Dao Article Discussion: Curzer on Harris

The latest in our series of discussion pieces on recent articles published in Dao, here we have Howard Curzer (Texas Tech), an Aristotle sepcialist who has also developed an interest in early Confucianism, commenting on Thorian Harris’s essay. For Harris’s piece, click here




Harris begins by combining and fleshing out Aristotle’s scattered, elliptical remarks about the sense of shame in an admirably charitable and plausible way….

The sense of shame is a disposition to feel pain at the disapproval of good people caused by disgraceful conduct (325). Like the moral virtues, the sense of shame is a mean disposition. People with the sense of shame attend to the opinions of good people, while the shameless are indifferent to what good people think, and the bashful care about everyone’s opinion (324). The sense of shame is unnecessary for perfect people, but it helps the rest of us avoid wrong actions in two ways. It negatively reinforces bad behavior so that we learn from our errors and are less likely to repeat them. The sense of shame is also forward-looking; we avoid wrongdoing because we fear disapproval. Thus, the sense of shame builds character, and could be used by legislators to encourage people to good actions (328).

[Harris mentions that the sense of shame is not a virtue because even people who lack practical wisdom such as people with merely civic courage can be motivated by concern for what good people will think (326). If the sense of shame merely lacked practical wisdom, then would be a natural virtue. Adding practical wisdom would make it a proper virtue. Aristotle offers a different reason for denying that the sense of shame is a virtue. He says that it is useless for already virtuous people (NE 1128b21-22). However, since virtuous people occasionally act wrongly, I think that Aristotle is mistaken. He should say what Harris says he says.]


Harris finds all of these points within the Confucian tradition, too (329-330). He thinks that Aristotle and the Confucian tradition differ about socioeconomics. According to Harris, Aristotle maintains that people who have not developed a sense of shame by adulthood are out of luck. The window of opportunity closes at the end of childhood (331). Developing a sense of shame takes leisure, so people who must work all of the time are also out of luck (332). These two factors are intertwined. A major reason that people fail to develop the sense of shame during childhood is that they are born into poor families, and must spend their time grubbing for necessities rather than developing the sense of shame and the (other) virtues (332). Thus, Aristotle considers the working class to be morally stunted.

By contrast, Harris maintains that the Confucian tradition allows that badly-raised, overworked people may nevertheless develop the sense of shame. These disadvantages hinder, but do not preclude moral development (332-333).

Harris sides with the Confucian tradition on these points. He endorses its egalitarian view that shame and general moral development is open to all, and criticizes Aristotle for looking down on the masses. Harris says, “Aristotle has rather disparaging things to say about those who are compelled by economic necessity to work as merchants, craftsmen, or wage-earning laborers” (332).

Harris offers an explanation of the divergence between Aristotle and the Confucian tradition. He suggests that the Confucian tradition takes desirable character traits (including the sense of shame) to be available to badly-raised and overworked people because, unlike Aristotle, the Confucian tradition accepts situationism. Aristotle insists that virtuous people behave consistently across situations. Courageous people behave courageously whenever facing physical risk; just people behave justly whenever opportunities for injustice present themselves; and so on. By contrast, the Confucian tradition takes behavior to be relative to situations. When risk is low and theft is hard, everyone behaves well. Conversely, when risk is high and theft is easy, everyone behaves badly (338-340). This is because traits are characteristic of people embedded in social networks rather than isolated individuals (340-341).

Comparative philosophy owes Harris a debt of gratitude for foregrounding and clarifying the Aristotelian and Confucian accounts of the sense of shame, and for pointing out the extent to which these accounts converge. I shall argue that the convergence is even greater than Harris believes. Aristotle is no elitist. He agrees with the Confucian tradition that badly-raised and overworked people can develop the sense of shame. And neither Aristotle nor the Confucian tradition is situationist.


Aristotle thinks that mentally defective people (natural slaves, barbarians (sic), women (sic), and the brutish) cannot develop the sense of shame. The Confucian tradition agrees that some mentally defective people cannot develop a sense of shame. What about mentally healthy people?

King Hui takes famines to be unavoidable natural phenomena. By contrast, Mencius takes droughts to be natural phenomena, but famines to be the avoidable result of inadequate governmental action (Mencius 1A3.1-2). Similarly, elitists take the inequality of virtue (even among the mentally healthy) to be an unavoidable natural phenomenon. “We aristocrats are naturally more virtuous than the masses.” By contrast, liberals take the inequality of virtue (among the mentally healthy) to be the avoidable result of inadequate governmental action. Unchecked, the free market obstructs the moral development of some people, but governments can mitigate this by providing such things as public education, food stamps, and minimum wage laws (so that people need not work multiple jobs in order to survive).

I suggest that Aristotle is a liberal rather than an elitist. Like the Confucian tradition, he acknowledges the existence of mental illness which blocks some people from becoming virtuous, but aside from that, he does not blame inequality of virtue on nature. He says that, “all who are not maimed as regards their potentiality for virtue may win it by a certain kind of study and care” (NE 1099b18-20). Overworked people have little or no opportunity to acquire virtue, and governments should prevent this from happening by arranging for excessive labor to be done by natural slaves and barbarians who will not be disadvantaged by it (Politics 1329a25-26). Aristotle even fantasizes about robots freeing all people from burdensome labor (Politics 1253b33-1254a1). When equalizing opportunity is not feasible, government should at least ameliorate inequality by providing such things as public education (Politics 1337a33ff) and subsidized meals (Politics 1272a12-21).

In general, a charitable reading of Aristotle would take him to sympathize with laborers rather than criticizing them. Just as Aristotle should not be blamed for observing that there are some non-natural slaves, for he laments this fact, and argues that it should be changed (Politics 1255a3-28), so he should not be blamed for observing that badly-raised and overworked people lack the opportunity to acquire virtue, for he laments this fact, and argues that it should be changed.


Harris sketches the Confucian tradition’s two-pronged method for developing the sense of shame. First, provide basic needs to the poor. Lack of basic needs poses a huge obstacle, but it is not insurmountable. A few sages succeed despite destitution (334-336). Second, improve character directly by instruction and example. Good examples are to be provided both by teachers, and by policies which put good people into positions of visibility and authority (336-338). Instruction and example are not panaceas; they help only those who are receptive. But even badly-raised people can gain the sense of shame, and embark on the moral development path.

Aristotle agrees with all of this. He does not say that badly-raised people are incorrigible, but only that they will not benefit from teaching (NE 1095a4-6, 1095b2-7, 1179b23-31). Indeed, Aristotle contrasts badly-raised people with the incorrigible.

The many obey necessity rather than argument, and punishments rather than what is noble. This is why some think that legislators ought to stimulate men to virtue and urge them forward by the motive of the noble, on the assumption that those who have been well advanced by the formation of habits will attend to such influences; and that punishments and penalties should be imposed on those who disobey and are of inferior nature, while the incurably bad should be completely banished. (NE 1180a4-10)

As Harris acknowledges, the Confucian tradition accepts that bad habits eventually become entrenched (332). Aristotle also thinks that repeated vicious acts make people vicious, and the vicious are incorrigible. The vicious are “the incurably bad [who] should be completely banished.” However, Aristotle does not endorse banishment for the badly-raised many. Those who have not “been well advanced by the formation of habits” need “punishments and penalties” because arguments won’t help them, but they are not incorrigible. The point of punishment is not merely to keep badly-raised people in line, but also to improve their character.

Aristotle often cautions that ethics is inexact (NE 1098a26-29, 1103b34-1104a5). He warns that he is speaking “roughly and in outline,” and that his claims are “only for the most part true” (NE 1094b11-22). Thus, although Aristotle claims that the goods of fortune are necessary conditions for virtue, he is not making an exceptionless claim. Instead, a charitable reading would take Aristotle to be agreeing with the Confucian tradition that lack of basic needs is typically, but not always an insurmountable obstacle.


Harris is right to observe that Aristotle is no situationist. Aristotle thinks that virtuous people reliably behave virtuously across a wide range of situations. But he also thinks that virtue is rare (NE 1156b24), so he would agree with the Confucian tradition that most people panic (or steal, or overeat, etc.) in some situations, but fight fiercely (or refrain from theft, or eat moderately, etc.) in other situations. Aristotle thinks that virtuous action is situationally relative (NE 1106a29-b7), so he would agree with the Confucian tradition that virtuous people perform different sorts of acts in different situations. Finally, because Aristotle thinks that people are political animals (NE 1097b8-11, Politics 1253a1-4), he would agree with the Confucian tradition that ethics is about people embedded in social situations rather than about isolated individuals.

Aristotle and the Confucian tradition not only agree on these points, they stand shoulder to shoulder against situationism. Both situationists and virtue ethicists allow that courageous people respond differently to risky situations of different sorts, temperate people respond differently to different sorts of temptations, etc. Situationists go further. They claim that people behave differently not only in situations which differ in morally significant ways, but also in situations which differ in trivial ways. But neither Aristotle nor the Confucian tradition is committed to behavioral differences of this sort.


I have no quarrel with what Harris says about the Confucian tradition, or with what the Confucian tradition says about the sense of shame. My only goal has been to show that Aristotle is even closer to the Confucian tradition, and to the truth than Harris allows.

Hannah Pang detail

13 replies on “Dao Article Discussion: Curzer on Harris”

  1. Many thanks to Howard for these comments! I thought I would throw some ideas out for consideration. Certainly I am also interested in hearing what Thor may think about Howard’s reading of Aristotle.

    I was suprised that Thor does not put more emphasis on the role of ritual in encouraging the transformed dispositions that he discusses. Instead, it seems play a marginal role in his account (see p. 336 on why ritual might be mentioned in 2:3). Seems to me that several people have called attention to the ways in which ritual might function in deeper ways than just keeping us in line. The account in Puett et al, Ritual and Its Consequences (2008) might be particularly friendly to other aspects of Thor’s account.

    In two ways, I think that future development of these ideas would benefit from an engagement with some of the recent work at the borders of philosophy and psychology. First of all, Thor’s account of “emulation” strikes me as too resolutely cognitive and active. People like Hagop Sarkissian have explored ways in which phenomena like “emotional contagion” can help to explain some of the Confucian claims. Second, and more generally, I think that some of the more recent discussions around situationism may help to resist the dichotomy with which Thor presents us: either individualist, robust trait-based, “virtue ethics,” OR virtues/sense of shame as practices by individuals within specific relationships (see his p. 340). For example, take a look at Nancy Snow, Virtue as Social Intelligence, and its reliance on “social-cognitive personality system” theory. I agree with Thor in 340n40 that Eric Hutton’s arguments do not settle any of the issues, but neither do Thor’s rebuttals. I think that much of what Thor says in this essay about dispositions and their relational contexts is very apt. But there is more to be worked out here, to ensure that we avoid a crude “situationism” of the type that Howard worries about.

    Two more things. I was puzzled by Thor’s seeming to translate “de” as “good example.” What’s wrong with, say, virtuosity? Because it’s NOT just about examples of good behavior: it’s about the quality and motivation and style and spontaneity and situational aptness of the behavior…and I take it, about what it is that underlies the consistency with which a given examplary individual acts in these ways.

    Finally, about the relevance of a “pragmatic notion of truth” (p. 335 and elsewhere). Since there’s been a lot of debate over issues in this vicinity over the years, I think we need to be as careful and precise as we can. I see that Thor has a different, forthcoming paper on related issues, so perhaps I will have to wait for that. But I see little evidence that the authors of the Analects — or anyone else in early China — had a sense of a “pragmatic coneption of truth” as opposed to some other “notion of truth.” On the other had, there is a strong case to be made — in some texts at least, with the Analects not always numbered among these texts — that speakers/authors were consciously aware of, and emphasizing, the pragamtics of language use. Is this what you mean, Thor?

    That’s quite a miscellaneous bunch of comments, which I hope might spur discussion of one or another thread!

    I want to thank Howard for his insightful objections. I am delighted to agree with his first objection; the second—concerning the nature of Confucian dispositions—is something that I believe deserves much time and thought. In what follows I can only offer my own initial thoughts on the issue.
    I agree with Howard that Aristotle is not an elitist in the sense specified—that is, Aristotle does not blame inequality of virtue on nature. The inequality of virtue is not a result one’s birth, but of equipage. Thus it follows that I was uncharitable to Aristotle when I described him as having “rather disparaging things to say about those who are compelled by economic necessity to work as merchants, craftsmen, or wage-earning laborers.” It is not those who are driven to live these lives, but rather the existence of such lives, that seems to be Aristotle’s true object of critique. I thank Howard for pointing this out.
    I wish to maintain a distinction between the Aristotelian and Confucian conceptions of praise-worthy dispositions. While a number of the Aristotelian virtues are other-regarding and may even require the participation of another person to be realized, this cannot be said of every Aristotelian virtue (temperance, courage, and wisdom come to mind as likely candidate for virtue that need not be other-regarding, nor require another’s participation). As far as our character is concerned, Aristotle would certainly agree that it depends upon others for its initial formation; but Aristotle would seem to support the idea that we can eventually establish a set of character traits that, for better or worse, resist the impact of others and are thus unaffected by our relationships. This is where the similarity between Aristotle’s description of us as “social animals” and the idea of relational selves comes to an end: for Aristotle, the other is necessary as a means perhaps, but we might come to have our character traits in isolation; a relational self—which I ascribed to the Confucian conception of praiseworthy dispositions—would have dispositions not simply cultivated by others but realized only with others.
    Let us take xiao as our example of a praiseworthy disposition. If this is a virtue, it is an other-regarding virtue. It is, however, a virtue that will need to attend to not only the specific role we have with the other, but also the other’s particularities. Furthermore, xiao, interestingly enough, seems to come in degrees (if only by comparison with the xiao of others). Perhaps it is a matter of reputation, or comparative exemplification, but the xiao of Shun takes the cake—and to do so, in part, because of his family members. His xiao is so easy to see (can we say ming ming de 明明德?) because of the depravity of his family members, because of Shun’s ability to take care of them and avoid being killed by them, but also because of how he transforms them. In other words, the xiao of Shun is a matter of his conduct, but it is also a matter of his family’s conduct—agent and patient are both significant contributors to the disposition. Another way of approaching the matter might be to look at the practices of praise and blame. On this note Shun’s practice (a sagely one at that) of blaming himself for (what we would want to describe as) the faults of his family members also supports a relational reading of dispositions. When Yao began considering Shun as a successor, he went about giving Shun a series of tasks (one of which involved marrying Yao’s daughters and living with Yao’s useless sons). Arguably Yao was testing Shun—trying to get a sense of Shun’s dispositions. But can we not also say that Yao was cultivating Shun—providing him with situations that would sustain the relevant relational dispositions?
    I agree with Howard that we should be cautious in describing Confucianism as endorsing situationism. The greatest danger in making this claim is that it might make us blind to the person’s own contribution to any given social situation—we run the risk of ignoring the person’s own impulses and the interactions of their numerous relational dispositions. However, I think that there are at least two things to commend the association. First, it can act as a corrective on our assumptions about individual agency, and thus bring us closer to the Confucian view of things. Second, from a pragmatic point of view—and this was my point of departure in this paper—situationism, informed by a Confucian account of the salient features of our social context, is likely to produce better results when it is endorsed by those who seek to foster a sense of shame.
    At NE 1128b21-22 Aristotle states that “shame is not even characteristic of a good man, since it is consequent on bad actions.” I believe Aristotle is right—shame should not be experienced by good persons (or at least not very often) because good persons should make few mistakes. However, a sense of shame—which Aristotle is not talking about in this passage—can be said to be a quality of good persons and does not occur only after a mistake is made. Much controversy exists over Aristotle’s claims about shame in the Nicomachean Ethics—controversy that could be overcome if we simply distinguished, as Aristotle does, between “shame” and the “sense of shame.”


    All virtues involve imagination to some degree. At the very least, virtuous people must imagine the likely outcomes of their choices in order to determine which choice is best.

    But the sense of shame is particularly imagination-intensive. The forward-looking aspect of the sense of shame (which I take to be the more important aspect) requires one to imagine (a) what others would think about one’s choices, and (b) how one would feel about the way others would think about one’s choices. That is, a person with the sense of shame (a) puts himself/herself in the shoes of various other people, and looks at himself/herself through their eyes. Then he/she takes the further step of (b) imagining his/her possibly painful reaction to the reactions of others. (Sartre describes this in excruciating length and detail in BEING AND NOTHINGNESS.) Of course, a person with the sense of shame is correctly affected by the judgments of the right people, at the right times, etc. But there is not much else to the sense of shame besides imagination. In particular, there are no actions peculiar to the sense of shame, for it just helps agents to perform courageous acts, temperate acts, liberal acts, etc.

    In light of the centrality of imagination to the sense of shame, I don’t see how ritual could help agents develop it. While ritual is useful in inculcating and maintaining social practices, I don’t see how it produces the sort of imagination required by the sense of shame. I can’t IMAGINE a mechanism linking ritual with imagination.


    The view that each person is a relational self doesn’t appeal to me. Let’s come at the notion of a relational entity from a different, and rather speculative direction. I do think that there are relational entities which can have character traits including virtues and vices. These are organizations ranging from small (e.g. families, committees, communities) to huge (e.g. corporations, NGO’s, nation states).

    Each of these organizations can make choices which are different from, and not straightforwardly reducible to the choices of its individual members/components. (The Citizens United decision gets this right.) Sometimes this happens through a compromise with which no one is fully satisfied; other times it happens through a sort of mob psychology in which everyone comes to see things through the same distorting lens. Presumably, there are yet other mechanisms. The nature of the organization (e.g. its organizational memory, its dysfunctional internal relationships, its mission statement, its corporate culture) often enables observers to predict its choices, and even the way it arrives at these choices. In other words, organizations have dispositions.

    Since organizations can make choices and have dispositions, both their choices and their dispositions can be morally admirable or horrible. Thus, it makes sense to talk about the moral and immoral acts of organizations, and the virtues and vices of organizations. Of course the character traits of organizations affect, and are affected by the character traits of the individuals which constitute them. But again they are not straightforwardly reducible to the character traits of their component individuals.

    Now I suggest that shame is a peculiarly individual passion; it is not the sort of emotion that organizations can have. Individuals which make up organizations can be ashamed of belonging to the organization, but the organization cannot be ashamed of its actions or nature. Thus, the sense of shame is not the sort of character trait that organizations can have.

    (Organizations can worry about principled boycotts, but this is not prospective shame, for it does not have moral overtones. The worry is not that the organization itself is morally flawed, but rather that its policies might lead to bad consequences.)

    If organizations cannot feel shame, and thus cannot have the sense of shame, then we should be reluctant to think of people as having relational selves. For people clearly can feel shame and acquire the sense of shame.


    One interpretation of Aristotle’s rather obscure argument in NE IX takes Aristotle’s claim to be that virtuous people need friends in order to exercise and preserve their virtue. If this interpretation is correct, then Aristotle thinks that our relationships with other people are necessary for maintaining, as well as developing virtue.

    • Howard—I’m curious to hear more regarding these last points. Why can’t organizations feel shame? Can they feel outrage? Can they be angry? If they can worry (which you grant) then it doesn’t seem a stretch to say that they can imagine “(a) what others would think about the organization’s choices, and (b) how the organization would feel about the way others would think about one’s choices.” (There’s an excellent literature on the general topic of mental states of collectives that has developed in the past few years, driven by experimental data, and largely conducted by philosophers.)

      I’m also not clear about the argument by analogy you present. Why can’t a relational self feel shame—say, shame in the context of a particular relationship and not another? I suppose we’d get bogged down trying to hash out what it means to be a relational self, but even on Thorian’s account (to keep things simple) I don’t see how it precludes shame. If having a sense of shame requires (a) and (b), then why not think that the relevant relationships are those with the ‘others’ imagined in the process of (a) and (b)? I feel as though I’m just missing something (I probably am).


      Hagop: Thanks for inviting me to expand on my speculative and preliminary (i.e. half-baked) ideas about the self. I think it is crucial to distinguish between a relational self and a self which includes relations. Everyone accepts that selves include relations, so in order to be interestingly different, a relational self must involve a significantly deeper, broader unity.

      Consider this example. I have been a member of the TTU Philosophy Department for 30 years, and it is very important to me – one of my ground projects. I would be a different person if I left that department. This does not make my self a relational self, however. For my self to be relational, it would somehow have to include the other members of the department, and not just my membership in the department and my friendships with its members. At least sometimes, WE would have to act, feel, think, and perceive as a unity. Some sort of group-think and group-feel (without the pejorative connotations), as well as some sort of collective action and responsibility are essential components of a relational self.

      Suppose the department faces a funding shortfall. To say that WE are anxious about it doesn’t mean each of us (individually) is anxious, or the majority of us are anxious, or even that ANY of us are anxious. However, in department meetings WE might compose a pleading letter to the Dean, engage in prolonged discussions about renaming our courses to increase enrollment, etc. But more importantly, when we are together (or thinking of ourselves as together, even if we are not in communication), there is a pervasive mood of anxiety. Concern is “in the air.” Pre-meeting jokes take a morbid turn, post-meeting post-mortems are grim, etc. But perhaps none of us are individually anxious. We are not losing sleep, filling Xanax prescriptions, or even feeling tense.

      ASIDE: I think I am describing a common phenomenon, so I hope you know what I mean. However, I find it hard to articulate what I mean without sounding like a New Age guru or pop-Heideggerian.

      Using this notion of collective thinking, feeling, perceiving, acting, etc., one might build a notion of character traits for relational selves. The idea that character traits of organizations are not reducible to the traits of individuals is hardly new. In the REPUBLIC, Plato is willing to attribute justice to cities on the basis of their class-relationships, even if almost no one in the city is just. I am suggesting a different, more phenomenological basis for attribution of character traits to relational selves.

      Since I allow that organizations and relational selves can feel passions and even have character traits which include dispositions to feel passions, why not allow that they can feel shame and develop the sense of shame? Well, I don’t have a good argument, just an inability to picture collective shame. When I try to tell a story about collective shame, it morphs into something else.

      Suppose one of my colleagues does something shameful. I might feel bad for him, or feel the same thing he feels, or feel ashamed to be associated with him, but these are not cases of a relational self feeling collective shame. To be collective shame, WE must feel shame together, a single feeling occurring in and between two bodies. But do such feelings actually occur?

      Suppose the Philosophy Dept. is concerned about dropping enrollment, and we discover that our average grade in undergrad Philosophy classes is significantly higher than the average grade in undergrad English, History, and Math, classes. In a faculty meeting, we consider inflating our grades in order to boost enrollment. (a) I can imagine someone objecting that such a policy would be unjust; we would be allocating goods to those who do not deserve them. (b) I can imagine someone objecting that such a policy would be imprudent; people and units around the university would change their policies in ways that would hurt the department. For example, the Honors program would stop sending us good students. (c) I can imagine the individual members of the department feeling prospective shame at the thought of being associated with such a policy. (d) But I cannot imagine an objection based on collective shame. I cannot imagine anyone saying, “Virtuous people and organizations would think we are acting unjustly, and our collective awareness of their opinion would be collectively painful, so we should not inflate our grades.”

      Now maybe I can’t imagine this because I lack imagination. I haven’t read any of the literature on mental states of collectives that Hagop mentions. But I suggest that shame is just too complex an emotion to be felt by an organization. Collective anxiety, anger, envy, empathy, pride, and many other passions seem possible because they are relatively simple, but shame requires the difficult feat of imagining how one would react, to how others would feel, if one did a certain thing. Perhaps the Borg could pull this off, but it seems to me that organizations built of looser connections just can’t do it.

    • Thanks, Howard. This is helpful. Let me just press this point:

      (d) But I cannot imagine an objection based on collective shame. I cannot imagine anyone saying, “Virtuous people and organizations would think we are acting unjustly, and our collective awareness of their opinion would be collectively painful, so we should not inflate our grades.”

      I suppose I don’t see the stark contrast. To be honest, I think you do a much better job of “selling” the anxiety case by discussing all the behavioral signals that might accompany it. Why not think the same could happen for collective shame? Before making the decision I can imagine someone saying “It would be shameful for the department,” and once the decision is made there might be signals: Perhaps people will be more reticent when representing the department, or the mood at a meeting will be tentative and withdrawn. Maybe certain activities will be postponed or cancelled. The triggering event (the behavior of some faculty, or a scandal involving a student) might affect the collective in such a way.

      Full disclosure: I don’t have a horse in this race. However, I do find what’s intuitive or not to often be shaky grounds for individuating emotions (even collective ones), and whether one finds something intuitive or not will often be hostage to simple things such as framing effects or past experience, and nothing inherent in the phenomenon under consideration.

    • I agree that intuitions are shaky grounds, and as I mentioned, I don’t have anything better to support my claim that collective shame does not occur. But maybe if I elaborate a bit, my intuition will come to seem more plausible.

      It is important to distinguish between prospective, collective SHAME and prospective, collective GUILT. Guilt requires: (a) recognizing that the act we are contemplating is wrong, and (b) predicting that if we perform the act, then we will feel guilty (a certain sort of psychological pain). Shame requires: (1) predicting that if we perform the act, others whom we respect would think that the act is wrong, and (2) predicting that if these others think the act is wrong, then we will feel ashamed (a different sort of psychological pain). Guilt requires making a judgment about (a) right and wrong, and about (b) how we will react to our judgment. Shame requires making a judgment about (1) other peoples’ judgments about right and wrong, and about (2) how we will react to their judgment. Guilt is personal; shame is interpersonal.

      At least to me, reacting to other people’s judgments seems much more complex than reacting to my own judgment. Shame requires adopting another person’s perspective, which is much more difficult than making my own judgment, because other people are PEOPLE rather than abstractions. I don’t mean merely that it is hard for me, as an Aristotelian, to determine what Anne will think because she is more-or-less a Kantian. I also have to keep in mind that Anne was raised by an archeologist and a community activist, which gives her a certain slant on various issues. And she has a history of conflict with certain people, and working well with others about certain matters. And she is a physician which has given her an unusual window into the inner lives of her patients, creating unusual sorts of both sympathy and skepticism for people in general. And so on. Before I can feel shame about how Anne will see my act, I have to determine how Anne will see my act, and that is hard.

      Shame also requires reacting to other people, which is more difficult than reacting to any judgment because I am related to those other people in concrete ways. They have a history and a future with me. They have bodies. They have relationships with third parties. They occupy roles. And so on. For example, if I am thinking of doing something wrong, when I think about how I would react to Anne’s disapproval, the fact that Anne is my wife, that we have had arguments about similar things, that we just shared a nice dinner, that we are planning to go to a concert next weekend, that Anne might tell our daughter about my wrongdoing, that Anne is a beautiful woman, etc. are all inseparable from how I would react to her disapproval.

      My hope is that elaborating this distinction between guilt and shame will clarify why COLLECTIVE shame seems implausible to me. While I can imagine collective guilt, and individual shame, I just can’t imagine a group assembling, integrating, and applying all of the facts about other people necessary in order to feel shame. Think of how long a committee meeting would have to be in order to accomplish this task.

  5. Some quick thoughts:

    I like Thorian’s use of Sunzi and Han Fei to tease out the widespread acknowledgment of situation-based virtues in the classical period. I don’t think the Confucians emphasize familial relationships more than ‘terrain and public attention’. The Analects and Xunzi clearly have examples of the latter (and the Hutton piece cited by Thorian presents nice bits of text from the Xunzi in this regard).

    I also would like to hear more about how the fact that certain traits are cultivated in certain relationships (like a sense of shame) necessarily tethers them to those relationships. Take de for example. The ruler’s de (or the nobleman’s de) seems to be something that radiates out of his person and influences and shapes the behavior of those below them in the social hierarchy. This seems to render de largely independent of any particular relationship. (The only salient qualifier is that one has de with regard to one’s inferiors, but the particulars of the inferiors seems less important. Indeed, for the ruler, the qualifier would still include the set of all people under Heaven.) In this sense, it seems to be an individualistic trait, and has the type of consistency (in Doris’s sense) to be more like an Aristotelian virtue (even if de is the result of cultivation that takes place across and within relationships). (Mencius seems to be the strongest example of individualistic virtue in this sense.) Do you think there is room on your account for more individualistic conceptions of traits such as de?

  6. Angle: I take “emulation” to cover a wide range of phenomena—the emotion of emulation (which will be cognitive in part) as well as the practice of “intelligent imitation”; but I see no reason why it might not also include non-cognitive and ‘less active’ aspects as well (I’m not quite sure I know what you mean by “active” in that context). As for the “pragmatic notion of truth”—yes, the language might suggest an anachronism; that was not intended.

    Howard: I like the idea of using our understanding of groups, such as philosophy departments or perhaps even corporations, in our attempt to think through what a relational self might involve. The problem, however, is that thinking of such “groups” as nothing more than a collection of individuals—as wholes composed of parts—would be to presuppose a non-relational conception of self, while a relation conception of self would not think of families, departments, or corporations as collections of individuals but as overlapping relational centers (here I am thinking of Fei Xiaotong’s sociological distinction between “organizational modes of association” [tuantigeju] and “differential modes of association” [chaxugeju]).

  7. Let me add another angle. I’ve been trying to use organizations as a model for thinking about relational-selves. My line of thought has been something like this. Shame is a virtue. Organizations can’t feel shame. Therefore relational-selves can’t feel shame. This is a strike against the view that the self is relational.

    But this line of thought does not take seriously enough the idea that shame is a learner’s virtue. Learners are not fully-formed selves seeking an infusion of morality. Rather they are partially-formed selves. Moral development is a process of self-development. Maybe shame’s role is to bind the partially-formed self to other people, CREATING the relational-self. As I feel ashamed before Anne, I am bound a bit more closely to her, and that is one of many steps toward forming a relational-self which will eventually incorporate both of us. If this is right, the problem with my earlier line of thought is that organizations are not good models for relational-selves.

  8. Thorian pushes my button when he writes on p.332,

    Confucius, for his part, acknowledges the tremendous impact one’s upbringing has upon the ethical quality of one’s relationships with others—including one’s relationship with political authorities (Analects 1.2).

    No other passage is cited for this point. But Analects 1.2 does not touch on upbringing; nor I think is it particularly about childhood. (Nor is it about the quality of one’s relationships as distinct from the qualities of one’s approach to certain relationships.)

    Further, Analects 1.2 claims to report the words of Youzi, not Confucius.

    Although Youzi’s explanatory discussions in 1.2 and 1.12 have been main seed-crystals of most Chinese and Western interpretations of Confucius, nevertheless Analects 1.2 is not evidence of Confucius’ views. Youzi’s views as reported in 1.2 are markedly out of harmony with the remarks the Analects actually does attribute to Confucius. The preponderance of the ancient evidence supports the view that Youzi was not an associate of Confucius at all, and indeed that Confucius was not aware of the ideas of 1.2. (I defend these points in PEW Oct. 2008.)

  9. Howard Curzer writes,

    [Harris mentions that the sense of shame is not a virtue because even people who lack practical wisdom such as people with merely civic courage can be motivated by concern for what good people will think (326). If the sense of shame merely lacked practical wisdom, then would be a natural virtue. Adding practical wisdom would make it a proper virtue. Aristotle offers a different reason for denying that the sense of shame is a virtue. He says that it is useless for already virtuous people (NE 1128b21-22). However, since virtuous people occasionally act wrongly, I think that Aristotle is mistaken. He should say what Harris says he says.]

    In that passage, Aristotle argues from the premise that virtuous people do not act wrongly. So I suppose he’s talking about a theoretical extreme case, so I think the question should be whether as real cases approach that, the value of a disposition to shame decreases faster.

    I imagine Aristotle’s picture must be something like this: the feeling of shame naturally tends to follow upon one’s noticing or expecting the disapproval of others. (Aristotle there says shame amounts to a “fear of dishonor,” i.e. of disrepute.) A moral advance would be for me to feel ashamed only at the disapproval of good others (i.e., those I take to be good). Now insofar as I am good, I understand and am therefore attracted by goodness. It would seem that this latter disposition can occasionally generate feelings at cross-purposes with the wish that people I think good not think I am acting badly. The former trait would seem capable of generating finer-grained preferences that more closely track what’s good. Where the two traits agree, the second is arguably superfluous, and where they disagree the second seems regrettable. So it could be that for very good people, the distraction of a disposition to shame would be greater than its guidance value.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.