Shared Ends: Kant and Dai Zhen on the Ethical Value of Mutually Fulfilling Relationships

1. Introduction

I start from an assumption that most people (in most times and places) will share, which is that ethics is substantially concerned with relationships between particular individuals. To give a full accounting of our day-to-day successes and failures as ethical agents, we would need to talk about how and in what respects we meet and fall short of ethical norms that arise in part because of our obligations to be good parents, children, siblings, friends, neighbors, supervisors, or colleagues. Furthermore, I assume that some of the distinctive goods that we realize through these relationships require what I will call shared ends. Roughly, these are outcomes or states of affairs that two or more ethical agents value noninstrumentally, goals that they both (or all) regard as valuable independently of whether or not they promote some further good or valuable state. Moreover, sharing ends with people is part of what it means to be in certain kinds of relationship with them. If I do not have some common interests with my would-be close friends, if their welfare and important projects did not have some final value for me, then there is a sense in which they’re not really my close friends. Something similar is true for relationships between family members, and perhaps even work colleagues and neighbors or members of the same club or community.

Many moral philosophers would agree that it is good for us to develop and enjoy shared-end relationships and that doing so may be ethically valuable, insofar as we must develop good shared-end relationships in order to be good parents, friends, colleagues, and so forth. However, few Western moral philosophers provide us with well-developed accounts of the nature and value of good shared-end relationships, and it is unclear whether they have the right conceptual apparatus to do so. With those issues in mind, this paper begins to develop a positive account of good shared-end relationships and their value. In the interest of highlighting some strengths and weaknesses of competing approaches, I contrast the accounts that are suggested by the Confucian philosopher Dai Zhen 戴震(1724–1777) and the influential moral philosopher Immanuel Kant (1724–1804).

My basic thesis is that Dai Zhen’s approach to thinking about the nature and value of good shared-end relationships is superior to Kant’s because it highlights the fact that good shared-end relationships must be motivated by ethically-shaped forms of other-concern and self-interest. As we will see, this view contrasts with Kant’s approach in part because Kant does not picture self-interest as an important source of morality or ethically valuable relationships. Of course, it is possible to revise Kant’s theory so as to better accommodate such motives. Nonetheless, Kant’s writings are so systematically devoted to clearly distinguishing natural inclination and self-love from what he regards as the proper source of moral motivation that such changes come at considerable cost to the integrity and core commitments of his moral philosophy. By contrast, Dai quite arguably built his entire ethical theory around what I will call good relationships of mutual fulfillment. At the same time, Dai was in certain respects a kindred spirit to Kant: he sought to articulate a system of thought based on the assumption that we should value others as we value ourselves, that treat the interests of others as mattering for their sake (rather than instrumentalizing or subordinating their interests to one’s own good or a greater good), and that conceives of moral norms as those which all parties can in principle affirm. Dai thus offers a way to account for good shared-end relationships that many Kantians would be inclined to endorse, or at least more inclined to endorse than, say, straight-forward utilitarian or other consequentialist justifications.

This article is organized as follows. Section 2 explains how self-interest factors in among the rich mixture of motives necessary for such relationships. Section 3 describes relationships based on shared ends, called “relationships of mutual fullilllment,” and notes some conditions that make some of them (a subset) good ones. Section 4 gives a brief overview of Dai Zhen’s ethics, one that aims both to highlight some of the more Kantian features of his framework and to show how it serves to justify the rich mixture of motives necessary for relationships that are mutually fulfilling and good. Section 5 explains why both Kant himself and a promising revision of Kant’s ethics cannot account for the comparative ethical value of these relationships.

2. Mutual Fulfillment and Self-Interest

In order to have meaningful relationships in the relevant sense, most of us will need a rich mixture of motives that include some self-interested ones. Consider close friendship again. To be a close friend with Jiaying, I should have some common interests with her. These might be a shared hobby or social cause or intellectual pursuit, but our common interests should also include (1) maintaining our friendship, and (2) one another’s well-being and success. Part of what makes me a close friend to Jiaying is that I take her happiness, contentment, and flourishing to have some final value for me, so that I derive gratification, from them. But if pressed to explain what motivates me to support Jiaying and cultivate our friendship, I would answer that it is a complex array of care for Jiaying, shared interests, and self-concern. I am partly motivated by my care for Jiaying for her sake, my love of the friendship as such, and the benefits that friendship has for me. If I am helping her in some ambitious project of hers, I reflect on how her success will strengthen our friendship, how we will someday reminisce about the time that she undertook the project and prevailed with support from friends like me, and this strengthens my resolve and commitment to helping her and enhances the joy that I derive from doing so. If Jiaying is in peril and needs my help to avoid catastrophe, I help her because I cannot bear to see her life fall apart, but it also helps to think (and seems constitutive of a good friendship to think) that she would do the same for me if I were in peril, and that gives me some satisfaction and contentment which figure prominently in the motivational set that helps to sustain our friendship.

Perhaps a saintly person would feel differently—she might care about her friend’s success and welfare entirely for the friend’s sake, with no thought of its benefits for herself, and care with as much passion and commitment to the friend’s success as I have for Jiaying’s success. Quite often, in moral philosophy and in public discourse about ethics, we see people praise and prioritize this sort of self-lessness or purity of motive. Whether they are right to prioritize it depends on its exact implications for the sorts of ethical character and behavior we should value and strive to emulate, but in most cases and for most purposes, it is a mistake to do so. First, vanishingly few of us are saints or potential saints. In almost all cases, we need a rich array of mixed motives in order to have the right sort of commitments to friends and derive the right sort of satisfactions from friendship.¹ And in those overwhelming number of cases, it is better to be a close and deeply invested friend than a distant but altruistic one.

Second, even if sainthood were a live option for us, I doubt that saintly friendships actually qualify as good relationships in the relevant sense. One characteristic of good relationships of many kinds is that they are characteristically and deliberately reciprocal and fair. Perhaps in some special cases the sacrifices can be asymmetrical or one-sided, as when one friend is in an ongoing crisis and in no position to help. But, characteristically, the sacrifices and willingness to sacrifice should be more symmetrical. If my friendship with Jiaying were largely organized around my own success and happiness and only incidentally or occasionally served Jiaying’s interests, there is a sense in which we are not really friends at all. Furthermore, imagine that Jiaying did not see the friendship as being good for her, and yet she continued to spend a good deal of her time and energy on maintaining it. If the friendship were to end, she would be sad and see it as a loss for my sake but not for hers. In short, she would not value the friendship for her own sake at all. That indifference to the relationship’s contribution to her own welfare is incompatible with good friendship. Even if I were similarly indifferent about my own welfare relative to Jiaying’s (so that our willingness to self-sacrifice were symmetrical), that’s still a problematic friendship. We would both effectively be regarding ourselves as instruments, as means to the other’s ends, and not as friends. Self-interest helps us guard ourselves from exploitation, and it also positions us as reciprocal partners in a friendship, a friendship that we care about for both of our sakes. In numerous ways, self-interest makes relationships better—closer, more invested in their shared ends, and both fairer to and more inclusive of all parties.

As described above, the rich mixture of motives necessary to sustain mutually fulfilling relationships has at least two places for self-interest. First, there are what we might call relationship-conducive self-interested desires, such as the desire to maintain a friendship because one sees it as having final value for oneself. Second, there are self-interested desires that help ensure fairness and reciprocity in relationships, so that both sides benefit from the relationship in ways that deepen and further cement bonds between them.

For purposes of this paper, I take it as a necessary condition for an ethical theory that it be able to justify a certain sort of moral claim—a claim about what I will call the comparative ethical value of good shared-end relationships. Here is a rough version of the claim: in many circumstances and for most people, it is ethically better that people be deeply and mutually invested in shared ends than not, and that type of deep and mutual investment requires self-interested motives. Purely altruistic motives have a certain sort of positive ethical value, and they are often admirable, but having good, mutually fulfilling relationships based on shared ends is a higher ethical priority. An adequate ethical theory should be able to make room for this claim, such that it determines that self-interested desires or inclinations can be ethically good if they play the right role in maintaining mutually fulfilling relationships.

3. Mutual Fulfillment and Shared Ends

Mutually fulfilling relationships come in different kinds, presupposing different sorts of motivation and ways of setting the goals to which they are committed. For purposes of my argument, I am most interested in a subset of these relationships that I characterize as ethically good ones. Let me say a little bit more about the broader range of possible relationships so as to highlight what is distinctive about the (ethically) good and mutually fulfilling ones in particular. We can think of mutually fulfilling relationships in terms of two different spectra or dimensions. One dimension concerns how much the parties to the relationship share the same ends and are motivated by those shared ends. At one extreme are relationships where the benefits that motivate each party are exclusively instrumental, and none are motivated by any shared ends (call these “transactional” relationships). Market exchanges often exemplify this sort of mutual fulfillment: one person gives up a product that she does not particularly need or want for a product that she does want, and vice versa. Each side gets some benefit, but for both, the other’s good is at best a means to her own. Neither derives any final value from realizing the ends valued by the other.²

At the other extreme are relationships where one person’s motivating end just is another person’s motivating end, where one-and-the-same outcome has final value for all parties (call these “common causes”). Examples are shared social causes (a group of people in a political club or advocacy group, for example), or the interests of one or more parties in intimate relationships like those between family members or close friends. Family relationships are the most notable sort of example for Confucians like Dai Zhen. For Dai, there’s a sense in which a child’s survival and development just is a good for the parents, insofar as their survival and development is itself the realization of the parents’ own interest in procreation, continuing the family line, and reshaping the world and continuing their legacies through their offspring.

I would like to mention a certain in-between state or middle degree of relationship on this first dimension, because this type of relationship is important too. There are some arrangements in which the goods that motivate each party are a mix of instrumental and final ones. An example of this might be relationships between members of the same professional sports team—if the players on the team are at all emotionally invested in the team’s success, they will derive some final benefit when the team wins a championship or major tournament. Of course, the team’s success will have implications for their careers and salaries, and they care about that too. I find that many of the committees and short-term administrative projects that I do at my university put me in similar relationships with other faculty and staff: we care enough about our university (or college, or department) that our objectives have some final value for us, but we derive some instrumental benefits from setting up new academic programs or finding ways to recruit more students. For lack of a better term, let us call these “collegial relationships.”

• Instrumental vs. Final Value
• 1. Transactional relationships (purely instrumental for each party)
• 2. Collegial projects (some instrumental and some final value for each party)
• 3. Common causes (final value for each party)

I am most interested in relationships based on shared ends (types 2 and 3), not transactional ones (type 1). Transactional relationships are fine and unavoidable, while the basic orbit of our ethical lives is set by a wide range of relationships which, given our history and circumstances, require that we develop shared commitments to the same ends. These are the deep relationships that provide most of us with indispensable sources of meaning and purpose. They also ask more of us, require that we meet more ethically demanding goals and instantiate virtues worthy of the name.

The second dimension by which to measure relationships of mutual fulfillment is subtler. It has to do with how much one’s shared end is mediated by concern for the other people in the relationship. This concerns not what motivates us to pursue the end, but what motivates us to set the end in the first place. Compare the sense in which two fans of the same sports team (fans who are total strangers to one another) have shared ends, and compare that with how close friends and family members have shared ends. For the two fans—call them Mary and Meihua—their ends just so happen to overlap. Mary does not want the Sharks to win because of any concern for Meihua, nor does Meihua care about the Sharks’ success because of any concern for Mary. And yet there is a sense in which they are in a kind of relationship of mutual fulfillment, as they would embrace the spontaneous camaraderie and sense of kinship that they feel if they showed up at the same parade to celebrate a victory. In this case, Mary and Meihua have overlapping ends, but there is no other-mediation of those ends. They just happen to overlap.

In contrast, consider how parents and children, spouses, and good friends adopt one another’s ends. Sometimes I adopt an end of my parents’ or my friends’ solely for their sake. They care about a project deeply, so I do too, so that their end comes to have final value for me too. In these cases, for one or more parties, the ends are purely other-mediated. Aristotle sometimes speaks of virtue friendships in this way: “. . . in loving their friend they love what is good for themselves; “for when a good person becomes a friend he becomes a good for his friend.”³ Perhaps it also helps to think of a loving parent or spouse who says she just wants her beloved to be happy, to live the life that they find most fulfilling, and nothing more. Often this is meant to indicate a purely other-mediated setting of ends.

As for the first dimension, for this second dimension there is also a middle degree. In what I will call “intertwined relationships,” multiple parties share the same end, which are goods that have final value for them all, but the ends are neither entirely other-mediated nor just incidentally overlapping. The setting of the ends is motivated in part by concern for both parties, or by concern for the relationship or the unit to which they belong. Often family interests mark out relationships of this sort: if moving a family to a new and better location with more opportunities is good for the family, each member of the family will probably be motivated by ends that are partially for the sake of the other family members and partially for their own sake. Most of my close friendships are instances of intertwining rather than purely other-mediated adoption of shared ends. I like it when my friends do well and are happy, though that’s not purely for their sake. It often seems good for our friendship too, and thus for me, insofar as a good friendship is a final good for me (and it is). Perhaps my emotional investment in my students’ success also suggests intertwining ends—it’s partially for their sake and partially for my own then I want them to learn and improve themselves.

• Self-Mediated vs. Other-Mediated End-Setting
• 1. Overlapping ends (purely self-mediated)
• 2. Intertwined ends (both self-mediated and other-mediated)
• 3. Purely other-mediated ends

There is a tendency to admire and lionize the sorts of relationships that are most extreme in both of the respects I’ve outlined here—that is, relationships where the setting of an end is entirely other-mediated (adopting them purely for someone else’s sake) and one regards those ends as noninstrumentally valuable. Aristotelian virtue friendships are an object of fascination for this reason, both in scholarship and in the classroom. In some ethical scenarios that attract the most public attention in the English-speaking world (in Anglophone news media, films, and stories that capture the imagination and move us to tears) we often find paeans to selfless parenthood. However, it is clear that I am not primarily concerned with the most other-mediated sorts of relationships—with relationships at the selfless extreme. Many people tend to find purely other-mediated end-setting to be quite admirable, and I agree. However, with Dai Zhen, I worry that such relationships also tend to be ethically inadequate, that they lack a certain quality of reciprocity and equality or evenness (ping 平) that takes account of all parties in the relationship. Ends should be set in such a way that each party is empathically or sympathetically taking the other’s point of view into account as an equal starting point, or so we believe. To be ethically good relationships of mutual fulfillment, parties to a relationship must at least share some meaningful ends (so they must be relationships of the second or third type in the first dimension), and those ends must be set by some concern for all parties (so they must be of the second, “intertwined” type in the second dimension). This is one way to spell out the difference between relationships based on shared ends in general and the ethically good shared-end relationships that matter most.

As explained in the previous section, the relationships of the relevant sort require a rich mixture of motives, some of which are self-interested and some of which are more altruistic. This typology helps to say with more specificity why we need that rich mixture. One relatively obvious way that self-interest can help deepen a relationship is by giving us stronger attachments to others and more sources of satisfaction in helping them (as when I value a friend’s success and my friendship with her because of their contributions to my own interests and well-being). Nonetheless, there are less obvious ways in which self-interest can serve a person. Most notably, it can help us to advocate for a fairer or more reciprocal setting of ends. Motives matter not just because they get us to pursue and derive satisfaction from shared ends. They also matter because they help us to set the shared ends fairly and reciprocally. This also gives us a glimpse of the sophisticated ways in which self-interest might factor in. It might serve as a higher-order motive that regulates first-order ones: if I perceive that a friendship with someone is becoming too exploitative, too much about the friend and not me, I will become less inclined to do the things that maintain the friendship. It might also determine the character or valence of other motives: if an end that I share with a friend is purely mediated by her and does not take me into account, then any desire I have to pursue it becomes an altruistic one rather than a self-interested or mixed motive. Often, the self-interested motives will be quiet and nonconscious but “standing guard,” as it were, to make sure that our sacrifices do not become too onerous or asymmetrical.⁴

I assume that in most ethically good and mutually fulfilling relationships, shared ends are constantly being revised and reconsidered (and sometimes renegotiated). This implies that we would do well to maintain self-interested motives on an ongoing basis. In one of Seneca’s more memorable letters on friendship, he seems to suggest that we should select our friends based on their potential to contribute to our own virtue and flourishing. Once we have committed to the friend, however, we should commit completely. That, he seems to suggest, is how we realize that Aristotelian ideal of a purely altruistic friendship based on unconditional trust (Seneca 1969, 35–36).⁵ I disagree. For most shared ends, life is too unpredictable to commit once and for all. Our commitments to friendships should be resilient, they should last even through periods where they harm us more than they benefit us, so long as they are organized in such a way that they would characteristically or normally be more reciprocal. However, there are circumstances in which we can and should reevaluate them, and it is better, ethically speaking, that we take some account of our own interests in doing so, rather than permit ourselves to become exploited or more distant and unfulfilled friends.

Finally, it is worth noting that this analysis suggests at least two ways in which a relationship can become bad in the sense of being exploitative or asymmetrical. First, it might be that relationships are really designed to serve one or more party’s interests far more than another’s, so that even under normal circumstances, one party is being used or treated as a mere means to the interests of others. This worry is familiar enough as it looms in the background of many of our relationships. Nevertheless, there is a second and subtler sort of exploitation to guard against, and it has to do with fairness in how the shared ends and thus shared interests are set in the first place. I assume that there is a very real sense in which a child’s growth or happiness could be a shared end for both the child and her parent, and narrower goals (such as getting a college education or finding a lifelong romantic partner) can come to have some final value for both. Despite this, if the specification of those goals is not “fair”—if it does not give roughly equal weight to both party’s points of view, for example—then the end can be exploitative even if it is shared. I would add (looking forward to the discussion of Kant) that it can be exploitative even if both parties consent or agree to the shared end. To assess a relationship for fairness in this sense we need to know not just whose individual interests are served but also how their shared interests are determined.

4. Dai Zhen on Mutually Fulfilling Relationships

Dai Zhen is a Confucian philosopher who has long been the object of a great deal of scholarly fascination. Intellectual historians are interested in him because he was, by most accounts, the most innovative and influential scholar of the Qing dynasty (1644–1912, China’s last imperial dynasty), a true polymath who made tremendous contributions to philology, mathematics, geography, and astronomy (Elman 2001; Hu 2015; Wu and Sun 2015). The pivotal intellectuals of China’s New Culture Movement in the early twentieth century were interested in him because he seemed to be the one major philosopher of the indigenous Confucian tradition whose thought was most compatible with the more modern, naturalistic, and egalitarian ideas they sought to embrace (Hu 1996).⁶ Those who study the history of Chinese philosophy are interested in him because of his tremendous sophistication as a philosopher, which of all of Dai’s intellectual pursuits he regarded as his greatest calling.⁷

If you were to ask a Confucian scholar of Dai’s era what was most important and distinctive about his ethical thought, he or she would likely say that it was his robust defense of human desire as a constituent of virtue. By Dai’s time, Confucian orthodoxy and even popular morality had been under the spell of a kind of asceticism and self-abnegationism for so long that most moral ideals made little room for many seemingly basic and legitimate desires. As I have tried to show in some of my previous work on Dai, what appears to be a dispute about desires in general turns out to be a dispute about self-interested desires in particular, understood as desires for one’s own life-fulfillment or well-being for one’s own sake. Confucian orthodoxy and popular morality were perfectly fine with desires so long as they were virtuous ones. The problem that troubled Dai most was that they understood virtuous desires too narrowly as largely selfless ones, such that many desires for one’s own life-fulfillment as such (and for one’s own sake) would count as vicious. In direct opposition to this, Dai thought that we could not have the right sort of attitude toward others (caring about other people’s life-fulfillment for their sake) unless we have well-developed, self-interested desires (caring about our own life-fulfillment for our own sake) (Tiwald 2010a, 2012). We care about others in the right way by replicating and emulating proper self-concern. In Dai’s terminology, proper self-concern is “humane love of self” (renqishen 仁其身).⁸

A Confucian scholar of Dai’s era would also have talked about Dai’s aversion to the metaphysical and metaethical views that had been popular in China for at least seven centuries, views that attributed to all people a well-formed moral nature that they sometimes called our “original nature” or “inherent nature” (benxing 本性).⁹ Dai thought that this view was implausible, that it was an appropriation of the worst parts of Buddhism and Daoism, and that it lent itself to fundamental mistakes about how we justify ethical norms. Dai thought that our nature is good, but not fully developed or well-formed. It has a natural propensity to become good, provided the right nurturance and education.

Other features of Dai’s thought stand out more in retrospect than they did in his day. First, as I noted Dai was very concerned with how ethical norms are justified. He ultimately arrived at the following position: for any occasion in which you do something to others or expect something of them, your action or expectation must be one that could win universal approval under certain idealized circumstances. They must appeal to “invariant norms” (buyizhize 不易之則) that all “hearts-and-minds would affirm in common” (xinzhi suotongran 心之所同然) (Dai 2009, secs. 4–6, 8). He did not provide a lot of examples as to how this process would work in practice, but he seems to have understood it primarily as an exercise in empathetic perspective-taking (imagining how you would think and feel if someone were treating you in some way) with reference to acceptable and unacceptable desires (so as to determine whether the desire that motivates the action or expectation would be universally approved).¹⁰ On my reading, Dai comes to this universalizability criterion for a couple of reasons. First, he wants an ethical foundation that gives due consideration to everyone’s interests, meeting a standard of “fairness” or “evenness” (ping) that he takes to be fundamental (Dai 2009, secs. 2, 30). Second, he wants a moral epistemology that allows for people to check or second-guess one another’s judgments. As historians have often noted, Dai himself was born to a poor family, one that was bullied by more powerful people who dressed their bullying in ethical language and may even have believed that their bullying was right. Dai thought the source of so much moral mischief in his time was an intransigent form of higher-order moral ignorance: not knowing what is right or good, not knowing that one does not know it, and having no reason in principle to take the criticisms of others into account. By insisting that a powerful person’s ethical judgments can in principle be assessed by many others, he puts a check on high-order ethical ignorance that he thought unavailable to others in his time, given their presuppositions (Dai 2009, secs. 4–5).

Dai also had a great deal of interest in the relation between ethical imperatives, or what he called “what’s necessarily so” (biran 必然) with natural dispositions and inclinations, or what he called “what’s naturally so” (ziran 自然). Roughly, he thought that we should see ethical imperatives as an improvement upon or perfection of our natural inclinations, not as fundamentally discontinuous with them. When we embody and express the virtues, we are more perfect versions of our natural selves, not some new thing that has jettisoned its nature. Dai thought that this distinguished his view from prevailing moral views, which saw our natural inclinations as fundamentally flawed and appealed directly to transcendent norms and sources of motivation to help us overcome the bad parts of our natural endowment.¹¹

I hope that this brief summary is enough to help us see how Dai both is and is not a kindred philosophical spirit with Kant. Like Kant, he thinks that moral actions must in principle be capable of passing a test of universalizability. Furthermore, also like Kant, he was dissatisfied with metaethical theories that engaged in a lot of handwaving about some transcendent lawgiver or moral source, seemingly more interested in grounding human ethics in human interests broadly construed. As we will see later in this section, I also think that Dai was searching for a stronger sort of pro-attitude toward human beings than mere concern about their welfare, something that might be usefully compared with regarding others as ends in themselves.

However, Dai’s philosophical inclinations are also quite different from Kant’s in other respects. For reasons that I find understandable but a bit ill-conceived, Kant was adamant that the only source of categorical imperatives should be pure practical reason and was thus reluctant to allow that contingent human desires could determine our unconditional moral demands. He seems to have concluded from this that desires—understood as contingent “inclinations” (Neigungen) that arise from our natural constitution—are also a bad or at least less-than-morally-worthy source of moral motivation. When people act morally, they should be acting out of respect for the moral law, not out of the contingent desires that at best only accidentally line up with the moral law (and usually do not).¹² In contrast, Dai Zhen can not conceive of a plausible notion of ethical imperatives that does not take our natural inclinations into account, and is exasperated with the many generations of prior philosophers who seemed to think that such a project is possible. Ethical imperatives are improvements upon our natural inclinations or they are nothing at all. Moreover, he is emphatic that desires play a central and indispensable role in motivating good behavior, even self-interested desires.

At this general level, my description of the differences between Kant and Dai is a bit impressionistic. How much of a tension there really is depends on some details that I do not have the time to delve into here (for example, does it make a difference that Dai only talks about moral imperatives in general and not categorical or unconditional imperatives more specifically?). Despite this, I hope that these broad brushstrokes will be enough to appreciate the different philosophical agendas and orientations that the two philosophers adopted.

For Dai Zhen, a central notion in ethics is that of “mutual life-fulfillment.” It is central in the sense that many other ethical notions are justified and specified in terms of it. For example, Dai thinks that certain core virtues are justified in large part by their ability to realize this value. One of Dai’s well-chosen paraphrases for the Confucian Way is “the way of mutual nourishment and growth” (xiangshengyang zhidao 相生養之道) (2009, secs. 11, 15). And he characterizes the central virtue of humaneness (ren 仁) as “the virtue of mutual life fulfillment” (shengsheng zhide 生生之德) (sec. 36). Dai perceives (I think correctly) that most people, left to their own devices, see relationships of mutual fulfillment as central to ethics (he blames Daoists and Buddhists for downplaying these sorts of relationships and thus distorting people’s natural sense of the core of ethics). Nonetheless, Dai pays more attention than most to the ethical and psychological structure of these relationships, and to the emotional dispositions and character traits needed to realize them. In this section I will talk about how he understands these relationships and what they require of us.

I take “fulfillment of life” (suisheng 遂生) to be Dai’s particular conception of well-being. He thinks certain desires are intimately concerned with satisfying needs and interests that help to define us as living beings. Here he uses the Chinese character sheng 生, which refers not only to the state of being alive, but also to growth and reproduction. From Dai’s view, to fulfill a person’s life is to satisfy desires that have the right relation to her interest in survival, development, and procreation. Some desires are intimately connected to these goods (such as the desires for protection from the cold, sustenance, having and raising children). Other desires might seem more peripheral or indirect but still clearly count as life-based because they are sufficiently rooted in survival, growth, or reproduction (such as desires for romantic love and marriage, intellectual development, and teaching the next generation of students). He assumes that most of these desires are naturally flexible and of very wide scope, but they become narrower or more focused on specific outcomes as we figure out how to adapt them to mutually beneficial relationships. I happen to think that Dai’s account of well-being stands up reasonably well to common criticisms of standard theories of well-being in the twentieth and twenty-first centuries (hedonism, desire theories, objective list theories, etc.). It certainly stands up better than simple hedonic theories or desire theories. But for present purposes, not a lot depends on Dai’s specific account of well-being. So long as your preferred account allows that people can share ends and benefit directly from realizing shared ends, it will be compatible with much of what Dai says about the ethics of mutual fulfillment.

Let me now say a bit about why Dai thinks that relationships of mutual fulfillment are central to ethics (such that the major virtues are defined and specified in reference to them). If one did not know any better, it might be tempting to think that Dai is focused on these sorts of relationships because they have a kind of quasi-economic efficiency or make a greater contribution to the sum of human interests. Often, when multiple parties have the same ends and value them deeply, they will be better positioned for “win-win” propositions, such that one and the same outcome benefits all sides rather than set up one side to win at the other’s expense. The net total of human well-being that we can get is greater when people’s ends are shared than when they are different or in competition with one another. This argument appeals to “net benefit” or “net well-being” as a justification for good relationships of mutual fulfillment.

No doubt, Dai (like most of us) thinks it an advantage of these sorts of relationships that they make possible greater total satisfaction of desires, greater fulfillment of important goals, and thus more well-being. However, Dai does not think that welfare simpliciter is the central good. His organizing principle is what I call “orderly life-fulfillment.” “Life-fulfillment” is indeed a particular conception of well-being, but life-fulfillment must be in good order. “Order” (tiaoli 條理) tracks such things as social proximity, social status by virtue of age, merit, or desert and makes provisions for sustainability or continuity of living things. In Dai’s account of the virtues, he says that some are more concerned with promoting life-fulfillment, but others—such as ritual propriety (li 禮) and righteousness (yi 義)—are more concerned with tracking norms of good order (2009, sec. 36). Promoting life-fulfillment is virtuous so long as it takes some account of proximity, status, merit, and desert, making sure that our loved ones are the top priority, that elders are treated more deferentially than youth, that we are not showering undeserved benefits on people who should be punished (and are not punishing people who did no wrong), etc. Furthermore, Dai believes that social worlds with good, mutually fulfilling relationships are just better than those without, and that people who facilitate such relationships are ethically better than those that merely provide for their own well-being, even if those relationships were to come at some cost to net well-being. It is better for us to live as parents, children, siblings, and friends than as detached strangers. Dai says that we can enhance our admiration of relationships of mutual fulfillment by noticing how the cosmic order (“Heaven and Earth”) is also engaged in ongoing mutual life-fulfillment and ongoing creation (sec. 36). In short, Dai is not a welfarist —he does not think that well-being is the sole source of moral norms and values. His central good is orderly life-fulfillment, captured most powerfully in the vision of reciprocal processes that provide for our continuity into future generations.

Let me now turn to one of the most intriguing facets of Dai’s moral psychology, which has to do with the self-interested desires which I have mentioned. For Dai, good relationships of mutual fulfillment require a certain kind of self-interest that he calls “humane love of self” (sec. 21).¹³ In an oft-quoted passages in his Evidential Analysis of the Meanings of Terms in the Mengzi, Dai contends that a certain kind of love of self is continuous with love of those near and dear to oneself, and falls within the territory of humaneness (renzhishu 仁之屬), a core Confucian virtue (sec. 21). In another memorable passage Dai argues that a certain amount of interest in one’s own life-fulfillment is necessary in order to properly and adequately empathize with others. As he says at one point, “If one lacked desires [to fulfill one’s own life], one would look apathetically on even the most destitute and dire of life’s circumstances” (sec. 10). There is no way to arrange our emotional dispositions such that we could care in the correct way for others without caring ourselves (sec. 10). Dai thinks that humane love of others simulates and draws on humane love of self, so that we must continue to humanely love ourselves in order to sustain humane love of others.

In one of my first articles I did a close reading of some passages in Dai’s philosophical work to reconstruct what I take to be some notable features of humane love of self and explain why this is necessary to have the right sort of empathy for others (Tiwald 2010b). Briefly, I take it that Dai thinks when we love ourselves humanely, we love ourselves independently of ethically irrelevant properties or relations. Most of us love ourselves (if we do) regardless of whether we are high status or low, healthy or sick, wealthy or poor. Dai likes to note that a certain kind of empathetic concern cuts through such differences of position and fortune, and on my reading, Dai thinks we need to draw on the unconditional love of self to recreate a similar sort of love for others. When we love the self humanely we love the bare particular and not any of its accidental properties or relations (Tiwald 2010b). A second feature of humane love of self is that one loves oneself for one’s own sake, not for the sake of someone else or some higher end. We do not care about our own interests just insofar as contributing to them contributes to the prosperity of our employer, family, or society. A third (and important) feature of this love is what properties it does treat as necessary and fundamental to our status as worthy of love. Dai is quite clear that what gives us ethical status and intrinsic considerability is a combination of three things: being a living creature, being a creature that loves its life and fears its death, and having conscious awareness (jue 覺) (secs. 21, 30). Therefore, it is not, for example, because we are legislators in the realm of ends or because of a certain power or capacity for autonomy that we are worthy of love. Humane love of self has these three features—it loves the bare particular, for one’s own sake, and considers us worthy of love because we are living creatures with awareness and a love of life and fear of death. On my reading of Dai Zhen, these are the very features that we replicate (but sometimes to a lesser degree) in humane love of others, and from which we construct a common point of view that can be inhabited by many different people with mature capacities for wisdom and empathy.

Let us now see in action how Dai’s framework can be used to account for the ethical value of deriving mutual fulfillment from shared ends. Consider the shared end of a child’s formal education. Let us say that Jiaying has a daughter, Chen, whom she wants to provide with an education in the interest of promoting Chen’s intellectual and ethical development. Both Jiaying and her daughter see her daughter’s education as having some final value—having a more mature and spiritually and intellectually well-rounded Chen is in some sense the very goal that both seek to realize. Of course, they also both see her education as having some instrumental benefits as well, insofar as it will expand Chen’s career opportunities, give her more or greater sources of gratification and so on. On my understanding, part of what makes Chen’s formal education a shared end is that for both her and her mother, getting this education is relatively high on the list of valuable priorities that they care about, such that both are willing to sacrifice other competing interests or goods for its sake. So both think that some of the family’s resources and luxuries are worth giving up in order to pay for Chen’s tuition, for example, and that they rightly give up time playing games or listening to music in the interest of giving her more time to study or work on school projects. Having her education as a shared end thus has implications for trade-offs that they ought to make at various points in their lives. Some of those trade-offs will be obvious and relatively uncontroversial: clearly, it would be better to pay for Chen’s college tuition than to build a new addition to their already spacious home. Other trade-offs will require more wisdom and ethical discernment—whether Jiaying should intervene on Chen’s behalf when, for example, she thinks one of Chen’s teachers is treating her unfairly.

On Dai’s account, each of these decisions will be resolved by adopting a certain empathetic point of view, one that asks them to imagine themselves in the position of various affected parties and tranquilly reflect on whether they could reasonably bear the treatment they are giving them or fulfill the demands that they are making on them (2009, sec. 2). Dai thinks that certain demands will seem more reasonable to fulfill and certain kinds of treatment will seem more reasonable to endure if they are motivated by and meant to fulfill certain common and widely-shared desires, desires that we as intelligent living beings can all understand and stand behind. It is in these senses that the relevant ethical norms are those which “all hearts-and-minds affirm in common.” They are norms that seem warranted from multiple, somewhat idealized, points of view, and they take as their object a state of affairs which reasonably mature people can empathically imagine as satisfying or fulfilling enough to justify the sacrifice of other goods (secs. 2-8).

In order to realize the shared end of Chen’s formal education, Jiaying will need certain things—resources, various habits, emotional attachments, powers to ignore impulses or delay gratification, aptitudes of judgment, and so on. Some of these things will be banal or not particularly admirable or notable. However a certain subset of these things will, when working in tandem, exhibit a certain excellence or ethical beauty (yi 懿 or mei 美) (secs. 3, 36). Those character traits and characteristic behaviors that both exhibit this sort of excellence and play the right sort of supporting role in realizing “orderly life fulfillment” are virtues, and insofar as Jiaying instantiates these virtues, she is much better or more ethical than she would be without them. Dai says that some of these virtues (such as humaneness and righteousness) are themselves constituents of orderly life-fulfillment, and others (such as courage) are just the means by which we bring about orderly life-fulfillment (sec. 36). Either way, so long as they fit together into a system of character traits and characteristic behaviors that constitute or help to sustain the central good, and so long as they exhibit excellence, they make Jiaying a much better person than she would otherwise be.¹⁴

5. Kant on Mutually Fulfilling Relationships

On a certain caricature of Kant, he thinks there is really only one kind of motive worth adopting and one kind of behavior worth emulating, which are sometimes jointly characterized as action from duty. This is the sort of thing that we do when we represent the moral law in our minds and act out of respect for that moral law. According to this caricature, action motivated by any desire is always problematic because all desires are contingent things with no built-in capacity to follow the moral law, which makes them at best conditionally and accidentally good. Nature does not guarantee that our desires will do as the moral law requires, only practical reason and the will can do that, insofar as they inspire in us respect for the moral law. Desires respond to the senses and sensible objects, not to a priori reason and moral norms—in Kant’s term, they are “pathological” (pathologische). Moreover, the desires are stubborn things, intransigent and generally difficult to change, and they only treat our own happiness or pleasure as their guiding aim, not morality.¹⁵ For all of these reasons, this story goes, an action has moral value if and only if it is motivated by the good will and not by desires of any kind.

However, this caricature is mistaken. Among other things, it ignores both the content and the prominent aims of his relatively late work, The Metaphysics of Morals, which includes considerable discussion of the virtues. There he suggests that natural dispositions can be trained and reshaped, as when he recommends that people visit sick rooms and debtors’ prisons in order to fine-tune their sympathetic appreciation of the suffering of others. He also claims that we should secure a moderate amount of happiness for ourselves lest we become so disgruntled that we start transgressing the moral law or ignoring our duties on a regular basis.¹⁶ So Kant clearly endorses a limited pursuit of one’s own welfare, and seems to think that pathological emotions and desires can be changed so that they better align with morality. What, then, are we to make of his strong claims to the effect that only action from duty (action motivated by will and representations of the moral law) is good?

In late twentieth- and early twenty-first-century Western philosophy, a great deal has been written on the place of virtue and the components of good character in Kant’s ethics.¹⁷ I cannot hope to summarize it all, but let me highlight the arguments that are most relevant for my discussion. The first thing to note is that when Kant talks about the distinctive value of action from duty or the good will, he does not say that this motivational structure is necessary for value in general but for “moral worth” in particular.¹⁸ This presumably leaves open the possibility that other sorts of action can instantiate a different and presumably lesser sort of ethical or moral value. Second, defenders of Kant have proposed that Kant intended duty and respect for the moral law not just as the direct impetus to action, but as a kind of background condition or commitment that governs or regulates the more direct springs of human behavior (Marcia Baron calls this background condition a “secondary motive” and the direct impetus a “primary motive”). So a person can still have or express moral worth if, for example, she helps someone from a desire to please, so long as that primary motive is checked or conditioned by an overarching and overriding commitment to morality, and provided that overriding commitment is sufficient to effect dutiful behavior (Baron 1995, 113-114, 188-193). So long as the will stands ready to intervene on behalf of a representation of duty, and so long as it has the power to override primary motives to that end, a person can act from duty even as the direct impetus to action is one or more of our pathological desires, including, presumably, self-interested desires.

These two corrections go a good distance toward addressing the challenge that I have set for Kant. The challenge was to provide an ethical framework that helps to explain why it is ethically better to be the sort of person with mixed motives to participate in and contribute to good relationships than not, to have the mixed motives that make the good sorts of mutual fulfillment possible. What the two clarifications show is that Kant can in principle explain how someone might be motivated to help a friend partially by self-interest and yet be a person that (or carry out an action that) instantiates or imparts moral worth. Someone can be partially motivated to help a friend by self-interest in the primary sense, and that is just fine so long as there is a potentially overriding will that stands ready to intervene on behalf of the moral law.

Nonetheless, this does not address the challenge completely. First, on Kant’s view, moral worth is still solely a function of the duty motive. Kant does not suggest that having additional, self-interested but relationship-conducive motives adds or enhances moral worth. If we take someone who helps a friend purely out of duty but does so grudgingly and compare him with someone who helps a friend wholeheartedly and with great personal interest, by Kant’s lights both are equal in moral worth. The challenge was to show how the latter could be ethically better than the former, as most people in fact believe. Second, it does not address another feature of Kant’s moral psychology, which is that he sees desires as unresponsive to the sorts of reasoning that tracks the moral law. Being “pathological,” desires and emotions only respond to the senses and sensible objects, and at least some passages of Kant’s moral writings suggest that our desires and emotions are too intransigent to be meaningfully modified or updated.

These clarifications also raise a more fundamental conceptual issue that we should pause to consider. On this more nuanced picture of Kant’s ethical theory, someone with the motivational set that enables her to derive mutual fulfillment from her relationships can have moral worth under those circumstances (or her relationship-conducive actions can have moral worth), but only if there is an overriding secondary motive the ensures compliance with the moral law. If there is no such motive, or if there is such a motive but it is not sufficient to override, she (or her actions) cannot have moral worth. However, that front loads a rather formidable requirement for moral worth from the very start. Do we really want to insist that so much volitional capacity for moral compliance is necessary for any moral worth at all? One reason that I think the most influential moral theories have done such a poor job accounting for the comparative ethical value of good relationships of mutual fulfillment is because they tend to be constructed around ideal agents or instances of agency. We want to know whether it makes significant ethical difference whether Jiaying is a relatively distant parent who cares about her children for their sake or a deeply engaged and multi-faceted parent who cares about them partially for their sake and partially out of pride, shared projects, and a personal interest in having and raising children. To most of us, it does not seem that it should only make a difference if Jiaying also has a commitment to the moral law that is sufficient to override the relevant desires. Quite possibly, most of us lack that commitment, and yet there is a world of difference between a cold and distant parent and a deeply attached parent who derives mutual fulfillment from parenting. Therefore, it is fundamentally mistaken to insist on the requirement that there be an overriding secondary motive of duty. Having that overriding secondary motive can be a good and productive ideal, something that people should aspire to, but it is too demanding and stringent to insist that it be a necessary condition for all moral worth.

Perhaps Kant would say that the weak-willed parent who derives proper mutual fulfillment from her parenting is ethically better in some sense that does not involve moral worth. I hinted at this sort of solution earlier when I mentioned that there may be values or goods other than moral worth that can help to justify certain impure motivational sets. Maybe Kant recognizes that it’s good for people to learn to care about their children in the rich mixture of ways required for mutual fulfillment, but he does not think these should count as moral improvements, or as bases for moral worth. However, they still count as improvements in some sense, and Kant thereby provides us with reasons to pursue love and sympathy.

My objections are difficult to spell out in detail because they depend on how we specify the special status of moral worth relative to other ethical goods or values. Despite this, here is a rough description of the concern: whatever it means for one thing to impart moral worth and for another thing to be a mere good, Kant’s ranking of moral worth above all other values must at least imply that moral worth should have some sort of clear priority. For example, perhaps he thinks that given an opportunity to do something of moral worth or to something that is merely ethically good, one should do something of moral worth. However, if we were to give clear priority to action from duty over the desires necessary for proper mutual fulfillment, this would have implications that most of the world’s people (at most times and places) would find counter-intuitive. Surely being a deeply engaged and caring parent who is deeply invested in one’s parenthood is of far higher priority than having the strength of will to rein in one’s parental desires in the interest of the moral law.

In pursuit of a better defense of Kantian approaches to relationships of mutual fulfillment, we have to set aside the historical Kant and consider a modified version of him, making some revisions to his view that, we can hope, will leave his core ethical commitments intact. Among the most promising revisions are those suggested by a forthcoming article by Kyla Ebels-Duggan, who attempts to clean up problems in Kant’s moral psychology by proposing that Kantians jettison the view that desires are pathological. Desires on this revised view can represent values and objects as valuable or choice-worthy, thus giving us a degree of control over them. Having a degree of control over them, they become “attributable to us,” so that the actions motivated by those desires can have moral worth. If a desire represents someone as worthy of respect and love, for example, then to act on it is to “act in response to the value of humanity” as surely as the will can so act (Ebels-Duggan, forthcoming). In principle, at least, someone who has the desires to derive mutual fulfillment from her relationships could count as having greater moral worth for that reason. Furthermore, Ebels-Duggan’s revisions appear to lower the price of admission for moral worth, for it is no longer necessary that one have the strength of will and commitment to moral law to override a desire in order to count as morally worthy. So long as the desire itself represents someone as worthy of respect and love, the resultant action qualifies.

If Kantian ethics is to be saved from the charge that it cannot explain the comparative ethical value of good relationships of mutual fulfillment, this sort of move offers the best hope for doing so. Kantians need to allow that desires can carry and impart moral worth. Still, even this revisionary picture of Kant’s ethics faces formidable challenges. Firstly, depending on what “representing someone as worthy of respect and love” requires, we could still end up with an ethical theory that attributes much higher priority to what are in fact comparatively trivial acts and motives and underestimates the ethical importance of human relationships. The desires of love, for example, seem good candidates for the sorts of desires that Ebels-Duggan has in mind, but it seems likely that love more often than not falls short of this ideal, certainly if representing someone as worthy of respect and love entails representing her as an end in herself and never as means only (as Kant rightly worried, love often instrumentalizes the beloved).

A second problem is this: even though Ebels-Duggan’s view about the range of motives that might impart moral worth is more permissive and inclusive, it is not permissive enough to be compatible with widely-shared intuitions about the motives necessary for good relationships of mutual fulfillment. On the view that I find most defensible, there is a wide array of dispositions or character traits that can impart ethical value or worth, whether or not they happen to represent people as worthy of love and respect. Dai Zhen’s view on this matter is more promising: he thinks that an ethical virtue is an admirable character trait that cooperates with other admirable character traits to promote or constitute the central good, which for him is orderly life-fulfillment. Some admirable character traits (such as courage) do not characteristically represent people as worthy of love and respect, and they can be virtues nonetheless, just by being admirable in their own right and fitting rightly into the constellation of other virtues. Similarly, there are habits of mind and mental and emotional dispositions in good parenting and teaching that seem good candidates for bearers of ethical worth, things like resourcefulness or capacity to listen thoughtfully and charitably, and yet I doubt that they themselves are constituted of desires that represent people as worthy of love and respect.

This second problem may be symptomatic of a deeper issue of theoretical orientation that I alluded to earlier. The deeper issue has to do with whether we see certain distinctive features of morality as minimum requirements or as one important factor among others. Imagine that we could reach a consensus about the criteria that a motive must meet in order to qualify as bestowing this important sort of moral value. For example, both Kant and Dai think that there there’s a special imperative to recognize all other people as similar to ourselves in certain respects, and to see others as having intrinsic worth, and as having interests that in some important way have equal purchase on the norms of interpersonal interactions. Even if we can reach an agreement about the criteria for this privileged set of norms, however, there is still the question of how those criteria operate. One possibility, which from my view is ubiquitous in both Kant’s ethics and much Kantian ethics, is that criteria serve as a necessary condition for any and all moral worth. Only if one represents or conceives of others and one’s moral obligations to others in a certain way will the relevant motives qualify as having this special moral status. But another option is that the criteria serve as a special kind of regulative ideal—something to aspire to, such that, ceteris paribus, we become better people (or have better character or exhibit better behaviors) the more closely we approximate it. The latter view is closer to Dai Zhen’s, and because he treats the special norms governing interpersonal interactions as a regulative, aspirational ideal in this way, he can make allowances for kinds of ethical improvements and trade-offs that Kant cannot. He can say, for example, that it’s preferable that a teacher always treat her students as having intrinsic value and as equal end-setters, and to the extent that she fails to do so, she’s a nonideal teacher, but he can also allow that there are other sorts of ethical improvement which might, under some circumstances, be of higher priority. For example, it is better for many teachers to take self-interested pride in the success of their students than to be distant or indifferent to their students’ success, even if this pride sometimes comes at the expense of regarding students instrumentally or as less-than-equal end-setters.

Finally, there are worries about whether the move to admit some value-regarding desires into the club of morally worthy motives can be integrated into Kant’s broader moral framework, one so systematically devoted to clarifying the proper source of morality in pure practical reason and purging morality of impure sources. For example, Kant is correct to think that natural desires often go wrong, inclining us toward courses of action that ride roughshod over the interests of others, instrumentalize others, or treat others unjustly. It seems like any reasonably conscientious Kantian (and Kant himself) would insist that we at least vet our desires so that only those that aptly respond to the value of humanity are trusted and acted upon. Nevertheless, there is little in Kant that will ground a plausible or enlightening story about how the desires can be vetted, since his framework so frequently depends on the intervention of an independent faculty of will acting on the representations of pure practical reason. In stark contrast to this, Dai Zhen provides a realistic and appealing account of the sources of interpersonal ethics, one that builds ethical motives on natural ones and yet makes a compelling case that these improvements make us more aptly responsive to the value of others. And as we have seen, at most every turn his system is concerned with grounding good relationships based on shared ends. If we concede that such relationships are a central or at least an important part of our ethical lives, as I think we must, then Dai’s framework seems the better starting point for an adequate theory of ethics.

Acknowledgments

I thank Princeton University’s Center for Human Values for supporting this research through a Laurance S. Rockefeller Fellowship. For comments on an oral presentation of this paper I thank the participants in the Conference on Confucianism, Buddhism, and Kantian Moral Theory (September 2019), as well as the American Council of Learned Societies, the Chiang Ching-Kuo Foundation and the Sungkyun Institute for Confucian Studies and East Asian Philosophy for supporting and hosting the conference. Most of all, I am indebted to Brad Cokelet and Philip J. Ivanhoe for astute and truly supererogatory feedback on an earlier draft of this paper.

References

• Angle, Stephen C., and Justin Tiwald. 2017. Neo-Confucianism: A Philosophical Introduction. Cambridge, UK: Polity Press.
• Aristotle. 2000. Nicomachean Ethics. Translated and edited by Roger Crisp. New York: Cambridge University Press. [https://doi.org/10.1017/CBO9780511802058]
• Baron, Marcia W. 1995. Kantian Ethics Almost Without Apology. Ithaca, NY: Cornell University Press.
• Betzler, Monika, ed. 2007. Kant’s Ethics of Virtue. Berlin: De Gruyter. [https://doi.org/10.1515/9783110209655]
• Dai, Zhen 戴震. 2009.“Mengzi ziyi shuzheng 孟子字義疏證” (Evidential Analysis of the Meanings of Terms in the Mengzi). In Daizhenji 戴震集(The Collected Works of Dai Zhen), 263–329. Shanghai: Shanghai guji chubanshe.
• Ebels-Duggan, Kyla. Forthcoming. “Bad Debt: The Kantian Inheritance of Empiricist Desire.” In Kantian Freedom., edited by Evan Tiffany and Dai Heide. New York: Oxford University Press.
• Elman, Benjamin. 2001. From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China. Rev. ed. Los Angeles: University of California Los Angeles Press.
• Herman, Barbara. 2007. Moral Literacy. Cambrige, MA: Harvard University Press. [https://doi.org/10.4159/9780674273665]
• Hu, Minghui. 2015. China’s Transition to Modernity: The New Classical Vision of Dai Zhen. Seattle: University of Washington Press. [https://doi.org/10.1515/9780295806068]
• Hu, Shi 胡適. 1996. Dongyuande zhexue 戴東原的哲學(The Philosophy of Dai Zhen). Reprint. Taibei: Taiwan shangwu yinshuguan.
• Ivanhoe, Philip J. 2000. Confucian Moral Self Cultivation. 2nd ed. Cambridge, MA: Hackett.
• Ivanhoe, Philip J. 2016. Three Streams: Confucian Reflections on Learnings and the Moral Heart-Mind in China, Korea, and Japan. New York: Oxford University Press. [https://doi.org/10.1093/acprof:oso/9780190492014.001.0001]
• Kant, Immanuel. 1996. Practical Philosophy. Translated and edited by Mary J. Gregor. New York: Cambridge University Press. [https://doi.org/10.1017/CBO9780511813306]
• Seneca, Lucius Annaeus. 1969. Letters from a Stoic. Reprint. Translated by Robin Campbell. New York: Penguin.
• Tiwald, Justin. 2010a. “Dai Zhen on Human Nature and Self-Cultivation.” In Dao Companion to Neo-Confucian Philosophy, edited by John Makeham. Dordrecht: Springer. [https://doi.org/10.1007/978-90-481-2930-0_19]
• Tiwald, Justin. 2010b. “Dai Zhen on Sympathetic Concern.” Journal of Chinese Philosophy 37.1: 76-89. [https://doi.org/10.1163/15406253-03701007]
• Tiwald, Justin. 2012. “Dai Zhen’s Defense of Self-Interest.” Journal of Chinese Philosophy 38s: 29-45. [https://doi.org/10.1111/j.1540-6253.2012.01687.x]
• Wu, Genyou 吳根友, and Sun Bangjin 孫邦金. 2015. Daizhen, ganjia wueshu yu zhongguo wenhua 戴震、乾嘉學術與中國文化(Dai Zhen, Quanjia Scholarship, and Chinese Culture).
• Zhu, Xi. 2019. Zhu Xi: Selected Writings, edited by Philip J. Ivanhoe. New York: Oxford University Press.