If no one is morally responsible, how should we respond to wrongdoing? Over the past twenty-five years, Derk Pereboom has grappled with this question with tremendous ingenuity, rigor, and generosity to his interlocutors. That responsibility skepticism is no longer regarded as a merely notional possibility, or the province of a handful of historical figures, is attributable to his efforts. In Wrongdoing and the Moral Emotions, Pereboom offers a new and wide-ranging account of what remains when we reject the idea that people are at least sometimes—and in at least one important sense—morally responsible for what they do.The book’s animating idea is that our responsibility practices employ a class of unjustifiable attitudes of moral anger (for example, resentment and indignation) that are retributive, in that they reflect the presumption that their targets deserve to suffer or experience pain. In earlier work, Pereboom held that this kind of blame should be replaced with “moral sadness” at unwarranted wrongdoing. Here he offers a less emotionally detached and more interpersonally assertive reform of our practices, calling for a stance of moral protest against unwarranted wrongdoing. This stance permits expressions of “measured aggression” directed at wrongdoers, a proportionate and controlled defensive “fury” (72).Chapter 1 summarizes Pereboom’s deservedly influential views about free will and moral responsibility—namely, that we lack free will of the sort that is required for our being morally responsible in the “basic desert” sense. He holds that deliberation and choice earn their keep by providing a solution to the pervasive fact of epistemic openness, even if determinism is true (25). The second and third chapters present his recommended revision in our moral practices. Chapter 2 presents a conception of blame that does not presuppose what he calls basic desert but instead is grounded in several broadly instrumentalist or forward-looking considerations (52–53). Chapter 3 provides an account of the moral psychology of this attitude of measured aggression; Pereboom argues it can support defensive harm without presupposing basic desert. The remaining chapters explore the ramifications of this account for questions of criminal justice (chapter 4), the possibility of forgiveness (chapter 5), love and relationships (chapter 6), and the role of hope (chapter 7). Despite the complexity of the issues and the scope of the involved literatures, the discussion is consistently accessible.In what follows, I focus on the book’s two overarching theses: anti-retributivism and what we might call anti-angerism. Anti-retributivism holds that “we can do without retribution, whether it be in justifying our responses to wrongdoing, or in the emotions employed in those responses” (1). Emotions are retributive when they have a presupposition of deserved pain or harm (3). Anti-angerism holds that “moral anger, whether or not it presupposes basic desert, has too prominent a place in our practice of holding morally responsible, and too central a role in many normative accounts of that practice” (3).The case for anti-retributivism depends on identifying what makes an attitude retributive and on showing that retributive attitudes cannot be vindicated—that is, be shown to be apt, justified, or true. The more demanding the notion of retributivism, the harder it will be to vindicate; the less demanding, the easier it will be to show that retributivism can be vindicated. Pereboom’s retributivism explicitly invokes the idea of deserved pain or suffering, even though there are prominent but less demanding characterizations of retributivism (for example, in David Brink’s recent work) that only require the deprivation of certain rights or goods that reflect the nature and gravity of culpable wrongdoing. This latter characterization does not obviously require the relatively strong form of agency on the part of the wrongdoer that Pereboom thinks is required for his version of retributivism. Pereboom says relatively little about why a stronger notion is the right one. The issue is perhaps complicated by his view that the content of our retributive commitments may not be readily discernible from ordinary thought (35). If so, it makes it unclear what considerations favor the more demanding notion of retribution. The non-error-theoretic retributivist might object that Pereboom has stacked the deck against less demanding and more plausible forms of retributivism.Pereboom’s repudiation of retributivism entails that no one can “basically deserve” suffering or pain. Basic desert holds that a wrongdoer would deserve blame or punishment “just by virtue of having performed the action with sensitivity to its moral status, and not, for example, by virtue of consequentialist or contractualist considerations” (12). However, Pereboom acknowledges that the theoretical options are not just retributive desert or forward-looking blame. There can be accounts of moral responsibility that rely on a nonbasic conception of desert. Consider the idea, sometimes associated with Rawls, that social practices can adopt a system of penalties—for example, fouls and fines—because of the penalty’s positive effects for the practice. In some of these instances, the conditions of application of the penalty are a matter of what the offender has done. That is, the propriety conditions on the penalty can be entirely backward-looking, even if justification for having penalties is instrumentalist.A first puzzle about Pereboom’s handling of nonbasic desert is dialectical: after rejecting basic desert retributivism and acknowledging that there might be nonbasic notions of desert, Pereboom says very little about it, instead focusing on the appeal of “measured aggression.” Yet, given that a nonbasic notion of desert is not excluded as a possibility, and given that it would involve less radical transformation of our practice, this available but unexplored option haunts his central argument. If there is a nonbasic retributive account available, as some are inclined to think there is, it seems no small advantage if it does not entail a radical transformation of moral anger’s psychology.A second puzzle concerns the explanatory stakes of the idea of basic desert itself. Pereboom claims that a test of basic desert is the Kantian thought that a criminal deserves to be punished, even after society ceases (31). Yet nonbasic desert accounts can pass that test. Recall that we can distinguish (1) forward-looking systemic considerations in favor of having a practice with a given structure from (2) internal to the practice, purely backward-looking considerations of the propriety (or application, or truth, or aptness) of first-order desert judgments. For a practice where there are penalties that are deserved only by satisfying (by stipulation) entirely backward-looking conditions, then even if a society dies off, leaving a lone criminal, it could still be true that the criminal deserves to be punished even if no one is there to punish and even if punishing will produce none of the systemic effects that justified having that practice in the first place. Statuses can persist even if the justification for having them disappears.If that’s right, then basic desert seems to combine two importantly different ideas at two different registers: the propriety of an individual ascription of deserved blame, internal to a practice; and the external, independent, and systemic question of the basis of having a desert base like so (for example, whether on contractualist, consequentialist, or other grounds). To the extent to which basic desert’s appeal is that it captures the Kantian intuition, it isn’t obvious that desert needs to be basic. So, there is an apparent explanatory puzzle about basic desert: Why should we accept its entanglement of a view about the basis of individual ascriptions with a view about a separate matter—that is, the basis of systemic justification?Turning to the second animating thesis of the book, anti-angerism, Pereboom holds that, irrespective of considerations of basic desert, the practice of holding people responsible “malfunctions in general and crucial respects” because of moral anger (3). He rightly notes that when people are angry, they misrepresent features of the situation, they are prone to defiance and seeking humiliation, and this alienates others and predisposes people to confrontation. Instead, he recommends compassion and, as we have seen, measured aggression.The apologist for anger should be unmoved. The fact that expressions of some attitude can distort a practice is unconvincing evidence for the claim that we are better off without that type of attitude. When people are in love, they misrepresent features of the situation. Loving attitudes also motivate confrontation, alienation, and humiliation. Even if most instances of blame go badly and most relationships end in heartbreak, it doesn’t follow that we rightly settle whether to keep responsibility or love by counting the success cases and subtracting the failure cases. The question is a systemic one, whether we are overall better off with the attitude.A compelling answer requires a clear picture of what roles and functions these play in our overall lives. Love, anger, and the like are oftentimes the price of admission into certain kinds of practices and particular ways of being. They are constitutive of other things. They enable other goods that we have overwhelming reason to want. Moreover, anger that is comparatively unmeasured, in being indifferent to whether it secures goods comparable to its cost, has a variety of anticipatory functions, including the shaping of our sensitivities. Attunement to the risk of angry blame is uniquely effective at shaping valuable kinds of character or moral sensibilities, and, as some psychologists think, it is required for effective social norms and the possibility of cooperation and coordination in creatures like us. Such facts can be decisive for keeping moral anger even if it often goes badly. Analogously, if love leads us to heartbreaking debacles and dubious country music, but the alternative was never to have had humans at all, perhaps we should choose country music?Pereboom also considers the persistent worry that we may simply be psychologically unable to cease having retributive moral anger. As evidence that we can abandon retributive moral anger, he points to changes over time in who we have held responsible and what kinds of punishments are permissible. He also appeals to a 1970s study of a group of Inuits who rarely express anger, which he treats as evidence for the possibility that human psychology is malleable enough to relinquish retributive moral anger.Again, though, it is hard to see how these considerations bear on the issue. That we have altered our dietary practices over time doesn’t mean that we can do without eating; that we no longer blame people for being mentally ill and that we no longer express anger in the ways that we did in the past is entirely compatible with moral anger being a fixed feature of certain kinds of relationships, relationships that perhaps we should not and perhaps cannot want to do without. Nor is it germane that we can find a group of people who do not frequently express angry blame. Not expressing angry blame is not the same thing as not blaming. Nor is the fact that one specific community does not express angry blame evidence that nonblaming can scale up and be sustained outside of that very particular historical, economic, and cultural milieu. Pereboom is surely right to invite us to consider whether there are alternatives to our current moral psychologies and what the trade-offs might be. Having more to say on its behalf might bolster the tenability of the proposal.This brings us to a set of questions about the positive proposal that we instead adopt controlled, defensive anger without retribution. Is this an injunction to do something we already do in a wider range of cases? Or is it a proposal for adopting a novel attitude that is available to us but not regularly deployed? Inasmuch as it is an argument for something we already do, the empirical evidence Pereboom offers for our ability to engage in nonretributive measured aggression is the testimony of one military officer and some putative examples from combat sports. In those cases, though, it isn’t clear that what is at stake is moral anger (the examples neither require nor are limited by engagement with a wrongful threat). More clear-cut cases of moral anger—for example, athletes motivated by an opponent’s disrespect, which is not uncommon in sports—suggest something retributive.A more promising case of measured aggression comes in the form of an anecdote of a confrontation between Teddy Roosevelt and some bullies that is settled by the bullies getting whupped and Roosevelt subsequently inviting them to a beer. But here, too, that story doesn’t obviously point to a case of nonretributive moral anger; it is just as easily read as an example of how retributive moral anger has proper limits and the wisdom of reestablishing good will in its wake. Even first-person reports wouldn’t be decisive—recall that Pereboom claims the suppositions of an emotion are not readily available to the one experiencing the emotion. To the extent to which Roosevelt-style examples are moral, they seem readily read as cases of retributive anger. To the extent to which they are measured responses, this, too, seems a hallmark of normatively appealing forms of retributivism.As always, Pereboom is inventive, nuanced, and scrupulously responsive to critics of his views. No book this radical in its aims can hope to secure widespread agreement, but it will undoubtedly be a landmark for future discussions of culpability, moral psychology, hope, and the philosophy of punishment.

中文翻译：

---

错误行为与道德情感


如果没有人承担道德责任，我们该如何应对不当行为？在过去的二十五年里，德克·佩雷布姆（Derk Pereboom）以极大的独创性、严谨性和对对话者的慷慨来解决这个问题。责任怀疑论不再被视为仅仅是一种理论上的可能性，也不再被视为少数历史人物的领域，这要归功于他的努力。在《错误行为与道德情感》一书中，佩雷布姆对当我们拒绝“人们至少有时——并且至少在一个重要的意义上——对自己的行为负有道德责任”这一观点时所剩下的内容提供了一种新的、广泛的解释。我们的责任实践采用了一类不合理的道德愤怒态度（例如怨恨和愤慨），这些态度是报应性的，因为它们反映了这样一种假设，即他们的目标应该遭受痛苦或经历痛苦。在早期的研究中，佩雷布姆认为这种责备应该被对无理不当行为的“道德悲伤”所取代。在这里，他对我们的做法提出了一种不那么情感上的超然、更加人际自信的改革，呼吁采取道德抗议的立场，反对无理的不当行为。这种立场允许对不法行为者表达“有节制的侵略”，一种适度且有控制的防御性“愤怒”（72）。第一章总结了佩雷布姆关于自由意志和道德责任的当之无愧的有影响力的观点，即我们缺乏这样的自由意志：这是我们在“基本应得”意义上承担道德责任所必需的。他认为，深思熟虑和选择通过为认知开放这一普遍事实提供解决方案而得以生存，即使决定论是正确的（25）。第二章和第三章介绍了他对我们的道德实践的建议修订。 第二章提出了一种责备的概念，它并不以他所谓的基本应得为前提，而是基于几个广泛的工具主义或前瞻性考虑（52-53）。第三章阐述了这种有节制的攻击态度的道德心理；佩雷布姆认为，它可以在不预设基本沙漠的情况下支持防御性伤害。其余章节探讨了这种解释对刑事司法（第 4 章）、宽恕的可能性（第 5 章）、爱与关系（第 6 章）以及希望的作用（第 7 章）问题的影响。尽管问题很复杂，涉及的文献范围很广，但讨论始终是容易理解的。 接下来，我将重点关注本书的两个总体主题：反报应主义和我们可能称之为的反愤怒主义。反报应主义认为，“我们可以不报应，无论是为了证明我们对不法行为的反应是合理的，还是在这些反应中所使用的情感方面”(1)。当情绪以应得的痛苦或伤害为前提时，就会产生报应性（3）。反愤怒主义认为，“道德愤怒，无论它是否以基本应得为前提，在我们承担道德责任的实践中占有太突出的地位，并且在这种实践的许多规范性描述中发挥着太核心的作用”（3）。反报应主义取决于确定是什么使某种态度具有报应性，并表明报应性态度不能被证明是正确的，即不能被证明是适当的、合理的或真实的。报应主义的概念越是要求严格，就越难得到证实。要求越低，就越容易证明报应主义是正确的。 佩雷布姆的报应主义明确地援引了应得的痛苦或苦难的概念，尽管报应主义有一些突出但要求不高的特征（例如，在大卫·布林克（David Brink）最近的作品中），只需要剥夺某些反映了报应性质和严重性的权利或商品。应受谴责的不当行为。后一种描述显然并不需要违法者采取相对较强的代理形式，而佩雷布姆认为这是他的报应主义版本所必需的。佩雷布姆相对较少谈论为什么更强烈的观念是正确的。他认为我们的报应性承诺的内容可能不容易从普通思想中辨别出来（35），这一观点可能使问题变得复杂。如果是这样，就不清楚哪些考虑因素有利于更苛刻的报复概念。非错误理论的报应论者可能会反对，认为佩雷布姆已经为反对要求较低且更合理的报应主义形式做好了准备。佩雷布姆对报应主义的否定意味着没有人“基本上应该”遭受痛苦或痛苦。基本应得认为，不法行为者“仅仅因为在采取行动时对其道德地位敏感，而不是出于结果主义或契约主义的考虑”就应该受到谴责或惩罚（12）。然而，佩雷布姆承认，理论上的选择不仅仅是报应性应得或前瞻性指责。道德责任的描述可能依赖于非基本的应得概念。考虑一下有时与罗尔斯相关的想法，即社会实践可以采用惩罚制度——例如犯规和罚款——因为惩罚对实践有积极的影响。 在其中一些情况下，处罚的适用条件取决于犯罪者的行为。也就是说，惩罚的正当性条件可能完全是向后看的，即使惩罚的理由是工具主义的。关于佩雷布姆处理非基本应得的第一个困惑是辩证的：在拒绝基本应得报应主义并承认可能存在非基本观念之后关于沙漠，佩雷布姆对此很少提及，而是专注于“有节制的侵略”的吸引力。然而，考虑到应得的非基本概念并没有被排除在外，而且考虑到它不会涉及我们实践的不太彻底的转变，这种可用但未经探索的选择困扰着他的中心论点。如果存在一种非基本的报应性解释（正如一些人倾向于认为存在的那样），那么如果它不需要道德愤怒心理的根本转变，那么它似乎有不小的优势。 第二个谜题涉及基本应得概念本身的解释利害关系。佩雷布姆声称，对基本应得的检验是康德思想，即即使社会停止了，罪犯也应该受到惩罚（31）。然而非基本的沙漠账户可以通过这个测试。回想一下，我们可以区分（1）有利于具有给定结构的实践的前瞻性系统性考虑与（2）实践的内部性，纯粹向后看的适当性（或应用，或真理或适当性）考虑一阶应得判断。 对于只有满足（规定的）完全向后看的条件才应受惩罚的做法，那么即使一个社会消亡，留下一个孤独的罪犯，罪犯仍然应该受到惩罚，即使没有人在那里惩罚，即使惩罚不会产生任何系统性影响，而这些系统性影响首先证明这种做法是合理的。即使拥有地位的理由消失，地位也可以持续存在。如果这是正确的，那么基本应得似乎在两个不同的领域结合了两种重要的不同思想：个人应受谴责的适当性，内在的实践；以及拥有这样的沙漠基地的基础的外部的、独立的和系统的问题（例如，是否基于契约主义、结果主义或其他理由）。基本应得的吸引力在于它抓住了康德式的直觉，但应得是否必须是基本的这一点并不明显。因此，关于基本应得有一个明显的解释性难题：为什么我们应该接受个人归属基础的观点与单独问题（即系统正当性的基础）的观点之间的纠缠？转向第二个生动的论点在《反愤怒主义》一书中，佩雷布姆认为，无论对基本应得的考虑如何，由于道德愤怒而让人们承担责任的做法“在一般和关键方面都失灵了”（3）。他正确地指出，当人们生气时，他们会歪曲情况的特征，他们容易反抗和寻求羞辱，这会疏远他人并容易导致人们发生对抗。相反，他提倡同情心，并且正如我们所见，提倡有节制的攻击性。愤怒的辩护者应该不为所动。某种态度的表达可能会扭曲实践，这一事实并不足以证明没有这种态度我们会过得更好的说法。当人们相爱时，他们会歪曲情况的特征。爱的态度也会引发对抗、疏远和羞辱。即使大多数责备的情况都很糟糕，大多数关系都以心碎而告终，但这并不意味着我们通过计算成功案例并减去失败案例来正确地决定是保留责任还是爱。这是一个系统性的问题，即我们的态度是否会在整体上变得更好。一个令人信服的答案需要清楚地了解这些在我们的整体生活中扮演什么角色和功能。爱、愤怒等等常常是进入某些类型的实践和特定的存在方式的代价。它们是其他事物的组成部分。它们使我们有充分理由想要的其他商品成为可能。此外，相对无法衡量的愤怒，因为对是否获得与其成本相当的商品漠不关心，具有多种预期功能，包括塑造我们的敏感性。适应愤怒责备的风险对于塑造有价值的性格或道德情感来说是独特有效的，而且正如一些心理学家认为的那样，它是有效的社会规范以及像我们这样的生物合作和协调的可能性所必需的。这些事实对于保持道德愤怒具有决定性作用，即使它常常变得很糟糕。类似地，如果爱情让我们陷入令人心碎的崩溃和可疑的乡村音乐，但另一种选择是根本没有人类，也许我们应该选择乡村音乐？佩雷布姆还考虑到人们一直担心我们可能在心理上无法停止报复性道德愤怒。作为我们可以放弃报应性道德愤怒的证据，他指出随着时间的推移，我们追究责任的人以及允许什么样的惩罚的变化。他还引用了 20 世纪 70 年代对一群很少表达愤怒的因纽特人的研究，他将其视为人类心理具有足够的可塑性以放弃报应性道德愤怒的可能性的证据。不过，同样，很难看出这些考虑因素如何承担。关于这个问题。随着时间的推移，我们改变了饮食习惯，但这并不意味着我们可以不吃东西。我们不再责怪人们患有精神疾病，我们不再以过去的方式表达愤怒，这与道德愤怒完全兼容，道德愤怒是某些类型关系的固定特征，这些关系也许我们不应该，也许离不开。我们能找到一群不经常表达愤怒指责的人也没有密切关系。不表达愤怒的责备与不责备并不是一回事。某个特定群体不表达愤怒的指责这一事实也不能证明不指责可以在特定的历史、经济和文化环境之外扩大并持续下去。佩雷布姆邀请我们考虑当前的道德心理是否有其他选择以及可能的权衡，这无疑是正确的。代表它说更多的话可能会增强该提案的可行性。这给我们带来了一系列关于积极提案的问题，即我们采取有控制的、防御性的愤怒而不进行报复。这是否是一项禁令，要求我们做一些我们已经在更广泛的情况下做的事情？ 或者它是采取一种我们可以使用但不经常部署的新颖态度的建议？由于它是对我们已经做过的事情的论证，佩雷布姆为我们进行非报应性、有节制的攻击的能力提供的经验证据是一位军官的证词和来自格斗运动的一些假定例子。然而，在这些情况下，尚不清楚所面临的问题是否是道德愤怒（这些例子既不需要也不受不当威胁的限制）。更明确的道德愤怒案例——例如，运动员因对手的不尊重而受到激励，这在体育运动中并不罕见——暗示了一些报复性的行为。一个更有希望的有节制攻击的案例是泰迪·罗斯福之间的对抗轶事。还有一些恶霸，他们被殴打，罗斯福随后邀请他们喝啤酒。但在这里，这个故事也没有明显指向非报复性道德愤怒的案例；它就像一个例子一样容易阅读，说明报应性道德愤怒如何有适当的限度，以及在其之后重建善意的智慧。即使是第一人称报告也不是决定性的——回想一下，佩雷布姆声称，经历某种情绪的人不容易获得对某种情绪的假设。就罗斯福式例子的道德程度而言，它们似乎很容易被解读为报复性愤怒的案例。就其反应的程度而言，这似乎也是具有规范吸引力的报应主义形式的一个标志。一如既往，佩雷布姆富有创造力、细致入微，并且对批评他观点的人一丝不苟地做出了回应。 没有一本目标如此激进的书能够获得广泛的认同，但它无疑将成为未来关于罪责、道德心理学、希望和惩罚哲学讨论的里程碑。