Agent-focused Moral Realism Defended: Responses to my Critics

Agent-focused Moral Realism Defended: Responses to my Critics Yong Huang The Chinese University of Hong Kong 1 Introduction Moral realism, as a metaethical theory, arises from philosophical reflections on one of the most fundamental issues, if not the most fundamental one, of normative ethics: objectivity of moral properties or facts. Until recently, normative ethical theories dominating modern Western philosophical discourse have been consequentialism and deontology, both of which are primarily concerned about moral properties of rightness and wrongness of actions. Understandably, thus, moral realism has been also action-focused, aiming to show the objectivity of these moral properties, and classical criticisms of moral realism have been, also understandably, largely directed to this action-focused moral realism. However, in the last a few decades, virtue ethics as a normative theory, which is primarily concerned with the goodness and badness of human persons, has experienced an impressive revival and become a powerful rival to deontology and consequentialism. Unfortunately, however, most of our metaethical discussions, including the debate between moral realism and anti-realism, are lagging behind, failing to reflect this fundamental shift of the scene in normative ethics. It is in this context, as a virtual ethicist in normative ethics, someone who thinks that virtue ethics is a more plausible normative theory, that I’m motivated to develop an agent-focused moral realism, reflecting on issues arising from virtue ethics, to argue for the objectivity of moral properties of goodness and badness of persons, heavily drawing on, indeed mostly explicating, the view of the neo-Confucian philosopher Zhu Xi. As it is a novel approach, of course, I expect and indeed welcome criticisms from my fellow meta-ethicists who are commenting on my paper, to whom I’m most grateful and to whose comments I am most happy to make the following responses. 2 Being Healthy and Being Good: Response to Professor Copp In his very sympathetic reading of my paper, Professor David Copp points out what he takes to be “certain gaps” in my (presentation of Zhu Xi’s) argument, “which create some lack of clarity as to whether Zhu Xi’s theory is genuinely a kind of moral realism and, even if it is an example of moral realism, whether it is relevantly naturalistic.” Indeed, most of the gaps perceived by Professor Copp point toward the same issue: “good” in its attributive use is not an ethically normative good. In Professor Copp’s view, “X is a good human being,” where “good” is an attributive adjective, can only show that X is an excellent example of human kind but doesn’t show that X is a morally excellent member of the kind. He provides two counter-examples: (1) a good example of as assassin is not a morally good assassin; (2) A Martian zoo keeper, who would want an excellent example of the human kind would not necessarily pick a morally good person for its collection. Here, Professor Copp seems to think that moral excellence or moral good is not human excellence or good or, rather, human excellence or good doesn’t include moral excellence or moral good, so that someone who is good as a human being can be morally bad. This may be the case if what makes humans human, what separates human beings from non-human beings, i.e., the human nature, doesn’t include what we consider to be moral components. For example, an exemplar of tiger, a good and excellent tiger, is not a morally excellent member of the kind of tigers; indeed, we would be misusing the term “morally” here, since there is no moral component in our conception of the nature of tiger. Zhu Xi, however, argues that human nature consists of benevolence, rightness, propriety, and wisdom, which are normally considered to be moral or ethical. We may try to show that Zhu Xi’s argument is not successful,1 but given his argument, an excellent example of human kind must also be an ethically excellent member of the species, and so the Martian zoo keeper, who wants to have an excellent example of the human kind, would fail to do his job if he picks a member of human kind who is not ethically excellent, as no excellent example of the human kind, in Zhu Xi’s conception, is not ethically excellent. Of course, on the one hand, as we have seen, not any excellent example of any species is an ethically excellent member of that species, as “ethical” or “moral” only apply to the human species, and so an excellent example of oak tree is not an ethically excellent member of the species of oak tree. On the other hand, some human groups, by their natures, consists of only ethically deficient human beings and so an excellent member of a group of such a nature cannot be an excellent member of human species and thus cannot be an ethically good member of that species. The social group of assassins is a group of such a nature, and thus no excellent members of this social group, i.e., no “excellent” assassins, can be excellent examples of human specials and thus also ethically excellent members of that species. There can be no ethically good assassins, just as there cannot be virtue, as the mean between two extremes in emotions and actions, regarding how much to steal, since any degree or amount of stealing already falls to one of the extremes and is thus already a vice, as Aristotle would say. This same point comes back when Professor Copp discusses my analogy between being good and being healthy. Indeed, in my discussion, health and goodness are synonymous. So strictly speaking, the analogy I’m drawing is not between goodness and health but between ethical health/goodness and physical goodness/health, both of which are parts of one’s overall health/goodness, which may or may not contain other parts, depending upon how you divide one’s overall health or goodness.2 Rosalind Hursthouse makes this point clear when she says that, when we evaluate someone as an ethically healthy/good or unhealthy/bad human person, we are only paying attention to this part of the person’s overall health/goodness, with its physical parts hived off into human biology and/or medicine (Hursthouse 1999: 206-207). In Zhu Xi, the ethical aspect of one’s overall health/goodness consists of such characteristics as benevolence, rightness, propriety, and wisdom. Professor Copp thinks that Zhu Xi needs to show that these characteristics are “morally good, or are human virtues.” To do so, Zhu Xi should not only make the case that they are natural states or properties, which Zhu Xi does, but “he also needs to show that the fact that these characteristics are morally good or are virtues is a natural fact.” For Zhu Xi, however, he has already shown that these characteristics are ones that, at least partially, make one a good or healthy human being, which is a natural fact, just as it is a natural fact that possession of the characteristic human sight makes one, partially, a good human being. Since this aspect of health/goodness that 1 Professor Copp does also identify a gap in Zhu Xi’s argument in this regard. For Zhu Xi, only immediate and spontaneous emotional responses can reveal human nature, and Professor Copp counters: “Suppose I spontaneously felt envious of a friend, but later overcame my envy and felt happy for my friend’s good fortune, it is doubtful that this would be evidence that envy is characteristic of human nature but that friendly fellow feeling is not.” However, for Zhu Xi, additional to being immediate and spontaneous, emotional response can reveal human nature only if they are also shared by all human beings, as I emphasize in my original paper. Cleary the type of envy Professor Copp mentions is not shared by all human beings. 2 For example, we may add mental goodness/health to the ethical and physical ones. Of course, if one is a physicalist, holding that the mental is nothing but physical or even is entirely supervenient upon the physical, then one’s mental goodness/health is only part of one’s physical goodness/health. these characteristics bring to a particular human being is not physical goodness/health but falls into the realm of what we normally consider to be the ethical or moral realm, it is entirely legitimate to regard them as “morally good”. For the same reason, since human virtues are characteristics that make one a healthy/good human being, and the four human characteristics that Zhu Xi discusses, benevolence, rightness, propriety, and wisdom, are ones that make one primarily not a physically good/healthy human being but an ethically/morally healthy/good human being, it is entirely legitimate to regard them as moral/ethical virtues. Still, Professor Copp sees a disanalogy between physical goodness/health and ethical health/goodness: “there is a good reason for anyone to choose to be a[n ethically healthy/] good person,” but “it is not necessarily the case that there is good reason for a person to choose to be [physically] healthy[/good].” For example, Professor Copp says, there are circumstances in which suicide is rational, and in such circumstances “a person might have no reason to choose to be [physically] healthy,” but “there are no circumstances in which a person would have no reason to choose to be a[n ethically healthy or] good person.” My view is that there is no such distinction between the ethical health/goodness and physical goodness/goodness as Professor Copp conceives. Absent of any dilemmas, just as it is necessarily the case that there is good reason for a person to choose to be ethically healthy/good, it is also necessarily the case that there is good reason for a person to choose to be physically good/healthy. Indeed, we may put it more strongly: Absent of any dilemmas, just as it is necessarily the case that there is no good reason for a person to choose to be ethically unhealthy/bad; it is necessarily the case that there is no good reason for a person to choose to be physically bad/unhealthy. With dilemmas, however, just as there may be good reasons for one to choose to be physically bad/unhealthy, there may also be good reasons to choose to by ethically unhealthy (bad). Assuming that one’s overall health or goodness consists solely of the ethical one and the physical one. The dilemma can take place (1) between different aspects of the physical goodness/health, (2) between different aspects of the ethical health/goodness, and (3) between one’s ethical health/goodness and one’s physical goodness/health. When any of these dilemmas indeed arises, one has no way but to sacrifice the health/goodness of one part for the sake of the goodness/health of another part. Whichever way one takes to resolve such a dilemma, however, one always have good reasons to choose the way to maximize (and one never has good reason to choose to minimize) one’s overall (physical or ethical) health or goodness. In situation (1), one may have good reason to choose to be bad/unhealthy (or at least no good reason to choose to be good/healthy) with one part of their body, but this is because they have good reason to choose to be good/healthy (or at least they don’t have good reasons to choose to be bad/unhealthy) with another part of their body, and the reason is good because by doing so they maximize their overall physical goodness/health or minimize the harm to their overall physical goodness/health. For example, when there is a dilemma between saving one’s toe and saving one’s foot, one may have good reasons to choose to be bad/unhealthy with the toe, and the reason is good because by doing so they choose to be good/healthy with the foot and thus with their body overall. So unless in situation (3), contrary to what Professor Copp says, there is indeed always good reason to choose to be physically healthy/good overall. In situation (2), where we encounter moral dilemmas, one may have good reason to choose to be unhealthy/bad (or at least have no good reason to choose to be healthy/good) with one aspect of their ethical life, but this is because they have good reason to choose to be healthy/good (or at least because they have no good reason choose to be unhealthy/bad) with another aspect of their ethical life, and the reason is good because by doing so they choose to maximize the overall health/goodness of their ethical life or minimize its harm. For example, when encountering the dilemma between keeping a promise and saving a life, one may have good reason to choose to be unhealthy/bad (or at least don’t have good reason to choose to be healthy/good) regarding the former: to break a promise), but this is because by doing so they choose to be healthy/good with regarding the aspect of their overall ethical life: to save a life, and the reason is good because in doing so they choose to maximize their overall ethical health/goodness or minimize its hearm. So unless in situation (3), consistent with what Professor Copps says, there is indeed always good reason to choose to be ethically healthy/good. Now in situation (3), where there is a conflict between one’s ethical health and physical health, generally speaking, one has good reason to choose to be physically bad/healthy in order to maintain one’s ethical health/goodness and thus promote one’s overall health/goodness. This is because, at least in Confucianism in general and in Zhu Xi in particular, one’s ethical life is what distinguishes human life from other life, and so when there is a conflict between one’s ethical life and one’s physical life, the former trumps the latter. It is in this sense that, for Zhu Xi, “everyone necessarily has reason to choose to have these characteristics,” i.e., benevolence, rightness, propriety, and wisdom, as he regards such characteristics as constitutive of human health/goodness. However, even in Confucianism this is not absolute or unconditional. While human life is human life and so one may even choose to sacrifice one’s whole physical life for the sake of being ethical, human life is also human life and so in some situations one may choose to make some small sacrifice of their ethical health/goodness for the sake of some significant aspect(s) of their great goodness/health. Mencius mentions one such scenario. Generally speaking, to observe moral propriety, which is related to one’s ethical health/goodness, is more important than eating, which is related to one’s physical goodness/health. However, what happens if one would starve to death by refraining from eating in order to observe a relatively insignificant aspect of moral propriety but would live by taking food with an insignificant violation, or a violation of an insignificant aspect, of moral impropriety? Mencius answers by using an analogy: “gold is heavier than feather, but this does not mean that a small piece of gold is heavier than a whole cartload of feathers. If you compare a case in which there is a conflict between eating, which is extremely important, and observance of moral propriety, which is insignificant, then of course eating is more important” (Mencius 6A1). In summary, in all the above three cases where we are encountering dilemmas, just in all cases where no such dilemmas exist, we always have good reasons to choose to be healthy/good (or at least we don’t have good reason to choose to be unhealthy/bad) in our overall life, which includes both our overall physical life and our overall ethical life. The only difference between situations with dilemma and situations without dilemma is this. In the latter, one always has good reasons to choose to be healthy/good (and one neve has good reasons to choose to be unhealthy/bad) not only with one’s overall life but also with any of its components. In the former, however, while one still always has good reasons to choose to be healthy/good (and one still has no good reasons to choose to be unhealthy/bad) with one’s overall life, one may have good reason to choose to be unhealthy/bad (or one may have no good reason to choose to be healthy/good) with some component of their overall life. So when Professor Copp says that “a person might have no reason to choose to be [physically] healthy,” this person must either encounters a conflict between different parts of their physical goodness/health (situation [2] above) or they may encounter a conflict between their physical goodness/health and their ethical health/goodness (situation [3] above). Independent of either of the two situations, the person always has good reason to choose to be physically good/healthy and never has good reason to choose to be physically bad/unhealthy.3 3 Innatism or Non-Innatism?: Response to Walker Matthew D. Walker, in his comments, asks two questions: “(1) Do we have good reason to reject Aristotelian naturalism? (2) Do we have good reason to hold that the virtues are objective in the way that Zhu Xi suggests, viz., as strongly constitutive of human nature?” In the process, he gives negative answers to both questions. I’ll start with the second one. Walker raises the second question by contrasting what he considers to be Zhu Xi’s view and Mencius’s view. As a matter of fact, though, they are merely two different interpretations of Mencius’s view in the commentary history, the interpretation Zhu Xi takes and the interpretation Walker favors. Mencius regards our emotional responses as duan 端 of our human nature. This character has two basic meanings. One is sign or clue or hint. For example, a bud growing above ground can be seen as duan in this sense, a sign for, a hint about, or a clue to, the root underground. This is the interpretation that Zhu Xi takes. So when Mencius says that commiseration is the duan of benevolence, Zhu Xi understands Mencius to be saying that this emotional response indicates that there is benevolence in our human nature. The other meaning of duan is the beginning of something that is yet to be developed. So the same saying of Mencius can be understood to mean that the emotional response of commiseration is the beginning of benevolence, i.e., not benevolence itself but can be developed into benevolence. This is the interpretation that Walker favors. Walker’s point is that his favored interpretation of Mencius’s view is not just a better interpretation to take, which I shall leave aside, but that it is a philosophically more sound position to hold, which I would dispute, at least with our immediate concerns here. Given this, Walker characterizes Zhu Xi as a latent virtue innatist, who “maintains that we possess virtue innately, but that we do not possess this virtue in a precarnate state. Instead, this virtue lies within us from birth, in latency, waiting to be manifested”, and Mencius as a dispositional innatist, who “denies that we already possess latent virtue. Instead, dispositional innatism holds that we possess only predispositions to acquire virtue”. After distinguishing these two views this way, Walker poses three sub-questions to me: “(1) Can Zhu Xi offer any argument for inferring latent virtue innatism instead of dispositional innatism from the child and the well case (and related cases)? (2) If we can infer only dispositional innatism from such cases, can a viable agentfocused moral realism still go through? (3) If Mengzi’s dispositional innatism suffices to ground agent-focused moral realism, are there other, independent, philosophical benefits of accepting Zhu Xi’s apparently bolder latent virtue innatism instead?” My simple answer to the first sub-question is “yes”, and thus the other two questions Walker raises become simply moot. So let me explain my answer to the first sub-question. Walker and I agree that both Zhu Xi and his version of Mencius are trying to provide some explanation to our common emotional responses to certain situations by appealing to our human nature. So let’s use the example of the emotional response, commiseration, that everyone will have upon seeing a child about to fall into a well that Mencius creates and Zhu Xi adopts. Why do all human beings have such an emotional response in such a situation? Here I highlight “everyone” that both Mencius and Zhu Xi emphasize. That means 3 This idea of Mencius that one’s ethical/moral consideration is not always overriding one’s non-ethical/moral consideration is echoed by Bernard Williams when he illustrates his point by using his example of a person deciding to rescue his wife instead of a strange simply because she is his wife (Williams 1981: 17-19) “everyone”, including those who are endowed with turbid qi, which Zhu Xi uses to explain why some people are selfish. Zhu Xi’s explanation is that there must be something in the nature of humans, which he characterizes as benevolence, that enables them to have such a response when encountering such a situation. Such an explanation is sufficient and seems to me quite natural. In contrast, if we adopt Walker’s Mencius, the explanation of such a common emotional response would be that the thing in human nature that causes everyone to have such a response is not benevolence but merely a disposition to acquire benevolence. It is difficult to see how merely a disposition to acquire benevolence would cause one to have the emotional response of commiseration upon seeing a child about to fall into a well. The contrast between Zhu Xi’s explanation and Walker’s Mencius’s explanation can be further illustrated by a number of analogies that Zhu Xi’s uses. When we see the bud growing above the ground, Zhu Xi says that that there must be something underground, its root, from which it grows, while Walker’s Mencius would say that there must be something underground with the disposition to become root, from which the bud grows; when we see a shadow of a tree, Zhu Xi would say that there must be a tree that casts the shadow, while Walker’s Mencius would say that there must be something with a disposition to become a tree that casts the shadow; when we see water flowing from a source is clear, Zhu Xi says that the water in the source must be clear, while Walker’s Mencius would say that the water in the source must have the disposition to become clear; when we see a child behaves so well, Zhu Xi says that their mother must be a very nice person, while Walker’s Mencius would say that their mother must have the disposition to become nice. Clearly, in each of these cases, Zhu Xi’s explanation is far better than Walker’s Mencius’s explanation. This leads to the first main question that Walker raises, a question originally I prefer to say nothing about. Although in my original paper, I do share the worries with contemporary philosophers such as Bernard Williams and John McDowell about Aristotle’s attempt to derive moral virtues from rationality, one of my main goals in the paper is to develop an agent-focused moral realism that can avoid some typical criticisms of moral realism from moral anti-realists that the action-focused moral realism finds it difficult to respond to. So if Walker is right that my worries about Aristotle’s ethics are unnecessary and Aristotle can do the same job as I think Zhu Xi can do, then all is good and well. After all, as I pointed out in my original paper, Zhu Xi and Aristotle take the same agent-focused approach to metaethics. However, it becomes clear that, by contrasting Zhu Xi’s latent virtue innatism with his Mencius’s dispositional innatism, Walker argues that, although the latter is better supported by our common emotional responses than the former, which I have shown is not the case, it is after all still not supported by such emotional responses. Instead, he argues that Aristotle, who rejects both, provides a better alternative, a weak non-innatism: “although we lack any innate predispositions toward (or against) virtue as such, we nevertheless innately possess other capacities and character dispositions that we share with non-human animals…. And we can habituate these dispositions into natural virtues, i.e., dispositions that approximate various ethical virtues. With additional habituation, further practical experience, and reciprocal harmonization with practical reason, our natural virtues can become ethical virtues in the strict sense”. I will not go into the detail of Walker’s presentation of Aristotle’s view except asking one question: is there anything in Aristotle’s conception of human nature that requires us to “habituate these dispositions into natural virtues” and further cultivate them into “ethical virtues in the strict sense?” If the answer is “yes”, then Aristotle is not a non-innatist, weak or strong. If so, he needs to provide arguments to show that there is such a thing in human nature, arguments similar to those made by Zhu Xi (his inference to best explanation arguments in relation to common human emotional responses and to the human and nonhuman difference in their moral perfectibility, which I present in my original paper) as well as the contemporary neo-Aristotelian Rosalind Hursthouse (her argument about human nature as consisting of four ends and five aspects). If the answer is “no”, an answer that Walker seems to make when he regards Aristotle as a non-innatist, however, then ethical virtues in Aristotle are certainly normative (though Walker is wrong in saying that such a normativity is not sufficiently argued for in either Zhu Xi or Mencius) but lacks objectivity. If so, it is difficult to argue that Aristotle is a moral realist. Indeed, if for Aristotle “human beings [only] possess virtue-neutral capacities and dispositions” by nature, then virtues are not objectively grounded. In my view, criticisms of Aristotle made by Bernard Williams and John McDowell should also be understood in this context. One possible response that Walker could make is that, while what is innate in human nature is rationality, not virtues or even disposition to acquire virtues, virtues are what make one’s rationality function well and, in this sense, are not only consistent with but also required by rationality as human nature. A problem with such a response is that, just as “good” in “a good human being” is an attributive adjective, whose meaning is dependent on our understanding of what human being is, “well” in “wellfunctioning of rationality” can be said to be an attributive adverb, whose meaning is dependent upon our understanding of the functioning of rationality. Moreover, the rationality here is clearly human rationality, as in contrast to divine rationality. So what is innate in human being cannot be just some kind of generic rationality but human rationality, with specific features that tell us when it functions well and when it functions badly, since the well-functioning of the human rationality may not be the well-functioning of the divine rationality. If so, Aristotle is also not a non-innatist, even in a weak sense. 4 Objectivity of Zhu Xi’s Conception of Human Nature: Response to Cokelet In his paper, Brad Cokelet presses me to face two forms of normative skepticism about (my presentation of) Zhu Xi’s conception of human nature. The first is about its objectivity, and the second is about its normativity. I shall mostly address the first, as I have already addressed the second in my response to Professor Copp regarding being healthy and being good. The first form of normative skepticism “draws attention to the numerous disagreements that philosophers and others have about human nature and about the extension of ‘good human being’ and ‘bad human being’.” I assume that Cokelet here is not a skeptic about my attributive use of “good” and “bad” when I apply them to human beings. In other words, if we know human nature is X, whatever it is, Cokelet would agree with me that a good human being is one in whom X is well manifested, while a bad human being is one in whom X is impaired. So Cokelet’s first form of skepticism is mainly about two other things: (1) whether disagreements about what this X is are tractable, and (2) whether Zhu Xi’s answer to the question of what this X is is defensible. I don’t have many things to add to what I’ve already said in my original paper regarding (1). While the disagreement about whether a particular person is a good human being or bad one can be solved by appealing to human nature, I acknowledge that there is currently still disagreement about what human nature is, but this is a disagreement about fact, even though a normative fact (as it can provide us with a criterion to distinguish between good human beings and bad ones). This is similar to the disagreement we once had about the nature of swan, which is also a disagreement about fact (as well as a normative fact, as it provides us with the criterion to distinguish between good swans and bad swans). When we disagree on the nature of swans, we can always trace such disagreement to disagreements on certain facts about the swan, and so such disagreement is tractable. Moreover, when such facts about the swan are made clear by further studies, such disagreements would disappear. Now there is no, and there cannot be any, qualitative difference between disagreement on the nature of human beings and disagreement on the nature of swans, except that human beings are much more complicated than swans. Human nature is what distinguishes human beings from other beings, just like the nature of swan is what distinguishes swan from other beings. If we disagree on human nature, we just need to do some further studies of human beings, just as when we disagree on the nature of swans, we just need to do some further studies of swans. This is different from our disagreement about, for example, the rightness or wrongness of abortion, a disagreement that arises for action-focused moral realism. In this case, even when both parties of the disagreement completely agree on every aspect regarding what abortion is, they will continue to disagree on whether abortion is right or wrong, and so this disagreement is intractable. Now I come to second part of Cokelet’s first form of skepticism, the one about whether Zhu Xi’s view on human nature is defensible. Cokelet claims that I have the pre-conceived principles according to which to select my favored theory of human nature, which of course is not true. The only criterion I judge the adequacies of a theory of human nature is to see whether what this theory claims to be human nature is indeed human nature, i.e., whether this conception of human nature, while normative (as it will inform us about how to determine whether a particular person is a good human being or a bad one), is also objective. So my response here will focus on Cokelet’s challenge about the objectivity of Zhu Xi’s conception of human nature as I presented in the original paper. In this regard, Cokelet raises the familiar and expected objections to Zhu Xi’s view that all and only human beings have the emotional responses such as commiseration upon seeing a child about to fall into a well. About the “all” part, Cokelet counters that “some people with autism and some sociopaths will lack feelings of commiseration if they encounter a child who has fallen into a well or pond”. Zhu Xi’s response is that in the sense and to the degree that these people inherently and incurably lack such emotional responses characteristic of human beings, they are indeed deficient ones. This is similar to using a counterexample of someone who is born with one leg to challenge the view that human beings are characteristically bipeds. About the “only” part, Cokelet says that “some non-human animals can feel compassion and commiseration” as shown in “current empirical science”. Contrary to what Cokelet imagines, this is actually not beyond the “assumptions of classical Confucian thinkers such as Zhu Xi”. Indeed, Zhu Xi also talks about the “benevolence of tigers and wolves, ritual of sacrifice of jackals and otters, and rightness of bees and ants” (Zhu 1986: 58). However, for Zhu Xi, on the one hand, each of these animals has only one emotional response similar to those of humans, and, on the other hand, they cannot extend such emotional response beyond their intimate circles. For both “all” and “only”, we need to remember that what Zhu Xi provides is a naturalistic conception of human nature, which is different from, for example, mathematic conceptions of the nature of a geometric figure. There is no bordering area between square and rectangle, for example. In naturalistic conceptions, there is always some bordering area between one type of tree and another type of tree, one type of animal and another type of animal. This, however, doesn’t prevent us from using the nature of one type of tree (animal) and that of another type of tree (animal) to distinguish between these two types of trees (animals). When Zhu Xi develops his naturalistic conception of human nature to mark human beings away from non-human beings, he also allows the existence of the bordering area and residents living there. So he says that there are animals that are close to human beings, and there are human beings that are close to animals (Zhu 1986: 58). One of the arguments that Zhu Xi provides to support his conception of human nature is the distinction between bad human beings and non-human beings, an argument that Cokelet also challenges. Although in their present state, they are not much different from each other, in terms of their essential nature, they are very different from each other: the former can be transformed into good human beings, while the latter cannot. This is clearly an empirical fact about which Cokelet should not have the doubt that he actually has, as we have numerous examples of the former but none of the latter. Indeed it is something that we all take for granted. Thus when a bad human being kills another person, we blame them, saying that they ought not to do it, and it is wrong for them to do it. If a tiger kills a human being, however, we don’t say the same thing to the tiger. Instead we may blame the person killed for not being careful enough or the society for not doing enough to protect people from tigers. This shows that there must be something in (even bad) human beings that is absent in animals, and this thing must belong to human nature: something that all and only human beings have. This thing for Zhu Xi is benevolence, etc., which has been also called sympathy by David Hume or fellow-feeling by Adam Smith. If this is the case, it is difficult to understand why Cokelet can accept that “all [I trust he also agree to add “and only” here] human beings are capable of become more moral” (I prefer to use “virtuous” or simply “benevolent”, etc., here) and yet think that “doing so will not make them better human beings or even make them into bad human beings”, since benevolence etc., is part of human nature. Precisely because of that, even bad human beings are capable of being more benevolent but animals are not. The second form of skepticism that Cokelet has for Zhu Xi’s agent-focused moral realism is directed to the normativity, instead of objectivity, of Zhu Xi’s conception of human nature. Assuming that human nature does consists of benevolence, etc., and a good human being is one in whom such a nature is well manifested, while the bad human being is one in whom such a nature is impaired, Cokelet claims that “[t]here remains space to question whether he [a bad man] therefore ought to be different and cultivate his sprouts to become a better human being.” He does so by focusing on my relating goodness to health, claiming that health is not always normative. Since Professor David Copp raised the same issue in his paper, to which I’ve already responded in Section 2, I would not repeat it here. 5 What about Moral Properties of Rightness and Wrongness of Actions?: Response to Liu Since the agent-focused moral realism is concerned to show the objectivity of moral properties of goodness and badness of a person, Liu’s first question is, quite naturally, what this agent-focused moral realism has to say about the moral properties of rightness and wrongness of action that action-focused moral realism is concerned with. One may complain that it is unfair as people never ask the action-focused moral realists what they have to say about the moral properties of goodness and badness of a person. This is not a problem, however, as the agent-focused moral realism should indeed answer the question and, fortunate, has a ready answer to it. As is well-known, in virtue ethics as normative ethics, right action is defined as an action that a good or virtuous person would characteristically do in circumstances. Since the agent-focused moral realism as a metaethical theory is a reflection on issues arising from virtue ethics, its view of the rightness and wrongness of action is also derived from its view of the goodness and badness of the person. So, in agentfocused moral realism, the statement “A (an action) is right (or wrong)” is true if it is the statement a good person would make. Moreover, a good person would make such a statement not because they have perceived some moral properties of rightness or wrongness in the action (if so, various difficulties that action-focused moral realism are inflicted with would come back). Instead, it is because the good person projects the value of rightness or disvalue of wrongness unto the action, which is basically an expressivist or projectivist view that Simon Blackburn advocates (Blackburn 1993). One may wonder how a moral realist on the moral properties of goodness and badness of a person can simultaneously hold such an apparently anti-realist view on the moral properties of rightness and wrongness of the action. I don’t see any inconsistencies here, as the moral realism that I’m developing and defending here is not a generic moral realism but, as I emphasized, an agent-focused moral realism. Still, as I argue more explicitly and in more detail in a forthcoming paper on Wang Yangming’s agent-focused moral realism, even though moral statements about actions are actually expressions of our attitude, they still have truth value, in a sense even stronger than Blackburn’s quasi-realism. In this sense, the agent-focused moral realism is still real realism, not a quasi one, even in terms of moral properties of rightness and wrongness of action. According to this view, a moral statement about rightness and wrongness of an action is true only if it is one that a good person would make, as it is nothing but the attitude or emotion the good person expresses, which makes the meta-ethical position looks anti-realistic. However, there is still an objective way to determine whether such a moral statement is true or whether the emotion/attitude expressed is appropriate, which is to see whether the person who makes the statement or expresses the emotion is a good person. Moreover, there is an objective criterion to determine whether the person making the statement or expressing the attitude is a good person or not, which is to see whether this person is one in whom the human nature is well manifested. Due to the transitive relations, it can be claimed that rightness of action, which is a value spread on the action by a good person, is also objective (for detail, see Huang 2024). I have some more things to say on this topic in my response to Sheng in section 6. I will not say much about the second area of issues that Liu has with my original paper, which concerns my presentation of Zhu Xi’s agent-focused moral realism. This is because she has fundamentally misunderstood the nature of Zhu Xi’s argument about human nature. As I present it in my original paper, Zhu Xi’s argument is neither deductive nor inductive; instead it is inference to the best explanation, as is accurately captured in Professor David Copp’s comments. Why do all human beings have the emotional response of commiseration upon seeing a child about to fall into a well? Zhu Xi’s explanation is that there must be something in human nature that enables humans to issue out such emotional response in such a circumstance, and this thing he calls benevolence. Why can even the most malevolent person be made into benevolent human being but no other animals can? Zhu Xi’s explanation is that there must be something in the nature of human beings, including the most malevolent ones, that doesn’t exist in the nature of other animals, and this thing Zhu Xi also identifies as benevolence. In similar ways, he argues that there must be rightness, propriety and wisdom in human nature. To argue against such an argument of Zhu Xi’s about human nature, one can either show that the two empirical facts that Zhu Xi aims to explain don’t exist or that, while they do exist, Zhu Xi fails to provide a good explanation or his explanation is not the best one. Liu, however, does neither of this, but instead misunderstands this inference to the best explanation argument as a deductive argument. With such a misunderstanding, she takes pain to construct a series of premises that she thinks Zhu Xi has to have in order to result in the desired conclusion, and then try to show why at least some of these premises are problematic and why the argument is weak or invalid or circular even if all the premises are true. Since this is not the type of argument that Zhu Xi makes and thus the problems that Liu thinks are associated with such an argument simply don’t exist with Zhu Xi’s argument, I shall not delve further into them. In my original paper, I argue that the agent-focused moral realism can meet many challenges that the action-focused moral realism finds it difficult to meet, and one of these challenges is related to the Is-Ought gap that is made famous by David Hume: people tend to jump, without logical deduction, as well summarized by Liu, “from the non-moral, descriptive statements concerning states of affairs, to statements concerning moral facts with normative properties”, and she wants to see whether I or my Zhu Xi “could provide such a logically valid deduction” or jump, and thus whether I successfully meet Hume’s challenge. Here I think Liu misses a point. Agent-focused moral realism meets Hume’s challenge not by providing the logical deduction that Hume demands but by showing that this very demand is based on the misconception that a descriptive statement can never be normative, which is further based on the misconception that facts are always pure facts and there can never be normative facts. As “good” in the statement “X is a good human being” is an attributive adjective, while the statement is clearly a normative one (X is what they ought to be), it is also a descriptive one (X well manifests the nature of human being); and this is because goodness as a moral property is both objective and normative. In this context, it is indeed strange for Liu to insist that I or my Zhu Xi needs to provide the logical deduction. Specifically, she wants us to provide the logical deduction from “you are a good person” to “you ought to be [a] good [person]” similar to the logical deduction from “you are healthy” to “you ought to always keep your body healthy”. However, “a good person” already means “a person one ought to be”, and so to say that “you ought to be a good person” is equivalent to saying that “you ought to be a person you ought to be,” which is tautological. Here Liu fails to see that there are at least two ways to express a normative claim: “X is a good person” and “X ought to be the person they are” or even “everyone else ought to be a person like X” (“X is right action” and “one ought to do X”). I indeed didn’t try to make the deduction from the former to the latter, but if I did, I would have not provided the deduction that Hume demands: since what he wants is to deduce something normative from something purely descriptive, but, as I said, “X is a good person” is not purely descriptive: it is both descriptive and normative, and, for that reason, there is indeed no longer the need to do the deduction.4 6 Other Issues: Responses to Sheng, Rooney, and Relador Sheng’s whole paper is devoted to one of the questions that Liu raised: what my agentfocused moral realism has to say regarding moral properties of rightness and wrongness of actions. Since I have already made some response to this question raised by Liu above, I shall limit my response here to some specific issues in Sheng’s paper. Just as Liu thinks that there is a generic moral realism, Sheng thinks that there is a general moral realism, which would affirm not only the objectivity of both moral properties of rightness and wrongness of actions and moral properties of goodness and badness of agents but also the identity of objectivity in these two cases. My contention is that there has never been a generic or general moral realist. Moral realism, as a metaethical theory, is reflective of one of the most important issues, if not the most important one, of normative ethics. Normative ethics, as we know it, is either actionfocused (consequentialism and deontology) or agent-focused (virtue ethics). In the former, the goodness or badness of an agent is derived from the rightness or wrongness of the action this person does, while in the latter, the rightness or wrongness of action is derived from the goodness and badness of its agent. So there cannot be a generic moral realism, if it aims to adequately reflect issues in normative ethics, that can put moral properties involving actions 4 Liu does also raise the question about normativity of health and thus also of goodness, a question I already responded to when I address the similar question raised by Copp and Cokelet, and so I’ll not repeat it here. and those involving agents on an equal footing. As deontology and consequentialism have been dominating ethical theories until recently in modern Western philosophy, moral realism we are familiar with has been mostly action-focused, even though the general or generic term “moral realism” has always been used to denote such a theory. Sheng demand from me such a generic moral realism, which is unreasonable, as I made it clear, in the subtitle of my original paper, that I take a virtue ethics approach to metaethics, i.e., I regard virtue ethics as the appropriate type of normative ethics, from which metaethical reflection ought to start. Still it is reasonable for Sheng to ask me what my agent-focused moral realism, which is primarily concerned with the goodness and badness of a person, has to say about rightness and wrongness of action, as I indeed didn’t say anything about it in my original paper, but I have just briefly said it now in my response to Liu’s first question, drawing on things I’ve said more extensively in one of my my forthcoming papers, which I shall not repeat here. What I want to point out here is that the position my agent-focused moral realism takes regarding moral properties involving action is not one of what Sheng considers to be the exhaustive three options: (1) keep silent; (2) take a non-realist position, and (3) take a realist stance, the same stance that the action-focused moral realism will take. Sheng claims that the first is “seriously inadequate”, which I agree. Should I take the second one, Sheng claims, while not incoherent, I would not be able to claim the position I take as a realist one, which I’m not sure, as it is an agent-focused, not action-focused, moral realism. Should I take the third option, Sheng argues that I would no longer be an agent-focused moral realist but a general or generic moral realist. As I indicate in my response to Liu, the option I take is first to take an expressivist account of moral properties of rightness and wrongness of action, which appears to be Sheng’s second option, and then provide a realist defense of such an expressivism, much stronger than Simon Blackburn’s quasi-realist defense, which appears to be Sheng’s third option, though still very different. So while my account of moral properties of rightness and wrongness is ultimately also realistic, it is still properly agent-focused, as the rightness and wrongness of the action is determined by the goodness and badness of the person who expresses such attitudes toward the action, and the goodness and badness of a person can be objectively determined. The main advantage of Zhu Xi’s agent-focused moral realism is that it eliminates the gap between “is” and “ought” and provides a plausible candidate for objective prescriptivity, which would otherwise appear queer from John Mackie’s point of view. This is because goodness in “X is a good human being”, from Zhu Xi’s point of view, while clearly normative or prescriptive, indicating that X is the type of human being they ought to be, is thoroughly objective, describing the fact that X is one who manifests human nature, which consists of benevolence, etc., well. James Rooney, however, claims that Zhu Xi, at least as I present him, fails to do so. How? Because I acknowledge that there are deficient human beings, who also possess human nature but don’t manifest it well, Rooney claims that this leaves intact the gap between possessing human nature and manifesting it well, or, to use his own expression, the gap “between being human essentially and exemplifying human nature well.” Because of that, Rooney claims that Zhu Xi’s version of agent-focused moral realism suffers the same problem that I claim exists in Aristotle and the neo-Aristotelian Hursthouse. Two things can be said in response. On the one hand, the problem I see with Aristotle and Hursthouse is not that there are people who possess human nature (rationality for Aristotle and the five aspects serving four ends for Hursthouse) and yet don’t manifest this nature well. Rather, it is that at least some ethical virtues that Aristotle and Hursthouse argue that all human beings ought to have cannot be derived from their respective conceptions of human nature. On the other hand, the fact that there are human beings who are not what they ought to be, whether in the Aristotelian sense or in the Confucian sense, doesn’t mean that there is a gap between Is and Ought. What a good human being is or what kind of human being one ought to be, the “ought”, is not determined by the particular “Is” of any particular human being but by the “Is” of human nature, the thing that sets human beings apart from other beings. It is precisely in this sense that I claim that Zhu Xi’s version of agent-focused moral realism has an advantage over the Aristotelian one, as human virtues (the “ought”) are from human nature (the “is”) and this is because human virtues are constitutive of human nature, and so human nature and human virtue are identical. While all human beings, as human beings, possess the same human nature, some manifests it well and they are regarded as good human beings, but others don’t manifest it well, if at all, and they are regarded as bad human beings, which is something about which Aristotelians and Zhu Xi have no disagreement about.5 In my original paper, after presenting Zhu Xi’s agent-focused moral realism, I argue that such a version of moral realism can avoid the four classical criticisms of (the action-focused) moral realism made by Hume, Moore, and Mackie. In his paper, Relador argues that I succeeded to do so only at the cost of normativity. He does so by showing that these critics indicate that a metaethical theory needs to capture three desiderata: “(i) the connection between moral judgment and motivation; (ii) how our moral explanations are unified; (iii) how our moral judgments are categorical and authoritative.” I’m not sure how Relador gets these desiderata from these critics in the objections they raise to moral realism. For one, he is obviously wrong to think that the point of Hume’s Is-Ought problem is that moral judgement is motivational. An “ought” statement is certainly normative but whether it is also necessarily motivational is at least subject to controversy. Indeed this is precisely where moral judgment internalism and moral judgment externalism disagree. The former claims that motivation is internal to moral judgment, while the latter holds that it is external to moral judgment. By equating normativity and motivation, Relador shows that he is a moral judgment internalist, which is fine. What is problematic is that he pictures David Hume, the best-known moral judgment externalist, as an internalist, and he does so partially by changing “morals” in Hume’s original text into “moral judgments” as the subject terms of the verbs “excite passions, and produce or prevent action.”6 For another, it is also not clear why Relador claims that the point of Mackie’s (and Gilbert Herman’s) argument [against moral realism] from relativity is that moral judgment is categorical and authoritative. Moral relativism for Mackie and Harman is a better explanation of the intractable moral disagreement than realism, and it is hard to see how a relativist moral judgment can be more categorial and authoritative than a realist moral judgement. Nevertheless, even if the three listed by Relador are indeed desiderata strictly coming from these critics, I don’t see any more difficulty for a version of 5 Rooney raises a couple of other issues with Zhu Xi’s agent-focused moral realism, arising from his misperceived problem of the same gap in Zhu Xi as in Aristotle that I have just dealt with here. He says that, in order to avoid the gap, I or my Zhu Xi have to adopt the view that one’s lack of virtues or possession of vices will “cause a change in species membership” so that they will have no “reason to become human, any more than a cat does”. Since neither to keep the gap as he (mis)perceives in my account of Zhu Xi’s conception of human nature nor to adopt the species crossing approach as an alternative is good, he further suggests that I should take his own initiative. Since I have already shown that he is mistaken about the gap, there is no need for me to say anything about the “remedies” he proposes for this perceived, mistakenly, problems in my account of Zhu Xi. 6 I myself agree with Hume that one is motivated to act with both belief and desire, but I disagree with him in the sense that belief and desire for me are not two separate mental states but a single one, which is both belieflike and desire-like and thus can be properly regarded as the mental state of besire. In other words, I can be regarded as a moral judgment internalist, but only because this moral judgement is both cognitive and emotive (see Huang 2020). moral realism, naturalist or non-naturalist, to satisfy them than a version of moral antirealism, naturalist or non-naturalist, does. This leads to the next point Relador makes: the three desiderata “give rise to the central problem running throughout all four critiques: that naturalness and normativity are incompatible” (emphasis added). First, if Relador is right, then these classical criticisms are not criticisms of moral realism or, for that matter, moral anti-realism, but moral naturalism, whether realist or anti-realist, and so all these critics must be non-naturalists, whether realists or anti-realist, which is obviously not the case. Perhaps G.E. Moore is the only non-naturalist among the three (or four, if we add Gilbert Harman, who I also discussed as a proponent of the argument from relativity in my original paper), but his open-question argument is not merely targeting naturalist but also non-naturalist, since he argues that the good cannot be defined in terms of what God commands (something non-natural) as much as it cannot be defined in terms of what produces happiness (something natural). Second, in moral realism, normativity is related to objectivity, not naturalness. In naturalist moral realism, both normativity and objectivity are natural, while in non-naturalist moral realism, both normativity and objectivity are non-natural. Moreover, a moral naturalist is not necessarily a moral realist, as you can also be a naturalist moral anti-realist.7 Does Relador think that the normativity and naturalness are also incompatible in naturalistic moral anti-realism? Third, while there is no ground for us to say that all critics of moral realism aim to show the incompatibility of normativity and naturalness, can we say instead that they all aim to show the incompatibility of objectivity and normativity? While Hume’s Is-Ought problem and Mackie’s argument from queerness may indeed try to show this incompatibility, it is difficult to say that it is also the central point of Moore’s open question argument and Mackie’s argument from relativity. In the latter case, as we have already pointed out, Mackie aims to show that moral realism cannot explain the intractable moral disagreement as well as moral anti-realism. In the former case, Moore himself is a moral realist, though a non-naturalist one, and he would certainly endorse the compatibility between objectivity and normativity. The substantive argument Relador makes against my agent-based moral realism, however, is that attributive good is not normative but merely evaluative. Apparently, Relador is not saying that (a) “John is a good human being” is merely evaluative because there is no “ought” there, since he would agree that this sentence is equivalent to (b) “John ought to be the kind of human being he is”, in which “ought” appears. So if the latter is normative, then the former is also normative. This is similar to the case of (c) “John’s action is right” and (d) “John ought to do what he does”, which are also equivalent, and so if the latter (where “ought” appears) is normative, then the former (where there is no “ought”) is also normative. The reason I think this is not what Relador means is that he makes the distinction between evaluative ought and normative ought. I’m not sure whether he will consider (d) as also merely an evaluative ought, but his view is certainly that (b) is merely an evaluative ought and not normative ought. He doesn’t tell us how he makes the distinction between these two senses of ought, except by making a reference to Mark Schroeder’s paper “Ought, Agents, and Actions.” Unfortunately, Relador has mistaken Schroeder’s view in the paper in a number of fundamental ways. First, Schroeder doesn’t contrast evaluative ought with normative ought. Instead, he contrasts evaluative ought with deliberative ought. Second, 7 Richard Joyce, for example, points out that both moral realism and anti-realism can be naturalistic. He gives the following example to illustrate the latter: “a simplistic non-objectivist theory that identifies moral goodness (say) with whatever a person approves of. Such a view would be a form of anti-realism (in virtue of its nonobjectivism), but since the phenomenon of people approving of things is something that can be accommodated smoothly within a scientific framework, it would also be a form of moral naturalism” (Joyce 2021: § 1). according to Schroeders’ distinction, our (b) above, which expresses attributive goodness, is deliberative ought and not evaluative ought. For him, the evaluative ought, such as “there ought to be world peace” or “the world peace ought to obtain,” does not express “a relation between an agent and action” (Schroeder 2011: 5); in contrast, deliberative ought, such as “Jim ought to Jam” or “Sal ought to sail,” expresses such a relation. Clearly, our (b), just like our (d), does express a relation between an agent and an action (“to be” in (b) as in contrast with “to do” in (d)), and so are deliberative ought and not evaluative ought. Third, most importantly, for Schroeder, both evaluative ought and deliberative ought are normative oughts. His discussion of these two kinds of ought, evaluative and deliberative, is after he claimed, with the epistemic ought excluded as non-normative, that “Henceforward, all uses of ‘ought’ under consideration are to be understood as normative uses” (Schroeder: 3; emphasis original); and, after listing a number of examples of evaluative ‘ought’, he says that “each is plausibly interpreted as making a broadly normative claim, in contrast to a merely epistemic claim about what is likely to be the case” (Schroeder: 5; emphasis original). Clearly, in all these three aspects, Relador gets Schroeder wrong. 7 Metaphysics and Essentialism in Zhu Xi?: Concluding Remarks I would like to conclude by addressing two additional issues, both related to Zhu Xi’s conception of human nature, not explicitly raised by my critics above but by a number of other friends and colleagues when I present my paper on several occasions. The first is whether Zhu Xi’s moral realism can be appropriately characterized as naturalistic as I did in my original paper. This is because, as I explained there, Zhu Xi considers human nature to be metaphysical (xingershang 形而上), in contrast to human emotions, which he regards as “physical” (xinger xia 形而下). When explaining their view that moral properties are objective but non-naturalistic, in additional to mathematical properties, which are both objective and non-naturalistic, non-naturalistic moral realists often use the analogy of metaphysical properties, such as Plato’s form, as also both objective and non-naturalistic. Since human nature for Zhu Xi is metaphysical, my critics say, his moral realism must be non-naturalistic. My response to this is that metaphysics can be naturalistic as well as nonnaturalistic. Plato’s is non-naturalistic, as his form is not part of the natural world, nor can it be known in the way we know things in the natural world or through our knowledge of things in the natural world. Zhu Xi’s metaphysics, however, is naturalistic, since human nature, which is metaphysical, is the nature of human beings, which are residents in the natural world, and can be known only through our knowledge of things in the natural world. So Zhu Xi’s conception of human nature as metaphysical doesn’t prevent it from being naturalistic at the same time. The second issue raised about Zhu Xi’s conception of human nature is that it is a kind of essentialism, which has long been discredited in contemporary philosophy, at least since Wittgenstein made his idea of family resemblances popular. My response to it starts with a similar distinction, one between non-naturalistic essentialism, which Wittgenstein is against, and naturalistic essentialism, which, while still different from the idea of family resemblances, is not something very unWittgensteinian. Horizontally, as I’ve shown in my response to Cokelet, naturalistic essentialism leaves room for the bordering area between human beings with one essence and non-human beings with different essences, so that residents in the bordering region may look like human beings in some aspects but non-human beings in some other aspects. Vertically, naturalistic essentialism can allow the change of human nature as the essence of human beings through evolution. It only states that at this stage of evolution, there is an essence of human beings, which helps to distinguish human beings from non-human beings. It doesn’t exclude the possibility, however, that in the next stage of evolution, which may take millions of years, if not more, to complete, human beings may acquire a somewhat different nature or essence. Then we will have a new way to distinguish between human beings and non-human beings. Thus, essentialism in the naturalistic sense is not that bad, and indeed it may be indispensable, as some radical antiessentialist views, perhaps surprisingly, also assume the truth of such an essentialism in one way or another. For example, Roger Ames, as an anti-essentialist in his interpretation of the Confucian conception of human nature, urges us to use “human becomings” instead of “human beings” to show that there is no fixed human nature. However, he has to agree that the fact that someone is a human becoming instead of a dog becoming or an oak becoming shows that there is some human essence in the human becoming, just as there is a dog essence in the dog becoming and an oak essence in the oak becoming, as the human becoming will not become a dog becoming or an oak becoming, and vice versa. For another example, essentialism, which claims that a thing’s essence precedes its existence, is traditionally in contrast to existentialism, which claims that a thing’s existence precedes its essence. However, even contemporary existentialists such as Martin Heidegger and Jean-Paul Sartre are also essentialist in another sense, since for them only in human beings does existence precede essence, while essence precedes existence in all non-human beings, which implies that the precedence of existence over essence is the very essence that distinguishes human beings from non-human beings. References Blackburn, Simon. 1993. Essays in Quasi-Realism. New York & Oxford: Oxford University Press. Huang, Yong. 2020. “Belief, Desire, and Besire: Slote and Wang Yangming on Moral Motivation.” In Yong Huang, ed., Michael Slote Encountering Chinese Philosophy: A Cross Cultural Approach to Ethics and Moral Philosophy. London: Bloomsbury, pp. 81-98. _____. 2024. “Two Defenses of Moral Expressivism: From Simon Blackburn’s QuasiRealism to Wang Yangming’s Agent-Focused Realism.” In Yong Huang, ed., Simon Blackburn Encountering Chinese Philosophy. London: Bloomsbury (forthcoming). Hursthouse, Rosalind. 1999. On Virtue Ethics. Oxford: Oxford University Press. Joyce, Richard. 2022. “Moral Anti-Realism.” The Stanford Encyclopedia of Philosophy, eds. by Edward N. Zalta & Uri Nodelman, URL = <https://plato.stanford.edu/archives/win2022/entries/moral-anti-realism/>. Williams, Bernard. 1981. Moral Luck: Philosophical Papers 1973-1980. Cambrdge: Cambridge University Press. Schroeder, Mark. 2011. “Ought, Agents, and Actions.” Philosophical Review 120.1: 1-41. Zhu, Xi. 朱熹. 1986. Classified Sayings of Master Zhu 朱子語類. Beijing: Zhonghua Shuju.