On the View that People and Not Institutions Bear Primary Credit for Success in Governance: Confucian Arguments

1. Introduction

Let us assume for the sake of argument that we live in a time of flawed governance. Many of the basic outcomes that it is reasonable to expect of our governments are not actually met by them, whether those are to keep the peace, to provide for basic needs and interests of the people, or to resolve potential conflicts in minimally just or fair ways. What should be done to improve governance so that more of those basic outcomes are realized? One familiar set of responses will point to flaws in the institutions of government. Construed broadly, this can include problems in the laws that are applied to citizens (e.g., tax codes, criminal statutes) or more procedural problems in the rules and regulations that governing institutions abide by (e.g., judicial protocols, rules determining how laws are established or amended). But sometimes we give a different sort of answer: we say that governance will be much more likely to improve if we reform the character of those who govern. See to it that those who make and execute laws and those who adjudicate cases are altruistic, conscientious, and not susceptible to bribery or corruption, and we will see a better government. Of course, both sorts of answers can be correct. It might be that we need both institutional reforms and people of better character in order to have better government. But sometimes people will argue that character reform should be regarded as primary, that the real work of improving government consists most fundamentally in ensuring that we have virtuous decision-makers rather than in improving laws and procedures. We might call this the “virtue-centered” theory or approach to improving governance.

For scholars of Confucianism and historians of East Asia, it is well known (almost a platitude) that many of the major political thinkers in the Confucian tradition endorsed virtue-centered approaches and frameworks. As many readers know, the commitment to virtue-centered approaches was particularly strong in the Song through Ming dynasties, at least in the period that came after the New Policies (Xinfa 新法) of Wang Anshi 王安石 (1021—1086). A familiar historical account makes this explicit. It says that institutional reforms were a major pre-occupation of Song government from the mid-eleventh century through the end of the Northern Song dynasty in 1126, starting with the Qinli Reforms (Qinli Xinzheng 慶曆新政) of Fan Zhongyan 范仲淹 (989—1052) in 1043—1045, continuing in the more dramatic and systematic New Policies of Wang Anshi in 1069—1076, and culminating in a more dogmatic and uncompromising period of reform carried out by Wang’s successors, who were restored to power in 1093 and had imperial favor until Jurchens conquered northern China and compelled the Song dynasty to move its capital to the south. The loss of northern China was a source of great humiliation for later Chinese thinkers and members of the political class, especially for Zhu Xi 朱熹 (1130—1200), who came to define Confucian orthodoxy after the Song. For Zhu and several other major Confucian philosophers in the Southern Song and in the later Ming dynasty, these attempts at institutional reform were largely responsible for the downfall of the Northern Song. While there were some problems with the reforms themselves, the greater mistake was in thinking that institutional changes alone would be sufficient to strengthen the state. These thinkers proposed that better governance depended primarily and more fundamentally on improving the character of the people most responsible for governing. And it happens that this view is easy to reconcile with a longstanding Confucian interest in virtuous rule. The considerable attention given in Confucian classics to the moral education and cultivation of rulers and ministers seems to suggest that the character of those who govern is more important than most anything else. Furthermore, virtue-centered approaches were in many ways a defining feature of China’s most influential thinkers, distinguishing mainstream Neo-Confucian philosophers like the Cheng brothers and Zhu Xi from what some modern scholars call the “utilitarian factions” (Gonglipai 功利派) in Confucian politics, whose members believed institutional reforms could be effective and valuable even without improving character.¹

If virtue-centered approaches are a defining characteristic of much of later Confucian political thought, however, it is striking how little philosophical attention they have received. Two sorts of philosophical attention in particular are lacking. First, as described above, what makes an approach to governance “virtue-centered” is vague, encompassing a variety of overlapping but analytically distinct ways in which improving character could be prior to or more fundamental than institutional reform. Furthermore, it takes some work even to see which of these interpretations is plausible. The most influential voices for virtue-centered approaches also advocated strenuously for major reforms to laws and procedures, and there are many ways in which the development of good character and the selection of virtuous leaders depends on reasonably good laws and institutional practices, as all major Confucian thinkers readily acknowledged. Political philosophers and political theorists need greater precision and clarity if they are to bring Confucian ideas about the fundamental importance of character to bear on contemporary political debates. Second, there is scant discussion of Confucians’ arguments for their views on these issues. The Confucians who embraced virtue-centered ideas made their reasoning relatively clear, sometimes through explicit justifications and other times through artful references to classical sources. However, we have yet to see a comprehensive review or reconstruction of those arguments.²

My explication proceeds in three parts. In the next section of the paper (Section 2), I offer several ways of understanding the claim that the character of those who govern bears primary credit for success in governance, highlighting those that I take to be more important for the Confucian political philosophers who embraced virtue-centered positions. There I rely on Zhu Xi and the comments of contemporary intellectual historians to refine my account. In Section 3, I reconstruct the major Confucian arguments for virtue-centered approaches, many of which either reference or build on Xunzi’s 荀子 (c. 310—219 BCE) famous claim that the proper sources of social order are “people” (ren 人) and not “rules” (fa 法) and Xunzi’s subsequent discussion of that claim.³ In Section 4, I look at what I take to be one of the most notable Confucian critiques of virtue-centered theories, which comes from Huang Zongxi 黃宗羲 (1610—1695), a reform-minded Confucian of the late Ming and early Qing. There I argue that he raises a different sort of objection than contemporary readers expect, and that his views are not uniformly at odds with the virtue-centered view. I conclude in Section 5 with some observations about how we might add more depth and texture to the debate as I have reconstructed it here.

2. Ways of Bearing Primary Credit for Successful Governance

There is a wide-ranging set of questions that Confucians interested in this debate often raise, with implications for several different points of dispute in Confucian political thought, from struggles over the content of the civil service exams, to concerns about the proliferation of laws and punishments, to questions about how much incentive structures should be built around more cynical assumptions about human beings. The issue also has implications for (but is different and more focused than) longstanding debates about how best to distinguish between wangdao 王道 (the way of the true king) and badao 霸道 (the way of the hegemon), or about fazhi 法治 (rule by law) and dezhi 德治 (rule by virtue). A larger project might survey all of these issues in their most notable historical manifestations, but for the sake of having a clear and well-defined point of entry into this debate, I will focus on one particular series of arguments that have to do with the credit-bearingness of office-holding people (ren 人) relative to that of the institutional rules (fa 法, sometimes translated as “laws” or “standards”) of their offices. In an influential chapter titled “The Way of the Ruler” (jundao 君道), Xunzi makes the case that where one finds political and social order, it is the people in positions of power who should get credit for effecting that order, with little credit left over for institutional rules—a point that Xunzi puts succinctly in the quotable line, “There are people who create order; there are no rules that create order” (youzhiren, wuzhifa 有治人, 無治法) (Xunzi 1988, ch. 12, 230; 2014, 117). Unlike some of Xunzi’s other well-known views, this one was generally well received by Confucians in the Song dynasty and thereafter, and was frequently invoked by virtue-centered political thinkers or criticized by more institutional thinkers in the Neo-Confucian era (Angle and Tiwald 2017, 191-194; Tiwald 2016, 458). When Xunzi and later proponents credited good social and political order to “people,” they largely had in mind their judiciousnes in moral matters and stable character traits that enabled them to remain steadfast in the service of the state and its people in the face of temptations to do otherwise, the right combination of which can be characterized as virtues. Moreover, they were largely (but not exclusively) concerned with the people who ran the institutions—rulers, ministers, clan leaders, and other people with the power to shape the social order. Thus, the guiding question of this study will be why and in what respects the virtues of those with power and influence rather than institutional rules should account for success in governance.

It is not obvious what it means for people’s virtues to be more (or more fundamentally) responsible for success in governance than institutional rules. Many of the most apparent interpretations turn out to be wrong on one of two counts: either they attribute to the virtue-centered thinkers a view they did not actually hold, or they focus on something about the efficacy of virtue that is relatively uncontroversial and widely conceded by everyone (including opponents and critics of the virtue-centered approach). For example, no serious virtue-centered theorist held that a virtuous ruler could completely overcome the effects of truly vicious or perverse rules. Imagine a state in which the rules generally reward people for cheating or harming one another, which promote the most despicable characters and punish those who are team players. Furthermore, no serious virtue-centered thinker believed that a sufficiently virtuous ruler could effectively bring about political and social order without having any rules at all. As we will see in the next section, defenders of the virtue-centered approach assume that rules are indispensable instruments of good governance.

Similarly, there are some positions that we can rightly attribute to virtue-centered thinkers, but which do not in themselves capture what is most important and controversial about their position. For example, even Confucian critics of the virtue-centered position believed that having rulers and ministers with at least a certain amount of virtue is necessary for decent governance, and many also agreed that optimal governance requires a ruler or influential advisor who is for all intents and purposes a moral paragon or sage. Thus, the distinctiveness of the virtue-centered position is not captured by saying that a certain amount of virtue (moderate or maximal) is a necessary condition for a certain amount of good governance (decent or optimal).

Finally, there are some conceptual issues that tend to obfuscate the debate between virtue-centered theorists and their critics. For example, there are various “chicken and egg” problems that arise from the fact that we need certain sorts of institutional rules in order to cultivate good character and put virtuous people in power in the first place. Among the New Policies advocated by the institutional reformer Wang Anshi, arguably he was most passionate about his changes to the civil service exam system and public education, precisely because he thought these changes would bring people of better quality into government ranks (Xiao 1982, 491-492). Historians sometimes point out that institutionalists like Wang seemed to care more about “talent” (cai 才, ability to get things done effectively) than “moral quality” or “virtue” (de 德) (Liu 1988, 154). However, it is clear enough that Wang thought a certain amount of virtue is prerequisite. A despicable and intemperate person like Cao Cao 曹操 will make for a bad ruler no matter how much he may excel at the arts of management and administration (Tillman 1982, 138). Similarly, virtue-centered thinkers like Zhu Xi conceded that social and political order would not be restored in China until the rules governing the exams and selection of ministers were reformed. Of course, systematic changes to the civil service laws and procedures would not be likely to happen without an emperor and ministers who are reasonably interested in changing those laws and procedures for the betterment of the people and political society. Thus, it seems that reforming rules is necessary for improving the moral character of state officials, and some improvement in their moral character is necessary for reforming the rules. This is one notable chicken and egg problem that appears in many common-sense or preliminary attempts to define the terms of the debate.

A careful examination of Zhu Xi’s arguments suggests that two claims distinguished virtue-centered views like Zhu’s from that of his critics. First, while Zhu concedes that both institutional reform and having rulers and state officials of good moral character stand to improve governance in China, he nevertheless insists that improving character is considerably more challenging, and thus more demanding than changing institutions.

This age suffers from two defects: defects in its institutional rules (fa) and defects in the current political situation. The defects in the rules can all be altered at once quite easily, but the defects in the current political situation all reside in people. How can they be changed when people go about their business with a selfish heart-mind! The rules of the last eight years of Emperor Renzong’s reign can be considered defective. [Wang Anshi] changed them all soon after [becoming the emperor’s prime minister], but this only gave rise to numerous new defects. This is because people are hard to change.⁴

Although Zhu does not make his reasoning fully explicit, he often responds to questions about the effectiveness of institutional versus characterological reform by noting that the latter is considerably more knotty, vexing, and demanding than the former. Perhaps it is not entirely clear why the relative difficulty of reforming people’s character or rules should make a difference in terms of which bears more credit, so let me clarify. To take a page from the playbook of Xunzi (whom Zhu followed closely on this issue), we might think of rules as being like standards by which good archery is measured and virtue as being like the strength and skillfulness of individual archers.⁵ The standards by which good archery is measured are relatively clear: one should hit the target with an arrow; the closer to the center, the better; one should use certain sorts of instruments (a bow of certain shape and structure) to do this, from a certain distance, etc. Learning these standards does not take long, and there is little achievement in doing it. When a person is a good archer, most of the credit goes to the strength of her arms and shoulders and her skill at holding a bow, aiming an arrow, accounting for the effects of wind and gravity, etc. Those abilities demand the most of us, and thus it is by conscientiously applying ourselves to them that our efforts will reap rewards, and not by conscientiously applying ourselves to the work of learning the standards by which good archery are measured. As Zhu says, “the defects in the rules can be altered at once quite easily,” but defects in the character of people—the defects of human vice—are formidable indeed.⁶

A second distinctive claim of Zhu’s is subtler, better characterized by what it denies than what it affirms. It says that having decent rules is all well and good, but that after rules have reached a certain minimum or floor of acceptability or “viability,” further improvements of those rules will make no further gains in the social and political order without concomitant improvements in the character of the leadership class. That is, this claim for virtue-centered politics is not so much about the efficacy of virtue alone but rather about the fruitlessness of institutional reform without virtue. Its aim is to show another political view and orientation wrong, one that we could characterize as “institutionalism”—the belief that (even after institutions have reached a certain minimum of decency) changing the rules of institutions will of itself yield meaningful improvements in the social and political order, without concomitant improvements in the character of those who govern. As noted earlier, Zhu and other virtue-centered thinkers do accept that it helps to amend truly perverse or ridiculous rules (rules that reward people for cheating and hurting one another, for example), so this objection to institutionalist solutions does not apply in all cases, but it does apply to most regimes where the rules have reached a certain minimum level of functionality to be viable. Zhu hints at this position in the passage quoted above, suggesting that Wang Anshi’s New Policies showed that fixing defects in the rules without fixing defects in people only leads to more defects. The only way forward, he suggests, is to get people to stop being so selfish.

I have characterized this negative position as the claim that institutional reforms alone are unlikely to yield “meaningful” improvements. There are a number of specific ways of interpreting that modifier. One is to say that institutional reforms alone will yield no improvements at all. A second is that it will yield small but relatively insignificant improvements, perhaps on a diminishing marginal utility model. A third is that it could yield some improvements, but that the improvements would be transitory and not sustainable (e.g., circumstances might improve for a time, but before long the old customs or practices will reassert themselves). A fourth is that any improvements will be a mixed blessing, reducing some problems while giving rise to several new ones (trading one sort of corruption for another, for example). I have used the phrase “meaningful improvements” so as to remain neutral between these four interpretations. Scholars of Song political thought sometimes hint at the third interpretation, suggesting that any improvements that come about from institutional reforms alone will be relatively short-lived. In the next section, I will propose that Zhu was struck by the idea that many types of improvements in laws invariably require trading away some other advantage, which suggests the “mixed blessing” view. In any case, in the interest of offering an ecumenical interpretation I will say that both Zhu’s position and the virtue-centered political view more generally allow that there can be some improvements, just not particularly meaningful ones.⁷

It may seem a bit surprising that a core commitment of virtue-centered Confucian politics would be more an objection to institutionalism than a positive claim for the independent efficacy of virtue, but in fact, that seems to be exactly what Zhu Xi had in mind, and also what intellectual historians of Confucian politics have often presupposed. Consider one historian’s succinct description of the intellectual consensus represented by Zhu after the failures of Northern Song reforms:

[In the Southern Song,] there was a growing perception among statesmen and thinkers alike that despite the most earnest attempts during the eleventh century by activist statesmen such as Fan Zhongyan 范仲淹 and Wang Anshi 王安石 to introduce specific political and social reforms, to advance practical measures intended to address the most pressing political, military, and economic problems facing the dynasty, the empire was nonetheless still in danger. In the late eleventh century and the early twelfth century, the barbarian menace to the north continued to loom large, now in the form of the Jurchen tribespeople. The country’s economy remained weak and overburdened, and the Chinese bureaucracy was embroiled in a bitter, paralyzing factionalism. Looking at the failed practical attempts at social and political reform, thinkers concluded that too little attention had been paid by men like Fan and Wang to the inner sphere, to matters of personal morality. . . . These thinkers believed that progress in political and social affairs depended on prior progress in the inner sphere or moral self-cultivation. (Gardner 2007, xxii–xxiii)⁸

Zhu makes his position clear when discussing one of the largest political issues of his day, which has to do with how political authority is apportioned and assigned to regional governors outside of the capital. According to the historical accounts Zhu shared with most Chinese scholar-officials, China had once enjoyed a relatively stable but decentralized system called fengjian 封建, sometimes translated as “feudalism” but which I translate here as the “enfeoffment system.” Under the system of enfeoffment, regional governors were appointed for life and their authority was passed down to their sons, unless of course they had committed crimes so egregious that the monarch or other lords saw fit to remove them. This had the advantage of creating multi-generational, lifelong ties between regional governors and their people, and regional power bases so that the empire would remain resilient even when the central government was weak or dysfunctional. However, that system collapsed, they thought, with the Qin 秦 (221—206 BCE), and in spite of some attempts in Han to re-instate it, it never again took hold. What replaced it was a less stable but more meritocratic system called junxian 郡縣, variously translated as the “county-district system,” the “imperial system,” or (as I prefer) the “commandery system.” The rules of the commandery system changed over the course of the several centuries, but the goal throughout was to ensure that regional governorships only go to those who had performed well on the civil service exams, that positions be rotated on a regular basis, and that governors be prohibited from overseeing districts that included their family or ancestral homes. The aims of this system were to ensure that only those with sufficient moral and practical education be put in positions of authority, and to see to it that regional authorities feel stronger ties and obligations to the emperor and his court than to local residents.⁹

Like most politically-informed scholars of Zhu Xi’s era, Zhu too had a well-considered position, and it appears to have been exactly what one would expect of a philosopher who embraced the second core commitment of virtue-centered politics. His view is that for regimes whose rules are essentially decent, further modifications of the rules will get no traction without getting virtuous people to lead and administrate:

The students were discussing the defects of the commandery and enfeoffment systems. Zhu Xi said, “In general established rules [fa] invariably have defects and no rules are without them. What’s really important in this matter is getting the right person for the job [of implementing and administering those regulations]. If the person is right then even if the rules aren’t good he will still amply make up the difference in score. If the person isn’t right and yet the rules are good, how could this have any benefit for the actual affairs?” (Zhu 1986, juan 108, 2680)¹⁰

Thus, Zhu thought that the rules could vary quite widely—as much as found in the yawning gap between the enfeoffment and commandery systems—and yet still make essentially no meaningful difference in effective governance unless they are accompanied by more virtuous leadership. To be clear, and more precise, Zhu’s mature and final view in this long debate was that reinstituting the enfeoffment system would cause a major upheaval in the social and political order, the costs of which made them more trouble than they were worth. But as the above passage shows, he nevertheless used this debate to reaffirm his core commitment to the idea that institutional reform is essentially fruitless or counter-productive without improvements in the character of the people leading those institutions (Zhu 1986, juan 108, 2682; Ivanhoe 2019, 62, passage 11). People and their virtues are prior to institutional rules in this sense.

With this analysis in mind, we can sum up the virtue-centered view by describing two of its “core commitments.” Both are meant to show how people and their virtues rather than institutional rules are the primary credit-bearing entities for success in governance:

C1: Reforming people is far more demanding than reforming institutional rules: if one wants to improve governance, by far the hardest task which calls for the most concerted effort is to see to it that state officials are virtuous.

C2: Merely reforming institutional rules is unlikely to be effective: given a range of viable institutional rules and less-than-optimal social conditions, merely changing the rules without concomitant improvement in the character of government officials is unlikely to make meaningful improvements to the social order.

By taking these to be the core commitments of the virtue-centered position, we avoid some of the interpretive issues mentioned earlier. For example, we observed previously that there is quite clearly a sense in which Zhu Xi admits that having good laws and regulations (good institutional rules) is necessary for successful governance. C1 admits that this is so, but notes that refining the laws and regulations is not the most demanding part of improving the social order—cultivating a virtuous officialdom is. Moreover, informed readers of Zhu Xi know that he devoted a great deal of effort to refining laws and institutional procedures that he regarded as flawed. C2 can explain how a virtue-centered thinker might find such enterprises justified. It makes sense to advocate for improvements in the rules when (1) the rules fall short of the minimum threshold of basic viability—that is, when they are so perverse or wrongheaded that they set leadership up for failure no matter what, or (2) when those improvements in the rules will be accompanied by concomitant improvement in the character of those who govern with them. This explains how Zhu could object strongly to laws that are so arcane or counter-intuitive that ordinary people find themselves unknowingly violating them—such laws do not even meet the minimum standard of viability (Zhu 1986, juan 108, 2683; Ivanhoe 2019, 65, passage 18). It also explains Zhu’s most notable experiment in regulatory and institutional reform, his famous institution of community granaries (Shecang 社倉), which he developed in the interest of saving lives and maintaining productive farms during periodic famines. These were meant to enlist and supersede the “ever-normal granaries” (Changpingcang 常平倉) and private charities that had failed repeatedly to alleviate the worst effects of periodic famines. Zhu devoted a great deal of time and political capital to establishing and perfecting the community granaries, even pressing wealthy friends and acquaintances to donate to them. In designing and defending his nimbler, and more locally-controlled alternative to the loan and price-stabilization program associated with the evernormal granaries, Zhu went out of his way to explain how his policies and procedures were importantly different from a notorious agricultural loan program of Wang Anshi (the “Green Sprouts” program).¹¹ But Zhu never thought that his reforms would succeed on their own. Quoting Xunzi’s famous line that “people” and not “rules” are the real sources of order, he reminded his readers that his granaries would not succeed without compassionate and honest members of the wealthy and ruling classes who build trust with the people.¹²

The two core commitments of the virtue-centered position also help clarify the different dimensions or ways of measuring virtuecentrism. For example, one way to be a very strong virtue-centric political thinker is to maintain, in the spirit of C1, that the requirements of improving moral character are so demanding as to make the challenges of institutional reform trivial by comparison. The easier institutional reform is relative to cultivating virtue, the more virtuecentric one is. Applying this metric, the historian Hoyt Tillman is right to characterize the Northern Song Confucian Cheng Hao 程顥 (1032—1085) as one of the most extreme representatives of the virtue-centric view. As Tillman writes (1982, 49), for Cheng Hao, “moral intentions were primary to all external and institutional considerations, for he assumed that, once the moral will was firmly established, all else would easily follow from this moral base.”

C2 is a significant core commitment because it suggests two ways of measuring the strength of a philosopher’s virtue-centrism. First, one can be robustly virtue-centric because one thinks that, when the rules meet or surpass the threshold of viability, it is highly improbable that institutional reform alone will make a difference in successful governance. By this standard, most of the famous Song dynastic virtue-centered political thinkers were strongly committed to the view, because they tended to allow that there might be some isolated successes of merely institutional improvements, but tended to doubt that these would have a lasting effect. Second, one could be robustly virtue-centric because one sets the floor level for “rule viability” very low. Remember, C2 says that once rules are good enough to be viable, it is really the moral quality of state officials and not further refinements in the rules that makes a difference. On this issue, many of the Song dynasty’s most famous virtue-centric political thinkers part ways. For example, perhaps one of the staunchest defenders of the virtuecentrism was Hu Hong 胡宏 (1106—1161). But Hu was also one of the most intransigent proponents of the enfeoffment system, proposing that this was the only way to create a state with lasting ties, peace, a kind of equality between members of the same class, and military strength and resilience in the face of poor leadership or succession crises in the central government.¹³ By contrast, as we have seen, Zhu thinks that the range of viable institutional regimes is much broader, such that even the commandery system can succeed with the right leadership. If the leader is right, as Zhu says, “then even if the rules aren’t good he will still amply make up the difference in score” (Zhu 1986, juan 108, 2680; Ivanhoe 2019, 61, passage 10). In contrast, Hu assumes that good leaders will succeed because they jettison the rules of the commandery system and replace them with enfeoffment, not because they can “make up the difference” between good rules and bad through virtuous management. The scope of viable rules is much wider for Zhu, and so, by this measure, Zhu turns out to be a considerably more virtue-centric political thinker.

3. Justifications for the Virtue-Centered View

The locus classicus for the Confucian defense of virtue-centered politics is in the Jundao 君道 (The Way of the Ruler) chapter of the Xunzi. The chapter opens with arguments that echoed through the subsequent two millennia of Chinese political thought and discourse. As these arguments were so familiar as to be frequently taken for granted by later Confucian political thinkers, it is worth looking at them in detail:

There are lords that create chaos; there are no states that create chaos. There are people who create order; there are no rules [fa] that create order. The rules of Archer Yi have not perished, but not every age has an Archer Yi who hits the target precisely. The rules of Yu still survive, but not every age has a Xia dynasty to reign as true kings. Thus, rules cannot stand alone, and categories cannot implement themselves. If one has the right person, then they will be preserved. If one loses the right person, then they will be lost. The rules are the beginning of order, and the gentleman is the origin of the rules. And so, with the gentleman present, even if the rules are sketchy, they are enough to be comprehensive. Without the gentleman, even if the rules are complete, one will fail to apply them in the right order and will be unable to respond to changes in affairs, and thus they can serve to create chaos. One who tries to correct the arrangements of the rules without understanding their meaning, even if he is broadly learned, is sure to create chaos when engaged in affairs. And so, the enlightened ruler hastens to obtain the right person. The deluded ruler hastens to obtain power. (Xunzi 1988, ch. 12, 230)¹⁴

Xunzi’s argument invokes several reasons for his claim that “there are people who create order” but “no rules that create order.” He notes that the rules are by themselves insufficient to effect order (“the rules cannot stand alone”). Rules are often sketchy (or “economical,” sheng 省) and thus leave a great deal to the discretion of state officials, but even when they are complete (ju 具), there will still be problems that call for invention and the good judgment of decision-makers, for there will be cases where one needs to prioritize between rules (as when two rules are at cross purposes, or when one lacks the resources to fully enforce them both, for example), and changes in circumstance will call for revisions of the rules. To be skillful in reprioritizing and revising the rules, one must have a good understanding of their “meaning” (yi 義), which calls for virtue.

In this passage and elsewhere, Xunzi often calls attention to the ways in which rules and models (fa) cannot fully determine good governance, so that state officials will invariably need to exercise personal discretion and judgment in order to execute their responsibilities well. We could call this the argument from underdetermination. Based on my limited experience as an administrator of an academic department, it seems obvious that the argument from underdetermination is correct. Consider the task of choosing, assigning, and scheduling courses for a given semester. When I have devised schedules for my department, I have heeded rules so numerous that they could fill a volume of the tax code, specifying orders of priority for fulfilling faculty entitlements, seat targets, ensuring that only qualified instructors are appointed, and avoiding schedule conflicts within major programs. And yet in spite of this abundance of rules, there are still thousands of possible configurations of course arrangements that would be permitted by the rules, some of them catastrophic for students. I could schedule courses at terribly inconvenient times, such that hundreds of students could be prevented from graduating on schedule, or compel faculty to teach late at night and then again early the next morning, all whilst following the letter of the law, as it were. Good governance of a state is similarly underdetermined but to a far greater degree: the number of rule-adhering options for any complex issue can be multiplied many times over and extended into many different dimensions of decision-making. Perhaps an advocate for a certain libertarian (and utopian) “minimal state” can envision a legal apparatus that leaves less to the discretion of decision-makers, with bright red lines defined by certain basic rights (e.g., private property, bodily autonomy) and little room for positive legislation. But in my view (and certainly on the Confucian view) that would not bring about good governance, for among other things it would not provide even minimally for the legitimate needs and interests of the people, nor support thick relationships and meaningful community bonds. Moreover, some administrative units have to adjudicate and enforce those bright red lines, which invariably involves managing people, making trade-offs between desirable goals, and choosing to prioritize some tasks over others. The problem of underdetermination is inescapable.

Xunzi has other ways of showing how rules are insufficient when they stand alone, without the aid of virtuous officials. He contends that the rules of the sage-king Yu have survived to his present age, and yet it is abundantly clear to him and his contemporaries that the Central States are not well governed. Perhaps he makes this remark only to establish that we cannot have ideal governance without an ideal ruler. But if that’s the extent of Xunzi’s insight into Yu’s contribution to good governance, this sounds suspiciously like a weaker claim for virtue-centered politics discussed in the previous section—the claim that virtue is necessary for optimal governance, a claim that most institutionalists (at least in the Confucian tradition) can readily accept. The real point of contention, as we saw, is whether changes to the rules alone can effect positive change after they have been made good and decent enough to be viable (C2). To address that issue we must turn to Zhu Xi, who adds the following argument: once a set of rules is viable, invariably it will have “defects.” There is a sense in which all sets of rules—even the very best—are defective, and the defects are such that it takes compassionate, wise, publicminded state officials to remedy them.¹⁵ Zhu does not elaborate, but I take it that he sees that once one has a viable set of rules, improving them in some respect will invariably diminish them in another. In the case that he is discussing, he sees disadvantages in both the enfeoffment system (less meritocratic, more difficult to replace bad political authorities) and in the commandery system (impossible to cultivate lasting bonds between the governor and the local community, difficult for the state to survive when the central government is dysfunctional). In the final analysis, Zhu seems to suggest, there is nothing to do but trade one set of disadvantages for the other; there is no significantly better system that can remedy the defects of both. Similarly, Xunzi says that some rules are “sketchy” and others are “complete,” the former being characterized by the fact that they leave a great deal to interpretation (think of rules meant to prevent bullying) and the latter come as close as possible to being fully determinative of state action (as for a tax table, perhaps). There are defects in both levels of permissibility. Sketchy rules are flexible but more easily abused; complete rules are harder to abuse but procrustean. Once the rules have reached a certain level of functionality, such that any further improvements require these sorts of tradeoffs, there is nothing to do but accept one set of defects or another and then turn to virtuous administrators to contain or mitigate them.¹⁶

Another argument is that the virtuous person is the “origin” (yuan 原) of the rules, and from this it appears to follow that most of the credit for successful rules should go to the people who originate them. As he says, “The rules are the beginning of order, and the gentleman is the origin of the rules” (法者, 治之端也; 君子者, 法之原也). It is tempting to read Xunzi as making the relatively obvious point that the gentlemen both precedes the rules in time and plays some part in bringing them about. But this is probably too superficial an interpretation of “origin.” Just because X precedes Y and plays some causal role in bringing it about, it does not follow that most of the credit for Y’s successes should go to X. If that were true, then history’s greatest villains and the law of gravity would get far too much credit for the things they caused. Moreover, if that were Xunzi’s argument then he would run headlong into another chicken and egg problem. Xunzi recognizes that good rulers do not spring from the ground. Good laws, carefully calibrated ritual protocols, and other finely-tuned social conditions must be in place first. If we construe “origin” so loosely then both people and rules originate one another, without any clear bearer of credit at bottom. Finally, Xunzi in this passage is not concerned with origins for the sake of making an historical point: his point in arguing that people are credit-bearing is to show that it is by means of improving people and not improving rules that the real work of good governance is accomplished.

I propose that Xunzi is better understood as appealing not to the mere temporal and causal priority of people to rules, but to a notion that people are originators in a more robust sense. Consider his own analogy: the standards by which Archer Yi measured his success are the same for nearly all archers, and yet it is intuitive to say that Yi’s strength and skill better accounts for his success than those standards. Of course, both the standards and strength and skill have some causal role in successful archery, but we might say that strength and skill are more causally responsible for the success. This notion of causal responsibility is notoriously vexing for philosophers who work on causality and action, but I think it is intuitive enough, and certainly would be intuitive to Xunzi and his audiences. Quite likely, strength and skill count as being more causally responsible because they are more demanding and thus greater achievements than mastery of the (relatively simple) rules of archery. Presumably, Xunzi means to suggest the same thing: acquiring the skills and character traits for governing and implementing rules well (fairly, compassionately, wisely) is considerably more demanding and thus a greater achievement than crafting the rules themselves.

This interpretation is consistent with how virtue-centered political thinkers in the Song tend to understand Xunzi. As noted in the previous section, one of the “core commitments” of later virtue-centered political thought was that reforming people is more demanding than reforming rules (C1). Furthermore, Song political thinkers seem to read Xunzi as proposing that the achievements of virtue are more causally responsible than reforming rules and used analogies to skill and craft to illustrate the point. Here is Hu Hong:

Xunzi said, “There are people who create order; there are no rules that create order.” I humbly submit that we illustrate this by drawing an analogy between wanting to restore order after a period of chaos, and trying to cross a river or lake [by boat]. The rules are like the boat and the people [i.e. the ruler and his officials] are like the steersman. If the boat is damaged and the rudder is broken, then even if [the steersman] has seemingly divine technique everyone nevertheless understands that the boat cannot get across. So whenever there is a period of great disorder it is necessary to reform the rules. There has never been a case where one could successfully restore order without reforming the rules. (Hu 1987, 23-24)¹⁷

Hu takes Xunzi to suggest that both good rules and virtuous people are necessary (“There has never been a case where one could successfully restore order without reforming the rules”). And yet he ultimately agrees with Xunzi that people and not rules are the proper sources of order. The reason seems to have to do with the fact that rule-implementation and rule-making are more demanding, more skill-like, more like steering through rough waters or building boats. In Xunzi’s passage he arguably attends more to the skill of the rule-implementers (the executives, the steersmen) than to the skill of the rule-makers (the legislators, the shipwrights), but that is a matter of emphasis or focus. Xunzi acknowledges that virtuous state officials play a role in making and revising rules as well.

4. Confucian Criticisms of the Virtue-Centered View

As we have seen, the two claims that capture what is most important and controversial about the virtue-centered understanding of politics are (C1) that improving the character of state officials is the more demanding work of governmental reform, and (C2) that after the institutional rules are good enough to be viable, further improvements in the rules will probably be ineffective without concomitant improvements in the character of state officials. We have also seen how Xunzi and the Song Confucians who follow him argue for these claims, which, in part, is to show that virtue is more causally responsible for success in governance than rules are, and to maintain that even viable laws invariably have defects which can only be mitigated by virtuous rule-makers and rule-executors. How could a critic respond to these claims and arguments? One interesting line of response goes as follows: Xunzi and his virtue-centric political disciples think it relatively obvious that people’s virtues have to do the real work of ensuring good governance, that the virtues are analogous to the strength and skill of archers and the institutional rules more closely resemble the standards of archery. Part of what makes this idea appealing is that virtues are needed to ensure that certain outcomes are moral—that governance is fair, compassionate, public-minded, and so on—and it is difficult to see how institutional rules can do that moral work. But maybe it is not so difficult. Perhaps some of the moral work can, through carefully-crafted rules, be offloaded to institutions after all.

One notable example of offloading moral work to institutions is in so-called “advocacy systems,” which appoint people to advocate for opposing sides of an issue (e.g., to represent the defendant or represent the state or people in a criminal case) and incentivizes each person to make the most persuasive argument for their assigned side. When done rightly, the advocacy system is supposed to guarantee a degree of procedural fairness and perhaps even substantive justice, replicating some of the very work which traditionally would have depended on the wisdom and righteousness of the virtuous and fair-minded magistrate. Another example is the system of checks and balances of constitutional governments. At least on some views, the net effect of that sort of such systems is to offload some of the work that would otherwise come from individual virtues (self-control, moderation, loyalty to a state or its people) and substitute structural incentives instead.

If this idea of moral offloading seems odd, then consider a relatively simple thought experiment. Imagine that a savvy player of games, Mei, is joining two other people in playing a game and that she is the sole author of the rules of the game—whatever she declares to be the rules really are the rules. In this context it makes sense to say that Mei herself, as a person, is the “rules authority” for the game, and she bears credit for their success or failure. But now imagine that we develop two different procedures for determining rules of the game, adopting one procedure for the first game and another for the second. The first procedure says that the three players have to agree unanimously to the rules, so that each has an incentive to devise rules that give no one player any special advantages; the other procedure says that each player will propose her own rules and the winner will be determined by a few coin tosses. In the latter case, there is no need to compromise, and self-interested players will be inclined to propose rules that most favor themselves. The players remain the same in personality and talent whether they implement the first procedure or the second, but the different procedures would very likely produce different results. In this light, we can see how it might make sense to give some credit for successful or unsuccessful rules-making to the procedure rather than the people who employ it. Similarly, a critic of virtue-centered politics might say that complex rules for the making of laws and regulations can make the procedures themselves, or the rules that govern those procedures, bear credit for their success.

Many scholars have suggested that Huang Zongxi is among the traditional Confucian political thinkers who comes closest to proposing a system of checks and balances.¹⁸ In his now famous work Mingyidaifanglu 明夷待訪錄 (Waiting for the Dawn), he seems particularly interested in creating systems that check one another by balancing power between competing offices and units of administration. For example, he proposes the reinstitution of a more powerful and independent prime minister (zaixiang 宰相) and establishing ritual protocols meant to nurture mutual respect and deference between the emperor and prime minister (Huang 2011, 27-36; 1993, 100-103). He was particularly concerned that the state provide more mechanisms by which to communicate public interests, which led to his most famous proposals to give greater independence, local control, and freedom of expression to the academies, and to compel the emperor to sit in attendance (as though a student or disciple) as scholars debated public affairs (de Bary 1993, 30-34; Huang 1993, 104-110; 2011, 37-54). He also recommended that crown princes (i.e. the emperors-to-be) be educated outside the walls of the imperial compound, so that they are “informed of real conditions among the people and be given some experience of difficult labor and hardship” and do not develop “false notions of their own greatness” (Huang 2011, 46).¹⁹

At first glance, reforms like these appear to be examples of offloading the moral work of personal virtues to institutions, so that rules carry some of the burden of (and get some credit for) ensuring a moral outcome. So it seems, but I admit that I am somewhat ambivalent about this way of characterizing Huang’s various attempts to balance powers. Firstly, there is some uncertainty about the degree to which Huang’s recommendations were meant as ground-level rules that constrain or delimit the emperor’s own activities, or recommendations that the emperor would be wise to take into account (insofar as he cares about the public good or wants to secure a lasting dynasty) (Chan 2018, 208-209). Secondly, at least some of the described reforms strike me as promising better governance not by offloading moral work from personal virtue to institutions, but instead by creating institutions which more reliably cultivate virtues in leadership. The recommendation that crown princes be educated outside of the imperial compound seems to be meant to make for humbler and more compassionate emperors. If the rules do not constrain rulers at the ground level and if their primary achievement is just to develop more virtuous rulers, this is largely consistent with the account of the proper sources of governance set out by the likes of Xunzi and Zhu Xi. After all, both Xunzi and Zhu Xi readily admit that state officials of good character are more likely to come about under better laws, methods of selection of civil servants, and systems of education, and they see this as quite compatible with the view that people and not rules are the proper sources of order.

Still, some of Huang’s reforms do seem to be genuine examples of moral offloading. By making the office of the prime minister more powerful, independent, and respected, Huang raises the price of certain abuses of power that would otherwise come more cheaply for the emperor—namely, abuses of power that undermine and work against competent prime ministers. Moreover, the rules requiring the emperor to attend relatively free and open discussions of public issues seem to be meant to make the emperor better informed and thus concerned about the interests of his people whether or not he had the inclination and good graces to discover these things for himself. Insofar as that is the case, these would be cases of having well-crafted rules do some of the work that would otherwise be left to individual virtues.

Furthermore, Huang’s political proposals read like the work of someone who appreciates the challenges of institution-building. His interest in the finer points of balancing power and redefining the responsibilities of offices suggest that he thinks that there is much skill and artistry in rule-crafting, that some rule-crafting is more closely analogous to an archer’s strength and skill than to the expectation that the archer hit a target from a certain distance with certain instruments. All of this suggests that Huang would reject the strong virtue-centered claim that reforming people is far more demanding than reforming institutional rules (C1).

Another point of contention is whether reforming rules alone is enough to improve governance without concomitant improvements in character, at least in regimes that already have viable systems of laws (C2). On the face of it, Huang Zongxi seems likely to reject C2 as well. Among the many sorts of reforms that he seems to think will gain traction on their own, two stand out. First, Huang argues that there are some systems of rules that are so “restrictive” and “profuse” (mi 密), and so consistently devoted to protecting the interests of the ruling families, that they create a culture that is inimical to virtuous governance. Huang calls these sorts of rules “unlawful” or “unruly” (feifa 非法) to emphasize that they tend to encourage rebellion and exploitation of the rules, creating an outcome directly opposed to the aims of having rules in the first place (2011, 23-24; 2014, 317). In these situations, fixing the rules must come before improvements in character. This reasoning leads to Huang’s memorable inversion of the Xunzian formula:

Some pundits say, “There are people who create order; there are no rules that create order.” To this I say, “Only if there are rules that create order can there be people that create order.” Since unlawful rules shackle people’s hands and feet, even those that are capable of creating order and unable to overcome the pushing and pulling or the suspicions and doubts that keep them constantly on the lookout. When there is something to be set up or implemented they just finish their own share. They are content to use the most expedient methods and thus unable to achieve anything beyond the sphere [defined by the letter of the law]. If the rules of the former kings still existed, all would have aspirations that go beyond what’s just legally required. . . . This is why I say, “Only if there are rules that create order can there be people that create order.” (Huang 2011, 25)²⁰

One way to read this argument is to say that Huang takes the Xunzian insight that governance is underdetermined by rules but uses it against the Xunzians. When there is a proliferation of restrictive rules meant primarily to protect the interests of the ruling families, few will be inclined to do more than what is minimally required by the letter of the law, and so the extra work that Xunzi thinks so crucial will be left undone.

Although I find Huang’s argument for this sort of reform powerful and poignant, I am not confident that his case is in direct opposition to the virtue-centered political views of Xunzi and Zhu Xi. For one thing, Huang’s argument presupposes that virtue makes a tremendous difference between good governance and bad. It is just that certain institutional changes need to take effect before state officials become capable of developing virtuous approaches to governing. This, as we have seen, is a point that Xunzi and Zhu Xi would readily concede. Furthermore, read charitably, virtue-centered political thinkers never meant to suggest that mere changes in institutional rules would always be ineffective. Rather, they meant to say that they only worked where the rules were fundamentally perverse or dysfunctional, not meeting the minimum threshold of viability. Quite arguably, Huang’s point about the rules of his era is precisely that they are not viable, for they serve primarily the interests of the ruling families and have the paradoxical effect of create disorder and exploitation.

A second set of reforms that Huang proposes work differently. They improve the social order without requiring a substantial improvement in the character of government officials to be effective. Huang sometimes characterizes these sorts of reforms as changes to the structural tendencies and incentives (shi 勢) of an institution, contrasting these with more ancillary changes that rely on prohibitions and punitive laws, which only work sporadically and change behavior superficially. For example, Huang says that corruption amongst lower-level officials (xuli 胥吏) would be better controlled by reinstituting the rotational draft service system (chaiyifa 差役法), which ensures that officials will rotate out before they amass too much knowledge and influence, and makes their interests more continuous with those of ordinary people. Surely this is more effective than relying on prohibitions, and it does not require any meaningful improvement in the internal character of the officials in question (Huang 1993, 162; 2011, 165-166). Here we have an example of a reform proposal that really does imply a rejection of one of the core claims of the virtue-centered view of politics. Defenders of Huang-style institutionalism would do well to build their arguments on proposals like it.

5. Conclusion

The debate about the relative power of rules and human virtues to effect good governance runs deep in the veins of Confucian political discourse, and yet it is easy to misconstrue both the positions and the major arguments offered up by its participants. Both the virtue-centered thinkers and their more institutionalist opponents recognize that personal virtues and good rules are necessary for optimal governance, and both understand that rules and virtues are mutually supporting in crucial respects. In this paper, I have attempted to bring some clarity to the debate by identifying what I take to be the two core commitments of the virtue-centered position—that improving the character of state officials is the more difficult and demanding work of improving governance, and that for systems of rules that are basically decent and viable, further reforms to the rules alone will be ineffective without improvements in character. As we have seen, these two core commitments are shared by the Confucian thinkers most closely associated with the virtue-centered view, notably by Xunzi, Hu Hong, and Zhu Xi. And the core commitments are rejected by great institutionalist Huang Zongxi, although careful examination shows that some of his memorable proposals and arguments are more clearly and unambiguously opposed than others.

I see this framework as no more than a start at imposing some systematicity and order on a large and unwieldy political discourse. To do justice to the depth and sophistication of the Confucian political thinkers, more work needs to be done. For example, the virtue-centered view depends in part on the claim that virtue is more causally responsible for good governance than rules, more like the strength and skill of an archer. Huang Zongxi and other thoughtful crafters of institutions compel us to ask how true this is, and whether there are not ways in which complex rule-making can be analogous to strength and skill in effecting a desired outcome. My reconstruction of the debate here also gives relatively short shrift to Zhu Xi’s interesting contention that all systems of rules have defects which only virtuous state officials can remedy. While I think this argument is quite plausible, it depends in part on how we conceptualize defects, and it is not clear that this pessimism has exactly the implications for virtue-centered politics that Zhu Xi thinks it does. Maybe the best way to mitigate defects in some rules is with higher-order rules.

There is a worry about the way that I have characterized the virtue-centered view. I have tried to show that the dispute is about what can be done to improve governance in a special range of cases—those where the rules are functional enough to be viable. But as we saw in looking closely at Huang Zongxi’s objections, this makes it difficult to pinpoint an actual critic or outsider to the virtue-centered tradition of Confucian political thought, because wherever there is an institutional or legal thinker demanding reforms, one can always try to frame those reforms as a matter of improving less-than-viable rules. Xunzi and Zhu Xi admit that there will be times that the rules are so bad that fixing them will help without a concomitant improvement in virtue. Maybe we should see Huang Zongxi’s proposals in the same light—as fixing deeply flawed rules so as to help them meet the minimum standard of viability.

I have two responses to this worry. First, a lot depends on how high a bar we set for rules that we deem “viable,” and the criteria that the rules must meet in order to count as viable. Surely for the rules to be viable they should at least be sustainable over a long period of time and capable of preventing massive social upheaval or civil war so long as competent administrators are in charge. We might want to stipulate that they should also provide all or nearly all of the people with the means for basic goods like food, shelter, and a livelihood, but I think neither Zhu nor Huang would set the bar so high. From their points of view in history, China had been beset with periodic famines for a long time, both were concerned about the large class of itinerant farmers and profound inequalities of property ownership in the countryside, which they saw as products in part of state corruption. And yet, I think both would have said that the rules of their era were essentially capable of sustaining themselves and maintaining the minimum of social order. So we can take their recommendations for improving governance in their own times as indicative of the sorts of things that should be done for states in the “viable” range. Nevertheless, I admit that there is a problematic ambiguity in my formulation, one that would take more space to resolve than I have here. A second point is that, whatever the ambiguities in my way of characterizing the dispute, there is little question that Confucians in the Song through Ming dynasties understood themselves to be in dispute. The Cheng brothers, Hu Hong, and Zhu Xi took themselves to be articulating a more plausible way forward from the status quo than Wang Anshi had provided, and in Zhu’s own day, more institutionalist thinkers like Chen Liang 陳亮 (1143—1194) and Ye Shi 葉適 (1150—1223) took themselves to be disputing the very proposal that changes in the rules had to be accompanied by improvements in character. Their aim was to show that institutional reform alone could make genuine progress in their day (Niu 1998; Tillman 1982; Xiao 1982, 493-513). Writing more than four centuries later, Huang Zongxi’s views about Zhu’s politics are more nuanced, but as we have seen there are some components of his grand political vision that assume Zhu was wrong about the sources of good governance.²¹

Finally, better elucidating the debate about the relative effectiveness of virtue and institutional rules stands to enrich contemporary political philosophy. There is no debate quite like it in contemporary political thought, and yet it is not hard, I think, to see how the arguments offered by Xunzi, Zhu Xi, and Huang Zongxi would have implications for most any approach to contemporary problems of governance. If we want to figure out how much our own problems are symptoms of defective people or defective institutions, these Confucian philosophers provide us with a much-needed framework.

Acknowledgments

I thank Stephen C. Angle, Eirik Lang Harris, Philip J. Ivanhoe, Halla Kim, Hannah H. Kim, Harvey Lederman, Yao Lin, A. P. Martinich, Gordy Mower, Michael Nylan, Jin Y. Park, Wendy Salkin, Weimin Sun, Jingcai Ying, Ya Zuo, and two anonymous reviewers for helpful discussion and feedback on earlier versions of this paper.

References

• Angle, Stephen C., and Justin Tiwald. 2017. Neo-Confucianism: A Philosophical Introduction. Cambridge, UK: Polity.
• Chan, Elton. 2018. “Huang Zongxi as a Republican: A Theory of Governance for Confucian Democracy.” Dao 17.2: 203-218. [https://doi.org/10.1007/s11712-018-9601-2]
• Chu, Hsi, and Tsu-ch’ien Lü. 1967. Reflections on Things at Hand, translated by Wing-tsit Chan. New York: Columbia University Press.
• De Bary, Wm. Theodore, ed. and trans. 1993. Waiting for the Dawn. New York: Columbia University Press.
• De Bary, Wm. Theodore. 2011. “Waiting for the Dawn: Huang Zongxi’s Critique of the Chinese Dynastic System.” In Finding Wisdom in East Asian Classics, edited by Wm. Theodore de Bary, 199-208. New York: Columbia University Press. [https://doi.org/10.7312/columbia/9780231153973.003.0016]
• Gardner, Daniel K. 2007. The Four Books: The Basic Teachings of the Later Confucian Tradition. Indianapolis: Hackett.
• Hao, Chang. 1974. “On the Ching-shih Ideal in Neo-Confucianism.” Ch’ingshih wen-t'i (Late Imperial China) 3.1: 36-61.
• Hsiao, Kung-ch’uan. 1979. A History of Chinese Political Thought, Volume One: From the Beginnings to the Sixth Century A.D., translated by F. W. Mote. Princeton: Princeton University Press.
• Hu, Hong 胡宏. 1987. Hubongji 胡宏集 (The Collected Works of Hu Hong). Beijing: Zhonghua shuju.
• Huang, Zongxi 黃宗羲. 1993. Waiting for the Dawn, translated by Wm. Theodore de Bary. New York: Columbia University Press.
• Huang, Zongxi 黃宗羲. 2011. Mingyi daifanglu 明夷待訪錄 (Waiting for the Dawn). Beijing: Zhonghua shuju.
• Huang, Zongxi 黃宗羲. 2014. “On Law,” translated by Justin Tiwald. In Readings in Later Chinese Philosophy: Han Dynasty to the 20th Century, edited by Justin Tiwald and Bryan W. Van Norden, 315-318. Cambridge: Hackett.
• Ivanhoe, Philip J., ed. 2019. Zhu Xi: Selected Writings. New York: Oxford University Press. [https://doi.org/10.1093/oso/9780190861254.001.0001]
• Liu, James T. C. 1988. China Turning Inward: Intellectual-Political Changes in the Early Twelfth Century. Cambridge, MA: Harvard University Press. [https://doi.org/10.1163/9781684172702]
• Metzger, Thomas A. 1977. Escape from Predicament: Neo-Confucianism and China’s Evolving Political Culture. New York: Columbia University Press. [https://doi.org/10.7312/metz91032]
• Niu, Pu. 1998. “Confucian Statecraft in Song China: Ye Shi and the Yongjia School.” PhD diss. Arizona State University.
• Qian, Mu 錢穆. 1966. Zhongguo jin sanbainian xueshushi 中國近三百年學術史 (A History of Scholarship in China’s Three Recent Centuries). Taipei: Taiwan shangwu yinshu guan.
• Struve, Lynn A. 1988. “Huang Zongxi in Context: A Reappraisal of His Major Writings.” Journal of Asian Studies 47.3: 474-502. [https://doi.org/10.2307/2056971]
• Tillman, Hoyt Cleveland. 1982. Utilitarian Confucianism: Ch’en Liang’s Challenge to Chu Hsi. Cambridge, MA: Harvard University Press. [https://doi.org/10.1163/9781684172351]
• Tiwald, Justin. 2016. “Xunzi among the Chinese Neo-Confucians.” In Dao Companion to the Philosophy of Xunzi, edited by Eric L. Hutton, 450-489. Dordrecht: Springer. [https://doi.org/10.1007/978-94-017-7745-2_15]
• Von Glahn, Richard. 1993. “Chu Hsi’s Community Granary in Theory and Practice.” In Ordering the World: Approaches to State and Society in Sung Dynasty China, edited by Robert P. Hymes and Conrad Schirokauer, 221-254. Berkeley and Los Angeles, CA: University of California Press. [https://doi.org/10.1525/9780520378223-008]
• Yu, Yingshi 余英時. 2004. Songming lixueyu zhengzhi wenhua 宋明理學與政治文化 (Song-Ming Lixue Confucianism and Political Culture). Taipei: Yunchen 允晨.
• Xiao, Gongquan 蕭公權. 1982. Zhongguo zhengzhi sixiangshi 中國政治思想史 (A History of Chinese Political Thought), vols. 1 and 2. Taipei: Lianjing 聯經.
• Xunzi 荀子. 1988. Xunzi jijie 荀子集解 (Xunzi, with Collected Explanations). Edited by Wang Xianqian 王先謙. Beijing: Zhonghua shuju.
• Xunzi 荀子. 2014. Xunzi: The Complete Text. Translated by Eric L. Hutton. Princeton: Princeton University Press.
• Zhu, Xi 朱熹. 1986. Zhuzi yulei 朱子語類 (The Classified Sayings of Master Zhu). Edited by Li Jingde 黎靖德. Beijing: Zhonghua shuju.
• Zhu, Xi 朱熹. 2000. Zhuzi wenji 朱子文集 (The Collected Writings of Master Zhu). Edited by Chen Junmin 陳俊民. Taipei: Defu wenjiao jijinhui.
• Zhu, Xi 朱熹, and Zuqian Lü 呂祖謙, eds. 2008. Jinsilu 近思錄 (Reflections on Things at Hand). Zhengzhou: Zhongzhou guji chubanshe.