To understand one another as individuals and to fulfill the moral duties that require such understanding, we must communicate with each other. We must also maintain protected channels that render reliable communication possible, a demand that, Seana Shiffrin argues, yields a prohibition against lying and requires protection for free speech. This book makes a distinctive philosophical argument for the wrong of the lie and provides an original account of its difference from the wrong of deception.
Drawing on legal as well as philosophical arguments, the book defends a series of notable claims-that you may not lie about everything to the "murderer at the door," that you have reasons to keep promises offered under duress, that lies are not protected by free speech, that police subvert their mission when they lie to suspects, and that scholars undermine their goals when they lie to research subjects.
Many philosophers start to craft moral exceptions to demands for sincerity and fidelity when they confront wrongdoers, the pressures of non-ideal circumstances, or the achievement of morally substantial ends. But Shiffrin consistently resists this sort of exceptionalism, arguing that maintaining a strong basis for trust and reliable communication through practices of sincerity, fidelity, and respecting free speech is an essential aspect of ensuring the conditions for moral progress, including our rehabilitation of and moral reconciliation with wrongdoers.
1. This asymmetry figures prominently in most nonconsequentialist moral theories but also figures in sophisticated consequentialist accounts that give special weight to harm minimization.
2. Although these issues are closely related to the significant role harm plays in law, I focus here only upon the moral concept of harm and I do not attempt an account of the proper role it should play in law. Still, I believe that an adequate account of the moral concept should help make sense of the significant role the concept of harm plays in the law. I hasten to add, though, that I do not take a showing of harm to be a necessary predicate for legal liability and hence I do not regard it as one of the obligations of an adequate account of harm that it can support a comprehensive account of legal liability.
3. For some authors, though, this orientation is explicit. See, e.g., Hershovitz, Scott, Two Models of Tort (and Takings), 92Va. L. Rev.1147 (2006), at 1166–1167 (arguing that harms are necessarily the products of wrongs). In a recent set of very thoughtful and inventive articles, Matt Hanser rejects both an act-focused and a state-focused account of harm, advocating an event-focused account of harm. Hanser, Matthew, The Metaphysics of Harm, 77Phil. & Phenomenological Res.421 (2008); Matthew Hanser, Harming and Procreating, inHarming Future Persons 179 (Roberts, Melinda A. & Wasserman, David T. eds., 2009); Hanser, Matthew, Still More on the Metaphysics of Harm, 82Phil. & Phenomenological Res.459 (2011). Judith Jarvis Thomson offers some powerful criticisms of his view as well as some convincing replies to his attack on state-based views; Thomson, Judith Jarvis, More on the Metaphysics of Harm, 82Phil. & Phenomenological Res.436 (2011). See alsoHarman, Elizabeth, Harming as Causing Harm, inHarming Future Persons 137 (Roberts, Melinda A. & Wasserman, David T. eds., 2009), at 148–150.
4. It seems correct to say that a storm, an avalanche, or treacherous waters may bring one harm and that this notion of harm is the same as the notion characterizing the conditions that result from an agent's infliction of similar bodily injury on another. Moreover, it need not have been possible to save the victim, i.e., it need not be an instance of allowing, for naturally inflicted injuries to be harm. Similar reasons to act or to refrain from acting in response to such harms arise whether the harms emanate from natural forces alone or from human action or inaction. I aim to explore the range of overlap and similarity without claiming that the overlap is complete. But even if harm is believed necessarily to involve human agency, whether positive or negative, the paper's main points still hold between doing and allowing harm, on the one hand, and providing and failing to provide or protect benefits, on the other hand. The reason for this must, I think, have something to do with the difference of condition that harms and benefits occasion.
5. I do not mean to rule out the possibility that some harms may be harms only because of their causal origins, whether, e.g., because they were caused by persons under particular circumstances or by persons with specific motives. Further, the severity of some harms may be far worse when they were caused by persons or when they are not accidentally caused but rather caused by, e.g., a negligent or malevolent motive. See text accompanying note 43infra. See alsoPerry, Stephen, Harm, History, and Counterfactuals, 40San Diego L. Rev.1283 (2003), at 1294–1295. But seeGoldberg, John C.P., Rethinking Injury and Proximate Cause, 40San Diego L. Rev.1315 (2003), at 1322–1323.
6. Others associate harm with a different moral significance. For example, whereas I focus on the close connection between harms and reasons to alleviate, prevent, and refrain from causing them, Judith Jarvis Thomson draws a tight connection between suffering harm and having a claim against someone that they not bring it about. Judith Jarvis Thomson, The Realm of Rights (1990), at 228–48. On this point, we differ. Although I think harms provide reasons to prevent them, to alleviate them, and to refrain from causing them, I do not believe that those reasons necessarily underwrite claimsagainst others. The relevant reasons may not be that strong, for instance. Further, Thomson's connection between harm and claims against othersmay divert our attention away from self-inflicted harm. Differences about the nature of harm's moral significance may in turn generate different judgments about what counts as harm. I mention a few such divergences in what follows, especially note 24infra and Part IV.D.
7. The idea of comparably sized benefits and harms is, I concede, more than a little obscure. For my purposes, strict comparability is not essential. We need only have in mind, e.g., the rough comparisons of minor benefits to minor harms.
8. That is, the recipient has not merited and has no special right to the benefit, the harm, or its prevention. The priority on harm is easiest to discern here but arises even with deserved harm, as I discuss in note 33infra.
9. Important moral issues turn on whether and when larger benefits generate stronger reasons for action than lesser harms. Nonetheless, the basic elements of harm's priority do not depend upon a specific stance on this matter. Other asymmetries I discuss, though, suggest that larger benefits do not per se generate stronger reasons than harms. The view that harm generally provides stronger moral reasons to act does not entail that it is always impermissible to pursue a benefit if there is (as there usually is) an opportunity to refrain from, avert, or alleviate harm. Legitimate personal relations, goods of character associated with benevolence, and personal prerogatives may support permissions to pursue some benefits. So this fairly weak characterization of harm's priority can evade the criticism of excessive stringency that are faced by other views closer to negative consequentialism that posit a strong, overriding obligation to avoid or relieve harm. See, e.g., Griffin, James, Is Unhappiness Morally More Important Than Happiness?, 29Phil. Q.47 (1979).
10. There are complications here with respect to non-anonymous giving, especially when it involves transformative benefits. Such bestowals may substantially alter personal relations, risking hierarchy and dependence and/or introducing duties of gratitude. These considerations may generate greater justificatory demands for non-anonymously bestowed benefits. Barbara Herman has recently written illuminating work on the subject. Barbara Herman, Being Helped and Being Grateful: Puzzles about Imperfect Duties, the Ethics of Possession, and the Unity of Morality (2012) (unpublished manuscript). (Of course, some of these dynamics may also arise with respect to preventions and alleviations of harm.)
11.Feinberg, Joel, Wrongful Life and the Counterfactual Element in Harming, 4Soc. Phil. & Pol'y145 (1986), reprinted inFreedom and Fulfillment 3, 27 (1992).
12. There may, of course, be limits to these justifications, depending upon how serious the lesser harm is, how minor the greater harm is, and how close the distance is between the lesser and greater harm. Nonetheless, in a large range of cases, it is permissible to inflict a lesser harm to avert a greater harm.
13.See, e.g., Feinberg, supranote 11; Frances Kamm, Creation and Abortion 173–174 (1992); McMahan, Jeff, Wrongful Life: Paradoxes in the Morality of Causing People to Exist, inRational Commitment and Social Justice: Essays for Gregory Kavka208 (Coleman, J. & Morris, C. eds., 1998); McMahan, Jeff, Cognitive Disability, Misfortune, and Justice, 25Phil. & Pub. Aff. 3 (1996); Bernard Williams, Resenting One's Existence, inMaking Sense of Humanity 224 (1995).
14. I discuss this point in Shiffrin, Seana Valentine, Wrongful Life, Procreative Responsibility, and the Significance of Harm, 5Legal Theory117, 121–126 (1999).
15. Frances Kamm has prompted me to note that it would likewise be wrong, without evidence of the patient's will, to perform the harm-inflicting surgery to preserve these benefits. (The fact that the patient had already undergone similar surgery to achieve these conditions, though, might constitute such evidence.) Still, there may be greater reasons to act to preserve an already vested pure benefit, perhaps because of a patient's strong personal identification with it, than to bestow a benefit. But unless the identification is so strong as to make the loss a harm, both nonconsensual surgeries would be impermissible.
16. A weaker version of the asymmetry claims that harming to bestow a benefit requires different, stronger justifications. It may demand that a benefit be of a greater magnitude than the magnitude of the greater harm that justifies lesser harm. Or, it may permit bestowal of the benefit only if the benefactor has a good motive and assumes responsibility and liability for harm caused in the delivery of the benefit.
17. Also, suppose the patient really does appreciate the benefit conferred. She may even be, all things considered, glad it was bestowed. Still, if she complained that its nonconsensual bestowal also imposed harm on her and demanded compensation, she seems entitled to such compensation, whereas the patient in the Rescue Case cannot reasonably demand compensation. This suggests that those who aim to expose themselves to harm or to avoid harm-reduction efforts operate under a larger burden to make their intentions known.
18. Here I bracket issues about socially irresponsible uses of medical resources and assume they do not arise in this case.
19. Some doubt that it does. See, e.g., Perry, supranote 5, at 1301.
20. An appeal to autonomy rights also fails to explain another closely related first-person/third-person asymmetry. As Scanlon observes, our assistance to others (whether morally mandated or not) may reasonably be targeted towards fulfilling their nutritional needs even if the recipients sacrifice their health to build a religious monument and would prefer us to support their mission. Scanlon, Thomas, Preference and Urgency, 72J. Phil. 655 (1975), at 660. At least one of the example's lessons is that in my terminology, our welfare efforts may reasonably favor preventing and alleviating harm (i.e., nutritional deficiencies) over providing pure benefits. Still, recipients of such assistance might reasonably choose to weight their expenditures toward the fulfillment of their projects at the expense of their full nutrition. It cannot be argued here, as it might have been in the medical cases, that harm prevention is preferred because of ignorance of the recipients’ wishes or because of limitations imposed by the recipients’ autonomy rights, for the donors may target their assistance as they like.
21.See, e.g., Thomson, supranote 6, at 261.
22. Feinberg is explicitly interested in legitimate principles of criminal liability and hence with the more narrow category of legal harm. Legal harm also requires as a component an element of wronging or wrongdoing that involves a rights violation. Feinberg, supranote 11, at 36. His conception of legal harm, however, rests on an underlying broader theory of morally significant harm.
23.See also Matthew Hanser's discussion of the relation between similar examples and the killing/letting die distinction. Hanser, Matthew, Why Are Killing and Letting Die Wrong?, 24Phil. & Pub. Aff. 175 (1995), at 183–186.
24. Because the shot causes her to suffer harm, I would also say that Jones harmed her. If his act was performed with the proper motive, e.g., to avert greater harm, his act would not merit criticism. But it still inflicted harm. Had Jones performed it for a poor motive, e.g., the intent to inflict suffering, we would condemn him for harming Black, even though the alternative would have made her worse off. Some may claim that “causing to suffer harm” and “harming” come apart and that it is a conceptual truth that for A to harm B, A must worsen B's position. Thomson discusses a related case concerning historical accounts. Thomson, supranote 6, at 263. She denies that you harm someone if you cause him (what would be a) lesser harm, e.g., amputate a leg, to eliminate greater harm (death or a more severe disability occasioned by a crushed leg). Her denial may be motivated by the permissibility of performing the amputation. But I do not think this excludes the possibility that the act harms. Even if harm were an essentially normative notion, the fact that it should be inflicted and that its infliction does not elicit charges of culpability or wrongdoing does not exhaust the forms of its significance. Thomson's position may also emanate from the tight connection she draws between suffering harm and having a claim against someone that they not bring it about, a connection I regard as too strong. Seesupranote 6. The amputated leg itself, prima facie, generates strong reasons to relieve the suffering associated with it, e.g., through pain relief and a prosthetic; further, it requires the same sorts of justificatory reasons associated with harm to bring it about, so I regard it as a harm even if one lacks a claim against the surgeon that she not amputate.
25. Why am I not worse off than before? Before I was about to be saved by Harry, and I am no longer in this anticipatory position. There are two difficulties with this maneuver. First, suppose I am left in the same position. That is, even after your intervention, Harry is still just about to help me. (And suppose you will continue to intervene to preclude his help from actualizing.) Your action does not eliminate Harry; you just preclude his putting his intentions into practice. It is unclear, then, in what way you have made me worse off than I was. Second, if we say that I am worse off because Harry was about to save me and be effective, then in essence the historical position has collapsed into the counterfactual account. For the account of why your action makes me worse off is not that I am in a worse position but, in essence, that I could have been better off but your action eclipsed this alternative. If interpreted that broadly, the historical account encounters the difficulties of the counterfactual account.
26. I do not suggest that comparative assessments are irrelevant for moral purposes. They are particularly relevant to choices between harm-inflicting or benefit-bestowing alternatives. If I am forced to harm one of two people and one person's harm is overdetermined, this gives me reason to harm her and to spare the one who may avoid harm entirely. But mere overdetermination will not necessarily excuse harm (or make a harmful activity nonharming). Other things being equal, it would be harmful and wrong to kill an innocent person even if she would be immediately killed by another were I to refrain. Further, I may have reason to provide a benefit even if others will provide it should I refrain. These reasons may include an obligation to contribute to a helping enterprise, to avoid free-riding on others’ beneficence, or to establish or contribute to a particular relationship between myself and the beneficiary. See also Hanser,Why Are Killing, supranote 23; Hanser, Matthew, A Puzzle about Beneficence, 58Analysis159–166 (1998).
27.See Shiffrin, supranote 14. Many discuss the problems associated with such cases. See alsoDerek Parfit, Reasons and Persons (1984), at 351–442; Feinberg, supranote 11; Hanser, Matthew, Harming Future People, 19Phil. & Pub. Aff. 47 (1990); Kavka, Gregory, The Paradox of Future Individuals, 11Phil. & Pub. Aff. 93 (1982); McMahan, Wrongful Life, supranote 13; McMahan, Cognitive Disability, supranote 13; Thomson, supranote 6, at 262.
28.See Derek Parfit's discussion in Parfit, supranote 27, at 366–371.
29.Seesupranote 24; see alsoPerry, supranote 5, at 1305.
30. Both Thomson's and Raz's accounts seem awkward in this way. They acknowledge that pain gives rise to reasons that are like the moral reasons associated with harm, but yet they deny that pain is a form of harm. Thomson, supranote 6, at 249–253, 264; Joseph Raz, The Morality of Freedom (1986), at 413–414.
31. Ronald Dworkin argues that pains frustrate my experiential interests but they do not frustrate my critical interests and therefore do not make my life overall go better or worse. Ronald Dworkin, Life's Dominion (1993), at 201.
32. One might modify the comparative account by appealing to thresholds in which only conditions that fall under a threshold of minimal overall interest satisfaction should count as harm. Such accounts may evade cases like the billionaire's monetary loss, but the price for evasion is too high. For it will then be difficult to acknowledge that the billionaire's broken leg is a harm to him should he remain above the minimal welfare threshold. One might then suggest that different thresholds cover different aspects of well-being. Ascents up the financial ladder do not compensate for physical harm.
Stephen Perry puts forward a more nuanced historical account that gestures in this direction, suggesting that historical worsening is a necessary but not sufficient condition of harm. Perry, supranote 5, at 1302. This would represent a step in the right direction, but without more elaborated details, it leaves important questions open and does not dispel the lingering suspicion that the significance criteria that infuse the necessary conditions will implicitly incorporate noncomparative considerations that do the fundamental work. Where to set the threshold reopens the question of what harm is that the standard account lacks the internal resources to answer.
33. This problem could be avoided by modifying the account to claim that one is harmed only if one is worse off than one deservedly was or would have been. This modification would, however, create a worse problem by eliminating the possibility of deserved harm. Michael Otsuka has suggested in conversation that the priority on harm might be understood as a priority on undeserved harm. Although this suggestion has some plausibility, there are problems with it. Even deserved harm is commonly thought to require special justification; witness the dissatisfaction with spare retributivist accounts and the desire to find further, positive reasons for punishment. See, e.g., Hart, H.L.A., Prolegomenon to the Principles of Punishment, 60Proc. Aristotelian Soc'y1 (1960); Goldman, Alan, The Paradox of Punishment, 9Phil. & Pub. Aff. 42 (1979). Perhaps one's purposes need be less important if the harm is deserved or if the recipient's superior position is undeserved, but the fact that harm is involved elicits special justificatory burdens. The burden of justification for inflicting deserved harm or harmful deprivation of undeserved goods seems higher than that associated with deserved failures to benefit or failures to protect undeserved benefits.
34.Thomson, Realm, supranote 6, at 227–271. She distinguishes pain and other forms of non-belief-mediated distress, such as nausea, but believes they may give rise to the same sorts of claims against others that harm does. By contrast, she argues that belief-mediated distress such as grief, indignation, disgust, anger, and disappointment do not themselves give rise to claims. I discuss this distinction in Section IV.Dinfra.
35.Kleinig, John, Crime and the Concept of Harm, 15Am. Phil. Q. 27 (1978), at 31, 33.
36. Thomson may emphasize worsening because she, like many others, is interested in providing an account of harm as that which we have a claim against others’ bringing about. Seesupranote 6. One might think, then, that worsening is not a primitive quality of the harmed state but is essential to causing harm in this sense. But if I fail to alleviate your suffering although I easily could, this could be plausibly described as a case in which my failure harms you and even I harm you although you are not made worse off; and further, you might have a claim against my doing this. But this issue lies outside my inquiry here. I mean to focus on the sorts of conditions that give rise to reasons to act and to refrain from acting, whether these conditions are brought about through human agency, its absence, or natural conditions, and whether or not these reasons are sufficiently strong to form the basis of a claim.
37.See, e.g., Harman, Elizabeth, Can We Harm and Benefit in Creating?, 18Phil. Persp.89 (2004), at 96–97; Harman, Harming,supranote 3, at 139. Harman, addressing what actionsharm people, identifies harmful states as ones that are worse (in some respect) than life with a “healthy bodily state,” i.e., states with cuts, burns, diseases, deformities, pain, and premature death. Although the characterization largely works for her purposes (except perhaps that it does not correctly capture what is harmful about rape), this approach does not explain why harm is limited to (physical?) health defects. Many harms we suffer do not necessarily affect our health. A more satisfying theory would accommodate this fact and explain the connection between states of compromised health and other sorts of harms.
38. I discuss related issues in Shiffrin, Seana Valentine, Autonomy, Beneficence, and the Permanently Demented, inDworkin and His Critics195 (Burley, Justine ed., 2004).
39. They may be most useful in explaining where and why autonomy rights generate strong positive duties. So we may have duties to contribute to creating conditions in which people may develop their capacities, by providing adequate options, sufficient educational training, and a tolerant atmosphere, etc. These duties may be linked to the character-based aspects of autonomy's value.
40. A related account of harms and benefits, but one with more developed and pronounced Aristotelian features (and one that takes a historical comparative form), is offered in David Velleman's fascinating and thoughtful three-part article on future people and procreation. See Velleman, J. David, Persons in Prospect: Parts 1, II, and III, 36Philosophy & Public Affairs221–288 (2008).
41. As Greg Keating observed to me, in some contexts, sexual and narcotics experiences may bring pleasure but yet be overwhelming in ways that shade or threaten to teeter into this characterization. In the good cases, the absence of control and the totalizing nature of the experience are sought and well realized; the experience is more liberating than invasive. In bad cases, the borderline is crossed, and though pleasurable in some respects, the experience becomes alienating and sometimes frighteningly so. And, of course, addiction and other sorts of dependence may be risked that involve a more thorough compromise of one's agency and autonomous capacities.
42. The claim that the contents of one's life exceed the boundaries of one's beliefs and experience has been widely discussed in the literature on death. See, e.g., Nagel, Thomas, Death, inMortal Questions1 (1979).
43. The frustration or failure of certain projects that essentially involve risk or competition raises further complications that I can only briefly describe. When such projects fail or are frustrated because the risk is realized or the competition is lost, we may have reason to deny that failure or frustration so caused represents a harm. Or in the alternative, we might regard their failure or frustration as self-inflicted harm. I am tempted to say that properly understood, such projects really involve attempting to surmount a risk or attempting to best others. Thus the failure to win, for example, therefore does not involve a failure or impediment to mounting the attempt. One might also say, perhaps, that the failure of that sort of project for that reason does not deposit one in a condition foreign to one's will; where the essential value of the project involves confronting risk or competition, the realization of the downside may be disappointing but is an aspect of the world one embraces, as opposed to the sort of destruction caused by a hurricane, earthquake, accidental plane crash, or malicious destruction by another. Thus I dispute that, as Scott Hershovitz seems to maintain, the intuition that a loss or setback caused by competition is not harm may only be explained by embracing the idea that harms are necessarily the product of wrongs. Hershovitz, supranote 3, at 1166–1167. A classic discussion of the relationship between harm and losses sustained through competition appears in Mill, John Stuart, On Liberty, inOn Liberty and Other Essays104 (Gray, John ed., Oxford University Press, 1988) (1859).
44. If we conceive of promises as transfers of power or decision-making authority, then we might be able to conceive of breach of promise (and breach of contract) as a harmwithout having to resort to a counterfactual comparative theory. For worries as to whether noncounterfactual theories can accommodate breach of contract, see Goldberg, supranote 5, at 1325.
45. Matthew Hanser suggests that a state-based approach cannot accommodate the idea that death is a harm because the person who has died no longer exists and so is not in a bad state per se. Hanser, Metaphysics of Harm, supranote 3, at 437–440. I disagree. First, being subject to death can represent a harm. One's mortality coincides with one's existence, even if one's death does not. That reply, however, does not capture the sense that not merely being subject to death but actually the fact of death is a harm to one. But if we think the boundaries of a person or her life may extend beyond her corporeal existence, then she may suffer harm even when she no longer exists. That is, Hanser's argument seems to presuppose an existence requirement that is in conflict with other familiar ideas, such as that some events after a person's death may harm them by frustrating their core projects. Importantly, the way death harms one is, on this account, unsurprisingly an extreme version of the way other harms operate; the possibility of exerting and implementing one's will is not merely impeded or obstructed, but eliminated. One might also take a different route and argue, as Judith Jarvis Thomson does, that death is reasonably thought of as a special case. Thomson,More on the Metaphysics, supranote 3, at 453–456. See alsoMatthew Hanser, Still More, supranote 3, at 462–465.
46. So this approach is not precluded from counting as harms natural deaths, ordinary inabilities, ordinary pains, naturally contracted diseases, or the susceptibility to them because of their naturalness or ordinariness. They impose states that generate strong reasons to act to avert or relieve them; their strength is not lessened just because they happen to everyone or are predictable. Still, the context may matter enormously as to which states function as harms; an atypical condition in a particular social context may operate as an inability, whereas in another social context differently designed, it might not; being deaf in a dominantly hearing community is often a harm, whereas being deaf in a predominantly deaf community or in nineteenth-century Martha's Vineyard (in which sign language was widely used by the hearing and the deaf alike) might not be (or might represent a significantly lesser harm, depending on whether one is inclined to regard a lack of access to music as a deprivation of a benefit or as significant form of estrangement from the world and therefore a harm). SeeNora Ellen Groce, Everyone Here Spoke Sign Language: Hereditary Deafness on Martha's Vineyard (1985). Although I regard it as a strength of the approach that it can recognize quotidian conditions such as ordinary pains and diseases as harms, others see this as a liability and worry that this approach overidentifies harms. SeeKamm, Frances, Baselines and Compensation, 40San Diego L. Rev.1367 (2003), at 1384–1385; Cohen, I. Glenn, Intentional Diminishment, The Non-Identity Problem, and Legal Liability, 60Hastings L.J.347 (2008), at 373–374.
47. In this way, a will-centered account may capture some of what is attractive about the standard, interest-based model. As I am arguing, interest-based accounts seem overbroad and threaten to veer too close to desire-fulfillment accounts that pose a challenge to a maintaining harm's priority. But still they exert a persistent pull. Part of their attraction, I think, is their connection, through the emphasis on what the agent has a stake in, to the agent's will. An approach that locates some essential connection between harm and the will may accommodate some of what drew us toward standard accounts in the first place. For example, it has the resources to acknowledge that certain severe losses are harms without claiming that all losses are harms. Some severe losses of position or comparative setbacks to interest will not land one in a position that is intrinsically harmful but may represent a harm for that individual because the loss will place her in a circumstance that is foreign to her, given her self-conception, expectations, and investments in her present situation.
48. Thus I am sympathetic to Joseph Raz's account of harms as impediments to autonomy. Raz, supranote 30, at 369, 448. But there are reasons to be cautious about this characterization. It can be understood in a way that is overbroad. Many but not all obstacles to self-authorship are harms. Impediments to full self-authorship may well be undesirable, and their removal will often represent a benefit. But if one's relation to one's experience, life, and circumstances is sufficiently cognizant and integrated, then it may be desirable (for some people) to be able to replace some of unchosen elements, but it might not be essential for that agent's avoiding harm. If every obstacle to complete self-determination were considered a harm, then the account would face renewed problems about explaining harm's special priority. Second, the view can be underinclusive. By focusing solely on the abilities and obstacles to developing and exercising one's autonomy, it tends to exclude noxious, intrusive, but temporary conditions such as pain.
49.Thomson, Realm, supranote 6, at 250.
50.Id. at 253.
51.See supranote 34 and accompanying text. See alsoThomson, Realm, supranote 6, at 253–259.
52.Thomson, Realm, supranote 6, at 254.
53.Id. One might have a claim that another not engage in threatening act that gives rise to the belief and the distress, but the claim does not emanate from the harm of distress but from something else that is wrong about the threatening itself.
54.Id. at 254–255.
55.Id. at 254.
56.Seesupranote 6 and accompanying text.
57. Greg Keating, in a thoughtful discussion, argues for considering some forms of emotional distress as forms of harm, at least for legal purposes. Keating, Gregory C., Is Negligent Infliction of Emotional Distress a Freestanding Tort, 44Wake Forest L. Rev.1131 (2009).
58. I would not suggest, however, that the harm of grief provides standing reasons not to inform someone of a death. How we avoid and alleviate these harms must be appropriately responsive to the values that underlie them.
59. Frances Kamm makes a related point that some conditions that normally give rise to reasons (and claims) may be irrelevant to certain choices and therefore fail to render a claim more stringent. For instance, if we must choose whether to divert a trolley onto one person rather than another, where the diversion will cause either victim to lose a leg, it should seem irrelevant that one victim would also suffer a sore throat (even if usually claims restrict us from causing sore throats). Her analysis suggests that some states may give rise to reasons in some circumstances but may be overwhelmed or rendered irrelevant in others. SeeFrances Myrna Kamm, 1 Morality, Mortality: Death and Whom to Save From it (1993), at 144–164.
60.Thomson, Realm, supranote 6, at 257–259.
61.See my related discussion in Shiffrin, Seana Valentine, Egalitarianism, Choice-Sensitivity, and Accommodation, inReason and Value: Themes from the Work of Joseph Raz270–302 (Pettit, Philip et al. . eds., 2004).
62. A somewhat similar story might be told about appropriate guilt. To recognize it as a harm associated with moral reasons not to evoke it would be in tension with the grounds that render it appropriate.
63. This issue of the degree to which what “harm” encompasses incorporates or may conflict with other foundational normative standards is fleetingly broached in the discussion above of comparative accounts’ implication that equality norms (of justice) are in necessary tension with harm's priority, at note 33 and accompanying text, in the idea that conditions caused by rights violations may be harms even if the condition as such would not otherwise be, and also in the discussion of belief-mediated distress in Section IV.D.
64. The answer may depend on the source and strength of the obstruction and how significant the endeavor is to the agent. Although certain significant discordances between the will and one's experience, life, or circumstances are the key components of harm, complete achievement of what one wants is not necessary to avert harm. Many preferences may go unfulfilled without making one's life foreign and without, in contrast with pain, presenting an intrusive, noxious, or importantly alien element into one's direct consciousness. Of course, persistently present preferences that one disapproves of or otherwise disidentifies with but that resist efforts at removal may themselves represent harm. It is their continued presence, rather than their nonfulfillment, that is harmful. On the other hand, persistent and widespread frustration of all or very many of one's preferences may deposit one within circumstances that render one's life significantly foreign and to be endured, watched from within. Likewise, not all obstacles to one's projects will represent harm, or at least, not overall harm. The ability to choose one's endeavors and to participate in efforts to promote them matters here, as does the fact that many endeavors reasonably involve risk and competition. To will them is to will participation in an endeavor whose success may depend upon the achievement of excellence through collective striving and competition. One may not desire frustration, but it will occur as part of a process to which one has committed and whose fruition more integrally depends upon competition for the achievement of its aims. With some projects, realistic appreciation of their nature means that one's reasonable pursuit of them is not necessarily aimed at success or achievement of their aims but at making a good-faith effort, so that some forms of failure are, at bottom, consistent with what one has willed. See also supranote 43.
* Largely, I wrote this paper more than a decade ago as an extension of some arguments made in Seana Valentine Shiffrin, Wrongful Life, Procreative Responsibility, and the Significance of Harm, 5 Legal Theory 117–148 (1999). I put it aside in dissatisfaction. I remain dissatisfied. Nonetheless, despite my lingering doubts, I have not found sufficient cause to reject its main ideas either. The manuscript has circulated a little over the years. Because some have cited (and criticized) it, Gregory Keating convinced me it might be useful if it were publicly available, despite its flaws. I publish it with hesitation, however. I wish it were better, clearer, and shorter. I am grateful to a number of people and audiences for patient, insightful help and critical encouragement. I owe particular thanks to Janet Broughton, Tyler Burge, Meir Dan-Cohen, Hannah Ginsborg, Matthew Hanser, Barbara Herman, Frances Kamm, Sanford Kadish, Gregory Keating, Herbert Morris, Michael Otsuka, Samuel Scheffler, Steven Shiffrin, Terry Stedman, Matthew Strawbridge, Judith Jarvis Thomson, J. David Velleman, Daniel Warren, and Stephen White.