Journal Archive

To describe my issue very roughly, I am interested in the ethical basis for caring about people with whom one is not personally acquainted and with whom one shares no national identity or citizenship. Let us say that Kai goes to great lengths to provide life-saving vaccines for thousands of people who live in a faraway country. Or consider a more minimal obligation not to inflict horrific harms on other human beings of any kind. For example, many of us think that we have ample moral reason to refrain from participating in genocidal acts against people in other countries, even when those people are not friends, family members, or the like. What makes Kai’s efforts toward total strangers in faraway lands ethically good and admirable, and what makes it wrong for us to participate in genocidal acts against people who have no special (political or personal) relationship with us? One tempting answer appeals to the mere humanity of the non-citizen strangers for whom we and Kai act. Just in virtue of the fact that these people are human beings or have certain features characteristically possessed by human beings, it is often good and obligatory for us to act to protect or promote their interests. For the most part, we and Kai do not need to have any particular relationship with the relevant people in order for the relevant ethical values to hold. It is not because we are related as members of the same family or belong to the same community that we should act to protect or promote their interests. But there is another temptation, which is to say that there is some sort of meaningful relationship with non-citizen strangers after all, and that this relationship provides some basis for treating them well. Perhaps Kai does not know the beneficiaries of his vaccination campaign personally, but maybe there is some sense in which he is nevertheless in a kind of relationship with them. Perhaps they are, somehow, “siblings” in some extended or attenuated sense, being both contemporaries and members of the “family” of the human species. Maybe we could identify some common goals or concerns that Kai and his beneficiaries share, such that we could characterize them as colleagues or even comrades or friends in a greater cause. Roughly speaking, there are two ways of going about accounting for the ethical value of caring for non-citizen strangers. One appeals to the non-citizen stranger’s mere humanity (or mere possession of features that humans characteristically have) and supposes that a potential beneficiary’s humanity alone (or mere possession of the characteristic features) suffices to establish the relevant ethical value. The other looks to account for the ethical value through meaningful relationships. Abstracting from some of the historical context, this is my issue.

Described in more historic terms, my issue is how a line of traditional Confucian thought accounts for the ethical value of care for non-citizen strangers. Confucianism is rightly said to prize human relationships more highly than some other major ethical worldviews, and some scholars even think that Confucians conceive of all ethical values in terms of relationships (Ames 1998, 2011; Rosemont 1991, 2015). This makes it an interesting challenge to explain how they can account for the ethical norms that seem to be for the sake of people with whom we do not have relationships in any obvious sense. It also potentially provides an opportunity to see more clearly how deeply or thoroughly “relational” Confucianism really is. If it turns out that some Confucians believe that there is not a meaningful relationship between people like Kai and his beneficiaries and yet also that Kai’s behavior is virtuous, admirable, or good, that suggests that Confucianism does not treat human relationships as all-encompassing of ethics or as the central ethical notion after all.

Since Confucianism is a vast and sprawling philosophical tradition with many branches, I will focus on one line of Confucian thought in particular, which for present purposes we can call the “Mengzian” line, taking after the canonical philosopher Mengzi 孟子 (also known as “Mencius,” fl. c. fourth century BCE). One idea that is more clearly articulated in the Mengzi than in other canonical texts—and later developed by other philosophers who take after Mengzi—is that our basic obligations toward others or the basic claim that other people’s interests hold over us varies according to how near and dear they are to us (see, e.g., Mengzi 3A.5 and 7A.45). For example, we have the strongest moral or ethical interest in protecting and promoting the welfare of immediate family members, slightly weaker interests in doing the same for close friends and more distant family members, still weaker interests in doing such for members of the community more broadly, and so on. Sometimes this view is called the Confucian doctrine of “graded love” or “care with distinctions.”¹ Care with dis tinctions shapes the ethical thought of a great many Confucian philosophers. However, among the classical and canonical Confucian philosophers, Mengzi is arguably the most explicit about it, and it is some later philosophers who see themselves as building on Mengzi that elaborate on it with the most subtlety and sophistication. I will focus in particular on the elaborations of latter-day Mengzians Wang Yangming 王陽明 (1472–1529) and Dai Zhen 戴震 (1724–1777).

To sum up, my aim is to describe the ethical basis for caring about non-citizen strangers, as the Mengzians see it. In the interest of making sure that we are on the same page, let me say a bit more about two italicized phrases as I will use them. By “ethical basis,” I just mean the normative patterns, principles, or standards in virtue of which some traits of character or courses of action are ethically good, bad, permissible, impermissible, etc. In construing the phrase in this open-ended way, I am trying to sidestep some thorny issues in the implicit metaphysics or metaethics of Mengzianism that would take us too far afield. If we can extract from Mengzi a theory which makes it good or virtuous to care for non-citizen strangers, that would qualify as an ethical basis. But the basis need not be so sophisticated and comprehensive as a theory, nor as precise as a principle. For example, perhaps the Mengzians just recommend a certain picture or conception of the virtue of humaneness or benevolence (ren 仁), which can account for the fact that caring for non-citizens strangers is right or good. That recommendation establishes a standard that could qualify as an ethical basis too, even if the conception of humaneness is roughly hewn and presupposes some open-ended form of know-how on the part of the people who adopt it. Another possible source of confusion is my use of “caring.” “Caring” can refer to various kinds of attitude (as in, “Lanfei cares about the people of Botswana”) and to various kinds of activity (as in, “Lanfei is taking time off from work to care for her injured mother”). For purposes of the main argument, I am most interested in the ethical basis for something like virtuous care, which is at minimum an attitude but often has direct implications for action as well. As a caring brother, I should at the very least care about the welfare or interests of my siblings, and care about them in part for their sakes and not entirely for my own. Often, this will entail that I should act to protect or promote their interests, but not always. Sometimes, it would be impossible, unhelpful, or excessively burdensome to act to promote or protect their interests, as when a younger sibling wants very much to complete a task in which she takes personal pride without the gratuitous and heavy-handed intervention of her older brother.

The interpretation of the Mengzian line that I will defend is a rela tively moderate and nuanced one. Over and against the strongest readings of Confucian ethics (including Mengzian ethics) as having an all-encompassing commitment to human relationships and relational conceptions of the self and its ethical values or concerns, I will argue that Mengzi and the later Mengzians do have ways of accounting for the value of caring for non-citizen strangers. More controversially, I will contend that no “social relationship” in any meaningful sense of the phrase is required to establish such norms. However, I will contend that forming human relationships out of other-directed care or concern is nevertheless a valuable goal and ideal by Mengzian lights, and that Mengzian ethics is centrally concerned with social relationships in that specific sense. Furthermore, I think that even if Mengzians do not require “‘social relationships’ in a meaningful sense” to do the relevant work, their ethics is still nevertheless relational in a thinner sense—namely, insofar as the basic ethical claims made on us by other creatures depend in part on what kind of creatures we and they are. Species sameness (as I will call it) is a relational property, and so Mengzian ethics is relational in this thinner sense.

Like me, some readers are highly motivated to have an accurate and nuanced account of specific Confucian philosophers, and for those readers, I hope that there will be plenty of novel material to digest here. Even for those who have no prior interest in understanding the Confucians on their own terms, however, I hope that this paper will be of use. The position that I attribute to the Mengzian line of thinkers is both more psychologically realistic and more consonant with most people’s considered judgments than other influential ethical theories and worldviews. For example, one of the influential ways of thinking about the basis of our obligations to non-citizen strangers is what we might call radical cosmopolitanism about statehood and citizenship. According to this view, when an individual is in great need of aid (for example, if they are in imminent danger of dying from starvation), it is ultimately morally arbitrary and irrelevant whether they happen to be citizens of the same state or of a different one. Just as we should intervene to save a dying child no matter her family name or hair color, so too should we intervene to save a dying child no matter her citizenship. But radical cosmopolitanism struggles to explain how to set reasonable limits to the demandingness of its views about citizenship and statehood. Famously, some cosmopolitans suggest that at least some of us (e.g., those of us who are capable of great acts of altruism) must give away all of our personal wealth and resources to the point of near destitution, if doing so can save distant famine victims from death. Insofar as the citizenship of the victims matters, it matters because it factors into considerations of what is motivationally or psychologically feasible or plausible, not because it alters the fundamental weight or priority of our obligations to others (Singer 1972; Unger 1996). Most people, however, seem to think that we really do have weaker fundamental obligations to non-citizen strangers than to fellow citizens, and it is psychologically unrealistic to expect that people will live up to the considerable demands of radical cosmopolitanism. In being both more psychologically realistic and more consonant with ordinary views, Mengzi and the Mengzians offer a framework that shows much more promise than radical cosmopolitanism. As I read the Mengzians, their framework achieves this without falling back on strongly nationalistic or tribalistic views that offer no fundamental basis for caring about non-citizen strangers. Furthermore, scholars of global justice in the twentieth and twenty-first centuries have not much indulged the temptation to spell out obligations to non-citizens in terms of human relationships, a temptation that I will indulge here for the purpose of pushing it to its limits. For readers interested in these more present-minded issues of civic identity, global justice, and relational ethics, there should be much to mull over as well.

In what follows, I will begin with an explication of an important feature of humane virtue known as “care with distinctions,” as interpreted by the Mengzian line of thinkers. I will then explore whether and in what respects the Mengzians assumed that relationships and relational ethical frameworks provide a basis for caring about non-citizen strangers. Finally I will conclude with a brief discussion of some strengths and weaknesses of the Mengzian position as well as some future avenues of inquiry on this topic.

Historical Confucians developed compelling arguments for a certain conception of virtuous care that is sometimes called “graded love” or “care with distinctions.” In most cases, this is seen a requirement or component of the virtue of humaneness (ren 仁), quite arguably the most important of the Confucian virtues. When someone instantiates humane or proper care with distinctions, they care most about and have the most demanding ethical obligations toward their immediate family members. They then care somewhat less for (and have somewhat less demanding obligations toward) good friends and more distant family members. They might have a smaller degree of care for and obligation toward members of their respective residential and work communities, and so on for people more distantly related to the agent in question. For most Confucians in what I will call the Mengzian line, the differences in degree of care and obligation are basic and not derivable from a deeper commitment to fundamentally impartialistic ethical norms. For example, it’s not the case that a Confucian should care more for her immediate family members just because she can make a greater contribution to human welfare overall by focusing on the people that she knows and interacts with most.²

Those of us who teach Confucianism on a regular basis sometimes use the image of concentric circles to illustrate this schema. Depending on the particular Confucian in question, at the outermost ring are non-human animals, flora, or even insentient artifacts like tiles (e.g., see Mengzi 7A.45; see also Wang Yangming, “Questions on the Great Learning” [Daxue wen 大學問], §1 [in Wang 1992, 2009]). But some steps in from the outermost ring are human beings in general, whom humane moral agents presumably care about in virtue of the fact that they are human and not in virtue of their nationality or citizenship. On my interpretation, that ring—the one for human beings in general—is the locus of cosmopolitan attitudes, on the Mengzian account.

If there is much hope of developing a basis for caring for non-citizen strangers that is consistent with core Confucian doctrines, it must be of the more moderate variant. “Care with Distinctions” seems to endorse some obligations to non-citizen strangers, but clearly weaker or less demanding ones than to strangers within one’s own nation-state (see Bai 2020, 175–78; Tsai 2020, 77–79). Confucians came to refine, embrace, and build an ethical worldview around care with distinctions over the course of many centuries of debate with philosophical and religious rivals, at first in opposition to Mohist advocates of impartial caring (jian’ai 兼愛) and later in contrast with Buddhist proponents of great compassion (daci 大慈) (see Mengzi 3A.5, 3B.9, and 7A.26; Tiwald 2018a). It is thus very much at the center of Confucian ethical and political thought.

I find at least two influential Mengzian arguments for care with dis tinctions. First, care with distinctions is necessary for special hu man relationships, which are a central and indispensable part of the human good and the ethical life. To treat someone as a parent just is to regard their interests as having a stronger fundamental ethical claim on oneself than the interests of strangers, and to live without such special relationships is to live without essential human goods. Mengzi hints at this view with reference to the idea of having a father—to have a father, he suggests, is to treat the father as being owed very special consideration and deference. If one regarded the interests of all people as making the same (basic) claim on one’s self as one’s father’s interests, there is a sense in which one really would not have a father and thus would be deprived of a relationship that makes one a proper human being (see Mengzi 3B.9).³ A similar point could be made about fellow citizens: to share citizenship with someone just is to be committed to being a reliable source of aid and comfort for them in the event of a disaster, to give them some priority over non-citizens when undertaking certain sorts of civic projects, and so on. Second, any system of ethical value that fails to take proper account of our natural inclination to care about those near and dear to us—especially immediate family members—simply will not be adequately grounded or “rooted” (ben 本) in human nature, which is sure to make for all kinds of ethical and perhaps metaethical mischief.⁴ Quite arguably, caring more about people with whom one shares a government and a political community comes more naturally than caring about non-citizen strangers. So if there is to be a way of accounting for obligations to non-citizen strangers that is compatible with the deep and historically durable grain of Mengzian ethics, it would need to treat non-citizen strangers as having a significant basic claim on moral agents, but also a weaker basic claim than fellow citizens or subjects of the same nation-state. To demand that we regard total strangers as having the same basic claim on us as fellow citizens or family members would run too strongly against our natural tendencies.

As noted in the introduction, it is a commonplace to say that for Confucians, special human relationships (such as family, close friends, neighbors, community members) are more central or important than they are for proponents of other ethical systems or worldviews. There are multi ple ways of spelling out this idea. One relatively obvious way is to say that for Confucians, special human relationships are a larger part of the good or virtuous life, such that they think we should devote more time, attention, or effort to developing close, personal relationships than to other endeavors. For example, it might be that Confucians should be more committed to supporting projects and enterprises that are good for their community members and fellow citizens than for people with whom they share no formal social or political bonds. On this view, Confucians should still have some concern for non-citizen strangers, but perhaps less concern than adherents of stronger sorts of cosmopolitanism. Maybe a cosmopolitan living in a more permissive and individualistic culture would say that they can be a moral paragon while devoting 80% of their energies to world peace or a global vaccination campaign, while a Confucian would demand of its moral paragons (sages) that they normally set aside at least 50% of their energies for family and more immediate community members and thus allocate no more than 50% for global projects. It is plausible that special human relationships are more important for Mengzians in roughly this way, although much will depend on many specifics that we cannot adequately address here. For instance, a great deal depends on how we specify what conditions count as “normal” and thus justify the normal allocation of the sage’s energies, and what conditions might be “extraordinary” enough to permit a sage to devote more to public goods. Mengzi appears to endorse the view that the legendary water engineer and tamer of floods Yu was so occupied with his work during the great floods that he passed the gates to his home three times without entering to see his family (Mengzi 3A.4). If the ethical demands made by the flooding of Yu’s time are comparable to the ethical demands of world peace or pandemic diseases in the present day, then perhaps the Mengzian sage would be willing to sacrifice nearly all of her time with family and the local community for more public goods. I doubt that the traditional texts will give us much guidance in this matter (I think the issue is underdetermined by the texts themselves). However, I share with many scholars the impression that most traditional Confucian philosophers (including Mengzi) saw personal human relationships and the virtues that sustain them as more ethically important than many modern peoples do.

There is a subtler way of spelling out the idea that Confucians regard special human relationships as more central or important, one that has different sorts of implications for the moderately cos mopolitan component of Confucian ethics. Rather than ask how much of one’s time, attention, and effort one can or should devote to the interests of non-citizen strangers, we could ask whether it is desirable or worthwhile to make them less strange, more familiar, and thus closer to one’s circle of “near and dear” compatriots in the first place. Perhaps good Confucians enact their commitment to close, personal relationships by striving to make their major projects more personal and relationship-oriented. For example, maybe Kai would strive to better acquaint himself and form social bonds with the non-citizen strangers that he aims to provide vaccinations to, and thereby make them more like friends or colleagues.5 There is a strand of Confucian thought which strives to build intimacy even with total strangers, evident, for example, in longstanding views about how hosts should properly treat their guests (Ivanhoe 2024).

To unpack the implications of this idea for more careful consideration, I find it helpful to look for historical parallels. In some respects, the idea of making colleagues or friends of non-citizen strangers mirror a strategy implicit in the surviving works of the second-century Greek philosopher Hierocles, usually regarded as a Stoic thinker and an early cosmopolitan. In Hierocles’s discussion of “familiarization” (oikeiôsis), he uses the image of concentric circles to represent the relationship of the self to humanity more broadly, with the self at the center, immediate family members in the next ring out, and so on until we reach human beings in general in the outermost ring. It seems that for Hierocles, it is natural that we react differently to people depending on their place in the concentric rings, but he nevertheless thinks it desirable and good to bring those in the outer rings closer, and even goes so far as to propose some ways of making that happen—for example, by using slightly more intimate terms of address for those more remotely related (calling male cousins “brothers,” calling people with no relation “cousins,” and so on).⁶ This strategy has the advantage of making major concessions to human nature while still holding up civic cosmopolitanism as a meaningful, action-guiding goal. Perhaps Mengzians could endorse it as well?

Inasmuch as Mengzians could adopt a strategy of familiarization like this one, it would need to be more fine-grained than Hierocles’s. At minimum, it would need to distinguish between attempts to familiarize non-citizen strangers by making them more like acquaintances and attempts to familiarize them by making them more like family. We can see this in part by looking at Mengzi’s response to a somewhat similar strategy proposed by a Mohist contemporary of his. In a proxy debate too subtle and contested to reconstruct here, a Mohist rival named Yi Zhi defends what appears to be a somewhat more sophisticated understanding of impartial caring, according to which “care is without distinctions, but is bestowed beginning with one’s parents” (愛無差等, 施由親始). On a plausible interpretation, he thinks that people learn to care for others by indulging their natural affection for family members, but then strive over time to care similarly for non-family (Mengzi 3A.5).⁷ This is the very ethical worldview that Mengzi criticizes as having “two roots” or two foundations and causing ethical mischief, probably because one of the foundations runs against the grain of human nature (Mengzi 3A.5; Nivison 1996; Van Norden 2007, 305–12). Mengzi seems to find the very aspiration to blur the distinctions between family and non-family most objectionable. So much of Confucian ethics is premised on the presupposition that we should harness and nurture rather than resist or undercut the subtly different ways in which we are naturally predisposed to treat members of our own families (Mengzi 3A.4). That said, it does not seem a stretch to think that Mengzians would endorse other, non-familial ways of familiarizing non-citizen strangers, such as Kai’s attempts to make the beneficiaries of his vaccination campaign more like friends or colleagues and less the faceless masses who receive the fruits of his beneficence from afar. Personal acquaintance and mutual reliance are important constituents of the good life as they see it, and also critical sources of more sincerely or wholeheartedly virtuous motives. Familiarization of this sort has a natural home in the ethics of Mengzi and the Mengzians.

In the previous section, I attempted to spell out the Mengzian way of accounting for the goodness of caring about non-citizen strangers. A specific concern was whether and to what extent this Mengzian approach is consistent with various claims about the greater importance of relationships and relationality for Confucianism (including Mengzianism) than for other ethical worldviews. So far, I have pointed to two ways in which it can be consistent with such claims: it might be the case that Mengzians believe a greater proportion of a person’s time, effort, or energy should be devoted to building and contributing to personal relationships than to more public or impersonal endeavors, and it might be that Confucians should strive to personalize or “familiarize” the non-citizen strangers that they work with or for. On at least one influential reading of Mengzian ethics, however, this does not go far enough. According to a strongly relational interpretation Mengzian ethics that is sometimes (in twenty-first-century Confucian scholarship) called “role ethics,” all of our ethical values and commitments are derived from our relationships or roles. On this view, there is no such thing as an obligation to care for a someone just in virtue of facts or features of that person alone, independently of their relationship to the moral agent. All of my ethical obligations (and other ethical norms) are derived from my relationship to them—say, as a work colleague, a teacher, a parent, a sibling, or a fellow citizen. This suggests that Mengzian ethics is more deeply relational or relationship-oriented than my two previous concessions have allowed. It also seems to pose challenges for the moderately cosmopolitan reading that I have advanced. I seem to read the Mengzian line of thought as suggesting that we have obligations not to participate in genocide against strangers in foreign lands in virtue of the fact that they are human beings. Is this consistent with the role-ethical interpretation of Confucianism? I turn to this issue next, both to explore the limits of the widespread claim that Confucian ethics is “relational” and to look more closely at some relatively neglected questions about whether relationships really can support a cosmopolitan concern for non-citizen strangers.

To spoil the conclusion, the view that I will defend is as follows. Mengzian ethics is relational in the thin sense that the relevant ethical values are derived from a relational property, which I will call “species sameness.” But it is not relational in a stronger sense that seem to be important to some interpreters of Confucianism, which banks on what I will call “social relationships” (e.g., teacher-student, friend-friend, members of the same team or community). For Mengzians who take seriously that humaneness requires care with distinctions, the relevant relation to non-citizen strangers is that they are fellow human beings— that is, members of the same species. I will explore the temptation to say that there is some sense in which virtuous Mengzian agents are in social relationships with non-citizen strangers, and explain why I think it is a mistake to indulge this temptation. We can only establish that we are in social relationships with the strangers that we should care about by effectively trivializing the idea of a social relationship, which we should not do.

Let me start with a crude description of relational ethics and then refine it a bit. We might say that an ethical worldview is relational just in case it takes all first-order ethical norms to depend on the ethical agent’s relationships to others. On this view, for example, there is no such thing as specifying what, exactly, Jun should do when he meets Mei at the door to his home without reference to the relationship between Jun and Mei. Is Jun a prospective client and Mei a salesperson? Is one a host and the other a guest? Are they son and mother? How Jun should greet Mei always depends on the answers to these questions. To be sure, there will be some situations where the question “what should Jun do in this situation?” can be safely answered without knowing much about his relationship. Let’s say that Jun opens the door to Mei and finds that Mei is having trouble breathing: whether Mei is Jun’s mother or a stranger or a business partner, his first reaction should probably be to administer or summon emergency aid. Still, one of the factors that will regularly determine how Jun responds to Mei will be his relationship to Mei. Perhaps the commonplace that Confucian ethics is “relational” just means “relational” in this rough sense.

This preliminary formulation of the view is too crude. When scholars say that Confucian ethics is fundamentally relational, they usually take this to be something distinctive about Confucianism, and the preliminary formulation does not yet say anything distinctive. For example, many non-Confucian philosophers, including the ones that are supposed to epitomize Western ethics (such as Kant and Plato) would have no difficulty conceding that a person’s relationships are among the factors that should be taken into account when spelling out the content of first-order ethical norms. Both acknowledge that the precise features of good or virtuous behavior will depend on whether one is interacting with a spouse, a son or daughter, a teacher, a work supervisor, etc. Furthermore, if we do not say more about what we mean by “relationship,” we risk reading it in a way that makes a platitude of the preliminary formulation. For example, many moral philosophers will say that my behavior toward person X should depend at least on whether I am aware of X’s existence or probability of existing in the future. If X has lived all their life on the far side of a distant planet, totally unbeknownst to me, that obviously has important implications for what I should be doing about X. In describing the fact that I am not even aware of X’s existence, I am describing something about my relationship with X . . . at least in some sense of “relationship,” and no one would deny that relationships in that minimal sense are relevant to ethics.

A more controversial and distinctive version of the claim says something much stronger about the basis for Confucian ethical norms: namely, that when describing ethical agents and their norms, the only relevant considerations are relational. There is nothing about the person independently of their relationships to others that will matter, except insofar as they bear on a person’s relationships. To go back to the scenario where Jun opens the door to Mei and finds that she is having trouble breathing. Perhaps in that case, he will be obligated to render emergency aid even before establishing how he is related to her, but that’s only because he can safely generalize across many different roles, not because there is some role-independent way of specifying how he should react. This stronger understanding of relationality contrasts most clearly with views that attribute great ethical significance to intrinsic features of individual persons, such as self-consciousness, various kinds of autonomy or capacity to will, or the capacity to engage in certain kinds of reasoning or reflection. Famously, Kant seems to think that people have moral standing in virtue of being autonomous and presumed capable of practical reasoning, and these appear to be intrinsic features of them as individuals (Kant 1998). So Kant, at least, offers an ethical framework that is incompatible with Confucian relational ethics so conceived.

This revised version of Confucian relational ethics is my attempt to give a charitable reading of what is sometimes called “Confucian role ethics.” According to the most prominent proponents of Confucian role ethics, Roger Ames and Henry Rosemont Jr., an important feature of Confucian role ethics is that it is based on what they call a “relational conception of persons” or “role-bearing persons.” Everything important about a person—including the ethical norms by which they should be assessed—follows solely from the fact that the person is (or maybe should be) in some sort of relationship with other people. For example, my obligations to other people follow solely from the fact that I am related to them as a son, sibling, spouse, colleague, teacher, neighbor, and so on, rather than from the mere fact that we are all human beings or agents with moral standing. Rosemont says that once these sorts of roles have been specified, there is “nothing left over with which to piece together an individual self” (2015, 49; see also 14, 94).

Earlier I pointed out that if one construes “relationship” broadly enough, the claim that Confucian ethics is fundamentally relational turns out to be trivially true—everyone thinks that relationships to others matter in some sense (at the very least we have to have the sort of relationship that allows us to know the probability that the others exist or will exist). Rosemont and Ames do not seem to have “relationships” in that very loose sense in mind. Rather, they have in mind relationships that can be characterized in terms of social roles, things like being a parent, child, sibling, student, teacher, coworker, or spouse. Ames describes these as “family roles and extended relations we associate with community” (2011, 168). When Rosemont gives examples of roles in the role-bearing person as he envisions that person, he mentions relationships like being a father and neighbor, and not relationships in the thin sense of merely knowing a person’s existence (Rosemont 2015, 49). Let us give the name social relationships to those relationships that depend on each party adopting a social role in this way. If the role-ethical account of Mengzian Confucianism is true, it poses a major challenge to the very idea of Mengzian cosmopolitanism. For moderate civic cosmopolitans, they see themselves as having at least some meaningful commitments to protecting or promoting the interests of non-citizen strangers, and yet it would seem that on the role-ethical reading of Confucianism, such commitments can only be warranted when the interest-bearer has some sort of relationship with the non-citizen strangers. This then leaves Mengzian cosmopolitans with a dilemma: either they would have to concede that they have no fundamental ethical norms at all with regard to non-citizen strangers, or they must contend that they are, in some sense, in social relationships with non-citizens strangers after all. Let’s examine each horn of this dilemma.

To be sure, Ames sometimes flirts with the suggestion that there are no direct ethical commitments or values between people who do not have any social relationship with one another. In one of his early papers on relational persons, he appears to acknowledge that by his lights Confucians would not be able to make sense of direct obligations to people outside of their communities (Ames 1988, 210–11). But I think it is relatively safe to say that for most scholars of Confucianism, this view is neither appealing nor plausible as an interpretation of traditional Confucian philosophy. On most any reasonable reading of the major Confucian texts, a virtuous person should be at least decent and somewhat kind to non-citizens strangers, and so should people who are merely moderately virtuous or superficially good. The very idea of a person constituted entirely of their social roles depends on an implausible reading of the metaphysics of personhood in Confucianism, one that overstates the significance of relationships at the expense of the virtues and human nature (Ivanhoe 2007). I also doubt that early Confucians like Kongzi and Mengzi had systematic views about an issue in metaphysics that was never thematized—never a point of debate and explicit theory-building—before or during their time (Tiwald 2022).

The other option is to say that when a Mengzian Confucian has ethical norms that pertain to non-citizen strangers, there is some “relationship” to that non-citizen stranger after all, appearances to the contrary. I find this option much more interesting, in part because, for all of the talk of relationships in twenty-first century ethics and Confucian philosophy, there is very little discussion of what counts as a relationship in the relevant sense. Perhaps we can tell a story to explain how I am in fact in a relationship with non-citizen strangers in faraway places. After all, we are contemporaries living in a world that is by most accounts increasingly interconnected and mutually dependent. In some attenuated but nevertheless meaningful sense, we depend on one another. Maybe Mengzian cosmopolitans could point to these facts (or facts like them) as a basis for caring about faraway peoples. If that is the case, maybe Mengzian cosmopolitanism turns out to be fundamentally relational.

This raises difficult questions about the nature of relationships that can be characterized as social ones (“relationships” in the thick sense that Rosemont and Ames have in mind). I think we can make a start at addressing them by considering what it is for two people to be in a social relationship. Here it might help to spell out the special significance of relationships in terms of what present-day philosophers sometimes call particularity and fungibility. If Jiaying is a close friend of mine, I value her in part for the bare particular that she is, rather than solely for her traits or other characteristics. That is why, if Jiaying dies before I do, I would mourn her loss even if I happen to find a new friend with exactly the same characteristics. I care about Jiaying for her sake, for the sake of the bare particular, at least to a significant extent. The same is true to an even greater degree for the immediate members of my family. In contrast, when I act for the sake of people with whom I am not in a social relationship, my ethical commitments and attitudes are more fungible. If I decide that it would be good to contribute to a relief fund for earthquake victims in a country with which I have no personal connection, it does not matter much which particular individuals I support. If my research shows that the philanthropic organization that supports community A is more wasteful than the organization that supports community B, it could well be rational to give my money to the second organization instead. In contrast, the fact that my time and resources might do more good for a stranger than for my sister does not make it rational for me to give them to the stranger in lieu of my sister. My care for my sister is not so fungible.⁸

It also helps to specify social relationships in terms of shared ends. To share ends with Jiaying (my close friend) is for both of us to take one and the same outcome or state of affairs as having final value for ourselves. As I have argued elsewhere, not all shared ends are indicative of relationships. Sometimes two people can share an end without being in any relationship with each other whatsoever, as when two fans of the same sports team both fervently want that team to win a championship. For people truly in relationships with one another, the shared ends are “other-mediated”—it is in part for the sake of the other person that each person adopts the shared end. For example, if I have a hobby that I pursue with Jiaying, and this is an end that we share in the relevant sense, I regard this hobby as having final value for me because (in part) it is something I can enjoy with Jiaying. I value the hobby in part for her sake and for the sake of our friendship. My sharing of ends with her is other-mediated in this way, unlike in the case of sharing ends with strangers who happen to be fans of the same sports team (Tiwald 2020, 112-17).

Here are two plausible criteria for my being in a relationship with someone: first, I should care about that person for the particular individual she is and not regard my commitments to her as fully fungible; second, I should share ends with her, and share ends in an other-mediated way. I suggest that most of the ways in which we are related to total strangers in other nations or states fail to meet these criteria. More to the point, in most cases where Mengzians do believe we have obligations to total strangers in other nation-states, it would be a stretch to say that we have other-mediated shared ends or care about particular individuals non-fungibly. Perhaps a humane Mengzian leader believes that she should join forces with leaders of every other nation to reduce the threat of nuclear annihilation. That is a shared end, surely, but it is not meaningfully other-mediated. Perhaps a humane Mengzian citizen donates to a philanthropy that provides temporary shelters to thousands of dislocated Syrian families. It is unlikely that care for the particular victims plays a meaningful role in providing the ethical motivation or justification for her doing so. Accordingly, I do not have much hope that social relationships alone can serve as a basis for Mengzian cosmopolitan norms.

That said, there may be another sense in which Confucian ethics is fundamentally relational, albeit one that is quite different from what Ames and Rosemont have in mind. On the interpretation that I find much more defensible, the Mengzian basis for caring about non-citizens strangers is that they are “fellow human beings.” It is in virtue of being fellow human beings that humane moral agents care about them. On its face, for someone to take a non-citizen stranger as a fellow human being is to see the stranger as having a relationship to them—specifically, it sees the stranger as a member of the same species as them. This is not a “relationship” in the particularist, non-fungible sense that scholars like Ames and Rosemont seem to have in mind, but it is a relationship insofar as it inheres in a relational property (species sameness).

There are two ways of construing the ethical significance of insight that non-citizen strangers are fellow human beings. On one construal, what matters most is just that they are human beings. The fact that I happen to be of the same species is incidental, and not very significant for purposes of figuring out what I should do about their suffering. If I were of another species, I would still have the same ethical norms with respect to them, because what matters is their being human and all that entails. On another way of construing it, however, the fact that they are fellow human beings—that we belong to the same species— is ethically significant. Our similarity in kind is part of what puts non-citizen strangers several rings in from the outermost of the concentric circles, what should distinguish my ethical commitments to them from my ethical commitments to non-human animals, plants, and so on.

The difference between these two construals is subtle. In my experience, people who are not mindful of the difference often drift or waver between both interpretations—sometimes suggesting, for example, that a person’s humanity taken on its own gives that person special moral standing relative to non-humans, and sometimes suggesting that we should treat fellow human beings in special ways because they are our fellows. Since it is easy to slide between the two interpretations, I am somewhat reluctant to attribute a consistent and systematic view about the matter to traditional Confucian philosophers. Still, it is striking how many of the major Confucians make a point of saying that fellow human beings are owed greater care from us, seemingly in virtue of their being members of the same species. Wang Yangming uses language to suggest that being a member of the same species is part of what justifies and motivates the right degree and kind of concern for strangers (“Questions on the Great Learning” [Daxue wen 大學問], §1 [in Wang 1992; 2009]). Dai Zhen says that humaneness is an outgrowth of our natural affinity for members of our own species, an affinity that arises when we notice similarities between ourselves and other humans. For example, Dai thinks that we rarely see intra-species cannibalism because sentient living creatures are more likely to care about the life of creatures that closely resemble themselves.⁹ Given the very important role that biological connections and resemblances tend to play in Confucian ethics more generally, it doesn’t seem a stretch to me that Confucian ethics might construe “fellow human beings” in the more relational way. At least for many of the authoritative Confucian philosophers who took themselves to be following Mengzi, it could be that all first-order ethical norms are deeply dependent on at least one important relational property, the property of “being of the same species” or “being of a different species.” If this is correct, then Mengzian ethics can build in concern for non-citizen strangers even while being fundamentally relational, although it is relational in a much thinner sense than the proponents of role ethics have suggested.

There are many things meant by “cosmopolitanism,” even within disciplinary philosophy and political theory (Brown and Held 2010, 1–14). And there are many things meant by the cluster of claims that Confucian ethics is more deeply or centrally relational than other influential ethical worldviews. Rather than try to untangle all of the conceptual threads and arguments in the space of a short article, I have tried to establish a foothold in these larger issues by getting clear about two important claims. The first has to do with how a particular strand of Confucian thought—the strand that theorizes about “care with distinctions” as Mengzi understood it—can go about accounting for the ethical value of caring for non-citizen strangers. Among the many overlapping notions of cosmopolitanism under discussion today, most share a commitment to the idea that the interests of non-citizen strangers can have some significant ethical claim on moral agents, and yet it is this very (minimal) claim that can potentially cause trouble for more relational and relationship-oriented ethical worldviews. The second issue has to do with the “relational” nature of this Mengzian strand. For reasons given in the previous two sections, I think that it qualifies as relational in certain very specific senses, and that very strong characterizations of its relationality—namely, those found in the “role ethics” account—overstate the degree to which Mengzian ethical thought is relational. Still, even if we take Mengzian ethics to be relational only in my thinner sense, my reading raises questions and challenges that call for more careful study. In this conclusion, I will briefly gesture at those questions and challenges.

To start, there are some advantages of the Mengzian approach to care for non-citizen strangers. Often, among philosophers who work on the ethics of special relationships, there is a temptation to divide up the normative terrain between the ethics of partiality and the ethics of impartiality, with the former covering cases where we have agent-neutral reasons to protect some person’s rights or interests regardless of our relationship to her (e.g., in providing life-saving aid to someone in great danger) and the latter covering special cases where we have agent-relative reasons to do something for someone with whom we are in a special relationship (e.g., in showing special affection for a lover, or in going to extraordinary lengths to help a close friend—these are things that moral agents who are lovers or friends have reason to do, but not random passers-by). But generally speaking, there are not well-integrated theories or models to explain when a departure from impartialist norms is justified. Even among the most sophisticated defenders of partialistic ethics, the partialistic norms tend to look like an arbitrary appendage to an otherwise impartialistic moral theory, without a theoretically or conceptually coherent picture to account for them both.¹⁰ In contrast, for the Mengzian view as I have explicated it here, there is a sense in which all ethical norms are partialistic— even supposedly public-minded obligations to render lifesaving aid to other human beings are based on species sameness. Other species, on the thinly relational view, have weaker obligations than we do to help humans in distress. Furthermore, there is a sense in which the conceptual apparatus invoked by Mengzian “care with distinctions” is one that most human beings find intuitive, even if the boundaries between the concentric circles (or the exact amounts of concern owed to each) are somewhat fuzzy. There is a good case for thinking that care with distinctions is a psychological natural kind, one that draws upon concepts (such as proximity, intimacy, social and familial closeness and distance) that most human beings are well equipped to grasp and apply.¹¹ There is a theoretical unity and elegance in the Mengzian view that is rarely found in other attempts to account for the ethics of special relationships.

I myself am interested in ethical worldviews that human beings are capable of adopting wholeheartedly, such that they experience their ethical commitments as ones that they are comfortable with, ones that they sincerely cherish and want to maintain, and not as ones that feel alien or require them to perform “moral” acts at the cost of great personal regret or loathing (as would be the case for ethical theories that require that people sacrifice their children’s welfare for the greater good, or capture and harvest the organs of one healthy person to save several others) (see Tiwald 2018b, 179–83).¹² It seems to me that Mengzianism is much more likely than other theories or worldviews to be something that we can wholeheartedly embrace. Most notably, Mengzian ethics is supposed to hew closely to our natural tendencies. More specifically, with regard to its approach to motivating humane behavior toward non-citizen strangers, it seems to recommend ways of “familiarizing” them, so that our concern for them comes more effortlessly and sincerely than would be the case for cosmopolitan duties toward faceless and unfamiliar others.

There are also some worries about the Mengzian account that warrant more investigation. Species sameness may seem a relatively clear and unproblematic way of accounting for the ethical value of caring for human beings that we are unacquainted with, but there is at least the specter of more problematic views at the periphery. One wonders what Mengzi or Dai Zhen would say if, for example, we discovered a species of creatures very much unlike ourselves that never theless have the complex inner lives and capacities that would, on less relational views, warrant special moral standing or considerability akin to humanity’s. By the lights of the Mengzians, would we rightly care far less about the lives of this other species than about the lives of other human beings? This is just one sort of objection among others that might be raised by people who doubt that ethics can, at the end of the day, rest on relational or agent-relative foundations alone, that some sort of direct accounting of the relevant non-relational ethical facts is necessary in order to be minimally responsible as a moral agent. Even so, present-day Mengzians may well have the resources to respond to such worries. In any case, the Mengzian approach to care for non-citizen strangers seems to have more promise than many competing attempts to explain when and why we should take an interest in the welfare or dignity of people who are, to one’s self, really just fellow human beings and no more.