given day, is experiencing a rare bout of good mood. Knowing that the boss is in a good mood you ask for a much-deserved raise and you get it without a problem. In such a case you would not hold the boss particularly praiseworthy for his action: you would attribute the action to the good mood rather than the boss himself.
Good moods can even ameliorate blame. A person who wakes her neighbors up in the middle of the night with an excited cry of “yesss” because she has just learned that she won the lottery is less blameworthy than a person who wakes her neighbors up in the middle of the night in a more cold-blooded way. On the other hand, if a person performs a brave and otherwise virtuous political act out of sheer anger at the government (real anger, veins bulging, shoulders shaking…) this person receives his praise. I don’t know if all these action deserve the term of art “autonomy”, but we do treat them as actions for which one is accountable.
I would like to respond quickly to Sarah’s response to my point about mania. I say that no amount of moral concern short of sainthood can prevent a person from some kinds of immoral behavior (say, throwing something in a fit of rage) if one is manic. Sarah says that one can substitute a character defect for mania and get the same result. This is wrong given my view of what character defects are. Character defects – the sort that damns, anyway – are deficiencies in moral concern (concern for the right or the good) or attitudes that oppose it (here my view is similar to the views promoted by Hurka and Adams when they say that virtue is about loving the good or about being for the good respectively). It is possible to say “even a saint would throw things in rage if she were manic”, but it makes no sense to say “even a saint would throw things in rage if he had a character defect” – not as an excuse, anyway. I will not defend my view of virtue and vice here.
oops. I got confused between versions there. I said that “I don’t know if all these action deserve the term of art ‘autonomy’, but we do treat them as actions for which one is accountable”, but obviously he angry action is the one we take to be fully accountable: the other two actions I described are less so.
Thank you very much for your response. I agree that speaking of what compromises someone’s autonomy might be too ambiguous. However, I’m not sure that speaking of a failure to respect someone’s autonomy captures the issue either. Or at least not respect simpliciter. To return to the example of the suicidal woman, let’s imagine that in her suicidal state she were to sign a contract. Now say she found herself in court due to a dispute over the contract and, given her emotional condition, the court concludes she lacked the capacity to enter into the agreement. Or to take another example, upon applying for a gun permit, her application is denied by the licensing authority due to knowledge of her condition. These would both seem to be cases of a failure to respect the woman’s autonomy. But I think there is an important difference in the cases of forcing or tricking her into taking medication, and is in the case of the mad scientist’s experiment. In the contract and permit examples, her intentions effectively determine her actions/activities [I introduce the latter term here only for lack of a better word, as an intervention doesn’t seem to properly be an action]. In those examples, it is just the actions/activities of other persons that her intentions fail to determine. But in the intervention cases, not only are the woman’s intentions disregarded, something is being positively forced on her. Her intentions have no say in the matter: it is due to solely to someone else’s will that she take the medication or undergo the mad scientist’s intervention, regardless of her intentions. This disconnect between the woman’s intentions and her activities (namely, taking the medication or undergoing the mad scientist’s intervention) is what I was attempting to refer to in saying that her autonomy is compromised. And because the intervention alters the conditions which govern the formation of her intentions with respect to suicide, it seems fair to say that the subsequent intentions she forms with respect to suicide are less truly “her own” than those intentions she would have formed had she consented to the intervention, or, indeed, even if the intervention had never taken place.
‘To give a more ordinary example, consider a normally nasty boss who, on a given day, is experiencing a rare bout of good mood. Knowing that the boss is in a good mood you ask for a much-deserved raise and you get it without a problem. In such a case you would not hold the boss particularly praiseworthy for his action: you would attribute the action to the good mood rather than the boss himself.’
Not if I held the boss responsible for his moods, Nomy. Do our moods just happen to us, and last just as long as “they” please? Or, as Sartre maintained, are we at least responsible for their sustenance? If moods are completely beyond our control, then, whether the boos is being his usual nasty self or not, his behavior isn’t up to him. If he can’t help being cranky when things don’t go his way, which is most of the time, then he isn’t blameworthy for whatever bit of rudeness then ensues. If there is nothing he can do about those rare moments of elation either, then praiseworthiness is out of the question in those instances as well.
Here is my third (and probably last) set of responses:
*I don’t understand Brad’s suggestion. So I’ll have to content myself with admitting as much.
*I’m also not sure what is bothering Mike. On my account, if an animal does things for reasons, then whether it is accountable for what it does depends on how it came to notice the facts that it does and to attribute to them a certain normative significance. If, on the other hand, an animal does not do things for reasons, then it is not capable of governing itself; and so it is not accountable for what it does. Doing something because you take yourself to have reason to do it is a necessary condition of autonomous agency. But it isn’t sufficient. The aim of my paper is to try to identify the additional condition. Does this help?
*Unfortunately, I can’t think of anything to add to my earlier response to Robert. Perhaps, however, it is worth stressing that, as I read Being and Nothingness, it makes the point I mentioned in my response to Dave’s first comment: when we are deciding what to do, what counts is whichever motives are“for consciousness.” This is perfectly compatible with the fact that there is a causal story to tell about these motives. I.e., it is compatible with the fact that my noticing X, Y, or Z and attributing to X, Y, and Z the normative significance I do are the effects of various psychological and nonpsychological causes. (How could they not be?) Again, the fact that these causes are not themselves “for consciousness” is precisely what makes it possible for some of them to be exculpating.
*I agree with Hilary and Nomy that there is important work to be done on the relationship between (1) disfunction that seems to depend on social attitudes and (2) disfunction that seems to be independent of such attitudes. One of these issues is the one Nomy flags in her comment: for any habit of thought and/or deed, we can always think of a social context in which these habits turn out to be self-destructive (and sadly, some of these possible social contexts are no mere possibilities). What conclusions should we draw from this possibility? What conclusions should we draw about someone whose inability to function minimally well is a function of attitudes and practices we take to be unwarranted, or worse? I suggested that in attributing someone’s “disability” to such attitudes, we are denying that she is sick. This suggests that our conception of pathology presupposes (at least in the case of mental illness) circumstances of action that are not unjust in ways that make it impossible for reasonable, morally upstanding human beings to thrive. More needs to be said about this. It seems to me, however, that what I said is correct, as far as it goes.
Another issue in this area comes up in discussions of selective abortion, preimplantation genetic diagnosis, and cosmetic surgery: Isn’t there a difference between, on the one hand, selecting for an embryo that will not have a disease — or performing surgery on someone with a cleft lip — and, on the other hand, selecting for an embryo that will not have traits that mark her out for discrimination, given widely held prejudices — or performing surgery on someone so that she will look more like a typical man’s ideal woman, or more like someone of the socially dominant race? No matter how much suffering would be prevented/ alleviated by the second sorts of medical interventions, it seems wrong to regard the conditions that are altered as “disabilities”; and so, it seems wrong to regard the changes as “cures.” There is lots to say about this. Here I merely want to note that many issues appear to be raised by the general worry about how to think about the relationship between individual disfunction and social disfunction.
Is it the fact that a person did a good deed because he was in a good mood that bothers Nomy’a students? Or is it the fact that his good mood — and hence his good deed — depended on his having found some coins in the phone booth? Since we are sharing intuitions, here are mine: I don’t believe that the causal role played by the coin-finder’s happiness undermines his responsibility for his deed. Similarly, I have no inclination to think that the boss’s good mood prevented him from making up his own mind. As I have stressed, I have no firm convictions about whether anger can reach such an intensity that it is reasonable to regard its influence on our judgments as accountability-undermining. (I would have thought that in the typical case of anger against injustice, it is the judgments that drive the anger.) I am simply committed to telling a particular story about what would have to be true in order for a given emotional state to be exculpating. As for that “yes!” my thought is that insofar as we believe that the person is not to blame, this is not because we believe that she is not accountable for her outburst, but because we believe that it is reasonable to respond to the super-duper news in this way. (If she were at the funeral of a dear friend, and was checking her phone messages, I doubt most people would think it was reasonable to respond in this way. Precisely for this reason, they would not excuse her. At any rate, this seems to me to be the correct response.)
My point about the distinction between someone with mania and someone with a character flaw is that the counterfactual condition is not the criterion for which of these is which; and so it cannot be the criterion of moral responsibility. We need an independent criterion for determining whether the disposition to which we can attribute a given act (e.g., the act of smashing a vase) qualifies as an instance of pathology or a character trait. Again, this will, necessarily, be a criterion for determining whether the intention that is caused by the disposition is attributable to the agent herself in a way that not every intention is. In short, it seems to me that the very distinction Nomy relies on (the distinction between mania and character flaws) presupposes an antecedent moral responsibility judgment. Of course, if having a given character flaw just is having too little moral concern on every given occasion (and if, therefore, it is not possible to have a flawed character which prevents one from acting on one’s moral concern in certain circumstances (does this rule out the possibility of weak-willed moral wrongdoing?)), then insofar as a person’s intention can be traced to this character flaw, we can say that if she had had more moral concern, then she would have done otherwise. But this is just to restate the fact that her action can be traced to her character.
*Is Chris’s point about the two women supposed to be that it was a mistake to conclude that they didn’t really exercise the capacity for autonomous choice? Or is it that they really were not capable of governing their own choices under the circumstances? In the latter case, it seems wrong to say that their autonomy was not respected. I’m not sure that I understand the examples — in part, because I’m not sure what Chris means in saying that “it is just the actions/activities of other persons that her intentions fail to determine.” In any case, I entirely agree that someone who is given medicine, or who undergoes surgery, without having given her consent to such interventions is denied an opportunity to so much as consider whether the intervention is one she endorses. And I agree that if anything is incompatible with respecting her autonomy, this is. I don’t think it necessarily follows, however, that the choices such a person makes as a result of this intervention are not her own. As I said, it seems to me that nothing rules out the possibility that such interventions could remove whatever was preventing a person from making up her own mind.
I’d like to thank all those who posted comments for prodding me to think harder about many issues I need to think harder about.
Sarah
One good Sartre reference deserves another. Thank you Professor Buss for a great discussion. I suspect, though, that your albatross, like most philosophical ideas, is likely to return.
Thanks so much for the excellent paper and discussion, Sarah. Sorry to have launched objections and run — I was traveling this past week — but hopefully we can take up some of these issues in person soon!